Blufford Hayes, Jr. v. Jill Brown, Warden of the California State Prison at San Quentin

TALLMAN, Circuit Judge,

with whom KLEINFELD, GOULD, and BYBEE, Circuit Judges, join, concurring in part and dissenting in part:

I concur in Parts I, II-A, II-B, and III— A of Judge Thomas’s opinion for the court. *989 There are few things more repugnant to the fundamental notions of fairness embodied in due process than a prosecutor allowing false evidence to go uncorrected when it appears. Napue v. Illinois, 360 U.S. 264, 269 , 79 S.Ct. 1173 , 3 L.Ed.2d 1217 (1959). In their solemn constitutional obligation “as[] representative^] of the government to protect the integrity of the court and the criminal justice system,” Northern Mariana Islands v. Bowie, 243 F.3d 1109, 1122 (9th Cir.2001), prosecutors have a “special duty commensurate with [their] unique power, to assure that defendants receive fair trials.” United States v. LaPage, 231 F.3d 488, 492 (9th Cir.2000). The majority appropriately rejects the State’s argument that Napue and its progeny prohibit only perjury. The prosecutor’s failure to correct the misimpression left on the jury and the court as to the full scope of the deal offered to secure the testimony of Andrew James, even though the witness himself was ignorant of all benefits he would realize from his cooperation, is as objectionable to due process as perjury.

However, in considering all of the evidence presented to this jury in determining Hayes’s guilt, I respectfully disagree with the court’s conclusion that there was a reasonable likelihood that the information the prosecutor withheld from James and the court could have affected the jury’s verdict. In concluding otherwise, the majority misapplies the test for materiality and ignores the substantial evidence introduced at trial otherwise impeaching James’s credibility. I fear that by reducing the threshold of what is reasonably likely to affect a juror’s judgment the majority has effectively endorsed a per se reversal rule. Therefore, I dissent from Part III-B. 1

I.

Despite the majority’s unqualified claim that “if it is established that the government knowingly permitted the introduction of false testimony reversal is ‘virtually automatic,’ ” Maj. Op. at 978 (citing United States v. Wallach, 935 F.2d 445, 456 (2d Cir.1991)), the Supreme Court has expressly directed otherwise:

We do not [] automatically require a new trial 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[.]” ... A new trial is required if “the false testimony could ... in any reasonable likelihood have affected the judgment of the jury[.]”

Giglio v. United States, 405 U.S. 150, 154 , 92 S.Ct. 763 , 31 L.Ed.2d 104 (1972) (citing Napue, 360 U.S. at 271 , 79 S.Ct. 1173 ). Thus, it is not enough to simply find constitutional error. Instead, contrary to the evidence before us in this record, even where testimony is demonstrably false, we are required to determine the materiality of the false evidence at issue by considering all of the evidence and determining whether there was a reasonable likelihood that the false testimony could have affected the jury’s judgment. Giglio, 405 U.S. at 154 , 92 S.Ct. 763 ; LaPage, 231 F.3d at 491 .

I disagree with the majority’s characterization of the materiality standard as an inquiry requiring the court to consider, not whether the verdict would have been dif *990 ferent, but whether the trial was fair and the verdict worthy of confidence. The majority’s citation to Hall v. Director of Corrections, 343 F.3d 976, 983-84 (9th Cir.2003) (Tallman, J. dissenting) is unavailing. In explaining the materiality standard in a case involving false evidence, the Hall majority improperly plucked the materiality standard from Kyles v. Whitley, 514 U.S. 419, 434 , 115 S.Ct. 1555 , 131 L.Ed.2d 490 (1995), an appeal involving a prosecutor’s alleged failure to disclose potentially exculpatory evidence. The standards, though similar, are not the same. See, e.g,, United States v. Agurs, 427 U.S. 97, 103 , 96 S.Ct. 2392 , 49 L.Ed.2d 342 (1976) (distinguishing the tests for materiality in cases alleging constitutional violation for the use of false evidence and cases involving the failure to disclose exculpatory evidence).

