FACTS

I.

GUILT PHASE EVIDENCE

The evidence presented at the guilt phase of Davenport’s trial was described by the California Supreme Court in Davenport I as follows: “The uncontradicted evidence showed that the victim, Gayle Lingle, spent the evening of March 26, 1980, at the Sit ‘N Bull Bar in Tustin, where she talked with [Davenport] and with Larry Richards, both of whom she knew. At one point in the evening she asked Richards to give her a ride home. Richards put her off and [Davenport] offered her a ride. She did not then accept the offer. Later in the evening she telephoned her boyfriend telling him that she had a six-pack of beer and a ride and would be home soon. A few minutes later between approximately midnight and 1 a.m., she

2 Undesignated statutory references are to the Penal Code.

3 and [Davenport] both left the bar. The victim’s body was found the next morning lying in a large, uncultivated field south of the I-5 freeway near Tustin. “The area where the body was found is bounded by Myford and Michelle Streets; a dirt berm two to three feet high lies between Myford Street and the field. Between the road and the berm was an area in which items from a woman’s purse were scattered among random, twisting footprints. This evidence suggested a struggle had occurred in that location. Behind the berm, about 20-30 feet from the body, was a large pool of bloodied water, a pair of woman’s pants, a woman’s underpants, and a pair of boots. There were two sets of motorcycle tracks in the area. One set left and re- entered Myford in the area of struggle. The second set of tracks led from Myford, over the end of the dirt berm within four to five feet of the body, and back to the street. “[Davenport]’s nickname was ‘Honda Dave.’ Police learned that he had left the Sit ‘N Bull with the victim on the night of her death and that he rode a 350 cc Honda motorcycle. On March 28 the officers, who had a warrant to arrest [Davenport] for an unrelated traffic offense, entered the open carport at [Davenport]’s home where they examined and photographed the tires on [his] bike. The examination indicated a possible match between the tracks at the murder scene and the tread on [Davenport]’s tires. The officers went to the house and arrested [Davenport]. “The prosecution produced three eyewitnesses who placed a motorcycle similar to one owned by Davenport at the murder scene between 12:30 and 1:30 a.m. on March 27. The first witness had left his workplace on Myford Road shortly after 1 a.m. As he passed by the intersection of Michelle and Myford he saw a motorcycle parked beside the road which looked similar

4 to a 350 cc Honda that his son owned. The second witness testified that she also had driven past the intersection about 12:30 a.m. and that she had seen a motorcycle parked off the road near a telephone pole. She had seen a man crouched on the far side of the dirt berm. The man wore a plaid shirt and had shoulder length hair. She could not ascertain the nature of his activity. The motorcycle was dark in color, and of a foreign, probably Japanese, make. This witness did not identify [Davenport]. A third witness testified that she had driven around that corner at 12:35 a.m. As her headlights swept the field she saw a parked motorcycle and a man who was crouched over and appeared to be digging. The motorcycle resembled the bike belonging to [Davenport,] which was in the courtroom as a prosecution exhibit. The man was white and had collar length hair. She saw his face when he briefly looked into her headlights. Prior to trial this witness had picked [Davenport]’s photo as one of three which could have been the man she saw. She was unable to select [Davenport] from a live lineup. At trial the district attorney displayed the photographic line-up from which the witness had originally selected several pictures. She selected a photograph of [Davenport] and testified that she thought that was the person she had seen in the field although she could not be positive that it was. “Three expert witnesses testified to facts that connected [Davenport]’s motorcycle to the crime. Bonnie Driver, a criminalist employed by the Orange County Sheriff’s Department, testified that she had examined vegetable matter taken from [Davenport]’s motorcycle and compared it with vegetation taken from the area where the victim’s body was found. Driver found the gross morphology of the plants in both samples to be consistent with each other. A forensic microscopist, Skip Pallinick, examined and compared the heavy mineral content of soil samples taken from [Davenport]’s

5 bike with samples taken at the murder scene. He testified that the samples were generally consistent with each other. One of the samples from the motorcycle contained sufficient similarity to the murder scene samples that the witness concluded they were virtually indistinguishable. Both of these witnesses admitted they had not compared the samples taken from [Davenport]’s bike with samples taken from other parts of Orange County. Dr. Stephen Dana, a geologist retained by the [defense], examined the same soil samples and found similarities and differences in all of them. Based on his knowledge of the geology of the area Dr. Dana opined that the samples could have come from anywhere in Orange County. “Jack Leonard, the production manager for the International Sport and Rally Division of Dunlop Tire Company, testified that [Davenport]’s motorcycle could have made the tracks at the scene. Mr. Leonard had examined plaster casts of the tracks found near the victim’s body and the actual tires on Davenport’s motorcycle. He found that the tracks of the rear tire had the same highly unique and distinctive characteristics as the rear tire of the motorcycle. Both were Dunlop brand motorcycle tires, size 4.00-18 with a K-70 tread pattern. Both were characterized by a defect in a portion of the tread pattern known as the cross-slot. The defect was caused by damage to the upper half of the manufacturing mold. Mr. Leonard testified that tires with such blemishes are routinely removed from the regular production line. They are sold as ‘blems’ at a reduced price but pursuant to trade restrictions are not shipped to the U.S. It was Leonard’s opinion that not more than two or three tires with this defect which were manufactured at the same time and place as the tire on [Davenport]’s bike would have eluded inspectors and been shipped to the U.S. Similar tires are currently manufactured in the U.S., but the steel utilized for the molds here is stronger

6 than the aluminum used in Britain and not subject to the kind of breakage which caused this defect. The witness also explained that it was unusual for a nine-year-old tire to show only 40 percent wear, as did the tire on [3] [Davenport]’s motorcycle. The vast majority (95 percent) of similar tires are completely worn out after five or six years of use. This degree of wear was consistent with the tracks at the scene. Finally, Leonard testified that the track of the front motorcycle tire at the scene showed a tread pattern which he recognized as a Bridgestone tire, similar to the front tire on [Davenport]’s motorcycle. “The nature of the charge against [Davenport] requires that we describe the grotesque details of the condition of the victim’s body and the cause of her death. The body, nude except for a sweater draped over the upper torso, had suffered numerous and extensive stabbing and slashing wounds about the neck, chin and jaw as well as defensive wounds to the hands and forearms. The victim’s carotid artery had been severed. Her breasts were bruised. A long wooden stake had been inserted into her rectum. An autopsy revealed that the stake was forced deeply into the body, injuring various internal organs and tissues before it came to rest just under the right armpit. “The autopsy pathologist, Dr. Walter Fischer, testified that the cause of death was the loss of blood from the carotid artery, and that the insertion of the wooden stake was a contributing factor. Dr. Fischer testified that it was his opinion that the victim was alive at the time the wooden stake was inserted, because the 200 cc’s of blood which he found in the pleural

3 “Code numbers on the rear tire of [Davenport]’s bike showed it had been manufactured in the spring of 1972 in Leicester, England and indicated the particular mold that had been used.”

7 cavity could only have been pumped out of the perforation in the lung by a beating—and thus living—heart. Dr. Fischer did not believe that the surprisingly small amounts of blood found at the point where the stake contacted the anus and liver contradicted his conclusion. He attributed the lack of further bleeding to the probability that the victim’s body had entered a state of shock after the carotid artery was cut and before the stake was inserted. When in shock, he explained, blood is channeled to the most vital organs—primarily the brain, heart and lungs—and the vessels in organs of lesser immediate importance, such as intestines, contract. “Dr. Fischer testified that the neck and hand wounds had probably occurred before the penetration of the stake. He testified that the victim could have lived from 5 to 20 minutes after the artery was severed. He could not offer a medical opinion that the victim was or was not conscious when the stake was inserted. “Dr. Renee Modglin, an experienced pathologist, reached different conclusions. Based on an examination of photographs of the victim’s body and Dr. Fischer’s autopsy report, Dr. Modglin formed the professional opinion that the stake had probably entered the victim’s body after her death. He suggested blood could have entered the lung either through one of the initial cuts in the carotid artery or during post-mortem movements of the body. He offered his opinion that a state of shock would not have prevented bleeding in all of the areas which the stake had contacted. “The prosecution also called as a witness John Farmer who testified to incriminating admissions [Davenport] made to him while both were in a holding cell at the Orange County Superior Court at various times between September 8, 1980, and December 19, 1980. According to the testimony, [Davenport] told Farmer he had killed a girl, that he had

8 sodomized her while she yelled for him to stop, that he had stabbed her and tried to cut off her head but could not succeed because his knife was too small. [Davenport] said that after he stabbed her, he had lowered the girl’s body onto a wooden stake because he wanted to make a scarecrow of her. When the stake came out of the ground, he had pushed and kicked it further into her body. According to Farmer, [Davenport] variously said he had killed her because she had sexual relations with another man and because she had led [Davenport] on. “[Davenport]’s testimony was consistent with that of the other witnesses who had been at the Sit ‘N Bull Bar on March 26. After cashing his paycheck at a local bowling alley, he had gone to the bar around 8:30. There he played pool, drank beer and visited with people. He was acquainted with the victim and with her boyfriend. Lingle had asked Larry Richards for a ride home but Richards said she would have to wait until he finished playing pool. [Davenport] had then offered her a ride but she had not replied. Some time later after she indicated she wished to purchase some beer to take home [Davenport] bought her a six-pack. His motive was to return the favor the victim and her boyfriend had done for [him] when he was too broke to buy himself a drink. He walked out of the bar with her and carried the beer. In the parking lot he gave her the six-pack and again offered her a ride; she again declined. [He] left on his motorcycle and went to a nearby coffee shop, alone, and drank two cups of coffee. He then went home to bed. The following morning [Davenport] got up at a quarter to seven, showered and headed for his job at Apri Cosmetics, in Orange. He was due to begin work at 8 a.m. but his motorcycle developed a flat tire in Tustin. [He] stopped and called his supervisor. At 8:30 a.m. he purchased a new inner tube from Tustin Honda and changed the tire. He arrived at work at 9:30 a.m.

9 “[Davenport]’s girlfriend testified that she had gone to sleep at 10 p.m. on March 26. [He] was not at home at that time but was in their bed when she awoke at 6 the following morning. [Davenport] was dressed only in his undershorts and she could see his arms, legs and torso. There were no cuts, scratches or blood on him. She had heard no sounds during the night which indicated he had showered or cleaned himself or his clothes. “Susan Stock, [Davenport]’s supervisor at Apri Cosmetics, testified that she had tried to verify [his] excuse for being tardy on March 27. She had examined the bike and formed the opinion that neither tire had recently been changed. Officer Veach, an investigator for the Irvine Police Department examined the tires on [Davenport]’s motorcycle after his arrest. He also formed the opinion that neither tire had been recently changed. Paul Jorgensen, the manager of Tustin Honda, testified that the business opened at 9 a.m. and that his sales records for March 27 showed no sale of a motorcycle tube such as that [Davenport] claimed to have purchased.” (Davenport I, supra, 41 Cal.3d at pp. 256–260.)

II.

