La Carl Martez DOW, Petitioner-Appellant, v. Tim VIRGA, Warden, Respondent-Appellee, 729 F.3d 1041

OPINION REINHARDT, Circuit Judge: La Carl Martez Dow’s state court trial for second degree robbery involved textbook prosecutorial misconduct, recognized as such by the California Court of Appeal (“state court”). In the course of the trial, the prosecutor knowingly elicited and then failed to correct false testimony. That testimony, by Detective Oglesby, was that Dow (rather than his attorney) made the request that each of the participants in a lineup wear a bandage under his right eye at the location at which Dow had a small scar under his.

Then, based on this evidence, the prosecutor told the jury during closing argument that Dow had demonstrated consciousness of guilt by trying to hide his scar in order to prevent the sole eyewitness from identifying him. Dow contended that the prosecutor’s eliciting of false testimony and failure to correct it violated his federal constitutional rights. The state appellate court held that misconduct had occurred. It stated: We find that misconduct occurred.

Although Detective Oglesby testified that defendant made the request to have “the band-aid placed beneath all the participants’ right eyes,” the prosecutor was aware that representation of the evidence was erroneous. More importantly, by asserting that defendant was attempting to “hide” his scar, the prosecutor was mischaracterizing the evidence. The prosecutor’s misconduct violates the basic tenet of Napue v. Illinois, which prohibits “soliciting false evidence,” and requires the prosecutor to not “allow[ ] it to go uncorrected when it appears.” 360 U.S. 264, 269, 79 S.Ct. 1173, 3 L.Ed.2d 1217 (1959).

Nonetheless, the state court upheld Dow’s conviction because it found that it was not reasonably likely that, absent the misconduct, Dow would have obtained a more favorable verdict; in other words, it held that the error was harmless. Dow’s claim comes to us on a petition for habeas corpus. Applying the deferential standards of the Antiterrorism and Effective Death Penalty Act of 1996 (“AEDPA”), we conclude that the state court’s decision upholding Dow’s conviction was “contrary to” and/or an “unreasonable application” of Supreme Court precedent.

Specifically, in rejecting Dow’s claim, the state court applied a harmlessness standard that is “contrary to” the harmlessness standard required by Napue v. Illinois. The Napue standard is different from the ordinary harmlessness standard, and is referred to in Na-pue and its progeny as a “materiality” standard. We so refer to it here. Even were we to presume, as the state urges, that the state court applied the Napue materiality standard when rejecting Dow’s claim, its application of that standard would have constituted an “unreasonable application” of clearly established Supreme Court law.

As the Napue error is without doubt “material,” we reverse the district court’s denial of Dow’s petition and remand with instructions to grant the writ of habeas corpus. FACTS & PROCEDURAL BACKGROUND I. The Robbery and Investigation Because our “review is limited to the record that was before the state court that adjudicated the claim on the merits,” Cullen v. Pinholster, — U.S. -, 131 S.Ct. 1388, 1398, 179 L.Ed.2d 557 (2011), we do not look to facts beyond the state court record. We therefore rely upon the California Court of Appeal’s statement of facts in People v. Dow, No. A105381, 2004 WL 2367997 (Cal.Ct.App.

Oct. 22, 2004) (unpublished). According to the state court: One night in November 2002, Felix Sa-blad was working as a cashier in the convenience store at the Olympian Gas Station in Daly City, California. An individual, later identified by Sablad as Dow, entered the store, approached the cash register, and asked for a certain type of medication that Sablad explained the store did not carry. After walking around the store as if he were looking for something, the perpetrator returned to the register, placed a package of gum on the counter, and briefly engaged Sa-blad in idle conversation.

As Sablad opened the cash register, the perpetrator produced a very small silver handgun and ordered him to hand over the money. Sablad was frightened, but placed the money on the counter with his left hand as he attempted to press the “panic button” at the bottom of the cash register with his right hand. The perpetrator yelled, “what the fuck are you doing, man,” and aimed the gun at Sablad’s head. Sablad turned away from the perpetrator and indicated that he should take the money, less than $300, from the register. The perpetrator did so and ran outside.

Sablad pushed the panic buttons on both cash registers and yelled to the assistant manager, “call the police. We were robbed.” He then went outside the store and observed the perpetrator driving away.

