In Re Scott

61 P.3d 402, 129 Cal. Rptr. 2d 605, 29 Cal. 4th 783, 2003 Daily Journal DAR 991, 2003 Cal. Daily Op. Serv. 800, 2003 Cal. LEXIS 525

• Court: California Supreme Court
• Decided: January 27, 2003
• Docket: S059739
• Status: Published

Opinions

Opinion

CHIN, J.

Petitioner James Robert Scott seeks relief on habeas corpus from the judgment of death entered against him in Los Angeles County Superior Court, Case No. A747321. We affirmed that judgment on direct appeal. (People v. Scott (1997) 15 Cal.4th 1188 [65 Cal.Rptr.2d 240, 939 P.2d 354] (Scott).) In this matter, we issued an order to show cause based on allegations that (1) trial counsel failed to investigate and to adequately present a mental defense at the guilt phase, (2) trial counsel failed to investigate and to adequately present mitigating evidence at the penalty phase, and (3) these failures rendered counsel’s advice to waive a jury uninformed and hence ineffective. We appointed the Honorable Howard J. Schwab, Judge of the Los Angeles Superior Court, as our referee, and directed him to take evidence and make findings of fact on specified questions. Judge Schwab has done so and has issued a report responding to the questions.

We adopt the referee’s factual findings. In accordance with those findings, we also conclude petitioner has failed to carry his burden of establishing ineffective assistance of counsel. Accordingly, we discharge the order to show cause and, in a separate order, deny the petition for writ of habeas corpus.

I. Facts

A. The Underlying Judgment

Petitioner assaulted and raped Wanda Jensen in her home, then set her on fire. In 1986, before she died, he pleaded guilty to her rape and attempted murder. In 1988, after she died, he was charged with her murder. He waived a jury trial. After a court trial in 1989, the court convicted him of first degree murder, found true special circumstance allegations of rape and burglary murder, and found that he intended to kill Jensen and used a deadly weapon. It then imposed the death penalty.

We described the underlying facts in our opinion in the direct appeal. (Scott, supra, 15 Cal.4th at pp. 1199-1200.) Around 2:00 a.m., on April 22, 1986, petitioner entered Jensen’s apartment in Palmdale, placed a screwdriver against her side, threatened to harm her five-year-old daughter, hit her with a baseball bat, raped her, beat and choked her into unconsciousness, set her on fire, and left. Jensen’s daughter rescued her, but she died of her bums on February 25, 1987. Petitioner confessed to the crimes in two statements to separate investigators. Glenn “Reran” Johnson, Jr., testified that petitioner came to his home shortly before the crime. He gave petitioner some socks and the two smoked cocaine. Johnson denied it at trial, but he had previously said that petitioner had asked him for a screwdriver and gloves, and he gave him a screwdriver but had no gloves. Petitioner told the police that before the crime, he purchased some cocaine and “did a couple lines of coke.”

At times during his two confessions, petitioner claimed the persona of “Tony.” At one point in the first confession, petitioner “glared at the investigator and said ‘he was now Tony and that Tony had taken charge of James’ body, and he said he had to destroy the girl because she was a shit bomb . . . .’ Before the second confession, [petitioner] signed the waiver card with the name, ‘Tony Adman.’ When questioned about the signature, he changed it to his true name. He told both investigators that he was ‘Tony’ and that ‘James was a wimp.’ ” (Scott, supra, 15 Cal.4th at p. 1199.)

Petitioner presented a defense that Jensen had received negligent medical treatment and would not have died had the treatment been competent.

At the penalty phase, the prosecution presented evidence that in 1983, petitioner assaulted Paula H. with a knife in her house and said he was going to rape and kill her. A struggle ensued in which petitioner cut and bit her, after which she escaped. For this incident, he pleaded guilty to assault with a deadly weapon. Three weeks before the assault on Jensen, petitioner assaulted Violet H. in her home with a knife and his fists, raped her, and choked her into unconsciousness. For this incident, at the same time that he pleaded guilty to Jensen’s rape and attempted murder, he pleaded guilty to rape with use of a knife and infliction of great bodily injury.

“The defense presented some of Paula H.’s testimony at the 1983 preliminary hearing of that prosecution and a psychological evaluation of [petitioner] prepared in conjunction with that prosecution.” (Scott, supra, 15 Cal.4th at p. 1200.) It presented no other mitigating evidence, although trial counsel argued in mitigation the medical malpractice evidence he presented at the guilt phase.

B. The Reference Hearing

1. Background

We asked the referee to take evidence and make findings of fact on these questions.

“1. What actions did petitioner’s trial counsel take to investigate a possible mental defense at the guilt phase? What were the results of that investigation? What additional evidence supporting a mental defense, if any, could petitioner have presented at the guilt phase? What investigative steps, if any, would have led to this additional evidence? How credible was this additional evidence? What circumstances, if any, weighed against the investigation or presentation of this additional evidence? What evidence rebutting this additional evidence reasonably would have been available to the prosecution?

“2. What actions did petitioner’s trial counsel take to investigate potential evidence in mitigation at the penalty phase? What were the results of that investigation? What additional mitigating evidence, if any, could petitioner have presented at the penalty phase? What investigative steps, if any, would have led to this additional evidence? How credible was this additional evidence? What circumstances, if any, weighed against the investigation or presentation of this additional evidence? What evidence rebutting this additional evidence reasonably would have been available to the prosecution?

“3. Did petitioner himself request the curtailment of or do anything to hinder or prevent the investigation or presentation of evidence supporting a mental defense at the guilt phase or in mitigation at the penalty phase? If so, what did he do or request?

“4. Why did trial counsel advise petitioner to waive a jury? Did any failure by trial counsel to investigate a mental defense or mitigating evidence render counsel’s advice to waive a jury uninformed? Would any additional investigation have affected the advice to waive a jury and, if so, what and how?”

At the evidentiary hearing, petitioner presented the testimony of his trial counsel, William Clark; three mental health experts; several friends and family members; Richard Vargas, who had been with petitioner the evening of the crime; and an attorney. The prosecution presented the testimony of one mental health expert and Burrell Ford, who had lived with petitioner’s mother for three years. After receiving briefing, the referee issued a detailed report. At the beginning, he summarized his findings: “The referee will find that petitioner’s trial counsel’s investigation of potential mental defenses and mitigating penalty phase evidence was minimal. However, the referee also will find that petitioner in the habeas corpus evidentiary hearing did not proffer any significant credible evidence that would have been discovered if there had been a more thorough investigation by trial counsel. Further, the referee will find that petitioner himself requested the curtailment of the investigation and presentation of evidence by requesting not to have his family involved. The referee also will find that trial counsel’s advice to petitioner to waive jury was based upon his concerns about the nature of the jury panels in the Antelope Valley (where the case would be tried), his desire to try the case before a certain highly skilled jurist, and his belief that a defense of intervening medical malpractice would be best presented before that particular judge rather than before a jury.”

The referee then went into detail. He discussed each witness’s testimony and made credibility determinations, then answered our specific questions. Because we find the report reliable and helpful and, as discussed below, we adopt its credibility determinations, we review it in detail, especially the testimony the referee found credible.¹

2. The Witnesses’ Testimony and Their Credibility

a. Trial Counsel

The referee found trial counsel, Clark, to be “totally credible.” Clark was first appointed to represent petitioner in 1986, before Jensen died, regarding the Jensen case and the crimes involving Violet H. “He read the felony complaints, the initial police reports and petitioner’s record. [^[] The initial reports stated that petitioner had taken on the persona of an individual named ‘Tony.’ Mr. Clark had his client checked for competency to stand trial. He was evaluated by Dr. Kaushal Sharma and Dr. Michael Maloney who found him competent. Based upon the doctors’ reports, Mr. Clark ‘largely’ determined he would not raise a mental competency issue. As Ms. Jensen at the time was taking a ‘very bad turn,’ Mr. Clark decided it would be best to waive a preliminary hearing and petitioner pled guilty to the attempted murder and rape of Ms. Jensen and the rape of [Violet H.] in hopes that this might deter the district attorney from seeking capital charges if Ms. Jensen should die.

“In 1986, petitioner never seemed to be physically agitated. He changed his demeanor upon his return from the competency hearing and would giggle when the name ‘Tony’ was mentioned. This was corroborated also by trial counsel’s investigator’s report. Mr. Clark was aware that petitioner’s mother had talked to him after he was in custody and told the police that he needed ‘psychological help.’ However, he did not request funding for an expert at that time.

“When petitioner was returned to stand trial on the capital charges, Mr. Clark had several discussions with him. Petitioner . . . was not talkative and there were no lengthy discussions. Petitioner did not offer or indicate anything to his lawyer about childhood physical abuse from any members of his family or any people with whom he was living and mentioned nothing remarkable about his childhood. Mr. Clark was familiar with the comment by petitioner in Dr. Maloney’s report that up to the age of five, his childhood was not ‘all that shabby.’ [Dr. Maloney’s report also stated, “Mr. Scott denies any physical abuse in the early family home. He states that he was never placed out of the home and suffered no significant traumas during his early years.”] Mr. Clark questioned petitioner about his early childhood and did not recall him offering any information about any childhood physical abuse. There was nothing ‘remarkable’ about petitioner’s childhood based upon those interviews. Mr. Clark did not recall being told by petitioner that he was molested by Mr. Douglas or Mr. Ford or physically beaten by his mother. Although Mr. Clark read in Dr. Michael Maloney’s report about petitioner being sexually molested when he was five years old by a stranger, petitioner did not dwell on that issue.

“In addition, petitioner told his lawyer that he had committed the crimes with which he was charged. He told Mr. Clark that he had engaged in forcible intercourse with Ms. Jensen, lit the fire (including how he had done it and where it started), and had a bat with him at the time of the assault. He also admitted raping [Violet H.], while having a knife and a bat with him at the time. Petitioner never blamed the crimes on anyone else but ‘Tony,’ and later, petitioner told Mr. Clark that Tony was a ‘con.’ [^|] Mr. Clark remembered inquiring of petitioner about contacting his family members. It was discussed at some length on more than one occasion. Mr. Clark’s recollection was that petitioner was adamant that he did not wish trial counsel to involve his family in any way. This was petitioner’s attitude in both 1986 and 1988.

“Trial counsel had spoken to petitioner’s mother, Ora Thornton, around the time of the ‘preliminary hearing of the trial phase of the case.’ She did not indicate that she had anything to offer. Although petitioner had told Mr. Clark not to get involved with family, trial counsel attempted to contact the mother for the penalty phase in order to test petitioner’s position of whether or not he wanted his family involved. However, Mr. Clark was not sure he would have called petitioner’s mother to the stand, but he was not in a position to follow up with her if there was anything to be followed up on as he could not find her. Mr. Clark also spoke to one of petitioner’s sisters who tried to offer an alibi inconsistent with what petitioner had told him. Trial counsel did not remember meeting any other family members. Mr. Clark had never told anyone they could not attend any of the court proceedings in the homicide case.