PENALTY RETRIAL EVIDENCE

A thorough description of the evidence presented at Davenport’s penalty retrial can be found in Davenport II, supra, 11 Cal.4th at pages 1190 to 1192. We provide a limited summary for context and as relevant to the issues in this appeal. At the penalty retrial, the prosecution presented much of the evidence from the guilt phase trial, including the evidence depicting how Lingle was killed and the evidence connecting Davenport to the murder. (Davenport II, supra, 11 Cal.4th at p. 1190.) The prosecution, however, did not present Farmer’s testimony concerning Davenport’s admissions.

10 Dr. Fischer, who performed Lingle’s autopsy and testified at Davenport’s guilt phase trial, was deceased by the time of the penalty retrial. Dr. Fukumoto, a pathologist who had been Fischer’s partner at the time of the Lingle autopsy, opined Lingle was alive when she was impaled by the stake. (Davenport II, supra, 11 Cal.4th at p. 1190.) The prosecution presented evidence of two instances of violent criminal activity by Davenport as aggravating factors warranting the death penalty. One incident involved a woman, who awoke in her apartment to find Davenport, a stranger, “straddling, smiling, and stabbing her.” (Davenport II, supra, 11 Cal.4th at p. 1191.) Davenport “continued smiling as he stabbed her in the throat, severing her jugular vein, and in her eye, breast, lung, stomach, arm, and hand.” (Ibid.) She suffered 23 stab wounds but survived. (Ibid.) As a result of this incident, Davenport was convicted of burglary and assault with a deadly weapon in 1975. He was released from prison in August 1979, seven months before Lingle was murdered. (Id. at p. 1191, fn. 2.) The other incident occurred while Davenport was in custody prior to the start of his guilt phase trial for Lingle’s murder. Davenport and other inmates were in a holding tank. After one of the inmates gave his sandwich to a cellmate, Davenport and others demanded he do something for them. The man was beaten, pulled into the bathroom, and forced to orally copulate several inmates, including Davenport. Davenport also forcefully sodomized the man. (Davenport II, supra, 11 Cal.4th at p. 1191.) The defense “presented evidence intended to create a lingering doubt as to the guilt phase jury’s finding that Lingle was alive when she was impaled by the stake, an element of the special circumstance of torture murder. Stephan Schliebe, a criminalist for the California Laboratory of Forensic Science, examined the stake. His tests revealed no evidence of blood

11 on the stake or the nails on the stake. He stated that the passage of time would not preclude a finding of blood. However, he conceded that he did not know the effect of silver nitrate, a chemical used to reveal fingerprints, on blood after a nine-year lapse. [¶] Dr. Ronald Kornblum, a medical examiner for Los Angeles County and pathologist, testified that in his opinion Lingle died before she was impaled by the stake.” (Davenport II, supra, 11 Cal.4th at pp. 1191–1192.) The prosecution presented rebuttal evidence the stake had been treated with silver nitrate. (Id. at p. 1192.)

III.

HABEAS CORPUS PROCEEDINGS

In 2017, in response to a postconviction discovery request by Davenport’s federal habeas counsel, investigators with the district attorney’s office interviewed Therrien. In the interview, Therrien contradicted Farmer’s trial testimony that Davenport admitted murdering a woman. Therrien stated Davenport did not confess to committing murder but only told Farmer of the accusations against him. After the district attorney’s office provided Davenport’s federal habeas counsel a tape recording of the investigators’ interview, Davenport filed a habeas petition in the superior court in January 2018. Davenport’s habeas petition included the audio recording and transcript of Therrien’s 2017 interview, as well as a declaration by Therrien (Therrien’s 2017 statement). In Therrien’s 2017 statement, he declared he was previously contacted by an investigator with the district attorney’s office in 1980 or 1981 (prior to Davenport’s guilt phase trial) and similarly told them Davenport did not confess (Therrien’s original statement). A declaration by Davenport’s original trial counsel was also included with his habeas

12 petition. Until provided Therrien’s 2017 statement, trial counsel was unaware of a statement by Therrien contradicting Farmer. Davenport contended his habeas petition was not “‘successive’” as defined in section 1509, subdivision (d), and even if it was, an exception applied because he could show by a preponderance of the evidence he was innocent of the special circumstance and therefore ineligible for the death penalty. He asserted “the torture-murder special circumstance was based on false forensic evidence regarding the time and manner of Gayle Lingle’s death.” In support, he submitted a declaration by forensic pathologist Dr. Terri Haddix, disagreeing with the pathologists’ testimony at his trials that Lingle was alive when the stake was inserted. Davenport presented five claims in his habeas petition. First, he argued Therrien’s original statement was material exculpatory evidence and its suppression violated the prosecution’s obligations under Brady v. Maryland (1963) 373 U.S. 83 (Brady). Second, Davenport asserted, in the alternative, if the prosecution did disclose Therrien’s original statement before trial, his trial counsel was ineffective for failing to investigate and use the evidence to impeach Farmer. Third, he contended the prosecution knowingly produced false evidence (Farmer’s testimony) in violation of his due process rights under Giglio v. United States (1972) 405 U.S. 150 (Giglio) and Napue v. Illinois (1959) 360 U.S. 264 (Napue). Fourth, he asserted he was entitled to relief based on the prosecution’s destruction of Therrien’s original statement as no memoranda or recordings concerning the interview now exist. Fifth, he asserted he was entitled to relief under section 1473 based on new evidence of innocence and the prosecution’s presentation of false evidence at trial. Within this last claim, he asserted the prosecution presented three categories of false evidence: (1) Farmer’s testimony

13 Davenport confessed to murdering Lingle; (2) pathologist testimony Lingle was alive at the time the stake was inserted; and (3) testimony the tires on Davenport’s motorcycle matched tire tracks found at the crime scene. In a written order, the superior court denied Davenport’s habeas petition, concluding he had not established a prima facie case for relief on his claims.4

DISCUSSION

Davenport contends he presented a prima facie case for relief on each of his claims and the superior court therefore erred by denying his habeas petition without issuing an order to show cause. We find no error.

I.

PROPOSITION 66

Prior to Davenport filing this habeas petition in the superior court, the electorate approved Proposition 66, the Death Penalty Reform and Savings Act of 2016 (Gen. Elec. (Nov. 8, 2016) § 1) (Proposition 66). Proposition 66 altered the procedures for collateral review of convictions and sentences in capital cases. (In re Friend (2021) 11 Cal.5th 720, 723 (Friend); In re Robinson (2019) 35 Cal.App.5th 421, 423, 425.) It added sections 1509 and 1509.1, which govern the processes for capital defendants adjudicating their petitions for writs of habeas corpus and seeking review if the petitions are denied in the superior court. (Friend, at pp. 725–726.) Section 1509 is “the exclusive procedure for collateral attack on a judgment of death.” (§ 1509, subd. (a).) This statute requires capital defendants to file their habeas petitions in the superior court that imposed

4 We discuss each of Davenport’s claims, the facts underlying them, and the superior court’s ruling in greater detail in the discussion section post.

14 the death sentence (rather than the California Supreme Court). (Ibid.) Section 1509 requires a superior court to dismiss an untimely initial petition or “a successive petition” “unless the court finds, by the preponderance of all available evidence, whether or not admissible at trial, that the defendant is actually innocent of the crime of which he or she was convicted or is ineligible for the sentence.” (Id., subd. (d).)5 Subdivision (d) further defines “‘[i]neligible for the sentence of death’” as meaning “circumstances exist placing that sentence outside the range of the sentencer’s discretion” and a claim of ineligibility includes “a claim that none of the special circumstances in subdivision (a) of Section 190.2 is true.” (§ 1509, subd. (d).) While Proposition 66 did not express what constitutes a “successive petition,” the California Supreme Court concluded the term should be defined consistent with its use in prior decisions. (Friend, supra, 11 Cal.5th at p. 731.) In Friend, the California Supreme Court explained: “‘[C]laims presented in a “subsequent” petition that should have been presented in an earlier filed petition will be barred as “successive” unless the petitioner “adequately explains” his or her failure to present all claims in the earlier filed petition.’” (Id. at p. 732.)6 The corollary is that “a claim will not be barred as successive, even though it may be presented in a second or subsequent habeas corpus petition, if the petitioner offers adequate justification for the failure to present a particular claim in an earlier petition.” (Id. at p. 731.) Thus, “Proposition 66’s procedural bar on successive

5 Section 1509, subdivision (c) also establishes a deadline for filing an initial habeas petition. We do not detail this provision as it is irrelevant to the matter before us.

6 A “‘subsequent’” habeas petition is not necessarily the same as a “‘successive petition.’” (Friend, supra, 11 Cal.5th at p. 731.)

15 claims only applies when a petitioner filing a second or subsequent habeas corpus petition has not adequately justified his failure to present the claims in an earlier petition.” (In re Maury (2024) 105 Cal.App.5th 645, 663.) When a capital habeas corpus petitioner files a subsequent petition, the superior court must determine whether the “successiveness bar” applies by examining whether the petitioner has “adequately justified” the failure to present the claim in a previous petition. (Friend, supra, 11 Cal.5th at p. 728.) “Adequate justifications include the inability to bring the claim earlier, as where the claim depends on newly available evidence or on a change in the law that has been made retroactively applicable to final judgments.” (Ibid.) If the petitioner is unable to provide an adequate justification, the successive claim will be dismissed unless the petitioner makes a showing of actual innocence or ineligibility for the death penalty. (§ 1509, subd. (d).) Section 1509.1 governs the appeal of a decision on a petition for writ of habeas corpus filed by a capital defendant and “imposes a related procedural hurdle for litigants seeking to appeal the denial of a successive petition.” (Friend, supra, 11 Cal.5th at p. 727.) This statute allows capital habeas petitioners to appeal the superior court’s decision “denying relief on a successive petition only if the superior court or the court of appeal grants a certificate of appealability.” (§ 1509.1, subd. (c).) Subdivision (c) of section 1509.1 provides: “A certificate of appealability may issue under this subdivision only if the petitioner has shown both a substantial claim for relief, which shall be indicated in the certificate, and a substantial claim that the requirements of subdivision (d) of Section 1509 have been met.” (Ibid.) This provision further specifies: “The superior court shall grant or deny a certificate of appealability concurrently with a decision denying relief on the

16 petition.” (Ibid.) If the superior court denies a certificate of appealability, “[t]he court of appeal shall grant or deny a request for a certificate of appealability within 10 days of an application for a certificate.” (Ibid.) The court of appeal’s jurisdiction “is limited to the claims identified in the certificate and any additional claims added by the court of appeal within 60 days of the notice of appeal.” (Ibid.)

II.

MOTION TO DISMISS

7 We begin with the Attorney General’s motion to dismiss. Raising the issue for the first time on appeal, the Attorney General contends all of Davenport’s claims in his habeas petition are successive. The Attorney General asserts Davenport was therefore required to obtain a certificate of appealability to appeal the superior court’s denial of his habeas petition. Because Davenport did not obtain a certificate of appealability, the Attorney General moves for dismissal of the appeal or, alternatively, for remand to the superior court for it to rule on a request for a certificate of appealability. We deny the motion for reasons we explain below. A. Background In the superior court, Davenport asserted his habeas petition was not successive for purposes of section 1509, subdivision (d) and even assuming it was, he could establish he was ineligible for the death penalty. In its informal response, the district attorney’s office did not argue any of Davenport’s claims should be denied as successive. Instead, it argued

7 In the superior court, the Orange County District Attorney’s Office filed an informal response on behalf of respondent to Davenport’s habeas petition. On appeal, the Attorney General’s Office represents respondent.

