Real Party in Interest.


First Appellate District, Division Four


San Francisco City and County Superior Court

2219428, 195198

August 30, 2021


Erwin F. Fredrich for Petitioner.

Benjamin Salorio, Public Defender (Imperial), and Darren Bean, Deputy Public Defender, for William Morse as Amicus Curiae on behalf of Petitioner.

No appearance for Respondent.

Xavier Becerra and Rob Bonta, Attorneys General, Lance E Winters, Chief Assistant Attorney General, Jeffrey M. Laurence, Assistant Attorney General, Seth K. Schalit, René A. Chacón and Moona Nandi, Deputy Attorneys General, for Real Party in Interest.

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

Chief Justice Cantil-Sakauye filed a concurring opinion.

Opinion of the Court by Cuéllar, J.

The Sexually Violent Predator Act (Welf. & Inst. Code, § 6600 et seq. (SVPA or the Act))[1] allows the state to petition superior courts for the involuntary civil commitment of certain convicted sex offenders whose diagnosed mental disorders make them a significant danger to others and likely to reoffend after release from prison. The purpose of the SVPA is to protect the public from a select group of criminal offenders (sexually violent predators, or SVPs), and to provide these offenders with the necessary treatment for their mental disorders. (Hubbart v. Superior Ct. (1999) 19 Cal.4th 1138, 1143–1144 (Hubbart).) Consistent with this goal, the Act relies on a number of procedural safeguards to ensure that only those offenders predisposed to criminal sexual violence can be committed, and only for as long as they need treatment. Section 6602, subdivision (a) of the Act provides one such safeguard: It requires the superior court to hold a “probable cause hearing” as an initial step in the judicial process for commitment. (§ 6602, subd. (a).) If the court determines that probable cause supports the state’s petition, it must then hold an offender over for trial. (Ibid.) Otherwise, the court must dismiss the petition. (Ibid.)

What concerns us in this case is what kind of evidence the trial court may consider in making its initial SVPA probable cause determination. Specifically, we must resolve whether superior courts can admit certain hearsay evidence in psychological evaluation reports in finding probable cause to commit individuals under the SVPA. Petitioner Jeffrey Walker challenges the Court of Appeal’s denial of his writ petition, arguing that the trial court admitted inadmissible hearsay in two evaluations in finding probable cause: factual details underlying two rape offenses that he had been charged with, but not convicted of, and resulted in convictions that did not qualify as predicate offenses for commitment under the SVPA. He contends that the trial court’s decision to admit this hearsay concerning nonpredicate offenses represented prejudicial error.

We agree. Contrary to the Court of Appeal’s reasoning, section 6602, subdivision (a) does not create an exception that allows hearsay regarding nonpredicate offenses to be introduced via evaluation reports. What we hold is that nothing in the statutory language, its legislative history, its place in the broader SVPA statutory scheme, or comparisons to other statutory provisions indicates the existence of a hearsay exception for such hearsay in expert evaluations. Nor does anything in the SVPA or our case law indicate that the Legislature — in creating the hearing as a safeguard for SVP candidates to test the sufficiency of the evidence supporting the state’s petition and prevent meritless ones from proceeding to trial — must have created an exception for hearsay on nonpredicate offenses to be introduced via evaluations. Under these circumstances, we decline to find that the Legislature explicitly or implicitly created a hearsay exception in section 6602, subdivision (a), for this evidence.

Because the inadmissible hearsay was foundational to the trial court’s probable cause determination, we must reverse and remand to the Court of Appeal, with instructions for it to remand the matter to the trial court so it can conduct a new probable cause hearing consistent with this opinion.


In June 2015, the District Attorney of the City and County of San Francisco filed a petition to commit Walker as an SVP.[2] At the time, Walker was nearing the end of a state prison term for a pandering conviction. (Pen. Code, § 266i.)

Two mental health evaluations supported the petition. Thomas MacSpeiden and Roger Karlsson, psychologists appointed by the Director of the State Department of State Hospitals (DSH), evaluated Walker shortly before the district attorney filed the petition. MacSpeiden and Karlsson were appointed pursuant to section 6601, subdivision (e), after the first two appointed psychologists disagreed whether Walker satisfied the statutory criteria to be an SVP. Both concluded that Walker satisfied the statutory criteria. MacSpeiden diagnosed Walker with “Borderline Personality Disorder” and “Other Specified Paraphilia, Sexual Activity with Non-consenting Persons”; Karlsson diagnosed him with “Antisocial Personality Disorder, augmented by a severe level of psychopathy.”

In their evaluation reports, the psychologists discussed Walker’s 1990 conviction for rape, a predicate “ ‘[s]exually violent offense’ ” under the SVPA. (Welf. & Inst. Code, § 6600, subd. (b).) They also discussed the alleged facts regarding two charged sex crimes that did not result in convictions qualifying as sexually violent offenses. First, Walker was charged in 1989 with raping a 16-year-old victim. The trial court dismissed the rape charge prior to trial, but it convicted Walker of unwanted sexual intercourse with a minor. (Pen. Code, § 261.5.) Second, Walker was charged with rape in 2005. A jury acquitted Walker of this charge (apparently, during the trial, it was determined the victim had lied), but it convicted him of pandering.

MacSpeiden and Karlsson obtained the details underlying the 1990 rape conviction from a September 1991 report from Walker’s probation officer. The evaluations related the following details regarding the offense: Walker unsuccessfully approached the victim at a nightclub. He eventually pulled her onto the dance floor and danced with her, though she attempted to push him away when he tried to pull her closer. He then pulled her to the club’s parking lot. She believed she could get in her car to drive away. When they arrived at her car, she pushed him away as he tried to pull her closer. She reluctantly agreed to give him a ride to his house. When they arrived at the location that he had directed them to, Walker reached across her and turned off the ignition. She rejected his attempts to kiss her and attempted to fight him off, but he raped her.

The psychologists obtained the details underlying the 1989 rape allegation from the 1991 probation report, and they obtained the details underlying the 2005 rape allegation from a police inspector’s affidavit in support of an arrest warrant. In his evaluation, MacSpeiden quoted the documents’ description of events, which in turn summarized and quoted the victims’ account of Walker’s conduct and statements leading up to, during, and after the alleged rapes. Karlsson also quoted the police affidavit, and he summarized the probation report’s description of events.

In particular, the evaluations conveyed the following about the 1989 rape allegation: Walker met the victim at a car show, and he told her he was a photographer for a company hiring models. They met up five nights later, and Walker drove her to a park and took photographs of her. He then asked her to change into a swimsuit she had brought along. When she went into the bathroom to change clothes, Walker followed her, refused to leave, pushed her against the wall face first, groped her, and called her a “ ‘bitch.’ ” While repeatedly pushing her face into the wall, he forcibly had sex with her; he then forced her onto the floor facedown and continued to rape her. Afterward, Walker drove the victim to her car, and he grabbed her and forcibly kissed her before she left.

Regarding the 2005 rape allegation, the evaluations indicated that Walker introduced himself to the 2005 alleged rape victim as the employee of a local radio station that was looking for help promoting the station at clubs. She readily conveyed her interest. They met a few days later, after he contacted her and informed her that he had some promotions at a few San Francisco clubs. He drove her up to San Francisco, and during the drive he instructed her on the procedures for working in a strip club and the prices to charge for certain sex acts. The victim had never worked at a strip club or as a prostitute. Once they arrived in San Francisco, Walker parked the car and told her he needed to show her “ ‘the game.’ ” He went around to the passenger door, entered the car, and placed his hands between her legs. The victim told him to stop and kept her legs closed, but Walker refused. He attempted to take her underwear off, and he digitally penetrated her. She continued to tell Walker to stop and was very upset, but Walker persisted and eventually forcibly had sex with her. He then took her to various strip clubs in an attempt to employ her. The victim went along out of fear. She turned over the money she made at the clubs to Walker. When they returned to his car, he forced her to orally copulate him. He then drove her home.

In February and March 2016, the trial court held a probable cause hearing spanning five sessions. At the hearing, the prosecution moved to admit the psychologists’ reports into evidence. Walker objected, arguing that the reports contained inadmissible hearsay regarding the 1989 and 2005 rape allegations, and that their admission would violate due process because of the unreliability of the hearsay evidence. He moved to exclude the reports entirely, or in the alternative to strike the portions of the reports containing inadmissible hearsay. The trial court overruled Walker’s objections and admitted the reports into evidence.

During the probable cause hearing, Walker’s attorney cross-examined the psychologists about their evaluations, including their reliance on the 1989 and 2005 rape allegations.