“Mr. Clark believed there was an absence of any presence of overwhelming psychiatric factors. Petitioner had told him there ‘ . . . was no Tony and he was a con.’ [^[] Mr. Clark had hired his father-in-law, Walter D. Hill, a former Los Angeles Sheriff’s officer, as an investigator. In a report prepared by Mr. Hill, dated November 30, 1988, it was noted that petitioner had told the investigator that he had taken various drugs the night of the crimes and did not remember anything when they took place. Mr. Clark also knew that petitioner had taken Mellaril, researched it, and believed it was something he could not ‘muster’ in terms of a defense. He did not remember subpoenaing petitioner’s California Department of Corrections records during the capital trial because he had seen some in the hands of the prosecution.

“Petitioner did not display any symptoms or signs of drug abuse to Mr. Clark. A 1986 probation report indicated that petitioner was not a habitual drinker or user of marijuana. This was consistent with what petitioner had told him in 1986. He was also aware of Dr. Sharma’s 1986 report relative to a history of drug abuse including PCP, marijuana and cocaine. Mr. Clark believed that Dr. Maloney’s report contraindicated a mental defense in the matter. In light of the reports of Dr. Sharma and Dr. Maloney, Mr. Clark believed it was in his best tactical interest to take the avenue of the defense of medical malpractice.

“As to the general investigation of the offenses, Mr. Clark recalled in his 1988-1989 investigation that he talked to various Los Angeles Sheriff officers he knew regarding the case. He did not contact any psychiatrists and had only a ‘nominal’ consideration of mitigating evidence other than intervening medical malpractice at the penalty phase. Mr. Clark planned to use the medical malpractice defense at both the guilt phase and the penalty phase. Mr. Clark believed that petitioner was not responsible for the death by reason of the medical malpractice. Counsel did not want to present a ‘cafeteria defense’ ....

“Mr, Clark had difficulty in locating Glenn ‘Rerun’ Johnson and did not interview him before the trial. When Mr. Johnson testified, Mr. Clark utilized his testimony as a basis of ‘diminished actuality’ although he did not remember such at the time of the evidentiary hearing. [(See Scott, supra, 15 Cal.4th at p. 1216.)] fl[] No individual, including family members, offered any information which Mr. Clark could utilize in the penalty phase.

“Trial counsel had several reasons why he wanted to waive jury in this matter. His practice was primarily in the Antelope Valley where he had tried many cases. He knew that the jurors there were conservative, and he was familiar with the attitude of the local populace. Mr. Clark was concerned about putting evidence of at least two or three rapes in front of a jury by reason of their ‘cross-racial’ nature. Mr. Clark had discussed the subject of the jury waiver with petitioner on multiple occasions and gave petitioner an opportunity to think about it for weeks. Trial counsel explained to his client the defense of intervening medical malpractice. The jury waiver was as to only one particular jurist, Los Angeles Superior Court Judge Margaret Grignon [now an Associate Justice of the Court of Appeal]. She had experience as a judicial officer both in municipal court and superior court. Mr. Clark had been in front of Judge Grignon numerous times. The judge appeared to be a very smart individual, a ‘ . . . skyrocket in the legal community.’ In addition, Judge Grignon was married to a physician and her father-in-law was a physician who practiced in the Lancaster/Antelope Valley area for a number of years. Counsel believed that she was better able to grasp intricacies of a medical malpractice defense than a jury.”

Trial counsel said he did nothing to investigate a possible mental defense in 1989, but “in 1986, he had a ‘general survey’ which was part of the reason that reports had been requested. He believed that once the 1986 proceeding was concluded, the possibility of a psychiatric defense had been ruled out, based upon the findings of the two doctors and based upon the settlement of petitioner’s symptoms. Mr. Clark noted that ‘ . . . each of these reports has come back and indicated that the defendant had manifested certain symptomology in front of them which they did not believe to be real, that they believed to be feigned.’ Mr. Clark further noted that the problems did not seem to rise to the level of being ‘clinically significant.’ He did not believe that the use of any experts was warranted by reason of the two reports that he had in his possession. He believed issues about ‘Tony’ were ‘not a viable place to go ... .’ When petitioner returned for trial, Mr. Clark did not further investigate a mental defense because of ‘[petitioner’s] initial comments that did not seem to be at all warranted, or criminal.’ ”

When asked what he did to investigate potential mitigating evidence, Clark said “that the mitigation as to petitioner was the fact that he was not ‘. . . responsible for what happened to Ms. Jensen.’ ” Clark said that he did not investigate other possible evidence because he believed “it did not seem to be where the case ‘needed to go.’ Although Mr. Clark believed he argued issues of mental impairment at trial, he did not mount a full-blown investigation in that direction, flj] He believed that such would be inconsistent with ‘ . . . where the case had gone in chief.’ Nothing in the experience of petitioner seemed to make those viable alternatives that could be established with any degree of certainty to the trier of fact, [f] Mr. Clark had not gone into petitioner’s background because he had discussed that issue in the framework of Dr. Maloney’s report and petitioner had not ‘ . . . complained about them.’ Petitioner did not want to involve his family in the case at all.”

When asked why he had advised petitioner to waive a jury, “Mr. Clark said that he felt that the nature of the arguments that were going to be raised in the trial and their complexity would not be well received by a jury ‘tactically’ and would be seen as an ‘excuse.’ He believed that such ‘ . . . would be better situated with someone like Judge Grignon.’ When asked if Mr. Clark would have changed his mind, in advising his client to waive jury, if he had investigated ‘the mental defense or other mitigating evidence,’ he responded ‘ . . . no, I felt that the tenor of the medical malpractice case was such that it warranted the jury waiver.’ ”

b. Petitioner’s Family and Friends

The referee found the testimony of petitioner’s family and friends generally had “very little credible value. The testimony on the whole of these individuals appeared to be exaggerated, histrionic and recently fabricated. It should be noted in this regard that Carrie MacMurray, petitioner’s sister, and Cynthia Ford Geiggar and Candius Ford, petitioner’s half sisters, when they testified at the evidentiary hearing, all gave an alibi as to petitioner’s whereabouts on the night of the crimes against Ms. Jensen. In contrast, petitioner admitted to the authorities and to his trial counsel his involvement with the rape and fire at the Jensen residence. In this regard, petitioner’s mother, Ora Thornton, also believed that her son was innocent of the rape and murder of Ms. Jensen.” The referee also noted that “[n]ot only did Ms. MacMurray, Ms. Geiggar and Ms. Ford proffer an alibi at the evidentiary hearing on behalf of petitioner, their versions thereof were inconsistent.” The report recounted these witnesses’ testimony. In general, they painted a grim picture of petitioner’s unloved childhood, replete with abuse by his parents and others, including Burrell Ford, and abuse of drugs and alcohol. Some reported that petitioner, when young, had attempted suicide.

The referee found that Natalie S., “who had been made pregnant by petitioner when he was 22 and she was 14, had no credibility.” Natalie S. was the daughter of Violet H., the victim of the rape petitioner committed shortly before the capital crime. “Her claim that petitioner had a consensual affair with her mother . . . was belied by petitioner’s admissions to the crime to his counsel and the blood found at the scene after the incident by [Ms. S.] herself. [Ms. S.’s] accusations that Mr. Clark told her that she was not going to testify because petitioner (using an offensive racial epithet) was going to ‘firy’ is nothing less than ludicrous.”

The referee found some of these witnesses credible. “Of all the family members, Leander Thornton appeared to be the most credible. He had married petitioner’s mother. He testified that the mother smoked marijuana in petitioner’s presence, that petitioner intervened [in] a fight between Mr. Thornton and the mother on the mother’s behalf, that petitioner and the mother smoked marijuana together, that petitioner loved his sisters and protected them in any way he could, and that the mother disciplined petitioner by striking him in the face with her hand, but that he did not remember ever seeing her abuse him.”

“Richard Vargas . . . did appear to have credibility with the referee. Richard Vargas testified that on the night of the Jensen crimes, petitioner had asked Mr. Vargas for a ride to purchase some rock cocaine and that petitioner appeared to be ‘under the influence.’ In Mr. Vargas’ opinion, petitioner appeared to he ‘high on cocaine’ as he was ‘real twitchy,’ ‘antsy-like’ and ‘hyper.’ ”

The referee also found to “have some credibility” a witness the People called, Burrell Ford, who had lived with petitioner’s mother for three years. “He testified that he neither physically nor sexually abused petitioner, and that he never saw the mother use drugs in petitioner’s presence. He also denied giving petitioner narcotics or alcohol. He noted that the mother would ‘take off and leave the family at times and that he never saw the mother do anything sexually ‘improper’ towards petitioner. He also testified that he never saw the mother beat petitioner [in the ways that some of the other witnesses testified] and in fact never saw the mother punish petitioner. Mr. Ford had punished petitioner once when he and Cynthia Ford set the trash can on fire [by spanking him with his belt].”

“In summary, the referee finds that the testimony of family and friends was in large part unbelievable and inferentially recently fabricated after the sentence of death had been imposed.”

c. The Experts

Three mental health experts, Dr. Dale Watson, Dr. Roderick Pettis and Dr. Julie Kriegler, testified for petitioner and one, Dr. Kaushal Sharma, testified for the People. The referee found the testimony of Drs. Watson, Pettis, and Kriegler had “very little weight,” as their opinions “were based in large part on the statements and declarations of petitioner’s family, and further, were based upon examinations done years after the crimes of which petitioner had been convicted.”

“Dale Watson, a clinical psychologist, had performed a neuropsychological evaluation in June or August of 1996 and met with petitioner for approximately 11 hours, giving him a battery of tests. Dr. Watson gave the opinion that petitioner operated in the low average to borderline range of intellectual ability, that there was ‘evidence’ that petitioner suffered from mental illness and posttraumatic stress disorder although he made no such specific findings, that petitioner had a history of substance dependence, that he had been a victim of physical and sexual abuse that had left ‘substantial signs of psychological damage,’ that he lived a chaotic life in early childhood, that he had suffered episodes of depression and attempted suicide and was an individual who experienced a great deal of trauma. However, much of Dr. Watson’s opinion was based upon what petitioner had told him, which was not admitted for the truth asserted but merely as a basis of Dr. Watson’s opinion. Dr. Watson testified that all the information he received about petitioner’s life was derived from petitioner himself and he did not verify the accuracy of that information. Dr. Watson testified that if the information that petitioner told him was not accurate, such would certainly affect his report.

“Dr. Roderick Pettis, a psychiatrist, . . . first examined petitioner 11 years after the crimes and obtained information derived from petitioner, family members, experts hired by petitioner and various records. He never personally interviewed petitioner’s family members; rather, he learned information from petitioner’s family through declarations.

