People v. Sotelo-Urena

Filed 10/26/16 CERTIFIED FOR PUBLICATION IN THE COURT OF APPEAL OF THE STATE OF CALIFORNIA FIRST APPELLATE DISTRICT DIVISION TWO THE PEOPLE, Plaintiff and Respondent, A144021 v. VLADIMIR SOTELO-URENA, (Sonoma County Super. Ct. No. SCR644374) Defendant and Appellant. In April 2016, George Lowery, a 50-year-old homeless man, was beaten to death in San Diego County. Two brothers were charged with his kidnapping, torture, and murder. That same month, the homeless 51-year-old John Gerald Holiday was murdered near a dumpster in downtown Fresno. The next month saw the murders of two more homeless individuals: 66-year-old Stephen Williams was found dead in a pond in Golden Gate Park. It was reported that two men, along with other assailants, had tortured Williams over a three-day period, eventually killing him and dumping his body in the pond. And 37-year-old Joshua William Clark died in Santa Rosa following a one-sided fight with another man over either cigarettes or a debt owed for marijuana. Both men were homeless and downtown Santa Rosa regulars. In July 2016, the police arrested a man suspected of committing a series of attacks in San Diego that left three homeless men dead and at least two others seriously injured. These tragedies in those few months are not anomalies—studies show that homeless individuals are the victims of crime at a significantly higher rate than housed individuals. 1 Here, on trial for the first degree murder of Nicholas Bloom, defendant Vladimir Sotelo-Urena—who was homeless—sought to introduce expert testimony to this effect, via an expert prepared to testify that as a result of this higher rate of victimization, homeless individuals experience a heightened sensitivity to perceived threats of violence. This evidence, defendant submitted, was relevant to his claim that he acted in self-defense or, at the very least, imperfect self-defense. The trial court excluded the testimony on the ground that it was irrelevant to defendant’s claim that he actually and reasonably believed he needed to use lethal force to defend himself—that homelessness was a “common experience with the jurors and not subject to expert testimony.” On appeal, defendant contends among other things, that the exclusion of the expert testimony was an abuse of discretion and deprived him of his constitutional right to present a complete defense, and that this error was prejudicial. We agree, and we reverse on that ground, without a need to address defendant’s other arguments. EVIDENCE AT TRIAL In December 2013, 31-year-old defendant had been homeless for over two years. He had been living in Santa Rosa for five or six months, having moved from Las Vegas to attend Santa Rosa Junior College. While in Santa Rosa, he frequently spent the night downtown behind the Santa Rosa public library. Nicholas Bloom was also homeless and camped in downtown Santa Rosa. Around 8:30 or 9:00 p.m. on December 24, Donovan Sweeden was in a downtown Santa Rosa homeless encampment that he shared with Bloom. He watched Bloom inject “a fat one”—an 80cc syringe of methamphetamine—described by a medical expert as a dosage large enough to kill someone who was not a regular user habituated to the drug. After shooting up, Bloom became “agitated,” “upset,” and “aggressive.” Sweeden described him as “hot. . . . [S]eemed to be somewhat out of it,” and said he “could be perceived” as violent and “explosive.” Around 10:00 p.m., Bloom left the encampment and walked down Jeju Way, an alley that runs behind the Russian River Brewing Company and the Santa Rosa public library. He soon encountered defendant, an encounter that ended when defendant 2 admittedly stabbed Bloom to death. Because defendant did not testify at trial, his version of what happened during that encounter came from two interviews with Santa Rosa police officers after his arrest. The recordings of those interviews were played for the jury, and were in substance as follows: Defendant generally camped in the back of the library, most nights sleeping “down at the bottom level.” Defendant, who was Muslim, was sitting alone on the rear steps of the library reading the Quran when Bloom approached him from the direction of the Russian River Brewing Company. From Bloom’s mannerisms, defendant believed he was “probably” intoxicated. With an aggressive demeanor, Bloom asked if defendant had a cigarette. Defendant did not respond and continued to read. With an angry voice, Bloom asked if he spoke English, and defendant responded that he did. Bloom again asked if he had a cigarette, and defendant answered that he did not. Bloom moved even closer with a belligerent attitude, as if he wanted to fight. Defendant had been stabbed a few weeks earlier, and he was “pretty sure” he recognized Bloom as one of his attackers: “[H]e was one of the people that, one of the individuals that, you know, tried to harm me, tried to kill me in the past . . . .”1 Bloom approached “like he’s gonna do something,” and defendant thought, “[O]h shit, another fight.” Bloom “went to grab something” from his pocket or waistband, and defendant believed he was grabbing a knife: “I didn’t see a knife but I’m, but I’m pretty sure he had something in his hand ‘cause he kept like, like he wanted, you know, and like I was getting stabbed again, essentially.” Defendant did not have good eyesight, but he thought he saw “something thin” with a “linear line” to it, and he believed Bloom was “gonna poke [him] or stab [him] or something.” As Bloom got closer, defendant felt he was in danger because he had been through this before: “I just felt like I was in a position of danger and it was, and it got my nerves 1 It was stipulated that Bloom was in fact in custody on November 11, the day defendant was stabbed. 3 rattled, ya know? My adrenaline just jumped up.” He grabbed his backpack and pulled out a kitchen knife he had bought for protection after he was stabbed. At that point, Bloom was “[p]retty close.” Defendant told Bloom to get away or he was “gonna send [him] straight to hell.” Bloom said, “oh really?” and laughed like he wanted to hurt defendant. They then “got tangling” and “just started going at it.” They “kinda walked . . . kinda waltzed” down to the end of the street, where defendant grabbed Bloom’s shirt and went for his throat, stabbing him. Bloom fell to the ground on his back, and defendant “got on top of him [and] made sure that he didn’t get up,” “made sure that he didn’t pose a threat . . . .” Defendant explained: “I wasn’t gonna wait for him to get stabbed. Last time it happened is because I waited. And because ya know, I let, you know, him get the best of me, you know, and I wasn’t gonna do that a second time.” Defendant continued to stab Bloom after he was down because he “had to make sure he didn’t . . . move.” After it was over, defendant was “pissed,” and he kicked Bloom in the head. According to defendant, the altercation started at the library steps and ended almost at the end of the street. He did not know how they got that far, although they were “lunging at each other . . . .” When asked if he knew how many times he had stabbed Bloom, defendant responded, a “Bunch of times,” and “A lot.” “Enough, enough to mess somebody up permanently.” When asked if he was trying to kill Bloom, defendant answered, “Essentially, yes. I wasn’t tryin’ to tickle him,” and “Obviously. I’m not gonna lie, yes.” After stabbing Bloom, defendant walked back to the steps, put the knife in his backpack, and “just sat there.” He felt “sick,” “sad,” “kinda weak,” and “disgusted.” He then waited for the police to arrive: “I mean, I wasn’t gonna run. I wasn’t gonna