The police arrived shortly thereafter and took a description of the suspect: “an African-American male, dark-skinned, medium build approximately five foot ten inches tall to six feet tall,” 25 to 35 years old, with a “scar somewhere on his face.” According to Sablad, the man was “wearing a plain gray sweatshirt and sweat pants and a ball cap.” Months later, Sa-blad told the investigating officers that “he just remembered that the robber was missing a tooth.” The police used the security video recording to produce still images of the perpetrator, but these images were not clear enough to show the robbery or to permit the identification of the perpetrator.

Other than two partial fingerprints that could not be used for identification, no physical evidence was retrieved from the scene of the crime. Dow became a suspect in the Olympian Gas Station robbery as a result of his alleged involvement in another robbery. After executing a search warrant at Dow’s home, the detective assigned to the case— Detective Oglesby—seized a generic gray sweatsuit that resembled the one worn by the perpetrator in the security video. Over two months after the robbery, Ogles-by prepared a photo lineup for Sablad to view that included Dow’s photograph among eight others.

Sablad selected Dow’s photograph as one that “resembles” the robber, although he mentioned, “I can’t see the scar on the photo.” Several months later, Sablad was shown a live lineup at the district attorney’s office using five individuals, including Dow. Dow’s lawyer, who was present at the lineup, expressed concern that Sablad might falsely identify Dow because Dow was the only lineup participant with a facial scar. As a result, counsel asked the district attorney’s office to ensure that each individual in the lineup wore a bandage to cover the area under his right eye, the area in which Dow had a small scar.

Sablad again identified Dow at the live lineup. He also acknowledged that Dow was the only person to appear in both the photo and live lineups. II. State Court Trials Dow had two jury trials in California state court. Dow’s first trial ended in a deadlocked jury, requiring the judge to declare a mistrial. At Dow’s second jury trial, however, he was convicted of second degree robbery. The judge then sentenced him to fifteen years of imprisonment. His claim of constitutional error arises from the prosecutor’s conduct at the second trial. A.

The Evidence The prosecution’s case against Dow turned on the reliability of Sablad’s identification of Dow as the perpetrator. Although Sablad testified that he was confident of his identification, he made a number of important, inconsistent statements over the course of the investigation and prosecution regarding his memory of the perpetrator’s appearance, including the location of a scar on the perpetrator’s face and the existence and location of a missing tooth.

Shortly after the robbery, Sablad told the officer that he was not sure where the scar was located on the perpetrator’s face; at the preliminary hearing, he said the scar was on the right jaw line; then, at the first trial, he testified that the scar was on the right side of the face adjacent to the eyebrow and going downward; finally, at the second trial, Sablad stated, looking at the defendant, that he could not see the scar he remembered on the perpetrator’s face. As to the missing tooth, Sablad remembered this fact only months after he gave his initial description of the perpetrator to the police.

Then, he gave inconsistent testimony as to which tooth was missing: at the first trial, Sablad testified that the perpetrator was missing a lower right tooth; then, at the second trial, he testified that the missing tooth might have been on the top. To corroborate Sablad’s identification, Detective Oglesby, who set up the photo and live lineups, testified that Dow has a small scar below his right eye that is visible from “a foot or two away,” but not from “a distance,” and that he has “a gap” between his two front teeth. Detective Oglesby did not mention a missing tooth, either upper or lower.

The defense sought to undermine Sa-blad’s identification by using expert testimony. Dr. Robert Shomer testified regarding the (un)reliability of eyewitness identification generally. He also testified that eyewitness testimony can often be influenced by numerous factors, including the amount of stress on the eyewitness at the time of the incident, the race of the perpetrator, whether the perpetrator used a weapon, and the lineup procedures used by the police. Shomer testified that it was his opinion that the lineup procedures used in Dow’s case were not reliable and were unduly suggestive. B.

The Constitutional Error Two relevant events form the basis of Dow’s claim that his second trial was constitutionally infirm. First, during the prosecutor’s direct examination of Detective Oglesby, she asked: “At whose request was the band-aid placed beneath all of the participants’ right eyes?” Oglesby responded, “Mr. Dow’s.” This testimony was false. In addition, as the state appellate court stated, the prosecutor knew at the time that this “representation was erroneous.” Nevertheless, the prosecutor did not correct the detective’s testimony.

