People v. Lemcke

486 P.3d 1077, 11 Cal. 5th 644, 278 Cal. Rptr. 3d 849

• Court: California Supreme Court
• Decided: May 27, 2021
• Docket: S250108
• Status: Published

Opinion

IN THE SUPREME COURT OF CALIFORNIA

THE PEOPLE, Plaintiff and Respondent, v. DESIRAE LEE LEMCKE et al., Defendants and Appellants.

S250108

Fourth Appellate District, Division Three G054241

Orange County Superior Court 14CF3596

May 27, 2021

Justice Groban authored the opinion of the Court, in which Chief Justice Cantil-Sakauye and Justices Corrigan, Liu, Cuéllar, Kruger, and Jenkins concurred. PEOPLE v. LEMCKE S250108

Opinion of the Court by Groban, J.

Defendant Charles Henry Rudd was convicted of assault and robbery. The prosecution’s primary evidence at trial was the testimony of the victim, who identified Rudd as her assailant and confirmed that she had previously identified Rudd during a photographic lineup. The trial court provided the jury an instruction modeled on CALCRIM No. 315 that listed 15 factors it should consider when evaluating eyewitness identification evidence. One of those factors stated: “How certain was the witness when he or she made an identification?” Rudd argues that the certainty instruction violated his federal and state due process rights to a fair trial (see U.S. Const., 5th & 14th Amends.; Cal. Const., art. I, §§ 7, 15) because empirical research has shown that a witness’s confidence in an identification is generally not a reliable indicator of accuracy. We reject Rudd’s due process claims. When considered in the context of the trial record as a whole, listing the witness’s level of certainty as one of 15 factors the jury should consider when evaluating identification testimony did not render Rudd’s trial fundamentally unfair. (See Salas v. Cortez (1979) 24 Cal.3d 22, 27 (Salas) [“The touchstone of due process is fundamental fairness”]; People v. Foster (2010) 50 Cal.4th 1301, 1335 (Foster) [in determining whether a jury instruction violated a defendant’s right to due process, the “ ‘instruction “may not be judged in artificial isolation,” but must be considered in the context of the instructions as a whole and the trial record’ ”

1 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

(italics omitted)].) As we explained in a prior decision approving CALJIC No. 2.92’s similarly worded instruction on witness certainty, the instruction does not direct the jury that “certainty equals accuracy.” (People v. Sánchez (2016) 63 Cal.4th 411, 461– 463 (Sánchez); see People v. Johnson (1992) 3 Cal.4th 1183, 1231–1232 (Johnson) [approving CALJIC No. 2.92’s certainty instruction].) Although the language may prompt jurors to conclude that a confident identification is more likely to be accurate, Rudd was permitted to call an eyewitness identification expert who explained the limited circumstances when certainty and accuracy are positively correlated. Moreover, the court provided additional instructions directing the jury that it was required to consider the testimony of the expert witness, that the prosecution retained the burden to prove Rudd’s identity as the perpetrator beyond a reasonable doubt, and that witnesses sometimes make honest mistakes. Despite the absence of a constitutional violation, we nonetheless agree with amici curiae that a reevaluation of the certainty instruction is warranted. Contrary to widespread lay belief, there is now near unanimity in the empirical research that “eyewitness confidence is generally an unreliable indicator of accuracy.” (State v. Henderson (N.J. 2011) 27 A.3d 872, 899 (Henderson); see Commonwealth v. Gomes (Mass. 2015) 22 N.E.3d 897, 912–913 (Gomes); State v. Guilbert (Conn. 2012) 49 A.3d 705, 721–723 (Guilbert); State v. Lawson (Or. 2012) 291 P.3d 673, 704 (Lawson).) As currently worded, CALCRIM No. 315 does nothing to disabuse jurors of that common misconception, but rather tends to reinforce it by implying that an identification is more likely to be reliable when the witness has expressed certainty. This is especially problematic because many studies have also shown eyewitness confidence is the

2 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

single most influential factor in juror determinations regarding the accuracy of an identification. (See Lawson, at pp. 704–705.) Given the significance that witness certainty plays in the factfinding process, we refer the matter to the Judicial Council and its Advisory Committee on Criminal Jury Instructions to evaluate whether or how the instruction might be modified to avoid juror confusion regarding the correlation between certainty and accuracy. (See Cal. Rules of Court, rule 2.1050(d).) Acting pursuant to our supervisory powers, we further direct that until the Judicial Council has completed its evaluation, trial courts should omit the certainty factor from CALCRIM No. 315 unless the defendant requests otherwise. I. BACKGROUND A. The Crime and Investigation 1. The assault and initial investigation In July of 2014, Monica Campusano traveled to a motel to visit a friend. While walking down a hallway, Campusano saw a woman standing outside room 216 and a man standing just inside the doorway. The woman asked if she could use Campusano’s cell phone. As Campusano began to retrieve her phone, the man in the doorway suddenly struck her in the face and pulled her into room 216. The man then punched and kicked Campusano repeatedly, causing her to lose consciousness. When Campusano regained consciousness she was alone and her purse and phone were gone. She immediately went to the motel lobby and called 9-1-1. Ricardo Velasquez, an officer of the Santa Ana Police Department, responded to the call and interviewed Campusano at the motel. Campusano described her assailant as an African-American male, between six feet

3 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

three inches to six feet five inches in height and weighing 260 to 300 pounds. Campusano described the female who had asked to use her phone as a “heavy set white female,” standing approximately five feet six inches in height and weighing over 200 pounds. After arranging an ambulance for Campusano, Officer Velasquez obtained records from the motel manager that showed room 216 was registered to a woman named Desirae Lee Lemcke. Officer Velasquez ran a records check on Lemcke and determined that she matched Campusano’s description of the female standing outside the motel room. The records check also revealed Lemcke had previously obtained a “no contact” order against a man named Charles Rudd, who the order described as an African-American male, six feet three inches in height and weighing approximately 250 pounds. Later that evening, Officer Velasquez created a six-pack photographic lineup that included an image of Rudd and then drove to the hospital where Campusano was receiving treatment. When Velasquez arrived, Campusano was “under anesthesia,” but stated that she could answer questions. Officer Velasquez showed her the photographic lineup and asked whether she saw the person who had attacked her. Campusano pointed to Rudd’s photograph and stated that she recognized his nose, mouth and jaw area. She signed her name by the photo but used an incorrect spelling. 2. Follow-up investigation Approximately three months after Campusano initially identified Rudd, Santa Ana Police Department Detective Adrian Silva contacted her to discuss the details of the assault. Silva showed Campusano a photographic lineup of several females

4 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

that included an image of Lemcke. Campusano selected Lemcke’s photograph. Campusano also mentioned for the first time that the male assailant had a tattoo on his neck. Based on the information Campusano had provided, Detective Silva prepared a photographic lineup showing images of two neck tattoos. The first photograph showed the neck tattoo of Rudd and the second showed the neck tattoo of Lemcke’s current boyfriend. The men’s faces were not shown in the photographs. Silva met with Campusano again six days later and asked whether either of the tattoos resembled the one she had seen on her assailant’s neck. Campusano selected the photograph of Rudd’s tattoo, explaining that it looked “more like the one that she remembered.” Silva then asked Campusano whether she recalled having previously selected an image of an African- American male during a photographic lineup. After Campusano said yes, Silva showed her the image of Rudd that Officer Velasquez had used in the original photographic lineup and asked whether “she remembered that this was [the] . . . photo . . . that she had already identified.” Campusano confirmed it was the same photograph. Silva then showed Campusano a second six-pack photographic lineup that did not include Rudd. Campusano stated that she did not recognize anyone, and then pointed back to the photograph of Rudd and said, “for sure it was [him].” In October of 2014, the Orange County District Attorney filed an information charging Lemcke and Rudd with second-