The proper standard, and the one set forth by the Supreme Court, requires us to determine whether, in the context of all the evidence, there was a reasonable likelihood that the false testimony could have affected the jury’s judgment. Giglio, 405 U.S. at 154 , 92 S.Ct. 763 ; see also LaPage, 231 F.3d at 491 . Employing that standard, we must use caution in trying to peer inside the minds of jurors who .had the distinct benefit of hearing all the testimony and seeing the evidence first-hand.

II.

To label the testimony of James “false” is a misnomer on these facts. Because James was never told of the secret side deal, James did not testify falsely when he declared that his prior felony charges were still pending and he truthfully testified that he was, unaware of any deals or promises made in exchange for his testimony. What offends due process is not James’s testimony, but the prosecution’s failure to correct that which the prosecutor knew to be false: James’s unwittingly untrue statement that he had no deal regarding his pending felony charges. Napue, 360 U.S. at 269 , 79 S.Ct. 1173 . The majority reasons that the prosecutor’s failure to correct James’s testimony kept from the jury impeachment evidence that it might have considered in weighing James’s credibility. But, there is no link here between the prosecutor’s misconduct and its effect on what James knew at the time he testified about the murder and theft by Hayes. To claim, as the majority does, that there is a reasonable likelihood that James’s testimony, in which he truthfully swore he was unaware of any deal, affected the jury’s judgment also ignores the substantial amount of impeachment evidence adeptly employed by the defense at trial to impugn James’s credibility. It also ignores the fact that the witness himself was unaware of a benefit the law cautions us to suspect might lead a potentially biased witness to testify untruthfully.

Critically, James testified that the State had given him immunity from all crimes related to Patel’s murder — a fact the majority completely discounts. There is no question the jury heard that James was testifying as part of a deal with the prosecution. That the deal also included future dismissal of different conduct is immaterial. The prosecution’s undisclosed agreement to provide James additional immunity is, at best, merely cumulative of evidence suggesting that James already had ample motive to color his testimony against Hayes. Cf. United States v. Cooper, 173 F.3d 1192, 1203 (9th Cir.1999) (additional item of impeachment evidence could not have damaged witness much more and there was ample evidence of defendant’s guilt without the witness’s testimony); United States v. Marashi, 913 F.2d 724, 732-33 (9th Cir.1990) (notes containing cumulative impeachment evidence were not material).

*991 Evidence that James had any kind of deal is relevant only to demonstrate that he had a motivation for testifying adversely against Hayes. However, any motivation James may have had to testify differently for the prosecution because the pending felony charges against him would later be dismissed was nullified because he did not know of that term of the deal. The testimony James offered simply could not have been influenced by a deal he knew nothing about. Thus, the wafer-thin likelihood that the jury would have been affected by James’s uncorrected statement vanishes altogether. Instead, the majority punishes the State for the due process violation without any regard to its lack of effect on the credibility of James’s actual testimony.

The majority reasons that “the fact that the .jury was apprised of other grounds for believing that the witness [ ] may have had an interest in testifying against petitioner [does not turn] what was otherwise a tainted trial into a fair one.” Napue, 360 U.S. at 270 , 79 S.Ct. 1173 . This, however, is not a case of “other grounds.” It is simply invalidating an otherwise credible verdict based on prosecutorial misconduct alone. The jury knew that James had accepted a generous deal from prosecutors giving him transactional immunity for the Patel murder and burglary in return for his testimony. James’s deal for immunity from prosecution for his other pending felonies is not “other grounds,” but the very same grounds for questioning his motive to color the truth when he agreed to testify against Hayes: a substantial prosecutorial concession to secure James’s testimony. Not only is this duplicative of grounds already used to call James’s credibility into question, and thus not “other grounds” within the meaning of Napue , but it added nothing to the jury’s assessment of James’s credibility since he himself was unaware of it.