Second, the prosecutor exploited her knowing presentation of false evidence by arguing that Dow had requested the placement of the band-aids in order to hide his scar, thus indicating consciousness of guilt. Defense counsel objected to this line of argument, but was overruled. The exchange during the prosecutor’s closing remarks in rebuttal was as follows: [PROSECUTOR]:.... But who knows the defendant’s face better than anyone else in this courtroom? The defendant. If there is no noticeable scar on his face, why did he demand that— [DEFENSE COUNSEL]: Excuse me, I’m sorry. Objection.

There’s no evidence as to whose initiative it was that band-aids were placed on the faces. THE COURT: Overruled. There was such evidence. [THE PROSECUTOR]: If he does not have a scar on his face, why did, as Detective Oglesby testified, why did he ask that a band-aid be placed under his right eye about the exact same location where you can see, in the pictures that Detective Cisneros took, what looks to be a scar. What was he trying to hide if there’s nothing. [THE PROSECUTOR]: ... The defendant’s actions speak for themselves.

He knows what’s on his face, and he knows what he was trying to hide when he had that band-aid placed under his right eye. [DEFENSE COUNSEL]: Objection. I need to phrase an objection. This is outside the scope of evidence. The implication that is being made— THE COURT: It simply is not outside the scope of the evidence. Of that I am sure.

But once again, the jury, if there’s any question in the jury’s mind about what the testimony was on that point, you can have it re-read. [DEFENSE COUNSEL]: I also object to the implication that this somehow represents a consciousness of guilt when, in fact, the very instructions that are given with respect to how to conduct these things say you should try to obstruct a scar. THE COURT: The objection is overruled. Go ahead. This is just argument based on the evidence. [THE PROSECUTOR]: Thank you, Your Honor.... The pictures at the live lineup, sure, they show the mouths of these people closed.

And you don’t know for sure what happened before or what happened after. But I suggest this to you. A person who is careful enough to have concealed a distinguishing mark on his face, do you think that that person would be showing a victim, a potential witness in this case, a prominent gap between his two front teeth. III. State Appellate Court Decision and Habeas Proceedings On October 22, 2004, the California Court of Appeal affirmed the trial court judgment in an unpublished opinion.

Dow had argued to the state court that the prosecutorial misconduct in this case violated his federal constitutional rights because “[w]here a prosecutor deceived a jury regarding the true nature of evidence, reversal ‘is required if the evidence could in any reasonable likelihood have affected the judgment of the jury,’ ” citing Giglio v. United States, 405 U.S. 150, 154, 92 S.Ct. 763, 31 L.Ed.2d 104 (1972), which in turn cites Napue v. Illinois, 360 U.S. 264, 271, 79 S.Ct. 1173, 3 L.Ed.2d 1217 (1959).

The state court found that “misconduct occurred” because “the prosecutor was aware that [Detective Oglesby’s] representation of the evidence was erroneous.” Nonetheless, the state court affirmed Dow’s conviction. It concluded that Dow had not shown that it was “ ‘reasonably probable that a result more favorable to the defendant would have occurred’ absent the misconduct,” citing People v. Welch, 20 Cal.4th 701, 753, 85 Cal.Rptr.2d 203, 976 P.2d 754 (1999). The standard that the state court applied is the state law standard for reviewing the harmlessness of non-constitutional errors.

The state court reasoned that the misconduct was harmless because defense counsel was able to object and present a contrary view from which the jury could have understood that “the reason for the bandages was to facilitate a fair lineup,” “Sablad’s identification testimony was strong” and corroborated by the gray sweatsuit, and “[t]he presence of a scar on the defendant’s face was only a minor aspect of the identification process, and was in fact not seen by the witness either in the lineups or at the trial.” As the state court concluded, “a more favorable verdict to defendant was not reasonably probable without the misconduct.” DISCUSSION Dow’s petition for habeas is governed by the strict standards of AEDPA.

