State v. Coe

109 Wn.2d 832 (1988) 750 P.2d 208 THE STATE OF WASHINGTON, Respondent, v. FREDERICK H. KEVIN COE, Appellant. No. 51600-6. The Supreme Court of Washington, En Banc. January 28, 1988. David Allen, Richard Hansen, Donald Roistacher, and Allen & Hansen, for appellant. Donald C. Brockett, Prosecuting Attorney, Patricia A. Thompson, Deputy, and Stephen R. Matthews, for respondent. DOLLIVER, J. Kevin Coe appeals his conviction of three counts of first degree rape. The primary issues on appeal are whether the admission of testimony by previously hypnotized witnesses was proper and whether the State proved the use or threatened use of a deadly weapon — one of the elements of first degree rape. We reverse the two counts in which witnesses were hypnotized, and affirm the third count which involved no hypnosis. We conclude the elements of first degree rape were met, and accordingly we affirm the conviction of one count of first degree rape. Frederick H. "Kevin" Coe was originally charged in 1981 with six counts of rape in Spokane. He was convicted of four counts of first degree rape. The convictions were *835 reversed by this court in State v. Coe, 101 Wn.2d 772, 684 P.2d 668 (1984). The primary basis for reversal was the admission of posthypnotic testimony of witnesses who had been hypnotized prior to Coe's arrest. The court held: Upon remand, the admissibility of the testimony of the previously hypnotized witnesses should be determined in accordance with our holding in State v. Martin [101 Wn.2d 713, 684 P.2d 651 (1984)]. Thus, testimony as to facts recalled during hypnosis would be inadmissible.... The testimony of the previously hypnotized witnesses would be admissible[, however,] if the State can show that the testimony consists solely of prehypnotic memory ... State v. Coe, at 786. A new trial began in King County, after a change of venue at the defendant's request, in January 1985. Before trial began, the court held a lengthy pretrial hearing on hypnotism. The trial court ultimately interpreted the phrase "prehypnotic memory" to include both (a) information given by the witnesses before being hypnotized, and (b) any evidence provided after hypnosis which the State could prove to be based on prehypnotic memories and untainted by the hypnosis. On the basis of this holding, the court ruled the victims could testify to their identifications of the defendant, made some time after they had been hypnotized. The jury found the defendant guilty of three of four rapes charged. The court dismissed count 2, involving Sherrill South, after the jury stated it was unable to reach a verdict. The counts at the second trial retained the numbering from the 6-count information in the first trial. Count 3 — Julie Harmia. In the evening on October 23, 1980, Julie Harmia was raped and beaten in Spokane by an attacker who told her he had a knife. Before he left, he also told her he might come after her later and use the knife. She gave a detailed description of her attacker to the police. Samples of semen were collected for testing. On March 10, 1981, Harmia identified the defendant in a lineup as the person who raped her. She was never hypnotized. *836 Count 5 — Mary Patricia Strange. On the morning of February 5, 1981, Mary Patricia Strange was raped on a playing field in Spokane where she had been jogging. After initial resistance by her, the attacker told her he had a knife. He inserted his fingers in her vagina, but no penile penetration occurred, so no semen sample was collected. She gave a detailed description of her attacker to police. A few days later, she was hypnotized by the police, but provided no different or additional evidence. Strange identified the defendant's picture out of a photo montage 2 or 3 weeks later. On March 10, 1981, she identified the defendant at a lineup. Charles Williams, the custodian at a school next to the field where Strange was attacked, noticed a car parked near the field on the same morning of the Strange attack. He observed the features of the car, and after the rape was reported in the paper the next day, he gave a description of the car to the police. The police later hypnotized Williams in an attempt to obtain the license plate number of the car, but without success. Count 6 — Diane Fitzpatrick. On the morning of February 9, 1981, Diane Fitzpatrick was raped in Spokane. The attacker told her he had a knife. She gave the police a description of her attacker. Semen samples were collected. Approximately 2 weeks after the attack, Fitzpatrick was hypnotized, but no different or additional evidence was obtained. On March 10, 1981, she identified the defendant at a lineup as the attacker. The defendant was convicted on these three counts of first degree rape and given sentences of 25 years for count 3, 30 years for count 5, and life for count 6, to run consecutively, for a total sentence of life, plus 55 years. The case is before this court on direct review. I The defendant first challenges the admission of the identification testimony of the previously hypnotized victims as *837 directly contrary to the holdings of this court in the prior Coe decision and two other cases. We have announced our rules regarding the admissibility of testimony by previously hypnotized witnesses in three companion cases, State v. Martin, 101 Wn.2d 713, 684 P.2d 651 (1984); State v. Laureano, 101 Wn.2d 745, 682 P.2d 889 (1984); and State v. Coe, supra. The Martin court stated: [A] person, once hypnotized, should be barred from testifying concerning information recalled while under hypnosis. Martin, at 722. The court also prohibited the admission of "testimony by a witness as to a fact which became available following hypnosis ..." Martin, at 714. The court established a separate rule regarding the admissibility of testimony as to facts recalled prior to hypnosis. A party seeking to admit such testimony has "the burden of establishing what the witness remembered prior to the hypnosis." Martin, at 722. The proponent should have some independent verification of the witness' prehypnotic memory, such as a record preserved prior to hypnosis. State v. Martin, at 722-23. In the second case, State v. Laureano, we specifically excluded a lineup identification made after hypnosis, even though the witness had given a description to the police prior to the hypnosis. [W]e hold that all posthypnotic testimony should be rejected, and only the prior recall of the witness, properly preserved and documented (as set forth in State v. Martin), should be allowed in evidence. State v. Laureano, at 753. The court held the process of hypnosis itself necessarily affects everything the witness recalls about the incident thereafter and stated, "[t]he plain fact is that such testimony is not and cannot be reliable." State v. Laureano, at 752 (quoting Diamond, Inherent Problems in the Use of Pretrial Hypnosis on a Prospective Witness, 68 Calif. L. Rev. 313, 348-49 (1980)). In the third opinion, the prior Coe decision, the court stated: *838 The unreliability of hypnosis as a means of restoring memory makes the use of hypnotically aided testimony unacceptable in the context of a criminal trial.... ... Thus, testimony as to facts recalled during hypnosis would be inadmissible. Coe, at 786. In the trial court below, the State successfully argued this specific language did not prohibit the admission of the identification testimony since the descriptions by the victims did not change during hypnosis, so the prehypnotic descriptions and later identifications were not "hypnotically aided", nor did they constitute "facts recalled during hypnosis". [1] This, however, contradicts the mandate of Laureano, which rendered all posthypnotic testimony inadmissible, treating the hypnosis as a time barrier after which no admissible identifications could be made. The trial court below did not address the Laureano decision in its ruling on hypnosis. It is impossible to distinguish the counts involving hypnotism in the present case from Laureano. In both cases witnesses gave detailed descriptions of their attackers prior to hypnosis, but made no identification until after hypnotism. Laureano held under these circumstances the posthypnotic testimony was inadmissible, and we must reach the same result here. We note the recent opinion of the United States Supreme Court in Rock v. Arkansas, ___ U.S. ___, ___ L.Ed.2d ___, 107 S.Ct. 2704 (1987), which found the prohibition of certain posthypnotic testimony to violate the constitutional right of criminal defendants to testify on their own behalf. Since the decision is expressly limited to the testimony of criminal defendants, it has no effect upon our decision here regarding the posthypnotic testimony of other witnesses. The trial court's ruling did not comply with the explicit rule set forth in Laureano. The admission of the posthypnotic identifications mandates reversal of counts 5 and 6. *839 The conviction of the defendant for count 3 is not affected, however, since no hypnotic testimony was presented regarding that count. II The defendant next challenges the loss of the vaginal swab samples from counts 2 (South) and 3 (Harmia), which had been used to perform semen identification tests. We conclude the defendant failed to prove a reasonable possibility that further testing would have been possible or would have refuted the State's tests. [2] The prosecution has a duty to preserve all potentially material evidence. State v. Wright, 87 Wn.2d 783, 793, 557 P.2d 1 (1976). The duty to preserve evidence does not, however, require the State to preserve evidence longer than reasonably necessary to provide the defendant a reasonable opportunity to obtain access to the evidence. The only Washington case which has reversed a conviction for destruction of evidence involved destruction before the defendant had any opportunity to view the evidence and before any testing by either party. State v. Wright, supra. See also Hilliard v. Spalding, 719 F.2d 1443 (9th Cir.1983) (no matching tests conducted before destruction of semen samples). In the present case, three separate samples were collected for count 2 (South), count 3 (Harmia), and count 6 (Fitzpatrick). The charge on count 2 was dismissed below and is not at issue in this appeal nor any evidence from it. The sample in count 3 (Harmia) was discarded some time after the first trial and conviction, and there is no showing it was due to any other cause than inadvertence or good faith. The sample from count 6 (Fitzpatrick) was never discarded. The defense received the count 6 sample but never retested it. The defendant has failed to prove the missing evidence would, with reasonable possibility, have helped to exculpate him. In