I dissent.
When this case was before us after defendant’s original trial, I would have affirmed the judgment because there was no “miscarriage of justice” within the meaning of article VI, section 13 of the California Constitution. (People v. Alcala (1984) 36 Cal.3d 604, 637 [205 Cal.Rptr. 775, 685 P.2d 1126] (dis. opn. of Mosk, J.).)
Since the matter has returned following retrial, I am compelled to conclude to the contrary. I would reverse the judgment because now there has indeed been a miscarriage of justice. Through several erroneous evidentiary rulings, the trial court improperly disturbed the scales on which guilt or innocence was to be weighed, allowing the People to introduce inadmissible evidence for inculpation and barring defendant from introducing admissible evidence for exculpation. As a result, it denied defendant the fundamental fairness guaranteed, by the due process clause of the Fourteenth Amendment to the United States Constitution.
In the analysis that follows, I shall focus on the four most significant evidentiary rulings.
I
First—and most egregious—the trial judge, on the People’s motion outside the presence of the jury, erroneously allowed the prosecution to introduce into evidence testimony that Dana Crappa had given at defendant’s original trial. That testimony was crucial, inasmuch as it alone placed defendant at the scene of the murder and did so by eyewitness observation. But it also contained “material inconsistencies.” (People v. Alcala, supra, 36 Cal.3d at p. 636.) Moreover, it presented itself as only one of several *813accounts Crappa had related, none of which was fully reconcilable with any other. Further, it was given in a highly unusual manner, with “substantial delays between the question and an answer, sometimes running perhaps close to a minute, perhaps even longer.” Finally, it originated in questionable interrogation by the police. Indeed, at the first trial “[d]efense and prosecution experts gave conflicting views on whether . . . Crappa had been ‘brainwashed’ or otherwise led, while in a suggestible state, to false memories.” (People v. Alcala, supra, at pp. 620-621.)
Crappa’s prior testimony was obviously hearsay. It constituted “evidence of . . . statement[s] that [were] made other than by a witness while testifying at the hearing and that [were] offered to prove the truth of the matter stated.” (Evid. Code, § 1200, subd. (a).) As hearsay, it was inadmissible. (Id., § 1200, subd. (b).)
The trial judge, however, did not so rule. He determined that Crappa’s prior testimony was not inadmissible hearsay because Crappa was assertedly “unavailable as a witness” and defendant was a “party” at his original trial and “had the right and opportunity to cross-examine [her] with an interest and motive similar to that which he ha[d]” on retrial. (Evid. Code, § 1291, subd. (a).) The judge found that Crappa was “unavailable” on the ground that she was allegedly “unable ... to testify . . . because of then existing . . . mental . . . infirmity.” (Id., § 240, subd. (a)(3).)
The trial judge’s ruling cannot survive scrutiny. The critical finding of unavailability is invalid. It matters not what standard of review is applied. The reason is plain. The finding is fictive, and not real. The majority miss this dispositive fact. As a result, their analysis fails.
In making the People’s motion immediately before Crappa was scheduled to take the stand, the prosecutor represented that “[s]he has . . . told me, in response to my direct question essentially, I have asked her, is it a situation where you can’t remember, or you don’t want to remember. And essentially it seems to be a situation where she just doesn’t want to remember so she is going to say she doesn’t remember. But it’s not a situation where she can’t remember.”
The trial judge chose to let the prosecutor’s representation go past as though it had never been made. Presumably, he did so because he was disinclined to take whatever “reasonable steps” would have been required “to induce [Crappa] to testify” if he found that she refused to do so. (People v. Sul (1981) 122 Cal.App.3d 355, 364-365 [175 Cal.Rptr. 893].)
I do not overlook the evidence introduced at the hearing. The People, of course, bore the burden to prove Crappa’s unavailability (People v. Stritzinger (1983) 34 Cal.3d 505, 516 [194 Cal.Rptr. 431, 668 P.2d 738]) by a *814preponderance of the evidence (Evid. Code, § 115, 2d par.). The evidence, however, was insufficient.
