I concur.
In light of the facts related in the majority opinion, there is a substantial possibility that the victim’s testimony on penetration was “confabulated” in the police-conducted hypnosis session, and, given the unreliability of that evidence, I do not think that the rape convictions can stand. To the extent that the Shirley decision embodies the principle that testimony “created” in this type of hypnosis session is not sufficiently reliable to be admitted in a criminal trial, I agree that it establishes a rule that goes to the heart of the accuracy of the factfinding process and should be applied retroactively.
But while I agree that the victim’s hypnotically created testimony on penetration was impermissibly admitted and should be excluded on retrial, I would make it clearer than the majority does that Shirley should not be applied retroactively to bar the victim from testifying on retrial about facts or memories that the trial court can reasonably determine were not created *430during the hypnosis session. The record before us suggests that, aside from the testimony on penetration, the victim’s posthypnosis testimony at trial was, at least in large part, consistent with the account she gave of the attack before undergoing hypnosis. Even if the testimony on penetration is disregarded, under her prehypnosis version of the night’s events the jury apparently could still have found defendants guilty of attempted oral copulation as well as attempted rape. There is no justification for effectively immunizing defendants from prosecution for those crimes—which would be the result if the victim is barred from testifying about any subject touched on in the hypnosis session—simply because the hypnotically created testimony on penetration is too unreliable to be admitted.
I should make it clear that I am not suggesting that the basic holding in Shirley be overruled. Although I still believe that that decision’s sweeping prophylactic rule—barring all testimony by a witness concerning any subject touched on during a hypnosis session—is too broad (see 31 Cal.3d at pp. 73-78 [conc. opn.]), the police and prosecution have been on notice since Shirley was filed of the consequences that flow under that decision if a potential witness is hypnotized. As a result, the prospective application of Shirley’s strict exclusionary rule may not impose substantial, adverse effects on the administration of justice in this state.1
*431But the same cannot be said for the retroactive application of the decision’s prophylactic sanction. We must remember that this is only one of many cases that will be affected by our determination of the appropriate remedy to apply to those witnesses who were hypnotized before Shirley. With respect to such cases, there is a special need to ensure that in our zeal to protect the citizenry from the hazards of hypnosis, we do not create a greater injustice by an after-the-fact disqualification of crucial witnesses who have relevant—frequently vital—information that is not tainted by the hypnosis.
Although it may not be immediately apparent from a reading of the scholarly majority opinion, the recent out-of-state decisions which are most in tune with Shirley do not support Shirley's conclusion that a witness who has been hypnotized may not testify with respect to any subject discussed in the hypnosis session. Indeed, virtually all of the other states that have adopted a “per se" rule excluding hypnotically induced testimony, have at the same time expressly declared that a witness is not necessarily barred from testifying to events which the witness recalled and related to others before undergoing hypnosis. (See State ex rel. Collins v. Superior Court, etc. (1982) 132 Ariz. 180 [644 P.2d 1266, 1295-1296]; Com. v. Kater (1983) 388 Mass. 519 [447 N.E.2d 1190, 1197-1200]; State v. Patterson (1983) 213 Neb. 686 [331 N.W.2d 500, 503-504]; People v. Hughes (1983) 59 N.Y.2d 523 [466 N.Y.S.2d 255, 453 N.E.2d 484, 495-496]; State v. Collins (1983) 296 Md. 670 [464 A.2d 1028, 1044]; State v. Blanchard (Minn. 1982) 315 N.W.2d 427, 430-431; Strong v. State (1982) - Ind. - [435 N.E.2d 969, 970-971]; Robison v. State (Okla.Crim.App. 1984) 677 P.2d 1080, 1085; State v. Peoples (1984) 311 N.C. 515 [319 S.E.2d 177, 188]; Com. v. Taylor (1982) 294 Pa.Super. 171 [439 A.2d 805, 808]; People v. Quintanar (Colo.App. 1982) 659 P.2d 710, 713-714.)2
These decisions recognize that while there are theoretical objections to even such testimony, the probable reliability and potential importance of the evidence justifies its admission. As the New York Court of Appeals explained; “A criminal trial for rape or assault would present an odd spectacle if the victim was barred from saying anything, including the fact that the crime occurred, simply because he or she submitted to hypnosis sometime prior to trial to aid the investigation or obtain needed medical treat*432ment. Even in cases dealing with the frailties of eyewitness identification some allowance must be made for practicalities (see, e.g., Stovall v. Denno [(1967)] 388 U.S. 293, 301-302 . . . .) [f] When confronted with suggestive pretrial identification, it has not been found necessary to preclude the witness from making an in-court identification on the basis of recollections prior to the suggestive procedure, if it is found as a fact that he can do so without relying on the improperly made identification .... A similar procedure would seem to be appropriate in cases involving the pretrial use of hypnosis . . . .” (People v. Hughes, supra, 453 N.E.2d at pp. 495-496.)
