In re
S258912
IN THE SUPREME COURT OF CALIFORNIA
April 3, 2023
First Appellate District, Division One A152748; Sonoma County Superior Court SCR32760
Chief Justice Guerrero authored the opinion of the Court, in which Justices Corrigan, Liu, Kruger, Groban, Jenkins, and Evans concurred.
Opinion of the Court by Guerrero, C.
In 2005, Rico Ricardo Lopez was tried along with several codefendants for the murder of Ignacio Gomez. A jury convicted Lopez and three of his codefendants (Peter Amante, Rogelio Cardenas, and Patrick Higuera, Jr.) of Gomez‘s first degree premeditated murder (
Later, following our opinion in People v. Chiu (2014) 59 Cal.4th 155 (Chiu), Lopez filed a petition for writ of habeas corpus challenging his conviction. Lopez alleged his jury had been instructed on the natural and probable consequences theory of aiding and abetting first degree murder, which we found invalid in Chiu, and this error was not harmless beyond a reasonable doubt. The trial court agreed with Lopez and granted relief. The prosecution appealed, and the Court of Appeal reversed. (In re Lopez (Sept. 25, 2019, A152748) [nonpub. opn.].)
The Court of Appeal relied on our then-recent opinion in People v. Aledamat (2019) 8 Cal.5th 1 (Aledamat). Aledamat discussed the standard of prejudice when a jury is instructed with two theories of an offense, one of which is legally valid and one of which is legally invalid, otherwise known as “alternative-theory” error. (Id. at p. 9.) We held that “no higher standard of review applies to alternative-theory error than applies to other misdescriptions of the elements. The same beyond a reasonable doubt standard applies to all such misdescriptions . . ..” (Ibid.) “The reviewing court must reverse the conviction unless, after examining the entire cause, including the evidence, and considering all relevant circumstances, it determines the error was harmless beyond a reasonable doubt.” (Id. at p. 13.)
The Court of Appeal held that the Chiu error here was harmless beyond a reasonable doubt based on the gang-murder special circumstance, which required the jury to find that an aider and abettor acted with intent to kill, and the “overwhelming” evidence against Lopez more generally. The Court of Appeal discounted the prosecutor‘s discussion of the natural and probable
We granted review to discuss the import of Aledamat on this record, including the jury‘s true finding on the gang-murder special circumstance and the potentially “overwhelming” nature of the evidence against Lopez. We conclude the gang-murder special circumstance here does not necessarily render the Chiu error harmless beyond a reasonable doubt. However, a reviewing court may hold the error harmless where it would be impossible, based on the evidence, for a jury to make the findings reflected in its verdict without also making the findings that would support a valid theory of liability. (Aledamat, supra, 8 Cal.5th at p. 15.) Indications in the record that the jury may have actually relied on an invalid theory, such as a prosecutor‘s closing argument or a jury note, do not preclude a finding of harmlessness if this standard is satisfied.
These principles reflect our holding in Aledamat that “no higher standard of review applies to alternative-theory error than applies to other misdescriptions of the elements.” (Aledamat, supra, 8 Cal.5th at p. 9.) Misdescriptions (or omissions) of the elements will invariably involve indications that the jury actually relied on an invalid theory, since the invalid theory was the only theory provided to it. Nonetheless, such errors may be found harmless if it would be impossible, based on the evidence, for a jury to make the findings reflected in its verdict without also finding the missing fact as well. (People v. Merritt (2017) 2 Cal.5th 819, 832 (Merritt).)
The Court of Appeal was therefore incorrect to hold that the gang-murder special circumstance, standing alone, showed that the Chiu error was harmless beyond a reasonable doubt. And, while “overwhelming” evidence may demonstrate harmlessness, a court‘s analysis of whether the evidence is “overwhelming” in this context is not as subjective or free-ranging as that term might imply. Instead, the analysis requires a court to rigorously review the evidence to determine whether any rational juror who found the defendant guilty based on an invalid theory, and made the factual findings reflected in the jury‘s verdict, would necessarily have found the defendant guilty based on a valid theory as well. (Aledamat, supra, 8 Cal.5th at p. 15.) Based on its short discussion, the Court of Appeal does not appear to have fully appreciated the proper standard for harmlessness in this context. We therefore reverse the judgment of the Court of Appeal and remand for reconsideration in light of the standards we describe in this opinion.
A. Trial Evidence
Sometime before midnight on June 26, 2002, Miguel and Rebecca S. stopped their car on a bridge in Santa Rosa, California because Miguel saw his father walking on the side of the road. Miguel got out to speak with him, while Rebecca remained in the car with their young children. A creek and bicycle path ran underneath the bridge. Miguel saw the victim, Gomez, riding his bicycle. Miguel knew Gomez was his father‘s friend, but he did not recognize Gomez at the time. Gomez said hello to Miguel and his father, and they whistled back and forth. Gomez turned, rode underneath the bridge, and continued along the bicycle path. Gomez was wearing blue clothing indicative of the Sureño criminal street gang. He was engaged to a Sureño associate and knew their distinctive whistle.
Miguel and Rebecca noticed a group of young men jumping over a fence that separated the creek from an apartment complex. Four men, two of whom Rebecca identified as Higuera and Ochoa-Gonzales, walked past. One of the unknown men said “Norte” or asked if Miguel “bang[ed] Norte.” The man showed Miguel what appeared to be a knife handle in his pants pocket. Miguel responded, “I don‘t bang nothing,” and the men kept walking. Miguel and Rebecca had been associated with the Norteño criminal street gang in the past, but no longer.