State v. Vaster, 99 Wn.2d 44, 659 P.2d 528 (1983), we held the inadvertent or good faith failure to preserve evidence does not abridge a defendant's right to a fair trial *840 unless there existed a reasonable possibility the particular evidence was material and favorable to the defense: Lost or destroyed evidence which does not rise to the level of establishing a "reasonable possibility" that it will exculpate a defendant will be deemed insufficiently material to constitute a due process violation. State v. Vaster, at 52. The results of the State's tests only served to indicate Coe and the attacker were within the same approximately 38 percent of the population. The defense has not shown that the samples could have been effectively tested several years later, nor did it indicate the State's tests were improperly performed. The defense expert conceded he believed the State's tests were properly performed. The defense also failed to test the one sample that did remain. The defendant's arguments on this issue do not present a basis for reversal. III Prior to trial, the defendant moved to exclude reference to the conviction of Ruth Coe, Kevin Coe's mother, for solicitation to commit the murders of the Spokane County judge and prosecutor following Kevin Coe's first trial. The trial court ruled the prosecutor could use the evidence of the conviction to impeach Ruth Coe if she testified. Following this ruling, the defense decided not to call Ruth Coe as a witness. The defendant contends the trial court erred in ruling the evidence could be admitted, depriving the defendant of an important witness. We conclude the evidence of the conviction was relevant, probative, and not unduly prejudicial to defendant. ER 609 provides: For the purpose of attacking the credibility of a witness, evidence that he has been convicted of a crime shall be admitted ... during cross examination but only if the crime (l) was punishable by death or imprisonment in excess of 1 year under the law under which he was convicted, and the court determines that the probative value *841 of admitting this evidence outweighs its prejudicial effect to the defendant ... [3] The central issue under this rule is the balancing of probative value and prejudicial effect. The defense apparently concedes the evidence had probative value and does not dispute it would show interest or bias of a witness. Nonetheless, the defense contends the prejudicial effect was to deprive the defendant of her testimony, given at the first trial, regarding his habitual schedule, his wardrobe and appearance during the relevant time period, and his relationship to his former girl friend. However, both the defendant himself and his father, Gordon Coe, remained available to testify to these matters. It does not appear the deprivation of Ruth Coe's testimony produced any unfair prejudice to the presentation of Coe's defense. IV The defendant contends the trial court erred in denying challenges for cause to jurors who were aware of the outcome of the first trial or Ruth Coe's conviction. We find no error, since the trial court relied on the jurors' assurances of impartiality. [4] The Supreme Court has stated a prospective juror who has prior knowledge of the defendant's prior conviction and reversal need not be automatically disqualified. Patton v. Yount, 467 U.S. 1025, 81 L.Ed.2d 847, 104 S.Ct. 2885 (1984). A trial court may decide, as a question of fact, whether a juror's "protestation of impartiality" should be believed, and the appellate court must then give great deference to that finding. Patton, at 1036. The defense primarily relies on United States v. Williams, 568 F.2d 464, 471 (5th Cir.1978). Williams, however, was relied upon by the Third Circuit in Yount v. Patton, 710 F.2d 956, 969 (3d Cir.1983), which the Supreme Court later reversed in Patton v. Yount, supra. Further, as the trial court below observed, the jury inevitably would learn of the prior trial during the course of the retrial. In State v. Latham, 100 Wn.2d 59, 64, 667 P.2d 56 (1983), the court noted: *842 Petitioner's argument assumes he was entitled to a jury which was totally ignorant of the subject matter of the case and the witnesses. This standard was rejected in Irvin v. Dowd, [366 U.S. 717, 6 L.Ed.2d 751, 81 S.Ct. 1639 (1961)]. There the Court noted: It is not required, however, that the jurors be totally ignorant of the facts and issues involved.... It is sufficient if the juror can lay aside his impression or opinion and render a verdict based on the evidence presented in court. Irvin, at 722-23. The refusal to excuse these jurors for cause was not error by the trial court. The defense also raises an issue regarding the seating of an alternate juror on the jury. Three alternate jurors had been selected before trial began. One of the original jurors had to be excused and was replaced with alternate juror number 1. A second juror was excused during the course of the trial and replaced with alternate juror number 2. Defense apparently argues the selection among the alternate jurors should have been random rather than sequential. CrR 6.5 provides: If at any time before submission of the case to the jury a juror is found unable to perform his duties the court shall order him discharged, and the clerk shall draw the name of an alternate who shall take his place on the jury. [5, 6] We conclude the constitutional requirement of a randomly selected jury was satisfied by the initial random selection of jurors and alternate jurors from the jury pool. See State ex rel. Murphy v. Superior Court, 82 Wash. 284, 144 P. 32 (1914); State v. Killen, 39 Wn. App. 416, 693 P.2d 731 (1985). Further, the defendant did not challenge the trial court's compliance with CrR 6.5 at any time before appeal. Since the issue is compliance with a procedural rule rather than a constitutional issue, it may not be raised for the first time on appeal. See Wilson v. Steinbach, 98 Wn.2d 434, 440, 656 P.2d 1030 (1982); State v. Portnoy, 43 Wn. App. 455, 465, 718 P.2d 805 (1986). V The defendant contests the admission of testimony by *843 Dr. Robert Wetzler, regarding an interview and evaluation he conducted of the defendant after the conviction in the first trial and before sentencing. During the interview, the defendant admitted he "did" the Fitzpatrick rape, count 6. The admission was apparently made as part of an unsuccessful strategy of the defendant to be sent to Western State Hospital instead of being sentenced. The defendant claims the interview was privileged. [7] In order for a privilege to apply, the communication must originate in confidence that it will not be disclosed. State v. Wilder, 12 Wn. App. 296, 299, 529 P.2d 1109 (1974); 8 J. Wigmore, Evidence § 2285 (1961). This court has stated a forensic examination, made not for purposes of treatment but for the publication of results, is not privileged. State v. Sullivan, 60 Wn.2d 214, 223-24, 373 P.2d 474 (1962). In the present case, both parties anticipated the communications would be revealed in court. Wetzler was employed specifically to testify. Wetzler testified he informed Coe three times the conversation was not privileged. We find the communication was not privileged, and the trial court did not err in admitting the testimony. VI The defendant contends the trial court abused its discretion in ruling inadmissible the expert testimony of Dr. Elizabeth Loftus regarding the fallibility of eyewitness identification testimony. [8] The specific issue of the admissibility of Dr. Loftus' testimony has arisen numerous times in the appellate courts of this state. The decisions have consistently found the refusal to admit the testimony to be within the discretion of the trial court. State v. Guloy, 104 Wn.2d 412, 430, 705 P.2d 1182 (1985), cert. denied, 475 U.S. 1020 (1986); State v. Jordan, 39 Wn. App. 530, 542, 694 P.2d 47 (1985); State v. Barry, 25 Wn. App. 751, 760-61, 611 P.2d 1262 (1980); State v. Brown, 17 Wn. App. 587, 596-97, 564 P.2d 342 (1977). *844 The defendant cites, however, a recent Court of Appeals decision, State v. Moon, 45 Wn. App. 692, 726 P.2d 1263 (1986), in which the court defined limited circumstances where the exclusion of Dr. Loftus' testimony would constitute an abuse of discretion: (1) an eyewitness identification is the principal issue at trial; (2) the defendant presents an alibi defense; and (3) there is little or no other evidence linking the defendant to the crime. Moon, at 697. Even if such a test were accepted by this court, the test has not been met here. Other evidence was presented linking the defendant to the crimes, including an admission by him to one of the rapes. We find the exclusion of Dr. Loftus' testimony was not an abuse of discretion. VII The defendant contends the trial court erred in allowing the crimes to be charged as first degree rape since the State presented no evidence of the actual presence of a deadly weapon. We conclude the element of first degree rape requiring use or threat of use of a deadly weapon is satisfied by the threat itself, without evidence of the actual existence of a deadly weapon. At the time of the crimes involved here, RCW 9A.44.040 defined rape in the first degree to occur when the attacker "[u]ses or threatens to use a deadly weapon". The issue of whether an actual deadly weapon must be present was raised but not decided by this court in State v. Hentz, 99 Wn.2d 538, 663 P.2d 476 (1983). In Hentz, the 4-judge lead opinion concluded the defendant need not possess an actual weapon to satisfy the elements of first degree rape. The defendant had apparently possessed a toy gun which he represented to the victim as real. The four Justices agreed with State v. Ingham, 26 Wn. App. 45, 612 P.2d 801, review denied, 94 Wn.2d 1008 (1980), which upheld a first degree rape conviction where the defendant threatened the use of a knife, although no knife was seen or found upon the defendant. Hentz, at 542. The fifth Justice necessary to create a majority concurred *845 only on the basis there was evidence the defendant had a real gun. Hentz, at 546 (Dore, J., concurring). The opinion thus did not fully resolve the issue where no evidence is presented of an actual deadly weapon. [9, 10] Our conclusion the statute does not require evidence of an actual deadly weapon is supported by several considerations. First, as the lead opinion in State v. Hentz, supra, noted, the effect upon the victim is the same whether the deadly weapon is actually seen or merely described, by removing the possibility of self-defense: The concern regarding a perpetrator of a rape threatening to use a deadly weapon is a legitimate one. The believable or credible threat to use a deadly weapon will likely