“Dr. Pettis interviewed petitioner for approximately three hours in 1997. In 2001 he interviewed petitioner twice, once for approximately four hours and the second time for ‘about’ two and a half hours. Dr. Pettis opined that petitioner was vulnerable to mental illness and substance abuse by reason of family history of the same. He further felt there was a very compelling history of universal trauma and very little opportunity to develop a healthy sense of ‘self.’ He believed that petitioner suffered from ‘hyper-arousal’ and longstanding mental health symptoms which included depression, substance abuse, disorganized thinking, psychosis and borderline cognitive functioning. He opined that petitioner suffered a combination of posttraumatic stress disorder plus psychosis based upon both genetics and an environment of sexual and physical abuse and the utilization of drugs, and further, had difficulty in abstract thought. Dr. Pettis believed that his opinions were corroborated by the fact that when petitioner was taken to County Jail he was placed on Mellaril, an antipsychotic drug, and placed in four-point security restraints. There was no indication that petitioner suffered from multiple-personality disorder, and further petitioner never mentioned ‘Tony.’ Dr. Pettis’ diagnosis was that on the night of the murder, petitioner was in a disorganized and dissociative state and that events on his mind were not going in a normal linear fashion but rather ‘things are popping up and down’ and going out of order. He therefore saw ‘ . . . things in terms of a motion picture which is not running on track.’ Petitioner was not capable of deciding whether or not he would satisfy his lust, whether or not to beat Ms. Jensen, or deciding whether or not to light the match.

“The problem with Dr. Pettis’ diagnosis of dissociative disorganized state is that it is not borne out by the evidence. The crimes of rape and the burning were committed in a logical sequence. In addition, Mr. Scott admitted committing the crimes to two Deputy Sheriffs and also to his attorney Mr. Clark. Dr. Pettis opined that either (1) the confessions were suggested by the authorities, or (2) the defendant filled in the gaps unconsciously, even if it harmed himself and even though he did not actually remember the facts, or (3) the confessions were true. Dr. Pettis conceded, however, that if the confessions were true, it might be inconsistent with being dissociative at the time. However, he stated that by reason of his state of intoxication petitioner’s weighing of the pros and cons or his ability to make a moral choice might have been impaired.

“The theory of Dr. Pettis is belied by the ‘Tony’ defense when petitioner confessed to the Sheriffs office and used the word ‘Tony’ in [an] attempt to create a false mental defense of a multiple personality. Petitioner later admitted to his trial counsel that ‘Tony’ did not exist and was a ‘con.’ As such, the referee finds that the statements of culpability made by petitioner were true and that when he spoke to the Sheriffs office, he deliberately tried to fabricate a multiple personality defense by the use of ‘Tony.’ It is believed that petitioner had knowledge that what he did was wrong by trying to falsify a ‘Tony’ defense when he confessed to the authorities. Dr. Pettis’ suggestion that the statements to the authorities were false by reason of their being suggested or unconsciously manufactured is belied by petitioner intentionally mixing the statements in a devious, well-thought-out manner with the ‘Tony defense.’ Even Dr. Pettis found no evidence of multiple personality disorder, a diagnosis rejected as well by Dr. Sharma.

“Dr. Julie Kriegler, a licensed psychologist, . . . had created a psychological social history. This history had encompassed the time up to early 1986. Dr. Kriegler interviewed petitioner on July 14, August 4, and October 16, 2000. The interviews each lasted up to three hours. Dr. Kriegler also interviewed the mother, Ora Mae Thornton, and the sister, Carrie MacMurray. She also referred to declarations of other individuals. Dr. Kriegler held the opinion that there was a pattern of significant mental disorders that had spanned through his great-grandparents’ generation and that petitioner was emotionally, medically, nutritionally and educationally neglected by an array of inconsistent caretakers who were not capable of appropriate parenting. Dr. Kriegler believed that petitioner suffered severe, protracted physical and psychological abuse and had few opportunities to develop outside systems of support. Dr. Kriegler also had the opinion that petitioner suffered ‘severe protracted sexual abuse,’ was raised in communities of violence, and experienced ‘high levels’ of violence in the home. She believed that petitioner had borderline intellectual functioning and impaired cognitive abilities which restricted his ability to cope. ft[] Based upon her evaluation, she found that petitioner suffered from substance abuse and was exposed to a significant amount of negative risks and a lack of positive factors. She believed that the cumulation of these factors contributed to a severe psychological decomposition. Her diagnosis of petitioner included posttraumatic stress disorder, multiple mood disorders and a chemical dependency.

“However, Dr. Kriegler’s investigation was based in large part upon family interviews and discussions with petitioner himself. As already noted, the referee finds that the family members’ testimony in large part was not credible. While petitioner himself gave details to Dr. Kriegler, he was very parsimonious in his discussions with Mr. Clark and gave no such information during trial counsel’s representation. Since the bulk of Dr. Kriegler’s opinion was based on information which the referee finds to be unbelievable, her opinions carry very little weight.

“Dr. Kaushal Sharma, a psychiatrist, ... is the only mental health professional who testified who had examined petitioner shortly after the crimes.[²] He believed that petitioner was presenting himself as more mentally impaired than suggested by authentic mental illness. He noted that petitioner feigned both confusion whenever the doctor asked questions about the charged crimes and a headache whenever the doctor questioned him regarding the facts of the case. He opined that petitioner did not appear to suffer from dissociative identity disorder. Dr. Sharma was not too concerned about the administration of Mellaril to petitioner while in custody in 1986. At that time the drug was used as a chemical restraint in institutional jail settings and Dr. Sharma noted that petitioner had been given the drug only after being involved in a jailhouse physical altercation. He also believed that there was no sign of psychosis or posttraumatic stress disorder. In this regard there was no evidence that petitioner had posttraumatic stress disorder flashbacks while in court, and he had not been diagnosed with the syndrome until the late 1990’s. Dr. Sharma had reviewed the declarations prepared by petitioner’s family members and friends and found that they lacked credibility, were inconsistent and based upon multiple hearsay .... Dr. Sharma rejected the reports prepared by Dr. Watson, Dr. Pettis and Dr. Kriegler as he believed the information relied upon [by] those professionals was lacking in credibility.”

The referee summarized his credibility findings regarding the mental health experts. “Since the testimony of the medical experts proffered by petitioner were based primarily on statements by family and friends and petitioner himself, they also carry little weight. This is especially true since petitioner’s statements to the health experts conflicted with what he told defense counsel whose testimony the referee finds to be credible in recounting their discussions.”

Petitioner also called Edward Rucker, an attorney, who “had great credibility in the referee’s eyes. He testified that social histories were in vogue with regard to penalty phases of capital cases in 1989. However, he also explained that in the late 1980’s information for a social history was gathered by an investigator and not amassed by health professionals and sociologists.”

3. Referee’s Answers to Questions

a. Whether Petitioner Hindered the Investigation or Presentation of Evidence

The referee first answered the question whether petitioner had himself requested curtailment or done anything to hinder or prevent the investigation or presentation of evidence supporting a mental defense or in mitigation, “because it affects the answers to all further questions.” He found petitioner did so in two respects.

First, “In 1986 and 1988, petitioner did not want to discuss or involve his family. Mr. Clark wanted to get into that area on more than one occasion, but it was not a place that petitioner ‘ . . . wanted to go.’ ” Clark could not remember specifically what petitioner had said in this regard in 1986, but he had the “impression” that petitioner did not want to involve his family, and his attitude was always consistent in 1988. In 1988, “petitioner was ‘adamant about not wanting to involve his family in any way.’ ffl] It was Mr. Clark’s normal practice to discuss with the defendant and advise that client of the importance of having family members who are involved cooperate and testify. . . . Mr. Clark abided by petitioner’s wishes in not contacting any family because ‘those were his wishes, those were the bounds of my representation.’ ”³

Second, “petitioner was not very talkative to trial counsel in their discussions relative to giving Mr. Clark information to be used at trial. [^J] Petitioner did not offer or indicate any information about childhood physical abuse and said nothing remarkable related thereto.”

b. Investigation into a Possible Mental Defense

The referee found that trial counsel’s investigation “as to possible mental defenses at the guilt phase was minimal.” Counsel “requested a report on petitioner’s competency to stand trial.” Based on the resulting reports of Drs. Sharma and Maloney, and their beliefs that petitioner had “feigned” certain symptomology, counsel did not believe that hiring other mental health experts was warranted. “Although counsel spoke to petitioner about ‘Tony’ in 1986, his client was vague and would ‘giggle’ whenever ‘Tony’ was mentioned, [f] In 1988 when trial counsel was reappointed, he attempted to accumulate the information initially gathered in 1986 and again interview his client. Trial counsel did not conduct any formal investigation as to mental defenses when petitioner was brought back for trial. However, he did have informal conversations with Sheriffs deputies whom he saw at the courthouse. Also, Mr. Clark spoke again with petitioner about ‘Tony’ and his client told him ‘ . . . that Tony was a con and there was no Tony.’ In addition, trial counsel in 1988 hired an investigator, . . . who interviewed petitioner. Counsel had interviewed petitioner on several occasions and investigated the use of Mellaril.”

“The results of the investigation by trial counsel were findings by two court-appointed mental health experts, Dr. Kaushal Sharma and Dr. Michael Maloney, on the issue of competency to stand trial. Their findings did not support a mental defense. Rather these reports reflected an absence of evidence of mental illness with the opinion that petitioner was a malingerer. Dr. Sharma’s report noted that petitioner had a tendency to express symptoms which ‘may not be’ based on ‘authentic’ mental illness. This was corroborated by the seeming feigning of a headache by petitioner when questioned about the facts of the case and the lack of exhibition of any signs of multiple personality disorder. Dr. Maloney’s report was corroborative of the findings of Dr. Sharma in his report. Although Dr. Sharma’s report indicated that petitioner had no conscious knowledge of the facts of the ‘instant crimes,’ petitioner had admitted to his lawyer committing the crimes against Ms. Jensen and confessed to the authorities as well.

“. . . The report of Mr. Hill [the defense investigator] noted that petitioner said he had drunk ‘Southern Comfort’ and had taken certain drugs with a Derrick Tibbs and a woman called ‘Mater’ until at least 11:30 p.m. and could not remember any events between the hours of 11:30 p.m. and the time he knocked on his sister’s door the following morning. However, this was contradicted ... by petitioner’s recounting to his lawyer and to the authorities the events of the crimes against Ms. Jensen. fl[] Based upon the conversations petitioner had with his lawyer, Mr. Clark did not see any signs his client suffered from any short or long-term substance abuse as he did not exhibit lapses of memory or recollection and was able to account for his activities. Further, Mr. Clark had the opinion that his client did not appear to be an individual who did not know where he had been, what he had done and what he had been doing.”

Regarding what additional evidence supporting a mental defense petitioner could have presented at the guilt phase, the referee noted, “[A] great deal of information [was] presented at the evidentiary hearing by family, friends and mental health experts relative to [petitioner’s] upbringing and mental state. Family members and friends testified as to childhood deprivation, drug and alcohol use, physical and sexual abuse, neglect, discrimination, repeated suicide attempts, and petitioner’s continuing efforts to protect and feed his siblings. This information, along with statements made by petitioner to the defense mental health experts and the mental health experts’ opinions themselves, formed the crux of the evidence presented at the habeas corpus hearing. However, the referee finds that petitioner would not have been able to present any such appreciable evidence at the guilt phase because it believes that the bulk of the matters testified to by family and friends was not credible in that they were recently fabricated and/or greatly exaggerated and that petitioner’s mental health expert opinions proffered at the hearing were based in large part upon this apocryphal material. The referee finds that trial counsel would have been unable to accumulate the great bulk of the materials presented by his able appellate and habeas corpus counsel.