17 Davenport’s claims should be denied on their merits. In denying the petition, the superior court did not find any of Davenport’s claims to be successive. Davenport filed a notice of appeal in May 2019. He did not request a certificate of appealability as to his claims, and the superior court did not issue one. The superior court clerk prepared the appellate record, and Davenport filed his opening brief in this court. More than a month after Davenport filed his opening brief, the Attorney General moved to dismiss Davenport’s appeal on the ground his claims were successive and he was required to obtain a certificate of appealability to appeal the denial of his habeas petition. Davenport opposed the dismissal motion, arguing: (1) the contention regarding successive claims was forfeited because it was not raised below; (2) the Attorney General should be estopped from pursuing a defense not raised in the superior court; (3) dismissal of his appeal would be unfair as he had not had an opportunity to counter the Attorney General’s assertion with additional evidence and argument; (4) his claims were not successive, but if they were, the petition meets the requirements for a certificate of appealability; and (5) the Attorney General’s “tactic” would potentially delay the appeal by injecting unadjudicated issues of fact into the proceeding well after the superior court’s ruling. This court issued an order indicating it would decide the motion in conjunction with the appeal. B. Analysis The Attorney General concedes no argument was made in the superior court that Davenport’s claims should be dismissed as successive. He also recognizes the superior court did not find Davenport’s claims to be successive. Nevertheless, the Attorney General asserts Davenport was required to obtain a certificate of appealability “as a condition precedent to

18 this appeal” and “[h]is failure to do so deprives this court of jurisdiction over the appeal, requiring dismissal.” We disagree. Because the superior court did not deny Davenport’s claims as successive, a certificate of appealability was not required under section 1509.1, subdivision (c). (In re Hill (2024) 104 Cal.App.5th 804, 823 & fn. 11 (Hill); Friend, supra, 11 Cal.5th at p. 747.) The prosecution did not argue Davenport’s claims should be denied as successive until after Davenport filed his opening brief. This was too late. If the respondent does not raise a procedural bar when responding to a petition for writ of habeas corpus, a court can find the argument forfeited on appeal. (See In re Moser (1993) 6 Cal.4th 342, 350, fn. 7 [People’s failure to raise issue concerning timeliness of petition for writ of habeas corpus forfeits the issue on review]; In re Rothwell (2008) 164 Cal.App.4th 160, 172, fn. 5 [concluding the prosecution forfeited argument the petition for writ of habeas corpus was untimely because the argument was not made in the informal response, which it agreed would be deemed the return].) Because the Attorney General did not argue in the superior court that Davenport’s claims should be denied as successive, he may not seek dismissal of Davenport’s appeal on this ground. Generally, a party is not permitted to change his or her defense and adopt a new and different theory on appeal that was not litigated below. “‘To permit him to do so would not only be unfair to the trial court, but manifestly unjust to the opposing litigant.’” (Cable Connection, Inc. v. DIRECTV, Inc. (2008) 44 Cal.4th 1334, 1350, fn. 12; see also In re Blake (1979) 99 Cal.App.3d 1004, 1022 [rejecting district attorney’s argument based on appellate principle “that a party may not change his position and adopt a new theory on appeal”].) In the superior court, the district attorney’s office, on behalf of respondent, took the position Davenport’s habeas petition claims

19 should be denied on their merits and did not argue Davenport’s claims should be denied as successive. It would be unjust to permit the Attorney General to argue at this stage, on behalf of respondent, that the superior court should have rejected Davenport’s claims as successive and Davenport’s appeal should be dismissed because he did not obtain a certificate of appealability to appeal successive claims. Had the issue of whether the claims in Davenport’s habeas petition were successive been raised in the superior court, it would have permitted both parties an opportunity to develop the record and the superior court to make findings concerning the successiveness of Davenport’s claims. This fact finding must take place first in the superior court as it is the initial arbiter of whether a petitioner’s claims are successive. (§ 1509, subd. (d).) Davenport further argues the Attorney General should be estopped from taking a position on appeal inconsistent with the position below. But we need not decide whether to exercise our discretion and apply the judicial estoppel doctrine, given the reasons we have already expressed to deny the Attorney General’s motion to dismiss. (See People v. Castillo (2010) 49 Cal.4th 145, 155 [“Application of the doctrine is discretionary”].) The Attorney General alternatively seeks remand to allow the superior court to determine whether Davenport’s habeas petition claims are successive. This would only further delay Davenport’s appeal. Because we affirm the superior court’s denial of Davenport’s claims on their merits, we need not further delay his appeal.

20

III.

STANDARD OF REVIEW

Although Proposition 66 changed several procedures for petitions for writs of habeas corpus in capital cases (§ 1509), the basic requirements remain unchanged. “Because a petition for a writ of habeas corpus seeks to collaterally attack a presumptively final criminal judgment, the petitioner bears a heavy burden initially to plead sufficient grounds for relief, and then later to prove them. ‘For purposes of collateral attack, all presumptions favor the truth, accuracy, and fairness of the conviction and sentence; defendant thus must undertake the burden of overturning them.’” (People v. Duvall (1995) 9 Cal.4th 464, 474; accord, In re Figueroa (2018) 4 Cal.5th 576, 587.) “The petition should both (i) state fully and with particularity the facts on which relief is sought [citations], as well as (ii) include copies of reasonably available documentary evidence supporting the claim, including pertinent portions of trial transcripts and affidavits or declarations. [Citations.] ‘Conclusory allegations made without any explanation of the basis for the allegations do not warrant relief, let alone an evidentiary hearing.’” (People v. Duvall, at p. 474.) Upon receiving a habeas petition, “a court must first determine whether the petition states a prima facie case for relief—that is, whether it states facts that, if true, entitle the petitioner to relief—and also whether the stated claims are for any reason procedurally barred. [Citation.] To assist the court in determining the petition’s sufficiency, the court may request an informal response from the petitioner’s custodian or the real party in interest.” (People v. Romero (1994) 8 Cal.4th 728, 737; Cal. Rules of Court, rule 4.573(a).) If “the court finds the factual allegations, taken as true, establish a prima facie case for relief, the court” must issue an order to show

21 cause (People v. Duvall, supra, 9 Cal.4th at p. 475), “directing the respondent to show cause why the relief sought should not be granted based on those allegations.” (In re Clark (1993) 5 Cal.4th 750, 781, fn. 16.) However, if the habeas petition does not state a prima facie case for relief, the court will deny the petition. (Ibid.) The parties agree we conduct a de novo review of the superior court’s denial of Davenport’s habeas petition. “Our standard of review is de novo with respect to questions of law and the application of the law to the facts. We accept as final the superior court’s resolution of pure questions of fact if they are supported by substantial evidence.” (In re Richards (2012) 55 Cal.4th 948, 960, superseded on other grounds by change in statute as explained in In re Richards (2016) 63 Cal.4th 291, 293 (Richards II); accord, Hill, supra, 104 Cal.App.5th at p. 825.)

IV.

THE SUPERIOR COURT PROPERLY DENIED DAVENPORT’S CLAIMS

CONCERNING THERRIEN’S STATEMENTS

Davenport presents several claims pertaining to Therrien’s statement Davenport did not confess to the murder when talking to Farmer. Davenport’s first two claims are related. First, Davenport contends the prosecution’s failure to disclose Therrien’s original statement prior to trial constitutes a Brady violation and the superior court erred by rejecting this claim. Second, Davenport asserts, in the alternative, his trial counsel rendered constitutionally deficient assistance by not investigating and presenting Therrien’s testimony at trial. Davenport argues he is entitled to relief on either his Brady or ineffective assistance of counsel claim. He asserts if Therrien’s potential testimony was not disclosed, he has satisfied the suppression requirement for a Brady claim. And if Therrien’s potential

22 testimony was disclosed, his trial counsel’s failure to investigate and present it constitutes deficient performance, one of the two elements for establishing constitutionally defective assistance of counsel under Strickland v. Washington (1984) 466 U.S. 668 (Strickland). The superior court was required to issue an order to show cause on these claims only if Davenport made a prima facie showing of a reasonable probability the result of his trial would have been different if the jury had heard Therrien’s statements or testimony. We conclude Davenport failed to make the required showing, and the superior court properly rejected these claims. As we explain below, while Therrien’s 2017 statement would have impeached Farmer’s testimony, it does not undermine our confidence in the outcome of Davenport’s trial. We similarly reject Davenport’s two other claims relating to Therrien’s statements: the prosecution knowingly presented false evidence at trial by presenting Farmer’s testimony and the loss of Therrien’s original statement violates his state and federal constitutional rights. A. Background

The documentary evidence in the habeas corpus proceedings below indicates Davenport, Farmer, and Therrien were in custody in the county jail and transported to the superior court for proceedings in their individual cases on November 21, 1980. 1. Farmer’s Testimony at Davenport’s Guilt Phase Trial At Davenport’s guilt phase trial in 1981, Farmer testified he saw Davenport on the transport bus to court and in the holding cells at court at various times between September and December 1980. On one occasion, while Farmer and Davenport were in the court holding cell, the two conversed after Farmer learned his girlfriend would not testify on his behalf

23 in his case. Sitting next to Davenport, Farmer complained his girlfriend was not going to testify for him. Davenport told Farmer not to worry about it and to kill her when Farmer got out in five years. Davenport told Farmer he had killed “a girl” and provided Farmer details about the murder. Farmer testified Davenport previously told him different stories about the murder and had changed his story at times. On this occasion, Davenport said “a lot of bizarre things he did.” Farmer testified Davenport said he left the bar with the woman, and while “he was walking with her, and she started to fight him, trying to resist or something, and he started hitting her and he stabbed her then, and he sodomized her then. [¶] So I guess that was at the time she was yelling.” According to Farmer, when Davenport sodomized the woman, she yelled and pleaded with him to stop. Davenport said he punched her in the face. Davenport told Farmer he tried to cut off her head but was unable to because he only had a small knife. Davenport also said he stuck a fence post or picket up her rectum. Davenport told Farmer he lowered the woman onto a fence post or stake that was in the ground. He wanted to make her a human scarecrow, but the post came out of the ground because of the weight of her body. Davenport stated he pushed and kicked the stake up her while she was on the ground. Davenport said he stabbed the woman before he inserted the stake, but he did not say she was dead when he did so. Farmer believed this conversation took place in November, but he was unsure of the date. Farmer denied being given any leniency or promises in exchange for his testimony, and he did not expect to receive anything for his testimony. However, in the past, he received a benefit for testifying against another inmate in Los Angeles County.