instill a greater fear in the victim than any other type of threat. If the defendant threatens to strangle his victim, she has at least an opportunity to defend herself; but the same does not apply to the threat to use a gun, knife or other deadly weapon. Hentz, at 544. The opinion stated the difference between the first and second degree rape statutes reflected the more serious nature of a credible threat to use a deadly weapon. This type of threat is equally terrifying and effective whether or not the perpetrator actually possesses a deadly weapon, in light of the personal nature of the crime and the inability of a victim to defend against a bullet or other deadly force. Hentz, at 544. Further, the first degree rape statute is unique among serious felony statutes in including threatened use of a deadly weapon within the definition of the offense. Definitions of other first degree felony offenses involving weapons do not include threatened use within the definition. See, e.g., RCW 9A.32.030 (first degree murder); RCW 9A.36.010 (first degree assault); RCW 9A.56.200 (first degree robbery). See Hentz, at 545. The history of the legislative response to Ingham and Hentz also supports our interpretation of the statutory language. After Ingham was decided, the Legislature met several times and did not change the language addressed in *846 Ingham. In fact, it specifically considered and reenacted the same language in the first degree rape statute several times. Laws of 1982, ch. 192, § 11, p. 804; Laws of 1982, ch. 10, § 3, p. 87; Laws of 1981, ch. 137, § 36, p. 533; Laws of 1981, ch. 136, § 57, p. 495. The Legislature is deemed to acquiesce in the interpretation of the court if no change is made for a substantial time after the decision. Hangman Ridge Training Stables, Inc. v. Safeco Title Ins. Co., 105 Wn.2d 778, 789, 719 P.2d 531 (1986); Nyland v. Department of Labor & Indus., 41 Wn.2d 511, 513, 250 P.2d 551 (1952). We therefore can conclude the legislative silence after Ingham was an indication of legislative approval of the Ingham interpretation of the statute. The defendant claims a later addition to the statute supports his position. The Legislature in 1983 amended the language to include use or threat of use of a deadly weapon "or what appears to be a deadly weapon", Laws of 1983, ch. 73, § 1, p. 433. The Legislature made this amendment after the Court of Appeals published its Hentz decision which had held evidence of a toy gun would not support a conviction for first degree rape. State v. Hentz, 32 Wn. App. 186, 647 P.2d 39 (1982), rev'd, 99 Wn.2d 538, 663 P.2d 476 (1983). We do not believe the purpose of the amendment, as the defendant asserts, was an indication the threat alone was not sufficient. Rather, the language was added by the Legislature in light of the conflicting results of Ingham and the Court of Appeals decision in Hentz to avoid disparity of treatment between a defendant who threatens with a displayed but ineffective deadly weapon and a defendant, as here and in Ingham, who threatens the use of, but never displays, a deadly weapon. Other jurisdictions have interpreted similar statutory language and also found no evidence of an actual weapon is necessary. In People v. Dodt, 61 N.Y.2d 408, 462 N.E.2d 1159, 474 N.Y.S.2d 441 (1984), the court interpreted a criminal statute with the language "using or threatening to use deadly force", and concluded: *847 Proof ... that defendant threatened the use of a gun is sufficient to establish ... threatened use of deadly physical force, even though there is no evidence that in fact he possessed an operable firearm.... ... ... Defendant argues ... that the proof was insufficient because it must be shown that the threat made was capable of present realization and there was no evidence at trial that he actually possessed a gun... The plain language of the statute requires rejection of defendant's argument. Dodt, at 411, 414. See also People v. Gallegos, 193 Colo. 108, 109, 563 P.2d 937 (1977) (threat of criminal violence satisfies statutory requirement of "felony involving the use of force or violence or the use of a deadly weapon"). The Hentz lead opinion concluded: [I]t is clear from the terms of RCW 9A.44.040(l)(a) that the threat which he made in perpetrating the rape, coupled with the circumstances surrounding that threat which lent it credibility, is the crux of his conviction for first degree rape. Hentz, at 545. In the present case, the circumstances surrounding the attacks also lent credibility to the threats made. The counts in the present case were properly charged against the defendant as first degree rape. VIII After Wetzler testified in the State's case in chief, the trial court permitted the State to recall him to rebut statements made by Coe. The defendant contends this was an improper use of rebuttal, allowing Wetzler to reiterate his testimony. We find the rebuttal testimony properly responded to new issues raised by the defense and was not repetitive. [11] The general standard for rebuttal testimony is stated in State v. White, 74 Wn.2d 386, 394-95, 444 P.2d 661 (1968): Rebuttal evidence is admitted to enable the plaintiff to answer new matter presented by the defense. Genuine rebuttal evidence is not simply a reiteration of evidence *848 in chief but consists of evidence offered in reply to new matters. The plaintiff, therefore, is not allowed to withhold substantial evidence supporting any of the issues which it has the burden of proving in its case in chief merely in order to present this evidence cumulatively at the end of defendant's case. Ascertaining whether the rebuttal evidence is in reply to new matters established by the defense, however, is a difficult matter at times. Frequently true rebuttal evidence will, in some degree, overlap or coalesce with the evidence in chief. Therefore, the question of admissibility of evidence on rebuttal rests largely on the trial court's discretion, and error in denying or allowing it can be predicated only upon a manifest abuse of that discretion. (Citations omitted.) The State has thoroughly demonstrated that each item of Dr. Wetzler's rebuttal testimony was presented in direct response to assertions made by Coe on the stand and was not repetitive of Dr. Wetzler's prior testimony. The trial court's ruling fails to demonstrate manifest abuse of discretion in admitting the rebuttal testimony. IX One of the reasons the court allowed Dr. Wetzler to be recalled was to confirm or deny Coe's assertion he had told Dr. Wetzler that Diane Fitzpatrick had lied about the knife. The court referred to this issue during the rebuttal testimony but outside the presence of the jury. The defense contends the court's comments improperly suggested a question to the prosecution. We conclude the court's conduct did not present undue bias. The verbatim report of proceedings indicates the court had initially expressed its concern regarding the complexities of rebuttal testimony and had allowed the State to ask leading questions in order to carefully define the scope of the testimony: THE COURT: I was doing that because I have a concern about a mistrial. I think we are all aware in this particular area if you don't walk very carefully, there are some real dangers of a mistrial in this area because of questions *849 being asked, and that's why with this particular witness I'm exceedingly careful about what questions are going to be asked and what the answers are going to be, because if you get certain answers, as you have indicated to me and I have indicated to you, there is a grave danger of a mistrial. I think that's something we all want to stay away from. The court's subsequent comments can be understood as its attempt to enforce the scope of the rebuttal testimony. The first comment by the court regarding the knife was part of its ruling allowing Dr. Wetzler to be recalled to testify on that subject. The second comment by the court referring to the questioning regarding the knife can also be explained by the trial court's intent to control the use of the rebuttal: THE COURT: It should be noted that this is entirely outside the presence of the jury ...; that the jury is not present; that we had this discussion about 15 minutes ago when the State indicated that they were going to bring up the subject of a knife. Thereafter, they did not bring up the subject of a knife, and it seemed to me that I cannot let this matter just vanish in sort of thin air. It is not my intention to give the State hints in that regard, but I think that just before about 15 minutes ago, they did mention the fact that they were going to bring it in. The question that I would have is then what happened to this evidence, and the inference could be drawn that there was some bad faith on the State's part, so I think it should be made a matter of record as to whether this is good or bad faith. ... It is not my intention to hint to the State, but to find out what had happened in the meantime. Although the court may have been unartful in its use of language, its focus was the prevention of a mistrial resulting from its rulings. The question discussed regarding the knife statement had already been raised by the prosecution more than once as a matter that it would raise during rebuttal. The court's statements were made outside the presence of the jury, so no bias or appearance of unfairness was presented to the trier of fact. It does not appear the statements made by the court were suggestive to the prosecution or presented an appearance of unfairness to the jury. *850 The conviction and sentence for first degree rape in count 3 is affirmed, and the convictions for counts 5 and 6 are reversed due to the admission of posthypnotic identification testimony. UTTER and GOODLOE, JJ., and CUNNINGHAM, J. Pro Tem., concur. PEARSON, C.J., dissents from the affirmance of count 3. GOODLOE, J. (concurring) I agree with the majority opinion in all respects. I write separately only to address the dissent's concern that Coe's conviction on count 3 (Harmia) should be reversed due to cumulative taint from the admission of improper posthypnotic testimony on counts 5 and 6. There is nothing in the record to suggest that the admission of this improper testimony in any way prejudiced Coe's conviction on count 3. The jury was told to consider each count separately, and a jury is presumed to follow the court's instructions. State v. Grisby, 97 Wn.2d 493, 499, 647 P.2d 6 (1982), cert. denied, 459 U.S. 1211 (1983). The lack of prejudice in this case is demonstrated by the fact that the jury failed to convict Coe on one of the charged counts. This demonstrates that the jury was able to consider each count independently as it had