State v. Worley

2020 NMSC 021, 476 P.3d 1212

• Court: New Mexico Supreme Court
• Decided: August 27, 2020
• Status: Published

Opinion

Office of the Director New Mexico 10:46:33 2020.12.16 Compilation '00'07- Commission

IN THE SUPREME COURT OF THE STATE OF NEW MEXICO

Opinion Number: 2020-NMSC-021

Filing Date: August 27, 2020

No. S-1-SC-37526

STATE OF NEW MEXICO,

Plaintiff-Petitioner,

v.

CURTIS WORLEY,

Defendant-Respondent.

APPEAL FROM THE DISTRICT COURT OF EDDY COUNTY Jane Shuler-Gray, District Judge

Released for Publication December 22, 2020.

Hector H. Balderas, Attorney General Martha Anne Kelly, Assistant Attorney General Santa Fe, NM

for Petitioner

Duncan Earnest, LLC Theresa M. Duncan Albuquerque, NM

for Respondent

OPINION

THOMSON, Justice.

The district court granted Defendant Curtis Worley’s petition for a writ of habeas corpus without providing a basis for granting relief. The State appealed. See Rule 5- 802(N)(1) NMRA (giving the state the right to appeal an order granting a writ of habeas corpus when a death sentence has not been imposed); see also Rule 12-102(A)(3) NMRA (requiring that “appeals from the granting of writs of habeas corpus” be taken to this Court). Defendant argues that the district court order granting the writ of habeas corpus was proper because (1) Defendant was actually innocent and new evidence in the form of recanted testimony, DNA evidence, and an allegedly suppressed statement undermines confidence in the verdict, and also because (2) the allegedly suppressed statement violated Defendant’s due process rights pursuant to Brady v. Maryland, 373 U.S. 83 (1963), undermining the fundamental fairness of the trial. Based on the analysis that follows, we disagree and reverse the district court order that granted the writ and a new trial.

I. BACKGROUND

Nancy Mitchell’s decomposing body was discovered on January 30, 1982, near the Pecos River in Eddy County, New Mexico. See State v. Worley, 1984-NMSC-013, ¶ 4, 100 N.M. 720, 676 P.2d 247 (citing State v. Case (Case I), 1984-NMSC-012, ¶ 4, 100 N.M. 714, 676 P.2d 241). The State charged Defendant with the murder and criminal sexual penetration of Mitchell and prosecuted Defendant based on testimony from multiple witnesses that he was one of a group of men that attacked and raped Mitchell on January 1, 1982. The jury convicted Defendant of first-degree felony murder, NMSA 1978, § 30-2-1(A) (1980), and first-degree criminal sexual penetration, NMSA 1978, § 30-9-11(A) (1975), and the district court sentenced him for both crimes to imprisonment for life plus eighteen years, see NMSA 1978, § 31-18-14(A) (1979); NMSA 1978, § 31- 18-15(A)(1) (1981). This Court affirmed Defendant’s convictions on direct appeal. Worley, 1984-NMSC-013, ¶¶ 1-2.

Defendant petitioned the Eddy County District Court in 2004 for a writ of habeas corpus. The petition alleged that “two of the three purported eyewitnesses to the alleged crime . . . disclose[d] for the first time to the defense that they fabricated their testimony,” and Defendant asserted that he was entitled to relief based on these “newly discovered recantations.”

Co-defendant Carl Case was convicted of the murder and criminal sexual penetration of Mitchell in a separate proceeding. Case I, 1984-NMSC-012, ¶¶ 1, 4; see Worley, 1984-NMSC-013, ¶¶ 1, 9. This Court affirmed Case’s convictions on direct appeal, see Case I, 1984-NMSC-012, ¶¶ 2, 28, and Case also filed a petition for writ of habeas corpus in the Eddy County District Court, Case v. Hatch (Case II), 2008-NMSC- 024, ¶¶ 1-4, 144 N.M. 20, 183 P.3d 905. In the habeas petition, Case argued that he was entitled to relief based on “newly discovered” recanted testimony and a 1982 statement by Bobby Autry that Case asserted was not disclosed in violation of Brady. 1

1Defendant’s answer brief in this appeal explains that investigators in Case’s habeas proceedings

discovered a statement that Autry made to law enforcement on February 3, 1982, which the prosecution had possessed and failed to disclose to Case’s trial counsel. Defendant maintains that this statement was favorable to his defense and that the prosecution failed to disclose the statement to his trial attorney as well. Defendant’s answer brief also discusses “new” DNA results favorable to his case, obtained from tests performed in 2005, during Case’s habeas proceedings, on physical evidence obtained from Mitchell’s body and clothing in 1982. Case II, 2008-NMSC-024, ¶¶ 1-4, 44, 48. After holding evidentiary hearings in 2005 and 2006, the district court denied Case’s habeas petition. Id. ¶ 4.

In late 2007, before the district court held an evidentiary hearing on Defendant’s habeas petition, Defendant agreed to stay the proceedings pending the outcome of Case’s appeal of the denial of his habeas petition because of “the overlap of both factual and legal issues between that case and this one.” In April 2008, this Court affirmed the denial of Case’s petition for a writ of habeas corpus. Id. Case filed a petition for a writ of habeas corpus in the United States District Court approximately two months later.

On September 2, 2008, Defendant filed a supplemental petition and memorandum in support of his previously stayed habeas petition, adding to his own case as additional grounds both the alleged suppression of the Autry statement and the allegedly new DNA evidence that Case had asserted as grounds for relief. Ultimately, the district court did not consider the merits of Defendant’s petition until after the resolution of Case’s federal habeas petition. See Case v. Hatch (Case III), 773 F. Supp. 2d 1070, 1071, 1149 (D.N.M. 2011) (conditionally granting habeas relief), vacated by Case v. Hatch (Case IV), 731 F.3d 1015, 1019, 1044 (10th Cir. 2013) (vacating the conditionally granted habeas relief).

The Tenth Circuit denied Case federal habeas relief, holding that the alleged Brady evidence failed to meet “the Brady materiality standard” and therefore the federal district court did not have jurisdiction to reach the merits. Case IV, 731 F.3d at 1042-44. The Tenth Circuit described the DNA evidence as “inconclusive” and agreed with the analysis of this Court in Case II that the recanted testimony did no more than “revert to the original statements [those witnesses] gave to the police.” Id. at 1043-44; see Case II, 2008-NMSC-024, ¶ 39.

Defendant relied on transcripts from the evidentiary hearings conducted for Case II and Case III rather than calling the recanting witnesses Audrey Knight and Paul Dunlap to testify in this matter. Additional relevant testimony from the evidentiary hearings on Case’s state habeas petition conducted for Case II concerned the result and effect of both the DNA evidence and the alleged failure to disclose the February 3, 1982, statement to Case’s attorney. Defendant simply “incorporated the evidence presented during the Carl Case hearings [conducted for Case II] as well as the Supreme Court’s [opinion] in [Case II]” and “supplemented the record with additional exhibits, including, inter alia, additional transcripts from the Carl Case federal proceedings, judicial opinions from those proceedings [Case III and Case IV], and a videotaped deposition of [Defendant].” Defendant argued, “[T]he real issue for this court is how [Defendant’s] trial was different from Carl Case. If this was exactly the same trial then you’d be governed by the different opinions, but it’s not, and that, from our perspective, is legally significant.”