I shall assume for argument’s sake only that this case is not subject to the general rule, which requires “expert testimony on the witness’s present condition . . . .” (People v. Stritzinger, supra, 34 Cal.3d at p. 517.) Nevertheless, I do note that it certainly does not come within the only recognized exception, which covers refusal to testify out of fear. (See id. at pp. 518-519, discussing People v. Rojas (1975) 15 Cal.3d 540, 547-552 [125 Cal.Rptr. 357, 542 P.2d 229, 92 A.L.R.3d 1127].)
Be that as it may, there was not sufficient evidence that Crappa was “unable ... to testify . . . because of then existing . . . mental . . . infirmity.” (Evid. Code, § 240, subd. (a)(3), italics added.) At best, Crappa and Dr. Anthony Staiti, a psychiatrist she had briefly consulted about a year earlier, made statements that might support an inference that she may have suffered from some “mental infirmity” of some sort in the past. That is all. To be sure, the trial judge stated that Crappa “seemed to wince and blink” under questioning. But he could reasonably have deduced little from such an observation. He may have been an experienced jurist. He was not a diagnostician.
The trial judge was apparently not troubled by the evidentiary void. He declared in the midst of the hearing that “I am never hampered by the law . . . .” Events proved him right. He had determined the course he would follow. He also declared: “Basically, what we are going to do is put [Crappa] on [before the jury], let her have her say-so, give it a shot, and get into the reading [of her prior testimony].” Events proved him right here too.
Indeed, the trial judge apparently sought to maintain the evidentiary void. That is the only reasonable explanation for his decision to deny a motion defendant made for an independent psychiatric examination of Crappa. I recognize that, in refusing the request, he declared that “I don’t think [such an examination] would add or detract.” Of course he did not. As noted, he had already determined the course he would follow. Earlier, he had stated that he was not “hampered” by the law. Here, he implied that he would not be “hampered” by the facts. The majority find the denial of the motion proper. I disagree. An independent psychiatric examination is indeed appropriate when the asserted “mental infirmity” is only “apparent”—as the majority themselves essentially concede to be the case here. Further, defendant sought the examination in question to determine whether Crappa was in fact suffering from a “mental infirmity,” and not to simply to impeach her credibility.
*815II
Second—and only slightly less egregious—the trial judge, denying a motion defendant made outside the presence of the jury, erroneously prohibited Dr. Ray William London, a psychologist, from giving any testimony whatsoever on behalf of the defense. “That gentleman,” declared the trial judge, “ain’t going to testify in this case . . . .” Dr. London’s testimony would have included expert opinion to the effect that Grappa had been led to false memories by suggestive police interrogation incorporating sophisticated psychological techniques. At defendant’s original trial, testimony of this sort was given. On retrial, Dr. London was a crucial defense witness because Grappa was a crucial prosecution witness.
Dr. London’s proposed testimony was plainly admissible.
Manifestly, the testimony was relevant. It had a strong “tendency in reason to . . . disprove [a] disputed fact that [was] of consequence to the determination of the action,” specifically, the credibility of Grappa. (Evid. Code, § 210.)
Moreover, the testimony was a proper subject of expert opinion. First, it was “[r]elated to a subject that is sufficiently beyond common experience” as to “assist the trier of fact.” (Evid. Code, § 801, subd. (a).) What the record reveals is not “stage hypnotism,” with theatrical display by a performer and humorous antics by a member of his audience. Rather, there was suggestive police interrogation incorporating sophisticated psychological techniques— techniques designed not to call attention to themselves. Jurors might be familiar with the former. They would have little acquaintance, however, with the latter. Second, the testimony was “[b]ased on matter. . . perceived by or personally known to the witness or made known to him at or before the hearing, . . . that is of a type that reasonably may be relied upon by an expert in forming an opinion upon the subject to which his testimony relates . . . .” (Id., § 801, subd. (b).) Dr. London reviewed materials including transcripts and audio-tape recordings of Grappa’s interrogation by the police and transcripts of her prior testimony.