MacSpeiden testified that the two rape allegations constituted key rationales for his evaluation, even though he knew neither resulted in a rape conviction. According to MacSpeiden, the two allegations and Walker’s 1990 rape conviction all had “essentially the same” “modus operandi.” Because of this modus operandi, and because the rape allegations resulted in charges, he determined it was important to describe them in his report, and he believed in reaching his evaluation and continued to believe during his testimony that the allegations were true and the documents relaying them were reliable and appropriate evidence for him to rely on. He did contend that he would have still arrived at the same evaluation even without the rape allegations, because Walker had 11 sex offense charges between 1988 and 2007 — a pattern of illegal sexual conduct. But he admitted that charges do not carry the same weight as convictions for purposes of his evaluation, and that he had none of the factual details underlying the charges besides the rape allegations.

Karlsson also testified that his evaluation was informed by the 1989 and 2005 rape allegations. He explained that he relied on the probation report and police affidavit relaying these allegations because the documents were from sworn officers, and he therefore had no reason to believe the records had untrue information. But he indicated that he had not considered whether the 1989 allegation involved force, stating, “I would need to read my report and recalibrate my opinion of that.” And he stated that had he not been able to factor either the 1989 or 2005 rape allegation into his evaluation, his overall opinion could have been different.

After cross-examining the psychologists, the defense called four witnesses at the probable cause hearing: (1) the 2005 victim’s ex-boyfriend, who testified that the victim admitted she had falsely accused Walker of rape; (2) Bruce Yanofsky, one of the initial psychologists to evaluate Walker, who testified that Walker did not qualify as an SVP; (3) the police officer investigating the 2005 rape allegation, who very briefly testified about producing the affidavit at issue; and (4) Walker, who testified that, as he had explained in his evaluation interviews, the 1989 and 2005 rape allegations were untrue.

Following the hearing, the trial court determined that probable cause existed to commit Walker as an SVP.

Walker then repeatedly but unsuccessfully sought to dismiss the petition. He first moved to dismiss the petition in September 2016, arguing that the psychological evaluations contained inadmissible hearsay in violation of our recent decision in People v. Sanchez (2016) 63 Cal.4th 665 (Sanchez). The trial court denied the motion. In March 2017, Walker unsuccessfully moved to have the trial court reconsider the denial of his motion to dismiss. Walker filed another motion to dismiss in October 2019, this time citing Bennett v. Superior Court (2019) 39 Cal.App.5th 862 (Bennett), a decision holding that facts regarding two dismissed rape allegations against the defendant, relayed by psychologists in their SVPA evaluation reports and probable cause testimony, were inadmissible hearsay under Sanchez. The trial court denied Walker’s motion. Walker filed a petition for writ of mandate with the Court of Appeal, but it summarily denied the petition. In January 2020, Walker filed yet another motion to dismiss, citing Bennett as well as Couthren, supra, 41 Cal.App.5th 1001, a decision that also applied Sanchez at an SVPA probable cause hearing to bar the admission of hearsay in psychological evaluations. But the trial court again denied Walker’s motion to dismiss.

Walker challenged the denial of his last motion to dismiss by filing another petition for writ of mandate in the Court of Appeal. After issuing an order to show cause, the court denied Walker’s writ petition. (Walker v. Superior Court (2020) 51 Cal.App.5th 682, 686 (Walker).) In reaching this holding, it disagreed with Bennett and Couthren: It created a split of authority over whether the SVPA permits the trial court at an SVPA probable cause hearing to admit hearsay regarding nonpredicate offenses contained in expert evaluations. (Walker, at pp. 694, 701–702.) We granted review to resolve the split.


To determine whether trial courts conducting SVPA probable cause hearings may admit hearsay concerning nonpredicate offenses in evaluation reports, we must apply the SVPA’s general requirements, including those governing probable cause hearings; and the SVPA’s hearsay rules, as established by the statute and decisional law. We examine these threads individually before proceeding to weave them together.


The SVPA provides for the involuntary civil commitment of certain sex offenders before the end of their prison or parole revocation terms. (§ 6601.) “In describing the underlying purpose” of the SVPA, “the Legislature expressed concern over a select group of criminal offenders who are extremely dangerous as the result of mental impairment, and who are likely to continue committing acts of sexual violence even after they have been punished for such crimes.” (Hubbart, supra, 19 Cal.4th at pp. 1143–1144.) “[T]o the extent such persons are currently incarcerated and readily identifiable,” the Legislature has indicated that “commitment under the SVPA is warranted immediately upon their release from prison.” (Hubbart, at p. 1144.) The Act provides these individuals with “treatment for mental disorders from which they currently suffer and reduces the threat of harm otherwise posed to the public.” (Hubbart, at p. 1144.) SVPs are committed “for an indeterminate term to the custody of [DSH] for appropriate treatment and confinement in a secure facility.” (§ 6604.)

In order to commit someone under the Act, the state must establish four conditions: (1) the person has previously been convicted of at least one qualifying “sexually violent offense” listed in section 6600, subdivision (b) (§ 6600, subd. (a)(1)); (2) the person has “a diagnosed mental disorder that makes the person a danger to the health and safety of others” (ibid.); (3) the mental disorder makes it likely the person will engage in future acts of sexually violent criminal behavior if released from custody (ibid.); and (4) those acts will be predatory in nature (Cooley v. Superior Court (2003) 29 Cal.4th 228, 243 (Cooley)). Civil commitment can commence only if, after a trial, the trier of fact finds beyond a reasonable doubt that each of these four requirements is met. (Ibid., citing §§ 6600, 6601, 6603, 6604.)

The trial represents the final step in the “complex administrative and judicial process” required to civilly commit an individual as an SVP. (Cooley, supra, 29 Cal.4th at p. 244.) The process leading up to a trial begins when the Department of Corrections and Rehabilitation screens inmates at least six months before their release date (§ 6601, subd. (a)), and refers any potential SVP to DSH for a “full evaluation” (id., subd. (b)). DSH then designates two practicing psychologists or psychiatrists to evaluate the inmate in accordance with a “standardized assessment protocol,” which requires “assessment of diagnosable mental disorders, as well as various factors known to be associated with the risk of reoffense among sex offenders.” (Id., subd. (c).) If the two mental health professionals agree that the inmate qualifies as an SVP (or if only one reaches this conclusion and two subsequently appointed professionals concur), the DSH Director forwards a request for a commitment petition, along with copies of the evaluation reports and other supporting documents, to the county in which the inmate was last convicted. (Id., subds. (d)–(f), (h).) If the county’s designated counsel agrees, the petition for commitment is filed in superior court. (Id., subd. (i).)

The superior court must review the petition once it’s received to determine whether probable cause exists to commit the individual as an SVP. As an interim step, the SVPA allows a potential SVP to request a review of the petition under section 6601.5. If the superior court determines from the face of the petition that probable cause exists, it shall order a probable cause hearing under section 6602. (§ 6601.5.) Regardless of whether alleged SVPs request a paper review of the petition under section 6601.5, they are entitled to a probable cause hearing under section 6602.

They are entitled to specific procedures at the hearing, too. The trial court “shall review the petition and shall determine whether there is probable cause to believe that the individual named in the petition is likely to engage in sexually violent predatory criminal behavior upon his or her release.” (§ 6602, subd. (a).) The alleged SVP is entitled to assistance of counsel at the hearing. (Ibid.) If the court determines that there is not probable cause, it must dismiss the petition; but if it determines that probable cause does exist, the court must order a trial to be conducted. (Ibid.)

Section 6602, subdivision (a) provides instructions — but only spare ones — for conducting the probable cause hearing. It does not delineate the “specific procedural requirements” governing the presentation and admission of evidence at the probable cause hearing. (Cooley, supra, 29 Cal.4th at p. 245, fn. 8.) Following In re Parker (1998) 60 Cal.App.4th 1453 (Parker), lower courts have generally construed the subdivision as requiring “something more than a facial review of the petition” (Parker, at p. 1464): that is, that it allows for the admission of both oral and written evidence (id. at p. 1469; see People v. Hayes (2006) 137 Cal.App.4th 34, 43 (Hayes)). In a few cases, we have briefly noted the procedural requirements that Parker developed, without resolving whether section 6602, subdivision (a) mandates them. (See, e.g., Cooley, at p. 248, fn. 8; People v. Cheek (2001) 25 Cal.4th 894, 899–900 (Cheek).)

The parties must comply with the rules of evidence. (Evid. Code, § 300 [“Except as otherwise provided by statute,” the Evid. Code applies in all actions other than those before a grand jury]; In re Kirk (1999) 74 Cal.App.4th 1066, 1071–1073 (Kirk).) That the hearsay rule applies at SVPA probable cause proceedings is the crux of the issue we resolve in this case.