“The fact that trial counsel would have been unable to glean information from petitioner’s family and friends is bolstered by petitioner being ‘adamant’ about his attorney not involving family. In fact, counsel for petitioner had attempted to contact the mother for purposes of the penalty phase to see if his client would change his mind in that regard but was unable to locate her. Furthermore, counsel for petitioner had spoken to his mother around the time of the preliminary hearing and she had nothing to offer. One of the sisters of petitioner who contacted trial counsel merely gave an alibi, which of course was contradicted by his client’s own admissions. Mr. Clark questioned the credibility of this sister. In addition, Mr. Clark did not remember meeting the other family members. In fact, most of the family members only attended the trial sporadically or not at all.” At this point, the report goes into detail about how often the various witnesses attended the trial.

“While petitioner was seemingly loquacious about his alleged past activities and alleged drug use to the health providers, especially after he had been sentenced to death, he was not talkative or informative to his trial counsel when he was being represented by him. As such, the referee finds that it was very doubtful, if not impossible, for trial counsel to have discovered the ‘plethora’ of material presented at the habeas corpus hearing, which formed the vast majority of the basis for the mental health expert opinions presented by petitioner at the evidentiary hearing.

“However, two issues may have been presented based upon further investigation.” First, “trial counsel was aware that his mother had suffered psychiatric difficulties.” Second was Richard Vargas’s testimony regarding petitioner’s condition the night of the crimes.

“Although the mother had a psychiatric background, counsel for petitioner was hampered by his client’s desire that he not involve family, and further, it is questionable as to what could have been discovered as to a possible mental defense by reason of psychiatric problems in the family. The same would apply as to any other alleged history of psychological disorder in petitioner’s family.”

Regarding what investigative steps would have led to this additional evidence, the referee found that “the bulk of [the] matters presented at the evidentiary hearing on behalf of petitioner would never have been discovered no matter what additional investigative steps [were] taken. As to the mother’s psychiatric history, it would have been difficult to obtain such information by reasons of petitioner’s request that family not be involved. Further, there is no showing that such would have led to any fruitful material absent what the referee finds to be recently fabricated material by family and friends, flj] Relative to Mr. Vargas, petitioner’s trial counsel could possibly have been more aggressive in attempting to contact him or persons who had seen petitioner on the night in issue. Such investigation may or may not have resulted in the information testified [to] by Mr. Vargas at the evidentiary hearing.”

Regarding the credibility of this additional evidence, the referee reiterated that he found the bulk of the evidence petitioner presented at the evidentiary hearing not credible. He found Vargas credible but noted that his testimony “would have been in large part duplicative of the testimony of Glenn ‘Rerun’ Johnson, Jr., who testified on behalf of the prosecution at the trial. Mr. Johnson testified that on the night in issue, he was sharing cocaine with petitioner and that ‘we were getting high.’ ”

Regarding what circumstances weighed against investigating or presenting this additional evidence, the referee reiterated much of his previous findings regarding the circumstances confronting trial counsel and noted that counsel wanted to maintain his credibility and not present a “cafeteria” defense. The referee also noted that “the various probation and prison records, investigative reports and other materials were contradictory and in conflict with the issues of drug use and psychiatric background.”

Regarding what rebuttal evidence would have been available to the prosecution, the referee found that “[c]ertain records had information contrary to a mental defense. However, since the referee finds that trial counsel would not have been able to retrieve the bulk of the evidence presented at the habeas corpus hearing, it is questionable what rebuttal evidence, if any, would even have been needed.”

c. Investigation into Potential Mitigating Evidence

The referee found that counsel’s investigation into potential mitigating evidence for the penalty phase “was minimal except for the utilization of intervening medical malpractice as both a frontal attack in the guilt phase and again in the penalty phase. Furthermore, Mr. Clark had attempted to speak to petitioner about his family and about his childhood . . . [and] attempted but was unable to locate the mother for the penalty phase in order to determine from petitioner if he would be allowed to use her.

“As to consideration of issues of mental impairment which did not reach a level of a defense at the guilt phase, Mr. Clark believed he argued those to some degree but did not mount a full-blown investigation in that direction. Mr. Clark did not go into any other issue of mental impairment because he felt it would be inconsistent. There was nothing in Mr. Clark’s experience with petitioner that seemed to make those viable alternative defenses that could be presented with any degree of certainty to the trier of fact. Mr. Clark also had not investigated too deeply into petitioner’s background or youth because he had discussed those issues with petitioner as raised within the framework of Dr. Maloney’s report. Petitioner did not complain about those issues and he had not wanted to involve the family in the case at all.”

Regarding the results of this investigation, the referee noted that counsel obtained “Dr. Marvin Ginsburg, a cardiopulmonary specialist, who testified on the issue of intervening medical malpractice at the trial.” Clark also spoke to petitioner’s mother, who had nothing to offer, to a sister of petitioner, who wished to give an alibi that was contrary to what petitioner had told his lawyer, and to “a young 14-year-old teenager claiming to be pregnant (thought to be Natalie [S.]) who offered no information .... Discussions with petitioner were not particularly fruitful as he was not talkative. He did not offer or indicate anything about childhood physical abuse, and never mentioned being molested. Mr. Clark was aware that petitioner had told Dr. Maloney that up to the age of five, his life was not ‘all that shabby.’ Although counsel was aware of a molestation by a stranger of petitioner when he was five years old by reason of Dr. Maloney’s report, it was not something that petitioner seemed to dwell upon as far as their conversation was concerned. The results of the investigation relative to petitioner’s mental defenses has been discussed in answer to the questions regarding mental defenses at the guilt phase.”

Regarding what additional mitigating evidence petitioner could have presented, the referee reiterated that the “great bulk of the evidence presented at the evidentiary hearing by petitioner’s family and friends was recently fabricated and would not have been available to trial counsel at the time of the investigation or for use at the penalty phase. fl¡] Relative to the mental defenses, the referee believes that the family and friends’ discussions of physical abuse, familial sexual abuse, use of drugs and alcohol, maltreatment, and other related matters were either nonexistent or so greatly embellished that one would never be able to ferret out the truth.”

“While the referee believes that petitioner was raised in a family which was not entirely stable and in which illicit drugs were no stranger, the extent of the alleged chaos of petitioner’s upbringing is greatly in question in light of petitioner’s own statements to his attorney relative to his youth. In this regard, the referee does not accept as true the allegations of physical abuse or sexual abuse within the family and friends, but rather believes them to be a belated attempt by the family and friends to aid petitioner in the habeas corpus proceedings. There are certain issues which may have possibly been raised by investigation, and while the referee has doubts as to their viability, they will be mentioned in this report.

“As discussed earlier, petitioner’s family testified that they loved him and the sisters had testified that petitioner was their protector. This is supported by the testimony of Leander Thornton, the most credible of petitioner’s family .... When Mr. Thornton and petitioner’s mother were having a fight, petitioner hit him. There possibly could have been information gleaned that petitioner supported his family and they in turn loved him. However, such is contraindicated by the fact that petitioner’s family, except for Candius Ford, rarely attended the court proceedings. Again, petitioner’s trial counsel was hampered by his client’s wish that he not get family involved.

“Another possible area of investigation would have been Natalie [S.], as she had testified of petitioner’s kindness to her and helping her with her algebra. However, such testimony would have been fraught with danger since Ms. [S.] was 14 years old at the time she was impregnated by the adult 22-year-old petitioner who knew her to be underaged. It is questionable what further investigation would have brought up regarding the alleged child molestation of petitioner when he was five years old by a stranger as set forth in Dr. Maloney’s report. Whether it was true or not is made questionable in light of the fact that it was not a subject which petitioner seemed to dwell upon. . . . [T]rial counsel may have been able to obtain information from Mr. Vargas relative to a possible drug intoxication on the night of the murder which, while not being a legal defense, would go to an apparent state of mind. However, whether such evidence would have been obtainable is not certain and was in many ways duplicative of the testimony of Glenn ‘Rerun’ Johnson, Jr., at trial.”

Regarding what investigative steps would have led to this additional evidence, the referee reiterated that the bulk of the evidence presented at the evidentiary hearing was recently fabricated and thus could not have been discovered before trial. “However, petitioner’s counsel could have inquired more as to the relationship of petitioner and his family relative to love and loyalty except for the fact that petitioner had told his lawyer that he did not want the family involved. He could also have made more inquiries of Natalie [S.] and attempted to investigate the knowledge of Mr. Vargas as to petitioner’s condition on the night of the murders.”

Regarding the credibility of the additional evidence, the referee largely reiterated his earlier findings. He added, “Natalie [S.j’s testimony would have been highly questionable in light of the fact that she had been impregnated at the age of 14 by petitioner when he was 22. The utilization of mental impairment not equivalent to mental defense would have had the same problems as the utilization of the mental defense itself in the guilt phase because of its basis primarily on fabrications, exaggerations, and embellishments of family and friends.”

Regarding what circumstances weighed against investigating and presenting this additional evidence, the referee reiterated previous relevant findings. “Furthermore, mental impairment relative to a mental defense might be questionable if voluntary use of drugs or alcohol was a primary cause of such alleged impairment. The same reasons why petitioner’s trial counsel did not raise a mental defense except to briefly mention the diminished actuality in argument would apply to this issue as well.”

Regarding what rebuttal evidence might have been available to the prosecution, the referee found it possible the prosecution would have presented evidence that “petitioner had been a member of the Raymond Crips Gang and, further, that he had sold marijuana in the past if family members had been called to testify.” If Natalie S. had testified, the prosecution might have presented evidence that petitioner had “impregnat[ed] a 14-year-old schoolgirl.” “Burrell Ford could have also been called in rebuttal relative to the lack of physical or sexual abuse by family and friends when he had been present.”

d. Reasons for Waiving a Jury

The referee credited Clark’s testimony as to why he advised petitioner to waive a jury. He summarized, “Thus, a primary motivation for the jury waiver was [to be] able to have the case decided by a particular jurist, Judge Grignon, because of the nature of the defense preferred and because of concern about the jurors in the area.” He also found “nothing relative to the investigation that would have changed the dimension of the trial. . . . The decision to waive jury was in no way compromised by any failure of trial counsel’s investigation. . . . ffl] Mr. Clark testified that he still would have waived jury even if he had investigated a mental defense or other mitigating evidence . . . .” The referee concluded, “In light of the nature of the jury pool, the ability to get a highly skilled jurist to hear the matter, and the desire to proffer a defense of intervening medical malpractice in a case involving horrific facts, the referee believes that no additional investigation would have affected the advice to waive a jury.”