Finally, the testimony’s “probative value,” which was high, was not “substantially outweighed by the probability that its admission” would “necessitate undue consumption of time" or “create substantial danger of undue prejudice, of confusing the issues, or of misleading the jury.” (Evid. Code, § 352.)
Here too, the trial judge ruled otherwise. He determined that Dr. London’s proposed testimony was not a proper subject of expert opinion. He found Dr. *816London’s opinion too “speculative,” essentially because the psychologist had not personally attended the pertinent police interrogation sessions involving Crappa. He also found Dr. London’s opinion substantially more prejudicial than probative because it “could give the jury no guidance but confusion,” “would confound, screw up, do nothing for this case as far as either side getting a fair trial,” and would entail “undue consumption of time.”
Here too, the trial judge’s ruling fails. More than a year earlier, in People v. McDonald (1984) 37 Cal.3d 351, 377 [208 Cal.Rptr. 236, 690 P.2d 709, 46 A.L.R.4th 1011], we had expressly held: “When”—as here—“an eyewitness identification of the defendant is a key element of the prosecution’s case but is not substantially corroborated by evidence giving it independent reliability, and the defendant offers qualified expert testimony on specific psychological factors shown by the record that could have affected the accuracy of the identification but are not likely to be fully known to or understood by the jury, it will ordinarily be error to exclude that testimony.” This case is no exception. The “judicial policy disfavoring attempts to impeach witnesses by means of psychiatric testimony” (maj. opn., ante, at p. 781), invoked by the majority to support the ruling, proves to be inapplicable. It generally opposes challenges to the veracity of a person who testifies. But it does not bar attacks on the accuracy of his testimony.
Contrary to the majority’s view, the trial judge’s determination that Dr. London’s proposed testimony was substantially more prejudicial than probative is unsound. At the hearing on the motion, he was able to prevent any confusion or delay that might have been occasioned by the psychologist’s opinion. He would surely have been able to do the same before the jury. Without doubt, similar testimony given at defendant’s original trial was not “unduly prejudicial” in any way. Dr. London’s would have had no untoward effect.
Again contrary to the majority’s view, the trial judge’s determination that Dr. London’s proposed testimony was not a proper subject of expert opinion is unsound as well. Indeed, the judge’s finding that the psychologist’s opinion was too “speculative” essentially because the psychologist had not personally attended the pertinent police interrogation sessions involving Crappa is practically ludicrous. As stated, Dr. London had reviewed materials including transcripts and audio-tape recordings of Crappa’s interrogation by the police and transcripts of her prior testimony. The judge’s implication is unsupported: there is simply no requirement that an expert must have personal perception or knowledge of all the matters on which his opinion is based. In any event, any such “requirement” would have to be waived in this case. When Crappa was interrogated by the police, there was no defense *817expert; if there had been, he would obviously not have been welcomed. Subsequently, as the prosecutor acknowledged, Crappa refused to talk to the defense: “she always said she wouldn’t talk to them . . . .”
Ill
Third, the trial judge, again denying a motion defendant made outside the presence of the jury, erroneously prohibited Tim Fallen from giving any testimony whatsoever on behalf of the defense. Fallen would have briefly and simply testified to the effect that he observed Robin Samsoe the day after Crappa assertedly saw the girl and defendant at the scene of the murder near a remote mountain ravine. At defendant’s original trial, Fallen had given substantially similar testimony. There, when he was shown a photograph of Samsoe on direct examination by defense counsel, he positively identified the subject as the girl he had seen. But when he was shown a photograph of another girl on cross-examination by the prosecutor, he made the same positive identification. On retrial, Fallen was a crucial defense witness because Crappa was a crucial prosecution witness.
Fallen’s proposed testimony was plainly admissible. It was relevant, having as it did a strong tendency to disprove a disputed, material fact, viz., the credibility of Crappa. Further, it was not “unduly prejudicial”: if credited, it would point to defendant’s innocence; if not credited, it would have no effect.