The district court summarily granted Defendant’s habeas petition, and the State appealed. II. ANALYSIS

Defendant argues that the district court properly granted a new trial because he is actually innocent and because no reasonable juror, considering what he asserts is new evidence, would have convicted him. Alternatively, Defendant argues that the State did not provide his defense counsel with a statement Autry made to law enforcement on February 3, 1982, which Defendant asserts constitutes a Brady violation and undermines the fundamental fairness of the trial. We first review Defendant’s allegation of a Brady violation based on Autry’s February 3, 1982, statement and then examine Defendant’s actual innocence claim.

A. Standard of Review

“When reviewing the propriety of a lower court’s grant or denial of a writ of habeas corpus, the trial court’s findings of fact concerning the habeas petition are reviewed to determine if substantial evidence supports the [trial] court’s findings.” Dominguez v. State, 2015-NMSC-014, ¶ 9, 348 P.3d 183 (alteration in original) (internal quotation marks and citation omitted). However, “[q]uestions of law or questions of mixed fact and law . . . are reviewed de novo.” Id. (omission in original) (internal quotation marks and citation omitted).

Although this Court will give deference to the district court in its role as “fact- finder” when the district court is a “first-hand observer,” this Court must nonetheless perform its “sanctioned role as arbiter of the law.” Duncan v. Kerby, 1993-NMSC-011, ¶ 7, 115 N.M. 344, 851 P.2d 466. A decision that is ordinarily discretionary but is “premised on a misapprehension of law” may be categorized as an abuse of discretion. N.M. Right to Choose/NARAL v. Johnson, 1999-NMSC-028, ¶ 7, 127 N.M. 654, 986 P.2d 450 (internal quotation marks and citation omitted); see also State v. Torres, 1999- NMSC-010, ¶ 28, 127 N.M. 20, 976 P.2d 20 (observing that the threshold question, whether the lower court applied the correct law to the facts, is reviewed de novo). This approach provides deference to the fact-finder as first-hand observer while ensuring that we perform our ultimate role, assuring proper application of the law. Duncan, 1993- NMSC-022, ¶ 7.

B. Alleged Brady Violation

Defendant alleges that the prosecution failed to turn over an audio recording or transcript of Autry’s February 3, 1982, statement to law enforcement and that the failure violated Defendant’s constitutional right to due process. See Brady, 373 U.S. at 87 (holding that “the suppression by the prosecution of evidence favorable to an accused upon request violates due process where the evidence is material either to guilt or to punishment, irrespective of the good faith or bad faith of the prosecution”); accord Case II, 2008-NMSC-024, ¶ 44.

1. Legal test and standard of review A district court’s conclusion that a Brady violation occurred is a legal determination of whether the prosecution violated the requirements of due process. See Smith v. Cain, 565 U.S. 73, 75 (2012) (“Under Brady, the State violates a defendant’s right to due process if it withholds evidence that is favorable to the defense and material to the defendant’s guilt or punishment.”); see also Cone v. Bell, 556 U.S. 449, 469 (2009) (observing that when the prosecution violates Brady it violates due process of law pursuant to the Fourteenth Amendment).

“Claims involving the denial of procedural due process are legal questions that [this Court] review[s] de novo.” Miller v. Tafoya, 2003-NMSC-025, ¶ 9, 134 N.M. 335, 76 P.3d 1092; see Fry v. Lopez, 2019-NMSC-013, ¶ 12, 447 P.3d 1086 (observing that this Court reviews constitutional challenges to convictions de novo). 2

Meeting the following three requirements establishes a Brady violation:

The evidence at issue must be favorable to the accused, either because it is exculpatory, or because it is impeaching; that evidence must have been suppressed by the State, either willfully or inadvertently; and prejudice must have ensued.

Strickler v. Greene, 527 U.S. 263, 281-82 (1999). We first analyze whether the February 3, 1982, statement was suppressed, then establish whether it was favorable to Defendant, and finally determine whether it was material (that is, whether suppression resulted in prejudice).

2. For purposes of this habeas proceeding we presume the evidence was suppressed

Under Brady and subsequent federal law, “the individual prosecutor has a duty to learn of any favorable[-to-the-defendant] evidence known to the others acting on the government’s behalf in th[e] case, including the police.” Strickler, 527 U.S. at 281 (internal quotation marks and citation omitted). Although “police investigators sometimes fail to inform a prosecutor of all they know[,] . . . procedures and regulations can be established to carry [the prosecutor’s] burden and to insure communication of all relevant information on each case to every lawyer who deals with it.” Kyles v. Whitley, 514 U.S. 419, 438 (1995) (second alteration in original) (internal quotation marks and citation omitted). Therefore, even if the evidence at issue was not disclosed because it was overlooked, we presume it was suppressed. See Giglio v. United States, 405 U.S. 150, 154 (1972) (“[W]hether the nondisclosure was a result of negligence or design, it is the responsibility of the prosecutor.”); accord Case II, 2008-NMSC-024, ¶ 45 (“[I]t is

2The federal district court suggested that this Court improperly applied a deferential abuse-of-discretion

standard in determining there was no Brady violation in Carl Case’s state habeas proceeding. See Case III, 773 F. Supp. 2d at 1135-36. Although Case II states, “An alleged Brady violation is a charge of prosecutorial misconduct [that] is reviewed for abuse of discretion,” 2008-NMSC-024, ¶ 47, it appears to defer only to the trial court’s factual determinations while reviewing legal questions de novo, id. ¶ 55 (holding that the suppressed Autry statement was not material and that the trial court did not abuse its discretion by denying habeas relief). irrelevant whether the prosecutor failed to disclose evidence either intentionally or negligently.”).

Defendant’s attorney Charles A. Feezer was deceased by the time Defendant filed his habeas petition in 2004. Defendant argues that “[a]lthough [his] trial counsel, Charles Feezer, is deceased and thus unable to testify, the transcript of the trial establishes that Feezer was also unaware” of the February 3, 1982, Autry statement.

Our review of the trial transcript determines that, during cross-examination of Autry, Feezer enumerated by date two of four recorded statements that Autry made to law enforcement in 1982 and made no mention of the February 3, 1982, statement. We observe that the prosecutor in redirect examination of Autry also omitted the February 3, 1982, statement in enumerating by date three of the four recorded statements Autry made to law enforcement in 1982 for Autry to confirm. 3

This Court in Case II ostensibly presumed suppression by proceeding further with the Brady analysis after stating, “Ultimately, we are not left with a clear answer about whether the taped statement was suppressed by the prosecution.” 2008-NMSC- 024, ¶ 49. Although the evidence presented in Case II supported a presumption that the February 3, 1982, statement was suppressed in that case and thus suggests that the evidence was also suppressed in this case, it does not establish that the prosecution did not disclose the evidence to Feezer.