Lindh v. Murphy, 521 U.S. 320, 326-27, 117 S.Ct. 2059, 138 L.Ed.2d 481 (1997). Under AEDPA, we ask whether the state court’s adjudication of the claim “resulted in a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States.” 28 U.S.C. § 2254(d)(1). A state court decision is “contrary to” clearly established Supreme Court precedent “if it ‘applies a rule that contradicts the governing law set forth in [Supreme Court] cases.’ ” Early v.

Packer, 537 U.S. 3, 8, 123 S.Ct. 362, 154 L.Ed.2d 263 (2002) (quoting Williams v. Taylor, 529 U.S. 362, 405-06, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000)); see also Frantz v. Hazey, 533 F.3d 724, 734 (9th Cir.2008) (en banc). A state court decision constitutes an “unreasonable application of’ Supreme Court precedent if it is “objectively unreasonable,” not merely if it applies that precedent “erroneously or incorrectly.” Williams, 529 U.S. at 409, 411, 120 S.Ct. 1495; Bell v. Cone, 535 U.S. 685, 694, 122 S.Ct. 1843, 152 L.Ed.2d 914 (2002). I.

The clearly established Supreme Court precedent, at the time of Dow’s state court decision, was that a Napue violation—a presentation to a fact-finder of false testimony knowing it to be false— results in the reversal of a conviction if “the false testimony could ... in any reasonable likelihood have affected the judgment of the jury....” Giglio v. United States, 405 U.S. 150, 153, 154, 92 S.Ct. 763, 31 L.Ed.2d 104 (1972) (quoting Napue v. Illinois, 360 U.S. 264, 271, 79 S.Ct. 1173, 3 L.Ed.2d 1217 (1959)).

In Napue, the prosecutor elicited and did not correct what he knew to be false testimony'—that the state’s principal witness had not been promised any consideration by the State in exchange for his testimony. 360 U.S. at 265, 267, 79 S.Ct. 1173. The Court explained that the principle that a prosecutor, working on behalf of the state, may not knowingly use false testimony to obtain a conviction is “implicit in any concept of ordered liberty.” Id. at 269, 79 S.Ct. 1173.

The Court held that “a conviction obtained through use of false evidence, known to be such by representatives of the State” violates the Fourteenth Amendment. Id. at 269, 79 S.Ct. 1173. The Court reversed Napue’s conviction on the ground that the false testimony “may have had an effect on the outcome of the trial.” Id. at 272, 79 S.Ct. 1173. As explained in subsequent opinions applying the Napue standard, “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, 153, 154, 92 S.Ct. 763, 31 L.Ed.2d 104 (1972) (quoting Na-pue, 360 U.S. at 271, 79 S.Ct. 1173); see also Sivak v. Hardison, 658 F.3d 898, 912 (9th Cir.2011); Libberton v. Ryan, 583 F.3d 1147, 1164 (9th Cir.2009); Jackson v. Brown, 513 F.3d 1057, 1076 (9th Cir.2008). Although the government’s knowing use of false testimony does not automatically require reversal, courts apply a less demanding materiality standard to Napue errors: whether “there is any reasonable likelihood that the false testimony could have affected the judgment of the jury.” United States v.

Agurs, 427 U.S. 97, 103, 96 S.Ct. 2392, 49 L.Ed.2d 342 (1976) (emphasis added). This materiality standard is, in effect, a form of harmless error review, but a far lesser showing of harm is required under Napue’s materiality standard than under ordinary harmless error review. See Smith v. Phillips, 455 U.S. 209, 220 n. 10, 102 S.Ct. 940, 71 L.Ed.2d 78 (1982) (describing the “materiality requirement” that applies to Napue and Giglio claims); see also Hayes v. Brown, 399 F.3d 972, 984 (9th Cir.2005) (en banc).

Napue requires us to determine only whether the error could have affected the judgment of the jury, whereas ordinary harmless error review requires us to determine whether the error would have done so. In short, prosecutorial misconduct of the kind that occurred here violates the constitutional rights of the defendant and requires a reversal of the conviction if (1) the testimony was actually false, (2) the prosecutor knew it was false, and (3) the false testimony was material (i.e., there is a reasonable likelihood that the false testimony could have affected the judgment).

See Napue, 360 U.S. at 271-72, 79 S.Ct. 1173. II. The state court found that the first two requirements of a Napue violation were met. It found that “the representation of the evidence [that Dow requested the bandages] was erroneous” and that “the prosecutor was aware” of this false representation. The state raises no objection to these findings.