Three of the men went down the path after Gomez, while Ochoa-Gonzales hung back. A fifth man, apparently Amante, came by afterward. He was with two young women. Miguel saw Amante drop a knife, pick it up, and run toward Gomez. From a distance, Miguel saw three men attacking Gomez. One man, wearing a white shirt, had a knife and was making stabbing motions. Miguel later identified that man as Amante.
After the attack, the group walked back past Miguel and Rebecca. Miguel saw blood on two of the men. Both were wearing white shirts, and one was Amante. Miguel later clarified that Amante was wearing a red San Francisco 49ers football jersey with a white tank top underneath; the blood was on the tank top but not the jersey. Miguel and Rebecca drove away and called police to report the attack.
Police officers responded, but they did not find anything at the time. Gomez‘s body was discovered the next morning, along the bicycle path. His pants were pulled down, and there was blood nearby. Police also found four pieces of a broken knife blade at the scene.
An autopsy revealed that Gomez had suffered approximately 40 to 44 sharp-force injuries, including 38 to 40 stab wounds. The majority of the stab
The two women in the group, Kacee Dragoman and Lindsay Ortiz, testified at trial. Both were granted immunity, and their accounts of the night‘s events were largely consistent. Dragoman had been in a relationship with Amante for several years, and they had a child together. They lived in the apartment complex near the creek. The backyard of their apartment faced the creek itself. Ortiz lived in the same apartment complex as Dragoman and Amante. Ortiz described Dragoman as her best friend, and she loved Amante like a brother. Ortiz had known Lopez for a couple years and did not care for him. Dragoman had met Lopez only a few nights before the attack.
On the night of the attack, the defendants here — Lopez, Amante, Higuera, Cardenas, and Ochoa-Gonzales — were drinking and socializing with Dragoman, Ortiz, and others in the apartment shared by Amante and Dragoman. Lopez, Amante, Higuera, Cardenas, and Ochoa-Gonzales were all active participants in the Norteño criminal street gang. Dragoman associated with the Norteño gang, and Ortiz was friends with many Norteños. As described further below, the Norteños were in a deadly rivalry with the Sureños. Amante himself had been stabbed and severely wounded by Sureños during a Cinco de Mayo celebration a couple months earlier.
During the party, Dragoman was outside on her patio with several other people. They heard some whistling, and Dragoman recognized it as a Sureño gang whistle. It was a “bad sign,” according to Dragoman, because “[u]sually if they whistled, more were coming. They‘re hollering for more people to come out. They‘re signaling.” Someone at the party said, “It‘s a Scrap whistle,” and Ochoa-Gonzales yelled some remarks over the fence. (“Scrap” is a derogatory term for Sureño.) Dragoman recalled everyone getting “antsy” and “mak[ing] each other excited.” The defendants ran into the kitchen. Dragoman and Ortiz heard sounds like drawers opening and closing.
The defendants ran outside. Dragoman and Ortiz followed. The women came upon Amante, who had tried to climb over the fence but got stuck. Amante appeared to be intoxicated. Dragoman and Ortiz thought the situation was somewhat humorous and helped him down. Amante kept walking toward the creek, while Dragoman and Ortiz took a longer way around.
Ortiz testified that Amante disappeared from view for about five minutes. She initially saw Lopez, Cardenas, and Ochoa-Gonzales coming back from the creek, without Amante and Higuera. Lopez was wearing a white Raiders football jersey with black lettering, and Ortiz noticed blood on the front of his shirt. Amante and Higuera emerged afterward. Higuera had a cut on his arm and appeared to be in pain. Dragoman recalled seeing Amante meet up with the other defendants, then walk farther down the path with Higuera. They were gone for about 15 or 30 seconds. She confirmed Lopez was wearing a white Raiders jersey and Amante was wearing a red 49ers jersey with a white shirt underneath. When Dragoman saw Lopez, he was wearing a dark blue beanie that he did not have before the attack.
The defendants, Dragoman, and Ortiz all walked back over the bridge to the apartment complex. Lopez had a black knife handle in his hand, which Dragoman recognized from a knife set in her kitchen. After the group arrived back at the apartment, Dragoman remembered that Lopez was “[v]ery bouncy” and “[h]appy.” He had the blue beanie on his head and “was kind of like bragging[,] like walking around with a little strut, stuff like that.” Ortiz saw Lopez wearing the blue beanie as well, which she remembered as having the word “Sur” on it. Ortiz thought Lopez was “excited” and “pretty happy.” He said “something about that guy dying” and told Amante that “this was for Cinco de Mayo.” Amante responded, “What the fuck are you talking about?” Lopez said something in response, and everyone got quiet.
Dragoman watched Ochoa-Gonzales flush a knife handle (apparently the one Lopez was carrying) down the toilet. Dragoman put the knife Amante was carrying back in a kitchen drawer. Dragoman did not see anyone else with a knife that night.
Dragoman asked Lopez and Ochoa-Gonzales to take their clothes off so she could wash them. Ortiz helped. Dragoman saw blood on Lopez‘s shoes. Ortiz saw Ochoa-Gonzales pacing back and forth. He looked scared. He kept saying there were “cops in the creek.” At one point, Ochoa-Gonzales said,
The next morning, Dragoman found the blue beanie in her kitchen. She put it in a brown paper sack and threw it away in a stranger‘s garbage across town. Dragoman also noticed that her knife set was no longer complete; several knives were missing. All of the knives had markings and serial numbers that would identify them as part of a set, so Dragoman and Amante decided to get rid of the remaining knives. They drove into the countryside and threw them away, along with the knife block. Police were able to recover several of those knives, and Dragoman identified them as part of her set. Dragoman identified the broken knife blade found near Gomez‘s body as part of her set as well.