Remaining mindful of this uncertainty and considering that the purpose of Brady is to reinforce the integrity of the judicial system, we presume without deciding that the statement was suppressed in this case. Cf. United States v. Cooper, 654 F.3d 1104, 1119 (10th Cir. 2011) (“Under Brady, the suppression by the prosecution of evidence favorable to an accused upon request violates due process where the evidence is material either to guilt or to punishment, irrespective of the good faith or bad faith of the prosecution.” (internal quotation marks and citation omitted)).

3. The evidence was favorable to Defendant

Evidence is “favorable to an accused” if its disclosure and effective use “may make the difference between conviction and acquittal” regardless of whether such evidence is impeachment evidence or exculpatory evidence. United States v. Bagley, 473 U.S. 667, 676 (1985) (internal quotation marks and citation omitted).

In his February 3, 1982, statement, Autry admitted that he had lied to officers about having intercourse with Mitchell because “I didn’t know if she was raped at the time or what, and I didn’t want nobody to point the finger at me cause I didn’t do it.” Autry told officers about a sexual encounter with Mitchell in Autry’s car, ostensibly in December of 1981:

3“Well, we have got several of them. You gave one on January the 30th, March the 5th and March the

12th, is that right?” I just – I just – we laid there foolin’ around, kissin’ and stuff, and I was playing with her breasts, and then she just – I – I tried to – I got fixed up about half way in, and she said no, she pushed back, and I said alright. Then she said let’s go on out to that party and I said well, let’s go. Got dressed up and we both crawled back in the front seat of my car and we went out to the – we went by and got Mike and we went out to the party.

When asked by officers if he was mad at Mitchell, Autry responded, “No. I got – I got mad, yeah, but not, not that mad. I said well hell, ain’t no great big loss to get turned down by a girl.”

When defense counsel attempted to cross-examine Autry about his relationship with Mitchell at trial, Autry admitted that he made a sexual advance toward her but stated that he did not remember when it had occurred.

The jury weighed each witness’s credibility to determine the facts of the case. If defense counsel had use of the February 3, 1982, statement to more effectively impeach Autry, it could have affected the jury’s assessment of Autry’s credibility, but this hypothetical difference between possible cross-examinations of Autry is one of degrees. Defense counsel did test Autry’s credibility concerning his relationship with Mitchell. Further, other evidence at trial corroborated Autry’s testimony concerning the gang rape and murder of Mitchell on January 1, 1982. Autry’s statement does not make a clear difference between conviction and acquittal, but because it could have affected the jury’s assessment of Autry’s credibility, the February 3, 1982, statement is favorable. See Case II, 2008-NMSC-024, ¶¶ 50, 53 (determining that Autry’s February 3, 1982, statement “has some impeachment value” and is therefore favorable evidence). We now turn to the question of whether the statement was material.

4. The statement was not material

A “real” Brady violation requires “nondisclosure . . . so serious that there is a reasonable probability that the suppressed evidence would have produced a different verdict.” Strickler, 527 U.S. at 281. The crux of the Brady analysis “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.” Id. at 289-90 (internal quotation marks and citation omitted). Concerning whether the suppressed evidence was material to the verdict, this Court has stated, “Materiality only exists if the suppressed evidence ‘creates a reasonable doubt that did not otherwise exist.’” Case II, 2008-NMSC-024, ¶ 54 (quoting United States v. Agurs, 427 U.S. 97, 112 (1976)).

“To make the materiality determination, we view the suppressed evidence’s significance in relation to the record as a whole.” Cooper, 654 F.3d at 1120 (10th Cir. 2011) (internal quotation marks and citation omitted); accord Case II, 2008-NMSC-024, ¶ 53 (“Evaluating materiality under Brady requires us to look at the entire trial to determine whether the defendant’s conviction was obtained by violating due process, whether his or her trial was tainted with fundamental unfairness because certain evidence was not disclosed to the defense.” (internal quotation marks and citation omitted)). The suppressed evidence is considered material only if it “could reasonably be taken to put the whole case in such a different light as to undermine confidence in the verdict.” Strickler, 527 U.S. at 290 (internal quotation marks and citation omitted). A new trial is not automatically required “whenever ‘a combing of the prosecutors’ files after the trial has disclosed evidence possibly useful to the defense but not likely to have changed the verdict. . . .’” Case II, 2008-NMSC-024, ¶ 54 (quoting Giglio, 405 U.S. at 154).

Defendant argues that the February 3, 1982, statement was material because not having the statement affected his ability to suggest and argue (1) that Autry fabricated his testimony and (2) that Autry had a motive to kill Mitchell based on her rejection of him. Defendant argues that his impeachment of Autry was less effective than it could have been. Even presuming that Defendant’s impeachment of Autry would have been more effective, the absence of the February 3, 1982, statement does not undermine confidence in the verdict within the context of the entire trial.

Autry was cross-examined on his prior inconsistent statements, the facts that he failed a polygraph test and had “told a lot of lies,” which Autry admitted. The February 3, 1982, statement was cumulative evidence that Autry did not always tell the truth or the whole truth. “[W]here the credibility of a witness has already been substantially called into question in the same respects by other evidence, additional impeachment evidence will generally be immaterial and will not provide the basis for a Brady claim.” Cooper, 654 F.3d at 1120 (internal quotation marks and citation omitted). “Where evidence ‘insignificantly impact[s] the degree of impeachment,’ it generally will ‘not be sufficient to meet the . . . materiality standard.’” Id. at 1120 (alteration and omission in original) (quoting Douglas v. Workman, 560 F.3d 1156, 1174 (10th Cir. 2009)); accord Case IV, 731 F.3d at 1041-42; Case II, 2008-NMSC-024, ¶ 54.

Defendant argues that the suppression of the statement is more material in his case than in Case’s habeas matter because “[a]lthough the tape would not have changed Curtis’s defense, it undeniably would have strengthened it.” As support, Defendant relies on his alibi and his own testimony that Mitchell “was scared that Bobby Autry was going to come back up there and bother her,” and that the last time he saw Mitchell she was with Autry on December 20, 1981. Further, Defendant asserts the February 3, 1982, statement had “unique impeachment value” because although at Defendant’s trial Autry admitted making sexual advances, he “denied having an intimate relationship with” Mitchell.