5 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

degree robbery (Penal Code, §§ 211, 212.5, subd. (c))1 and aggravated assault by means of force likely to result in great bodily injury (§ 245, subd. (a)(l)). The information also charged Rudd with battery with serious bodily injury. (§ 243, subd. (d).) B. Trial 1. Prosecution’s witnesses At trial, Campusano testified about her recollection of the assault. Campusano stated that she had a good view of her assailant during the attack and “remember[ed] his face well.” The prosecution asked Campusano whether she saw the person who had attacked her in the courtroom. Campusano said yes, explaining that he was sitting at the defense table. The prosecutor asked Campusano to describe what the person was wearing. Campusano then stood up and stated that the individual was wearing a blue shirt with a black shirt underneath. When asked what features of his face she remembered, Campusano stated: “I remember his face, his tattoo and his look, like he was looking with anger.” Campusano also confirmed that she had selected Rudd from a photographic lineup Officer Velasquez had shown to her at the hospital. On cross-examination, Campusano admitted that she could not see the defense table when she had first identified Rudd during her direct examination, which is why she had to stand up to describe his clothing. Campusano explained that she had identified Rudd before seeing him because “it [was] logical that he was going to be in the courtroom.” She reiterated

1 Unless otherwise noted, all further statutory citations are to the Penal Code.

6 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

that she “remember[ed] [Rudd’s] face well” and that it was “impossible for [her] not to recognize his face.” Defense counsel questioned Campusano about various inconsistent statements she had made throughout the investigation, including her initial failure to inform Officer Velasquez that her attacker had a neck tattoo, her description of what the perpetrator had said to her during the attack, where she was when she regained consciousness and the extent of her injuries. Defense counsel also elicited testimony in which Campusano admitted she had previously been convicted of soliciting prostitution and lying to a police officer. Campusano further admitted that after having spoken with law enforcement about her assault, she applied for a “U visa,” a form of temporary visa that provides legal status for noncitizen victims of serious crimes who assist in the investigation. (See People v. Morales (2018) 25 Cal.App.5th 502, 506; 8 U.S.C. § 1101(a)(15).) The defense introduced a copy of Campusano’s visa application, which stated that she had been kidnapped during the incident, an allegation she never made to the investigating officers. Officer Velasquez and Detective Silva also testified at trial regarding their role in the investigation. Both officers explained what Campusano had told them during their interviews and described how they had administered the photographic lineups. (See ante, at pp. 3–5.) 2. Defense’s expert witness on eyewitness identifications Mitchell Eisen, a psychology professor, testified on behalf of the defense as an expert on eyewitness identifications. Eisen discussed a wide range of subjects related to the reliability of eyewitness identifications including the procedures law

7 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

enforcement should follow to ensure a fair, nonsuggestive identification process. Several of the procedures Eisen described differed from the procedures that Velasquez and Silva had used when administering Campusano’s identifications. Eisen also identified various factors that can affect a witness’s identification, including what Eisen referred to as the “commitment effect.” Eisen explained that once a witness has made an initial identification of a person, the witness is likely to select that person again in any future identification regardless of the accuracy of the initial identification. Eisen emphasized that the commitment effect is particularly problematic for in- trial identifications because a witness who has agreed to testify is normally prepared to “reassert their belief” as to whatever identification he or she had previously made. According to Eisen, a witness’s statements at trial regarding an identification are not reflective of memory, but rather reflect “their honest belief that they’ve come to believe after a long process that starts at the very first viewing, and every identification procedure that follows and every discussion and every piece of information they get . . . until [the trial].” Eisen also testified about the correlation between a witness’s level of certainty and the accuracy of the identification. Eisen stated that research on that issue had “come a long way,” explaining: “Just a few years ago, if you had an expert in my position sitting here, they might say something like confidence is not related to accuracy in any regard, but that’s not really true.” According to Eisen, current research suggested that “confidence” can be “useful” when there has been a “fair lineup soon after the event.” However, “once outside that window and you go forward, that moment in time when [the witness] made

8 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

an [identification], once you get past that, confidence is not related to accuracy in any regard.” Eisen further explained that an expression of certainty is only useful when the identification is made soon after the event because, as the investigation progresses, a witness will normally receive additional information that can unconsciously bolster his or her confidence. Eisen clarified that this additional information may come in the form of positive feedback from law enforcement or even the prosecutor’s decision to file charges against the suspect: “If you inform [the witness] inadvertently [or] purposefully . . . that they chose the right picture, we can drive confidence through the ceiling. [A witness can] become very, very confident in their choice even when they weren’t originally.” Eisen continued, “So what these studies show . . . is really the fundamental point and indisputable, that . . . [b]y the time they get to trial, they learned at least the government believes that they are the right person and that’s why they are prosecuting. It gives people sort of confirming feedback you can manipulate and drive confidence, irrespective of memory. . . . If we are going to look at confidence at all, only be confident in the moment they make the selection from a fair and unbiased identification test.” 3. Jury instructions on witness certainty After the close of evidence, the trial court heard argument regarding the parties’ proposed jury instructions. The prosecution requested an instruction modeled on CALCRIM No. 315 that listed 15 factors the jury should consider when evaluating the credibility and accuracy of eyewitness identification evidence. Counsel for Rudd requested that the court strike the eleventh factor, which directed the jury to

9 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

consider: “How certain was the witness when he or she made an identification?” Counsel explained that his objection was based on a concurring opinion in Sánchez, supra, 63 Cal.4th 411, that questioned the continuing validity of the certainty instruction in light of empirical research finding little correlation “between witness confidence and witness accuracy.” The trial court denied Rudd’s request, explaining that the certainty language set forth in CALCRIM No. 315 was consistent with the defense expert’s testimony that confidence can be reliable under certain circumstances. The court acknowledged the expert had cast doubt on the usefulness of certainty more generally but concluded that the defense could raise those points at closing argument. 4. Closing argument At closing argument, the prosecution focused on Campusano’s eyewitness testimony, asserting that she had consistently identified Rudd and Lemcke as the perpetrators and that the “essentials of her testimony” about the event had never changed. Acknowledging that the accuracy of Campusano’s identification was “obviously an issue,” the prosecutor directed the jury to CALCRIM No. 315, explaining that it provided a “tool to evaluate an eyewitness identification.” The prosecutor also reminded the jury that it had heard an expert testify about the subject “for quite a while.” The prosecutor then presented argument as to why various factors in CALCRIM No. 315 suggested Campusano’s identification was accurate, explaining that she had provided an accurate description of Rudd prior to the identification, she had picked him out of a group of people, and she had never failed to identify him. Regarding Campusano’s level of certainty, the prosecutor

10 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

asserted that Campusano “was certain the entire time. And she came in here and saw [Rudd] when she walked into court and said, ‘that’s him.’ ” Rudd’s closing argument focused primarily on Campusano’s credibility as a witness. Defense counsel argued that Campusano had made numerous inconsistent statements about the event and been argumentative throughout her cross- examination. Counsel also emphasized that Campusano had been convicted of prostitution and lying to the police, arguing that these prior convictions “show[ed] a readiness to lie to a police officer, a readiness to lie to the judge, and a readiness to lie to you folks.” In addition to questioning Campusano’s credibility, the defense presented alternative theories about what had actually occurred at the motel. First, counsel posited that Campusano may have gone to the motel to solicit prostitution and then got into a fight with a dissatisfied customer. Second, counsel theorized that Campusano had “to claim she was a victim of a violent felony in order for her to get that U visa to clear up her immigration status.” Defense counsel then addressed Campusano’s identification of Rudd, asserting that the procedures the police had used were suggestive and unreliable. Counsel noted that the first identification had occurred while Campusano was under anesthesia, and that the second identification was “extremely suggestive” because Detective Silva had merely shown Campusano the same photograph she had selected during the initial lineup that Officer Velasquez had conducted. Finally, defense counsel argued that Eisen’s expert testimony showed Campusano’s in-court identification was of no value, and that “confidence does not equal reliability [or] . . . accuracy.”