Dragoman and Ortiz initially lied to police about what happened that night. A few months later, Dragoman started speaking with the police again and disclosed additional details. Ortiz eventually did as well, apparently at Amante‘s request. Dragoman, Ortiz, and Amante discussed what they remembered about that night before Ortiz came forward. Ortiz also received police reports (or summaries) from Dragoman about the case. By the time of trial, neither Dragoman nor Ortiz had a relationship with Amante.
According to a pathologist, it was difficult to tell with any degree of certainty how many stabbing instruments were involved in the attack. The knives police recovered from Dragoman‘s knife set (apparently steak knives) could have made any of the wounds on Gomez‘s body. Other knives of a similar size could have made the wounds as well. The knife whose blade was found broken at the scene could have inflicted some of the stab wounds, including two of the wounds to Gomez‘s chest. In the pathologist‘s opinion, a single person could have inflicted all of the wounds in less than a minute.
A criminalist tested the broken knife blade for traces of blood. He obtained a presumptive positive result, but further testing could not confirm the presence of blood. The criminalist also examined the knife‘s serrated edges. Based on his experience, the criminalist would have expected to see more blood, tissue, or other material in the serrated areas of the knife if it had been used to stab a person.
A police gang expert also testified. He was familiar with both Norteños and Sureños. In his opinion, the Norteños were a criminal street gang under California law. (
The Norteños claimed all of Northern California, and they came into conflict with Sureños who moved into the region. The expert described several murders and attempted murders committed by Norteño gang members against Sureño gang members in Sonoma County.
Sureños claimed various areas in Santa Rosa, and the creek where Gomez was killed was disputed territory. The expert explained, “At the time this crime occurred . . . Sureños had pretty much claimed that area just north of there as theirs. Quite a few Norteño gang members were living just south of that area, also living east and west of there and using that path to go to and from places.” Both gangs had painted graffiti in the creek and crossed out graffiti painted by their rival. These “crossouts” are a sign of serious disrespect to a rival gang.
A common Norteño graffiti mark is “SK” or ” ‘Scrap killa.’ ” Sureño graffiti includes, ” ‘We Chap killas,’ ” with ” ‘Chap’ ” being a derogatory term for Norteños. Along the same road that spanned the creek here, Sureño graffiti mourned the death of a Sureño gang member killed by a Norteño. Nearby, someone had written, ” ‘Fuck Whacky Die Slow.’ ” ” ‘Whacky’ ” was Amante‘s gang moniker. The expert explained, “This is very significant to me in that it shows that Sureño gang members know Mr. Amante as a rival gang member, that they‘re well aware he was assaulted and seriously injured, and that they want him to die slowly. He is an enemy, and they want him to suffer.”
The expert explained each defendant‘s relationship to the Norteños. He opined that each defendant was an active participant in the Norteño criminal street gang. For Lopez, the expert showed photos with Lopez together with other Norteño gang members displaying gang symbols and writing. Lopez had a facial tattoo of one dot and four dots, for the number 14, which showed his affiliation with the Norteño gang. Lopez had Norteño tattoos on his hands as well, though he may have removed one. Lopez admitted to police officers that he was a Norteño gang member and wore clothing associated with the Norteño gang.
The prosecutor presented the expert with a hypothetical mirroring the facts of the prosecution‘s case and asked whether such a murder would have been committed for the benefit of or in association with a criminal street gang. The expert responded that it would, based on the clothing worn by the victim, which was consistent with clothing worn by Sureño gang members or affiliates; the act of defendants arming themselves together; the confrontation
In testimony to be considered against Lopez only, a jail inmate named Richard Smith recounted an argument with Lopez while they were both incarcerated. During the confrontation, Lopez turned to Smith and said, ” ‘I‘ll kill you just like I killed the guy in the creek.’ ” Smith had several previous felony convictions. He was a longtime heroin addict and was taking methadone. On cross-examination, Lopez‘s counsel introduced various letters Smith had written to the prosecutor requesting favors. Smith said he hoped his testimony would help him at sentencing in a pending criminal case, although the prosecution had not promised anything.
B. Closing Arguments
In his closing argument, the prosecutor did not expressly name the person or persons who stabbed Gomez. He explained, “In this case, there is no burden on the People to establish who the actual stabber was. . . . Simply that there was a stabber and that the defendants here on trial were either that stabber or an aider and abettor in the crime of that stabber.” The prosecutor discounted the possibility that either the broken knife found at the scene or the large butcher knife Amante carried were the murder weapon. Both knives were too large and, as to the broken knife, it did not have the expected amount of blood on its blade. The prosecutor suggested that the wounds were entirely consistent with the smaller steak knives from Dragoman‘s knife set. He argued that at least one other defendant must have had a knife and used it against Gomez. Given this uncertainty, the prosecutor focused on two theories of aiding and abetting, either directly or under the doctrine of natural and probable consequences. He explained each theory to the jury.
Regarding Lopez specifically, the prosecutor noted that he was wearing a white Raiders jersey, so he could have been the stabber Miguel identified. He also had blood on his shirt, according to Ortiz, and on his shoes, according to Dragoman. The prosecutor argued that Lopez personally wielded the knife that was found broken, “because there‘s no other evidence of another broken knife out at the scene or anywhere else.” It must have broken against a rock or the asphalt path, and “[t]hat tells you that [Lopez] was there and he was actively participating in the attack on [Gomez]. Whether or not that knife
The prosecutor said it was possible Lopez personally stabbed Gomez, but it was not probable given the size of the broken knife and the absence of any substantial amount of blood. So, according to the prosecutor, Lopez was a direct aider and abettor to murder: “He knew of the unlawful purpose of the person right there with him. Maybe it was the person who pantsed [Gomez]. Maybe it was the person who stabbed [Gomez] in the head splitting his skin to his skull or driving the knife into his chest. But he was there. He was right there. And he shattered his knife during the attack. Absolutely had knowledge. Intended to commit or encourage or facilitate the crime? Beyond any question. By act or advice did he aid, promote, encourage or instigate its commission? Of course he did. Whether or not he‘s the actual stabber.”