24 Farmer passed away in 1995. 2. Therrien’s 2017 Statement In 2017, two investigators from the district attorney’s office interviewed Therrien. Therrien told them he was in the holding tank with 8 Farmer and Davenport on one occasion. Therrien denied hearing Davenport confess to murdering a woman. Therrien stated Davenport said he had been accused of killing a woman, but Davenport never admitted killing her. Therrien accused Farmer of saying Davenport confessed to the murder so Farmer could get a reduced sentence. Therrien recalled Davenport did say something about “dragging” a woman “through the desert” on a motorcycle. Therrien told the investigators he was questioned by a district attorney’s office investigator in 1980 or 1981, after he got out of custody. Therrien said he spoke to the investigator on the telephone for about a half- hour. He told the investigators in 2017 he remembered what Davenport said after so many years because of the prior conversation he had with the investigator. Therrien said he would have told the investigator back then if Davenport had actually admitted killing the woman. Davenport’s habeas petition included as an exhibit Therrien’s declaration, signed under penalty of perjury. Therrien’s declaration included many of his statements from his 2017 interview, including Davenport did not confess to committing the murder and only told Farmer the accusations against him. Therrien declared Davenport “talked about what he was accused of in what [Therrien] felt was a ‘braggy’ way and sort of laughed about the

8 Therrien was shown photographs of Farmer and Davenport during the interview. He positively identified Farmer as one of the two men in the holding cell with him at the courthouse and was “pretty sure” about his identification of Davenport.

25 accusations against him.” Therrien avowed he would have testified against Davenport if he heard him confess to the murder. Therrien stated the two times he was contacted by the district attorney’s office about Davenport’s case, he told them the same things as in his declaration. 3. Therrien’s Prior Interview In February 1981, Farmer, who was in state prison, wrote a letter to Richard Ybarra, a district attorney’s office investigator, requesting Ybarra come see him. In the letter, Farmer stated he remembered who else was present during his conversation with Davenport. On June 19, 1981, Ybarra sent a letter to a lieutenant at the prison where Farmer was incarcerated.9 Ybarra’s letter included a photograph of Therrien and requested the photograph be shown to Farmer to determine if Therrien was the third person in a conversation with Farmer and Davenport where Davenport talked about the murder. The letter explained Therrien was living in another state and Ybarra wanted to confirm Therrien was the correct person before he traveled to interview him. The documentary evidence submitted by the parties in the habeas proceedings indicates Therrien was interviewed by an investigator with the district attorney’s office prior to Davenport’s trial in 1981. Unfortunately, any recordation of Therrien’s prior interview has been lost. Neither the prosecution nor Davenport’s current counsel have a record of Therrien’s 1981 interview. In 2017, the district attorney’s office contacted Ybarra concerning the 1981 interview of Therrien. Due to the intervening passage of time,

9 Ybarra’s letter was provided to Davenport’s federal habeas counsel in 2017, in response to a postconviction discovery request.

26 Ybarra did not recollect interviewing Therrien. Ybarra remembered attempting to call Therrien but was “pretty sure” he never contacted him and said he never traveled out of state to contact Therrien. He believed the deputy district attorney handling the case told him there was no need to contact the witness. 4. Davenport’s Trial Counsel Davenport’s original trial counsel provided a declaration for the habeas petition. Trial counsel was unaware of a statement by Therrien contradicting Farmer until counsel was provided Therrien’s 2017 statement. Trial counsel considered Therrien’s 2017 statement to be exculpatory and useful to impeach Farmer’s testimony. He declared if he had known about Therrien’s statement contradicting Farmer, he would have interviewed Therrien and likely presented his testimony at trial. Counsel considered Farmer “a significant witness,” and undermining Farmer’s credibility was “[a]n important part” of his trial strategy. The billing records for Davenport’s trial counsel indicate that on July 22, 1981, in preparation for trial, counsel listened “to tapes of conversations between DA investigator and possible witness James Therrien.” However, counsel had “no memory of a possible witness named James Therrien.” Trial counsel believed he received some tape relating to Therrien but did not recall receiving statements from Therrien that contradicted Farmer’s testimony.

27 5. Superior Court’s Ruling on the Brady and Ineffective Assistance of Counsel claims The superior court concluded Davenport did not meet his burden of setting forth a prima facie case warranting habeas relief on either his Brady or ineffective assistance of counsel claim. The superior court assumed, without deciding, favorable evidence was suppressed by the prosecution or Davenport’s trial counsel unreasonably failed to investigate and use Therrien’s testimony to impeach Farmer. As to Davenport’s Brady claim, the superior court concluded Davenport had not shown the suppressed evidence to be material. Addressing Davenport’s ineffective assistance of counsel claim, the court concluded Davenport had not shown counsel’s alleged deficient performance was prejudicial. The superior court found Davenport “does not establish there is a reasonable probability the outcome of [his] trial would have been different had the suppressed evidence been disclosed to the defense or had trial counsel investigated Therrien and used Therrien’s testimony to impeach John Farmer.” Explaining its reasoning, the court stated Therrien’s statements did “not conclusively refute Farmer’s testimony concerning [Davenport]’s incriminating statements” because: (1) Therrien did not state Davenport denied the offense; and (2) Farmer testified he had multiple conversations with Davenport while they were in custody and Davenport gave varying accounts of the murder at different times, but Therrien was only present for one conversation. After reviewing the other evidence presented at Davenport’s trial, the superior court reasoned there was “strong” evidence establishing Davenport’s guilt for first degree murder and the special circumstance, even without Farmer’s testimony. The court concluded its confidence in the outcome of Davenport’s trial was not undermined by the

28 alleged suppression of Therrien’s statement or by trial counsel’s alleged failure to investigate and present Therrien’s testimony to impeach Farmer. B. The Superior Court Properly Denied Davenport’s Brady Claim 1. Governing Principles The United States Supreme Court explained in Brady and its progeny the government violates the federal constitution’s due process clause “‘if it withholds evidence that is favorable to the defense and material to the defendant’s guilt or punishment.’” (Turner v. United States (2017) 582 U.S. 313, 315 (Turner).) Under the Fourteenth Amendment’s guarantee of due process, “‘the prosecution has a constitutional duty to disclose to the defense material exculpatory evidence, including potential impeaching evidence.’” (In re Jenkins (2023) 14 Cal.5th 493, 504; accord, Strickler v. Greene (1999) 527 U.S. 263, 280 (Strickler) [duty to disclose “encompasses impeachment evidence as well as exculpatory evidence”].) “There are three components of a true Brady violation: The evidence at issue must be [1] favorable to the accused, either because it is exculpatory, or because it is impeaching; [2] that evidence must have been suppressed by the State, either willfully or inadvertently; and [3] prejudice must have ensued.” (Strickler, supra, 527 U.S. at pp. 281–282; accord, People v. Superior Court (Johnson) (2015) 61 Cal.4th 696, 710; In re Sassounian (1995) 9 Cal.4th 535, 545 (Sassounian); People v. Salazar (2005) 35 Cal.4th 1031, 1043 (Salazar).) “Prejudice, in this context, focuses on ‘the materiality of the evidence to the issue of guilt or innocence.’” (Salazar, at p. 1043; accord, Strickler, at p. 282.) A defendant is entitled to habeas corpus relief when all three components of a Brady violation—favorability, suppression, and materiality —have been shown. (Strickler, supra, 527 U.S. at pp. 281–282; Sassounian,

29 supra, 9 Cal.4th at p. 545.) We discuss the materiality element of Davenport’s Brady claim as it is determinate and the main focus of the parties’ briefing.10 “‘[E]vidence is “material” within the meaning of Brady when there is a reasonable probability that, had the evidence been disclosed, the result of the proceeding would have been different.’ [Citations] ‘A “reasonable probability” of a different result’ is one in which the suppressed evidence ‘“undermines confidence in the outcome of the trial.”’ [Citations.] In other words, petitioners . . . are entitled to a new trial only if they ‘establis[h] the prejudice necessary to satisfy the “materiality” inquiry.’” (Turner, supra, 582 U.S. at p. 324; accord, Strickler, supra, 527 U.S. at p. 280; Salazar, supra, 35 Cal.4th at pp. 1043, 1050; Sassounian, supra, 9 Cal.4th at p. 544.) “‘The question is not whether the defendant would more likely than not have received a different verdict with the evidence, but whether in its absence he received a fair trial, understood as a trial resulting in a verdict worthy of confidence.’” (Strickler, supra, 527 U.S. at pp. 289–290; accord, Hill, supra, 104 Cal.App.5th at p. 853.) “[T]he materiality inquiry is not 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. [Citation.] Rather, the question is whether ‘the favorable evidence could reasonably be taken to put the whole case in such a

10 As Therrien’s statement impeached Farmer’s testimony concerning Davenport’s confession, it was favorable evidence and satisfies the first component of a Brady claim. We do not address Davenport’s argument the prosecution suppressed the evidence because we conclude the evidence was not material. (See Salazar, supra, 35 Cal.4th at pp. 1035, 1049 [declining to remand for consideration of whether the prosecution had suppressed the evidence because Supreme Court found evidence allegedly suppressed was not material].)

30 different light as to undermine confidence in the verdict.’” (Strickler, at p. 290.) In performing the materiality inquiry, “[w]e must examine the trial record, ‘evaluat[e]’ the withheld evidence ‘in the context of the entire record,’ [citation], and determine in light of that examination whether ‘there is a reasonable probability that, had the evidence been disclosed, the result of the proceeding would have been different[.]’” (Turner, supra, 582 U.S. at pp. 324–325.) This can be a “legally simple but factually complex” task. (Id. at p. 324.) 2. Analysis Davenport contends Farmer’s testimony he confessed to the murder was important to the prosecution’s case, and therefore, evidence impeaching this testimony was material. He asserts the other evidence against him was weak, circumstantial, and inadequate to provide confidence in the verdict. We disagree. Our confidence in the outcome of Davenport’s guilt phase trial is not undermined by Therrien’s 2017 statement.11 Evaluating Therrien’s statement in the context of the entire record, we agree with the superior court it is not reasonably probable the result of the proceedings would have been different if the statement had been disclosed to the defense prior to Davenport’s trial. There was overwhelming evidence of Davenport’s guilt and Therrien’s statement had limited exculpatory value. Therrien’s 2017 statement did not undermine the evidence Davenport murdered Lingle—only whether he confessed to doing so. It did

11 Because Farmer’s testimony was not presented at the penalty phase retrial, our confidence in its outcome is not undermined either and does not require further discussion.