11 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

The jury convicted Lemcke and Rudd on all counts. C. Rudd’s Appeal Rudd raised a single issue on appeal, contending that his “state and federal due process rights were violated when the court instructed the jury . . . to consider an eyewitness’s level of certainty when evaluating an identification.” 2 The Court of Appeal acknowledged that numerous scientific studies had found that a “witness’s certainty does not make the identification any more likely to be accurate.” The court explained, however, that Johnson, supra, 3 Cal.4th 1183, and Sánchez, supra, 63 Cal.4th 411, had expressly approved the use of similarly worded instructions on witness certainty. While noting that a concurring opinion in Sánchez questioned the continuing validity of such an instruction, the court concluded that it was “bound by the decisions in Sánchez and Johnson.” Rudd filed a petition for review seeking resolution of the following question: “Does instructing a jury with CALCRIM No. 315, which directs the jury to consider an eyewitness’s level of certainty when evaluating an identification, violate a defendant’s federal and state due process rights?”3

2 Codefendant Lemcke filed a separate appeal that raised an unrelated instructional error claim. The Court of Appeal rejected her claim, and we denied her petition for review. 3 Rudd’s petition also listed a second, broader question: “Are rules that assign the trial court a stronger gatekeeping role in the admission of eyewitness identification required or advisable?” Rudd’s briefing, however, does not address this “gatekeeper” issue and he did not raise the issue in either the trial court or the Court of Appeal. We therefore decline to address the merits of this claim, deeming it both abandoned and

12 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

II. DISCUSSION A. The Instruction on Witness Certainty Did Not Violate Rudd’s Due Process Rights Rudd has not challenged the procedures that Officer Velasquez and Detective Silva used to conduct the photographic lineups that preceded Campusano’s identifications, nor has he challenged the admission of any of the identification evidence. (See generally People v. Ochoa (1998) 19 Cal.4th 353, 412 [assessing whether lineup procedures rendered identification evidence inadmissible].)4 Instead, his sole claim is that the trial court violated his right to due process by listing the witness’s

forfeited. (See People v. Tanner (1979) 24 Cal.3d 514, 518, fn. 2 [issue raised in the notice of appeal, but not addressed in the briefing, deemed abandoned]; Barker v. Lull Engineering Co. (1978) 20 Cal.3d 413, 422, fn. 3 [courts generally “assume” appellant has “abandoned any claim” that is not briefed]; In re Sheena K. (2007) 40 Cal.4th 875, 880 [“Ordinarily, a criminal defendant who does not challenge an assertedly erroneous ruling of the trial court in that court has forfeited his or her right to raise the claim on appeal”].) 4 As discussed in more detail below (see post, at pp. 31–32 & fn. 16), the Legislature recently adopted a statute that requires law enforcement agencies to enact regulations mandating the use of certain procedures when administering “photo lineups and live lineups with eyewitnesses.” (§ 859.7, subd. (a).) The legislation was not in effect when Campusano made her identifications and Rudd has not raised any claim with respect to the new statute.

13 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

level of certainty as one of 15 factors the jury should consider when evaluating eyewitness identification testimony. 5

5 The full instruction the trial court provided to the jury in this case, which is essentially identical to CALCRIM No. 315, stated as follows: “You have heard eyewitness testimony identifying the defendant. As with any other witness, you must decide whether an eyewitness gave truthful and accurate testimony. In evaluating identification testimony, consider the following questions: • “Did the witness know or have contact with the defendants before the event? • “How well could the witness see the perpetrator? • “What were the circumstances affecting the witness’s ability to observe, such as lighting, weather conditions, obstructions, distance, and duration of observation? • “How closely was the witness paying attention? • “Was the witness under stress when he or she made the observation? • “Did the witness give a description and how does that description compare to the defendants? • “How much time passed between the event and the time when the witness identified the defendants? • “Was the witness asked to pick the perpetrator out of a group? • “Did the witness ever fail to identify the defendants? • “Did the witness ever change his or her mind about the identification? • “How certain was the witness when he or she made an identification? • “Are the witness and the defendant of different races?

14 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

Rudd’s claim is based on empirical research showing that confidence in an identification is generally not a reliable indicator of accuracy. (See Gomes, supra, 22 N.E.3d at pp. 912– 913; Guilbert, supra, 49 A.3d at pp. 721–723; Lawson, supra, 291 P.3d at pp. 704–705; Henderson, supra, 27 A.3d at pp. 898–899.) Rudd contends that the certainty instruction set forth in CALCRIM No. 315 implies just the opposite, effectively causing jurors to “equat[e] certainty with accuracy” and to “place more value than merited on the eyewitness’s confidence.” In Rudd’s view, if a trial court elects to instruct a jury on witness certainty, “[d]ue process requires that the [instruction be] accompanied by information reflecting scientific research.” “The touchstone of due process is fundamental fairness.” (Salas, supra, 24 Cal.3d at p. 27; see Gagnon v. Scarpelli (1973) 411 U.S. 778, 790 [“[F]undamental fairness [is] the touchstone of due process”].) A jury instruction may “ ‘so infuse[] the trial with unfairness as to deny due process of law.’ ” (Estelle v. McGuire (1991) 502 U.S. 62, 75 (Estelle).) However, “ ‘not every ambiguity, inconsistency, or deficiency in a jury instruction rises to the level of a due process violation. The question is “ ‘whether the ailing instruction . . . so infected the entire trial that the

• “Was the witness able to identify other participants in the crime? • “Was the witness able to identify the defendant in a photographic or physical lineup? • “Were there any other circumstances affecting the witness’s ability to make an accurate identification? “The People have the burden of proving beyond a reasonable doubt that it was the defendant who committed the crime. If the People have not met this burden, you must find the defendant not guilty.”

15 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

resulting conviction violates due process.’ ” ’ ” (People v. Mills (2012) 55 Cal.4th 663, 677, quoting Estelle, p. 72.) “ ‘It is well established that the instruction “may not be judged in artificial isolation,” but must be considered in the context of the instructions as a whole and the trial record.’ ” (Foster, supra, 50 Cal.4th at p. 1335, italics omitted; see People v. Haskett (1990) 52 Cal.3d 210, 235.) “ ‘If the charge as a whole is ambiguous, the question is whether there is a “ ‘reasonable likelihood that the jury has applied the challenged instruction in a way’ that violates the Constitution.” ’ ” (Mills, p. 677.) Over the past 30 years, we have repeatedly endorsed the use of instructions that direct the jury to consider an eyewitness’s level of certainty when evaluating identification evidence. In People v. Wright (1988) 45 Cal.3d 1126 (Wright), we approved CALJIC No. 2.92, a predecessor to CALCRIM No. 315, that includes similar language regarding witness certainty.6 Rejecting arguments similar to those Rudd raises here, we also concluded that an “instruction on eyewitness identification factors should . . . . [¶] . . . not take a position as to

6 The language in CALJIC No. 2.92 directs the jury to consider “[t]he extent to which the witness is either certain or uncertain of the identification.” Although this language differs somewhat from CALCRIM No. 315, which directs the jury to consider “[h]ow certain was the witness when he or she made an identification,” neither party has argued that this slight difference in phrasing is material. For purposes of Rudd’s due process claims, we find no material distinction between the two instructions. In effect, the instructions set forth two ways of saying the same thing: that jurors should consider the witness’s level of certainty when assessing the credibility and accuracy of the identification testimony.