been instructed. Absent a showing of prejudice, I see no reason to reverse Coe's conviction on count 3 for the rape of Julie Harmia. DORE, J. (dissenting) Kevin Coe was originally convicted of four counts of first degree rape. This court, however, reversed his convictions and remanded the case for retrial, in part because much of the evidence against him came from witnesses who had previously been hypnotized. State v. Coe, 101 Wn.2d 772, 684 P.2d 668 (1984). This court gave the trial court explicit instructions on what portion of the hypnotized witnesses' testimony could be admitted at the subsequent retrial. "Upon remand, the admissibility of the testimony of the previously hypnotized *851 witnesses should be determined in accordance with our holding in State v. Martin, [101 Wn.2d 713, 684 P.2d 651 (1984)]." Coe, at 786. Unfortunately, this mandate was not followed, and I feel this court has no choice but to reverse all three convictions against Coe. I write separately both to clarify further our prior ruling in State v. Martin, supra, and State v. Coe, supra, and why I believe the majority's decision to affirm the first rape conviction is legally impossible. INADMISSIBILITY OF HYPNOTIC TESTIMONY The majority bases its decision to reverse the second and third convictions on our decision in State v. Laureano, 101 Wn.2d 745, 682 P.2d 889 (1984). While I do not quarrel with the majority's analysis, I believe that since our decision in State v. Coe, supra, expressly mandated that the trial court follow our decision in State v. Martin, supra, the majority's analysis should be confined to that opinion. Martin became the law of the case and Laureano, which is somewhat different in its conclusions than Martin, should not be relied on as dispositive. The recent decision of the United States Supreme Court in Rock v. Arkansas, ___ U.S. ___, ___ L.Ed.2d ___, 107 S.Ct. 2704 (1987) does not apply to this case. Rock held that Arkansas' statute excluding the posthypnotic testimony of a criminal defendant infringes impermissibly on the defendant's due process right to testify on his or her own behalf. There are no such constitutional considerations surrounding the issue of posthypnotic memories of witnesses against Coe. In State v. Martin, supra, we held that hypnotism-aided testimony cannot be admitted into evidence. Many of the problems associated with hypnotically induced testimony make its use in trial particularly dangerous. After hypnosis, neither subject nor expert observer is able to distinguish between confabulations and accurate recall in any given case, absent corroborating evidence. See Beaver, Memory Restored or Confabulated by Hypnosis — Is it Competent?, 6 U. Puget *852 Sound L. Rev. 155, 199 (1983). The subjective conviction in the truth of the memory after hypnosis eliminates fear of perjury as a factor ensuring reliable testimony. Additionally, effective cross examination is seriously impeded, as the witness cannot distinguish between facts known prior to hypnotism, facts confabulated during hypnosis to produce pseudomemories, and facts learned after hypnosis. Finally, jury observation may be adversely affected, as the witness, as a result of the hypnosis, will have absolute subjective conviction about a particular set of events, whether or not his perceptions are objectively accurate. Beaver, at 200-01. It is this tendency toward immunization from meaningful cross examination in particular that leads us to conclude that a person, once hypnotized, should be barred from testifying concerning information recalled while under hypnosis. Hypnosis in its current state simply does not meet the [Frye v. United States, 293 F. 1013 (D.C. Cir.1923)] standards of reliability and accuracy. (Italics mine.) State v. Martin, supra, at 721-22. The effect of the hypnotic testimony in this case was especially prejudicial to Coe. Only one of the four raped women identified Coe without hypnosis, and her original identification was not definite. While the three other women were more positive of their posthypnotic identification of Coe, their certainty was tainted by hypnotic sessions. Sherry South, one of the complaining witnesses, testified as follows: Q. [By MR. ALLEN] Sherry, before you went into hypnosis when you weren't clear on a lot of things, they told you that they were trying to see why you couldn't remember a lot of things and that's why they recommended hypnosis to you, is that right? A. I don't know what things you mean that I don't remember, so I really — Q. But it is true that you were unsure of a lot of things before you went into hypnosis, is that right? A. No, I don't believe so. Q. Sherry, I'm going to ask you to take a look at a page of the transcript from the last trial. This is page 407, and ask you to take a look at line 14, and I'll read that along to you. This is page 407, line 14: "Before I *853 went into hypnosis, when I wasn't clear on a lot of things, they were trying to see why I couldn't remember a lot of things. They wanted to narrow it down. So they figured maybe hypnosis." And then the question is, "Who told you that?" And you said, "I don't remember." And the question was, "One of the officers?" And you answered, "I think it was Joan." Is that right? Is that what is down there? A. Yes, that's what is there. ... Q. And the next question is, "And, it is true that there were a lot of things that you were unsure of before you went under hypnosis, isn't that true?" And you answered, "True," isn't that right? A. Yes. (Italics mine.) Verbatim Report of Proceedings, at 1890-91. Similarly Fitzpatrick testified: Q: And, at that time didn't you state that you didn't get that good of a look? A: [FITZPATRICK] Well, I didn't really remember at that time. I was kind of confused. But, after I was hypnotized, I kind of remembered that I was looking at him when he — Q: You were hypnotized also? A: Yes. ... Q: [By MR. GIGLER] Now, your memory evidently improved after you were hypnotized, is that correct? A: Yes. I guess. Q: Did you separate what you remembered before you were hypnotized from what you remembered after you were hypnotized? A: Well, I remembered most of, you know — Q: [MR. GIGLER] Ok. Go ahead. A: (Continuing) — What I told the police officers. But, I remembered more of — I remembered that I looked at him when I was getting my money out. But, I could see more, remember more of what he looked like. (Italics mine.) Brief of Appellant, at 21-22. Moreover, the inconsistencies in their testimony was very apparent. For example, prior to being hypnotized, one of the victims, Diane Fitzpatrick, selected a different individual *854 from a series of photographs shown to her. Verbatim Report of Proceedings, at 1733. Another victim, Mary Patricia Strange, chose two individuals other than Coe from a montage shown to her prior to her being hypnotized. She also described her attacker as having acne or pockmarks even though ample testimony at trial indicated that Coe never had facial blemishes. Verbatim Report of Proceedings, at 2029, 2820. A third victim, Sherry South, identified her assailant as having shoulder length blond hair, and possibly driving a yellow car. Verbatim Report of Proceedings, at 177, 1853. Coe neither had long hair nor owned such a car. Considering the inconsistency of these women's testimony before and after the hypnosis, it would be imperative for Coe's counsel to attempt to impeach these victims' credibility. However, with the effectiveness of cross examination reduced or eliminated because of the apparent subjective conviction the witnesses achieved as a result of the hypnosis, it is clear in my mind that Coe did not receive a fair trial. Three of the four victims' identification of Coe as the rapist, the crucial issue in this case, was rendered unreliable by the hypnosis. The trial judge erred, therefore, by allowing these women to testify about events they remembered after the hypnosis session. Although in State v. Coe, supra, we had specifically allowed for a witness to testify as to facts recalled prior to hypnosis, we prohibited evidence provided after the hypnosis even if the trial court believed it to be untainted by the hypnotic session. So long as the trial court is careful to limit testimony to the witness' prehypnotic memory of events, and opposing counsel has the opportunity to demonstrate to the jury that the witness has been subjected to hypnosis, the danger of prejudice is minimized. (Italics mine.) Coe, at 722. Since hypnosis taints all memories of a given event, unless those prehypnotic memories are recorded prior to the hypnotic session (i.e., in a police report, conversation to officers or family, signed statement, *855 etc.), no posthypnotic testimony can be admitted. The trial court totally disregarded this mandate and, after a 7 1/2-day hearing designed to separate pre- and posthypnotic memories, allowed the victims to testify about identifications made after hypnotic sessions. This was reversible error. The Martin case indicated that such an approach would give trial courts the impossible task of deciphering a maze of conflicting and confusing data. Martin, at 722; see Beaver, Memory Restored or Confabulated by Hypnosis — Is it Competent?, 6 U. Puget Sound L. Rev. 155, 194 (1983). The evidence presented against Coe was not reliable because of the hypnosis, and Coe's counsel could not effectively cross-examine the victims to impeach their testimony. At a minimum, therefore, Coe will have to be retried on the second and third rape charges (counts 5 and 6). The victims will have to relive those horrible moments yet again, although this time their testimony will be confined to those prehypnotic memories. It is with great regret that I agree to remand this case, but trial court error leaves this court no other choice. SECOND RAPE CHARGE Both the majority and I agree that the second rape charge (Strange) must be reversed. Conspicuously absent from the majority opinion, however, is whether that charge and for that matter the third rape charge (Fitzpatrick) should be dismissed for lack of sufficient admissible evidence, or remanded for retrial. Because of the earlier difficulties with this case, I believe we abrogate our duty to the public, the victims and to the defendant by not clarifying this issue. The second rape charge (count 5) against Kevin Coe stemmed from the heinous attack against Mary Patricia Strange. Ms. Strange was attacked on February 5, 1981, and later that day, she described her assailant to the police. This description differed in some aspects from Coe's actual appearance. Ms. Strange described her attacker as having blond hair and pockmarks on his face (Verbatim Report of *856 Proceedings, at 2025, 2035), while