A core premise of evidence law is that not all statements are created equal. The hearsay label applies to an out-of-court statement offered to prove that its assertion is true. (Evid. Code, § 1200, subd. (a).) A familiar feature of the law of evidence, in California and beyond its borders, is that hearsay is generally inadmissible unless it falls under a specific exception that justifies its admission. (Id., subd. (b).) Documents like reports and records are generally hearsay if they are offered for their truth, and indeed, may contain further instances of hearsay, each of which is inadmissible unless also covered by an exception. (Sanchez, supra, 63 Cal.4th at pp. 674–675; see also Evid. Code, § 1201.)

No one disputes that the evaluation reports at issue here are hearsay and contain hearsay. As with many SVPA evaluation reports, and as the People concede, the MacSpeiden and Karlsson reports were offered for their truth. The People sought their admission as competent evidence of the experts’ opinions and the facts on which they relied. (See Couthren, supra, 41 Cal.App.5th at p. 1010.) The experts expressly drew upon secondary sources — most relevantly, a probation report and police affidavit — for their contents, including the truth of out-of-court statements like victim statements. (See id. at pp. 1010–1011.) Each level of out-of-court statement, from the evaluation reports to the probation and police reports to the victim statements, ordinarily must fall under a hearsay exception to be admitted into evidence. (Id. at p. 1011.)

Trial courts have conducted many probable cause hearings since Parker was decided. They’ve generally understood Welfare and Institutions Code section 6602, subdivision (a) as permitting the state to introduce into evidence the evaluation reports, despite their hearsay nature. (See, e.g., Parker, supra, 60 Cal.App.4th at pp. 1469–1470; Hayes, supra, 137 Cal.App.4th at p. 43.) Under the Parker procedure, which the trial court here followed, the state may present the reports as evidence — in support of or in lieu of the experts’ testimony on direct examination — subject to the alleged SVP’s right to cross-examine the experts. (Parker, at pp. 1469–1470; see also Kirk, supra, 74 Cal.App.4th at pp. 1071–1073 [the evaluations must be properly certified under Evid. Code, §§ 1530–1531].) The working assumption appears to be that the subdivision’s command to “review the petition” (Welf. & Inst. Code, § 6602, subd. (a)) also allows for the review of “the facts on which the petition was filed, i.e., the underlying . . . experts’ evaluations” (Parker, at p. 1468). This assumption does not strike us as an unreasonable inference to draw from the subdivision’s language. The subdivision directs courts to review the petition and determine “whether there is probable cause.” (Welf. & Inst. Code, § 6602, subd. (a).) The separate command to determine probable cause contrasts with the language of Welfare and Institutions Code section 6601.5, which simply directs courts to engage in a prehearing, facial review of the petition for its adequacy. This contrast suggests that Welfare and Institutions Code section 6602, subdivision (a) does not limit the probable cause inquiry to consideration of the four corners of the petition, and instead also contemplates review of other evidence such as the evaluations that necessarily support the petition.

Parker’s hearsay rule permits the admission of the evaluations. Although the rule plausibly effectuates section 6602’s sparse language, we need not definitively resolve its legal validity. (Cf. Cooley, supra, 29 Cal.4th at p. 248, fn. 8 [noting Parker’s hearsay rule in passing].) Walker does not challenge the admissibility of the reports; nor does he challenge the admissibility of hearsay in the reports writ large, including content otherwise admissible via hearsay exceptions outside the SVPA. (See Walker, supra, 51 Cal.App.5th at pp. 695–696 [similar].)[3] The dispositive question here concerns a narrower question: the admissibility of particular hearsay content in the reports. (Cf. Bennett, supra, 39 Cal.App.5th at p. 883 [the issue is not “whether the prosecution may present an expert’s conclusions at the probable cause hearing through the introduction of the expert’s report,” but instead experts relating particular inadmissible hearsay].) More specifically, we must determine whether hearsay about nonpredicate offenses — otherwise inadmissible hearsay — may be admitted through expert reports under section 6602.

When the appellate courts decided the cases that led us to grant review here, they understood their disagreement to be about the implications of Sanchez, supra, 63 Cal.4th at page 686 (an expert may not testify to case-specific hearsay facts, about which the expert has no personal knowledge, “unless they are independently proven by competent evidence or are covered by a hearsay exception”). (Walker, supra, 51 Cal.App.5th at pp. 694–695; Couthrensupra, 41 Cal.App.5th at pp. 1006, 1019–1021; Bennettsupra, 39 Cal.App.5th at pp. 878–880.) Though the parties’ briefing emphasized Sanchez, we don’t need to further parse that case to decide this one.

We can instead resolve this case as a straightforward question of statutory interpretation: whether the SVPA or decisional law on the statute create a hearsay exception covering expert report content like what’s at issue here. (Evid. Code, § 1200, subds. (a), (b); People v. Otto (2001) 26 Cal.4th 200, 207 (Otto) [“ ‘[E]xceptions to the hearsay rule [in Evidence Code section 1200] . . . may . . . be found in other codes and decisional law’ ”]; Sanchez, supra, 63 Cal.4th at p. 686.) The parties agree that this case turns on whether either source creates a hearsay exception for content in an expert report concerning offenses that are not predicate offenses.

Bennett and Couthren answered this question in the negative. But the Court of Appeal here disagreed. It held that section 6602 and prior interpretations of the SVPA establish a hearsay exception that permits trial courts at probable cause hearings to broadly admit any hearsay in evaluations. What we conclude from our analysis of the SVPA and cases interpreting the Act is that, as Bennett and Couthren reasoned and Walker argues, no hearsay exception allows for the wholesale admission of SVPA evaluations, with any hearsay that they may contain. More specifically, neither the Legislature nor our case law has created a hearsay exception allowing admission of hearsay accounts involving prior, nonpredicate allegations or convictions at SVPA probable cause hearings.

For the reasons that follow, the Court of Appeal erred in determining otherwise.



We interpret the SVPA de novo. (Kirby v. Immoos Fire Protection, Inc. (2012) 53 Cal.4th 1244, 1250.) As with any question of statutory construction, our core task here is to determine and give effect to the Legislature’s underlying purpose in enacting the SVPA and any particular provisions at issue. (California Teachers Assn. v. San Diego Community College Dist. (1981) 28 Cal.3d 692, 698; Calatayud v. State of California (1998) 18 Cal.4th 1057, 1065; Goodman v. Lozano (2010) 47 Cal.4th 1327, 1332.) We first consider the words of the statute, as statutory language is generally the most reliable indicator of legislation’s intended purpose. (In re H.W. (2019) 6 Cal.5th 1068, 1073.) We consider the ordinary meaning of the relevant terms, related provisions, terms used in other parts of the statute, and the structure of the statutory scheme. (Larkin v. Workers’ Comp. Appeals Bd. (2015) 62 Cal.4th 152, 157.) If the relevant statutory language is ambiguous, we look to appropriate extrinsic sources, including the legislative history, for further insights. (In re H.W., at p. 1073.)

Nothing in the language of the SVPA indicates the Legislature created an explicit hearsay exception to allow hearsay in evaluation reports, regarding an SVP candidate’s prior nonqualifying offenses, to be admitted at a probable cause hearing. In describing the requirements for the hearing, Welfare and Institutions Code section 6602, subdivision (a) states simply that the trial court “shall review the petition and shall determine whether there is probable cause,” and that the defendant has a right to the assistance of counsel. We find in this limited language no indication that the Legislature created an express exception for evaluation reports that covers hearsay content regarding nonpredicate offenses. (Couthren, supra, 41 Cal.App.5th at p. 1012; cf. Kirk, supra, 74 Cal.App.4th at pp. 1071–1072 [SVPA is silent as to whether documentary evidence must comply with the certification requirements of Evid. Code, §§ 1530–1531, and therefore does not create an exception to these requirements].) What the subdivision focuses on is the petition: It directs the superior court to make its determination based on its review of the petition, with no mention of the experts, their psychological evaluation reports, or any documentary evidence those evaluations relied upon. (See Couthren, at p. 1014.)

Even assuming section 6602, subdivision (a)’s spare language allows consideration of experts’ opinions and conclusions contained in their reports (see ante, at p. 14), the Court of Appeal’s elaboration on this premise — that the language creates a broad hearsay exception for all instances of hearsay contained in those reports, including facts regarding nonpredicate offenses — goes too far.