The prosecutor also touched on the natural and probable consequences doctrine: “And finally, nonhomicide target crimes. Even if you think that he was down there just trying to stab a Scrap, maybe. Maybe, despite all the evidence, he just wanted to really seriously wound the guy. It doesn‘t matter. He aided and abetted in that serious attack someone that was right there with him. Murdered [Gomez]. And of course under these circumstances, that was inevitable. [¶] So whether he is an actual stabber or not, whether he aided and abetted with the intent to kill or not, he‘s guilty of murder as a natural and probable consequence of his act.” But the prosecutor did not believe the jury needed to reach that theory: “I would submit to you that he is either an actual stabber, which is possible, or he‘s an aider and abettor to murder, period. You don‘t even need to get to this theory as to Rico Lopez.”
Lopez‘s counsel criticized the prosecutor‘s reliance on three alternate theories. He repeatedly attacked the credibility of Smith, the inmate who testified to Lopez‘s admission. He noted that the prosecutor seemed almost to concede that the knife Lopez was carrying was not used to stab anyone, so Lopez‘s alleged admission to Smith was inconsistent with the evidence. Lopez‘s counsel attacked the prosecution‘s arguments as speculative, and he argued the prosecution‘s witnesses lacked credibility. He contended that Ortiz and Dragoman were accomplices to murder, and their testimony was insufficiently corroborated. He criticized Miguel for testifying to an attack that was far away, at night, without any lighting.
Lopez‘s counsel emphasized that the defendants were all drinking, and Amante was drunk, so no one could have formed the specific intent necessary to commit first degree murder or otherwise intend to kill. He argued there
In his rebuttal argument, the prosecutor denied that Dragoman and Ortiz were accomplices and posited that even if they were, their testimony was amply corroborated by other evidence. He asserted that the prosecution‘s witnesses were credible, persuasive, and largely consistent. He disagreed that any of the defendants were so intoxicated that they could not intend to kill or premeditate and deliberate a killing, especially in light of the circumstances of the attack and the number of wounds inflicted on the victim. The prosecutor also noted that, under the natural and probable consequences doctrine, a defendant did not have to have the intent to kill to be found guilty of murder. He denied that the two theories of aiding and abetting were somehow fallback positions. They were, instead, “the law of the State of California.”
C. Jury Instructions
The trial court instructed the jury that a principal in a crime includes both a person “who directly and actively commit[s] the act constituting the crime” and a person “who aid[s] and abet[s] the commission of the crime.” (See CALJIC No. 3.00.) The court described direct aiding and abetting as follows: “A person aids and abets the commission of a crime when he or she: [¶] . . . [w]ith knowledge of the unlawful purpose of the perpetrator, and [¶] . . . [w]ith the intent or purpose of committing or encouraging or facilitating the commission of the crime, and [¶] . . . [b]y act or advice aids, promotes, encourages or instigates the commission of the crime.” (See CALJIC No. 3.01.) The court went on to explain the doctrine of natural and probable consequences, as it was understood at the time: “One who aids and abets another in the commission of a crime or crimes is not only guilty of those crimes, but is also guilty of any other crime committed by a principal which is a natural and probable consequence of the crimes originally aided and abetted.” (See CALJIC No. 3.02.) The elements were (1) the commission of a target crime, here breach of the peace, an assault, a battery, an assault with a deadly weapon, or an assault by force likely to produce great bodily injury; (2) “[t]he defendant aided and abetted one of those crimes“; (3) “[a] co-principal in that crime committed the crime of murder“; and (4) “[t]he crime of murder was a natural and probable consequence of the commission” of the target crime. (Ibid.) The court instructed the jury on the elements of the target crimes, as well as first degree murder, second degree murder, and manslaughter.
D. Jury Deliberations and Verdict
The jury deliberated for approximately four and a half days before reaching a verdict. During deliberations, the jury requested that Miguel‘s testimony be read back. It also asked two legal questions. First, the jury requested instructions for the criminal street gang sentencing enhancement. The court directed the jury to the appropriate instruction, which it had already provided. Second, the jury asked for clarification on premeditation and deliberation in the context of the natural and probable consequences doctrine. Its note stated, “We are having difficulties with the sentence[,] ‘To constitute a deliberate and premeditated killing, the slayer must weigh and consider the question of killing and the reasons for and against such a choice and, having in mind the consequences, he decides to and does kill,’ versus deliberated and premeditated breach of peace or assault that results in a killing. [¶] We need more clarification of premeditation and deliberation [CALJIC No. 8.20] and how to relate it to [CALJIC No. 3.02, regarding natural and probable consequences].” After a lengthy discussion with counsel, the court provided the following response: “The term ‘deliberate and premeditate[d]’ refers only to First Degree Murder. First Degree Murder is defined by jury instruction 8.20. [¶] The term ‘deliberate and premeditate[d]’ is not an element of any of the following: Breach of the Peace, Assault, Battery, Assault by Means of Force [L]ikely to Produce Great Bodily Injury, or Assault with a Deadly Weapon. Those crimes are defined elsewhere in the Court‘s instructions[.] [¶] . . .