It acknowledges that “the state appellate court found the first two prongs of the federal test satisfied,” and that “the court of appeal found the prosecutor had committed misconduct violative of Napue.” (emphasis added) Once the first two requirements of Na-pue are met, the court must determine whether the error is material, that is, whether “there is any reasonable likelihood that the false testimony could have affected the judgment of the jury.” Agurs, 427 U.S. at 103, 96 S.Ct. 2392.

It is clear from the state court’s opinion, however, that it applied a state law standard for harmless error review that is more difficult for the defendant to meet than the standard prescribed by the Supreme Court. The state court considered “whether it is ‘reasonably probable that a result more favorable to the defendant would have occurred’ absent the misconduct,” People v. Dow, 2004 WL 2367997, at *8 (Cal.Ct.App. Oct. 22, 2004) (citing People v. Welch, 20 Cal.4th 701, 753, 85 Cal.Rptr.2d 203, 976 P.2d 754 (1999)) (emphasis added).

Under that stricter standard, the court found that “a more favorable verdict to [Dow] was not reasonably probable without the misconduct.” On this basis, the state court denied relief. The state court applied a harmlessness standard that is “contrary to” clearly established Supreme Court precedent because it should have applied the materiality standard required for cases involving Napue errors, i.e., it should have determined whether “there [was] any reasonable likelihood that the false testimony could have affected the judgment of the jury.” Agurs, 427 U.S. at 103, 96 S.Ct. 2392.

Application of the wrong standard constituted an error of law that was contrary to clearly established Supreme Court precedent. See Napue, 360 U.S. at 271-72, 79 S.Ct. 1173; see also Caliendo v. Warden, 365 F.3d 691, 698 (9th Cir.2004) (holding that “AEDPA’s presumption of correctness does not apply to state court findings arrived at through the use of erroneous legal standards”); see also Bains, 204 F.3d at 975-76 (holding that the California Court of Appeal erred under AED-PA by applying California harmless error standard rather than federal constitutional harmless error principles).

Because the state court’s application of a stricter standard than is permissible in the case of Napue error was “contrary to” clearly established Supreme Court law, the “contrary to” prong of AEDPA, § 2254(d)(1), has been satisfied. See Tow-ery v. Schriro, 641 F.3d 300, 307 (9th Cir.2010) (citations omitted); see also Early v. Packer, 537 U.S. 3, 8, 123 S.Ct. 362, 154 L.Ed.2d 263 (2002) (per curiam). Thus, we must next “resolve the claim without the deference AEDPA otherwise requires.” Crittenden v. Ayers, 624 F.3d 943, 954 (9th Cir.2010) (quoting Panetti v.

Quarterman, 551 U.S. 930, 953, 127 S.Ct. 2842, 168 L.Ed.2d 662 (2007)). That is, we apply de novo review to Dow’s federal constitutional claim. See Bains, 204 F.3d at 976. III. Applying de novo review, we conclude that Dow prevails on his Napue claim because he meets the materiality standard. This standard, which requires us to determine whether “there is any reasonable likelihood that the false testimony could have affected the judgment of the jury,” Agurs, 427 U.S. at 103, 96 S.Ct. 2392 (emphasis added), is easily met here.

The evidence against Dow was weak and the prosecutor’s arguments undoubtedly had an effect on the jury’s decision. Thus, Dow was deprived of his constitutional right to due process of law. Dow’s first trial resulted in a deadlocked jury, proof that his case was a close one. See Kennedy v. Lockyer, 379 F.3d 1041, 1056 n. 18 (9th Cir.2004) (fact that jury was deadlocked in first trial showed that the question of “guilt or innocence was a close one”); Caliendo, 365 F.3d at 699 (fact that jury was deadlocked in first trial indicates error in second trial was prejudicial).

At the second trial, like the first, the prosecution’s case against Dow consisted principally of Sablad’s identification and a generic gray sweatsuit recovered from Dow’s apartment. By that point, Sablad’s identification had been severely undermined by his inconsistent recollection of Dow’s appearance. During the investigation, Sablad told the detective that he did not know the location of the scar that he remembered having been on the perpetrator’s face. At the preliminary hearing, he said that the scar was on the right jaw line.