16 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

the impact of each of the psychological factors listed” (Wright, at p. 1141), and that “[the] explanation of the effects of those factors is best left to argument by counsel, cross-examination of the eyewitnesses, and expert testimony where appropriate” (id. at p. 1143; see People v. McDonald (1984) 37 Cal.3d 351 (McDonald) [approving the use of expert testimony describing how psychological factors can affect the reliability of eyewitness identifications]). Four years later, we rejected a claim that the trial court erred when it refused to strike the witness certainty factor set forth in CALJIC No. 2.92. (Johnson, supra, 3 Cal.4th 1183.) Citing Wright, supra, 45 Cal.3d 1126, we reiterated that an instruction on eyewitness identification testimony “should not take a position as to the impact of each of the psychological factors listed” (Johnson, at p. 1230) and that CALJIC No. 2.92 “normally provides sufficient guidance on the subject . . . .” (Johnson, at p. 1230). Most recently, in Sánchez, supra, 63 Cal.4th 411, we rejected a claim challenging CALJIC No. 2.92’s certainty language based on “scientific studies [finding] that . . . there is, at best, a weak correlation between witness certainty and accuracy.” (Sánchez, at p. 461.) We explained that the studies defendant had cited were “nothing new” (id. at p. 462), noting that our decision in McDonald, supra, 37 Cal.3d 351, decided 32 years earlier, had “cited some of [those studies] . . . [in support of its] holding that the trial court has discretion to admit expert testimony regarding the reliability of eyewitness identification” (Sánchez, at p. 462). We further explained that our prior decisions had “specifically approved” the use of the “certainty factor,” and that the instruction was facially neutral, “telling the jury only that it could consider [a witness’s level of certainty]. It

17 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

did not suggest that certainty equals accuracy.” (Ibid.) Finally, we observed that the instruction was at least partially beneficial to the defendant because some of the trial witnesses had expressed uncertainty in their identification and concluded that “[a]ny reexamination of our previous holdings in light of developments in other jurisdictions should await a case involving only certain identifications.” (Ibid.) In a concurring opinion, Justice Liu disagreed with the majority’s assertion that the instruction was neutral, arguing that the language in CALJIC No. 2.92 “naturally ‘prompt[ed] the jury to conclude that an eyewitness identification is more reliable when the witness expresses greater certainty.’ ” (Sánchez, supra, 63 Cal.4th at p. 495 (conc. opn. of Liu, J.).) Although Justice Liu agreed that any instructional error was harmless based on the additional evidence of guilt presented at trial, he contended that this court should reevaluate “whether it is proper . . . to instruct that witness certainty is a factor bearing on the accuracy of an identification that juries should consider.” (Id. at p. 498.) Rudd disagrees with our prior decisions in Wright, Johnson and Sánchez. He contends that instructing the jury to consider an eyewitness’s level of certainty, without clarifying the limited correlation between certainty and accuracy, violates due process in two ways. First, the instruction “lowers the prosecution’s burden of proof” by causing jurors to “equat[e] certainty with accuracy, when science establishes otherwise.” Second, the instruction denies the defendant “a ‘meaningful opportunity to present a complete defense’ ” as to “why the identification was flawed . . . .”

18 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

Regarding Rudd’s first contention, we find nothing in CALCRIM No. 315’s instruction on witness certainty that operates to “lower the prosecution’s burden of proof.” As Sánchez, supra, 63 Cal.4th 411, explained with respect to CALJIC No. 2.92’s similarly worded instruction, the instruction does not direct the jury that “certainty equals accuracy.” (Sánchez, at p. 462.) Nor does the instruction state that the jury must presume an identification is accurate if the eyewitness has expressed certainty. (Cf. Francis v. Franklin (1985) 471 U.S. 307, 316 [instruction that “directs the jury to presume an essential element of the offense” violates due process].) Instead, the instruction merely lists the witness’s level of certainty at the time of identification as one of 15 different factors that the jury should consider when evaluating the credibility and accuracy of eyewitness testimony. The instruction leaves the jury to decide whether the witness expressed a credible claim of certainty and what weight, if any, should be placed on that certainty in relation to the numerous other factors listed in CALCRIM No. 315. Indeed, even Rudd acknowledges that, on its face, the instruction is “superficially neutral.” Although the wording of the instruction might cause some jurors to infer that certainty is generally correlative of accuracy (see post, at pp. 26–39), Rudd was permitted to present expert witness testimony to combat that inference. Rudd’s expert (Eisen) testified that the only time certainty may be useful in assessing accuracy is when the identification is made in close temporal proximity to the event and law enforcement has utilized nonsuggestive procedures. According to Eisen, “outside that window, . . . confidence is not related to accuracy in any regard.” Eisen emphasized that in-trial identification testimony is particularly meaningless because it does not “reflect[]

19 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

memory.” Eisen also described the procedures law enforcement should follow to ensure an accurate identification and answered a series of hypothetical questions that were designed to show those procedures were not followed in this case. (See People v. Vang (2011) 52 Cal.4th 1038, 1045 [expert may render opinion testimony based on hypothetical question that are rooted in the facts shown by the evidence].) Nothing in CALCRIM No. 315 suggested that the jury should ignore Eisen’s expert opinion on witness certainty. To the contrary, the jury received a separate instruction on expert testimony (CALCRIM No. 332) directing that it “must consider th[ose] opinions.” (Italics added.) The jury also received a general instruction on witness testimony explaining that “[p]eople sometimes honestly . . . make mistakes about what they remember” and that the jurors were responsible for “judg[ing] the credibility or believability of the witnesses.” The jury “thus remained free to exercise its collective judgment to reject what it did not find trustworthy or plausible.” (Cupp v. Naughten (1973) 414 U.S. 141, 149 (Cupp).) Additional instructions the jury received in this case further undercut Rudd’s contention that the certainty language lowered the prosecution’s burden of proof. (See Foster, supra, 50 Cal.4th at p. 1335 [“ ‘the instruction “may not be judged in artificial isolation,” but must be considered in the context of the instructions as a whole . . .’ ”].) The trial court expressly directed the jury that Rudd was presumed innocent, and that the prosecution had the burden of proving all elements of the crime beyond a reasonable doubt. The instruction on eyewitness identification evidence reiterated that requirement with respect to Rudd’s identity, stating: “The People have the burden of proving beyond a reasonable doubt that it was the defendant

20 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

who committed the crime. If the People have not met this burden, you must find the defendant not guilty.” (See Cupp, supra, 414 U.S. at p. 149 [instruction directing that witnesses were presumed to be truthful did not lower the prosecution’s burden of proof where the court provided additional instructions that “fully and explicitly [charged the jury] about the presumption of innocence and the State’s duty to prove guilt beyond a reasonable doubt”].) Our conclusion that CALCRIM No. 315’s certainty instruction did not operate to lower the prosecution’s burden of proof under the facts presented here finds substantial support in federal case law.7 In Cupp, supra, 414 U.S. 141, the prosecution called two witnesses who identified the defendant as the perpetrator of an armed robbery; the defendant called no witnesses. The trial court instructed the jury that “ ‘[e]very

7 While the protections afforded under the due process clauses of the California Constitution and the federal Constitution are not coterminous (see People v. Ramos (1984) 37 Cal.3d 136, 151–154 [jury instruction found to violate state due process clause despite having been found to comport with federal due process clause]; People v. Buza (2018) 4 Cal.5th 658, 684), we have previously acknowledged that, as with the federal Constitution, the “essence” of our state due process clause is “fundamental[] fairn[ess in the] decision-making process.” (Ramos, at p. 153; cf. Lassiter v. Department of Social Services (1981) 452 U.S. 18, 24 [due process “expresses the requirement of ‘fundamental fairness’ ”].) Moreover, Rudd has failed to present any separate argument or analysis with respect to the federal and state due process clauses. Indeed, the portion of his brief addressing his due process claims cites almost exclusively to federal authorities.