The jury returned verdicts against each defendant, as described above. The court sentenced Lopez to life imprisonment without the possibility of parole, and the judgment was affirmed on direct appeal.
E. Writ Proceedings
As noted, following our opinion in Chiu, Lopez filed a petition for writ of habeas corpus alleging that the jury was improperly instructed on the natural and probable consequences theory of aiding and abetting first degree murder. In her return, the district attorney argued that any error was harmless because the jury did not rely on the doctrine of natural and probable consequences in reaching its verdict. The district attorney pointed to the prosecution‘s closing argument as to Lopez, which did not emphasize the natural and probable consequences doctrine, and the jury‘s gang-murder special-circumstance finding, which showed the jury believed Lopez intended to kill.
After hearing argument, the trial court granted relief. It believed that the Chiu error was not harmless beyond a reasonable doubt. The court found persuasive the jury‘s note regarding the natural and probable consequences instruction. In the court‘s view, it showed the natural and probable consequences doctrine “was on the jury‘s mind shortly before it rendered its verdict, and . . . the jury was potentially grappling with its applicability as to all defendants.” The court believed that the evidence at trial was consistent with liability under the natural and probable consequences doctrine. It therefore vacated Lopez‘s first degree murder conviction.
The prosecution appealed. While the appeal was pending, we decided Aledamat, supra, 8 Cal.5th 1, which addressed the harmlessness standard for alternative-theory error, where the jury was instructed with valid and invalid theories of liability. Relying on Aledamat, the Court of Appeal concluded that the error here was harmless beyond a reasonable doubt. The Court
of Appeal highlighted the jury‘s gang-murder special-circumstance finding. It explained, “A true finding as to this circumstance required proof beyond a reasonable doubt that Lopez acted with an intent to kill, as opposed to the intent to commit one of the target crimes.” The Court of Appeal further held that the jury‘s note regarding natural and probable consequences was not tied to Lopez, so its import was unclear, and the evidence against Lopez was “overwhelming.” We granted review.
A. Valid and Invalid Theories
“There are two distinct forms of culpability for aiders and abettors. First, an aider and abettor with the necessary mental state is guilty of the intended crime. Second, under the natural and probable consequences doctrine, an aider and abettor is guilty not only of the intended crime, but also ‘for any other offense that was a “natural and probable consequence” of the crime aided and abetted.‘” (Chiu, supra, 59 Cal.4th at p. 158.) Chiu eliminated the latter form of aiding and abetting for first degree premeditated murder: “[A]n aider and abettor may not be convicted of first degree premeditated murder under the natural and probable consequences doctrine. Rather, his or her liability for that crime must be based on direct aiding and abetting principles.” (Id. at pp. 158-159, italics omitted.) Chiu is retroactive and may be raised, as here, in a petition for writ of habeas corpus. (In re Martinez (2017) 3 Cal.5th 1216, 1222 (Martinez).)2
Applying Chiu, the parties agree the trial court erred by instructing the jury that it could find Lopez guilty of first degree murder based on the theory of natural and probable consequences. They likewise agree (or at least do not contest) that the trial court properly instructed the jury that it could find Lopez guilty of first degree murder as a direct aider and abettor or as the actual perpetrator of the first degree murder.
For a defendant to be liable as a direct aider and abettor, “the prosecution must show that the defendant aided or encouraged the commission of the murder with knowledge of the unlawful purpose of the perpetrator and with the intent or purpose of committing, encouraging, or facilitating its commission. [Citation.] Because the mental state component — consisting of intent and knowledge — extends to the entire crime, it preserves the distinction between assisting the predicate crime of second degree murder and assisting the greater offense of first degree premeditated murder. [Citations.] An aider and abettor who knowingly and intentionally assists a confederate to kill someone could be found to have acted willfully, deliberately, and with premeditation, having formed his own culpable intent. Such an aider and abettor, then, acts with the mens rea required for first degree murder.” (Chiu, supra, 59 Cal.4th at p. 167.)
For a defendant to be liable as an actual perpetrator, the prosecution must prove that the defendant unlawfully killed a human being, or a fetus,
B. Standard of Prejudice
Where a jury is instructed on alternate theories of liability, one legally valid and one legally invalid, a federal constitutional error has occurred. The defendant has been deprived of his or her right to “a jury properly instructed in the relevant law.” (Martinez, supra, 3 Cal.5th at p. 1224; see
In Aledamat, we explored the meaning of the harmless beyond a reasonable doubt standard in this circumstance involving alternative theories of liability. We noted that “[t]his harmless error rule applies in a variety of contexts, such as . . . error in omitting entirely one or more elements of a charged offense.” (Aledamat, supra, 8 Cal.5th at p. 9.) We held that the application of this standard for alternative-theory error should be consistent with, and not different from, the application of the same standard for other misdescriptions of the charged offense. (Ibid.) Specifically, we rejected the argument that alternative-theory error could be found harmless only where “there is a basis in the record to find that ‘the jury has “actually” relied upon the valid theory.’ ” (Ibid.) Instead, a reviewing court may “examine[] what the jury necessarily did find and ask[] whether it would be impossible, on the evidence, for the jury to find that without also finding the missing fact as well.” (Id. at p. 15.) In other words, if ” ‘[n]o reasonable jury’ ” would have found in favor of the defendant on the missing fact, given the jury‘s actual verdict and the state of the evidence, the error may be found harmless beyond a reasonable doubt. (Ibid.)