be like well, I didn’t do it. ‘Cause, I mean, I’m not gonna lie, I’m not supposed to lie about anything.” Defendant denied that he tried to rob Bloom, that Bloom asked for money, or that they had been engaged in any kind of transaction. Bloom had not solicited a sexual encounter with defendant, and defendant denied having a sexual relationship with him. 4 He denied that he had wanted payback for having been stabbed a few weeks earlier. Defendant claimed he did not know anything about money that was found on the ground and in Bloom’s hands. Defendant also told the police officers about another incident that occurred about six days before he was stabbed. It was a verbal altercation with someone else, and he did not know why but “they wanna hurt me. They wanna kill me.” Jeffrey Jones, a passerby, testified at defendant’s trial. Around 10:30 p.m. on December 24, 2013, Jones left the Russian River Brewing Company and walked through Jeju Way on his way home. He passed by defendant, who was sitting on the library steps 10 to 12 feet away. As he continued walking, he noticed a crumpled $20 bill on the ground. About 30 steps away in a very dark area of the alley, he saw a body on the ground and blood everywhere. Next to the body, there was a wallet and four crumpled $20 bills. In Bloom’s hand, there were crumpled $20 bills and a tube of lip balm. Jones immediately called 911. While he was on the phone, he turned back in the direction of the steps and saw defendant still sitting there with a bag next to him. He walked back towards defendant, who stood up, picked up his bag, and then sat back down again. Jones stopped about 60 feet away from defendant and called out to ask if he had seen anything. The police arrived almost immediately. When they arrived, Jones pointed out defendant sitting on the steps. Defendant was arrested shortly thereafter. An officer on the scene asked him what had happened, and he responded, “I got them before they got me.” Another officer asked defendant if he was okay,2 and he answered, “No, he was trying to kill me.” Asked where the knife was, defendant said it was in his backpack. A large kitchen knife covered in blood was found in defendant’s backpack. Raymond Hoey, who was nearby at the time of Bloom’s stabbing and overheard part of the assault, also testified at trial. Hoey was homeless and generally slept in a 2 Defendant suffered a cut on his hand from his own knife. 5 stairwell behind the public library, a spot he chose because it was below eye level, it seemed less likely that anybody would see him down there, and for safety reasons, to avoid being seen while bedding down for the night. He had been settling in to sleep when he heard what sounded like gears of a 10-speed bike and then scuffling in the parking lot area above him. He then heard two distinct voices. The first one made a guttural noise, “almost a growl,” that sounded “very malicious and malevolent in nature,” “animalistic,” “as if someone was not well disposed towards another.” He heard a second voice yell, “Stop, stop,” and then moments later, “Help, help,” as if in distress. After the verbal exchange, it was quiet for a few moments and then the police arrived. Because he did not know what the situation was, he stayed put until the police found him. In a police interview, Hoey said he first heard two people arguing and that it immediately got heated and one of the voices started violently yelling. When asked at trial to elaborate, he testified, “What I heard sounded like—like two subjects disagreeing in the form of what I interpret to be a scuffle, such as when two males push each other or threaten one another in some physical way.” He described the first voice—the one that made the guttural noises—as the more aggressive of the two. Bloom’s body was found 135 feet from where defendant had been sitting on the steps. He suffered 70 to 80 sharp force injuries that could have been inflicted by a knife. He had eight stab wounds to his chest and abdomen, which caused severe internal injuries. The majority of the knife wounds were to his neck area, causing a 9.75 inch gaping wound. The primary cause of death was multiple sharp force injuries to his torso. Additional contributing factors were sharp force injuries to his neck, head, and extremities, as well as acute methamphetamine intoxication. Bloom’s body showed signs of intravenous drug use. His blood and urine tested positive for methamphetamine, amphetamine, THC, and opioids. Defendant’s blood sample was devoid of any intoxicants. 6 PROCEDURAL BACKGROUND Defendant was charged with Bloom’s first degree murder. (Pen. Code, § 187, subd. (a).) It was also specially alleged that he personally used a knife (id., § 12022, subd. (b)(1)) and personally inflicted great bodily injury (id., § 12022.7). On September 30, 2014, a jury convicted defendant of first degree murder and found the two special allegations to be true. Defendant was sentenced to 26 years to life, comprised of 25 years to life for the murder conviction plus a consecutive, one-year term for the personal use of a knife enhancement. The enhancement for inflicting great bodily injury was stayed pursuant to Penal Code section 654. Defendant filed a timely notice of appeal. DISCUSSION 1. Defendant’s Arguments Defendant asserts three arguments on appeal: (1) the trial court abused its discretion and deprived him of his constitutional right to present a complete defense by excluding expert testimony concerning chronic homelessness; (2) there was insufficient evidence to support the jury’s first degree murder verdict because there was no substantial evidence of premeditation and deliberation; and (3) the trial court erred in enhancing defendant’s sentence pursuant to Penal Code section 12022.7. For the reasons detailed below, we conclude defendant’s first argument is well taken. Because we reverse on that ground, we need not address his two remaining contentions. 2. Background Defendant’s defense to the charge that he murdered Bloom was that he believed he needed to use lethal force to defend himself from an attack by Bloom because Bloom had (defendant incorrectly believed) stabbed him a few weeks earlier and was going to do it again if defendant did not get him first. In other words, he acted in self-defense or, alternatively, imperfect self-defense. In support, defendant’s motion in limine no. 32 sought to admit the testimony of the Honorable Robert C. Coates (Retired) as an expert 7 witness on “Homelessness, and Substance Abuse.”3 More particularly, he proposed that Judge Coates would testify on the following two topics: “The Experience of Homelessness, specifically as it relates to different classifications of homelessness, as well as recent data and studies showing that, nationwide and in the County of Sonoma, homeless individuals experience higher incidences of victimization and violence than do non-homeless persons,” and “[H]ow a person experiencing homelessness would react in a hypothetical situation tracking the defense version of the facts of this case, as shown by the evidence.” Judge Coates’s testimony, defendant submitted, was relevant to his defense because it would demonstrate why he actually believed he needed to use lethal force to defend himself from Bloom. Further, he argued, it would be relevant to the jury’s assessment of his credibility. In a supporting memorandum, defendant identified the following topics as potential subjects of Judge Coates’s testimony: empirical studies indicating that homeless people are exposed to violence at a substantially higher rate than housed individuals, and that the risk of violence has increased over recent years due to a number of factors, including the release of approximately 37,000 individuals with felony convictions on parolee status, many of whom have become homeless; new and more dangerous drugs that cause the users (many of whom are homeless) to become aggressive and violent; and veterans returning from service with post-traumatic stress disorder joining the homeless populations. His expertise included research on violence experienced by homeless individuals in the City of San Diego and the County of Sonoma, and he also relied on national and international studies addressing fear among and victimization of homeless individuals. Based on this research, Judge Coates was prepared to testify that the vulnerability to violence experienced by homeless people tends to create a greater than normal sensitivity to perceived threats of violence. 