The Court of Appeal’s reading of the subdivision relied on a chain of dubious inferences. First, it determined that courts must consider the psychological evaluations because the SVPA requires the evaluation reports as the basis for the petition. (Walker, supra, 51 Cal.App.5th at pp. 686, 694–695.) Based on this “necessary” role, the court inferred that the reports’ facts are “ ‘impliedly intended to be pleaded by averments or proper attachment to the petition’ ” and the reports “must be deemed incorporated into the petition” (id. at p. 695) — meaning “section 6602’s directive for a trial court to ‘review the petition’ at a probable cause hearing necessarily requires the court to review the evaluations, as well” (Walker, at p. 696). Second, it reasoned that trial courts may, as part of their review of the evaluations, consider hearsay contained therein. (Id. at pp. 686, 688; see also id. at p. 696 [§ 6602, subd. (a)’s directive to review the petition includes review of the evaluations in their entirety, and even if the subdivision’s language is ambiguous, the SVPA’s “structure and purpose [citation] confirms that section 6602[, subdivision ](a) excepts the evaluations and any information contained within them from the hearsay rule”].) We’re not persuaded.

To begin, the SVPA does not appear to require consideration of the evaluation reports. It requires only that these reports be prepared as a predicate to filing an SVPA petition and then be made available to the county’s designated counsel, who then decides whether to file a petition and what to include in it. (§ 6601, subds. (d), (h)(1), (i).) Although the evaluations are often attached as exhibits to the petition (see Hubbart, supra, 19 Cal.4th at p. 1149; Walker, supra, 51 Cal.App.5th at p. 695; cf. People v. Superior Court (Ghilotti) (2002) 27 Cal.4th 888, 913 (Ghilotti) [in describing the process for how a party can challenge an expert evaluation for material legal error, we advised that the evaluations “should . . . be attached to the petition”]), the statutory provisions governing the evaluations do not dictate how the county’s counsel should present them to the court or even require the attorney to do so. The People may choose to establish the facts underlying the petition by other means. In view of these considerations, which tend to suggest that the evaluation reports largely play a “discrete and preliminary” gatekeeping role in the SVPA commitment process (People v. Superior Court (Preciado) (2001) 87 Cal.App.4th 1122, 1130; see also In re Wright (2005) 128 Cal.App.4th 663, 672), the Court of Appeal likely went too far in reasoning that the reports must be deemed incorporated into the petition and therefore considered — along with any hearsay tucked inside — in the trial court’s probable cause determination.

Even if the petition does incorporate the underlying reports, that doesn’t mean that courts “review[ing] the petition” under Welfare and Institutions Code section 6602, subdivision (a) have carte blanche to admit and consider any hearsay the experts include. (See Couthren, supra, 41 Cal.App.5th at p. 1012.) The logical extension of this argument: Anything the experts put in their reports can come in too. All of the other rules of evidence, like foundation (Evid. Code, § 402), relevance (id., § 350), or undue prejudice (id., § 352) wouldn’t apply either — essentially removing the trial court from any role in discerning what is admissible from inadmissible in the evaluations. That cannot be correct.

The legislative history also fails to buttress the Court of Appeal’s interpretation. Legislators have left unchanged the relevant language in the subdivision since the first version of the SVPA bill. (See Parker, supra, 60 Cal.App.4th at p. 1465, citing Assem. Bill No. 888 (1995–1996 Reg. Sess.) § 3, as introduced Feb. 22, 1995; § 6602, subd. (a).) The legislative history is silent on what procedural requirements govern the probable cause determination. (Parker, at p. 1465 [the legislative history lends “little assistance”].) Instead, it simply emphasizes that the probable cause hearing serves as one important safeguard for defendants’ liberty interests, including by preventing unfair or arbitrary involuntary confinements. (Assem. Com. on Public Safety, Analysis of Assem. Bill No. 888 (1995–1996 Reg. Sess.) as amended Apr. 17, 1995, p. 7.) Nothing else from the legislative history has material bearing on the hearsay question before us.

Nor have our prior interpretations determined that the Legislature, in enacting section 6602, explicitly or implicitly created a hearsay exception covering the evidence at issue here. Our case law has explained that subdivision (a)’s limited language, legislative history, and place in the broader structure of the SVPA all establish that the probable cause hearing functions much like a criminal preliminary hearing. It serves to “ ‘ “ ‘weed out groundless or unsupported charges.’ ” ’ ” (Cooley, supra, 29 Cal.4th at p. 247 [explaining this in the context of holding that a court must test the sufficiency of the evidence of all four elements required for commitment, and not just a single element]; see also id. at p. 252 [adopting the same burden of proof as in the criminal context].) Nothing about this evidence-screening function indicates that the Legislature necessarily meant for its limited instructions — “review the petition” and “determine probable cause” (§ 6602, subd. (a)) — to create a hearsay exception covering facts about nonqualifying offenses contained in an evaluation report.

This conclusion comes into sharper focus when we compare Welfare and Institutions Code section 6602, subdivision (a) to rules governing an analogous context: probable cause hearings involving criminal charges. The comparison is not a perfect one. As the Court of Appeal and the People identify, the governing statutes for the two hearings appear different in a number of ways. (Walker, supra, 51 Cal.App.5th at p. 700 [unlike Welf. & Inst. Code, § 6602, subd. (a), the statutes governing criminal preliminary hearings explicitly “contemplate that the prosecution will present its case by examining witnesses in the presence of the defendant” (citing Pen. Code, § 865)].) But we can still compare the two hearings because of their similar evidence-screening functions, and because the Evidence Code similarly governs how this screening occurs for both hearings, absent specified exceptions. (Evid. Code, § 300.)

It’s telling that an explicit exception to the Evidence Code’s hearsay rule does exist for criminal preliminary hearings. Proposition 115, adopted by the voters in 1990, amended Penal Code section 872, subdivision (b) to create a “limited” hearsay exception (Whitman v. Superior Court (1991) 54 Cal.3d 1063, 1074 (Whitman)), allowing “a properly qualified investigating officer to relate out-of-court statements by crime victims or witnesses” (id. at p. 1072). As we explained in Whitman, the subdivision clearly contemplates that the testifying officer has sufficient experience, expertise (id. at pp. 1073–1074 [at least five years in law enforcement or special training]), and “knowledge of the crime or the circumstances under which the out-of-court statement was made so as to meaningfully assist the magistrate in assessing the reliability of the statement” (id. at p. 1072). In view of this clear and carefully delineated hearsay exception in a relatively analogous context, we are not persuaded by the People’s interpretation. The People ask us to read section 6602, subdivision (a)’s spare language as an indication that the Legislature created an even broader hearsay exception than what Proposition 115 created: one that, in the absence of any apparent legislative determination of, or requirements for, their expertise or knowledge to do so, allows any psychologist to relate hearsay as true accounts of nonpredicate offenses from investigating officers’ reports, including any victim and witness statements to these officers. (Cf. Whitman, at pp. 1072, 1074 [declining to read the limited exception the voters enacted as embracing “ ‘reader’ ” testimony: “whereby a noninvestigating officer, lacking any personal knowledge of the matter, nonetheless . . . relate[s] not only what the investigating officer” described in his or her investigatory report, “but also what the other witnesses told the investigating officer”].)

Other SVPA provisions reinforce our qualms about reading into section 6602, subdivision (a) an exception for hearsay about nonpredicate offenses contained in expert reports. Section 6600, subdivision (a)(3) and section 6605, neither of which apply to the evidence in dispute here, both contain hearsay exceptions — in stark contrast to section 6602.

Consider what section 6600, subdivision (a)(3) allows. It permits the prosecution to show the existence of and details underlying the first element of the SVP determination — a predicate sex-offense conviction — “by documentary evidence, including, but not limited to, preliminary hearing transcripts, trial transcripts, probation and sentencing reports, and evaluations by the [DSH].” (Ibid.) As originally enacted, the subdivision did not provide for the admission of documentary evidence. But for understandable reasons, the Legislature amended it to relieve victims of the burden and trauma of testifying about the details of the predicate convictions. (Otto, supra, 26 Cal.4th at p. 208; see also ibid. [Legislature acted in response to prosecutors’ complaints about having to “ ‘bring victims back to court to re-litigate proof of prior convictions’ ”]; Whitman, supra, 54 Cal.3d at pp. 1070, 1072 [comparable reasoning for Proposition 115 context].) In light of its purpose, we have interpreted the provision as allowing the prosecution to prove the facts of a defendant’s prior qualifying convictions not just with certain documents (like evaluations) but also with multiple-level-hearsay statements contained therein (like police and probation reports, and victim and witness statements they include). (Otto, at pp. 207–208.)