For that last point, Aledamat quoted our earlier opinion in Merritt, supra, 2 Cal.5th at page 832, which considered a relatively more straightforward instructional error involving the omission of one or more elements of
This test is exacting, and it requires much of a reviewing court. “[S]afeguarding the jury guarantee will often require that a reviewing court conduct a thorough examination of the record. If, at the end of that examination, the court cannot conclude beyond a reasonable doubt that the jury verdict would have been the same absent the error — for example, where the defendant contested the omitted element and raised evidence sufficient to support a contrary finding — it should not find the error harmless.” (Neder, supra, 527 U.S. at p. 19.) “A reviewing court making this harmless-error inquiry does not, as Justice Traynor put it, ‘become in effect a second jury to determine whether the defendant is guilty.’ [(Traynor, The Riddle of Harmless Error (1970) p. 21.)] Rather a court, in typical appellate-court fashion, asks whether the record contains evidence that could rationally lead to a contrary finding with respect to the omitted element. If the answer to that question is ‘no,’ holding the error harmless does not ‘reflect a denigration of the constitutional rights involved.’ [Citation.] On the contrary, it ‘serves a very useful purpose insofar as [it] blocks setting aside convictions for small errors or defects that have little, if any, likelihood of having changed the result of the trial.’ ” (Ibid.)
Although Aledamat seemingly settled the issue, Lopez contests the relevance of Neder and Merritt to the alternative-theory error here. Relying largely on California authorities that predate Aledamat, and federal authorities that predate Neder, Lopez proposes a multistep “protocol” for a reviewing court considering whether an alternative-theory error is harmless beyond a reasonable doubt. Under the proposed protocol, if a jury‘s verdicts do not themselves show that the jury necessarily relied on a valid theory in convicting a defendant, a reviewing court “should then proceed to consider the entire cause, but it should consider first those aspects of the record that most clearly indicate the basis on which the jury actually rested its verdict.” If there are indications the jury considered the invalid theory, either because the prosecutor relied on the invalid theory during argument or because the jury referenced the invalid theory during deliberations, then reversal is required regardless of any other circumstances.
There are sound reasons for performing the same type of harmless error analysis in cases involving an alternative-theory error. In Aledamat, we recognized that ” ‘drawing a distinction between alternative-theory error and the instructional errors in [several cases including Neder] would be “patently illogical,” given that such a distinction “reduces to the strange claim that, because the jury . . . received both a ‘good’ charge and a ‘bad’ charge on the issue, the error was somehow more pernicious than . . . where the only charge
Lopez notes that our earlier decisions have referenced circumstances such as a prosecutor‘s argument or a jury‘s note during deliberations as one factor supporting our conclusion that an alternative-theory error was not harmless. (See, e.g., Martinez, supra, 3 Cal.5th at pp. 1226-1227; Chiu, supra, 59 Cal.4th at pp. 167-168.) But, as we explained in Aledamat, these decisions do not reflect a higher or different standard of review. “In both Chiu and Martinez, we examined the record and found that it affirmatively showed the jury might have based its verdict on the invalid theory. Because no other basis to find the error harmless beyond a reasonable doubt was at issue, we did not explore whether other ways of finding the error harmless existed. Those cases merely provide one way in which a court might evaluate harmlessness. They do not preclude other ways.” (Aledamat, supra, 8 Cal.5th at p. 13, italics added.) Chiu and Martinez did not consider the role of the prosecutor‘s argument or a jury note if the standard for harmlessness under Neder and Merritt (and later Aledamat) had been met.4
Lopez relies on Sullivan v. Louisiana (1993) 508 U.S. 275, which considered whether a defective jury instruction on reasonable doubt was subject to harmless error review. In this context, Sullivan explained, “Harmless-error review looks, we have said, to the basis on which ‘the jury actually rested its verdict.’ [Citation.] The inquiry, in other words, is not whether, in a trial that occurred without the error, a guilty verdict would surely have been rendered, but whether the guilty verdict actually rendered in this trial was surely unattributable to the error. That must be so, because to hypothesize a guilty verdict that was never in fact rendered — no matter how inescapable the findings to support that verdict might be — would violate the jury-trial guarantee.” (Id. at p. 279.) Because Sullivan did not believe that a jury verdict rendered with a defective reasonable doubt instruction was “a jury verdict within the meaning of the Sixth Amendment, the entire premise of Chapman review is simply absent.” (Id. at p. 280.)
Neder, however, sharply limited the application of Sullivan‘s broad language. It acknowledged that “this strand of the reasoning in Sullivan does
Lopez also relies on the more recent Court of Appeal opinion in People v. Thompkins (2020) 50 Cal.App.5th 365, which considered the standard of prejudice for alternative-theory error in light of Aledamat. The discussion in Thompkins reflects the complexity of this area of law. (Id. at pp. 398-401.) We need not examine this discussion in detail, except to note that Thompkins was incorrect to elevate Sullivan over more recent discussions of harmless error in Neder, Hedgpeth, and Merritt. (Thompkins, at pp. 400-401.) We disapprove Thompkins to the extent it is inconsistent with our description of the standard of prejudice in this opinion.
In sum, as we held in Aledamat, no higher standard applies to alternative-theory errors than applies to other misdescriptions or omissions of the elements of an offense. (Aledamat, supra, 8 Cal.5th at p. 9.) A reviewing court must determine whether any rational jury would have found the defendant guilty based on a valid theory if the jury had been properly instructed. “The reviewing court examines what the jury necessarily did find
C. Harmlessness Beyond a Reasonable Doubt
The Attorney General bears the burden of showing that the error was harmless beyond a reasonable doubt. (Chapman, supra, 386 U.S. at p. 24; Martinez, supra, 3 Cal.5th at p. 1227.) He argues that the jury‘s verdicts, combined with the evidence at trial, leave no reasonable doubt a rational jury would have convicted Lopez of first degree murder based on a valid theory if it had been properly instructed.