3 In his opening brief, defendant represents that Judge Coates served as a judge on the San Diego County Superior Court from 1982 until his retirement in 2011. He then resumed his law practice and subsequently joined the San Diego law firm of Olins, Riviere, Coates & Bagula. 8 The hearing on defendant’s motion began with the trial court expressing its “initial thoughts” that “what homeless people do or what they’re capable of doing or factors that may affect them” is “a common experience with the jurors and not subject to expert testimony.” In response, defense counsel argued that Judge Coates’s expert testimony concerning homelessness was relevant to the issue of defendant’s subjective belief that he was in danger and needed to use lethal force to defend himself, which went to his claim that he acted in imperfect self-defense. The prosecutor disagreed, arguing that the evidence was irrelevant to the charges and had the potential to lead the jury astray. The court stated that the issue was not whether defendant was homeless, but rather “what risk did he face that anybody would face sitting behind the library on Christmas eve at night?” And, in the court’s opinion, “Homelessness has nothing to do with his possible victimization.” Defense counsel countered, “Judge Coates has been intimately involved with the issue of homelessness as well as substance abuse issues as they affect the homeless and—to the point where he wrote a book on it and was involved in several boards down in San Diego and has developed opinions about this particular demographic and how it has changed and gotten more dangerous for people on the street and the sort of context of fear that exists out there among the homeless due to the increasing violence inflicted on the homeless by persons unknown, just for the fact that they’re homeless, and the evidence of growing numbers of homeless vets with PTSD issues, other mental disorders and, again, substance abuse problems, particularly methamphetamine, being ever more rampant on the streets, adding to the—the culture of fear and dangerousness.” The court was unpersuaded, observing, “Everyone is subject to the same risks. I don’t think it’s a subject for expert opinion, and I don’t think this judge has particular knowledge that’s relevant to the issues in this case. I am not going to allow the testimony.” Defense counsel queried whether that rationale also applied to defendant’s assertion of imperfect self-defense, “where the relevance, it would seem, to his subjective 9 beliefs is apparent?” The court agreed that defendant’s “state of mind is relevant” but believed that “some general opinion of a judge who’s dealt with homelessness” was not, explaining: “You know, we all deal with homelessness and substance abuse at one time or another in our work as judges. Perhaps all of us don’t write books on it, but I don’t think there’s specialized knowledge here that is relevant to specific issues in the case. You know, if you can point out something specific that the judge was going to testify to that was relevant, I would consider it. But his generalized testimony that you’ve outlined, and possible testimony, you know, I think is—it’s not directly applicable to this community. It’s his experiences in another community. “For instance, the number of homeless people that have drug abuse problems, you know, that’s not the population that was possibly a threat to Mr. Sotello. It’s the number of criminals that are on the streets in Santa Rosa after dark.” The court then ruled that it was not going to permit Judge Coates’s testimony. 3. The Trial Court Abused Its Discretion in Excluding Judge Coates’s Expert Opinion Concerning Chronic Homelessness A. The Law of Self-Defense and Imperfect Self-Defense A homicide is considered justified as self-defense where the defendant actually and reasonably believed the use of deadly force was necessary to defend himself from imminent threat of death or great bodily injury. Under such circumstances, the killing is not a crime. (People v. Elmore (2014) 59 Cal.4th 121, 133–134; Pen. Code, § 197; 1 Witkin & Epstein, Cal. Criminal Law (4th ed. 2012) Defenses, § 67 et seq., p. 507 et seq.) Where the defendant kills while actually but unreasonably believing the use of deadly force was necessary, defendant is considered to have acted in imperfect self-defense. Imperfect self-defense is not a complete defense to a killing, but negates the malice element and reduces the offense to voluntary manslaughter. (People v. Elmore, supra, 59 Cal.4th at p. 134; People v. Blakely (2000) 23 Cal.4th 82, 88; People v. Flannel (1979) 25 Cal.3d 668, 672.) “The subjective elements of self-defense and imperfect 10 self-defense are identical. Under each theory, the [defendant] must actually believe in the need to defend . . . against imminent peril to life or great bodily injury.” (People v. Viramontes (2001) 93 Cal.App.4th 1256, 1262.) As the California Supreme Court summarized it in People v. Humphrey (1996) 13 Cal.4th 1073, 1082 (Humphrey) (fn. omitted): “For killing to be in self-defense, the defendant must actually and reasonably believe in the need to defend. [Citation.] If the belief subjectively exists but is objectively unreasonable, there is ‘imperfect self-defense,’ i.e., ‘the defendant is deemed to have acted without malice and cannot be convicted of murder,’ but can be convicted of manslaughter. [Citation.] To constitute ‘perfect self-defense,’ i.e., to exonerate the person completely, the belief must also be objectively reasonable. [Citations.]” As this well-settled law applied at defendant’s trial, if the jury found that defendant actually and reasonably believed he needed to use lethal force to defend himself from Bloom, then he acted in self-defense, a complete defense to the murder charge. If it found he actually but unreasonably believed he needed to use lethal force, then he lacked the malice required for a murder finding and was guilty only of voluntary manslaughter. But if the jury rejected defendant’s claim that he actually believed he needed to use lethal force to defend himself from Bloom, then a murder conviction would result, as it did. Defendant’s actual belief in the need to use lethal force, and the reasonableness of that belief, were thus squarely at issue in his trial. The trial court ruled that Judge Coates’s expert opinion was inadmissible on these questions for two primary reasons. First, the court did not believe the evidence was relevant and, second, it did not believe the subject matter was sufficiently beyond common experience to warrant expert testimony. We review this ruling for abuse of discretion. (People v. Lucas (2014) 60 Cal.4th 153, 226; People v. Manriquez (1999) 72 Cal.App.4th 1486, 1492.) And conclude there was such abuse here. 11 B. Judge Coates’s Expert Opinion Was Relevant to Defendant’s Actual Belief in the Need to Use Lethal Force to Defend Himself Evidence is relevant if it has “any tendency in reason to prove or disprove any disputed fact that is of consequence to the determination of the action.” (Evid. Code, § 210.) Under Evidence Code section 351, all relevant evidence is admissible unless specifically excluded by statute. (People v. Riggs (2008) 44 Cal.4th 248, 289–290, citing Evid. Code, § 351.) A defendant claiming self-defense or imperfect self-defense is required to “prove his own frame of mind.” (People v. Davis (1965) 63 Cal.2d 648, 656; see also People v. Minifie (1996) 13 Cal.4th 1055, 1065 [“ ‘The defendant’s perceptions are at issue . . . .’ ”].) Because defendant’s frame of mind was at issue, the jury in defendant’s trial was expressly instructed that “In evaluating the defendant’s beliefs, consider all the circumstances as they were known and appeared to the defendant.” (CALCRIM No. 571.) In other words, the jury was to evaluate defendant’s belief in the need to use lethal force from his perspective. Evidence that would assist the jury in evaluating the situation from defendant’s perspective was thus relevant. Defendant’s perspective was that of a man who had been living for years on city streets. He had been stabbed a few weeks earlier (as corroborated by medical records), and he had also been involved in a verbal altercation just days before his encounter with Bloom. According to defendant’s account, Bloom approached him in an aggressive manner, was hostile when asking defendant if he spoke English or had a cigarette, and continued to belligerently approach while he sat on the steps. Defendant also told the police that when Bloom approached him, he believed Bloom was one of the assailants that had stabbed him and that he “was in a position of danger” because Bloom was going to do it again. Judge Coates was prepared to testify that individuals who are chronically homeless, like defendant, are subjected to a high rate of violence by both housed and unhoused individuals, and that the experience of living for years on the streets instills a perpetual fear of violence that would have affected defendant’s belief in the need to defend himself with lethal force. The relevance of this testimony to defendant’s actual 12 perception of the situation is evident: it would have helped the jury understand the situation from defendant’s perspective, that is, from the perspective of a chronically homeless man who had recently been violently assaulted and was aggressively approached by someone he believed to be the assailant. It would have explained his heightened sensitivity to aggression and why he was inclined to react more acutely to the perceived threat. The relevance of expert testimony to show a defendant’s perception of a threat of imminent harm has long been recognized in a different context that we find instructive here—cases involving intimate partner battering.4 The first case to recognize the admissibility of such expert opinion was People v. Aris (1989) 215 Cal.App.3d 1178 (Aris). There, defendant was found guilty of second degree murder after she shot and killed her abusive husband while he slept. The trial court excluded expert testimony that would have informed the jury that defendant was a victim of domestic violence and explained “ ‘how the psychological impact of being a battered woman affected her perception of danger at the time she shot her husband.’ ” (Id. at p. 1193.) The Court of Appeal reversed, holding that the expert testimony was “highly relevant to . . . defendant’s actual, subjective perception that she was in danger and that she had to kill her husband to avoid that danger.” (Aris, supra, 215 Cal.App.3d at p. 1197.) The expert would have testified that one of the symptoms of intimate partner battering was “a greater sensitivity to danger. ‘[T]hey don’t misperceive it; they perceive it very honestly, but it’s faster than somebody who [has not been] battered. And we call that a hypervigilance to cues of any kind of impending violence. That makes them . . . just a little bit more edgy, a little bit more responsive to situations than somebody who has not been battered might be.’ ” (Id. at p. 1194.) And, the court explained, “The relevance to the defendant’s actual perception lies in the opinion’s 4 Previously referred to as “battered women’s syndrome” (see, e.g., Humphrey, supra, 13 Cal.4th at p. 1076), the preferred terminology is now “intimate partner battering.” (Evid. Code, § 1107, subd. (f); People v. Wright (2015) 242 Cal.App.4th 1461, 1492, fn. 11.) 13 explanation of how such a perception would reasonably follow from the defendant’s experience as a battered woman.” (Id. at p. 1197; see also Humphrey, supra, 13 Cal.4th at pp. 1085–1087, 1096 (conc. opn. of Brown, J.) [“hypervigiliance generated by” repeated physical abuse]; In re Walker (2007) 147 Cal.App.4th 533, 553; People v. Day (1992) 2 Cal.App.4th 405, 419–420; Paine v. Massie (10th Cir. 2003) 339 F.3d 1194, 1199; Ibn-Tamas v. United States (D.C. 1979) 407 A.2d 626, 634.) The same rationale applies here. According to Judge Coates, a homeless individual who has repeatedly been subjected to violence and the threat of violence will experience a heightened sensitivity to such threats and will have a reduced threshold at which he or she subjectively perceives an imminent threat. As in Aris, supra, 215 Cal.App.3d at p. 1197, this opinion would elucidate defendant’s perception of the threat Bloom posed in the darkened alley. In excluding Judge Coates’s testimony as irrelevant to defendant’s actual perception of the situation, the court rejected the fundamental premise that homeless individuals are subjected to a higher rate of violence, as evidenced by its statements that “Homelessness has nothing to do with [defendant’s] possible victimization” and “Everyone is subject to the same risks.” Studies on chronic homelessness demonstrate that the court’s statements were incorrect: as Judge Coates proposed to testify, homeless individuals are victims of violent crime at a much higher rate than the general population. (See, e.g., In re Eichorn (1998) 69 Cal.App.4th 382, 386 [“Homeless individuals were 10 times as likely to be victimized by crime than the average population.”]) The National Coalition for the Homeless (NCH) conducts an annual survey tracking violent attacks on homeless individuals by housed individuals. In its June 2014 report entitled Vulnerable to Hate: A Survey of Hate Crimes & Violence Committed Against Homeless People in 2013 (Vulnerable to Hate), NCH reported that from 1999 to 2013, there were 1,437 documented acts of serious violence against homeless individuals by housed perpetrators, 375 of which were fatal. (Id. at p. 4.) In 2013 alone, NCH documented 109 attacks, 18 of which resulted in death. (Id. at p. 6.) That year, 30 percent of the attacks occurred in California. (Id. at p. 8.) The attacks included 14 unprovoked shootings, stabbings, and beatings with fists, sticks, tools, bottles, and other deadly weapons, and sometimes involved torture. (Id. at pp. 10–21.) The data depicted a pattern of violence whereby the offender arbitrarily—but intentionally—chose the victim because he or she was homeless.5 NCH also reported on a 2010 survey designed and administered by the National Consumer Advisory Board of the National Health Care for the Homeless Council. According to NCH, in that survey, “516 individuals experiencing homelessness over the age of 18, located in Detroit, Fort Lauderdale, Nashville, Houston, and Worchester, experienced violence 25 times more frequently than the general