II. Discussion

A. General Legal Principles

“Petitioner had the right to the effective assistance of counsel at trial, and thus was ‘entitled to the reasonably competent assistance of an attorney acting as his diligent and conscientious advocate.’ [Citation.] A defendant claiming ineffective representation bears the burden of proving by a preponderance of the evidence both (1) that counsel’s performance was deficient, i.e., that the representation fell below an objective standard of reasonableness, and (2) that there is a reasonable probability that, but for counsel’s unprofessional errors, the result would have been more favorable to defendant, i.e., a probability sufficient to undermine confidence in the outcome. [Citations.]” (In re Ross (1995) 10 Cal.4th 184, 201 [40 Cal.Rptr.2d 544, 892 P.2d 1287]; see also Strickland v. Washington (1984) 466 U.S. 668, 688, 694 [104 S.Ct. 2052, 2068, 80 L.Ed.2d 674] (Strickland))

The United States Supreme Court has recently reemphasized that, in applying these principles, “a court must indulge a ‘strong presumption’ that counsel’s conduct falls within the wide range of reasonable professional assistance because it is all too easy to conclude that a particular act or omission of counsel was unreasonable in the harsh light of hindsight.” (Bell v. Cone (2002) 535 U.S. 685, 702 [122 S.Ct. 1.843, 1854, 152 L.Ed.2d 914].) Accordingly, a court must “view and assess the reasonableness of counsel’s acts or omissions . . . under the circumstances as they stood at the time that counsel acted or failed to act.” (People v. Ledesma (1987) 43 Cal.3d 171, 216 [233 Cal.Rptr. 404, 729 P.2d 839]; also quoted in Scott, supra, 15 Cal.4th at p. 1212.)

“Any conclusions of law, or of mixed questions of law and fact, are subject to independent review. Mixed questions include whether counsel’s performance was deficient and whether the deficiency prejudiced the defense. Because the referee can observe the demeanor of the witnesses and their manner of testifying, findings of fact, though not binding, are entitled to great weight when supported by substantial evidence.” (In re Ross, supra, 10 Cal.4th at p. 201.)

B. Preliminary Matters

Before the evidentiary hearing was held, petitioner moved this court to vacate certain rulings of the referee. We granted the motion in part and denied it in part without prejudice to the parties’ rearguing it, if they chose, when the matter came before us on the merits. The parties have not reargued any part of that motion. “Nevertheless, because our minute orders did not set out the reasoning of the court, and cannot serve as precedent to guide future decisions, we believe it appropriate to explain the basis for our decisions.” (California Assn. of Psychology Providers v. Rank (1990) 51 Cal.3d 1, 9 [270 Cal.Rptr. 796, 793 P.2d 2].)

1. Ruling on Scope of Discovery

The Attorney General moved the referee to obtain the following discovery from petitioner:

“1. All discovery allowed under Penal Code section 1054.3, including reports, handwritten notes, and test data from experts petitioner intends to call at the hearing.

“2. All written or recorded statements by petitioner or reports of petitioner’s statements relevant to his three claims of ineffective assistance of counsel.

“3. All real evidence petitioner intends to offer in evidence at the hearing.”

In response, petitioner, through current counsel, agreed to provide “all non-privileged reports or statements of experts made in connection with the case, including the results of physical or mental examinations, scientific tests, experiments, or comparisons which either side intends to offer in evidence at the hearing,” and all “real evidence” that he intended to use. He also agreed he had waived the attorney-client privilege regarding communications with trial counsel to the extent they were relevant to the ineffective assistance of counsel claims, but he asserted the privilege as to communications with current counsel.

At the hearing, the referee ruled that petitioner must provide discovery of the kind allowed in Penal Code section 1054.3⁴ and, generally, of statements petitioner made to persons other than current counsel and investigators. The referee agreed that communications with current counsel, “as well as any statement [petitioner] may have made to any expert that will not be called,” are privileged. “However,” the referee ruled, “all other statements by Mr. Scott relevant to these issues are discoverable, as well as any statements made to any expert that was called or any expert that the original [trial counsel] may have utilized but decided not to call himself, because that will be an issue in this case. So far as the expert that [current counsel] called that he may use for investigative purposes but he will not call, those are not discoverable. That’s attorney-client privilege.”

Petitioner challenged this ruling in this court. He argued, primarily, that Penal Code section 1054.3, part of the reciprocal discovery provisions of Proposition 115, enacted in 1990, does not apply to a habeas corpus proceeding. We concluded the referee’s ruling came within his discretion and denied petitioner’s motion to vacate this portion of the ruling. Petitioner is correct that Penal Code section 1054.3 does not apply to habeas corpus proceedings. It applies instead to the underlying criminal proceeding. (See Izazaga v. Superior Court, supra, 54 Cal.3d at p. 364.) But it appears the referee merely cited that section as a shorthand way to describe and limit the discovery he was ordering.

The nature and scope of discovery in habeas corpus proceedings has generally been resolved on a case-by-case basis. (See Comparet-Cassani, Evidentiary Hearings in California Capital Habeas Proceedings: What Are the Rules of Discovery? (1999) 39 Santa Clara L.Rev. 409, 425-426.) We have indicated that discovery is available once we have issued an order to show cause, but we have not discussed the question further. (People v. Gonzalez (1990) 51 Cal.3d 1179, 1258-1261 [275 Cal.Rptr. 729, 800 P.2d 1159]; In re Avena (1996) 12 Cal.4th 694, 730 [49 Cal.Rptr.2d 413, 909 P.2d 1017].)

If, as Proposition 115 provided, discovery is reciprocal at the criminal trial itself—where the defendant is presumed innocent and has no burden of proof—it certainly should be so on habeas corpus, where guilt has been established and the petitioner bears the burden of proof. (See People v. Duvall (1995) 9 Cal.4th 464, 474 [37 Cal.Rptr.2d 259, 886 P.2d 1252].) (Petitioner received discovery from the People relevant to the issues before the referee.) Penal Code section 1054.3 was a logical place for the referee to look to fashion a fair discovery rule. It requires the defendant to provide the names, addresses, and statements of witnesses, expert reports, and real evidence the defendant intends to offer. This requirement is not onerous and could greatly facilitate the reference hearing. Additionally, by claiming trial counsel provided ineffective assistance, petitioner waived the attorney-client privilege to the extent relevant to the claim. (Evid. Code, § 958; In re Gray (1981) 123 Cal.App.3d 614, 616 [176 Cal.Rptr. 721].) Accordingly, the referee also properly ordered discovery of petitioner’s relevant statements to his trial attorney. The referee agreed that petitioner’s statements to current counsel and current experts whom petitioner did not intend to call as witnesses were not discoverable, and he therefore excluded those statements from the discovery order. We find the referee acted within his discretion in ordering this limited discovery.

2. Ruling on Petitioner’s Testimony

The Attorney General also requested, and the referee made, two related orders regarding calling petitioner as a witness: (1) the prosecution may call petitioner as a witness, although petitioner may assert the right against self-incrimination as to specific questions; and (2) if petitioner does exercise the right against self-incrimination, the referee will “make findings of fact adverse to petitioner as to each question that he is ordered to answer but declines to answer on the grounds of self-incrimination.” The referee based both parts of this ruling on orders we had issued several years earlier in another case in which we had ordered a reference hearing. The referee recognized that those orders were not binding on him, but he found them persuasive and followed them.⁵ Petitioner challenged the order in this court, and we agreed it was erroneous in one respect.

The first part of the order was correct. Californians have both a constitutional and a statutory right not to be called as a witness against themselves in a criminal case. (Cal. Const., art. I, § 15 [“Persons may not ... be compelled in a criminal cause to be a witness against themselves”]; Evid. Code, § 930 [“a defendant in a criminal case has a privilege not to be called as a witness and not to testify”].) This privilege extends to proceedings that are “criminal in nature,” such as a civil contempt proceeding. (In re Witherspoon (1984) 162 Cal.App.3d 1000, 1001 [209 Cal.Rptr. 67].) But it does not extend to proceedings that are essentially civil in nature. Thus, in Cramer v. Tyars (1979) 23 Cal.3d 131, 137-138 [151 Cal.Rptr. 653, 588 P.2d 793], we held the privilege did not apply in commitment hearings of mentally retarded persons. (See also People v. Whelchel (1967) 255 Cal.App.2d 455, 460-461 [63 Cal.Rptr. 258] [narcotics rehabilitation program]; Conservatorship of Bones (1987) 189 Cal.App.3d 1010, 1015-1016 [234 Cal.Rptr. 724] [involuntary postcertification treatment].)

We believe a habeas corpus proceeding like this one is civil in nature for these purposes.⁶ The Legislature has labeled it a “Special Proceeding[] of a Criminal Nature” (Pen. Code, pt. 2, tit. 12, ch. 1, before §§ 1473-1508), but the label is not dispositive. (In re Head, supra, 42 Cal.3d at p. 226; see Pen. Code, § 10004.) It is not itself a criminal case, and it cannot result in added punishment for the petitioner. Rather, it is an independent action the defendant in the earlier criminal case institutes to challenge the results of that case. (France v. Superior Court (1927) 201 Cal. 122, 126-127 [255 P. 815, 52 A.L.R. 869].) Accordingly, the petitioner does not have the privilege not to be called as a witness. However, if called as a witness, the petitioner does have the right to assert the privilege against self-incrimination as to individual questions, as does any witness in any proceeding. (Evid. Code, § 940; see Cramer v. Tyars, supra, 23 Cal.3d at pp. 137-138.)

We vacated the second part of this ruling, however, and directed the referee “not to draw any inference if petitioner properly asserts the privilege against self-incrimination.” Evidence Code section 913, subdivision (a), provides as relevant: “If in the instant proceeding ... a privilege is or was exercised not to testify with respect to any matter, . . . neither the presiding officer nor counsel may comment thereon, no presumption shall arise because of the exercise of the privilege, and the trier of fact may not draw any inference therefrom as to the credibility of the witness or as to any matter at issue in the proceeding.” (Italics added.) The comments of the Assembly Committee on Judiciary about this statute states that it “eliminates any remaining basis for applying a different rule in civil cases”; it makes “clear that, in civil cases as well as criminal cases, inferences may be drawn only from the evidence in the case, not from the claim of privilege.” (Assem. Com. on Judiciary com., 29B pt. 3 West’s Ann. Evid. Code (1995 ed.) foll. § 913, p. 168.) The Evidence Code “applies in every action before the Supreme Court . . . including proceedings in such actions conducted by a referee . . . .” (Evid. Code, § 300.) Accordingly, it applies to this action. In a reference hearing, as in other hearings, the trier of fact may not draw adverse inferences from any witness’s invocation of the right against self-incrimination.⁷

C. Exceptions to the Referee’s Report

The parties have filed numerous exceptions to the report. The Attorney General’s only exception is to the referee’s description of trial counsel’s investigation into a possible mental defense and potential mitigating evidence as “minimal.” He argues the investigation was adequate under the circumstances. We need not resolve this question as an exception to the report, for we will consider it in resolving the legal issues.

As might be expected, because the referee’s report was generally unfavorable to him, petitioner states many exceptions. All lack merit.