The valid theory of direct aiding and abetting required the jury to find “that the defendant aided or encouraged the commission of the murder with knowledge of the unlawful purpose of the perpetrator and with the intent or purpose of committing, encouraging, or facilitating its commission.” (Chiu, supra, 59 Cal.4th at p. 167.) ” ‘When the offense charged is a specific intent crime, the accomplice must “share the specific intent of the perpetrator“; this occurs when the accomplice “knows the full extent of the perpetrator‘s criminal purpose and gives aid or encouragement with the intent or purpose of facilitating the perpetrator‘s commission of the crime.” [Citation.]’ [Citation.]” What this means here, when the charged offense and the intended offense — murder or attempted murder — are the same, i.e., when guilt does not depend on the natural and probable consequences doctrine, is that the aider and abettor must know and share the murderous intent of the actual perpetrator.” (People v. McCoy (2001) 25 Cal.4th 1111, 1118 (McCoy), fn. omitted.) “An aider and abettor who knowingly and intentionally assists a confederate to kill someone could be found to have acted willfully, deliberately, and with premeditation, having formed his own culpable intent. Such an aider and abettor, then, acts with the mens rea required for first degree murder.” (Chiu, at p. 167.)6
The Court of Appeal below first held that the jury‘s verdict, standing alone, shows that the jury made the findings necessary for first degree murder. It has long been established that an alternative-theory error is harmless beyond a
The Court of Appeal noted that the jury did not merely find Lopez guilty of first degree murder. The jury also found true the gang-murder special circumstance. (
The Court of Appeal focused on the latter alternative, and it does appear far more likely the jury believed Lopez was an aider and abettor rather than an actual killer. The only direct evidence of the identity of the stabber came from Miguel, who identified Amante. Miguel also said the stabber was wearing a white shirt, which could include Lopez because he was wearing a white Raiders jersey, but Miguel did not identify Lopez specifically. The prosecutor argued that Lopez could have stabbed Gomez, but he described this possibility as unlikely because the knife Lopez carried was too large to make most of Gomez‘s wounds and the broken knife blade found at the scene did not have any confirmed blood or other tissue that would indicate the blade was used for stabbing. The prosecutor therefore contended Lopez was “an aider and abettor to murder.” Likewise, in his own briefing, Lopez argues that “this scenario [that he was an actual killer] is undermined by the record, as the prosecutor felt compelled by the weakness of the evidence to argue explicitly to the jury that [Lopez] was an aider and abettor to murder rather than an ‘actual stabber.’ ”
In any event, we need not decide whether the jury could have found that Lopez was an actual killer. Even assuming the jury found the gang-murder special circumstance true based on its belief that Lopez “intentionally killed the victim,” the Attorney General concedes the special circumstance does not itself establish the elements of first degree premeditated murder under either a direct perpetrator or an aiding and abetting theory. Based on the special circumstance instructions alone, if the jury found Lopez was an actual killer, it is reasonably possible the jury could have believed he did not personally premeditate and was liable only for second degree murder.
If the jury did not find that Lopez was an actual killer, it must instead have determined that Lopez “with the intent to kill aided, abetted, counseled, commanded, induced, solicited, requested, or assisted any actor in the commission of the murder in the first degree.” (See CALJIC No. 8.80.1.) The
The Attorney General argues that the special circumstance finding still demonstrates harmlessness because, in addition to intent to kill, it shows the jury made all of the remaining findings necessary to support the valid theory of direct aiding and abetting first degree murder. This argument is unpersuasive. While the relevant language evokes similar concepts, it does not cover all of the elements of direct aiding and abetting.
As noted, for a defendant to be liable for first degree murder as a direct aider and abettor, “the prosecution must show that the defendant aided or encouraged the commission of the murder with knowledge of the unlawful purpose of the perpetrator and with the intent or purpose of committing, encouraging, or facilitating its commission.” (Chiu, supra, 59 Cal.4th at p. 167.) The jury here was likewise instructed more broadly that “[a] person aids and abets the commission of a crime when he or she: [¶] . . . [w]ith knowledge of the unlawful purpose of the perpetrator, and [¶] . . . [w]ith the intent or purpose of committing or encouraging or facilitating the commission of the crime, and [¶] . . . [b]y act or advice aids, promotes, encourages or
The Attorney General relies on People v. Beck and Cruz (2019) 8 Cal.5th 548, but that case is distinguishable because it involved conspiracy to commit murder, not the gang-murder special circumstance. In Beck, we noted the defendants “were charged with conspiracy to murder, not conspiracy to commit a lesser crime that resulted in murder.” (Id. at p. 645.) We therefore held there was “no possibility they were found guilty of murder on a natural and probable consequences theory.” (Ibid.) Beck has little relevance here because, among other things, the findings necessary for conspiracy to commit murder and the findings necessary for the gang-murder special circumstance are materially different. Conspiracy to murder requires not only intent to kill, but also intent to agree and actual agreement. (
This conclusion, however, does not end the harmlessness inquiry. “In determining . . . whether the error was harmless, the reviewing court is not limited to a review of the verdict itself.” (Aledamat, supra, 8 Cal.5th at p. 13.) The Court of Appeal below nodded to this further inquiry by describing the evidence against Lopez as “overwhelming.” While it is common (including in this court) to shorthand the analysis in this way, such a description may obscure the specific question a reviewing court must answer in order to find an omitted element or alternative-theory error harmless beyond a reasonable doubt.