U.S. population. While 49 percent of homeless individuals report being victims of violence, only 2 percent of the general population does the same.” (Vulnerable to Hate, supra, p. 22.) While the NCH report addresses only attacks on homeless individuals by housed individuals, there is also a high rate of homeless-on-homeless crime, likely due to proximity, vulnerability, and untreated mental illness and substance abuse. In one article on the subject, the author reported: “Despite pressures toward under-reporting (due to embarrassment, an inability to document incidents, and the like), over one-half of all homeless [survey] respondents say that they have been victims of crime, primarily theft but also beatings and sexual assault. Results from other studies demonstrate substantial victimization rates for homeless women, youth, seniors, and shelter occupants. . . . [¶] Homeless persons are easy marks for domiciled predators and unscrupulous business operators (e.g., labor contractors who withhold pay, liquor store clerks who overcharge), but they also victimize each other. Close physical proximity, limited guardianship, 5 Because its survey in part relies on self-reporting, NCH believes that its results understate the violence inflicted on homeless individuals by housed individuals. (See Vulnerable to Hate, p. 5 [data is collected from published news reports, service providers, and self-reported incidents], p. 6 [“While this report provides alarming statistics, it is important to note that people experiencing homelessness are often treated so poorly by society that attacks are forgotten or unreported”; “many violent acts against homeless populations go unreported and therefore, the true number of incidents is likely to be substantially higher”].) 15 retaliation, pre-emptive displays of ‘toughness,’ and a low probability of sanctions are conducive to homeless-on-homeless crime. In general, street and shelter settings give rise to a vicious cycle in which some homeless people alternate between victim and offender roles.” (Barrett A. Lee, et al., The New Homelessness Revisited, Ann. Rev. Sociology, Vol. 36 (Aug. 1, 2010) pp. 506–507.) This empirical evidence is consistent with Judge Coates’s proffered testimony. And contrary to the trial court’s concern that his “generalized testimony” was not “directly applicable to this community,” Judge Coates’s expertise included Sonoma County, the homeless population of which both defendant and Bloom were members. In fact, the City of Santa Rosa, the very city in which defendant killed Bloom, declared an emergency of homelessness in August 2016, indicative of the extent of the homeless crisis in that city. The People counter that the exclusion of Judge Coates’s expert testimony did not preclude defendant from introducing evidence that would help the jury consider the relevant circumstances from defendant’s perspective. As the People put it: “The jury was made aware that appellant had been attacked in the past. Indeed, according to appellant he had even purchased the murder weapon—an actual kitchen knife that was figuratively a double-edged sword—to defend himself from future attacks. And the jury was allowed to hear appellant’s statements attempting to justify his actions and attempting to explain his state of mind.” But it has long been recognized that the introduction of the defendant’s narrative does not necessarily satisfy the defendant’s right to introduce relevant evidence, illustrated, for example, by People v. Smith (1907) 151 Cal. 619 (Smith), where the trial court excluded testimony of a doctor whom defendant called to testify about his physical condition on the day defendant allegedly committed a murder. Rejecting the People’s argument on appeal that defendant was not prejudiced because he testified himself about the subject, the Supreme Court observed that defendant was entitled to not only have the jury consider his own words “on the subject, but to introduce [expert testimony] to corroborate” his own narrative. (Id. at 16 p. 629.) This would permit him to “present the matter to the jury fully and under the most favorable circumstances.” (Ibid.) Likewise here. The People also cite two cases—United States v. Abdush-Shakur (10th Cir. 2006) 465 F.3d 458 (Abdush-Shakur) and People v. Romero (1999) 69 Cal.App.4th 846 (Romero)—in an attempt to persuade us that expert opinion on chronic homelessness was irrelevant to defendant’s self-defense and imperfect self-defense claims. Neither case is availing. In Abdush-Shakur—described by the People as “on point”—defendant was a prison inmate on trial for attempted murder of a prison employee. He unsuccessfully sought to introduce the expert testimony of a “ ‘corrections consultant,’ ” who would have testified to the “ ‘culture of violence’ in federal penitentiaries [to] explain why an inmate who is ‘disrespected’ by a corrections officer might retaliate in a violent manner.” Defendant challenged the exclusion of the testimony on appeal, arguing that the evidence went directly to his motive for attacking the victim and supported his claim that he only intended to wound him. (Abdush-Shakur, supra, 465 F.3d at p. 466.) The Court of Appeals affirmed, agreeing that the evidence was inadmissible. It noted that the proffered testimony would not excuse defendant’s attack on a corrections officer nor would it negate any elements of the charged crime. Rather, it simply highlighted a possible motive for his action. (Abdush-Shakur, supra, 465 F.3d at pp. 466–467.) In contrast, Judge Coates’s testimony, if persuasive to the jury, may have supported defendant’s self-defense claim or negated the element of malice. The second case, Romero, supra, 69 Cal.App.4th 846, is no more helpful. Romero involved a defendant who was charged with the murder of a car driver after a minor traffic incident resulted in a verbal exchange that escalated into lethal street violence. Defendant sought to introduce expert testimony on “the sociology of poverty, and the role of honor, paternalism, and street fighters in the Hispanic culture.” (Id. at p. 848.) More specifically, the expert would have testified that “(1) street fighters have a special understanding of what is expected of them; (2) for a street fighter in the Hispanic culture, there is no retreat; (3) the Hispanic culture is based on honor, and honor defines a person; and (4) in this culture a person ‘would be responsible to take care of someone,’ i.e., 17 defendant had a strong motivation to protect his younger brother.” (Id. at p. 853.) The trial court excluded the evidence, finding it irrelevant to whether defendant actually believed he was in imminent danger of death or great bodily injury, and whether such a belief was objectively reasonable. (Id. at p. 848.) The Court of Appeal affirmed. It held that the proffered testimony was irrelevant to whether deadly force was warranted under the circumstances. (Romero, supra, 69 Cal.App.4th at pp. 854–855.) Specifically, “whether a person should or should not retreat from a ‘street fight,’ has no bearing on whether that person may lawfully use deadly force.” (Id. at p. 854.) In other words, the evidence merely described defendant’s motive and was irrelevant to his self-defense claim. That is not the situation here. In short, we conclude that Judge Coates’s expert testimony on chronic homelessness was relevant to the issue of defendant’s actual belief in the need to use lethal force to defend himself from Bloom. C. Judge Coates’s Expert Opinion Was Relevant to the Reasonableness of Defendant’s Belief in the Need to Use Lethal Force to Defend Himself On appeal, defendant represents that he did not