31 not point to a third party or provide an alternative theory of who killed Lingle. Therrien’s statement did not counter the evidence of Lingle’s brutal murder nor undermine the evidence Davenport did it. Lingle was last seen alive leaving the bar with Davenport. (Davenport I, supra, 41 Cal.3d at p. 256.) Shortly thereafter, three witnesses saw a motorcycle that appeared similar to Davenport’s in the field where Lingle’s body was discovered the next morning. One of these witnesses selected Davenport’s photograph as the person she thought she saw in the field, but she was not certain. (Id. at pp. 256–257.) Lingle died from multiple knife wounds, and Davenport always carried a knife. Davenport’s motorcycle connected him to the crime scene based on vegetation and soil matches, as well as the motorcycle’s tires. (Davenport I, supra, 41 Cal.3d at pp. 257–258.) The motorcycle tire tracks at the scene of Lingle’s murder indicated the motorcycle had two different brands of tires, the same as Davenport’s motorcycle. Further solidifying the evidence against Davenport, the rear tire on his motorcycle had a unique and distinctive tread pattern, which was seen at the crime scene. (Ibid.) There was also evidence of Davenport’s consciousness of guilt, including his failed attempt to create an alibi and his lie about having a flat tire the morning after the murder. After he was arrested, Davenport told his girlfriend that the night of Lingle’s murder he left the bar and went to a coffee shop with some friends, whom Davenport’s girlfriend identified by name. However, these two men denied going to a coffee shop with Davenport. Because they did not corroborate his false alibi, Davenport testified at trial he went to the coffee shop by himself. He also testified he was late to work the morning after the murder because his motorcycle had a flat tire and he had to change it. The prosecution presented evidence indicating the

32 motorcycle tire had not been changed. (Davenport I, supra, 41 Cal.3d at p. 260.) Therrien’s 2017 statement only concerned Farmer’s testimony Davenport admitted killing Lingle. Davenport asserts Therrien’s statement impeaching Farmer was material because “evidence of a defendant’s confession is inherently potent.” We agree a confession is powerful evidence. (Arizona v. Fulminante (1991) 499 U.S. 279, 296 [“confession is probably the most probative and damaging evidence that can be admitted against” a defendant].) But when considered in the context of all the evidence presented at Davenport’s trial, Farmer’s testimony concerning Davenport’s confession had limited weight. Farmer’s credibility was impeached in several respects: Farmer made numerous prior inconsistent statements concerning Davenport’s confession; Farmer had been provided information concerning the murder by Ybarra; and Farmer previously benefited by testifying against a different inmate. Farmer’s inconsistent statements concerning Davenport’s confession were significant. Farmer told Ybarra that Davenport said he killed his girlfriend because she had sex with another man. Farmer also made this statement to the defense investigator. However, Lingle was not Davenport’s girlfriend. At trial, Farmer explained the reason for this discrepancy was Davenport told him different stories and about two different crimes. Farmer testified that when he made his prior statements, he was unsure which incident was the one pending against Davenport and was unable to keep the details straight. Given the evidence the jury heard concerning Farmer’s prior inconsistent statements about Davenport’s confession, Therrien’s statement Davenport did not admit committing the murder would have added little to the deterioration of Farmer’s credibility. The jury would not have been

33 surprised to hear Therrien’s statement Davenport did not confess and Farmer misunderstood what Davenport was saying as Farmer misunderstood and was confused about several points. It is likely the jury was already skeptical of Farmer’s rendition of his conversation with Davenport. The superior court noted another issue undercutting the materiality of Therrien’s statement. It is possible the conversation in which Davenport confessed to Farmer occurred at a different time or on a different day than the conversation to which Therrien was privy. Farmer, Davenport, and Therrien were transported to court on November 21, 1980. Farmer believed the pertinent conversation occurred in November but was unsure of the date. Farmer testified that in prior conversations Davenport gave him “different stories” about the murder. While Therrien’s statement impeaches the evidence Davenport confessed to Farmer on November 21, 1980, it does not mean Davenport did not confess at all. It is possible the confession occurred when Therrien was absent from the holding cell, he did not hear the whole conversation between Davenport and Farmer, or simply Davenport confessed to Farmer on a different day. Therrien’s 2017 statement would have also had negative implications for Davenport’s defense. Davenport testified he never spoke to Farmer, but this testimony is inconsistent with Therrien’s statement. While Therrien did not hear Davenport admit to killing Lingle, Therrien did hear Davenport admit to dragging a woman through the desert. Thus, Therrien’s statement he never heard Davenport confess the murder to Farmer had limited value. Davenport further asserts Farmer’s testimony was material because it was important to the prosecution’s case as demonstrated by how the prosecution addressed it in argument. A review of the prosecutor’s closing

34 argument demonstrates otherwise. For the first 50 transcript pages of his closing argument, the prosecutor discussed the plethora of evidence proving Davenport murdered Lingle, before the prosecutor discussed Farmer’s testimony Davenport confessed to the murder. The prosecutor then spent about five transcript pages discussing Farmer’s testimony and the anticipated defense argument Farmer’s testimony was not credible. As to the evidence proving the torture-murder special circumstance, the prosecutor mentioned Farmer only once in discussing the sequence of events and Davenport’s actions of sodomizing Lingle before stabbing her. In his rebuttal closing argument, the prosecutor spent less than 10 pages addressing the defense’s argument concerning Farmer’s testimony. Davenport contends Therrien’s statement was significant impeachment of Farmer’s testimony and not cumulative of the other evidence impeaching Farmer. We disagree. Impeaching Farmer with Therrien’s statement would not have materially impacted the jury’s assessment of Farmer’s credibility or Davenport’s guilt. We find an analogous situation in Sassounian, supra, 9 Cal.4th 535. There, the prosecution called a witness who testified the defendant confessed to the murder while they were in custody in the county jail. (Id. at p. 539.) The prosecution’s witness was subjected to extensive impeachment. (Id. at pp. 540–541.) After the defendant’s special circumstance murder conviction was affirmed on appeal, the defendant filed a petition for writ of habeas corpus in the superior court with a declaration by the prosecution’s informant witness recanting his testimony. (Id. at p. 542.) The defendant sought relief on the ground the prosecution failed to disclose favorable and material evidence. (Id. at p. 543.) The California Supreme Court concluded the defendant “failed to carry his burden of alleging that the introduction at

35 trial of false evidence that was substantially material or probative on the . . . special-circumstance finding.” (Id. at p. 547.) Separate from the prosecution witness’s testimony, there was overwhelming evidence of the defendant’s involvement in the murder and support for the special circumstance finding. (Id. at pp. 548–549.) Addressing the testimony of the prosecution’s informant witness, the Supreme Court stated: “[W]hatever effect his testimony might have had depends on whether the jury found it believable. Whether it did is open to question. Certainly, [the prosecution’s witness] was extensively impeached. [Citation.] As a consequence, the ‘testimonial thread’ that he plied became severely ‘tattered.’” (Id. at p. 548.) The Supreme Court concluded its confidence in the outcome of the defendant’s trial was not undermined because it was not reasonably probable the defendant could have obtained a different result in the absence of the testimony of the prosecution’s informant witness. (Id. at p. 550.) The same is true here. Even without Farmer’s testimony, there was overwhelming evidence at Davenport’s trial proving him guilty of first degree murder. “‘[T]here was nothing “close” about this’” issue. (Sassounian, supra, 9 Cal.4th at p. 548.) Farmer was impeached extensively with his prior inconsistent statements, admitted he previously acted as a jailhouse informant in a different case, and admitted to receiving some information regarding the case from Ybarra. Therrien’s statement only provided further impeachment of Farmer’s testimony. The most the further impeachment could have done was render Farmer’s testimony a nullity. (Id. at p. 550.) It is not reasonably probable Davenport would have obtained a different result in the absence of Farmer’s testimony in its entirety, and therefore, it is not reasonably probable he would have obtained a different result if Farmer’s testimony had been subjected to further impeachment.

36 Examining all the evidence at Davenport’s trial, we conclude it is not reasonably probable the result of Davenport’s guilt phase trial would have been different if Therrien’s statement had been disclosed to the defense prior to trial. Our confidence in the jury’s verdicts is not undermined by Therrien’s statement. (Turner, supra, 582 U.S. at p. 328.) Davenport did not make a prima facie showing there is a reasonable probability the jury would have returned a different verdict if Farmer’s testimony had been further impeached by Therrien, or even if Farmer’s testimony was excluded in its entirety. Accordingly, the superior court properly denied Davenport relief on his Brady claim. C. The Superior Court Properly Denied Davenport’s Ineffective Assistance of Counsel Claim

Davenport alternatively asserts that if the prosecution did disclose information concerning Therrien’s exculpatory statement to his trial counsel, then his counsel rendered ineffective assistance by failing to investigate and present Therrien’s testimony. The Attorney General argues Davenport failed to establish a prima facie claim of ineffective assistance by trial counsel. We agree with the Attorney General. 1. Governing Principles “An ineffective assistance of counsel claim has two elements: a defendant must show that their counsel’s performance was deficient, and that this deficient performance prejudiced the defense. [Citation.] A reviewing court can begin an ineffective assistance of counsel inquiry with either element and need not address both elements if one is not satisfied. [Citations.] Indeed, it is often preferable for a court to dismiss an ineffective assistance of counsel claim solely for lack of prejudice. [Citations.] To satisfy Strickland’s prejudice prong, a defendant ‘must show that there is a

37 reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.’” (In re Tellez (2024) 17 Cal.5th 77, 88.) 2. Analysis We begin (and end) our analysis of Davenport’s ineffective assistance of counsel claim with the prejudice prong. The showing required for Strickland’s prejudice prong is the same as the Brady materiality standard—it is reasonably probable that, absent the error, the jury would have had a reasonable doubt as to guilt. (United States v. Bagley (1985) 473 U.S. 667, 682; People v. Hernandez (2012) 53 Cal.4th 1095, 1108.) Davenport asserts his trial counsel’s failure to present Therrien’s testimony was prejudicial “for all of the same reasons discussed in regard to materiality for the Brady claim.” We have already concluded Davenport did not make a prima facie showing of materiality for his Brady claim, and we reach the same conclusion as to the prejudice prong of his ineffective assistance of counsel claim. Even if Davenport’s trial counsel had called Therrien as a defense witness to impeach Farmer’s testimony, it is not reasonably probable the jury would have reached a different result. There was overwhelming evidence Davenport brutally murdered Lingle. Therrien’s exculpatory statement only impeached Farmer’s testimony Davenport confessed to committing the murder; it did not discount all the other evidence establishing Davenport’s guilt. Thus, Davenport did not make a prima facie showing sufficient to undermine confidence in the jury’s verdict. (Strickland, supra, 466 U.S. at p. 694.) Davenport’s ineffective assistance of counsel claim fails under the prejudice prong. Therefore, we conclude the superior court properly denied this claim.

38 D. The Superior Court Properly Denied Davenport’s Claim the Prosecution Knowingly Presented False Evidence Davenport contends he is entitled to habeas relief because the prosecution presented Farmer’s testimony at his guilt phase trial knowing it was false. He further asserts it is reasonably likely this false testimony affected the jury’s verdict. (Giglio, supra, 405 U.S. 150; Napue, supra, 360 U.S. 264.) We disagree. 1. Background Raising this false evidence claim in his habeas petition, Davenport relied on Therrien’s 2017 statement that he was interviewed by an investigator with the district attorney office around 1981, he told the investigator he was present during the conversation between Davenport and Farmer and Davenport did not confess to the murder. Davenport asserted the prosecutor therefore knew Farmer’s testimony about the confession was false. He further argued “Farmer’s false testimony was reasonably likely to have affected the verdict for the same reasons discussed in . . . [his] Brady claim.” The superior court denied this claim, concluding Davenport failed to set forth a prima facie case for relief. The court found Davenport did not demonstrate Farmer’s testimony was false as Therrien was present during only one conversation between Davenport and Farmer and Farmer was unsure when the conversation occurred in which Davenport confessed. The court further reasoned, even if Farmer provided false testimony, Davenport had not shown it was reasonably likely the verdict would have been different. 2. Governing Principles “A defendant’s due process rights are violated if a prosecutor knowingly presents false testimony or fails to correct such testimony after it has been elicited.” (In re Masters (2019) 7 Cal.5th 1054, 1089 (Masters).) “Put

39 another way, the prosecution has the duty to correct the testimony of its own witnesses that it knows, or should know, is false or misleading. [Citations.] This obligation applies to testimony whose false or misleading character would be evident in light of information known to the police involved in the criminal prosecution [citation], and applies even if the false or misleading testimony goes only to witness credibility [citations].” (People v. Morrison (2004) 34 Cal.4th 698, 716, citing Napue, supra, 360 U.S. at p. 269 & Giglio, supra, 405 U.S. at pp. 153–154.) To prevail on a Napue claim, the defendant must show “‘(1) [the] testimony (or evidence) was actually false, (2) the prosecution knew or should have known that the testimony was actually false, and (3) . . . the false testimony was material.’” (Dickey v. Davis (9th Cir. 2023) 69 F.4th 624, 636.) A Napue claim, like a Brady claim, requires a showing of materiality; however, the standard is not the same for both. The materiality standard for a Napue claim requires reversal of a conviction if “there is any ‘“reasonable likelihood that the false testimony could have affected the judgment of the jury.”’ [Citations.] This standard is equivalent to the harmless beyond a reasonable doubt standard for determining whether constitutional error is prejudicial. [Citations.] ‘A strict standard is appropriate because, as the Supreme Court has explained, false testimony cases involve not only “prosecutorial misconduct,” but also “a corruption of the truth-seeking function of the trial process.”’ [Citations.] ‘[A] Napue claim fails if, absent the false testimony or evidence, the petitioner still “received a fair trial, understood as a trial resulting in a verdict worthy of confidence.”’” (Hill, supra, 104 Cal.App.5th at p. 835; accord, United States v. Agurs (1976) 427 U.S. 97, 103–104.)