But the Legislature carefully limited the scope of this hearsay exception to one category of proof: establishing predicate convictions. (See, e.g., Bennett, supra, 39 Cal.App.5th at p. 877; see Otto, supra, 26 Cal.4th at p. 211 [hearsay under § 6600, subd. (a)(3) is sufficiently reliable for this purpose, because “some portion, if not all, of the alleged conduct will have been already either admitted in a plea or found true by a trier of fact after trial”].) Courts agree that section 6600, subdivision (a) does not broadly permit the wholesale admission of an evaluation report: Any hearsay to prove the details of nonpredicate convictions, like Walker’s 1989 and 2005 offenses, would be inadmissible under the subdivision. (See, e.g., Couthren, supra, 41 Cal.App.5th at p. 1012; Burroughs, supra, 6 Cal.App.5th at pp. 410–411.) As we commented in People v. Stevens (2015) 62 Cal.4th 325, 338, the subdivision shows that “the Legislature knows how to adopt special rules of evidence to govern commitment proceedings.” It clearly has not elected to do so under section 6602 for hearsay evidence regarding nonpredicate offenses. (Couthren, at pp. 1012–1013.) And if section 6602 already permitted courts at probable cause hearings to broadly admit hearsay like this via evaluation reports, the need to amend section 6600, subdivision (a)(3) is, for the probable cause context, not immediately obvious, and potentially surplusage. (Cf. Couthren, at p. 1015.)

The Court of Appeal’s argument that section 6600, subdivision (a)(3) represents an inappropriate comparison point fails to persuade. The court pointed to two apparent differences between the “function and purpose” of the hearsay exception in the subdivision and its counterpart in section 6602. (Walker, supra, 51 Cal.App.5th at p. 701.) As it explained, section 6600, subdivision (a)(3) “functions as a hearsay exception that not only applies at SVP probable cause hearings, but also extends to SVP trials” and was intended to “ ‘relieve victims of the burden and trauma of testifying about the details of the crimes underlying the prior convictions.’ ” (Walker, at p. 701.) By contrast, the court reasoned, the section 6602 hearsay exception applies only at probable cause hearings, and it serves to allow the People to make an initial showing without putting on a mini trial. (Walker, at p. 701 [victims and witnesses may be spared from testifying at the hearing, but that does not represent the hearsay exception’s rationale].) Despite these considerations, Couthren’s point still stands: If section 6602, subdivision (a) already provided a broad hearsay exception for probable cause hearings, the Legislature could have aimed section 6600, subdivision (a)(3) specifically at the trial context, instead of covering the probable cause context as well. And in any event, it is notable that the Legislature has created a hearsay exception that permits “[t]he details underlying the commission of an offense that led to a prior [predicate] conviction” to be established through documentary evidence (§ 6600, subd. (a)(3)), but has not done the same for the details underlying the commission of nonpredicate offenses, despite the similar potential burden on victims who are called to testify.

A reading of section 6605 in context also offers a telling comparison to section 6600, subdivision (a)(3). That section provides that when a committed SVP defendant petitions for unconditional release, the court must order “a show cause hearing.” (§ 6605, subd. (a)(1).) In Cheek, we explained that section 6605 “resembles” section 6602 because the provisions use “parallel language” and both provide hearings that are pretrial in nature and afford a defendant the right to be present and represented by an attorney. (Cheek, supra, 25 Cal.4th at pp. 899–900.) After Cheek, Proposition 83 amended section 6605 to provide that “the court . . . can consider the petition and any accompanying documentation provided by the medical director, the prosecuting attorney, or the committed person” at the show cause hearing. (§ 6605, subd. (b) [now subd. (a)(1)] as amended by Prop. 83, § 29, as approved by voters, Gen. Elec. (Nov. 7, 2006); see § 6604.9, subd. (f).) This amendment, like the section 6600, subdivision (a)(3) amendment, shows it is possible to adopt special hearsay rules for SVPA proceedings, but that we have no comparable indication that such rules operate in probable cause hearings. (Cf. Couthren, supra, 41 Cal.App.5th at p. 1016, fn. 6.)

That the Legislature can suspend evidence rules in analogous contexts — but chose not to do so in SVPA probable cause hearings for hearsay concerning nonpredicate offenses — is reinforced by provisions in the Welfare and Institutions Code. Consider, for example, the Lanterman–Petris–Short Act (LPS Act; § 5000 et seq.), the general civil commitment statute governing the treatment of mentally ill persons in California. (Conservatorship of Susan T. (1994) 8 Cal.4th 1005, 1008 (Susan T.).) The LPS Act provides that an appointed hearing officer must conduct a “certification hearing” in order for the state to commit an individual beyond the initial 72-hour evaluation and treatment period (unless the detainee has already filed a petition for writ of habeas corpus). (Susan T., at p. 1009.) The certification hearing, similar to the section 6602 hearing, determines whether probable cause exists to detain individuals because they remain a danger to themselves or others. (Susan T., at p. 1009; § 5256.4.) But unlike the section 6602 hearing, the certification hearing “shall be conducted in an impartial and informal manner in order to encourage free and open discussion by participants. The person conducting the hearing shall not be bound by rules of procedure or evidence applicable in judicial proceedings.” (§ 5256.4, subd. (b), italics added.)

The People raise a variety of arguments that largely track the Court of Appeal’s reasoning (Walker, supra, 51 Cal.App.5th at pp. 695–699) and fail to persuade. They first argue that the subdivision establishes an implied hearsay exception — covering any hearsay in evaluation reports — by mandating that courts “review the petition” and thereafter assess probable cause. (Welf. & Inst. Code, § 6602, subd. (a).) This argument fails not only because it relies on the same dubious chain of inferences the Court of Appeal relied on, but also by comparison to the chief authority the People rely on, In re Malinda S. (1990) 51 Cal.3d 368 (Malinda). There, we interpreted Welfare and Institutions Code section 281’s language directing juvenile courts to, in reaching a jurisdictional determination, “ ‘receive and consider’ ” social studies prepared by probation officers or social workers. (Malinda, at p. 376, italics added by Malinda.) We concluded this language created an implied hearsay exception reaching multiple-level hearsay in the reports. (Id. at pp. 376–379; see id. at p. 379 [“Because the reports must include, inter alia, a statement of the minor’s feelings and thoughts concerning the pending action (Civ. Code, § 233, subd. (b)), these reports necessarily contain hearsay and even multiple hearsay”].) Section 6602, subdivision (a) lacks comparable language: There’s no direction that the courts consider expert evaluations. But even assuming the Legislature contemplated that courts would consider such evaluations, nothing in the SVPA definitively indicates that the subdivision permits courts to consider all of the multi-level hearsay contained in such reports, including hearsay concerning nonpredicate offenses.

The People also contend that the Legislature specifically contemplated that the evaluations would contain hearsay like accounts regarding nonpredicate offenses, because the “standardized assessment protocol” in Welfare and Institutions Code section 6601, subdivision (c) requires consideration of a broad array of historical information in hearsay sources. Quoting the court below (Walker, supra, 51 Cal.App.5th at pp. 696–697), the People urge that the Legislature “ ‘clearly intended’ ” for evaluators to rely on these hearsay sources in their evaluations, “ ‘as the alternative would be to require’ ” evaluators to embark on the “ ‘near-impossible task’ ” of “ ‘reinvestigat[ing] a lifetime worth of historical information comprising the person’s “criminal and psychosexual history.” ’ ”

But both strands of this argument fail. The standardized protocol merely assures that the experts offer their professional medical judgments within the “specified legal framework” establishing statutory criteria for committing an individual as an SVP. (Ghilotti, supra, 27 Cal.4th at p. 910, italics added by Ghilotti.) In no way does it allow or direct admission of hearsay in expert reports regarding the facts associated with offenses that didn’t lead to predicate convictions. And the absence of a hearsay exception for such evidence at probable cause hearings does not necessarily impose a near-impossible burden on experts or the People. First, the facts that certain offenses are alleged might be admissible for nonhearsay purposes. Also, at least some of the hearsay documents identified by the People and Court of Appeal — e.g., probation reports, as well as court, prison, and medical records — may still come in without too much difficulty, provided they don’t include further inadmissible hearsay material. (See, e.g., Evid. Code, §§ 1271 [business records], 1280 [official records], 452.5, subd. (b)(1) [properly certified records of conviction].) Moreover, nothing precludes the experts from, in forming their opinions, relying on inadmissible hearsay “that is of a type that reasonably may be relied upon by” those experts. (Evid. Code, § 801, subd. (b); see Sanchez, supra, 63 Cal.4th at p. 685.)[4] And, of course, the Legislature can also enact reforms to address any further practical concerns.