Then, at the first trial, Sablad described it as a longer scar near the eyebrow and slanting downward. When questioned at the second trial, he looked at Dow and said that he could not even see the scar about which he had previously testified. In contrast, when Detective Oglesby was called by the prosecution to explain the lineups, he testified that the scar was a small one under Dow’s right eye. He added that the scar could be seen from a foot or two away, but not from a distance.

As to Sablad’s belated recollection of a missing tooth, Sablad first said it was a lower right tooth and then that it might have been on top. Detective Ogles-by, however, testified that Dow (like many others) has only a gap between his two front teeth. He did not mention any missing tooth, top or bottom. Following this inconsistent testimony from the prosecution’s central witnesses, Dr. Shomer, the expert witness for the defense, provided extensive testimony as to the unreliability of eyewitness identifications generally and in the case of Dow in particular.

He also emphasized the suggestiveness of the police procedures used throughout the investigation. Here, it is reasonably likely that the false testimony and the prosecutor’s arguments based on that testimony had a material effect on the outcome of the jury’s deliberations. The case was a weak one that hinged almost entirely on Sablad’s inconsistent eyewitness testimony. The prosecutor argued on the basis of the evidence admitted in violation of Napue that Dow had acted in a manner consistent with a consciousness of guilt.

This argument bolstered the prosecution’s case that Dow was guilty by interjecting a new reason for the jury to convict him. The jury may well have concluded that the questionable identification was validated by Dow’s supposed self-incriminating act. Moreover, the Napue violation was particularly egregious because it adversely affected Dow’s due process interest in a lineup that was not unduly suggestive, see Manson v. Brathwaite, 432 U.S. 98, 107-09, 97 S.Ct. 2243, 53 L.Ed.2d 140 (1977); Neil v.

Big-gers, 409 U.S. 188, 198, 93 S.Ct. 375, 34 L.Ed.2d 401 (1972), and negatively affected his counsel’s effort to ensure a fair trial by portraying it as an acknowledgment of guilt. Furthermore, the prosecutor made her improper argument during her rebuttal, leaving the defense with no opportunity to respond, beyond the limited response counsel made by objecting. The fact that the objection may have made the jury aware of another inference to be drawn from the false evidence does not turn “what was otherwise a tainted trial into a fair one.” Napue, 360 U.S. at 270, 79 S.Ct. 1173.

Finally, the jury rendered a guilty verdict after hearing the false testimony while the first jury, which did not hear that testimony, failed to do so. Thus, we conclude that there is a “reasonable likelihood that the false testimony could have affected the judgment of the jury.” Agwrs, 427 U.S. at 103, 96 S.Ct. 2392. Because the prosecutor’s actions during his state court trial violated Dow’s constitutional rights to due process, reversal is required and we are compelled to grant the writ of habeas corpus. IV.

Nonetheless, the state asserts that the state court’s decision is entitled to AEDPA deference because it implicitly applied the Napue materiality requirement and implicitly rejected Dow’s claim as not material.

This contention has no merit because, as the state conceded, “the state appellate court found the first two prongs of the federal test satisfied,” and “the court of appeal found the prosecutor had committed misconduct violative of Napue.” It would stretch our credulity to believe that, having explicitly applied the traditional state court harmless error standard to the violation, and having done so erroneously, the state court also implicitly applied the proper Napue standard.

Our conclusion is confirmed by the fact that the state court reached a result that may well have been appropriate under Welch, but that would have been unreasonable to reach under Napue. Even if we were to assume, however, that the state court did reject Dow’s Na-pue claim on materiality grounds after applying the Napue standard, we would reach the same ultimate outcome, because finding the Napue violation in this case to be immaterial would be “objectively unreasonable” and, therefore, satisfy the “unreasonable application” prong of § 2254(d)(1). Williams v.

Taylor, 529 U.S. 362, 409, 411, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000). In light of the Supreme Court’s clearly established precedent on Napue violations, there was no “reasonable basis for the state court to deny relief’ on materiality grounds. Harrington v. Richter, — U.S. -, 131 S.Ct. 770, 784, 178 L.Ed.2d 624 (2011). Our conclusion follows from our earlier analysis of Dow’s claim.