Coe has light brown hair and no facial blemishes. Furthermore, there were no blood or semen samples available to link Coe to this rape. Moreover, contrary to the majority's assertion that she gave no different evidence concerning the rapist after hypnosis, she gave a police artist additional data with which to sketch a second, and in her opinion, better composite drawing. When she eventually identified Coe after being hypnotized, she qualified her identification with the words "very possibly." Of course, none of this evidence obtained after hypnosis can be admitted, because it lacks sufficient reliability due to the hypnotic session's taint. I would hold that the evidence the State offered is insufficient to prove to a jury beyond a reasonable doubt that Coe committed the Strange rape. Ms. Strange's prehypnotic description of the rapist did not match Coe's appearance. No chemical evidence linked him to the crime. The attack on Ms. Strange differed from the other two rape convictions in that no sexual intercourse took place. Therefore, because Ms. Strange's posthypnotic recollections and identifications cannot be admitted pursuant to our decision in State v. Martin, supra, and no other competent testimony is available, I would dismiss this charge against Coe for lack of sufficient evidence. THIRD RAPE CHARGE A detailed analysis of whether the third rape charge (Fitzpatrick) (count 6) against Coe should be dismissed is unnecessary. Ample evidence exists to show that Diane Fitzpatrick was raped on February 9, 1981. Coe admitted to Dr. Robert A. Wetzler after he had been found guilty of this attack that he did commit this rape. I agree with the majority that this is a nonprivileged interview. Evidence of a crime shown by independent evidence, coupled with an extrajudicial confession by the defendant, is sufficient to sustain a conviction. State v. DePriest, 16 Wn. App. 824, 560 P.2d 1152 (1977). Of course, on remand, a jury will have to decide whether Coe's confession was truthful or, *857 since he had already been convicted of the crime, merely a ploy to reduce the sentence he believed would be imposed on him. Moreover, Ms. Fitzpatrick may only testify about her recollections before her hypnotic session. Nevertheless, this charge should not be dismissed for lack of sufficient evidence. FIRST RAPE CHARGE The majority fails to address the issue of whether the first rape conviction (Harmia) should be affirmed in light of this court's decision to reverse the second and third rape convictions. The majority simply affirms the first conviction "since no hypnotic testimony was presented regarding that count." Majority, at 839. I believe the majority has misstated the issue, and that this first conviction cannot be affirmed in light of all of the other inadmissible evidence which was erroneously introduced at trial. Of crucial importance in all three of these charges was the identity of the rapist. Inadmissible posthypnotic testimony which indicated that Coe committed the second and third rapes necessarily influenced the jury's perception that Coe in fact committed the first attack. This taint caused by the cumulative evidence is exacerbated by the fact that the first victim (count 3), Julie Harmia, who never was hypnotized, was not definite about her identification of Coe. When she identified Coe at the police lineup she wrote "close" on her identification form. Verbatim Report of Proceedings, at 1607. She also described him as having dark brown hair and a dark complexion. Verbatim Report of Proceedings, at 1582. Coe has light brown hair and a fair complexion. I believe that because the identity of the attacker is so crucial to this trial, the cumulative evidence of Coe's identity improperly introduced at trial was highly prejudicial. In fact, I fail to see how any juror in this situation could not be influenced by the improper testimony of the second and third victims. I believe this court, therefore, has no alternative but to reverse the first conviction and remand *858 for a new trial in which no posthypnotic testimony can be introduced which would influence the jury's decision as to whether Coe was the person who committed any or all of these offenses. CONCLUSION It is with regret that I would remand this case for yet another trial. For the victims to have to relive these traumatic events a third time is tragic. However, to have a man spend the rest of his life in jail on the basis of unreliable evidence of identity cannot be countenanced by this court under any circumstances. I would dismiss the second rape charge (Mary Patricia Strange). I would remand the first (Harmia) and third (Fitzpatrick) charges for retrial without the admission of posthypnotic testimony. ANDERSEN, J. (dissenting) The defendant, Kevin Coe, was tried in accordance with the instructions laid down by this court in its decision reversing his first conviction. His second trial was conducted with scrupulous fairness. I would affirm his jury conviction on all three counts of rape in the first degree. Accordingly, I dissent from the majority opinion which reverses the Strange and Fitzpatrick counts (as well as from Justice Dore's "dissenting" opinion which would reverse all three counts — Strange, Fitzpatrick, and Harmia). SUMMARY OF DISSENT In State v. Coe, 101 Wn.2d 772, 684 P.2d 668 (1984), this court reversed Kevin Coe's Spokane conviction of four counts of first degree rape on various grounds. In remanding the case to the trial court for retrial, it instructed the trial court as follows: The testimony of the previously hypnotized witnesses would be admissible if the State can show that the testimony consists solely of prehypnotic memory, . .. *859 (Italics mine.)[1] At the retrial in Seattle, pursuant to this instruction, the trial court meticulously reviewed the evidence in two detailed written orders, totaling some 37 pages. It carefully excluded all statements made by witnesses during hypnosis and all facts recalled during hypnosis. The trial court, as distinguished from this court, is the fact finder. As such, it found as a fact that the State had sustained its burden of proving that the victims' identifications of Kevin Coe were based on their prehypnotic memories. Consequently, at the retrial, the trial court admitted their pretrial identifications and permitted them to identify Coe in a court as their attacker. At that retrial, the jury found Coe guilty of three counts of first degree rape and he was subsequently sentenced thereon. For each of three separate reasons, I would affirm the defendant's convictions on all counts. First, the trial court followed this court's specific instructions in the first Coe case and therefore did not commit error in admitting the victims' eyewitness identifications of Kevin Coe as their attacker. The trial court very carefully followed the guidelines expressly laid down in the first Coe case in allowing such eyewitness identification testimony as it was required to do.[2] It should, therefore, be affirmed on that basis. Second, unlike the majority, I do not read this court's decision in State v. Laureano, 101 Wn.2d 745, 682 P.2d 889 (1984) as compelling a different result, but even if it were to be so read, I would revise that holding as we have the authority to do. The mandate in the first Coe decision does not even mention this court's decision in Laureano even though it was handed down the same day as Coe. Yet the majority herein now reverses two of the defendant's convictions on the basis that Laureano "rendered all posthypnotic testimony inadmissible, treating the hypnosis as a *860 time barrier after which no admissible identifications could be made."[3] Justice Dore's "dissenting" opinion would go even further and reverse all three of the counts on essentially the same basis. Totally aside from the fact that Laureano was never so much as mentioned in the first Coe opinion, I simply do not read it as declaring the blanket testimonial prohibitions that the majority discovers in it. To the extent that Laureano could possibly be considered as so holding, I would overrule it as improvidently broad at least as it applies to this case which has already been before us. This we have the right to do under our court rules. The appellate court may at the instance of a party review the propriety of an earlier decision of the appellate court in the same case and, where justice would best be served, decide the case on the basis of the appellate court's opinion of the law at the time of the later review. RAP 2.5(c)(2).[4] Third, even if I were to assume that the majority's reading of Laureano is correct, and that it represents the current view of this court, I would nevertheless hold that in view of the substantial other evidence of the defendant's guilt that exists in this case any error committed by the trial court in this regard was harmless error. For each of these separate reasons, therefore, I would affirm Kevin Coe's convictions on all counts. DISSENT To then turn to a more detailed discussion of the bases of my dissent. Section VII of the first Coe decision addressed the admissibility of testimony by witnesses who had been hypnotized.[5] Therein, this court cited its holding in State v. Martin, 101 Wn.2d 713, 722, 684 P.2d 651 (1984) (also *861 filed the same day as the first Coe opinion), that "`a person, once hypnotized, should be barred from testifying concerning information recalled while under hypnosis.'"[6] The first Coe decision also cited the Martin exception to this prohibition: witnesses may testify as to facts recalled before hypnosis if certain procedural safeguards are met.[7] One such safeguard included preserving a detailed record of the witness' prehypnotic memory, as the party offering the testimony would have the burden of establishing what the witness remembered prior to the hypnosis.[8] If the testimony were admitted, other Martin safeguards would include giving the opponent the opportunity to show the possible effect of the hypnosis on the witness' testimony, and the manner in which the hypnosis was conducted.[9]Martin focused on an alleged victim of statutory rape who was hypnotized because she had no memory of any sexual abuse. During hypnosis, she stated that her stepfather had raped her. In court, the child stated that she remembered the offense only because of the hypnosis.[10] Since the victim in that case had no prehypnotic memory of her attack or her assailant, this court held her posthypnotic description of both to be inadmissible.[11] This court in the first Coe decision ordered that on remand the admissibility of the testimony of the previously hypnotized witnesses should be determined in accordance with Martin.