40 3. Analysis Davenport failed to make a prima facie showing of a Napue violation as he did not establish the prosecution knowingly presented false evidence to the jury. As the Attorney General argues, Therrien’s 2017 statement did not render false Farmer’s testimony concerning Davenport’s confession because it was possible Therrien was not privy to all the conversation between Farmer and Davenport. While Therrien’s 2017 statement undermines the prosecution’s evidence Davenport confessed to Farmer when all three men were in the holding cell on November 21, 1980, it does not show Farmer lied about Davenport confessing. Farmer testified he had multiple conversations with Davenport and Davenport gave different stories about the murder at different times. At best, Therrien’s 2017 statement created conflicting evidence and a credibility issue for the jury to resolve concerning whether Davenport confessed when all three men were in the holding cell. It did not prove Farmer testified falsely. The presentation of two conflicting versions of an incident does not establish one version was false. (United States v. Geston (9th Cir. 2002) 299 F.3d 1130, 1135.) Even assuming the prosecutor was aware Therrien’s original statement conflicted with Farmer’s testimony, it does not establish the prosecutor knowingly presented false testimony. The prosecutor may have had doubts as to Farmer’s veracity, but that does not mean the prosecutor presented false evidence. “The existence of evidence tending to contradict testimony the government elicits at trial does not conclusively show that either the witness perjured himself or (if he did) that the government knew or should have known of the perjury.” (Lambert v. Blackwell (3d Cir. 2004) 387 F.3d 210, 252.)

41 Lastly, even if we assume the prosecution presented Farmer’s testimony concerning Davenport’s confession knowing it was false, we still have confidence in the verdict and conclude Davenport received a fair trial. (Hill, supra, 104 Cal.App.5th at p. 835.) As discussed ante, there was overwhelming evidence establishing Davenport’s guilt for first degree murder and the torture-murder special circumstance. (See Phillips v. Ornoski (9th Cir. 2012) 673 F.3d 1168, 1190 [finding prosecution’s Napue violations “‘pernicious’ and ‘reprehensible’” but not material given the overwhelming evidence of guilt].) Considering all the evidence presented at trial, we are convinced beyond a reasonable doubt the admission of Farmer’s testimony did not affect the jury’s verdict. One week after oral argument was heard in this appeal, the United States Supreme Court issued its decision in Glossip v. Oklahoma (2025) ___ U.S. ___ [145 S.Ct. 612, 221 L.Ed.2d 90], wherein it reversed a capital defendant’s conviction based on prejudicial Napue error. At Davenport’s request, we permitted the parties to present supplemental briefs addressing Glossip. Davenport attempts to equate his case with the circumstances in Glossip and asserts it supports his Napue claim. However, we find Glossip readily distinguishable. First, in Glossip, the Attorney General conceded the trial prosecutor committed prejudicial Napue error by failing to correct the testimony of the prosecution’s main witness that the prosecutor knew was false. (Glossip, at p. ___ [145 S.Ct. at p. 618].) Here, the Attorney General has made no such concession. Second, the witness at issue in Glossip was the linchpin of the prosecution’s case, which “rested centrally on [his] credibility.” (Id. at p. ___ [145 S.Ct. at p. 629].) As the United States Supreme Court explained, “[T]he jury could convict Glossip only if it believed Sneed.” (Id. at p. ___ [145 S.Ct. at p. 628].) Here, the jury’s verdict was not

42 dependent on Farmer’s testimony as the prosecution presented a strong case against Davenport and any Napue error was harmless beyond a reasonable doubt. Accordingly, we affirm the superior court’s denial of this claim. E. The Superior Court Properly Denied Davenport’s Destruction of Evidence Claim Davenport argues the prosecution’s present inability to produce a recording or report of Therrien’s original statement entitles him to habeas relief. Davenport contends the state’s failure to preserve Therrien’s original interview interfered with the exercise of his state right to habeas corpus and violated his rights under California v. Trombetta (1984) 467 U.S. 479 (Trombetta), his right to fundamental fairness, and his federal constitutional right of access to the courts. He asserts the superior court erred by denying his claim without conducting a hearing and “the appropriate remedy is to provide” him with “court-ordered fact development” as “a meaningful substitute for that of which he has been deprived.” We disagree. 1. Background When raising this issue in his habeas petition, Davenport relied on Therrien’s 2017 statement he was interviewed by someone from the district attorney’s office near the time of Davenport’s original trial. The billing records for Davenport’s trial counsel indicate a tape was disclosed by the prosecution prior to the original trial containing conversations between Therrien and an investigator with the district attorney’s office. In 2017, after learning of the prior interview, Davenport’s federal habeas counsel sought a copy of the recording or memorandum reflecting the interview. It appears a recordation of Therrien’s original interview has not been retained by the prosecution or Davenport’s trial counsel.

43 In his habeas petition, Davenport argued the prosecution’s loss of all records concerning Therrien’s original interview deprived Davenport of his ability to review this evidence and interfered with his ability to pursue viable claims of constitutional error in his petition. He asserted the prosecution’s destruction of this evidence violated his due process rights under Trombetta, as well as his rights to fundamental fairness and access to the courts. The superior court denied Davenport’s claim stating: “Assuming without deciding that what Therrien represents is true, Therrien’s statements are not exculpatory particularly in view of Farmer’s testimony concerning [Davenport]’s incriminating statements made at various times. Therrien represents [Davenport] did not confess to the murder but does not claim [Davenport] denied committing the offense either. Moreover, trial counsel’s billing records indicate the defense was made aware of Therrien prior to trial and therefore had the ability and opportunity to obtain comparable evidence via reasonably available means. The fact Therrien’s 1981 recorded interview appears to be no longer available alone does not lead to a conclusion that the prosecution intentionally destroyed exculpatory evidence in bad faith. ‘The mere possibility that an item of undisclosed information might have helped the defense, or might have affected the outcome of the trial, does not establish “materiality” in the constitutional sense.’” 2. Governing Principles

The prosecution’s duty to retain potentially exculpatory evidence is different than its duty to disclose it. (People v. Alvarez (2014) 229 Cal.App.4th 761, 771.) “With respect to retention, . . . the prosecution’s obligation is narrower.” (City of Los Angeles v. Superior Court (2002) 29 Cal.4th 1, 8.) “‘“‘Law enforcement agencies have a duty, under the due

44 process clause of the Fourteenth Amendment, to preserve evidence “that might be expected to play a significant role in the suspect’s defense.” [Citations.] To fall within the scope of this duty, the evidence “must both possess an exculpatory value that was apparent before the evidence was destroyed, and be of such a nature that the defendant would be unable to obtain comparable evidence by other reasonably available means.”’”’” (People v. Chism (2014) 58 Cal.4th 1266, 1299 (Chism); accord, Trombetta, supra, 467 U.S. at p. 488.) “[U]nless a criminal defendant can show bad faith on the part of the police, failure to preserve potentially useful evidence does not constitute a denial of due process of law.” (Arizona v. Youngblood (1988) 488 U.S. 51, 58 (Youngblood); City of Los Angeles v. Superior Court, supra, 29 Cal.4th at p. 8.) 3. Analysis The Attorney General argues Trombetta and Youngblood “‘do not apply to claims that evidence was lost or destroyed after trial.’” (Tyler v. Purkett (8th Cir. 2005) 413 F.3d 696, 703.) Davenport does not address this argument in his reply brief. We could interpret this as an implicit concession of the argument’s merits. (See Rudick v. State Bd. of Optometry (2019) 41 Cal.App.5th 77, 89–90 [appellants implicitly conceded by “failing to respond in their reply brief to the [respondent’s] argument”].) Nevertheless, even assuming Davenport did not concede this point, we conclude the superior court properly denied his claim. The record in this habeas proceeding indicates the prosecution disclosed to the defense prior to trial a recording concerning Therrien’s original statement. Unfortunately, by the time this habeas petition was filed, approximately 36 years after the statement was made, neither party still retained a copy of Therrien’s original statement. Even assuming Therrien’s

45 original statement possessed an exculpatory value that was apparent before it was lost or destroyed, Davenport’s claim fails because he had access to it or comparable evidence obtained by other means. (Chism, supra, 58 Cal.4th at p. 1299.) Davenport obtained a declaration from Therrien in 2017, in which Therrien stated he previously made the same statements to a district attorney investigator. Moreover, even assuming Therrien’s original statement included “potentially useful” information not reflected in his 2017 statement, nothing suggests the prosecution lost or destroyed that information in bad faith. (Youngblood, supra, 488 U.S. at p. 58.) Indeed, the district attorney’s office sent investigators to interview Therrien in 2017 after a request for postconviction discovery by Davenport’s federal habeas counsel for information concerning the use of jailhouse informant Farmer at Davenport’s trial. Davenport asserts the prosecution’s failure to preserve Therrien’s original statement interfered with his ability to pursue viable claims of constitutional error in his habeas petition. Davenport cannot show the prosecution’s failure to preserve Therrien’s original statement impaired his ability to pursue his claims. We and the superior court have assumed for purposes of these proceedings that Therrien’s original statement was the same as his 2017 statement when considering Davenport’s claims concerning a Brady violation, ineffective assistance by trial counsel, and the prosecution knowingly presenting false evidence. Thus, there has been no interference with Davenport’s ability to exercise his state right to habeas corpus or his federal constitutional right to access to the courts. Accordingly, the superior court properly denied this claim without conducting an evidentiary hearing.