1. Petitioner argues that Judge Schwab was ineligible to act as our referee. After we appointed him, Judge Schwab informed the parties that he had been the trial judge in an earlier capital trial. (See People v. Cudjo (1993) 6 Cal.4th 585 [25 Cal.Rptr.2d 390, 863 P.2d 635].) Petitioner’s trial counsel had been the defense attorney in that case also, and the defendant, like petitioner, later claimed the attorney had represented him ineffectively. (In re Cudjo (1999) 20 Cal.4th 673, 679 [85 Cal.Rptr.2d 436, 977 P.2d 66].) Based on this circumstance, petitioner asked this court to appoint a new referee, which we declined to do. Petitioner argues that Judge Schwab was disqualified under Code of Civil Procedure section 170.1. He does not contend that any of the specific grounds for disqualification under that section exist. Instead, he relies on the general provision requiring disqualification whenever “a person aware of the facts might reasonably entertain a doubt that the judge would be able to be impartial.” (Code Civ. Proc., § 170.1, subd. (a)(6)(C).) He argues, “Because Judge Schwab had an extensive prior relationship with trial counsel and the opportunity to observe [trial counsel] defending Mr. Cudjo in a capital proceeding, which led to unsuccessful challenges both directly and collaterally, a person aware of the fact that Judge Schwab presided at the Cudjo case might reasonably entertain a doubt that Judge Schwab would be able to be impartial as these evidentiary proceedings unfolded.” We disagree.

It appears the provisions of Code of Civil Procedure section 170.1 apply to a judge we have appointed to act as our referee. That section applies to any “judge,” which Code of Civil Procedure section 170.5 defines as “judges of the superior courts, and court commissioners and referees.” (Italics added; see People v. Gonzalez, supra, 51 Cal.3d at pp. 1247-1248, fn. 44.) The standard for disqualification under Code of Civil Procedure section 170.1, subdivision (a)(6)(C), is objective, but potential bias and prejudice must clearly be established. Courts must apply with restraint statutes authorizing disqualification of a judge due to bias. (Roitz v. Coldwell Banker Residential Brokerage Co. (1998) 62 Cal.App.4th 716, 723-724 [73 Cal.Rptr.2d 85].)

No one could reasonably doubt the ability of a judge to impartially determine facts regarding the performance of an attorney solely because that attorney had previously appeared before the judge. Judges are routinely called on to determine a variety of questions involving attorneys who appear before them, including claims of ineffective assistance of counsel. Indeed, in People v. Fosselman (1983) 33 Cal.3d 572 [189 Cal.Rptr. 855, 659 P.2d 1144], we held that criminal defendants may raise claims of ineffective assistance of counsel in a new trial motion, to be decided by the trial judge. “It is undeniable that trial judges are particularly well suited to observe courtroom performance and to rule on the adequacy of counsel in criminal cases tried before them.” (Id. at p. 582.) Here the basis for the disqualification motion is not Judge Schwab’s observation of the attorney in this case, but in a different case. That circumstance does not make him any less able to act as a referee than a trial judge who decides a new trial motion claiming counsel acted ineffectively in the same action.

2. Petitioner contends that the referee’s actions at the hearing showed he “lacked the neutrality required” of a referee. The contention is based in part on petitioner’s evidentiary arguments discussed below. In addition, he complains that the referee “took an active role in questioning witnesses.” The referee did, in fact, question many of the witnesses, but doing so was proper. In general, judges are entitled to question witnesses as long as they are temperate, nonargumentative, and scrupulously fair. (People v. Hawkins (1995) 10 Cal.4th 920, 947-948 [42 Cal.Rptr.2d 636, 897 P.2d 574]; see Evid. Code, § 775.) When the judge is acting as a referee seeking answers to factual questions a reviewing court has posed, and thus no jury is involved, questioning witnesses is especially appropriate. We encourage our referees to take an active role in asking whatever questions they believe will assist in fully and accurately determining the facts. Referees should permit the parties fully to present, and examine and cross-examine, witnesses, as this referee did. But referees should also feel free to ask additional questions, as this referee also did. Petitioner asserts that the referee’s questioning was aimed at obtaining agreement with the referee’s own views. We disagree. All questioning was fair and designed to determine the facts, not reach a predetermined result.

Petitioner also claims the referee misunderstood his role. At one point the referee stated, “I may ask questions myself since this case is quasi-inquisitorial.” Later, he clarified that he did not mean the case was an “inquisition,” but rather it was “almost like a European procedure where you have one court that is given questions by another court, and makes findings, and the other court makes the final decision.” This description was correct. We ask our referees only to make findings on disputed factual questions; we then resolve the legal issues ourselves. (In re Ross, supra, 10 Cal.4th at p. 205.) The record demonstrates that the referee correctly understood his role and acted as an impartial fact finder for this court.

3. Petitioner challenges certain evidentiary rulings. First, he complains that, except for the limited purpose of impeaching trial counsel’s testimony, the referee refused to permit him to present expert legal testimony concerning the standard of care for capital trial counsel. The referee acted correctly. The questions we asked him were factual only and do not involve the standard of care or whether trial counsel’s performance was deficient. We reserved those questions for ourselves. (See In re Ross, supra, 10 Cal.4th at p. 205.) The referee was not required to consider expert testimony irrelevant to the specific questions. (See id. at pp. 214-215.) It is true that referees have sometimes heard and considered expert legal testimony, but in those cases the testimony came within the scope of our questions. (E.g., In re Avena, supra, 12 Cal.4th at p. 720 [whether counsel’s investigation was “inadequate”]; In re Neely (1993) 6 Cal.4th 901, 908 [26 Cal.Rptr.2d 203, 864 P.2d 474] [what “should” counsel have done, and “should” counsel have taken certain steps]; People v. Mayfield (1993) 5 Cal.4th 142, 198, fn. 9 [19 Cal.Rptr.2d 836, 852 P.2d 331] [“should” counsel have discovered potential defense witnesses]; In re Jackson (1992) 3 Cal.4th 578, 584 [11 Cal.Rptr.2d 531, 835 P.2d 371] [whether counsel “failed to provide adequate representation”]; In re Fields (1990) 51 Cal.3d 1063, 1068 [275 Cal.Rptr. 384, 800 P.2d 862] [whether defendant was “deprived of his right to effective assistance of counsel”].) Because of the nature of the contentions, we do not require, and did not ask for, such testimony in this case. Instead, having received and considered the referee’s factual report, this court will make the ultimate legal determination whether petitioner received effective assistance of counsel based on our judgment and experience in reviewing criminal matters.

Petitioner also complains that the referee did not permit him to present evidence that he did not commit the crime. The question arose when petitioner, through current counsel, cross-examined trial counsel regarding the exact facts of the crime. The referee expressed doubt about the relevance of this cross-examination. Petitioner responded it was relevant to show whether his confession “might have been attacked.” The referee noted that the evidence showed petitioner had twice confessed to the police and, later, also to trial counsel. He stated that petitioner could cross-examine trial counsel regarding petitioner’s statements, but otherwise, he said, “I don’t see any issue as being relevant as to looking for third party culpability in this case, ft[] It is not a question asked to me by the California Supreme Court. I don’t see how it applies herein, especially in light of the fact that there were two confessions to the police and he confessed to” trial counsel.

The proffered cross-examination was not directly relevant to the questions we asked the referee to decide, although factual innocence, if established, could be considered mitigating. But in this case, the trial testimony established that petitioner fully confessed to the crime, not once but twice to separate investigators. The evidence at the reference hearing, which the referee credited, additionally established that petitioner also told trial counsel he committed the crimes and supplied details. So far as the record of the trial and this proceeding shows petitioner has never repudiated those confessions. In stating his exceptions to the report, petitioner repeatedly refers to his “alleged” or “purported” confessions. But we are not dealing with mere allegations, but confessions proven at trial and at the reference hearing. The referee allowed petitioner to fully litigate whether his statements to trial counsel actually occurred and their precise nature, and to challenge counsel’s testimony in this regard. Petitioner could have testified and contradicted trial counsel’s testimony, but he did not do so. Trial counsel’s uncontroverted testimony fully supports the referee’s finding that petitioner did, in fact, confess yet again to trial counsel. This confession confirmed the confessions to the two investigators.

Petitioner claims that confessions are sometimes false and suggests they might have been false here. Dr. Pettis testified that the confessions to the investigators may either have been suggested by the investigators or confabulated by petitioner.. What is missing, however, is any suggestion from petitioner himself that these confessions—or his later confession to trial counsel—are false, or that he is, in fact, innocent. Under the circumstances, the referee acted within his discretion in limiting petitioner’s cross-examination of trial counsel regarding the truth of the confession. (In re Johnson (1998) 18 Cal.4th 447, 459 [75 Cal.Rptr.2d 878, 957 P.2d 299].)

Petitioner argues the refusal to permit evidence of factual innocence was especially unfair because the referee found some of the family members’ testimony incredible due to their proffering an alibi inconsistent with petitioner’s confessions, and the referee also noted that petitioner’s mother believed he was innocent. The referee based his disbelief in the proffered alibis both on the fact they were inconsistent with petitioner’s repeated confessions and that they were inconsistent with each other. Moreover, as he made clear when the parties argued the matter after the evidentiary hearing, the referee based his credibility determinations largely on the witnesses’ demeanor when testifying: “I looked at the witnesses carefully. Based upon my examination of their demeanor, I just believed that the testimony of the family and friends was exaggerated, was histrionic, and was recently fabricated. That just seemed to permeate this courtroom.” Under these circumstances, and especially absent any claim by petitioner himself that his repeated confessions, including those to trial counsel, were in fact false, we see no unfairness in the referee’s fact finding.

Petitioner claims the “referee had an erroneously narrow view of the scope of mitigating and mental health evidence, and, as a result, improperly excluded and/or refused to consider important mitigating and mental-health-related evidence.” Throughout the hearing, the referee limited mitigating evidence to matters involving petitioner himself, and not merely his family or others. Early in the hearing, the referee ruled, “In a death penalty case, what is admissible in the penalty phase are the facts of the crime and the special circumstances and anything dealing with the background and character of the defendant, including sympathy for the defendant. ... [1] We cannot go beyond that to sociological aspects that have nothing to do with the defendant personally. You can link things to the defendant personally, of course. Then they are admissible. But things of general social ills in and of themselves cannot be the basis of mitigating factors. They have to be related in some way specifically, and fact-specifically, to Mr. Scott.” In accordance with this general ruling, the referee refused to admit certain evidence regarding petitioner’s family and conditions at home that were not linked to petitioner. These rulings were correct. “[T]he background of the defendant’s family is of no consequence in and of itself. That is because under both California law [citation] and the United States Constitution [citation], the determination of punishment in a capital case turns on the defendant’s personal moral culpability. It is the ‘defendant’s character or record’ that ‘the sentencer . . . [may] not be precluded from considering’— not his family’s. [Citations.] [10 To be sure, the background of the defendant’s family is material if, and to the extent that, it relates to the background of defendant himself.” (People v. Rowland (1992) 4 Cal.4th 238, 279 [14 Cal.Rptr.2d 377, 841 P.2d 897].) The referee admitted evidence linked to petitioner personally. For example, he admitted and discussed in his report evidence that trial counsel knew that petitioner’s mother had had “psychiatric difficulties.” He merely excluded evidence not linked to petitioner at all. We see no abuse of discretion.