Not only does Dow prevail on his claim under Napue because there is a “reasonable likelihood that the false testimony could have affected the judgment of the jury,” Agurs, 427 U.S. at 103, 96 S.Ct. 2392, but to conclude that the violation was not material under Napue would, in light of our earlier description of the facts, be beyond the scope of “possibility for fairminded disagreement.” Harrington, 131 S.Ct. at 787. As we have explained, the evidence in Dow’s case was extremely weak.

It depended solely on Sablad’s identification of Dow as the perpetrator (and the fact that Dow possessed a generic gray sweatsuit), but Sablad made materially inconsistent statements regarding both the perpetrator’s scar and a purportedly missing tooth—at one point stating that the scar was on the right jaw line, at another that it ran down his face from the eyebrow. Dow’s scar, however, is small and located directly under his eye. Nor does Dow have a missing tooth.

Moreover, the prosecutor’s Napue violation gave the jury an independent (and improper) basis for finding Dow guilty—that his actions were evidence of his consciousness of guilt. Given these facts, any conclusion that the Napue violation in Dow’s case did not meet the materiality requirement would be objectively unreasonable. Thus, even the “unreasonable application” prong of § 2254(d)(1) is satisfied here.

CONCLUSION Because it is clear that the constitutional violation at issue here requires reversal under Napue, we REVERSE the judgment of the district court, and REMAND with directions to issue a writ of habeas corpus, releasing Dow from detention unless the state retries him within a reasonable period of time to be determined by the district court. REVERSED and REMANDED. . Because we grant the writ on Dow’s claim of knowingly eliciting false testimony, we do not reach his claim of ineffective assistance of counsel based on counsel's failure to object to that testimony. .

The only other evidence linking Dow to the crime was the generic grey sweatsuit found in Dow's residence. Sablad identified the gray sweatsuit, seized from Dow’s home, as consistent with the clothing worn by the perpetrator. . On a petition for federal habeas, we review the last reasoned state-court decision on Dow’s claim, in this case the California appellate court’s unpublished opinion. See Ylst v. Nunnemaker, 501 U.S. 797, 806, 111 S.Ct. 2590, 115 L.Ed.2d 706 (1991). . Whether an appellate standard of review is relatively "less” or "more” demanding typically depends on one’s point of view.

We consider the difference between the Napue materiality standard and the California harmless error rule from the perspective of Dow, who is challenging his conviction. Because the California harmless error rule applied by the state court would require Dow to demonstrate a higher degree of prejudice in order to merit relief, we consider that standard to be more strict than the materiality standard under Napue. See, e.g., Bains v. Cambra, 204 F.3d 964, 976 (9th Cir.2000). .

California state appellate courts apply this harmless error standard in reviewing "non-constitutional magnitude, trial type errors.” See Bains v. Cambra, 204 F.3d 964, 971 n. 2 (9th Cir.2000) (describing the Watson standard of review, which is the same as the standard applied by the state court in Dow's case). . That the state standard is stricter is also reflected in its use of the term “reasonably probable,” in contrast to Napue’s use of the term "any reasonable likelihood.” .

We note that the state does not contend that there were any other material differences between the first and second trials, and the state court opinion does not suggest any. . The "fairminded jurist” standard is an objective standard of law, not a reference to the quality of the judge making the decision.

The standard, therefore, does not require us to evaluate whether the individual jurists are "fairminded” in the sense that they are generally impartial and honest adjudicators, but rather whether there could objectively be fair-minded disagreement as to the outcome dictated by the Supreme Court's clearly established law. Fairminded jurists can make mistakes in legal reasoning or judgment, and if such a mistake is beyond reasonable legal disagreement, the "fairminded jurist" standard is satisfied.

Were we to apply a fairminded jurist standard literally, a federal court could never reverse a state court’s habeas decision. For every state appellate court contains at least one fairminded jurist, if not a majority of its supreme court or appellate court members who voted to reject the petitioner’s arguments. When we reverse a state court’s habe-as decision we are surely not saying that all the state court justices whom we are reversing are not fairminded jurists, but rather that objectively the answer is one that a fairmind-ed jurist should reach.