Lopez objects to this mode of analysis because the jury could only have rendered its special circumstance finding after finding Lopez guilty of first degree murder. He argues that, regardless of the content of the special circumstance finding, the jury‘s first degree murder verdict could have been based on the invalid theory of natural and probable consequences. Lopez‘s argument reflects a misplaced focus on what the jury subjectively thought, rather than what a reviewing court can determine, beyond a reasonable doubt, that a rational jury would have found if it had been properly instructed. We rejected that focus in Gonzalez, Merritt, and Aledamat, just as the United States Supreme Court did in Neder and Hedgpeth.
Lopez also contends various circumstances in this matter preclude a finding of harmlessness. For example, he asserts that the natural and probable consequences theory was the “easiest way” for the jury to find him guilty, but this contention again reflects an unduly restrictive focus on what the jury actually thought, rather than what a reviewing court can determine, beyond a reasonable doubt, that any rational jury would surely have found based on the verdict and the evidence here. The proper analysis under Aledamat does not rest on ” ‘the likelihood that the jurors would have applied the erroneous instruction,’ ” but whether the jury could have found what it did find without also making the findings necessary for a valid theory. (People v. Glukhoy (2022) 77 Cal.App.5th 576, 598, fn. 42, review granted July 27, 2022, S274792.) Lopez also argues the error cannot be held harmless because the evidence was “sufficient” to support a conviction based on the invalid theory. But this circumstance merely shows it was possible for a rational jury to rely on the invalid theory; it does not foreclose a conclusion
Along the same lines, Lopez focuses on what he characterizes as “indications” the jury actually relied on an invalid theory, including the prosecutor‘s closing argument and the jury‘s note during deliberations. Lopez is correct that our review for harmless error should encompass “the entire cause, including the evidence” and consider “all relevant circumstances.” (Aledamat, supra, 8 Cal.5th at p. 13.) But if a reviewing court can conclude the error was harmless beyond a reasonable doubt under Neder and Aledamat because any rational juror who made the findings reflected in the actual verdict and heard the evidence at trial would also have made the findings necessary to support a valid theory, such “indications” will not generally be significant. The prosecutor‘s mere reliance on an invalid theory will not overcome a showing of harmlessness under Neder and Aledamat. Indeed, as discussed, if the prosecutor‘s mere reliance on an invalid theory were dispositive on the issue of harmlessness, trial court errors of omission and misdescription of an offense‘s elements could almost never be held harmless because the prosecutor almost invariably argues based on the theories in the trial court‘s instructions. For similar reasons, a jury note showing mere consideration of an invalid theory will not overcome a showing of harmlessness under Neder and Aledamat. Every jury presented with instructions that omit or misdescribe the elements of an offense will have considered — and actually relied to some extent on — an invalid theory of liability. These “indications” are no bar to a finding of harmlessness under Neder, Merritt, and Aledamat.9
The Court of Appeal here correctly based its analysis on Aledamat, but it appears to have misapprehended certain aspects of the harmless error analysis. As noted, the Court of Appeal described the evidence against Lopez as “overwhelming” and found the Chiu error harmless on that basis. It explained
First, the Court of Appeal does not appear to have considered whether a rational jury, having rendered the verdicts at issue here, would necessarily have to believe the cited testimony regarding Lopez, especially his bragging about the stabbing. The evidence a court may consider under Aledamat does not necessarily extend to the whole body of evidence supporting the verdict. Rather, a court must determine what evidence a jury would rationally have to believe. For example, as Lopez correctly points out, “Although jurors evidently believed Dragoman and Ortiz‘s testimony that [Lopez] joined in the attack on the victim, one or more jurors may have doubted the credibility of those witnesses’ assertions portraying [Lopez] as the most culpable attacker.” It is well settled that the jury has wide latitude to believe or disbelieve witnesses, or even specific portions of their testimony, as it sees fit. ” ‘[T]he jury properly may reject part of the testimony of a witness, though not directly contradicted, and combine the accepted portions with bits of testimony or inferences from the testimony of other witnesses thus weaving a cloth of truth out of selected available material.’ ” (Stevens v. Parke, Davis & Co. (1973) 9 Cal.3d 51, 67-68; accord, People v. Crooker (1956) 47 Cal.2d 348, 355 [“The jury may accept as true a portion of the testimony of a witness and disbelieve the remainder or have a reasonable doubt as to its correctness“].)
Second, the Court of Appeal‘s reference to the sufficiency of the evidence supporting Lopez‘s first degree murder conviction is contrary to the applicable standard under Aledamat. The question here is not the sufficiency of the evidence to support a valid theory, but its opposite. To determine the sufficiency of the evidence, a reviewing court essentially asks whether any rational juror could have made the findings necessary to convict the defendant on a valid theory. To determine harmlessness under Aledamat, a reviewing court essentially asks whether any rational juror who made the findings reflected in the verdict and heard the evidence at trial could have had reasonable doubt regarding the findings necessary to convict the defendant on a valid theory. “The reviewing court examines what the jury necessarily did find and asks whether it would be impossible, on the evidence, for the jury to find that without also finding the missing fact as well.” (Aledamat, supra, 8 Cal.5th at p. 15.)
To summarize, under Aledamat, “no higher standard of review applies to alternative-theory error than applies to other misdescriptions of the elements.
Finally, while we have the discretion to apply the correct standard to the facts here (see, e.g., Aledamat, supra, 8 Cal.5th at pp. 13-15), we decline to do so given the size and complexity of the underlying trial record. (See
CONCLUSION
We reverse the judgment of the Court of Appeal and remand the matter for further proceedings consistent with this opinion.
GUERRERO, C. J.
We Concur:
CORRIGAN, J.
LIU, J.
KRUGER, J.
GROBAN, J.
JENKINS, J.
EVANS, J.