46 V.

THE SUPERIOR COURT PROPERLY DENIED DAVENPORT’S

CLAIMS UNDER SECTION 1473

Davenport contends the superior court erred by denying claims he made under section 1473 based on the presentation of false evidence at trial and new evidence of innocence. We disagree. A. Background In his habeas petition, Davenport argued he was entitled to relief under section 1473 based on new evidence of innocence and the prosecution’s presentation of false evidence at trial. He asserted the prosecution presented three categories of false evidence: (1) Farmer’s testimony Davenport confessed to killing Lingle; (2) expert testimony concerning whether Lingle was alive when the stake was inserted; and (3) testimony the tires on Davenport’s motorcycle matched the tire tracks at the crime scene. He also asserted these three categories of false evidence constituted new evidence of innocence. The first category was supported by evidence concerning Therrien’s 2017 statement, discussed ante. Dr. Haddix’s declaration, discussed post, formed the basis for the second category. For the third category, Davenport relied on a report issued by the President’s Council of Advisors on Science and Technology (PCAST) entitled Forensic Science in Criminal Courts: Ensuring Scientific Validity of Feature-Comparison Methods, which is also discussed post. 1. Dr. Haddix’s Declaration Haddix, a forensic pathologist, submitted a declaration after being retained by Davenport’s federal habeas counsel “to review materials associated with” Lingle’s death and the prosecution of Davenport. In

47 rendering her opinion, Haddix reviewed several reports, autopsy photographs, as well as the testimony of the prosecution’s pathology experts Fischer and Fukumoto and the defense’s pathology experts Modglin and Kornblum. Haddix concluded “a fundamentally flawed methodology utilized during the autopsy conducted on Ms. Lingle resulted in erroneous conclusions by all reviewing and testifying pathologists to some degree.” Haddix explained “Fischer utilized a fundamentally flawed approach at the very start of the autopsy examination by blindly removing the stake.” She indicated “Fischer should have left the stake in situ and then documented [its] path.” This would have prevented or “minimized the production of artifactual injuries and their resulting misinterpretation.” Haddix disagreed with the opinion Fischer rendered at Davenport’s guilt phase trial that Lingle was alive when the stake was inserted. Fischer testified that when the stake went through Lingle’s lung, it tore pulmonary blood vessels, causing the cumulation of 200 cubic centimeters (cc) of blood in her right pleural cavity. He explained if the stake had been inserted after death, then there would not have been “this amount of bleeding into the body” because “dead people don’t bleed.” He indicated Lingle’s heart had to be pumping in order to cause the bleeding in the chest cavity. Haddix declared the amount of blood in Lingle’s “right chest cavity [wa]s not indicative of antemortem insertion of the stake.” Haddix indicated two publications since Davenport’s trial “have reported on the volume of blood loss that can occur into a chest cavity from injuries known to have occurred postmortem.” She cited to Forensic Medicine: A Guide to Principles, 3d edition (I. Gordon, H.A. Shapiro, and S.D. Berson, Churchill Livingstone, 1988) and Amount of Postmortem Bleeding: An Experimental

48 Autopsy Study in the American Journal of Forensic Medicine and Pathology in 2004. Haddix stated: “These studies demonstrate that bleeding can and does occur from wounds inflicted post-mortem. In fact, they demonstrate that bleeding of a substantially greater volume than the 200 cc observed in this case can be expected in such circumstance. In sum, these studies conclusively disprove Dr. Fischer’s testimony that the 200 cc of blood found in the pleural cavity demonstrated that Ms. Lingle had to have been alive when the stake was inserted.” When testifying at Davenport’s trials, Fischer and Fukumoto cited to the partial collapse of Lingle’s right lung as evidence she was alive when the stake was inserted. In Haddix’s opinion, the partial collapse of Lingle’s right lung did “not provide guidance in distinguishing ante from postmortem injury.” Haddix opined the removal of the stake could “have artifactually produced the appearance of an antemortem partial collapse” and Fischer’s removal of the stake could not be ruled out as the cause of the partial lung collapse. While Fischer testified the absence of blood in other areas of Lingle’s body impacted by the stake did not establish the stake was inserted after Lingle was dead, Haddix disagreed. Haddix stated: “Had this stake been inserted while Ms. Lingle was alive, blood should have collected in multiple areas and throughout its course in the body. There should have been blood emanating from the anus as well as present in the peritoneal cavity where there were injuries of the colon, mesentery and liver; yet, no external blood drainage and no collections of blood were found in the retroperitoneum or peritoneal cavity.” Haddix described Fischer’s theory as “wholly without merit and contrary to basic principles of anatomy and physiology.”

49 Haddix disagreed with Fukumoto’s opinion the appearance of blood in microscopic slides of sections of Lingle’s lung indicated she was alive when the stake was inserted. Haddix opined: “[B]lood that drains from vessels injured postmortem will have a similar microscopic appearance as blood that drained immediately prior to death. The gross observations . . . rather than the microscopic observations offer the greatest guidance in distinguishing antemortem from postmortem injury in this case.” Haddix also disagreed with Fischer’s testimony Lingle’s death could have taken up to 15–20 minutes following the multiple sharp force injuries to her neck that resulted in an incision of the right carotid artery. Haddix stated: “[B]leeding from this injury would be brisk and a life- threatening amount of blood lost in well less than 15–20 minutes. A more reasonable estimate of time for a life-threatening loss of blood would be approximately five minutes.” 2. PCAST Report The PCAST report focused on “forensic ‘feature-comparison methods’” “that attempt to determine whether an evidentiary sample (e.g., from a crime scene) is or is not associated with a potential ‘source’ sample (e.g., from a suspect), based on the presence of similar patterns, impressions, or other features in the sample and the source.” The report focused on six specific feature-comparison methods (i.e., bitemark comparisons, DNA, latent fingerprint, firearms, footwear, and hair analysis) but indicated “the basic analysis can be applied to assess the foundational validity of any forensic feature-comparison method,” including tire track analysis. 3. Superior Court’s Ruling The superior court concluded Davenport’s claim under section 1473 failed to set forth a prima facie case warranting habeas relief.

50 Regarding Therrien’s 2017 statement Davenport did not confess to the murder, the court concluded the statement did “not undermine the validity of the judgment of conviction and sentence.” After discussing the evidence, the court stated Therrien’s 2017 declaration “[a]t best” impeaches Farmer’s testimony but does not exonerate Davenport. As to Haddix’s declaration, the court similarly held it did not undermine the validity of the judgment. The court concluded Haddix’s conclusions did not constitute new evidence because they were cumulative and corroborative of defense expert Modglin’s testimony. The court noted the literature cited by Haddix in her declaration “arguably contradicts” but does not “conclusively refute” Fischer’s testimony the stake was inserted prior to the victim’s death. The court concluded the PCAST report was “not relevant to the issues” and did not “undermine the validity of the judgment.” The court stated: “While the report focuses on ensuring scientific validity of feature- comparison methods used in forensic science in criminal investigations and prosecutions, the report does not address tire track identification analysis in general nor in particular with reference to testimony given concerning tire track identification in [Davenport]’s case. Furthermore, limitations inherent in testimony given by prosecution witnesses with respect to this issue was elicited and brought to the jury’s attention.” The court noted Jackson, the lead forensic specialist from the Orange County Sheriff’s Department, did not testify the tire tracks at the crime scene were created by Davenport’s motorcycle. Leonard, the product manager for Dunlop tires, did not testify the tire tracks at the crime scene were created by the rear tire on Davenport’s motorcycle, only that the tire tracks were made by a tire of the same type as found on Davenport’s motorcycle. The court concluded the PCAST report “is

51 not directly relevant to the testimony given during [Davenport]’s trial and therefore does not constitute new evidence.” Addressing all three (Therrien’s 2017 statement, Haddix’s declaration, and the PCAST report), the superior court stated when considered individually or collectively, they are not material nor “of such decisive force and value that they would more likely than not have changed the outcome of [Davenport]’s trial.” The court reasoned the outcome would not have changed because the prosecution’s evidence establishing Davenport’s guilt for special circumstance murder was strong. The court found equally without merit Davenport’s claim based on false evidence, rejecting his assertion Farmer’s testimony was false or that Fischer’s testimony regarding the time and manner of Lingle’s death was false. As to the latter issue, the court stated Davenport “does not establish that Dr. Fischer’s expert testimony was patently false or conclusively undermined by later scientific research or technological advances. Dr. Haddix’s expert opinion is largely cumulative and corroborative of testimony given at trial by the defense’s own expert. As such, it does not conclusively undermine Dr. Fischer’s testimony nor establish that his testimony was false.” Addressing the PCAST report, the court found it did not undermine the testimony given by the prosecution witnesses on the tire track identification issue or demonstrate that testimony was patently false. B. Governing Principles In its current form, section 1473 provides a petition for writ of habeas corpus may be used to pursue claims of false evidence (id., subd. (b)(1)(B)), new evidence (id., subd. (b)(1)(C)), or new developments relating to forensic, medical, or scientific expert testimony that was presented at trial

52 12 (id., subd. (b)(1)(D)). The statute sets forth the applicable legal standards for these claims. Section 1473, subdivision (b)(1)(C)(i) provides habeas corpus relief may be granted when “[n]ew evidence exists that is presented without substantial delay, is admissible, and is sufficiently material and credible that it more likely than not would have changed the outcome of the case.” The statute defines “‘new evidence’” as “evidence that has not previously been presented and heard at trial and has been discovered after trial.” (§ 1473, subd. (b)(1)(C)(ii).) Habeas corpus relief on false evidence may be granted when “[f]alse evidence that is material on the issue of guilt or punishment was introduced against a person at a . . . trial relating to the person’s incarceration.” (§ 1473, subd. (b)(1)(A).) Section 1473, subdivision (b)(2) provides: “‘[F]alse evidence’ includes opinions of experts that have either been repudiated by the expert who originally provided the opinion at a hearing or trial or that have been undermined by the state of scientific knowledge or later scientific research or technological advances.” “[O]nce a defendant shows that false evidence was admitted at trial, relief is available under section 1473 as long as the false evidence was ‘material.’ [California Supreme Court] case law further explains that false evidence is material ‘“if there is a ‘reasonable probability’ that, had it not been introduced, the result would

12 Since 2019, when the superior court was considering Davenport’s habeas petition, the Legislature has amended section 1473 multiple times—the latest effective January 1, 2025. (Stats. 2020, ch. 317, § 4; Stats. 2022, ch. 982, § 1.5; Stats. 2023, ch. 381, § 1; Stats. 2024, ch. 495, § 8.) Davenport asserts the current version applies, and the Attorney General does not disagree. We assume, without deciding, the current version applies in this appeal.