[12] Thus, testimony as to facts recalled during *862 hypnosis would not be admissible.[13] Facts based on prehypnotic memory would be admissible, however, provided that Martin's procedural safeguards were substantially complied with. Obviously, since Martin had not yet been decided by this court, the authorities responsible for hypnotizing the Coe witnesses were unaware of Martin's procedural safeguards applicable to prehypnotic memory; therefore precise compliance with these requirements would not have been possible.[14] The court in Coe then went on to instruct (or mandate) whichever trial court retried the case as follows:[15] Upon remand, the admissibility of the testimony of the previously hypnotized witnesses should be determined in accordance with our holding in State v. Martin, [101 Wn.2d 713, 684 P.2d 651 (1984)]. Thus, testimony as to facts recalled during hypnosis would be inadmissible. Obviously, the authorities responsible for hypnotizing these witnesses were unaware of Martin's procedural safeguards applicable to prehypnotic memory; therefore precise compliance with these requirements was impossible. On remand, in evaluating the admissibility of prehypnotic memories, the trial judge should determine whether there has been substantial compliance with these safeguards. The testimony of the previously hypnotized witnesses would be admissible if the State can show that the testimony consists solely of prehypnotic memory, thus assuring that the purposes of Martin's procedural safeguards are satisfied. In its 7-page order denying the motion to suppress the hypnotized witnesses' identifications, and in its 30-page supplementary order, the trial court repeatedly referred to the mandate in Coe and Martin, as it carefully outlined its reasons for admitting the witnesses' identifications. The trial court first described the impact of Coe and Martin in *863 the following terms:[16] It is clear that as a result of the Washington State Supreme Court's decisions in State v. Coe [101 Wn.2d 772, 684 P.2d 668 (1984)] and State v. Martin, 101 Wn.2d 713, [684 P.2d 651] (1984) all statements made by the witnesses during hypnosis and all facts recalled by the witnesses during the hypnosis are excluded.... The Washington State Supreme Court has ruled, however, that the pre-hypnotic memories of witnesses may be admissible if it can be shown that there has been substantial compliance with the procedural safeguards set forth in Martin, supra. ... It is my finding that there has been substantial compliance with the procedural safeguards of Martin, supra, and that the State has borne the burden of showing what the memory was of each hypnotized witness prior to the date that witness was hypnotized. The State had preserved the witnesses' prehypnotic memories by means of police reports prepared before hypnosis and they showed in detail what each witness recalled before being hypnotized. Furthermore, tape recordings of the hypnosis sessions amply demonstrated what it was that the witnesses had recalled during those sessions. The central question for the trial court was whether or not the lineup identifications of Coe by the previously hypnotized victims should be excluded as being based on postrather than prehypnotic recollections. In resolving this question, the trial court considered the facts of the case important. First, since all of the hypnotic sessions occurred before Coe became a suspect, it would have been virtually impossible for the hypnotist or the police to suggest, even unintentionally, that he was the assailant. The court reviewed the hypnosis tapes and also found that they were not suggestive. Those who performed the hypnosis knew only that a rape was involved, its date and location, and the victim's name. Any chance of suggestiveness was yet further reduced by the fact that the witnesses' identifications of *864 Coe were based in large part on his voice. The trial court also observed that each of the complaining witnesses gave a detailed description of the rape and of her assailant before being hypnotized. "This is not a case where the victims had difficulty recalling the details of the crime", said the trial court, "or the description of the assailant prior to hypnosis."[17] It was likewise important to the trial court's determination that each of the victims testified to having had a clear memory of her assailant prior to hypnosis and to making her identification of Coe on the basis of her memory prior to hypnosis. The trial court then reviewed the evidence to see if there were any suggestive police procedures before or after hypnosis. The trial court found nothing suggestive about the photo montage from which Ms. Strange identified Coe or about the lineup at which all three complaining witnesses identified Coe. The trial court's examination of the evidence also showed substantial corroboration of the witnesses' identifications of Coe (I will shortly refer in more detail to this corroborating evidence). It was only after its detailed factual findings and determinations that the trial court concluded that the witnesses should be permitted to testify regarding their lineup identifications and should also be allowed to make in-court identifications. Then, in its supplementary order, the trial court carefully examined and described the prehypnotic statements of each witness as well as the comments made during hypnosis. Thereupon the trial court found as a fact that the Martin procedural safeguards had been substantially complied with in each case. The trial court also very carefully listed the bits of information recalled during hypnosis that under its ruling would not be admissible as being based on posthypnotic rather than prehypnotic memory. In both of these orders, the trial court carefully and meticulously complied with the mandate of this court in Coe and with the guidelines set forth by this court in Martin, *865 as it was required to do. The trial court specifically noted the State's substantial compliance with the Martin procedural safeguards (formulated by this court long after the rapes and hypnosis in this case had occurred), and admitted the identifications because, as the trial court carefully explained, they were based on each victim's prehypnotic memory of her assailant. Even the majority herein observes that neither the Strange nor Fitzpatrick hypnotic sessions provided different or additional evidence from that supplied by the victims before hypnosis. Thus, I am totally unable to comprehend how their later identifications of Coe can now reasonably be said to be based on facts recalled during hypnosis and hence not admissible under Coe and Martin. The facts of the case before us are clearly distinguishable from those in Martin, where the victim could not even recall her attack until she was hypnotized. Each victim in this case gave a detailed description of her rape and attacker before hypnosis, and then later testified under oath that her subsequent identifications of Coe were based on prehypnotic memory alone. The only reason that the photo and lineup identifications were posthypnotic was that Coe did not become a suspect until after the hypnosis sessions had occurred and for reasons totally unconnected with those sessions. Thus, as I view it, the trial court did not err when it ruled that the Fitzpatrick and Strange identifications were not based on facts recalled during hypnosis and admitted them as well as the subsequent in-court identifications pursuant to Coe and Martin. Once admitted, the weight of that testimony was for the jury to determine. Another section of the Coe opinion reinforces my conviction that the trial court's reading of the Coe mandate on hypnosis was sound. In section V of its Coe opinion, this court examined the propriety of ordering Coe to read aloud in court several statements allegedly made by the rapist.[18]*866 This court then held that it was not error to require him to repeat those statements.[19] Because of the distinctive nature of the rapist's voice and because the identification of defendant by many of the rape victims was based in large part on his voice, the sound of Coe repeating the words spoken by the rapist was highly relevant. This court in Coe then agreed with the first trial court's ruling that the prejudicial impact of the sound of the defendant's voice repeating the rapist's words was substantially outweighed by the probative value of the evidence.[20] The defendant was required to read these words in order to assist the victims in making an in-court identification. The jurors did not witness the rapes or hear the rapist, so the reading could not have been for their benefit. Given this court's approval of the admission into evidence of this reading by Coe, the logic of the majority in now concluding that the Coe opinion's statements on hypnosis somehow made the victims' posthypnotic identifications absolutely inadmissible escapes me. Such a reading of the Coe opinion renders one of its sections totally irrelevant and even misleading. Certainly one would reasonably assume that if Coe's reading of the rapist's words to the victims was admissible, then the victims' identifications based in part on that reading would also be admissible. That the trial court at the second trial so read the Coe opinion is both understandable and supportable. Indeed, the majority today finds it necessary to turn to yet a third case to find support for reversing the trial court's ruling admitting the victims' eyewitness identification of this defendant. The majority now relies on a decision not even mentioned in the Coe opinion or its mandate to the trial court — State v. Laureano, 101 Wn.2d 745, 682 P.2d 889 (1984). Laureano focused on the hypnosis of a woman whose *867 husband was killed when the couple was robbed at home. She was hypnotized for the express purpose of enhancing her ability to identify the perpetrators, and before the identities of the robbers were known or presumed. Three months later, she identified the defendant from a photo montage. She picked him out of a lineup 6 months later, and subsequently identified him at trial. In that case, the defendant argued that the three identifications were inadmissible because they had been improperly induced by hypnosis. This court agreed, holding that "all posthypnotic testimony should be rejected, and only the prior recall of the witness, properly preserved and documented (as set forth in State v. Martin [101 Wn.2d 713, 684 P.2d 651 (1984)]) should be allowed in evidence."[21] The Laureano dissent argued that the identifications were admissible on the grounds that the hypnosis was distinguishable from that in Martin, and was without suggestion.