argue self-defense below, claiming that he only asserted imperfect self-defense and conceding here that his “statements to the police certainly did not describe a reasonable response to Bloom’s belligerent approach . . . .” We read the record differently: while defendant’s primary theory was imperfect self-defense, his counsel also argued perfect self-defense. And the jury was instructed on both theories. Because defendant could assert both theories on remand, we address the admissibility of Judge Coates’s expert testimony on the reasonableness of defendant’s belief in the need to use lethal force to defend himself from Bloom. And conclude expert opinion was also relevant to the reasonableness of his belief. As discussed above, Aris, supra, 215 Cal.App.3d 1178 held that expert testimony on intimate partner battering was relevant to the defendant’s actual belief in the need to use lethal force to defend herself from an imminent threat. (Id. at p. 1197.) It also held, however, that “upon request whenever the jury is instructed on perfect self-defense, trial 18 courts should instruct that [expert] testimony is relevant only to prove the honest belief requirement for both perfect and imperfect self-defense, not to prove the reasonableness requirement for perfect self-defense.” (Id. at p. 1199.) However, in Humphrey, supra, 13 Cal.4th 1073, the Supreme Court overruled Aris on this point, holding “that evidence of battered women’s syndrome is generally relevant to the reasonableness, as well as the subjective existence, of defendant’s belief in the need to defend, and, to the extent it is relevant, the jury may consider it in deciding both questions.” (Humphrey, at pp. 1088–1089.) In other words, where defendant claimed self-defense, “ ‘expert testimony on battering and its effects’ ” was relevant not only to whether the defendant actually believed in the need to defend herself from imminent harm, but also to whether that belief was objectively reasonable. (Id. at pp. 1083, fn. 3, 1088–1089.) Concerning the jury’s role in assessing the reasonableness of defendant’s belief in the need to defend, the Humphrey court explained: “Although the belief in the need to defend must be objectively reasonable, a jury must consider what ‘would appear to be necessary to a reasonable person in a similar situation and with similar knowledge . . . .’ [Citation.] It judges reasonableness ‘from the point of view of a reasonable person in the position of defendant . . . .’ [Citation.] To do this, it must consider all the ‘ “ ‘facts and circumstances . . . in determining whether the defendant acted in a manner in which a reasonable man would act in protecting his own life or bodily safety.’ ” ’ [Citation.] As we stated long ago, ‘. . . a defendant is entitled to have a jury take into consideration all the elements in the case which might be expected to operate on his mind . . . .’ [Citation.]” (Humphrey, supra, 13 Cal.4th at pp. 1082–1083.) By failing to permit the jury to consider the evidence as it pertained to the reasonableness element of self-defense, the trial court “failed to consider that the jury, in determining objective reasonableness, must view the situation from the defendant’s perspective.” (Id. at p. 1086.) For the same reasons, expert testimony explaining why a chronically homeless individual would experience a heightened fear of aggression would assist a jury in weighing the reasonableness of defendant’s belief of imminent harm, as was central to defendant’s self-defense claim. Given the earlier assault on defendant and heightened 19 sensitivity to violence experienced by the chronic homeless, the jury may have concluded that defendant was “ ‘justified in acting more quickly and taking harsher measures for [his] own protection in event of assault, than would a person’ ” who was not chronically homeless. (People v. Moore (1954) 43 Cal.2d 517, 528.) Paraphrasing the Humphrey court, “Evidence of [chronic homelessness] not only explains how a [chronically homeless individual] might think, react, or behave, it places the behavior in an understandable light.” (Humphrey, supra, 13 Cal.4th at p. 1088.) Contrary to the People’s assertion, this does not turn an objective standard—how a reasonable person would have perceived the risk—into a subjective one, creating a reasonable homeless person standard. Indeed, the Supreme Court rejected a similar argument in People v. Ochoa (1993) 6 Cal.4th 1199, where the court considered whether the defendant’s state of mind was relevant to a charge of gross vehicular manslaughter while intoxicated. The charge requires gross negligence, which is evaluated by an objective standard, namely, whether a reasonable person in the defendant’s position would have been aware of the risk involved. Rejecting defendant’s claim that his subjective state of mind was irrelevant and prejudicial, the court held: “In determining whether a reasonable person in defendant’s position would have been aware of the risks, the jury should be given relevant facts as to what defendant knew, including his actual awareness of those risks.” (People v. Ochoa, supra, 6 Cal.4th at p. 1205.) The Humphrey court found the same rationale applicable to the defendant’s self-defense claim: “What we said in Ochoa about the defendant’s actual awareness applies to this case. Although the ultimate test of reasonableness is objective, in determining whether a reasonable person in defendant’s position would have believed in the need to defend, the jury must consider all of the relevant circumstances in which defendant found herself.” (Humphrey, supra, 13 Cal.4th at p. 1083.) This did not replace the reasonable person standard with a reasonable battered woman standard, because the question remained what a reasonable person would have believed about the need to use lethal force, taking into consideration defendant’s situation and knowledge. (Id. at p. 1087.) 20 The same can be said here. A question before the jury was what a reasonable person would have believed about the need to use lethal force, taking into consideration defendant’s situation and knowledge. Judge Coates’s expert opinion would have shed light on this question. D. Judge Coates’s Expert Opinion Was Relevant to Defendant’s Credibility Judge Coates’s expert opinion was relevant to a third issue: defendant’s credibility. Although defendant did not testify at trial, the jury heard his account of the incident as he related it to the police just hours later and then again a few days later. By finding him guilty of first degree murder, the jury clearly did not accept defendant’s statements that he killed Bloom because he believed he posed an imminent threat of death or serious injury. Judge Coates’s opinion may have bolstered defendant’s credibility by providing a context within which the jury could assess defendant’s claims that he felt threatened by Bloom’s aggressive and belligerent demeanor and voice in the dark alleyway—and why defendant’s “adrenaline just jumped up” in response. (See People v. Day, supra, 2 Cal.App.4th at p. 415 [“[I]f the jury had understood [defendant’s] conduct in light of [intimate partner battering] evidence, then the jury may well have concluded her version of the events was sufficiently credible to warrant an acquittal on the facts as she related them.”]; accord, Humphrey, supra, 13 Cal.4th at p. 1087 [expert testimony on battered women’s experience was relevant to the victim’s credibility because it would assist the jury “by dispelling many of the commonly held misconceptions about battered women”].) E. The Subject of Judge Coates’s Expert Opinion Was Sufficiently Beyond Common Experience