Thus, when the scope of the recent out-of-state decisions is properly understood, it becomes clear that the numerous authorities relied on by the majority to accord retroactive application to Shirley do not support the proposition that a witness hypnotized before Shirley should be totally barred from testifying at trial. Rather, the cases suggest that such a witness should generally be permitted to testify at trial as to prehypnosis memories so long as there is satisfactory evidence from which the trial court can determine that the witness did in fact recall and relate the statements before undergoing hypnosis. (See, e.g., State v. Collins, supra, 464 A.2d 1028, 1044-1045; People v. Hughes, supra, 453 N.E.2d 484, 496; State v. Patterson, supra, 331 N.W.2d 500, 504; People v. Quintanar, supra, 659 P.2d 710, 713; Com. v. Kater, supra, 447 N.E.2d 1190, 1197-1200.)
Accordingly, following the lead of the out-of-state decisions, I believe we should advise the trial court that, on retrial, the fact that the victim has undergone hypnosis does not necessarily bar her from testifying to events which the court finds that she recalled and related before the hypnosis session. Although defendants should be permitted to impeach her testimony on those matters by bringing to the jury’s attention the fact and possible effect of her hypnosis, they should not gain effective immunity from prosecution by a rule that would bar the victim from the witness stand.
While I recognize that the majority has chosen not to resolve the issue of the admissibility of the witness’ testimony as to events recalled and related before hypnosis, I join in the judgment on the understanding that it does not foreclose the trial court from admitting such testimony.
Grodin, J., concurred.
This is particularly true inasmuch as the Legislature has recently enacted a statute, effective January 1, 1985, which modifies Shirley by authorizing a witness who has previously undergone hypnosis to testify to matters which the witness recalled and related prior to the hypnosis, so long as a number of specified safeguards are followed. The new provision, Evidence Code section 795, reads in full: “(a) The testimony of a witness is not inadmissible in a criminal proceeding by reason of the fact that the witness has previously undergone hypnosis for the purpose of recalling events which are the subject of the witness’ testimony, if all of the following conditions are met: [t] (1) The testimony is limited to those matters which the witness recalled and related prior to the hypnosis. [H] (2) The substance of the prehypnotic memory was preserved in written, audiotape, or videotape form prior to the hypnosis, [t] (3) The hypnosis was conducted in accordance with all of the following procedures: [Ü] (A) A written record was made prior to hypnosis documenting the subject’s description of the event, and information which was provided to the hypnotist concerning the subject matter of the hypnosis, [t] (B) The subject gave informed consent to the hypnosis. [f] (C) The hypnosis session, including the pre- and post-hypnosis interviews, was videotape recorded for subsequent review. [H] (D) The hypnosis was performed by a licensed medical doctor or psychologist experienced in the use of hypnosis and independent of and not in the presence of law enforcement, the prosecution, or the defense. [1] (4) Prior to admission of the testimony, the court holds a hearing pursuant to Section 402 of the Evidence Code at which the proponent of the evidence proves by clear and convincing evidence that the hypnosis did not so affect the witness as to render the witness’ prehypnosis recollection unreliable or to substantially impair the ability to cross-examine the witness concerning the witness’ prehypnosis recollection. At the hearing, each side shall have the right to present expert testimony and to cross-examine witnesses. [H] (b) Nothing in this section shall be construed to limit the ability of a party to attack the credibility of a witness who has undergone hypnosis, or to limit other legal grounds to admit or exclude the testimony of that witness.”
I say “virtually all” because the high court of one state, Michigan, simply reserved judgment on this question in People v. Gonzales (1983) 417 Mich. 968 [336 N.W.2d 751], modifying 329 N.W.2d 743. The intermediate appellate courts in Michigan which have passed on the issue—both before and after Gonzales—have concluded that a witness may testify to facts recalled and related prior to hypnosis. (See People v. Perry (1983) 126 Mich.App. 86 [337 N.W.2d 324, 325]; People v. Wallach (1981) 110 Mich.App. 37 [312 N.W.2d 387, 404-405].)