Defendant’s argument that in December 1981, Autry had a motive to kill Mitchell minimizes his alternative theory that Mitchell was still alive in mid-January. Defense witnesses testified that they saw Mitchell alive on January 6, or January 16, 1982, days after she was sexually assaulted and murdered. Although the jury evidently discredited that theory, it undercut Defendant’s own speculative theory that Autry could have killed Mitchell around December 20, 1981. Defendant’s argument that the February 3, 1982, statement is material exculpatory evidence also fails because “evidence that another person had a motive to commit the crime for which a defendant is on trial is generally inadmissible, absent direct or circumstantial evidence linking the third person to the crime.” Case II, 2008- NMSC-024, ¶ 55; see State v. Rosales, 2004-NMSC-022, ¶ 10, 136 N.M. 25, 94 P.3d 768 (discussing the direct-evidence requirement in other jurisdictions for consideration of third-person motive evidence). “For Brady purposes, exculpatory evidence cannot be purely speculative.” Case II, 2008-NMSC-024, ¶ 55; accord Case IV, 731 F.3d at 1041 (quoting United States v. Fleming, 19 F.3d 1325, 1331 (10th Cir. 1994) (“‘The mere possibility that evidence is exculpatory does not satisfy the constitutional materiality standard.’”)). Although Defendant argues that Autry had a motive to kill Mitchell based on Autry’s February 3, 1982, statement, it does not appear that the statement would have been admissible, let alone materially exculpatory, based on the record.

The jury convicted Defendant primarily on the basis of testimony from multiple eyewitnesses whose stories reinforced each other despite that they did not perfectly match. Defendant now challenges the credibility of some trial witnesses, insinuating that they lied and questioning their motives for testifying. But Autry’s February 3, 1982, statement is cumulative and speculative in light of all the evidence presented at trial. See Case II, 2008-NMSC-024, ¶¶ 54-55. Further, the jury appears to have disregarded the credibility (1) of testimony concerning Defendant’s alibi defense, (2) of testimony that Mitchell was alive after January 1, and (3) of Defendant, in testifying on his own behalf.

When viewed in context, we decline to conclude that justice has not been done or that there is a reasonable probability that the February 3, 1982, statement, if it had been produced and effectively used by defense counsel, would have delivered a different verdict. Thus, there was no Brady violation. See Strickler, 527 U.S. at 281. We now turn to Defendant’s claim of actual innocence.

C. Defendant’s Actual Innocence Claim Is Not Supported by New Evidence

Evidence used in a habeas proceeding to support a claim for actual innocence must be new evidence. Montoya v. Ulibarri, 2007-NMSC-035, ¶ 30, 142 N.M. 89, 163 P.3d 476 (“[A habeas] petitioner asserting a freestanding claim of innocence must convince the court by clear and convincing evidence that no reasonable juror would have convicted him in light of the new evidence.”). Defendant argues that the new evidence includes the recanted testimony, the DNA evidence, and the nondisclosed statement from the State’s eyewitness Autry who testified at trial.

1. The recanted testimony is not new evidence

“[A] petitioner seeking a new trial through a writ of habeas corpus because of recanted testimony must prove, based upon the entire record, including the original trial proceedings at issue, that the recantation is credible and was significant to the original verdict.” Case II, 2008-NMSC-024, ¶ 17. To determine that the recantation is credible, the district court must analyze the following factors, “none of which is dispositive” on its own:

(1) [T]he original verdict was based upon uncorroborated testimony; (2) the recantation is corroborated by additional new evidence; (3) the recantation occurred under circumstances free from suspicion of undue influence or pressure from any source; (4) the record fails to disclose any possibility of collusion between the defendant and the witness between the time of the trial and the retraction; and (5) the witness admitted the perjury on the witness stand and thereby subjected himself or herself to prosecution.

Id. In addition to these factors, the district court must also weigh the credibility of the witness. Id.

To determine that the recantation is significant, the district court must conclude that the testimony meets all of following requirements:

(1) [The testimony] must have been discovered since the trial; (2) it could not have been discovered before the trial by the exercise of due diligence; (3) it must be material; (4) it must not be merely cumulative; (5) it must not be merely impeaching or contradictory; and (6) the court is left with a firm belief that but for the perjured testimony, the defendant would most likely not have been convicted.

Id. ¶ 17 (“With the exception of the firm belief standard which we announce today, the remaining factors are already a part of New Mexico habeas corpus jurisprudence.” (internal quotation marks and citation omitted)).

Whether a recantation is significant is a mixed question of fact and law because the ultimate determination of significance is based on interim determinations, some of which are legal. For example, whether the recantation is material is a legal determination. See State v. Carabajal, 1920-NMSC-086, ¶ 8, 26 N.M. 384, 193 P. 406 (holding that a trial court’s statement concerning the materiality of testimony subsequently contradicted by current statements of the same witness was not a “comment upon the weight of the evidence,” but instead “it was a statement of a legal objection to the evidence”). As this Court previously acknowledged, we give deference to the district court’s role as “fact-finder” when it is a “first-hand observer” while we perform our “sanctioned role as arbiter of the law.” Duncan, 1993-NMSC-011, ¶ 7; see New Mexico Right to Choose/NARAL, 1999-NMSC-028, ¶ 7 (characterizing a discretionary decision “premised on a misapprehension of law” as an abuse of discretion). In habeas appeals, a mixed question of fact and law is subject to de novo review. Miller, 2003-NMSC-025, ¶ 9.

The district court necessarily relied on transcripts of evidentiary hearings in Case’s state and federal habeas proceedings to grant habeas relief, but did not provide any findings related to the credibility or significance of the recantations. This is problematic in part because the district court could not perform its function as a first- hand observer to evaluate the credibility of the recanting witnesses.

We presume the district court deferred to the first-hand observations of the respective courts that heard the testimony in Case’s state and federal habeas proceedings. The state district court ostensibly did not analyze credibility “because a finding that the recantations were not newly-discovered evidence is dispositive.” Case II, 2008-NMSC-024, ¶ 34 (“[A] negative finding of any of the six factors under the significance prong is dispositive.”). The federal district court determined that both Knight and Dunlap were credible after that court held a hearing that was specifically limited as to the witnesses’ credibility. See Case III, 733 F. Supp. 2d at 1111.

The credibility of the witness is however only one of the factors that a court must consider to determine whether the recantation is credible. See Case II, 2008-NMSC- 024, ¶ 17 (establishing that five factors must be considered in addition to the credibility of the witness). Because our determination that the recantations are not significant is dispositive, we do not review the credibility prong of the analysis here. See id. ¶ 34. Nonetheless, “rather than presuming the credibility of the recantations,” we strongly urge trial courts to “make findings with respect to each factor . . . before addressing the significance prong,” id., to facilitate our review and to minimize the issues for review in each case.

Concerning the significance prong, a defendant must prove that the recanted statements were “qualitatively different” from prior statements and testimony to constitute new evidence. Id. ¶ 35. Defendant urges this Court to overrule our prior determination that Knight’s and Dunlap’s recantations were an effort “to revert to the original statements they gave to the police that they did not know anything about the events leading to Mitchell’s death” and that “[t]he recantations were also cumulative.” Id. ¶¶ 39-40. Defendant suggests that we should treat the recantations of Knight and Dunlap as new evidence because these witnesses now swear that their trial testimony was a lie, despite having sworn at trial that their inconsistent trial testimony was true.

On cross-examination, Knight was questioned and she answered:

Q. [ ] Do you remember . . . giving those answers on this statement of yours on the 14th of March?

A. No, I don’t.

Q. Well, were you lying, then, or were you telling the truth?

A. I don’t remember.

....