That It Would Have Assisted the Jury In the case of expert testimony, it is not enough that it is relevant to an issue in the case. It must also satisfy the criteria of Evidence Code section 801, which limits such testimony to that “[r]elated to a subject that is sufficiently beyond common experience that the opinion of an expert would assist the trier of fact.” (Evid. Code, § 801, subd. 21 (a).)6 The People contend Judge Coates’s opinion was inadmissible under section 801 because “[h]omelessness is not a subject beyond the knowledge of the average citizen.” The trial court was of the same view, stating that “what homeless people do or what they’re capable of doing or factors that may affect them” is “a common experience with the jurors and not subject to expert testimony.” We conclude otherwise: jurors in a case involving homeless violence may have a rudimentary understanding of the hazards of life on the street, but the realities of being homeless for a long period of time are beyond the understanding and life experience of the average juror. Judge Coates’s testimony may have dispelled “ ‘commonly held misconceptions’ ” or misguided “ ‘ “ ‘common sense’ conclusions” ’ ” that may have prevented the jury from understanding the circumstances as defendant actually perceived them. (Humphrey, supra, 13 Cal.4th at pp. 1086–1087.) In order for an expert’s opinion to be admissible, the subject matter need not be completely unfamiliar to the jury. Rather, expert testimony has been held admissible in a range of cases where the general subject matter of the expert’s testimony may be familiar to the average juror, yet critical aspects of that subject “are not likely to be fully known to or understood by the jury.” (People v. McDonald (1984) 37 Cal.3d 351, 377 (McDonald)). One such subject matter, touched on above, is the behavior of victims of domestic violence. In such cases, courts have recognized that leaving jurors to rely solely on their personal experiences and common sense about domestic relationships—and how the average person evaluates and reacts to a threat of imminent danger—will tend not to result in reliable fact-finding by the jury. Instead, it will create a grave risk that the jury will decide the case based on commonly held misconceptions and intuitive but erroneous assumptions about domestic violence victims. As explained in Humphrey, expert testimony on intimate partner battering was “ ‘aimed at an area where the purported 6 Evidence Code section 801 provides that expert testimony is admissible if it is “(a) [r]elated to a subject that is sufficiently beyond common experience that the opinion of an expert would assist the trier of fact; and [¶] (b) [b]ased on matter (including his special knowledge” “that is of a type that reasonably may be relied upon by an expert in forming an opinion upon the subject.” 22 common knowledge of the jury may be very much mistaken, an area where jurors’ logic, drawn from their own experience, may lead to a wholly incorrect conclusion, an area where expert knowledge would enable the jurors to disregard their prior conclusions as being common myths rather than common knowledge.’ ” (Humphrey, supra, 13 Cal.4th at p. 1099 (conc. opn. of Brown, J.), quoting State v. Kelly (1984) 478 A.2d 364, 378; see also People v. Brown (2004) 33 Cal.4th 892, 904–907; People v. Day, supra, 2 Cal.App.4th at pp. 415–418, overruled on other grounds in Humphrey, supra, 13 Cal.4th at p. 1089.) Similarly, in McDonald, supra, 37 Cal.3d 351, the Supreme Court recognized the relevance of expert testimony on the reliability of eye witness identification, despite acknowledging that jurors generally know that eyewitness identification can be unreliable. There, defendant appealed a decision of the trial court excluding expert testimony of a psychologist on psychological factors affecting the accuracy of eyewitness identification. (Id. at p. 363.) Reversing, the Supreme Court highlighted numerous cases recognizing the “ ‘well-known’ ” “ ‘vagaries of eyewitness identification . . . .’ ” (ibid.), and noted a proliferation of “empirical studies of the psychological factors affecting eyewitness identification . . . .” (Id. at p. 364.) It then went on to discuss Evidence Code section 801, observing, “The statute does not flatly limit expert opinion testimony to subjects ‘beyond common experience’; rather, it limits such testimony to such subjects ‘sufficiently beyond common experience that the opinion of an expert would assist the trier of fact’ (italics added). The emphasized words . . . make it clear that the admissibility of expert opinion is a question of degree. The jury need not be wholly ignorant of the subject matter of the opinion in order to justify its admission; if that were the test, little expert opinion testimony would ever be heard. Instead, the statute declares that even if the jury has some knowledge of the matter, expert opinion may be admitted whenever it would ‘assist’ the jury. It will be excluded only when it would add nothing at all to the jury’s common fund of information, i.e., when ‘the subject of inquiry is one of such common knowledge that men of ordinary education could reach a conclusion as intelligently as the witness’ [citation].” (McDonald, at p. 367.) 23 Applying that test to expert testimony on eyewitness identification, the court acknowledged that “from personal experience and intuition all jurors know that an eyewitness identification can be mistaken, and also know the more obvious factors that can affect its accuracy, such as lighting, distance, and duration.” (McDonald, supra, 37 Cal.3d at p. 367.) It went on to note, however, that according to professional literature “other factors bearing on eyewitness identification may be known only to some jurors, or may be imperfectly understood by many, or may be contrary to the intuitive beliefs of most.” (Id. at pp. 367–368.) The proffered expert testimony, the court concluded, would have assisted the jury in evaluating these many factors. (Id. at p. 368.) In Delia S. v. Torres (1982) 134 Cal.App.3d 471, 478–480 (disapproved on other grounds in Christensen v. Superior Court (1991) 54 Cal.3d 868, 905, fn. 28), the court held that expert testimony regarding “the reactions of rape victims and the characteristics and motivations of rapists” was properly admitted because, while most jurors hold opinions on those topics generally, expert testimony could clear up misconceptions that many jurors likely harbor and allow the jury to more accurately assess the credibility of the witnesses. Similar reasoning warranted the admission of expert testimony regarding the behavior of the parents of abused children in People v. McAlpin (1991) 53 Cal.3d 1289. There, the Supreme Court recognized that, while virtually all jurors have experience, direct or indirect, with parenthood, they tend to hold the intuitive but erroneous belief that the “parent of a molested child, naturally concerned for the welfare of the child and of other children, would promptly report the crime to the authorities.” (Id. at p. 1302.) According to the court, expert testimony was properly admitted to assist the trier of fact by clearing up this common misconception and allowing the jury to more reliably evaluate the credibility of a critical witness. (Id. at pp. 1300–1302; see also People v. Housley (1992) 6 Cal.App.4th 947, 955–956 [expert testimony allowed to disabuse the jury of commonly held misconceptions about the behavior of abused children]; People v. Bowker (1988) 203 Cal.App.3d 385, 390–394 [same].) 