We’re also not persuaded that the experts’ role justifies admitting every single line in their reports. The People contend that because the experts are “neutral” evaluators applying the standardized protocol, the evaluation reports have a degree of reliability and trustworthiness that supports an implied hearsay exception for their full admission, including any hearsay they contain regarding nonpredicate offenses. As support, the People analogize to Malinda, supra, 51 Cal.3d at pages 375–378, 385. Setting aside the markedly different statutory language in that case (ante, at p. 27), Malinda also critically differs in terms of the nature of the hearsay evidence and expertise at issue. There, we did agree with similar arguments that the Legislature implicitly created a hearsay exception, but we did so in part based on a judgment about the reliability of the hearsay evidence: The relevant experts, social workers, would bring knowledge and expertise to bear in the use of the hearsay information in the social study reports they authored. (Id. at p. 377.) In particular, the social workers prepared the social studies on the basis of direct interviews with the minor and her parents (id. at pp. 373–374), and they related the contents of these interviews as part of their statutory role: as “ ‘a special arm of the court to investigate the status of the children and report’ ” back (id. at p. 377, fn. 8; see also id. at pp. 377–379).

Here, no similar justification exists for concluding the Legislature has implicitly allowed psychologists to relate, via their reports, hearsay accounts of nonpredicate criminal offenses. Yes, these experts offer diagnoses that fall within a properly qualified mental health professional’s expertise, and in doing so they often do draw insight from a comprehensive array of sources. (Couthren, supra, 41 Cal.App.5th at pp. 1010–1011.) But these circumstances surrounding the validity of the experts’ medical diagnoses, and the experts’ apparent objectivity and neutrality, are beside the point. Nothing about these circumstances indicates the Legislature has impliedly determined the experts have relevant expertise to be able to relate the reliability either of (a) hearsay accounts in law enforcement documents like police or probation reports, which may have been prepared years or even decades ago, or (b) further levels of hearsay, like victim statements, contained therein. (Malinda, supra, 51 Cal.3d at p. 377; see Couthren, at p. 1018, fn. 7; Bennett, supra, 39 Cal.App.5th at p. 884, fn. 6; cf. Whitman, supra, 54 Cal.3d at pp. 1072–1074.)

To begin with, this evidence presents some inherent reliability concerns. As the People admit, “the reliability of victim hearsay statements in [police and probation reports] is lessened where, as here, the defendant has not been convicted of the crimes to which the statements relate.” (Cf. Otto, supra, 26 Cal.4th at p. 211.) And, more importantly, we have no particular reason to believe it would be consistent with the legislative design to conclude the mental health evaluators bring any professional judgment to bear in assessing the veracity of these hearsay statements — as the facts of this case underscore. As we’d expect for any psychological expert, it doesn’t appear that either MacSpeiden or Karlsson had any meaningful basis to assess the reliability of the two dismissed rape allegations in the probation and police reports, including what the alleged victims told investigating officers. The experts readily admitted that they simply assumed these documents had accurate information, and they presented the information as accurate in their reports. Given the reliability concerns, we think it implausible that it was within the ambit of the legislative purpose to allow the admission of this information as evidence merely because experts chose to include it in their evaluation reports.

Finally, the People’s analogy to Conservatorship of Manton (1985) 39 Cal.3d 645 fails, too. Manton addressed an LPS Act provision relating to conservatorship proceedings for gravely disabled persons. That provision, section 5354, subdivision (a), directs a county officer to investigate alternatives to conservatorship and render a written report to the court prior to the initial conservatorship hearing; and it provides that the court “may receive the report into evidence” at the hearing “and may read and consider the contents thereof in rendering its judgment.” We held that this section does not permit the subsequent use of the report at trial, explaining in part: “If the report were admissible at both the initial hearing and a subsequent court trial, the two proceedings would be essentially identical in terms of the acceptable range of evidence to be considered. We believe that the better interpretation is one avoiding such redundancy in the absence of clear legislative intent to the contrary.” (Manton, at p. 651.) The People focus on this reasoning, arguing that, as in Manton, the language and structure of the SVPA indicate that it does not require duplicative evidence at the probable cause hearing and trial.

Neither the language nor the structure of the SVPA compels us to apply Manton’s reasoning here. In contrast to section 5354 of the LPS Act, Welfare and Institutions Code section 6602 does not mention the reports. Without legislative guidance to the contrary, the same evidentiary rules, i.e., the Evidence Code, govern the probable cause hearing and trial — which therefore expectedly leads to the potential for some duplicative evidence at the proceedings. The same holds true for criminal preliminary hearings, for example. (See LaFave et al., 4 Criminal Procedure (4th ed. 2015) § 14.3(a), p. 365, fn. 35 (LaFave) [California’s criminal preliminary hearing is relatively akin to a “mini-trial hearing,” even in the wake of Prop. 115, in that its rules potentially increase the rigor of its screening function by generally limiting the prosecution to the use of evidence that would be admissible at trial]; but cf. LaFave, § 14.4(b) at p. 383 [unlike California, “perhaps a majority” of jurisdictions “start from the premise that the rules of evidence,” including hearsay rules, do not apply to the preliminary hearing, and they entrust magistrates to take appropriate account of the reliability and weaknesses of such evidence].)

The Legislature can, of course, create a hearsay exception that prevents any duplication of evidence. (Cf. Parker, supra, 60 Cal.App.4th at p. 1469 [noting that the Legislature can “fill the procedural gap in section 6602”].) It can choose to permit hearsay involving prior nonpredicate crimes to come in through evaluation reports. Nothing in our analysis should be understood as taking a position on whether such an exception ought to exist. That is a distinct question from what we must resolve: whether a hearsay exception does, in fact, exist in section 6602. The Court of Appeal blurred the distinction between the two inquiries. It concluded that a hearsay exception supports the practical role of the probable cause hearing, and then reasoned backward, without any statutory basis, to conclude that such an exception must therefore be implied. If we embraced this reasoning, we’d risk imposing an arrangement that essentially requires consideration of hearsay statements in the reports — a result unmoored from the SVPA’s language, its legislative history, and other indicia of statutory purpose.

True: Allowing experts to relate hearsay accounts of nonpredicate offenses would “streamline the People’s ability to make [its] initial showing without having to duplicate” trial evidence and track down victims and witnesses. (Walker, supra, 51 Cal.App.5th at p. 701; see conc. opn., post, at p. 4.) But Walker raises countervailing considerations: defendants’ interest in especially rigorous evidence testing to bar any potentially unreliable hearsay from being used to prop up a weak petition. (See Couthren, supra, 41 Cal.App.5th at pp. 1014, 1020.) Reasonable minds can certainly disagree on whether hearsay like this, which would be inadmissible at the eventual SVP trial, should come in at a preliminary hearing. (Cf. LaFave, supra, § 14.4(b) at pp. 387–389.) But setting aside these policy judgments, the Legislature’s decision to not carve out an exception for the evidence in dispute in this case — hearsay accounts of nonpredicate offenses introduced via expert reports — reasonably aligns with the hearing’s evidence-testing function, as the Legislature has presently designed it and our prior cases have interpreted it.


We separately address one argument the Court of Appeal presented, and to which the People briefly allude. The Court of Appeal argued that two prior decisions have, consistent with section 6602, already recognized a hearsay exception covering the hearsay report content at issue: Parker, supra, 60 Cal.App.4th at pages 1469–1470, and Cooley, supra, 29 Cal.4th at page 245, footnote 8. (Walker, supra, 51 Cal.App.5th at pp. 691–694, 699–700.) Although exceptions to the hearsay rule may be found in decisional law (Otto, supra, 26 Cal.4th at p. 207), courts rarely exercise their power to create these exceptions, and for good reason (see In re Cindy L. (1997) 17 Cal.4th 15, 27; see also ibid. [“The general rule that hearsay evidence is inadmissible because it is inherently unreliable is of venerable common law pedigree”]). Courts exercise this power only “for classes of evidence for which there is a substantial need, and which possess an intrinsic reliability that enable them to surmount constitutional and other objections that generally apply to hearsay evidence.” (Id. at p. 28.) But nothing in Parker or Cooley can be applied to this case and taken to mean that hearsay accounts of nonpredicate offenses, relayed by expert evaluations, meet these stringent requirements (see ante, at pp. 17–19, 29–31), or that the cases even sought to recognize a relevant hearsay exception.

Parker addressed, as a matter of first impression, “the nature” of the probable cause hearing under section 6602. (Parker, supra, 60 Cal.App.4th at p. 1455; see id. at pp. 1461–1462.) It determined from its statutory analysis that section 6602 affords defendants “a hearing at which [they] could be heard, not merely by counsel pointing out legal deficiencies on the face of the petition, but also by being able to effectively challenge the facts on which the petition was filed, i.e., the underlying attached experts’ evaluations.” (Parker, at p. 1468.) The court explained that the probable cause hearing consequently should allow for the admission of oral and written evidence. (Id. at p. 1469.) It then made the following passing reference to hearsay: “While we believe the prosecutor may present the opinions of the experts through the hearsay reports of such persons, the prospective SVP should have the ability to challenge the accuracy of such reports by calling such experts for cross-examination.” (Id. at pp. 1469–1470.)