21 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

witness is presumed to speak the truth,’ ” but clarified that the presumption “ ‘may be overcome’ ” by other evidence. (Id. at p. 142.) After being convicted, the defendant filed a habeas petition arguing that “the presumption-of-truthfulness charge [violated his due process rights by] shift[ing] the State’s burden to prove guilt beyond a reasonable doubt and forc[ing] [him] instead to prove his innocence.” (Id. at p. 143.) The Supreme Court rejected the claim. Although the court acknowledged that numerous federal circuits had disapproved similar instructions on the basis that they tended to “ ‘shift’ the prosecution’s burden of proof” (Cupp, supra, 414 U.S. at p. 145), it noted that none had found the instruction violated due process. (Id. at p. 146.) The court explained that the mere fact several circuits had deemed the instruction “[un]desirable from the viewpoint of sound judicial practice” was “not, without more, authority for declaring . . . the giving of the instruction . . . invalid [under the Due Process Clause].” (Ibid.) The court went on to hold that, viewed in the context of the record as a whole, the presumption-of-truthfulness instruction did not “impliedly” (id. at p. 148) shift the burden to defendant to prove his innocence. In support, the court emphasized that the language of the instruction did not compel the jury to accept the testimony of any witness and that other instructions had “explicit[ly] . . . affirm[ed]” that the prosecution had the “obligation . . . to prove guilt beyond a reasonable doubt.” (Id. at p. 147.) The instruction in Cupp was found not to violate due process despite having effectively directed the jury that the prosecution’s eyewitnesses were presumed to speak the truth. We therefore fail to see how a due process violation could be found in a case like this one, where the instruction merely

22 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

directed the jury that it should consider the eyewitness’s level of certainty as one of 15 enumerated factors and where the defendant was permitted to present expert testimony explaining that certainty is generally not predictive of accuracy. Rudd’s argument that the certainty instruction violated his due process rights by denying him “a ‘meaningful opportunity to present a complete defense’ ” as to why the identification was flawed fares no better. The record shows that Rudd was permitted to put on a vigorous defense on the issue of identity. As explained above, Rudd called an eyewitness identification expert who testified at length about the weak correlation between certainty and accuracy, particularly with respect to in-court identifications. Defense counsel emphasized that testimony at closing argument, explaining: “A lot of people think the more confident you are in the eyewitness identification, the more accurate it can be, but nothing could be farther from the truth. Just listen to Dr. Eisen’s testimony.” 8 In addition, Rudd had the opportunity to cross-examine Campusano and the investigating officers regarding her identifications and the procedures used during the photographic

8 Rudd argues that jury instructions are a more effective mechanism than expert testimony to educate the jury about the limited correlation between certainty and accuracy. As discussed in more detail below, there appears to be an emerging dispute in the research regarding the efficacy of enhanced jury instructions on eyewitness identifications. (See post, at pp. 36– 37.) But even if we assume jury instructions can be a more effective method than expert testimony alone, Rudd has cited no authority suggesting that the due process clause entitles a defendant to a jury instruction summarizing empirical research that can otherwise be presented through expert testimony.

23 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

lineups. On cross-examination, Campusano admitted she could not see Rudd when she made her in-court identification, explaining that “it was logical” the person who had committed the crime would be in the courtroom. Defense counsel also elicited numerous inconsistencies in other aspects of Campusano’s recollection of the crime, including her statements as to when she had regained consciousness, the extent of her injuries and when she had told law enforcement that the perpetrator had a tattoo. Defense counsel’s cross-examination of the investigating officers explored problematic aspects of the identification procedures. Officer Velasquez admitted that the first identification had occurred while Campusano was receiving treatment in the hospital. Detective Silva admitted that during the second identification, he showed Campusano the same photograph of Rudd that Velasquez had used at the first identification. Given the expert testimony and cross- examination that occurred in this case, we find no merit in Rudd’s claim that he was denied the opportunity to present a complete defense on the issue of identity. Although unable to cite any California or federal authority that has rejected the type of certainty instruction set forth in CALCRIM No. 315, Rudd argues that several state courts have done so. As discussed in more detail below, however, none of those jurisdictions rejected the instruction on due process grounds. (See post, at pp. 26–29.) Instead, each of those courts acted pursuant to their supervisory powers, concluding the instruction should be avoided or supplemented to avoid the possibility that jurors might wrongly assume there is generally

24 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

a strong correlation between certainty and accuracy. 9 While an enhanced or modified version of the certainty instruction might well be advisable (an issue we examine in more detail below), that alone does not establish a due process violation. (See Mills, supra, 55 Cal.4th at p. 677 [“ ‘not every ambiguity, inconsistency, or deficiency in a jury instruction rises to the level of a due process violation’ ”]; People v. Huggins (2006) 38 Cal.4th 175, 192 [“even if . . . the trial court’s instruction created ambiguity, it did not infringe on defendant’s due process rights”]; People v. Engelman (2002) 28 Cal.4th 436, 445 (Engelman) [although instruction regarding jury misconduct was “inadvisable and unnecessary” because it might mislead jurors, the instruction did not violate any constitutional right]; Cupp, supra, 414 U.S. at p. 147.)

9 The few cases Rudd cites that do actually address a due process claim have little relevance to the instructional error claim he raises here. In Neder v. United States (1999) 527 U.S. 1, 4 and People v. Mil (2012) 53 Cal.4th 400, 409–410, for example, the courts found a due process violation where the trial court’s charge to the jury had omitted an essential element of the crime. In Crane v. Kentucky (1986) 476 U.S. 683, the court found a due process violation where the trial court excluded evidence regarding the coercive circumstances under which the defendant’s confession was made. In Estelle, supra, 502 U.S. 62, the court held that the admission of evidence related to the defendant’s uncharged criminal conduct did not violate due process. Finally, in Clark v. Brown (9th Cir. 2006) 450 F.3d 898, 908, the Ninth Circuit found a due process violation where the trial court had refused to instruct the jury on the defendant’s theory that he lacked an independent felonious intent at the time he started a fire, which would have precluded a conviction on the charged offense. All these cases are far afield from the situation presented here.

25 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

In sum, when considered “ ‘in the context of the instructions as a whole and the trial record’ ” (Foster, supra, 50 Cal.4th at p. 1335, italics omitted), we conclude that listing the witness’s level of certainty as one of 15 factors the jury should consider when evaluating an eyewitness identification did not render Rudd’s trial fundamentally unfair or otherwise amount to a due process violation. B. Reevaluation of CALCRIM No. 315’s Instruction on Witness Certainty Amici curiae argue that even if CALCRIM No. 315’s instruction on witness certainty does not amount to a due process violation, we should utilize our supervisory powers to either “strike the certainty factor” or direct trial courts to provide an “enhanced jury charge” summarizing the empirical research regarding the correlation between certainty and accuracy. The Attorney General acknowledges that the absence of a due process violation “does not necessarily mean that a better instruction could not be devised” and that courts should “periodically examine procedures relating to eyewitness identification, to see whether they can be improved.” The Attorney General contends, however, that “the appropriate way forward” for any such changes “is through the procedure prescribed by the Rules of Court: making suggestions to the committee that advises the Judicial Council as to instructional ‘improve[ments],’ and circulating proposed changes for ‘public comment.’ (Cal. Rules of Court, rule 2.1050(d).)” Several jurisdictions have concluded that the type of witness certainty instruction at issue in this case is potentially misleading and can be improved upon. The Supreme Court of