53 have been different.”’ [Citation.] The remedial purpose of the statute is to afford the petitioner relief if the ‘false evidence [was] of such significance that it may have affected the outcome of the trial . . . .’ [Citation.] Thus, the crucial question is whether the false evidence was material—not whether, without the false evidence, there was still substantial evidence to support the verdict.” (Richards II, supra, 63 Cal.4th at p. 312.) “‘Materiality is shown if there is a reasonable probability the result would have been different without the false evidence.’” [Citation.] “‘This required showing of prejudice is the same as the reasonably probable test for state law error established under People v. Watson (1956) 46 Cal.2d 818, 836. [Citation.] We make such a determination based on the totality of the relevant circumstances.’” (Masters, supra, 7 Cal.5th at p. 1078.) Section 1473, subdivision (b)(1)(D) allows a person to bring a petition for writ of habeas corpus where “[a] significant dispute has emerged or further developed in the petitioner’s favor regarding expert medical, scientific, or forensic testimony that was introduced at trial or a hearing and that expert testimony more likely than not affected the outcome of the case.” (Ibid.) It sets forth multiple elements a petitioner must establish by a preponderance of the evidence to obtain relief. (Id., subd. (b)(1)(D)(vi).) This provision provides: “[T]he expert medical, scientific, or forensic testimony includes the expert’s conclusion or the scientific, forensic, or medical facts upon which their opinion is based.” (Id., subd. (b)(1)(D)(i).) A significant dispute may concern “the reliability or validity of the diagnosis, technique, methods, theories, research, or studies upon which a medical, scientific, or forensic expert based their testimony.” (Id., subd. (b)(1)(D)(ii).) The statute provides: “[A] significant dispute can be established by credible expert testimony or declaration, or by peer reviewed literature showing that experts

54 in the relevant medical, scientific, or forensic community, substantial in number or expertise, have concluded that developments have occurred that undermine the reliability or validity of the diagnosis, technique, methods, theories, research, or studies upon which a medical, scientific, or forensic expert based their testimony.” (Id., subd. (b)(1)(D)(iii).) The statute further provides: “In assessing whether a dispute is significant, the court shall give great weight to evidence that a consensus has developed in the relevant medical, scientific, or forensic community undermining the reliability or validity of the diagnosis, technique, methods, theories, research, or studies upon which a medical, scientific, or forensic expert based their testimony or that there is a lack of consensus as to the reliability or validity of the diagnosis, technique, methods, theories, research, or studies upon which a medical, scientific, or forensic expert based their testimony. (Id., subd. (b)(1)(D)(iv).) C. Therrien’s Statement and Farmer’s Testimony We assume Therrien’s 2017 statement meets the definition of new evidence in section 1473, subdivision (b)(1)(C)(ii) because it was evidence that was not previously heard at trial and was discovered after trial. But that is only the first hurdle to relief. (See Masters, supra, 7 Cal.5th at p. 1078 [discussing the two hurdles in § 1473 for false evidence].) Davenport’s claim fails to clear the second hurdle as he has not presented a prima facie case Therrien’s 2017 statement “is sufficiently material and credible that it more likely than not would have changed the outcome of the case.” (§ 1473, subd. (b)(1)(C)(i).) As we have already discussed ante, Therrien’s 2017 statement was not material, and based on the evidence in this case, it is not more likely than not Therrien’s statement would have changed the outcome of Davenport’s trial.

55 Davenport asserts Therrien’s 2017 statement is proof Farmer testified falsely at trial. We agree with the Attorney General that Therrien’s statement does not establish Farmer’s testimony was false. In analyzing Davenport’s claim under Napue and Giglio, we concluded Davenport did not present a prima facie case Farmer’s testimony was false evidence, and we reach the same conclusion here. Therrien’s 2017 statement leaves open the possibility he was not present for or did not hear the entirety of Davenport and Farmer’s conversation. Also, as the Attorney General notes, Therrien’s account of the conversation is conflicting evidence, which is insufficient to establish Farmer’s testimony was false. Therrien’s 2017 statement presented a conflict in the evidence for the jury to resolve; it would not have resulted in Farmer’s testimony being excluded as false. (See Masters, supra, 7 Cal.5th at p. 1089.) Accordingly, the superior court properly denied this portion of Davenport’s section 1473 claim. D. Expert Testimony Concerning the Time and Manner of Lingle’s Death Davenport argues he is entitled to habeas relief under section 1473 because Haddix’s declaration is new evidence that presents a significant dispute and shows the testimony of the prosecution’s pathology experts Fischer and Fukumoto was false. We disagree. The prosecution presented expert opinion testimony at Davenport’s trial by Fischer and at the penalty retrial by Fukumoto that Lingle was alive when Davenport inserted the stake into her rectum. This issue was contested at both trials as the defense presented expert opinion testimony by Modglin at the guilt phase trial and Kornblum at the penalty retrial that Lingle was dead when impaled with the stake. Relying on Haddix’s declaration, Davenport asserts the testimony of prosecution experts

56 Fischer and Fukumoto was false or at least a significant dispute as defined in section 1473 has emerged. Davenport asserts the prosecution’s expert testimony concerning the timing and manner of Lingle’s death is false in four respects. We address the first three before turning to the fourth. First, Davenport contends Fischer falsely testified Lingle could have lived anywhere from 4 to 20 minutes after the injury to her carotid artery. Second, he argues Fukumoto’s reliance on the partial collapse of Lingle’s right lung “does not provide guidance in distinguishing ante from postmortem injury.” Third, Davenport asserts the condition of Lingle’s body is inconsistent with the theory the stake was inserted while she was alive as the absence of blood at points where the stake caused internal injuries indicates postmortem impalement. In each argument, Davenport relies on Haddix’s declaration, wherein she offered opinions contrary to those of Fischer and Fukumoto. These three are no more than disagreeing expert opinions— Haddix’s contrary opinion on these issues does not render Fischer’s or Fukumoto’s opinion false. (See People v. Johnson (2015) 235 Cal.App.4th 80, 91 [analyzing new evidence and false evidence claim under former § 1473].) For the most part, Haddix’s declaration merely reiterated or elaborated on the testimony of the defense experts who testified at Davenport’s trials. Based on the wounds to Lingle’s neck, Modglin opined Lingle would have died within five minutes and before that she would have entered a coma state in which she did not feel pain. Modglin testified, in contrast to Fischer’s testimony, the stake was inserted after Lingle died and not before and explained this conclusion was based on the lack of bleeding associated with the stake’s passage through Lingle’s body. Haddix theorizes Fischer’s

57 removal of the stake may have caused the partial lung collapse, but this is nothing more than supposition and does not render false the pathology expert testimony relying on the partial lung collapse. As to these three aspects of the prosecution’s pathology experts’ testimony, Haddix’s declaration does not show a significant dispute has emerged or further developed in Davenport’s favor. (§ 1473, subd. (b)(1)(D).)

Regarding the fourth aspect, Davenport asserts Fischer’s testimony that “dead people don’t bleed” has been undermined by later scientific research. This assertion is based on Haddix’s declaration and two medical studies cited in it. He therefore argues “the presence of 200 [cc] of blood in Lingle’s pleural cavity does not prove that she was alive at the time the stake was inserted.” Davenport has not made a prima facie showing these medical reports render false the testimony of the prosecution’s pathology experts (§ 1473, subd. (b)(2)) or reflect a significant dispute has emerged or further developed in his favor regarding the trial testimony (id., subd. (b)(1)(D)). Haddix summarizes the 2004 study on postmortem bleeding as showing “[a]n average of 400 cc of blood (range 100-1300 cc) were measured in the left pleural cavities after the passage of a minimum of five minutes.” At Davenport’s trial, the pathologists disagreed as to whether the presence of 200 cc of blood in Lingle’s right pleural cavity evidenced Lingle was alive when impaled. Quite simply, the 2004 postmortem bleeding study has little value in Davenport’s case. The circumstances under which the study was conducted and the postmortem bleeding measured bear no resemblance to the circumstances here. In the 2004 study, the causes of death did not involve a severed artery or the sudden loss of blood. Instead, the causes of death for the

58 study subjects were asphyxial (hanging, drowning, carbon monoxide intoxication, and electrocution) and sudden natural deaths (sudden cardiac deaths and intracerebral hemorrhages). (American Journal of Forensic Medicine and Pathology, Vol. 25, No. 1 (March 2004) p. 21.) In these instances, the blood remained intact in the body at the time of death. In this case, all the experts (both prosecution and defense) agree Lingle died as a result of the bleeding from the cut in her carotid artery. Haddix does not address how this study is relevant given the substantial difference in the cause of death.

Haddix also relies on an experiment reported in 1988 in the third edition of Forensic Medicine: A Guide to Principles (I. Gordon, H. A. Shapiro and S. D. Berson, Churchill Livingstone). According to Haddix, this study indicates the volume of blood found in “experimentally produced postmortem wounds in recently deceased individuals” ranged from 50 to 1000 cc. First, we know nothing about how this study was conducted. Second, even assuming its methodology rendered it relevant, its result revealed instances of blood seepage less than 200 cc. Thus, it does not render false Fischer’s or Fukumoto’s testimony. Haddix’s opinions, which merely reiterate those of the defense pathology experts who testified at trial, do not provide a prima facie showing of new evidence under section 1473, subdivision (b)(1)(C). Moreover, the two studies Haddix relies upon in formulating her opinions are not new studies as one was published in 1988 (prior to the penalty retrial) and the second in 2004 (13 years prior to the filing of the instant habeas petition). Thus, to the extent Davenport is relying on these studies as new evidence under section 1473, this portion of his claim fails as they were not “presented without substantial delay.” (§ 1473, subd. (b)(1)(C)(i).)

59 Therefore, the superior court properly denied this portion of Davenport’s section 1473 claim.

E. Tire Track Evidence Davenport contends the PCAST report constitutes new evidence that renders false the prosecution’s evidence matching the tires on his motorcycle to the tire track impressions at the crime scene. Although the PCAST report did not discuss tire tread comparison, Davenport asserts the report demonstrates the prosecution’s tire track evidence has been undermined by later scientific research. He asserts Leonard and Jackson’s work was a subjective forensic feature-comparison method and because neither employed the techniques recommended in the PCAST report, the evidence matching his motorcycle tire tread to the impressions at the scene was “‘scientifically meaningless.’” Even assuming the PCAST report constitutes new evidence under section 1473, subdivision (b)(1)(C)(ii), it is not sufficiently material that it more likely than not would have changed the outcome of the case. The PCAST report has little relevance to the tire track evidence in Davenport’s case. Most of the evidence presented at Davenport’s trial concerning his motorcycle tires and the tracks at the crime scene involved the tread design and wear. The powerful testimony was that concerning the types of tires on Davenport’s motorcycle (two different brands) and that the rear tire had a defect causing a unique tread pattern. (Davenport I, supra, 41 Cal.3d at p. 258.) The PCAST report specifically states it does not address comparison of “‘class characteristics’ (such as design, physical size, and general wear)” because “it is not inherently a challenging measurement problem to determine class characteristics, to estimate the frequency of shoes having a particular class characteristic, or (for jurors) to understand the nature of the

60 features in question.” It was not difficult for the jurors in Davenport’s trials to understand the features at issue in the tire track evidence. Davenport’s motorcycle was an exhibit at trial and the jurors could themselves compare the treads on his tires with the impressions from the tracks at the scene. Thus, even if the defense had used the PCAST report to challenge the tire track evidence at Davenport’s trial, it is not likely the jury would have reached a different verdict. Moreover, even apart from the tire track evidence, there was overwhelming evidence of Davenport’s guilt. Davenport’s assertion the PCAST report renders the tire tread analysis false evidence fares no better. It does not clear the first hurdle of establishing the trial testimony concerning the tire treads was false. As the Attorney General notes, Davenport’s argument at most is that the testimony of Leonard and Jackson concerning the tire treads was “‘scientifically meaningless’” because their techniques were not validated, standardized, or automated, and they did not report error or false-positive rates. But this does not equate to their testimony being false. Accordingly, the superior court properly denied this portion of Davenport’s section 1473 claim.

61

CONCLUSION

The order denying the petition for writ of habeas corpus is affirmed.

MOTOIKE, ACTING P. J.

WE CONCUR:

DELANEY, J.

GOODING, J.

62