24 To this line of cases recognizing subject matters “sufficiently beyond common experience that the opinion of an expert would assist the trier of fact” (Evid. Code, § 801), we add chronic homelessness and its psychological impact. Most, if not all, jurors have likely encountered homelessness to some degree in their daily lives and, through these encounters, have an elementary understanding that life on the streets is unpleasant and more challenging than life as a housed individual. But, based on this court’s “personal experience and intuition” (MacDonald, supra, 37 Cal.3d at p. 367), we believe that the extent to which the homeless are confronted with persistent danger, as demonstrated by the evidence of the violence experienced by the homeless, and the psychological consequences of experiencing chronic homelessness, is beyond the ken of most. Judge Coates’s testimony would have dispelled the commonly held misconception—shared by the trial court—that “Everyone is subject to the same risks.” F. Defendant Was Prejudiced by the Exclusion of Judge Coates’s Expert Opinion Defendant urges that the exclusion of Judge Coates’s expert testimony deprived him of his constitutional right to present a complete defense, and Chapman v. California (1967) 386 U.S. 18, 24 mandates reversal unless the error was harmless beyond a reasonable doubt. But “the routine application of provisions of the state Evidence Code law does not implicate a criminal defendant’s constitutional rights.” (People v. Jones (2013) 57 Cal.4th 899, 957; accord, People v. Robinson (2005) 37 Cal.4th 592, 626–627 [“ ‘[A]s a general matter, the ordinary rules of evidence do not impermissibly infringe on the accused’s [state or federal constitutional] right to present a defense.’ ”].) This is so because “only evidentiary error amounting to a complete preclusion of a defense violates a defendant’s federal constitutional right to present a defense.” (People v. Bacon (2010) 50 Cal.4th 1082, 1104, fn. 4.) As such, the trial court’s error was one of state evidentiary law only (People v. McNeal (2009) 46 Cal.4th 1183, 1203), and the proper standard of review is whether it is reasonably probable that defendant would have obtained a more favorable result in the absence of the error. (People v. Watson (1956) 46 Cal.2d 818, 25 836; see also People v. Bacon, supra, 50 Cal.4th at p. 1104, fn. 4; Humphrey, supra, 13 Cal.4th at p. 1089.) Under this standard, we conclude the error was prejudicial. The jury found defendant guilty of first degree murder, meaning it did not believe his statement to the police that he believed Bloom was going to attack him, and so he “got” Bloom before Bloom could get him. Thus, the question of defendant’s actual belief in the need to use lethal force to protect himself from Bloom was critical. A reasonable probability exists that, if presented with Judge Coates’s expert testimony on chronic homelessness, the jury would have found defendant guilty of a lesser included offense of first degree murder (either voluntary manslaughter or second degree murder) or would have found the killing justifiable homicide. Judge Coates would have testified regarding empirical data showing that the homeless population experiences violence and the threat of violence at a much greater rate than the general population, and that the resulting vulnerability tends to create a greater sensitivity to threats of violence. Such testimony would have bolstered the credibility of defendant’s statements in his two police interviews that he actually perceived an imminent threat of death or great bodily harm when Bloom aggressively approached him in the darkened alleyway wielding an object that defendant mistook for a knife. Judge Coates’s testimony on the experience of chronic homelessness would also have tied into the testimony of witness Raymond Hoey, the homeless man who heard, but did not see, the fatal fight between defendant and Bloom. First, Hoey’s testimony provided anecdotal evidence regarding the dangers of being homeless when he described the fear of violence he generally experienced while living on the streets—a theme, not incidentally, defense counsel touched on during closing argument. Judge Coates’s expert opinion would have provided a broader context within which the jury could connect Hoey’s statements regarding the dangers he routinely faced with the experiences of the homeless more generally. Within such a broader context, the jury may have found defendant’s statements that he actually perceived an imminent threat from Bloom more credible. (See Smith, supra, 151 Cal. at pp. 627–629.) 26 The second way that Judge Coates’s testimony would have reinforced aspects of Hoey’s corroborating testimony related to Hoey’s description of how the altercation between defendant and Bloom began. Hoey heard the fight start with two men arguing, then quickly begin fighting each other, or “scuffling.” This testimony was entirely consistent with defendant’s statement that he was sitting on the steps when Bloom aggressively approached him, and that they argued before Bloom “came at” defendant holding an object that, in the darkness of the alley, defendant thought was a knife. Moreover, it was undisputed that Bloom was under the influence of an extremely large dose of methamphetamine, his friend Donovan Sweeden having testified that Bloom injected a large dose of methamphetamine and acted “agitated” and “aggressive.” And, defendant told the officers, he believed Bloom was on drugs as soon as he saw him approaching. This aspect of the case may have gained a valuable contextual background from Judge Coates’s proposed testimony regarding the prevalence of substance abuse among the homeless, which would have corroborated defendant’s assertion that he perceived Bloom as an imminent threat. Thus, had the jury heard Judge Coates’s expert opinion, there is a reasonable probability of a more favorable result, namely that the jury would have concluded that defendant actually believed he needed to use lethal force to defend himself from Bloom and would have accepted his theory of either perfect or imperfect self-defense. There is also a reasonable probability that even if the jury still rejected defendant’s claim that he actually believed he needed to kill Bloom to protect himself—and thus rejected his claims of perfect or imperfect self-defense—it might have returned a verdict of second degree rather than first degree murder. During deliberations, the jury made only two requests: the first, for the transcripts of defendant’s statements to the police, the second, for a further explanation of the meaning of premeditation and deliberation. From this, we can reasonably infer that the elements of premeditation and deliberation were difficult questions for the jury. (See People v. Carrera (1989) 49 Cal.3d 291, 314, fn. 14.) Judge Coates’s expert opinion on chronic homelessness might have provided sufficient context for defendant’s 27 hypersensitivity to the threat of violence such that the jury would not have found the premeditation and deliberation necessary for a first degree murder conviction. DISPOSITION The judgment of conviction is reversed. _________________________ Richman, Acting P.J. We concur: _________________________ Stewart, J. _________________________ Miller, J. 28 Trial Court: Sonoma County Superior Court Trial Judge: Hon. Rene Auguste Chouteau Counsel: Geoffrey M. Jones, under appointment by the Court of Appeal for Defendant and Appellant. Kamala D. Harris, Attorney General, Gerald A. Engler, Chief Assistant Attorney General, Jeffrey M. Laurence, Senior Assistant Attorney General, Catherine A. Rivlin, Supervising Deputy Attorney General, Gregg E. Zywicke, Deputy Attorney General for Plaintiff and Respondent. 29