Q. Counsel labors with you about your lies and says that there were 42 and 16 pages of statements, essentially all lies, is that right? A. No.

Q. A lot of them lies?

A. I don’t know I haven’t read them. I don’t remember them.

On cross-examination, Dunlap was questioned and he answered:

Q. And you stayed in jail and said nothing for several months?

A. Yes, sir.

Q. Until finally you had enough and you decided to change your story?

Q. Okay. So, on September the 2nd you finally decided to change your story after they offered you the immunity?

Q. Because prior to that time you had told them what you are saying now were a bunch of lies?

Q. All right. Now, when you gave the statement on 3-10-82, before the polygraph, were you telling the truth or were you lying?

A. I was lying and telling the truth, too.

Q. A little of each?

The fact that Defendant continues to assert he is innocent does not alter the nature of the recantations, which are merely contradictory and impeaching because at trial, while giving the inconsistent testimony that they now recant, Knight and Dunlap admitted that they previously lied to the police. Knight’s and Dunlap’s recantations do not appear to be qualitatively different from their original statements, and instead the recantations go to the question of the weight of their testimony concerning conflicting statements they previously made. Further, other testimony corroborated the facts supporting Defendant’s conviction. This other testimony has not been recanted.

Jimmy Ray Hood testified, without recanting, that regardless of the photographs of Defendant at a New Year’s Eve gathering in Loving, New Mexico, Hood saw Defendant at the Carlsbad area party location on New Year’s Eve, December 31, 1981, which undermined Defendant’s alibi.

Autry testified that he and Defendant, Case, Mike Tweedy, Dunlap, and Joe Brown were with Mitchell immediately prior to the sexual assault on the following day but that he got scared and took off toward his house about a mile away when the other five commenced the sexual assault of Mitchell. Autry never recanted his testimony, which the jury ostensibly credited.

Randy Davis testified that he overheard Defendant tell Tweedy “he wanted something” as payment for everything he was doing for Mitchell and that Defendant thereafter admitted having had sexual intercourse with Mitchell. But contrary to what he originally told law enforcement, Davis denied on cross-examination that he had asked to be part of the sexual assault and that Defendant and Tweedy would not let him.

Knight’s and Dunlap’s recantations do not differ qualitatively from their original declarations that they did not know anything about the assault and murder of Mitchell. Therefore, the recantations are simply indicators of the credibility of these statements and of the witnesses themselves. See Case II, 2008-NMSC-024, ¶ 17. The jury assessed the credibility of the statements at Defendant’s 1983 trial. If the district court granted habeas relief based on Knight’s and Dunlap’s recantations, then the district court impermissibly reweighed the evidence. See State v. Garcia, 2011-NMSC-003, ¶ 5, 149 N.M. 185, 246 P.3d 1057 (observing that “appellate courts will not invade the jury’s province as fact-finder by second-guess[ing] the jury’s decision concerning the credibility of witnesses, reweigh[ing] the evidence, or substitut[ing] its judgment for that of the jury.” (alterations in original) (internal quotation marks and citation omitted)). The recantations are not new evidence.

2. The February 3, 1982, statement and the DNA results are not new evidence

The DNA evidence and the February 3, 1982, statement are subject to a slightly different standard than the significance standard for recanted testimony. The standard for newly discovered evidence requires the evidence to fulfill all of the following requirements:

“1) [The evidence] will probably change the result if a new trial is granted; 2) it must have been discovered since the trial; 3) it could not have been discovered before the trial by the exercise of due diligence; 4) it must be material; 5) it must not be merely cumulative; and 6) it must not be merely impeaching or contradictory.”

State v. Garcia, 2005-NMSC-038, ¶ 8, 138 N.M. 659, 125 P.3d 638 (citation omitted). We reiterate that although this Court generally defers to a district court’s ultimate determination concerning the grant or denial of habeas relief, like our determination of the significance of recantations, our determination of new evidence also considers mixed questions such as materiality for which we review the application of law to the facts de novo. See Carabajal, 1920-NMSC-086, ¶ 8; see also Miller, 2003-NMSC-025, ¶ 9.

As we concluded in the Brady analysis, the February 3, 1982, statement is not material and is cumulative and merely impeaching or contradictory. Therefore, the February 3, 1982, statement fails to fulfill all the requirements for new evidence.

The new DNA test evidence 4 also fails to fulfill all the requirements for new evidence. At trial, the jury convicted Defendant of criminal sexual penetration despite expert testimony that the testing of samples taken from in or on Mitchell’s body revealed no sperm cells (“no evidence of sexual intercourse”) in any samples. Because criminal sexual penetration could have been accomplished “with any object” and did not require “emission,” see § 30-9-11(1975), the lack of sperm (male DNA) evidence may be relevant, but evidence of its absence is not conclusive. Although the sensitivity of the more recent DNA testing is likely an improvement over the testing performed for the original trial, the new DNA test result is qualitatively the same type of evidence—the absence of male DNA evidence. The unavailability of DNA testing in 1982 does not make the more recent test result new evidence.

The jury verdicts indicate that Knight’s and Dunlap’s trial testimony was credible, that Autry’s trial testimony was credible, and that the absence of male ejaculate did not create reasonable doubt. Based on our analysis here, the recantations, the February 3, 1982, statement, and the cumulative evidence of absent male DNA do not meet the standard of clear and convincing evidence on which no reasonable juror would have convicted Defendant. If the district court granted habeas relief based on Defendant’s argument that new evidence supported his claim of actual innocence, then the district court abused its discretion by misapplying the law.

III. CONCLUSION

Based on the foregoing, we reverse the district court order granting Defendant the writ of habeas corpus and a new trial.

IT IS SO ORDERED.

DAVID K. THOMSON, Justice

WE CONCUR:

4Defendant’s 2008 petition for a writ of habeas corpus describes the “new” DNA result from more

“sophisticated” testing of 1982 samples taken from Mitchell’s body and clothing as showing the “absence of male DNA.” MICHAEL E. VIGIL, Chief Justice

BARBARA J. VIGIL, Justice

C. SHANNON BACON, Justice

JUDITH K. NAKAMURA, Justice (specially concurring)

NAKAMURA, Justice (special concurrence).

I concur in the result reached by my colleagues. Respectfully, I am compelled to write separately, to 1) explain the impact of the district court’s failure to enter findings of fact and conclusions of law; 2) accurately set forth the standard of review, particularly for habeas claims based on alleged newly discovered evidence; 3) clarify the law regarding Brady claims; and 4) discuss Defendant’s actual innocence claim—an independent claim requiring a distinct analysis.

I. FINDINGS OF FACT AND CONCLUSIONS OF LAW IN HABEAS PROCEEDINGS

The first issue is our approach to the factual record in this case. Generally, in habeas appeals, we review the district court’s findings for an abuse of discretion— meaning, we ask whether substantial evidence supports the findings. Miller, 2003- NMSC-025, ¶ 9; see also Duncan, 1993-NMSC-011, ¶ 7 n.2 (noting that, on habeas review, “our degree of deference is . . . a review for support by substantial evidence as that standard is applied in New Mexico”). But here, despite the State’s timely request, the district court failed to enter findings of fact and conclusions of law, and instead issued only a brief order in support of its judgment on Defendant’s habeas petition.