26 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

New Jersey and the Massachusetts Supreme Judicial Court both modified their instructions on witness certainty after having convened special proceedings to assess the scientific evidence regarding eyewitness identifications. (See Henderson, supra, 27 A.3d 872 [implementing recommendations of a special master appointed to investigate empirical research on eyewitness identification evidence]; Gomes, supra, 22 N.E.3d 897 [implementing recommendations of a study group appointed to investigate identification procedures].) Both courts concluded that, contrary to common belief, empirical research has consistently shown that “ ‘under most circumstances, witness confidence or certainty is not a good indicator of identification accuracy.’ ” (Gomes, at p. 912; see Henderson, at p. 899 [“eyewitness confidence is generally an unreliable indicator of accuracy”].) The courts further found, however, that the correlation is stronger under some circumstances — most notably when the witness expressed high confidence at the initial identification and law enforcement utilized proper lineup procedures. (See Gomes, at p. 912; Henderson, at p. 899.) In light of those findings, both courts elected to incorporate aspects of the scientific research into their model instructions on an eyewitness’s level of certainty.10 Rather than

10 Both courts also adopted substantial modifications to other factors listed in their eyewitness testimony instructions, including, for example, whether the witness received information that may have influenced his or her recollection, whether the witness and suspect were of different races and whether the witness was under stress at the time he or she observed the event. (See Gomes, supra, 22 N.E.3d at pp. 911– 918; Henderson, supra, 27 A.3d at pp. 925–926.) In this case,

27 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

merely telling the jury to consider a witness’s level of certainty, the section of New Jersey’s model instruction on eyewitness identification addressing witness confidence now provides: “Confidence and Accuracy: You heard testimony that (insert name of witness) made a statement at the time he/she identified the defendant . . . concerning his/her level of certainty that the person/photograph he/she selected is in fact the person who committed the crime. . . . [A] witness’s level of confidence, standing alone, may not be an indication of the reliability of the identification. Although some research has found that highly confident witnesses are more likely to make accurate identifications, eyewitness confidence is generally an unreliable indicator of accuracy.” (New Jersey Courts, Model Criminal Jury Charges, Identification: In-Court and Out-of-Court Identifications (May 18, 2020) p. 8; see id. at p. 3, fns. omitted.) The section of Massachusetts’s model instruction that addresses witness certainty, in contrast, directs the jury as follows: “Expressed certainty. You may consider a witness’s identification even where the witness is not free from doubt regarding its accuracy. But you also should consider that an eyewitness’s expressed certainty in an identification, standing alone, may not be a reliable indicator of the accuracy of the identification, especially where the witness did not describe that level of certainty when the witness first made the identification.” (See Massachusetts Superior Court Criminal Practice Jury Instructions (Mass.Cont.Legal Ed. 3d ed. 2018) §

however, we have only been asked to consider CALCRIM No. 315’s language relating to the certainty factor.

28 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

6.2; Commonwealth v. German (Mass. 2019) 134 N.E.3d 542, 555; Gomes, supra, 22 N.E.3d at p. 923.)11 The high courts of Kansas and Georgia have taken a different approach, directing their trial courts to refrain from instructing on witness certainty altogether. (See State v. Mitchell (Kan. 2012) 275 P.3d 905, 912 (Mitchell) [instruction “encourages jurors to give more weight to identifications by a certain witness than an uncertain one and does nothing to inform the jury that certainty evidence may be unreliable”]; Brodes v. State (Ga. 2005) 614 S.E.2d 766, 771 [“In light of the scientifically-documented lack of correlation between a witness’s certainty in his or her identification . . . and the accuracy of that identification, . . . we can no longer endorse an instruction authorizing jurors to consider the witness’s certainty in his/her identification as a factor to be used in deciding the reliability of that identification”].)12

11 As Rudd notes in his briefing, the model instructions of other states include language that “alerts jurors to the possibility that certainty does not correlate with accuracy.” (See, e.g., Model Utah Jury Instructions (2d ed. 2014) No. CR 404 [“A witness’s level of confidence in (his) (her) identification of the perpetrator is one of many factors that you may consider in evaluating whether the witness correctly identified the perpetrator. However, a witness who is confident that (he) (she) correctly identified the perpetrator may be mistaken”]; State v. Hansley (Me. 2019) 203 A.3d 827, 831 [approving instruction directing that “there may not be a correlation between the reliability of an eye witness identification and the amount of certainty expressed by the witness in making that identification”].) 12 While not directly addressing the question of jury instructions, other state courts have rejected witness certainty

29 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

Consistent with the findings of those jurisdictions, the California Commission on the Fair Administration of Justice13 has recommended that the state judiciary reevaluate how juries are instructed on eyewitness testimony “in light of current scientific research regarding . . . the relevance of the degree of

as an appropriate factor to consider when assessing the admissibility of eyewitness testimony. (See Lawson, supra, 291 P.3d at pp. 745, 759, 777–778; Young v. State (Alaska 2016) 374 P.3d 395, 426–427; State v. Guzman (Utah 2006) 133 P.3d 363, 366.) Numerous federal courts have also acknowledged that empirical research has “undercut[] the hypothesis that there is a strong correlation between certainty and accuracy.” (Haliym v. Mitchell (6th Cir. 2007) 492 F.3d 680, 705, fn. 15; see U.S. v. Greene (4th Cir. 2013) 704 F.3d 298, 309, fn. 4 [“We observe that . . . (witness certainty) . . . has come under withering attack as not relevant to the reliability analysis”]; Young v. Conway (2d Cir. 2012) 698 F.3d 69, 88–89; U.S. v. Bartlett (7th Cir. 2009) 567 F.3d 901, 906; U.S. v. Brownlee (3d Cir. 2006) 454 F.3d 131, 142–144.) 13 The California Legislature established the California Commission on the Fair Administration of Justice to “study and review the administration of criminal justice in California to determine the extent to which that process has failed in the past, resulting in wrongful executions or the wrongful conviction of innocent persons”; “[t]o examine ways of providing safeguards and making improvements in the way the criminal justice system functions”; and “[t]o make any recommendations and proposals designed to further ensure that the application and administration of criminal justice in California is just, fair, and accurate[.]” (Sen. Res. No. 44 (2003–2004 Reg. Sess.).)

30 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

certainty expressed by witnesses in court.” (California Commission on the Fair Administration of Justice, Final Report (Jun. 30, 2008) at p. 11.)14 The commission has explained that the current model instruction — presumably a reference to CALCRIM No. 315 — “offers no guidance as to the potential significance, if any” (Final Report, at p. 32) of a witness’s expression of confidence and noted that several other jurisdictions have disapproved similar instructions. The commission has recommended that the Judicial Council’s Advisory Committee on Criminal Jury instruction look further into the issue. (Id. at p. 32.) To date, however, our judiciary has taken no action in response to the commission’s recommendation. Although our Legislature has not expressed any views regarding how jurors should be instructed on eyewitness testimony,15 it has taken other actions in response to the large

14 Available at (as of May 17, 2021). All Internet citations in this opinion are archived by year, docket number, and case name at . 15 In other circumstances, our Legislature has mandated the use of jury instructions to combat traditional assumptions that have been discredited by empirical research. (See People v. Jones (1990) 51 Cal.3d 294, 315 [statute mandating jury instruction that child testimony cannot be discounted based solely on the ground of youth was adopted in response to empirical studies that undermined “traditional notions regarding the unreliability of child witnesses”]; People v. Catley (2007) 148 Cal.App.4th 500, 507 [statute mandating instruction that juries cannot discount testimony based solely on a witness’s disability was adopted to combat “ ‘ “ ‘traditional assumptions’ ” [that] may previously have biased the factfinding process’ ”].)