Again, the trial judge ruled otherwise. He determined in substance that Fallen’s proposed testimony was irrelevant because it was “not going to add any probative value.” He also determined that, even if it were relevant, it was “unduly prejudicial” because of the “danger of confusing the jury” and the risk of “undue consumption of time.”
Again, the trial judge’s ruling fails. The majority themselves concede the error. I shall not belabor the point. Suffice it to say that the brief and simple testimony that Fallen would have given on what was manifestly a critical issue was altogether relevant and not at all “unduly prejudicial.” That he would have been subject to impeachment because of his positive identification of the two photographs is of no consequence for present purposes. In passing, however, I note that the two girls, as pictured, were not dissimilar in appearance.
IV
Fourth, the trial judge, yet again denying a motion defendant made outside the presence of the jury, erroneously prohibited Gerald Crawford and Raul *818Vasquez from giving any testimony whatsoever on behalf of the defense. Crawford and Vasquez would have testified to the following effect: Crawford was a police officer and Vasquez was a convicted murderer; about 11 p.m., two days after Crappa assertedly saw Samsoe and defendant at the scene of the murder in a remote mountain ravine, Crawford found Vasquez “acting suspiciously” in that very area. At defendant’s original trial, Crawford and Vasquez had given substantially similar testimony. On retrial, Crawford and Vasquez were crucial defense witnesses because Crappa was a crucial prosecution witness. They were also crucial defense witnesses in their own right because they offered exculpatory evidence.
The proposed testimony by Crawford and Vasquez was plainly admissible. It was relevant: it had a strong tendency both to prove one disputed, material fact—the innocence of defendant—and to disprove another disputed, material fact—the credibility of Crappa. Further, it was not “unduly prejudicial”: if credited, it would point clearly to defendant’s innocence; if not credited, it would have no effect.
Yet again, the trial judge ruled otherwise. He determined in substance that the proposed testimony by Crawford and Vasquez was irrelevant because the “probative value of this type of evidence is zero . . . .” He also determined that, even if it were relevant, it was unduly prejudicial because it “would do nothing but confuse the jury . . . .”
Yet again, the trial judge’s ruling fails.
In People v. Hall (1986) 41 Cal.3d 826 [226 Cal.Rptr. 112, 718 P.2d 99], we held that evidence of third party culpability—like the proposed testimony by Crawford and Vasquez—is admissible if relevant. (Id. at p. 834.) Relevance requires “direct or circumstantial evidence linking the third person to the actual perpetration of the crime”; “evidence of mere motive or opportunity . . . , without more, will not suffice . . . .” (Id. at p. 833.) The proposed testimony linked Vasquez to the crime itself and went beyond mere motive or opportunity. Vasquez was found near the scene of the murder in a remote mountain ravine; he was there about 11 p.m. on a day not long after the crime was apparently committed; and, significantly, he was “acting suspiciously.” The trial judge missed the point entirely. How else to explain his statement that the “probative value of this type of evidence is zero”? For their part, the majority fail to discern the critical importance of Vasquez’s suspicious activity.
In Hall, we also held that evidence of third party culpability is not excludable unless “unduly prejudicial.” (41 Cal.3d at p. 834.) Contrary to the *819trial judge’s conclusion, there was simply nothing confusing about the proposed testimony.
V
In sum, through the four erroneous evidentiary rulings discussed above, among others, the trial court improperly disturbed the scales of justice and thereby denied defendant the fundamental fairness guaranteed by the due process clause of the Fourteenth Amendment. It is incomprehensible that the majority can conclude “defendant received a fair and untainted trial” (maj. opn., ante, at p. 810) and on that basis send him on his way to death in the gas chamber.
Accordingly, I would reverse the judgment.
Appellant’s petition for a rehearing was denied March 10, 1993, and the opinion was modified to read as printed above. Mosk, J., was of the opinion that the petition should be granted.