As a threshold matter, I would hold that the district court’s failure to enter findings and conclusions violated Rule 1-052(A) NMRA, which provides as follows:

In a case tried by the court without a jury, or by the court with an advisory jury, the court shall enter findings of fact and conclusions of law when a party makes a timely request. Findings of fact and conclusions of law are unnecessary in decisions on motions under Rules 1-012, 1-050, or 1-056 NMRA or any other motion except as provided in Paragraph B of Rule 1- 041 NMRA.

(Emphasis added). Therefore, upon timely request by a party, a district court must, at the conclusion of a bench trial, enter findings and conclusions in support of the judgment. See, e.g., Aguayo v. Vill. of Chama, 1969-NMSC-005, ¶ 10, 79 N.M. 729, 449 P.2d 331 (“The trial court must, when requested, find one way or the other upon a material fact issue, and failure to do so constitutes error.”); see also Whorton v. Mr. C’s, 1984-NMSC-080, ¶ 2, 101 N.M. 651, 687 P.2d 86. This Court held that the prior version of Rule 1-052 NMRA applied to post- conviction proceedings. State v. Hardy, 1967-NMSC-203, ¶ 17, 78 N.M. 374, 431 P.2d 752 (holding that former Rule 52 of the Rules of Civil Procedure, now contained in Rule 1-052 NMRA, “requiring the making of findings of fact,” applied to petitions for post- conviction relief); see also Smith v. Maldonado, 1985-NMSC-115, ¶ 6, 103 N.M. 570, 711 P.2d 15 (holding the same). In a more recent case, where the district court granted habeas relief but entered no findings and conclusions in support of its order, we observed that such findings would have been helpful, noting that “[h]abeas corpus proceedings are civil in nature; it is thus entirely appropriate for the district court to make findings of fact and conclusions of law under Rule 1–052 NMRA . . . .” Lopez v. LeMaster, 2003-NMSC-003, ¶ 29 and ¶ 29 n.2, 133 N.M. 59, 61 P.3d 185 (also observing that no party had requested that the district court enter findings and conclusions).

It is reasonable to apply Rule 1-052’s requirements to habeas proceedings—civil collateral proceedings for relief from unjust detention, for which evidentiary hearings are frequently required. 5 Here, the record encompasses almost forty years of evidence. The district court may have granted relief on all or only one of Defendant’s claims; however, in the absence of findings, we are required to address all the possibilities and scour an extensive evidentiary record. I urge the district courts to comply with the requirements of Rule 1-052, to illuminate for the parties and any reviewing court the basis for their decisions. However, neither party before us prefers a remand to the district court; accordingly, we resolve the State’s appeal in the absence of findings and conclusions.

II. STANDARD OF REVIEW

A. Review of Newly Discovered Evidence Claims in Habeas Appeals

The State asserts (with virtually no basis in the law) that we should, under the circumstances of this habeas appeal, review the factual record de novo. Defendant argues that we must rather ask whether substantial evidence supports the findings necessary to the district court’s judgment. The majority’s approach is unclear. Specifically, the majority suggests that, because the district court failed to enter findings of fact; because it relied primarily upon transcripts of prior witness testimony; and because the new trial standard includes some questions of law, our review of the record for purposes of Defendant’s recantation-based claim is de novo. Maj. op. ¶¶ 40-41.

In my view, this is incorrect. I would agree with Defendant. Asking whether substantial evidence supports the findings necessary to the judgment is consistent with our treatment of, for instance, habeas proceedings where only the trial judge’s comments are available, Lopez, 2003-NMSC-003, ¶ 31 (holding that, in the absence of findings, the appellate court looks to the district court’s remarks and construes them to uphold the judgment), or jury verdicts in civil cases, where (of course) no findings or

5Notably, when the district court grants a defendant’s motion for a new trial in a criminal proceeding, the

district court is required to set forth the reasons for its decision, even though the decision is a discretionary one. State v. Griffin, 1994-NMSC-061, ¶ 11, 117 N.M. 745, 877 P.2d 551 (“The trial court must clearly set forth the grounds for grant of a new trial.”). conclusions are entered, see, e.g., Perschbacher v. Moseley, 1965-NMSC-068, ¶ 9, 75 N.M. 252, 403 P.2d 693 (holding that “[t]he verdict of the jury will not be disturbed unless unsupported by substantial evidence. An appellate court will not reverse, unless, after viewing the evidence in the light most favorable to . . . the verdict, it is convinced that such verdict cannot be sustained either by the evidence or permissible inferences therefrom”). We then “resolve all disputed facts in favor of the successful party, indulge all reasonable inferences in support of a verdict, and disregard all evidence[] and inferences to the contrary.” Clovis Nat’l Bank v. Harmon, 1984-NMSC-119, ¶ 7, 102 N.M. 166, 692 P.2d 1315 (also noting that substantial evidence is “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion” (internal quotation marks and citation omitted)).

Furthermore, fact-finders routinely consider transcripts of prior testimony in evidence—and the State did not apparently object to the admission of such testimony here. When considering a transcript of prior sworn testimony, the fact-finder is not able to gauge, for instance, a witness’s demeanor; however, the judge or jury is able to evaluate the credibility of the testimony by comparison with other evidence, evaluation of internal consistency, and similar considerations. If the majority believes that a district court cannot make credibility findings on such a record, then there was insufficient evidence for the district court’s judgment with respect to Defendant’s recantation-based claims. See Case II, 2008-NMSC-024, ¶ 17 (holding that when seeking a new trial based on newly-discovered evidence, the defendant must prove that the evidence is both credible and significant). But the majority has not so held. In any event, I am unconvinced that we should employ a different standard of review here, where the record encompasses evidence from Defendant’s trial, dozens of transcripts of witness testimony (including testimony from the recanting witnesses) over multiple hearings in the related Case litigation, a video deposition of Defendant, and the live testimony of an alibi witness, Mary Helen Tweedy.

With respect to the majority’s third point—that the new trial standard includes questions of law, and is therefore “a mixed question of fact and law . . . subject to de novo review,” Maj. op. ¶ 40—this seems to me incorrect. While the materiality element of the new trial standard arguably presents a question of law (i.e., assessing the significance of the evidence in light of the crime(s) at issue), the other elements present factual questions, and I see no reason why we would not consider these under a deferential standard of review, as we have in other cases. E.g., State v. Moreland, 2008-NMSC-031, ¶ 21, 144 N.M. 192, 185 P.3d 363 (affirming the trial court’s grant of a new trial under an abuse of discretion standard, holding that “the trial court’s finding that [the d]efendant’s condition was not discoverable prior to trial with the exercise of due diligence was not ‘against the logic and effect of the facts and circumstances of the case’” (citation omitted)). The United States Supreme Court recently held that “mixed questions [of law and fact] are not all alike” and that when so-called mixed questions “immerse courts in case-specific factual issues—compelling them to marshal and weigh evidence, [and] make credibility judgments,” appellate courts “should usually review a decision with deference.” U.S. Bank N.A. ex rel. CWCapital Asset Mgmt. LLC v. Vill. at Lakeridge, LLC, 138 S.Ct. 960, 967 (2018) (holding that the standard of review for a mixed question depends “on whether answering it entails primarily legal or factual work”). Such an approach is consistent with analyzing the fact-based elements of the new trial standard under an abuse of discretion standard, particularly given that the absence of any one element is dispositive (and therefore the “ultimate question” is really an element-by-element inquiry). Case II, 2008-NMSC-024, ¶ 34. In sum, there is no basis to conclude that we ought to conduct de novo review of all the new trial elements; I would employ a deferential standard for those that present questions of fact.