31 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

body of research exploring how certain variables can affect the accuracy of an identification. Senate Bill 923, which went into effect on January 1, 2020, requires law enforcement agencies to adopt regulations mandating the use of specified procedures when administering “photo lineups and live lineups with eyewitnesses.” (§ 859.7, subd. (a); Stats. 2018, ch. 977, § 2.) The accompanying legislative findings explain that these newly mandated “best practices” are derived from a “large body of peer- reviewed research . . . demonstrat[ing] that simple systematic changes in the administration of eyewitness identification procedures by law enforcement agencies can greatly improve the accuracy of identifications.” (Stats. 2018, ch. 977, § 1, subd. (d).) The mandated procedures include, among other things, blind administration of the lineup and obtaining a statement regarding the witness’s level of confidence at the time of the identification. (See § 859.7.)16 These new requirements suggest the Legislature has accepted empirical findings that: (1) an

16 Rudd argues that several of the procedures Officer Velasquez and Detective Silva used during the photographic lineups, such as Silva’s use of the same image of Rudd that Campusano had selected during the first photographic lineup (see ante, at p. 5), conflict with the procedures mandated under section 859.7. (See, e.g., § 859.7, subd. (a)(2), (3) [requiring blind administration or written explanation why blind administration was not utilized], (9) [“Nothing shall be said to the eyewitness that might influence the eyewitness’ identification of the person suspected as the perpetrator”], (10)(A) [if the witness identifies a suspected perpetrator, “[t]he investigator shall immediately inquire as to the eyewitness’ confidence level in the accuracy of the identification and record in writing, verbatim, what the eyewitness says”], (11) [“An electronic recording shall be made . . . of the identification procedures”].) He acknowledges, however, that section 859.7 was not yet in effect when the officers conducted their investigation.

32 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

expression of certainty is less likely to be a reliable indicator of accuracy if the witness failed to express certainty at the initial identification; and (2) suggestive lineup procedures can have a substantial effect on the accuracy of an identification. Although CALCRIM No. 315’s instruction on witness certainty did not violate Rudd’s due process rights (see ante, at pp. 13–26), we now join other jurisdictions (and the California Commission on the Fair Administration of Justice) in acknowledging that this form of instruction has the potential to mislead jurors. There is near unanimity in the empirical research that “ ‘under most circumstances, witness confidence or certainty is not a good indicator of identification accuracy.’ ” (Gomes, supra, 22 N.E.3d at p. 912; see Henderson, supra, 27 A.3d at p. 899; Guilbert, 49 A.3d at p. 721; State v. Cabagbag (Hawaii 2012) 277 P.3d 1027, 1036.) The research has also consistently shown that “jurors . . . tend to overvalue the effect of . . . certainty . . . in determining the accuracy of eyewitness identifications.” (Lawson, supra, 291 P.3d at p. 705; see Gomes, at p. 913 [“it is necessary to inform a jury about th[e] tenuous relationship [between certainty and accuracy] because there is a near consensus that jurors tend to give more weight to a witness’s certainty in evaluating the accuracy of an identification than is warranted by the research”].) Indeed, many studies have “show[n] that eyewitness confidence is the single most influential factor in juror determinations regarding the accuracy of an eyewitness identification.” (Lawson, at p. 705 [citing studies]; see Wells & Bradfield, “Good You Identified the Suspect”: Feedback to Eyewitnesses Distorts Their Reports of the Witnessing Experience (1998) 83 J. Applied Psychol. 360, 361.) Although the language in CALCRIM No. 315 does not state that a certain identification is more likely to be accurate,

33 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

the instruction does nothing to disabuse jurors of the common misconception that such a correlation exists. Indeed, merely directing the jury to consider a witness’s level of certainty, without any further caveats, effectively operates to reinforce that misconception. (See Mitchell, supra, 275 P.3d at pp. 912– 913 [language “encourages jurors to give more weight to identifications by a certain witness” and “prompts the jury to conclude that an eyewitness identification is more reliable when the witness expresses greater certainty”].) That raises particular concerns in a case like this one, where the conviction was based almost entirely on the testimony of a single witness who expressed certainty in her identification and had no prior relationship with the defendant. (See McDonald, supra, 37 Cal.3d at p. 363 [“ ‘Centuries of experience in the administration of criminal justice have shown that convictions based solely on testimony that identifies a defendant previously unknown to the witness is highly suspect’ ”]; Sánchez, supra, 63 Cal.4th at p. 462 [“[a]ny reexamination of our previous holdings [regarding the witness certainty instruction] . . . should await a case involving only certain identifications”].) The risk of juror confusion is heightened by the structure of CALCRIM No. 315, which lists witness certainty among numerous other factors the jury should consider when assessing the eyewitness testimony. As written, the instruction implies that each of these factors have a direct, linear bearing on accuracy. For instance, “How well could the witness see the perpetrator” implicitly prompts the jury to believe that if the witness could see the perpetrator well, the identification should be given more weight, and vice versa; “How closely was the witness paying attention,” “Was the witness under stress when he or she made the observation,” “Did the witness ever fail to

34 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

identify the defendants,” all do the same. Hearing the certainty instruction in this context increases the risk that the jury will infer certainty operates the same way — as having some direct relationship with the accuracy of the identification. Having acknowledged the current version of the instruction might confuse jurors about the relationship between confidence and accuracy, that leaves the difficult question of determining what information trial courts should provide to the jury about witness certainty. While there is general agreement that witness certainty is not a good indicator of accuracy under most circumstances, that “does not mean that eyewitness certainty is never correlated with accuracy.” (Gomes, supra, 22 N.E.3d at p. 912.) Rather, as Justice Liu explained in his concurring opinion in Sánchez, 63 Cal.4th 411, the research suggests that “ ‘the strength of the confidence-accuracy relationship varies, as it depends on complex interactions among [numerous] factors.’ ” (Id. at p. 497 (conc. opn. of Liu, J.), quoting Nat. Research Council, Identifying the Culprit: Assessing Eyewitness Identification (2014) p. 108; see also Gomes, supra, 22 N.E.3d at p. 912 [“the existence and strength of the correlation depends on the circumstances”].) The large body of research conducted in this area has identified numerous factors that can affect the correlation between witness certainty and accuracy including (among other things): (1) whether the confidence statement occurred before or after the identification; (2) the temporal proximity between the event and the identification; (3) whether the witness provided an expression of certainty at the initial identification; (4) whether the witness was highly confident; (5) the use of suggestive identification procedures; and (6) information witnesses receive after the identification that might increase

35 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

their level of confidence. (See Gomes, supra, 22 N.E.3d at p. 912; Henderson, supra, 27 A.3d at pp. 896–900, 923, fn. 7; 2019 Report of The United States Court of Appeals for the Third Circuit Task Force on Eyewitness Identifications (2019) 92 Temp. L.Rev. 1, 53–54, 56, 99–100 (Third Circuit Task Force Report); Lawson, supra, 291 P.3d at p. 695; Guilbert, supra, 49 A.3d at pp. 722–723; ante, at pp. 8–9.) The relevance of the last two factors, in turn, requires further understanding of the type of law enforcement conduct that may be suggestive or confirmatory. (See Third Circuit Task Force, at pp. 53–55 [discussing types of potentially confirmatory or suggestive conduct].) Even among those states that have chosen to modify their instructions on witness certainty, there is no consensus as to what specific factors merit inclusion in the charge. New Jersey’s instruction, for example, notes that while “eyewitness confidence is generally an unreliable indicator of accuracy,” some research has shown that “highly confident witnesses are more likely to make accurate identifications . . . .” (New Jersey Courts, Model Criminal Jury Charges, Identification: In-Court and Out-of-Court Identifications, supra, at p. 8.) Massachusetts, on the other hand, emphasizes the importance of whether the witness expressed certainty at the initial identification. (Gomes, supra, 22 N.E.3d at p. 923.) Connecticut, in contrast, has approved the use of an enhanced jury charge on witness certainty, but has given its trial courts discretion to determine the content of such an instruction on a case-by-case basis. (Guilbert, supra, 49 A.3d at p. 727, fn. 27.) Complicating matters further, there appears to be an emerging dispute in the research over the efficacy of highly detailed instructions on eyewitness identifications. (See State v.