Then in Cooley, where we addressed “the scope and substance of the probable cause determination” under section 6602 (Cooley, supra, 29 Cal.4th at p. 235), we remarked on Parker’s hearsay reference in passing. Citing Parker, supra, 60 Cal.App.4th at pages 1469–1470, as part of a footnote in our general overview of the SVPA, we stated: “Although the petitioner is allowed, despite their hearsay nature, to present the contents of any reports that form the basis of the petition as evidence, the alleged sexual predator is allowed to cross-examine the expert concerning the evaluation and can call the expert to the stand for that purpose.” (Cooley, at p. 245, fn. 8.)

The Court of Appeal seized on the two cases’ brief references to hearsay, urging that the “Parker/Cooley rule” allows evaluation reports to be fully admitted at a probable cause hearing, despite their hearsay contents. (Walker, supra, 51 Cal.App.5th at pp. 693, 700.) But neither Parker nor Cooley establish a judicially created exception that would cover hearsay content regarding nonpredicate offenses.

Parker turned on whether due process requires something more than a facial review of the petition under section 6602. In answering this question, it briefly observed that prosecutors should be allowed to present the opinions of the experts through their hearsay reports. Its focus, though, was not on the admission of the reports’ hearsay contents, but instead on fleshing out what due process requires at the hearing to allow defendants a meaningful opportunity to challenge the basis of the petition. Parker therefore “contains no discussion regarding the competency of the multiple hearsay necessarily contained within . . . expert evaluations.” (Couthren, supra, 41 Cal.App.5th at p. 1017.)

Although Cooley did cite to Parker’s statement on the admission of hearsay reports, it did so in a single dictum footnote. Moreover, Cooley addressed the subject “in the context of describing matters which were not disputed by the parties and therefore not analyzed by the court.” (Couthren, supra, 41 Cal.App.5th at p. 1017.) As in Parker, Cooley provided no analysis supporting the admission of the reports or their hearsay contents as competent evidence, or concerning the application of the Evidence Code to SVPA probable cause hearings more generally. Consistent with the opinion as a whole, its focus in the footnote was generally laying out the procedural requirements that protect defendants at SVPA hearings.


The admission of the contested hearsay in the MacSpeiden and Karlsson evaluation reports represented prejudicial error under the standard set forth in People v. Watson (1956) 46 Cal.2d 818, 836 (for a statutory error, we must determine whether it is reasonably probable the result would have been more favorable to appellant absent the error). As described in Cooley, “a determination of probable cause by a superior court judge under the SVPA entails a decision whether a reasonable person could entertain a strong suspicion that the offender is an SVP.” (Cooley, supra, 29 Cal.4th at p. 252, italics added by Cooley.) We believe it is reasonably probable that, absent the erroneously admitted hearsay, the trial judge would not have entertained a strong suspicion that Walker qualified as an SVP.

On the one hand, some of the properly admitted evidence supports the existence of probable cause. Walker’s qualifying offense was a forcible rape of a stranger. MacSpeiden and Karlsson diagnosed him with particular mental disorders predisposing him to commit sex offenses — with MacSpeiden diagnosing him with “Other Specified Paraphilia, Sexual Activity with Non-consenting Persons” and “Borderline Personality Disorder,” and Karlsson concluding he had “Antisocial Personality Disorder, augmented by a severe level of psychopathy.” And both experts scored him as having a high risk of sexual reoffense under various diagnostic tools.

On the other hand, some of the properly admitted evidence cut against the existence of probable cause. Yanofsky, one of the initial psychologists appointed by DSH to evaluate Walker, concluded that Walker did not qualify as an SVP. In his evaluation report, which the trial court admitted into evidence, he diagnosed Walker as suffering from “Other Specified Personality Disorder (Mixed Features),” i.e., antisocial and narcissistic personality traits. Yanofsky determined these traits did not affect Walker’s “emotional and volitional capacity to such a degree” that it predisposed him to commit criminal sexual acts that would endanger the health and safety of others. He ruled out paraphilia as a differential diagnosis, explaining that Walker’s criminal sexual history, although reflecting a “sexual preoccupation,” did not necessarily appear driven by “deviance” or “to be a sustained pattern” of inappropriate conduct. Although he did score Walker as having a moderate-to-high risk of sexual reoffense under various diagnostic tools, he determined the absence of a predisposing mental health condition was dispositive. His testimony at the probable cause hearing aligned with the conclusions in his report.

Nothing in the record tells us exactly how the trial court settled on its probable cause determination by weighing the competing evidence. But the nature and role of the inadmissible hearsay make it likely that this evidence prejudicially affected the trial court’s determination. (See Bennett, supra, 39 Cal.App.5th at pp. 884–885.)

First, the lurid hearsay details regarding the 1989 and 2005 rape allegations depicted Walker as an individual with a strong propensity and modus operandi for violent sex offenses. (See ante, at pp. 5–6.) In particular, they portrayed him as someone who preyed on women by falsely promising them entertainment and promotion employment; coerced and lured them to isolated locations; suddenly groped them; and ignored their pleas to stop and raped them. The resulting impression enhanced the experts’ credibility and materially supported their conclusions that Walker’s mental health predisposed him to sexual criminal acts and made it likely he would reoffend with predatory behavior. (Cf. Burroughs, supra, 6 Cal.App.5th at p. 412.) We cannot discount the possibility that the nature of the rape allegation evidence impermissibly factored into the trial court’s probable cause determination. Similarly, even though Walker could (and did) cross-examine the experts regarding particular deficiencies of the rape allegation evidence, that did not adequately protect against the error here: the full admission of the reports, and the chance that the trial court would rely on this substantive evidence in its probable cause analysis.

Moreover, as in Bennett, the inadmissible hearsay that the court admitted critically supported the evaluation reports’ conclusions. (Bennett, supra, 39 Cal.App.5th at pp. 884–885.) Had that content been excluded, the state’s case would have been materially weakened.

MacSpeiden emphasized the hearsay in his report. He indicated in the diagnosis section of his report — which contained the hearsay accounts of the 1989 and 2005 rape allegations — that Walker’s “history amply demonstrates that he is sexually disordered with recurrent, intense sexually arousing fantasies and urges which he is unable to exclude from overt sexual behavior with non-consenting persons.” According to MacSpeiden, this “history” included a long track record of illegal sexual behavior, as shown in Walker’s numerous arrests and charges between 1988 and 2007. But the facts and circumstances underlying the two alleged rapes were the only two offenses in this track record, outside of Walker’s predicate conviction, that MacSpeiden had any real details to support his diagnosis. Moreover, as part of the diagnostic scoring to determine Walker’s risk of reoffense, MacSpeiden discussed how the 1989 and 2005 prior rape allegations helped show Walker was “inclined to engage in sexually violent predatory behavior directed toward a stranger, a person of casual acquaintance . . . or an individual with whom a relationship has been established or promoted for the primary purpose of victimization.”

MacSpeiden’s cross-examination testimony aligned with his report. He testified that the 1989 and 2005 rape allegations, which he assumed were true, constituted a central part of his evaluation. He explained that the allegations, along with Walker’s predicate offense, showed Walker had a modus operandi of telling the victims “ ‘I’m going to make you an important person,’ ” suddenly attacking them, and degrading them and acting with anger toward them. And relatedly, at several points MacSpeiden invoked the allegations as part of a “where there is smoke there is fire” type of logic for why he believed Walker qualified as an SVP.

Karlsson similarly indicated in his report that the hearsay details regarding the 1989 and 2005 rape allegations shaped his evaluation. He described these allegations as two of the three rapes Walker committed, the other being the rape underlying Walker’s predicate conviction. In diagnosing Walker with “Antisocial Personality Disorder,” Karlsson explained that the most prominent feature of these sex offenses was “a pattern of pandering/pimping, involving introducing women to the world of prostitution and strip teasing by acting as an adult entertainment promoter,” and the use of “manipulation and coercion to make . . . victims compliant” — all things clearly drawing on the hearsay accounts of the dismissed rape allegations. On cross-examination, he confirmed that the 1989 and 2005 rape allegations informed his report, and without these allegations his opinion could have been different.