James’s testimony, moreover, was far from pristine for reasons unrelated to further prosecutorial immunity. Specifically, James testified that he had received money, airline tickets, and other aid from the State in exchange for his testimony. He also admitted that he was still facing unresolved criminal charges that had been pending against him for the previous two years, and that he had also been convicted of petty theft, grand theft, and receiving stolen property. Yet, with all this evidence unquestionably undermining James’s credibility, the majority insists that had jurors heard that James also would later receive a dismissal of the pending felony charges — in addition to his truthful testimony that he received immunity for all crimes related to Patel’s murder — there is a reasonable likelihood that their judgment could have been affected. I cannot agree with the court’s conclusion that this redundant evidence of his motive to testify against Hayes was the straw which broke an already crippled camel’s back. 2

The majority’s conclusion characterizes James as a “key witness,” without whom Hayes could not have been convicted. That conclusion, however, ignores Michelle Gebert’s testimony, corroborating James’s story, that Hayes told her that he had robbed Patel, and that Hayes assured her *992 that Patel would not say anything to the police. It also ignores identical modus operandi evidence from James Cross, who testified that Hayes had previously beaten him, demanded money, and tied his hands and feet with coat hangers — -just as Hayes had done to Patel. It ignores the fact that after binding Cross, Hayes took additional money from Cross’s pockets and then sent his girlfriend to Cross’s room to look for more money.

The court also departs from and misapplies the materiality inquiry required by Giglio . The court’s analysis goes critically astray by considering the effect on the jury, not of James’s actual testimony, but of excluding his entire testimony. Instead of considering what effect James’s truthful testimony could have had on the jury, the majority concludes that without James’s testimony and credibility, it is reasonably likely that the remaining circumstantial evidence, resulting in a weaker inference of burglary-murder, could have affected the jury’s judgment. See, e.g., Maj. Op. at 987 (“Without the testimony of James, an entirely different trial would have occurred.”); see id. at 986 (“Without[James’s testimony and credibility], there was only circumstantial evidence of the burglary[.]”).

That is simply not the test for materiality. Giglio, 405 U.S. at 154 , 92 S.Ct. 763 (“A new trial is required if ‘the false testimony could ... in any reasonable likelihood have affected the judgment of the jury[.]’ ”) (quoting Napue, 360 U.S. at 271 , 79 S.Ct. 1173 ). We must only consider the marginal effect that the testimony would have had on the jury’s verdict, if given fully and completely; not, as the majority proposes, consider the effect that exclusion of James’s entire testimony would have had on the jury. Id.; see also LaPage, 231 F.3d at 491 . No rational juror, considering the baggage James carried to the witness stand, would have based his or her verdict entirely on what James said. There was ample other evidence to permit the jury to convict Hayes of the heinous crime for which he faces death. Considered in total, the case against Hayes is far stronger than the majority is willing to admit.

Curiously, the court also postures its analysis as if James’s testimony would have suddenly become completely incredible if he had testified that he also had a leniency deal for his pending felony charges. Even without revelation of the secret deal, James’s credibility had already been called into substantial question in several ways. Yet, we cannot now say whether jurors disregarded James’s testimony and relied on Gebert’s testimony that Hayes told her he had robbed Patel, or whether they believed James’s story because it was corroborated by Gebert, the autopsy, and the physical evidence at the crime scene. An appellate judge’s place is not in the jury box, post hoc. We must only consider whether it is reasonably likely that this marginal addition to the evidence could have affected the judgment of the jury. I am satisfied on this record it could not.

Though the prosecutor violated Hayes’s due process rights by failing to correct the misimpression left by the unwitting, but truthful testimony of Andrew James, because the majority ignores all of the evidence considered by the jury and misapplies the test for materiality, I respectfully dissent from the court’s conclusion that Hayes is automatically entitled to a new trial for a crime he committed more than 20 years ago.

. I express no opinion as to the majority’s opinion in Part II-C. Because I would hold that there was no reasonable likelihood that knowledge of a more favorable deal for James could have affected the jury’s judgment, there is no need to consider whether this is a new rule of constitutional procedure that cannot be applied under Teague v. Lane, 489 U.S. 288 , 109 S.Ct. 1060 , 103 L.Ed.2d 334 (1989).

. The majority’s assertion that James’s credibility was measured against the credibility of his peers is unsupported by any of the evidence. Additionally, there is absolutely no evidence in the record that James's transactional immunity was valueless in impugning his credibility. In making these inferential leaps that purport to delve into the minds of jurors sitting in the courtroom over twenty years ago, the majority does here that which it should not — substitute .its own judgments for that of the jury by making express but unsubstantiated assumptions about the jury's consideration of the evidence.