[22] Since the identities of the perpetrators of the crime were neither known nor presumed, the purpose of the hypnosis of [the witness] was to gain investigative leads. ... No indication is made there was any improper suggestion. Indeed, since there were no known suspects at the time of the hypnosis, there would be little, if any, risk of suggestiveness. To the extent that the majority in the present case considers Laureano as establishing an absolute bar to posthypnotic identification, the absence of suggestiveness has become irrelevant. Likewise irrelevant, according to that view, would be the extent of such highly relevant facts as how long the eyewitness viewed the assailant during the crime, the purpose of the hypnosis, or the timing of a suspect's arrest as compared with the timing of the hypnosis. This result, if the majority's present reading of it is proper, *868 is incomparably broader than anything suggested by the language of Coe and Martin, would overrule them sub silentio and is completely indefensible. The absurdity of considering the holding in Laureano as an absolute prohibition of all posthypnotic identifications is very well illustrated by the recent Court of Appeals decision in State v. Yapp, 45 Wn. App. 601, 726 P.2d 1003 (1986). In Yapp, a 16-year-old girl was forced into a truck, raped once in the truck by her kidnapper, raped three more times at his home, and forced to commit fellatio. She subsequently reported her abduction to the police and described her attacker, his truck and his house. Because of inconsistencies in her statements, she was asked to take a polygraph examination. That examination detected deception in three areas: (1) the point of abduction; (2) her failure to admit the rape in the truck; and (3) her failure to admit she was forced to commit fellatio.[23] The victim then underwent hypnosis on her own accord. She had earlier requested hypnosis, but her request had been refused. After the hypnosis, she explained the reasons for her previous inconsistencies. She had lied about the point of abduction because she did not want her father to know she was a half mile from home late at night. She had not mentioned the rape in the truck because it severely embarrassed her, and had not mentioned being forced to commit fellatio because that embarrassed her even more. Her description of her attacker, his home and his truck was the same both before and after hypnosis.[24] The victim in Yapp identified the defendant from a photographic lineup subsequent to being hypnotized. She subsequently testified at his trial, after which he was found guilty of rape in the first degree and unlawful imprisonment. *869 On appeal, the defendant argued that the trial court erred in allowing the victim to testify despite the fact she had been hypnotized. The Court of Appeals cited Martin, Coe and Laureano in holding that the fact that a person has been hypnotized before trial does not absolutely preclude that witness from testifying.[25] "A witness may testify to that which was known before hypnosis, provided the appropriate procedural safeguards are present".[26] The Court of Appeals then described the Martin safeguards and concluded that they had been satisfied given the State's "ample corroboration and independent verification" of the victim's prehypnotic memory. The Court of Appeals in Yapp then discussed the admissibility of the posthypnotic identification of the defendant:[27] Although K.H. did not identify Mr. Yapp from the photographic lineup until after being hypnotized, 9 days before being hypnotized she told the police she would recognize the suspect if she saw him again, and she testified at trial that prior to being hypnotized she had a clear picture in her mind of the person who kidnapped and raped her. The hypnotist was not given any information by the police, he did not know the details of the abduction and kidnapping, he did not hypnotize K.H. in order to prompt an identification of the assailant, and he did not ask questions during the time K.H. was under hypnosis, other than to tell her to narrate what occurred. Thus, although K.H. did not actually identify Mr. Yapp until after hypnosis, the hypnosis did not affect her identification of Mr. Yapp; it was based on her prehypnotic memory. The Court of Appeals then went on to distinguish the hypnosis in Yapp from the hypnosis session in Laureano:[28] *870 The hypnosis [in Yapp] was conducted to clear up confusion in K.H.'s mind as to certain embarrassing details and not as an aid to identification. Further, K.H., unlike the witness in State v. Laureano, [101 Wn.2d 745, 682 P.2d 889 (1984)] testified that her identification was based on her prehypnotic memory. The court in Laureano found that a witness' posthypnotic identification of a suspect is permissible as long as the identification is based on prehypnotic memory and procedural safeguards are satisfied. The trial court here meticulously applied the proper procedural safeguards. Thus, the court properly admitted K.H.'s posthypnotic identification. For all the same reasons, that should be the holding here as well. However, according to the majority's view of Laureano, the Court of Appeals in Yapp was dead wrong, the posthypnotic identification was not admissible, the victim's extensive prehypnotic memories and descriptions are irrelevant, and the fact that the victim was hypnotized because she was too embarrassed to describe certain details of her attack is also irrelevant. The foundation for a victim's identification of a defendant is no longer important, according to the majority; if the identification occurs after hypnosis, it will be automatically excluded however certain it may be! Yapp concerned a victim who was raped four times over a 5-hour period and subjected to other abuses. What about a victim who is kidnapped and raped repeatedly over a period of several days, and then hypnotized to see if she can recall the make of her assailant's car? Under Laureano, as here interpreted by the majority, once the victim is hypnotized for any reason, she can never identify her assailant in court, no matter how engrained his face may be in the victim's memory. Such an absolutist view is not realistic or, to my mind, legally or logically defensible. Particularly is this true in view of the pronouncements on this subject by the United States Supreme Court handed down subsequent to the Martin, Laureano and Coe decisions. The United States Supreme Court recently criticized a similar per se exclusion of posthypnotic testimony in Rock *871 v. Arkansas, ___ U.S. ___, ___ L.Ed.2d ___, 107 S.Ct. 2704 (1987). In Rock, the issue before the Court was the constitutionality of Arkansas' per se rule excluding a criminal defendant's hypnotically refreshed testimony.[29] The Court observed that Arkansas' absolute prohibition operated to the detriment of any defendant who undergoes hypnosis, without regard to the reasons for it, the circumstances under which it took place, or any independent verification of the information it produced.[30] The Court also noted that many courts have eschewed such per se prohibitions and allow the admission of hypnotically refreshed testimony.[31] The Supreme Court then examined hypnosis itself, finding that responses to hypnosis vary greatly, with hypnosis often having no effect at all on memory. "The most common response to hypnosis, however, appears to be an increase in both correct and incorrect recollections."[32] The Court then observed that despite the unreliability that hypnosis may introduce, the procedure has been credited as instrumental in obtaining investigative leads or identifications that were later confirmed by independent evidence. The inaccuracies the process introduces can be reduced by the use of procedural safeguards. "Such guidelines do not guarantee the accuracy of the testimony, because they cannot control the subject's own motivations or any tendency to confabulate, but they do provide a means of controlling overt suggestions."[33] Furthermore, the more traditional means of assessing accuracy of testimony remain:[34] Certain information recalled as a result of hypnosis may *872 be verified as highly accurate by corroborating evidence. Cross-examination, even in the face of a confident defendant, is an effective tool for revealing inconsistencies. Moreover, a jury can be educated to the risks of hypnosis through expert testimony and cautionary instructions. In Rock, the Supreme Court declined to unqualifiedly endorse the use of hypnosis as an investigative tool, but it did conclude that Arkansas was not justified in excluding all of a defendant's testimony that could not be proven to be the product of prehypnosis memory:[35] A State's legitimate interest in barring unreliable evidence does not extend to per se exclusions that may be reliable in an individual case. ... The State would be well within its powers if it established guidelines to aid trial courts in the evaluation of posthypnosis testimony and it may be able to show that testimony in a particular case is so unreliable that exclusion is justified. But it has not shown that hypnotically enhanced testimony is always so untrustworthy and so immune to the traditional means of evaluating credibility that it should disable a defendant from presenting her version of the events for which she is on trial. (Italics mine.) Nor do I think that the majority in the present case, or in any previous decision by this court, has demonstrated that a witness' posthypnotic identifications are always so untrustworthy that they must be totally and automatically banned from all in-court consideration. Such an absolute testimonial exclusion makes no sense to me, particularly under the facts of the case before us. Here, the only posthypnotic testimony at issue consists of eyewitness identifications, not an entire recitation of the facts of the crimes. The majority admits that the hypnotic sessions revealed no real new information. Kevin Coe was not even a suspect when the victims were hypnotized, so suggestiveness is absent. Furthermore, the identifications were based in large part on Coe's voice, and the majority has not yet suggested *873 that hypnosis can affect one's memory of a voice. In addition, victims Fitzpatrick and Strange faced extensive cross examination, and the defense presented its own expert who testified at length concerning the problems and pitfalls of posthypnotic testimony. I would hold that the trial court followed the Coe mandate precisely, and that the majority's overbroad interpretation of this court's holding in State v. Laureano, 101 Wn.2d 745, 682 P.2d 889 (1984) should not be allowed to override the express mandate by this court in Coe to the trial court. Even if this court insists on adhering to an absolute ban on all posthypnotic identifications, however, I would still hold that the admission of such identifications in this case was harmless error. In a recent first degree murder case, the California Supreme Court took this approach in holding that the admission of posthypnotic testimony was harmless error in light of the strong evidence against the defendant.