In other words, without the inadmissible hearsay, the trial court would have lacked critical evidence to establish the diagnosis and reoffense elements of the SVP determination. (Bennett, supra, 39 Cal.App.5th at p. 885; cf. People v. Yates (2018) 25 Cal.App.5th 474, 487.) For that reason, and because of the inflammatory nature of the hearsay evidence, its admission prejudiced Walker.[5]


When the Legislature enacted the SVPA, it provided safeguards to ensure that only a select group of dangerous sex offenders may be involuntarily committed — safeguards reflecting the Legislature’s judgment with regard to balancing risks to community safety and the liberty interests of individuals facing the prospect of long-term confinement. The probable cause hearing serves as a critical safeguard in this scheme. The provision governing the probable cause hearing, section 6602, subdivision (a), provides for an adversarial hearing and clearly establishes that the superior court must review the petition to determine whether the state has met its evidentiary burden to proceed to trial. What it does not provide, however, is a hearsay exception allowing the prosecution to introduce hearsay regarding nonpredicate offenses via expert evaluations. Nothing in the language of the subdivision, its legislative history, its place in the broader SVPA statutory scheme and relationship with other provisions, or comparisons to other analogous Welfare and Institution Code provisions indicates the existence of a hearsay exception for such hearsay in expert evaluations. The introduction of this hearsay prejudicially affected Walker’s ability to challenge the basis of the state’s petition and the sufficiency of the evidence to proceed to trial.

We reverse and remand with instruction to the Court of Appeal to, in turn, remand the matter to the superior court for a new probable cause hearing consistent with this opinion.


We Concur:







Concurring Opinion by Chief Justice Cantil-Sakauye

I concur in Justice Cuéllar’s majority opinion, which holds that the lack of an express hearsay exception in the statute governing sexually violent predator (SVP) probable cause hearings precludes the admission of hearsay regarding nonpredicate crimes contained in expert evaluation reports. I am concerned, however, that our ruling will complicate, if not frustrate, the intended screening function of SVP probable cause hearings, and I write separately to urge the Legislature to provide additional guidance addressing the proper conduct of such hearings.

A hearsay exception is not the only provision missing from Welfare & Institutions Code section 6602 (section 6602), the statute governing SVP probable cause hearings under the Sexually Violent Predator Act (SVPA; Welf. & Inst. Code, § 6600 et seq.). The majority is generous in characterizing the guidance provided by that statute regarding the conduct of such hearings as “spare.” (Maj. opn., ante, at pp. 11, 17, 22.) Section 6602 instructs the trial judge tasked with conducting a probable cause hearing merely to “review the petition” in order to “determine whether there is probable cause to believe that the individual named in the petition is likely to engage in sexually violent predatory criminal behavior upon his or her release.” (Id., subd. (a).) Taken on its own terms, this instruction is problematic. A trial judge can no more discern probable cause to believe a person may be an SVP from reviewing the commitment petition than a judge conducting a preliminary examination hearing can determine whether there is probable cause to believe a crime was committed by reviewing the criminal complaint. Both pleadings presumably contain the allegations necessary to support a claim of criminality or SVP status, but neither provides the evidentiary foundation necessary to a finding of probable cause. Additional proceedings are required.

Although section 6602 anticipates that a probable cause hearing will occur (id., subd. (a)), it provides no guidance about the nature of the hearing, as we recognized in Cooley v. Superior Court (2002) 29 Cal.4th 228, 245, fn. 8 [“The SVPA does not provide any specific procedural requirements for the probable cause hearing”]. Applying constitutional principles in an early decision, the Court of Appeal in In re Parker (1998) 60 Cal.App.4th 1453 (Parker) formulated an outline for the conduct of SVP probable cause hearings that has stood unchallenged, at least by this court, until our decision today. Other recent developments in the law, however, had already threatened to undermine the Parker procedures. Parker, ratified by our subsequent decision in Cooley, assumed that an SVP probable cause hearing would focus on the evaluation reports prepared by the two psychiatric professionals whose concurrence is required before an SVP commitment petition can be filed. (Welf. & Inst. Code, § 6601, subd. (d).) It permitted introduction of such reports at the probable cause hearing, notwithstanding their status as hearsay, as well as any hearsay they might contain. (Parker, at pp. 1469–1470 [“the prosecutor may present the opinions of the experts through the hearsay reports of such persons”]; see also Cooley, at p. 245, fn. 8 [“the petitioner is allowed, despite their hearsay nature, to present the contents of any reports that form the basis of the petition as evidence”].)

At the time of Parker, supra, 60 Cal.App.4th 1453, admission of the contents of the evaluation reports was uncontroversial because expert witnesses were permitted to testify concerning the basis for their opinions, even if those opinions were premised on hearsay. (E.g., People v. Montiel (1993) 5 Cal.4th 877, 918.) Parker therefore had no reason to opine separately on the admission of the type of hearsay considered today and, as the majority notes, did not do so. (Maj. opn., ante, at pp. 34–35.) That practice came to an end with People v. Sanchez (2016) 63 Cal.4th 665, which held that case-specific hearsay on which an expert relies is offered for its truth and must be supported by admissible evidence. (Id. at pp. 682–683.) Relying in part on Sanchez, one Court of Appeal has already ruled that the absence of a hearsay exception in section 6602 requires the exclusion of all otherwise inadmissible hearsay at an SVP probable cause hearing, including any such hearsay in the expert evaluation reports. (People v. Superior Court (Couthren) 41 Cal.App.5th 1001, 1010 [“We conclude that the rules of evidence apply in an SVP probable cause proceeding and therefore the admissibility of documentary evidence such as expert evaluations will be governed by the hearsay rule and any applicable exceptions”].) That would also appear to be the logical consequence of the rationale employed in our decision today, although the opinion disavows ruling on the issue. (Maj. opn., ante, at p. 14.) At a minimum, our decision requires the redaction from the evaluation reports of any accounts of prior nonpredicate crimes committed by the alleged SVP, assuming no admissible evidence is presented at the hearing to support that hearsay. But if the absence of an express hearsay exception requires exclusion of this type of hearsay, consistency likely implicates the exclusion of all other inadmissible hearsay as well.

I am concerned that these rulings will prevent the SVP probable cause hearing from serving its intended purpose, which I understand to be an efficient screening function to determine whether a trial is required. Like a probable cause hearing before a criminal trial, the purpose of a section 6602 hearing is, in general terms, to permit the court to ensure that there is an adequate evidentiary foundation for an SVP finding. By requiring the county to present admissible evidence of any nonpredicate offenses that form a part of that foundation, and potentially of any other hearsay found in the evaluation reports, our decision will convert the probable cause hearing into a proceeding barely distinguishable from a subsequent trial on the merits. Although such a hearing can, of course, serve the screening function, it will do so at the cost of time consuming and unnecessary efforts, imposing a potentially sizable burden on counsel and courts that will likely be duplicated at trial.

It was this concern for the efficient conduct of SVP probable cause hearings that led the Court of Appeal below to imply a hearsay exception into section 6602. Although I agree with my colleagues that we lack legal authority to recognize such an exception in these circumstances, I am otherwise wholly sympathetic to the Court of Appeal’s well-articulated concerns. Our present decision appears to be the beginning of the end for the time-tested Parker procedures.

The only solution for this problem is a legislative one. The Parker procedures have served as a fair and efficient guide to the conduct of SVP probable cause hearings for more than 20 years. I encourage the Legislature to make the statutory amendments necessary to preserve those procedures, beginning with an exception for hearsay contained in the expert evaluation reports. Even better, I hope the Legislature will reexamine SVP probable cause hearing procedures and formulate clear statutory guidelines for the conduct of such hearings. Our polestar is to implement our Legislature’s intent, but reliable implementation is difficult when, as in section 6602, there is little statutory indication of that intent.



  1. Further unspecified statutory references are to the Welfare and Institutions Code. ?
  2. We grant Walker’s request that we take judicial notice of the SVPA petition, the docket and the People’s writ petition in People v. Superior Court (Couthren) (2019) 41 Cal.App.5th 1001 (Couthren), and Department of Corrections and Rehabilitation forms.  (Evid. Code, § 452, subds. (d), (h).) ?
  3. Though the parties agree that the Parker rule properly allows for the admission of the expert reports, they disagree on its second step: the SVP’s right to cross-examine the report authors. On the one hand, this step, like Parker’s first step, strikes us as a potentially reasonable extrapolation to make regarding the conduct of the probable cause hearing — particularly since (a) the hearing requires more than section 6601.5’s “paper review,” and (b) cross-examination enables defendants to easily test the basis of the experts’ reports, regardless of whether the People have introduced this basis into evidence in the first instance via the reports. But we need not decide this particular issue, since this case turns on the first-order question of the admissibility of specific report content. ?
  4. Under these circumstances, the SVP might still challenge the basis of the experts’ evaluation, including by cross-examination (under the Parker procedure). (See People v. Valencia (2021) 11 Cal.5th 818, 838, fn. 16; but see ante, at pp. 11–12, 14 [this case does not require us to review the Parker procedure].) ?
  5. We decline to reach the second issue briefed by the parties: whether defendants in SVPA proceedings have a due process right to confront and cross-examine witnesses presenting contested hearsay evidence. Based on our state law holding, we need not further consider what due process requires. (People v. Williams (1976) 16 Cal.3d 663, 667.) ?