B. Defendant’s Habeas Claim Based on Newly Discovered Recantation Evidence

Even under a deferential approach, our holding in Case II—that Knight’s and Dunlap’s recantations were not newly discovered because they were not qualitatively different from those witnesses’ pretrial statements to police—governs here. 2008- NMSC-024, ¶¶ 35-43. Defendant attempts to distinguish our holding in Case II on grounds that 1) Knight and Dunlap have now testified twice, under oath; 2) Case (one of the witnesses who placed Defendant at the scene) did not testify in Defendant’s trial; and 3) Defendant asserted an alibi supported by witnesses who did not testify in Case’s trial. But these factual distinctions, viewed in the light most favorable to the grant of habeas relief, bear on materiality; that is, the relevance and weight of the recantations in Defendant’s case. They do not alter whether the recantation testimony is newly- discovered under the first element of the significance analysis. Defense counsel effectively conceded this at oral argument. As in Case II, Knight’s and Dunlap’s recantations were qualitatively the same as their pretrial statements, on which they were cross-examined during Defendant’s trial; as such, the recantations were cumulative. Id. ¶¶ 39-40. Thus, there is no substantial evidence for the finding that the recantations are newly-discovered evidence, and the district court, to the extent it granted habeas relief on this basis, abused its discretion. Id. ¶ 34-43 (noting that the failure of the defendant to prevail on any one of the new trial factors is dispositive).

III. BRADY CLAIM

Although I agree with the result of my colleagues’ analysis of Defendant’s Brady claim, I have a different view on the applicable law.

With respect to the three elements of a Brady violation, the majority asserts that the second element—whether the suppressed evidence is “favorable to the accused”— is defined in Bagley as evidence whose “disclosure and effective use ‘may make the difference between conviction and acquittal’ regardless of whether such evidence is impeachment evidence or exculpatory evidence.” Maj. op. ¶ 23 (citing Bagley, 473 U.S. at 676). Then, despite concluding that Autry’s February 3, 1982 statement “does not make a clear difference between conviction and acquittal,” the majority somewhat inexplicably holds that “because it could have affected the jury’s assessment of Autry’s credibility,” the statement is favorable. Maj. op. ¶ 26.

I agree that the statement is favorable, but the quoted provision from Bagley is not defining “favorable” evidence. The excerpt is explaining that “impeachment evidence, . . . as well as exculpatory evidence,” is “favorable to an accused” and “may make the difference between conviction and acquittal.” 473 U.S. at 676. In short, because both exculpatory and impeachment evidence may change the outcome of a trial (i.e., may be material), both are covered by the Brady rule. Id. Some courts have offered a definition of “favorable,” though our courts have not. See, e.g., People v. Clark, 261 P.3d 243, 343 (Cal. 2011) (“Evidence is favorable to the defense if it helps the defense or hurts the prosecution.” (internal quotation marks and citations omitted)); United States v. Trevino, 89 F.3d 187, 189 (4th Cir. 1996) (holding that “favorable” evidence “includes not only that evidence tending to exculpate the accused, but also any evidence adversely affecting the credibility of the government’s witnesses”). In Case II, we simply stated that “[t]he second Brady element is whether the suppressed evidence was favorable to the accused, either as impeachment or exculpatory evidence.” 2008-NMSC-024, ¶ 50. In my view, this suffices.

As to the third Brady element—materiality—the majority includes in its discussion a standard that is no longer good law. The majority quotes Case II, 2008-NMSC-024, ¶ 54, stating that “[m]ateriality only exists if the suppressed evidence ‘creates a reasonable doubt that did not otherwise exist.’” Maj. op. ¶ 27. This statement quotes Agurs, 427 U.S. at 112, a 1976 case in which the United States Supreme Court defined the materiality element of Brady claims according to three case types. The first are cases in which the government has knowingly withheld Brady material; the second are cases in which the defendant has made a specific request for Brady material; and the third are cases in which the defendant has made only a general request for Brady material, or no request at all. 427 U.S. at 103-114; Bagley, 473 U.S. at 678-81 (discussing the three case types). Agurs described materiality for purposes of the first class of cases as “any reasonable likelihood” that the undisclosed evidence “could have affected the judgment of the jury.” 427 U.S. at 103-104. It offered a similar definition for the second class of cases. Id. at 104-106 (implying that evidence is material in this context if it “might have affected the outcome” of trial); Bagley, 473 U.S. at 681 (noting that Agurs did not expressly offer a definition for this class of cases). With respect to the third class of cases, however, Agurs articulated a standard imposing a greater burden of proof on the defendant, defining material evidence as that which “creates a reasonable doubt that did not otherwise exist.” 472 U.S. at 108-113; Bagley, 473 U.S. at 680-681 (acknowledging that this definition imposes a “stricter” standard than the “reasonable possibility” standard applicable to the first class of cases). It is this third, stricter standard that the majority quotes. Maj. op. ¶ 27.

In Bagley, the United States Supreme Court discussed the “reformulation” of Agurs in subsequent cases, and then expressly replaced the Agurs tests with a single test: “evidence is material only if there is a reasonable probability that, had the evidence been disclosed to the defense, the result of the proceeding would have been different.” 673 U.S. at 681-82 (citing and quoting Strickland, 466 U.S. at 694, and deeming the new test “flexible enough” to cover the three types of cases discussed in Agurs). Accordingly, the “reasonable doubt” test quoted by the majority is no longer applicable when evaluating the materiality of alleged Brady evidence. See also, e.g., State v. Lykens, 271 Neb. 240, 250 (Neb. 2006) (holding that the “‘reasonable doubt’ standard set forth” in Agurs “was superseded” by Bagley, and disapproving a subsequent case “to the extent it [applied] the ‘reasonable doubt’ standard of Agurs rather than the Bagley standard . . . to analyze whether evidence was wrongfully withheld by the prosecution”); United States v. Shaffer, 789 F.2d 682, 688 (9th Cir. 1986) (noting that the district court had applied Agurs, and had “held that under both the ‘reasonable possibility’ test for specific requests, and the more stringent ‘reasonable doubt’ test for general requests, a new trial was required,” but that Bagley, which had been decided during the pendency of appeal, created a “new standard,” and applying Bagley instead); State v. Spurlock, 874 S.W.2d 602, 619 (Tenn. Crim. App. 1993) (holding that Bagley “discarded” the Agurs tests and created a single standard). My colleagues, in effect, apply the Bagley standard (Maj. op. ¶ 35), and therefore I concur in the result on this aspect of Defendant’s claim. But I submit that we should no longer include the Agurs “reasonable doubt” standard in our discussions of whether a claimed Brady violation is material.