Finally, petitioner argues the referee erred in excluding expert testimony regarding the impact of petitioner’s “mental state upon his purported Miranda waivers [(Miranda v. Arizona (1966) 384 U.S. 436 [86 S.Ct. 1602, 16 L.Ed.2d 694, 10 A.L.R.3d 974])], and upon the reliability of any statements he may have made.” During the direct examination of Dr. Pettis, the referee stated they were not “going to get into the issues of Miranda waivers,” because “the issues, as I said before, will be those raised by the California Supreme Court. Nothing further.” Accordingly, although petitioner presented extensive evidence regarding his mental state at the time of the crimes and as mitigating evidence, he did not present evidence specifically relating to how his mental state impacted his Miranda waivers. The referee did not err. He did not exclude evidence regarding the reliability of petitioner’s confessions. His ruling was limited to issues involving his Miranda waiver. Although the question of how petitioner’s mental state related to his waiver was closely intertwined with his mental state as a possible guilt phase defense, our order to show cause, and resultant reference order, did not include this question. The referee correctly limited himself to taking evidence relevant to, and issuing a report answering, the specific questions we asked, not other questions. In any event, the referee found Dr. Pettis’s testimony not credible for reasons that would have applied to any testimony regarding the Miranda waivers. We see no basis on which even to speculate that testimony from Dr. Pettis that petitioner was unable to, or did not, validly waive his Miranda rights, if he would have so testified, would have been any more credible than the extensive testimony he did give regarding petitioner’s mental state at the time of the crime.

4. Petitioner urges us to reject the referee’s credibility determinations. As a general matter, he complains that he “had no notice and no opportunity to be heard on the theory of ‘recent fabrication’ introduced in the referee’s findings and report.” On the contrary, he had full notice that all aspects of the witnesses’ credibility were at issue in this matter. He could have presented any admissible evidence supporting (or negating) any witness’s credibility. No requirement exists that a fact finder somehow tell a party in advance every basis on which it might disbelieve a witness. Moreover, at the beginning of oral argument after the hearing, the referee did give notice of his tentative findings, including that he found much of the testimony recently fabricated. Petitioner had full opportunity to argue the point. He cannot complain of lack of notice.

Petitioner claims that in making his credibility determinations, “the referee erroneously applied a purely personal, subjective standard, rather than addressing the relevant question, i.e., whether jurors were reasonably likely to have accorded weight to, and been moved by, the evidence.” He argues, in essence, that a jury might have found credible the witnesses the referee found incredible. We see no error. The referee’s conclusion that the lay witnesses’ testimony was incredible and recently fabricated is certainly relevant to whether counsel could have discovered that evidence before trial. But assuming counsel could have, and should have, presented the evidence, whether petitioner has shown prejudice is a legal question for this court to determine, not a factual question for the referee. (In re Ross, supra, 10 Cal.4th at p. 205.) We asked the referee to determine how credible the evidence was, and he has done so. Petitioner is correct that the question of prejudice hinges on whether it is reasonably probable the jury or, if, as here, a jury was waived, the trial court would have reached a different verdict, not whether the reviewing court (or the referee) would have reached a different verdict. (See, e.g., Smith v. Stewart (9th Cir. 1998) 140 F.3d 1263, 1270.) But the referee’s credibility findings help us make that determination. Although the weight to be given credible mitigating evidence is for the jury (or trial court) to decide, it is highly unlikely that a reasonable juror would credit evidence that the referee finds incredible. But such a possibility is for this court, which makes the ultimate legal determination, to consider, not the referee answering factual questions that we have posed.

Petitioner also contends the “referee improperly disregarded a large quantity of documentary evidence, which independently proved facts in support of petitioner’s claims, provided necessary foundation for the expert opinions, and corroborated the testimony of lay witnesses.” We disagree. The referee admitted into evidence all of the documentary exhibits, although he did not consider hearsay declarations of persons who did not testily for the truth of the matters asserted. Ultimately, he based his credibility determinations on the witnesses themselves and their testimony, which was proper. “A reference hearing following issuance of an order to show cause is subject to the rules of evidence as codified in the Evidence Code. (See Evid. Code, § 300.) Under those rules, an out-of-court declaration is hearsay, and unless subject to some exception permitting it to be admitted, should be excluded upon timely and proper objection. (See Evid. Code, § 1200.) . . . An expert witness, however, may base an opinion on reliable hearsay, including out-of-court declarations of other persons. [Citations.]” (In re Fields, supra, 51 Cal.3d at p. 1070, in. omitted.) The referee allowed the expert witnesses to rely on any document or other source they believed proper in forming their opinions.

Petitioner also argues the “referee improperly disregarded the ability of experts to evaluate the consistency and reliability of evidence underlying their professional opinions.” We see no error. The fact finder determines the facts, not the experts. Indeed, the fact finder may reject even “a unanimity of expert opinion. ‘To hold otherwise would be in effect to substitute a trial by “experts” for a trial by jury . . . .’ (People v. Wolff (1964) 61 Cal.2d 795, 811 [40 Cal.Rptr. 271, 394 P.2d 959].) ‘The chief value of an expert’s testimony . . . rests upon the material from which his opinion is fashioned and the reasoning by which he progresses from his material to his conclusion.’ [Citation.]” (People v. Samuel (1981) 29 Cal.3d 489, 498 [174 Cal.Rptr. 684, 629 P.2d 485].) Although experts may testify about their opinions, the fact finder decides what weight to give those opinions. This is especially important when the witnesses are not neutral court-appointed experts but experts hired by a party specifically seeking evidence supporting that party’s position. (See People v. Kelly (1992) 1 Cal.4th 495, 543-544 [3 Cal.Rptr.2d 677, 822 P.2d 385].) Here, because the referee heard testimony from many witnesses, and thus did not have to rely on their hearsay statements, he was particularly well situated to evaluate the reliability of the materials on which the experts relied. He was entitled to do so independently. Moreover, his findings were generally consistent with, and supported by, the testimony of one expert—Dr. Sharma—the only one who was originally court appointed and who had examined petitioner shortly after the crimes.

Petitioner challenges as illogical the referee’s rejection of much of the evidence provided by lay witnesses. We disagree. The referee did credit the testimony of some of the lay witnesses, including two—Leander Thornton and Burrell Ford—who contradicted some of the disbelieved testimony. Petitioner claims the referee failed to “understand and give due weight to the natural hesitancy of capital defendants and their families to disclose embarrassing details of their personal histories and the efforts customarily made by capital defense counsel to gain the trust needed to permit disclosure of such information.” Whether any such “natural hesitancy” exists is for the fact finder to consider, not for an appellate court to judicially notice. The significance of the witnesses’ failure to speak before trial and their willingness to speak after the death verdict was for the referee to consider in conjunction with all other relevant factors, including the witnesses’ demeanor and contradictory testimony of other witnesses.

Petitioner discusses the testimony of each witness in detail and asks us to overturn each of the referee’s credibility determinations that are adverse to him. “Deference to the referee is called for on factual questions, especially those requiring resolution of testimonial conflicts and assessment of witnesses’ credibility, as the referee has the opportunity to observe the witnesses’ demeanor and manner of testifying.” (In re Johnson, supra, 18 Cal.4th at p. 461.) Indeed, the reason we require habeas corpus petitioners to prove their disputed allegations at an evidentiary hearing, rather than merely decide the merits of the case on declarations, is to obtain credibility determinations. We have compared the referee’s report to the transcript of the evidentiary hearing. The factual discussion and record citations are accurate and reliable. The report, including its credibility discussion, is thorough and convincing, and fully responsive to our questions. We see no reason to overturn the referee’s credibility determinations. Accordingly, we adopt his factual findings. The significance of those findings to the legal issues is for us to determine, which we now proceed to do.

D. The Claims on Which We Issued an Order to Show Cause

1. Whether Trial Counsel Provided Ineffective Assistance by Failing to Investigate and Present a Mental Defense at Trial

On direct appeal, we rejected various guilt phase claims of ineffective assistance of counsel. We noted that “[wjhile judging counsel’s performance, we must keep in mind the circumstances he faced, and particularly the fact that his client had already confessed fully to a hideous crime. [Citation.] These circumstances did not bode well for a successful defense and may have justified tactical decisions that might seem unreasonable under other circumstances.” (Scott, supra, 15 Cal.4th at p. 1212.) We also noted that it is particularly difficult to establish ineffective assistance on appeal. (Ibid.) Accordingly, petitioner has raised other claims in this habeas corpus petition. We have now had a lengthy evidentiary hearing. Having reviewed the transcript of that hearing and the referee’s report, we see no reason the result should be different.

The referee found that counsel had requested a report on petitioner’s competency to stand trial, which resulted in examinations and reports by Drs. Sharma and Maloney that found petitioner competent and that did not support a mental defense. The doctors believed that petitioner had feigned certain symptomology. Counsel also knew that during his confessions, petitioner had occasionally assumed the persona of an alter ego, “Tony,” but petitioner had told counsel that Tony was a “con.” Counsel interviewed petitioner, observed nothing suggesting a mental defense, and investigated the use of Mellaril as a possible defense. Counsel did nothing further to investigate a mental defense.

Under these circumstances, it is a close question whether competent counsel was required to investigate further. “Criminal trial counsel have no blanket obligation to investigate possible ‘mental’ defenses, even in a capital case.” (People v. Gonzalez, supra, 51 Cal.3d at p. 1244.) There was evidence petitioner used drugs the night of the crime—evidence that was presented at trial (Scott, supra, 15 Cal.4th at p. 1216)—and he invoked the persona of “Tony” during his confessions. Both of these circumstances suggested a possible mental defense. But other evidence available to counsel, including the reports of Drs. Sharma and Maloney, contraindicated a mental defense. Drs. Sharma and Maloney investigated and reported on petitioner’s competency to stand trial, not possible mental defenses, but the two questions were closely intertwined, and the reports were relevant to both questions. They issued their reports within two months of the crime. Moreover, counsel had information, including statements from petitioner himself, that petitioner had feigned a mental defense when he assumed the persona “Tony.” Evidence that petitioner had feigned one mental defense would undercut an attempt to present a different mental defense. Moreover, petitioner described the crimes to counsel—as well as the investigators—in detail, which also undercuts a mental defense.