36 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

Booth-Harris (Iowa 2020) 942 N.W.2d 562, 578 [“a growing body of academic literature . . . questions the efficacy of certain provisions in such jury instructions on eyewitness identifications. In fact, recent studies have shown that the more comprehensive jury instructions like New Jersey’s Henderson instruction can actually overcorrect the problem”]; State v. Clopten (Utah 2009) 223 P.3d 1103, 1110–1111 [citing studies showing that enhanced instructions “do[] little to help a jury spot a mistaken identification” and are “less effective than expert testimony”]; Third Circuit Task Force Report, supra, 92 Temp. L.Rev. at p. 97 [task force minority view participants arguing that studies have found enhanced instructions “do not assist jurors in evaluating the evidence, but rather cause jurors to question all eyewitness identification testimony, thereby increasing the rate of acquittal regardless of the circumstances”]; Guilbert, supra, 49 A.3d at p. 726 [citing research finding that jury instructions are “less effective than expert testimony in apprising the jury of the potential unreliability of eyewitness identification testimony”].) Given the complexities described above, we agree with the Attorney General that the Judicial Council and its Advisory Committee on Criminal Jury Instructions, which is comprised of jurists, scholars and practitioners specializing in criminal law (see Cal. Rules Court, rule 10.59), are best suited to reevaluate whether or how CALCRIM No. 315’s instruction can be modified to remedy potential confusion regarding the correlation between certain and accuracy without “being unduly long or

37 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

argumentative.”17 (Wright, supra, 45 Cal.3d at p. 1143; see Mitchell v. Gonzales (1991) 54 Cal. 3d 1041, 1053, fn. 9 [directing the Committee on Standard Jury Instructions to consider whether model instruction defining proximate causation “could be improved” in light of empirical research showing that the current wording confused a substantial portion of jurors].) The goal of reevaluating the instruction should be to improve the jury’s ability to properly evaluate an eyewitness’s expression of certainty in an identification while avoiding any interference with the jury’s fact-finding function. (See Cal. Rules Court, rule 2.1050(a) [goal of “instructions is to improve the quality of jury decision making”]; Watson v. Damon (1880) 54 Cal. 278, 279 [instructions of court “cannot interfere with the exclusive prerogative of the jury in passing upon the facts”].) In assessing what modifications might be appropriate, the Judicial Council should remain mindful that while there is now general agreement in the research that witness certainty is not a good indicator of accuracy under most circumstances, research has also shown the correlation tends to be stronger when certain factors are present. (See ante, pp. 27–28, 35–36.) The Judicial Council should also consider research showing that highly detailed jury instructions may further confuse the jury or overcorrect the problem. (See ante, at pp. 36–37.)18

17 Referring this issue to the Judicial Council will also provide the public an opportunity to weigh in on any amendments the Council may propose. (See Cal. Rules of Court, rule 2.1050(d) [“Amendments to these instructions will be circulated for public comment before publication”].) 18 Given the complexity of the issue, the Judicial Council might also consider inviting its Criminal Law Advisory

38 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

As we have explained, Rudd has failed to establish that the trial court’s decision to include the certainty factor in CALCRIM No. 315 violated his due process rights or otherwise constituted error under the circumstances presented here. Nonetheless, for the reasons described, we believe there is a risk that the current version of the instruction will prompt jurors to infer that an eyewitness’s certainty in an identification is generally a reliable indicator of accuracy. Accordingly, in the exercise of our supervisory powers, we direct our trial courts to omit the certainty factor from CALCRIM No. 315 until the Judicial Council has the opportunity to consider how the language might be better worded to minimize juror confusion on this point. (See Engelman, supra, 28 Cal.4th at p. 444 [finding no error but exercising supervisory powers to disapprove potentially misleading instruction regarding juror misconduct]; People v. Burgener (2003) 29 Cal.4th 833, 861 [finding no error but exercising supervisory powers to disapprove “race- conscious” jury assignment procedure]; People v. Brigham (1979) 25 Cal.3d 283, 292 [exercising supervisory powers to disapprove “inartfully drawn” jury instruction that presented risk of misleading jurors about the meaning of the reasonable doubt standard].) Trial courts, however, retain discretion to include the factor when the defendant requests that it do so.19

Committee and the Appellate Advisory Committee to aid in evaluating how jurors should be instructed on the issue of witness certainty. (Cf. People v. Sivongxxay (2017) 3 Cal.5th 151, 169–170, fn. 4.) 19 We have previously distinguished the effect of the certainty instruction in cases where a witness has expressed doubt, rather than confidence, about the accuracy of the

39 PEOPLE v. LEMCKE Opinion of the Court by Groban, J.

III. DISPOSITION The judgment is affirmed.

GROBAN, J.

We Concur: CANTIL-SAKAUYE, C. J. CORRIGAN, J. LIU, J. CUÉLLAR, J. KRUGER, J. JENKINS, J.

identification. (See Sánchez, supra, 63 Cal.4th at p. 462 [“ ‘[certainty] instruction has merit in so far as it deals with the testimony of a witness who expressed doubt about the accuracy of her identification . . . .’ ”].) The misleading effect we are concerned with here — that the jury is prompted to believe there is a strong correlation between certainty and accuracy despite empirical research showing just the opposite — is not present when a witness has expressed doubt regarding the identification.

40 See next page for addresses and telephone numbers for counsel who argued in Supreme Court.

Name of Opinion People v. Lemcke __________________________________________________________________

Procedural Posture (see XX below) Original Appeal Original Proceeding Review Granted (published) Review Granted (unpublished) XX NP opn. filed 6/21/18 – 4th Dist., Div. 3 Rehearing Granted __________________________________________________________________

Opinion No. S250108 Date Filed: May 27, 2021 __________________________________________________________________

Court: Superior County: Orange Judge: David A. Hoffer

__________________________________________________________________

Counsel:

Sylvia W. Beckham, under appointment by the Supreme Court, for Defendant and Appellant Desirae Lee Lemcke.

Jeanine G. Strong, under appointment by the Supreme Court, for Defendant and Appellant Charles Henry Rudd.

Kramer Levin Naftalis & Frankel, Hannah Lee, David S. Frankel, John M. McNulty and Aaron L. Webman for The California Innocence Project, The Project for the Innocent at Loyola Law School and The Northern California Innocence Project as Amici Curiae on behalf of Defendant and Appellant Charles Henry Rudd.

Mary K. McComb, State Public Defender, Barry P. Helft, Chief Deputy State Public Defender, and Kathleen M. Scheidel, Assistant State Public Defender, as Amicus Curiae on behalf of Defendant and Appellant Charles Henry Rudd. Xavier Becerra, Attorney General, Michael J. Mongan, State Solicitor General, Janill L. Richards, Principal Deputy State Solicitor General, Julie L. Garland, Assistant Attorney General, Joshua A. Klein, Deputy State Solicitor General, Steve Oetting and Minh U. Le, Deputy Attorneys General, for Plaintiff and Respondent. Counsel who argued in Supreme Court (not intended for publication with opinion):

Jeanine Strong Strong Appellate Law 316 Mid Valley Center #102 Carmel, CA 93923 (831) 277-3904

Aaron Webman Kramer Levin Naftalis & Frankel LLP 1177 Avenue of the Americas New York, NY 10036 (212) 715-9100

Joshua A. Klein, Deputy State Solicitor General 1515 Clay St., Suite 2000 Oakland, CA 94612 (510) 879-0756