I also wish to acknowledge Defendant’s arguments that, in spite of our conclusion in Case II that Autry’s suppressed February 3, 1982 statement would not have been admissible as exculpatory evidence, Defendant’s case is factually distinguishable. Specifically, Defendant contends that sightings of Mitchell with Autry in the days and weeks prior to the murder, and testimony that Mitchell was afraid of Autry “bothering” her, were sufficient evidence of Autry’s motive to commit the crimes against Mitchell, such that the suppressed statement would have been admissible at trial. I disagree. As we discussed in Case II, direct or circumstantial evidence linking a third person to the crimes charged cannot be merely speculative. Id. ¶ 55. For instance, in State v. Rosales, 2004-NMSC-022, ¶ 13, 136 N.M. 25, 94 P.3d 768, which we discussed in Case II, 2008-NMSC-024, ¶ 55, a suppressed statement would have been admissible as exculpatory evidence where the statement, which occurred only a couple of weeks before the murder, indicated the third person’s motive to commit murder, and “[a]t trial, other evidence showed that [the third person] had recently threatened the life of the victim, the victim was killed in [the third person’s] vehicle, and the murder weapon might have been a pocket knife owned by [the third person].” The proffered distinguishing facts here do not demonstrate comparable “links” between Autry and the crimes charged. Sightings of Mitchell with Autry, and Mitchell’s fear of being “bothered” by Autry, while evidence of a relationship between the two, are not—without more— evidence linking Autry to Mitchell’s rape and murder. Accordingly, I agree with the majority’s conclusion that the February 3, 1982 statement would not have been admissible as exculpatory evidence, and was therefore not material to the defense.

IV. ACTUAL INNOCENCE CLAIM

I part ways from my colleagues on the framework necessary for our review of Defendant’s actual innocence claim. The majority misstates an important aspect of the actual innocence standard, Maj. op. ¶ 36 (“Evidence used in a habeas proceeding to support a claim for actual innocence must be new evidence,” citing Montoya, 2007- NMSC-035, ¶ 30), and then applies that standard only to conclude its discussion of Defendant’s claims for a new trial. Maj. op. ¶ 57. Actual innocence is a distinct basis for habeas relief—one not limited by the new trial requirements. We held in Montoya that a habeas petitioner “asserting a freestanding claim of innocence must convince the court by clear and convincing evidence that no reasonable juror would have convicted him in light of the new evidence.” 2007-NMSC-035, ¶ 30. However, we then stated as follows:

[T]he fact that the evidence presented by Petitioner does not fit within our rubric for newly discovered evidence does not mean that Petitioner’s claim will necessarily fail. When examining a freestanding claim of actual innocence, we will not be constrained by the requirements applicable to motions for a new trial. Instead, we examine the evidence presented and evaluate any reliable evidence. See Cole, 765 N.Y.S.2d at 481, 485 (examining petitioner’s claim of actual innocence despite the fact that he could not meet the criteria for a new trial on the grounds of newly discovered evidence because he could not establish that the evidence could not have been discovered prior to trial with due diligence). This is because the focus of our inquiry is on actual innocence rather than when the evidence could have been discovered or procedural error. We conclude, however, that the factors which determine whether evidence is newly discovered under our motion for a new trial standard remain relevant as we review whether or not the evidence presented by Petitioner is reliable.

2007-NMSC-035, ¶ 32 (emphasis added). In other words, even if the proffered evidence fails to meet all the criteria for a new trial based on newly discovered evidence, we still consider it together with other reliable evidence to assess whether any reasonable juror would have convicted the defendant. Id. ¶¶ 30-32.

Here, even viewing the evidence in Defendant’s favor, he did not make the requisite showing by clear and convincing evidence, and therefore the district court abused its discretion to the extent it granted the habeas petition on the basis of Defendant’s actual innocence. This is because, contrary to Defendant’s contention, the proffered newly-discovered evidence fails to meet the new trial criteria not merely because of the “timing” of the evidence, but because it is largely cumulative, and—in the case of the DNA evidence—immaterial. As discussed in Case II, the recantations fail to meet the new trial factors because they are cumulative. See 2008-NMSC-024, ¶¶ 35- 41. Autry’s statement was also largely cumulative of his admission that he had made unsuccessful advances toward Mitchell, and cumulative of other evidence challenging Autry’s credibility. Maj. op. ¶¶ 30, 32. The absence of male DNA on Mitchell’s body and clothing, while not supportive of the State’s case, does not exonerate Defendant. In Defendant’s original trial, forensic tissue examination showed the presence of no sperm cells on or in Mitchell’s body. However, as the majority rightly points out, forcible penetration does not necessarily entail ejaculation, nor penetration with the penis; accordingly, the absence of male DNA or sperm cells does not rule out forcible penetration. Maj. op. ¶ 56. Relatedly, although the later DNA test was more sophisticated, Defendant’s expert acknowledged that Mitchell’s body had decomposed for a month before it was discovered, and therefore the finding of no sperm cells did “not necessarily” mean that there were never sperm cells on or in Mitchell’s body. Thus, the DNA evidence is only further inconclusive proof of the absence of evidence.

With respect to other reliable evidence in the record, although the reliability of Autry’s testimony is shaky, I cannot say that a jury would be compelled to discredit his testimony that Defendant was present just prior to Mitchell’s sexual assault, and that Autry observed Defendant strike Mitchell, and rip her pants off. Testimony from witness Davis indicated that Defendant had admitted he wanted to get “something” in return for helping Mitchell; later, Defendant stated that he “took [Mitchell] out and fucked her.” Testimony from witness Hood undermined Defendant’s proffered alibi for the evening in question. This evidence, the evidence of Mitchell’s physical injuries (indicative of assault), and her open pants, with a torn zipper (consistent with testimony that Defendant ripped off her pants) leave me unable to say that “no reasonable juror” could have convicted Defendant in light of all the evidence.

V. CONCLUSION

We are better able to render just decisions when the basis for a district court’s judgment, here derived from a forty-year record, is illuminated. Had the district court rendered findings of fact and conclusions of law on Defendant’s habeas petition, these would have informed our review and substantially aided our resolution of his appeal. Relatedly, I emphasize that, with respect to recantation evidence, district courts should make credibility determinations before addressing the significance of the proffered evidence. Case II, 2008-NMSC-024, ¶ 34 (“[W]e believe it would inform our review if trial courts, rather than presuming the credibility of the recantations, would make findings with respect to each factor relating to the credibility prong before addressing the significance prong.”). Nevertheless, for all the reasons set forth hereinabove, I join my colleagues in concluding that the district court’s grant of habeas relief must be reversed.

JUDITH K. NAKAMURA, Justice