[36] Strong evidence also corroborates the victims' identifications of Coe as their assailant in this case, and is sufficient even without those identifications to establish Coe's guilt beyond a reasonable doubt. Let us look carefully at that evidence. As pointed out above, the majority does not challenge Ms. Harmia's identification of her assailant. Nor does it mention that the Harmia, Strange and Fitzpatrick rapes are strikingly similar in numerous respects. Both Ms. Harmia and Ms. Fitzpatrick were raped shortly after stepping off a bus, and both Ms. Strange and Mrs. Fitzpatrick were raped early in the morning. (Coe's former girl friend testified that she was often awakened by Coe entering their house early in the morning after what he said was jogging.) In each case, the rapist, dressed as a jogger, grabbed his victim from behind and attempted (and succeeded in the Harmia and Fitzpatrick rapes) to shove a gloved or mitted *874 hand down her throat. Ms. Fitzpatrick and Ms. Harmia suffered similar extensive throat injuries. Ms. Strange and Ms. Harmia said their attacker wore gloves; Ms. Fitzpatrick said hers wore oven mitts. Ms. Fitzpatrick said her nose bled during the attack, and she was sure that blood must have gotten on the mitts. At trial, Coe's former girl friend testified that on the Friday before his arrest, she found him washing oven mitts at 6:30 a.m. A policeman testified that the mitts seized from Coe's home were damp. The former girl friend also testified that Coe wore black leather gloves when he went jogging, summer or winter. Coe denied wearing or even owning gloves as an adult. His neighbor, his car repairman, and the repairman's wife all corroborated the former girl friend's testimony that Coe often wore gloves. In all three rapes of which Coe was convicted at the second trial, the victims were told not to scream, not to look at the rapist's face, and to take their clothes off. They were also told that the rapist had a knife, though none of the three ever saw one. In each instance, the rapist helped his victim disrobe, rubbed her genital area and used the same vulgarity to describe it, penetrated her with his finger, masturbated, used the same vulgarity in asking her when she last had sex, and asked if she enjoyed sex and how often she had sex. Sperm taken from Ms. Harmia and Ms. Fitzpatrick (Strange was not subjected to ordinary intercourse) showed that the rapist had blood type A. Ms. Harmia, her husband and Ms. Fitzpatrick all have O positive blood; Coe has type A blood. Kevin Coe became a suspect not as a result of the victims' statements or the hypnotic sessions, but because three school custodians reported seeing a car parked by Hart Field in Spokane at 5:30 a.m. on February 5, 1981. Ms. Strange was attacked while jogging on the Hart Field track at about 6:30 that morning. The custodians testified that they saw the car parked in a no-parking zone and that its hood was warm when they touched it. The car was a fairly new 4-door silver and gray Chevrolet Citation with a sun roof. The custodians saw books, binders and a coat inside *875 the car. The license plate had a yellow covering but was a Washington plate. The car was gone by 7:30 to 8:00 that morning. The three reported the car to the vice-principal after hearing about Ms. Strange's rape. The vice-principal, in turn, called the police. Ms. Strange had already told the police about a small two-toned car that she saw park briefly by Hart Field 2 days before her rape. The police proceeded to compile a list of everyone in the Spokane area who had bought a 1980 4-door Chevrolet Citation with a factory installed sun roof. They narrowed down the list of owners and found Gordon Coe's name. They knew the rapist was a younger man, and discovered that Gordon Coe had a son, Kevin Coe, who lived in the South Hill area. The police then showed a group of photos to Ms. Strange, who picked out Kevin Coe as her assailant. Kevin Coe became a suspect on February 25, whereupon the police began following him. He was seen driving his father's Citation on February 26. The police took photos of the car showing that it contained real estate books and binders. The car also had yellow plastic over the license plate. Kevin Coe's best friend later testified that Coe liked to put gold transparent plastic over his license plate to make it more distinctive. The police followed Kevin Coe for approximately 15 days before arresting him on March 10, 1981. During those 15 days, they observed him following buses and bus routes, watching women jog around Hart Field, and driving by parks and other jogging areas with his attention on those areas rather than on the road. They saw him dressed in jogging clothing but never saw him jog. Ms. Strange said that her attacker wore a red quilted ski jacket; the police saw Coe wearing a red quilted ski jacket on more than one occasion. After Kevin Coe was arrested, a friend visited him in jail. Coe told the friend he had put a sweater in a home listed for sale and insisted that the friend find it and give it to the Goodwill organization. The friend tried twice but could not get the sweater because the home was occupied. Coe's former girl friend testified that Coe always took a change of *876 clothing when he left to go jogging, even though Coe testified that he showered at home after he jogged. While in jail, Coe also told his friend to verify his story that the two men (Coe and his friend) were looking for the South Hill rapist, Coe's purpose being to justify his actions in following buses and bus lines. The friend testified that in fact they had never looked for the rapist. Finally, following his first conviction, Coe confessed to the rape of Ms. Fitzpatrick. His confession was made to a psychiatrist who had told him three times beforehand that their conversation was not privileged. Coe now claims the confession was false and part of his strategy to avoid prison. The psychiatrist testified that when Coe confessed, he at no time mentioned a fear of prison. Coe also told the psychiatrist that 90 percent of what he said to him was the truth. Thus, even if this court insists on interpreting Laureano as imposing a ban on all posthypnotic identifications, despite the absence of such a requirement in the Coe and State v. Martin, 101 Wn.2d 713, 684 P.2d 651 (1984) decisions, then it should at least hold the Strange and Fitzpatrick identifications to be harmless error.[37] To bring this already lengthy dissent to a close, let me just add this in conclusion. I have read the entire trial record and in my judgment Kevin Coe was fairly tried and his three convictions of first degree rape should be affirmed. I fully agree with the learned trial judge who allowed the victims to testify that Kevin Coe was the one who had raped them. In the final analysis, perhaps the most eloquent justification for admitting such testimony into evidence at the trial was that provided by one of the victims. Asked how she could possibly identify Kevin Coe as her rapist, when she had seen him only once and while being sexually assaulted, she replied: *877 I became angry and I just felt to myself, I was going to memorize his face as close as I could because I didn't want this to have to happen to anyone else. Another of the victims expressed it this way: What you know, you know! The victims were entitled to identify their rapist in open court and the jury was entitled to hear and weigh that testimony, then determine how much credence should be given to it. In our legal system, the jury is the body to weigh such evidence, not this court. And the jury in this case proved itself most perceptive in this regard. It was not stampeded by emotions; it declined to return a "guilty" verdict on the fourth rape charge against Coe, the one on which the available evidence against him was weakest. I emphatically disagree with the majority's analysis of prior Washington law as to the effect of hypnosis on a witness. That law does not — or at least until today it did not — establish an absolute bar to all posthypnotic identifications. It is particularly ironic for the majority to so rule in this case where, after more than a week of hearings, the trial court found as a fact that the victims' identification testimony consisted solely of prehypnotic memory. There is simply no way that the hypnosis of the rape victims could have been suggestive as to the identity of the rapist, since the defendant was not even a suspect at the time of the hypnosis. And finally, even aside from the victims' identifications of their attacker, the evidence of the defendant's guilt is so great that by itself it establishes his guilt of these crimes beyond a reasonable doubt. For these reasons, I would affirm the convictions of Kevin Coe on all three counts of rape in the first degree. CALLOW and DURHAM, JJ., concur with ANDERSEN, J. Reconsideration denied April 20, 1988. NOTES [1] State v. Coe, 101 Wn.2d 772, 786, 684 P.2d 668 (1984). [2] RAP 12.2; Monroe v. Winn, 19 Wn.2d 462, 465, 142 P.2d 1022 (1943). [3] Majority opinion, at 838. [4] See Washington State Bar Ass'n, Washington Appellate Practice Handbook vol. 1, § 24.3(f) (1980 & Supp. 1984). [5] Coe, at 785. [6] Coe, at 786. [7] Coe, at 786. [8] Martin, at 722. [9] Martin, at 722-23. [10] Martin, at 717. [11] Martin, at 723. [12] Coe, at 786. [13] Coe, at 786. [14] Coe, at 786. [15] Coe, at 786. [16] Order Denying Defendant's Motion To Suppress Identification of the Hypnotized Witnesses, at 2. [17] Order, at 4. [18] State v. Coe, 101 Wn.2d 772, 781, 684 P.2d 668 (1984). [19] Coe, at 782. [20] Coe, at 782. [21] State v. Laureano, 101 Wn.2d 745, 753, 682 P.2d 889 (1984). [22] Laureano, at 770-71 (Dolliver, J., dissenting). [23] State v. Yapp, 45 Wn. App. 601, 602, 726 P.2d 1003 (1986). [24] Yapp, at 603. [25] Yapp, at 603. [26] Yapp, at 603-04. [27] Yapp, at 604-05. [28] Yapp, at 605. [29] Rock v. Arkansas, ___ U.S. ___, ___ L.Ed.2d ___, 107 S.Ct. 2704, 2707-08 (1987). [30] Rock, 107 S.Ct. at 2711-12. [31] Rock, 107 S.Ct. at 2713. [32] Rock, 107 S.Ct. at 2713. [33] Rock, 107 S.Ct. at 2714. [34] Rock, 107 S.Ct. at 2714. [35] Rock, 107 S.Ct. at 2714. [36] See People v. Brown, 40 Cal.3d 512, 726 P.2d 516, 230 Cal. Rptr. 834 (1985). [37] See People v. Brown, 40 Cal.3d 512, 726 P.2d 516, 230 Cal. Rptr. 834 (1985).