But, as the referee noted, counsel’s investigation was “minimal,” and some indication did exist of a potential mental defense. We need not decide whether competent counsel should have investigated further, for we find no prejudice in the failure to do so. (In re Ross, supra, 10 Cal.4th at p. 204.) The only possible mental defenses would have been (1) insanity, i.e., that at the time of the crime, petitioner either could not know or understand the nature and quality of the act, or could not distinguish right from wrong (Pen. Code, § 25, subd. (b); People v. Skinner (1985) 39 Cal.3d 765 [217 Cal.Rptr. 685, 704 P.2d 752]); or (2) that petitioner did not intend to kill Wanda Jensen (Scott, supra, 15 Cal.4th at p. 1216). As discussed in the referee’s report, petitioner has produced no credible evidence showing either insanity or absence of intent to kill. Intent to kill is not a difficult mental state to achieve. We find the referee’s reasons for rejecting Dr. Pettis’s opinion that petitioner was in a “disorganized and dissociative state” the night of the crime convincing, and believe a jury or trial court would also reject that opinion for those reasons. That opinion was the strongest evidence of a mental defense petitioner presented at the hearing. Moreover, at the trial, the trial court heard the evidence of drug usage, and counsel argued the prosecution had not established intent to kill. (Scott, supra, 15 Cal.4th at p. 1216.) Nevertheless, the court found petitioner intended to kill. We see no reasonable probability the outcome would have been different had counsel further investigated a possible mental defense.

2. Whether Trial Counsel Provided Ineffective Assistance by Failing to Investigate and Present Mitigating Evidence at the Penalty Phase

On appeal, in rejecting different penalty phase ineffective assistance of counsel claims, we described the case that counsel actually made against a death verdict. “He argued that the happenstance of medical malpractice that made this a murder case rather than attempted murder mitigated against the death penalty; he portrayed the crimes against Paula H. as less serious than her trial testimony suggested; he portrayed [petitioner] as an ‘angry young man’ who committed rash acts rather than a merely evil person who committed calculated crimes; he argued that the crimes against women were not the type to be repeated in prison; he argued that [petitioner] admitted his wrongdoing at an early stage; he cited [petitioner’s] ‘split personality’ as evidenced by the references to ‘Tony’; and he urged that [petitioner’s] conduct was not the type for which the death penalty was designed.” (Scott, supra, 15 Cal.4th at p. 1225.) We said that “[o]n this record, we cannot say that some other approach was so clearly superior that counsel’s representation was incompetent.” (Ibid) Now that there has been an evidentiary hearing, we see no basis on which to conclude otherwise.

The referee found that counsel’s investigation of potential mitigating evidence was “minimal.” Counsel utilized the medical malpractice defense at the penalty phase as well as at the guilt phase, attempted to speak to petitioner about his family and childhood, attempted to speak with his mother for purposes of the penalty phase, and argued mental impairment. The referee also found that petitioner hindered the investigation in two respects: he was “adamant” about not wanting to involve his family, and he was not talkative to counsel and did not provide any useful information for the penalty phase.

Under the circumstances, it is also a close question whether competent counsel was required to further investigate potential mitigating evidence. “As we have repeatedly explained, an attorney representing a defendant at the penalty phase of a capital case is not required to present potentially mitigating evidence over the defendant’s objections.” (People v. Kirkpatrick (1994) 7 Cal.4th 988, 1013 [30 Cal.Rptr.2d 818, 874 P.2d 248]; also quoted in In re Andrews (2002) 28 Cal.4th 1234, 1254 [124 Cal.Rptr.2d 473, 52 P.3d 656].) Petitioner did not object to all mitigating evidence, but he did not want his family involved, which precluded mitigating evidence that only the family can provide. “Moreover, ‘ [t]he reasonableness of counsel’s actions may be determined or substantially influenced by the defendant’s own statements or actions. Counsel’s actions are usually based, quite properly, on informed strategic choices made by the defendant and on information supplied by the defendant. In particular, what investigation decisions are reasonable depends critically on such information.’ (Strickland, supra, 466 U.S. at p. 691 [104 S.Ct. at pp. 2066-2067]; Burger v. Kemp [(1987) 483 U.S. 776, 795 [107 S.Ct. 3114, 3126, 97 L.Ed.2d 638]].)” (In re Andrews, supra, at p. 1255.) In this case, petitioner “did not inform [counsel] of the conditions he endured thereby alerting [him] to the need for further investigation of possible mitigation.” (Ibid.) Our reasonableness inquiry must also assess “the decision to forgo further investigation in light of the defense strategy counsel ultimately adopted.” (Ibid.) Here, counsel knew that two experts believed that petitioner was feigning mental symptoms, and he knew from petitioner’s own mouth that petitioner had feigned the “Tony” defense. Nevertheless, he made the “split personality” argument based on the guilt phase evidence. By simply arguing the point rather than attempting to present additional evidence of mental impairment, counsel was able to make the “split personality” argument without risking expert testimony that “Tony” was feigned. An attempt to present a different mental impairment defense could have been undercut by evidence that petitioner had already feigned one mental condition, which in turn would have undercut counsel’s penalty phase argument that petitioner had admitted his wrongdoing at an early stage by confessing.

On the other hand, counsel could well have conducted a more thorough investigation than he did. It appears he made little effort to elicit any mitigating evidence from petitioner or anyone else. As with the failure to investigate a possible mental defense, we find no prejudice even assuming counsel should have investigated further. Counsel was able to argue petitioner’s supposed “split personality,” and he had argued petitioner’s cocaine usage at the guilt phase. Presenting additional evidence of mental impairment that, as the referee found, lacked credibility and that differed substantially from the “split personality” defense that petitioner presented in his confession would not have aided the defense cause. In addition, the referee found that additional investigation would not have discovered the bulk of the proffered mitigating evidence of petitioner’s alleged harsh childhood because most of the evidence was recently fabricated. Even if we assume it could have been discovered earlier, and counsel could have presented this evidence, the circumstance remains that the referee found most of it incredible. This credibility determination is especially significant here because petitioner waived a jury and the case was tried to a court. We see no reason to suspect the court would have found these witnesses any more credible than did the referee. Moreover, this evidence could have been weakened still further by the testimony of one witness the referee found credible—Burrell Ford—who, the referee found, “could have also been called in rebuttal relative to the lack of physical or sexual abuse by family and friends when he had been present.”

In stun, the evidence showed, and the court or a jury may well have believed, that petitioner had a difficult childhood and a troubled family history, although not as harsh as he claimed at the evidentiary hearing. But the circumstances of this crime were particularly egregious, and petitioner had a history of brutal sexual violence. We see no reasonable probability the outcome would have been different had counsel introduced the new evidence that petitioner has now presented instead of taking the approach he actually took at trial.

3. Whether Trial Counsel’s Advice to Waive a Jury Was Uninformed and Hence Ineffective

On appeal, we rejected petitioner’s claim that counsel was ineffective in advising him to waive a jury trial. We did “ ‘not assume inadequacy of representation unless counsel had no conceivable legitimate tactical purpose’ for waiving a jury.” (Scott, supra, 15 Cal.4th at p. 1213, quoting People v. Diaz (1992) 3 Cal.4th 495, 558 [11 Cal.Rptr.2d 353, 834 P.2d 1171].) We noted a number of possible valid tactical reasons for waiving a jury in this case. (Scott, supra, at pp. 1213-1214.) Petitioner claims, however, that counsel’s alleged failings in not investigating a possible mental defense and potential mitigating evidence rendered the advice to waive a jury uninformed and hence ineffective.

After reviewing the evidentiary hearing and the referee’s report, it is apparent that counsel did indeed have valid tactical réasons for advising petitioner to waive a jury: the desire to have a particular highly regarded judge decide the case, the nature of the particular jury pool from which a jury would have been selected, the nature of the proffered medical malpractice defense, and the horrific facts of the crime. The referee also found that the “decision to waive jury was in no way compromised by any failure of trial counsel’s investigation.” Accordingly, the referee found that “no additional investigation would have affected the advice to waive a jury.” We agree. The evidentiary hearing does not negate but bolsters the reasons we rejected this claim on appeal. The record shows “that further investigation would not reasonably have influenced that decision.” (In re Andrews, supra, 28 Cal.4th at p. 1266.)

The decision to waive a jury was unusual, but so too was one of the tactical decisions upheld against a similar ineffective assistance challenge in Bell v. Cone, supra, 535 U.S. at pages 690-691 [122 S.Ct. at page 1848]—to waive argument at the penalty phase. As with deciding whether to waive argument, whether to waive a jury under the circumstances was “a tactical decision about which competent lawyers might disagree.” (Id. at p. 702 [122 S.Ct. at p. 1854].) Accordingly, making that decision was not ineffective. (Ibid.)

III. Other Claims

Because our order to show cause and subsequent reference hearing were limited to specific claims of ineffective assistance of counsel, we do not discuss the other claims stated in the petition. As is our usual practice, we will resolve the petition for writ of habeas corpus by a separate order. (In re Andrews, supra, 28 Cal.4th at p. 1266.)

IV. Conclusion

The order to show cause is discharged.

George, C. J., Baxter, 1, Werdegar, J., Brown, J., and Moreno, J., concurred.

1 The referee provided record citations for his factual discussion, which we have deleted. Except as otherwise indicated, footnotes are also deleted. The substance of some of the footnotes is stated in brackets.

2 In a footnote, the referee noted that “[t]he reports of Dr. Sharma and Dr. Michael Maloney are very relevant to explain the nature of trial counsels’ actions and decision making and what may or may not have been discovered upon further investigation. However, neither Dr. Sharma nor Dr. Maloney could have been called in rebuttal at trial by the prosecution by reason of People v. Arcega (1982) 32 Cal.3d 504, 522-523 [186 Cal.Rptr. 94, 651 P.2d 338], as they had been appointed relative to the issue of competency to stand trial.”

3 In a footnote, the referee noted, “Although the investigative report of Mr. Hill mentioned that petitioner gave the investigator an address where he believed his sister, Carrie ‘Scott’ lived, it was in the context of his alleged whereabouts on the night of the Jensen crimes and not a request as to where to find his sister. Such is therefore not in conflict with trial counsel’s assertions that petitioner did not want his family involved.”

4 Penal Code section 1054.3 provides: “The defendant and his or her attorney shall disclose to the prosecuting attorney:

“(a) The names and addresses of persons, other than the defendant, he or she intends to call as witnesses at trial, together with any relevant written or recorded statements of those persons, or reports of the statements of those persons, including any reports or statements of experts made in connection with the case, and including the results of physical or mental examinations, scientific tests, experiments, or comparisons which the defendant intends to offer in evidence at the trial.

“(b) Any real evidence which the defendant intends to offer in evidence at the trial.” (See generally Izazaga v. Superior Court (1991) 54 Cal.3d 356 [285 Cal.Rptr. 231, 815 P.2d 304].)

5 We caution that our minute orders apply only to the specific case and do not establish binding precedent for other cases. (Thompson v. Department of Corrections (2001) 25 Cal.4th 117, 125 [105 Cal.Rptr.2d 46, 18 P.3d 1198]; California Assn. of Psychology Providers v. Rank, supra, 51 Cal.3d at p. 9.)

6 We need not, and do not, decide whether a habeas corpus proceeding is civil or criminal for other purposes. (See In re Head (1986) 42 Cal.3d 223, 226, fn. 4 [228 Cal.Rptr. 184, 721 P.2d 65].) It is a special proceeding and not entirely analogous to either category. (People v. Duvall, supra, 9 Cal.4th at pp. 477-478, fn. 4.)

7 Because petitioner ultimately did not testify at the reference hearing, we do not consider any other issues that might arise regarding a testifying petitioner’s possible invocation of the privilege against self-incrimination.