concurring:
I concur in the result despite profound misgivings about the procedures employed by the police in administering the hypnosis. The dangers peculiar to reviving recollection by hypnosis, and introducing such evidence at trial, have been treated at some length by the majority. It suffices to say, in summary, that hypnosis poses two distinct problems for the fact-finding process. First, the witness’ actual recollection can be altered by suggestive aspects of the hypnotic session, or by “confabulation,” the filling in of details with manufactured, though believable, recollections. Simply going through the hypnotic session may have the effect of artificially heightening the witness’ subjective belief in his own recollection. Second, the jury may be misled by the “scientific” nature of hypnosis to give undue credence to the verity of the witness’ testimony.
Although hypnosis in certain circumstances functions as a useful investigative tool, its effects are unreliable and occasionally damaging to the fact-finding process. Many jurisdictions have rejected outright the admissibility of testimony based on, hypnotic enhancement.2 For example, the Court of Appeals of Maryland concluded that the effects of hypnosis are so unreliable that it barred introduction at trial of recollection originated during and after the hypnotic session (allowing only testimony which originated prior to the hypnosis or testimony on subjects not addressed in the hypnotic session). See State of Maryland v. Collins, 296 Md. 670, 464 A.2d 1028 (1983). The decision came too late for Harker’s criminal trial in Maryland state court.
On appeal from denial of a habeas corpus petition, however, we may only grant relief if the petitioner is being held in violation of a federal statutory or constitutional right. 28 U.S.C. § 2254(a); Cupp v. Naughton, 414 U.S. 141, 146, 94 S.Ct. 396, 400, 38 L.Ed.2d 368 (1973). At least two federal circuits have identified potential constitutional objections to introduction of testimony that is the product of hypnosis. The Fifth Circuit, in United States v. Valdez, 722 F.2d 1196 (5th Cir.1984), emphasized two areas of concern. First, the “hardening” effect of hypnosis, the tendency for a subject to become artificially convinced of the verity of a recollection, may render the subject so confident in testifying that the defendant’s Sixth Amendment right to confront and cross-examine witnesses is effectively nullified. The Valdez court noted that a witness, after hypnosis,
*446may have an unshakeable subjective conviction that gives his account on the witness stand the imprimatur of absolute confidence. Indeed, in a criminal trial, the witness’ resultant undue confidence might violate the defendant’s constitutional right to confront and cross-examine witnesses, for an absolute conviction in the accuracy of his memory might make it ‘impossible to cross-examine [the] witness.’
722 F.2d at 1201, quoting State of Minnesota v. Mack, 292 N.W.2d 764, 769 (Minn.1980).
The court went on, however, to focus on due process objections; circumstances surrounding the identification of the defendant through hypnosis may be so suggestive that the trial and resulting conviction is fundamentally unfair.3 In Valdez, the witness hypnotized was a state police officer who had staked out two locations where money was dropped to pay off a blackmailer. The officer was hypnotized to enhance his recall of individuals who passed near the drop locations; the hypnosis occurred, however, after the investigation had narrowed to the defendant, and the officer had interviewed the defendant on at least two prior occasions. The Valdez court reversed the conviction, holding that uncorroborated personal identification, made only after hypnosis, of a person clearly singled out for suspicion must be excluded. The process of identification was so suggestive under the totality of circumstances that the defendant’s due process rights were violated. 722 F.2d at 1203.
The First Circuit has also recognized constitutional difficulties with use of post-hypnotic recollection in a criminal trial. See Clay v. Vose, 771 F.2d 1 (1st Cir.1985), cert. denied, — U.S.-, 106 S.Ct. 1212, 89 L.Ed.2d 324 (1985). Although the court addressed the danger to Sixth Amendment rights present in the use of testimony given after hypnosis, the court upheld the conviction of the defendant because (a) the witness had identified the defendant prior to hypnosis, and the hypnosis increased his certainty by only a small margin; (b) the defendant had been identified by another eye-witness; and (c) the judge had issued extensive cautionary instructions regarding the use and limitations of hypnosis. 771 F.2d at 4.
Given the serious scientific debate over the effects of hypnosis on memory, and the potential impact of hypnotically induced testimony on the constitutional rights of the accused, it is simply unacceptable for the government to conduct hypnotic sessions under potentially suggestive circumstances that cannot be examined or challenged at trial because of inadequate recording and other procedures. Hypnosis can have both the effect of a dye and of a catalyst: the witness’ memory can be altered or augmented through suggestion or through confabulation; once the recollection has been colored, the whole memory can be “hardened,” the experiences, both real and imagined, fixed more securely in the mind by the experience of hypnosis itself. If hypnosis is to be employed on a potential witness, at the least, appropriate measures should be taken to ensure that the session has been conducted under circumstances which minimize the possibility of suggestion, and that the defense has access to a complete record of the session for purposes of safeguarding defendant’s constitutional rights at trial.
Undoubtedly, hypnosis is a valuable tool of investigation, and it would be altogether too onerous to deprive police departments of the ability to enhance the recollection of witnesses. If the state wishes to introduce testimony about recollection, the subject of which was addressed in a hypnosis session, it should bear the burden of proof to show that the hypnosis was conducted and state*447ments elicited under tightly controlled, neutral conditions.
The leading case in structuring guidelines for police hypnosis of witnesses is State of New Jersey v. Hurd, 86 N.J. 525, 432 A.2d 86 (1981). The Supreme Court of New Jersey required that all hypnosis sessions be conducted under the following conditions: (1) a psychiatrist or psychologist trained in the use of hypnosis must conduct the session, (2) the professional conducting the hypnosis must be independent of the prosecution, the investigation and the defense, (3) all information regarding the subject and the witness given to the hypnotist prior to the session must be recorded, (4) the hypnotist should begin the session by obtaining a complete account of the pre-hypnosis memory from the witness, (5) all contacts between the hypnotist and subject must be recorded and videotaping is strongly recommended, and (6) only the hypnotist and the witness may interact just prior to, during and just after the hypnosis. Hurd, 86 N.J. at 545-47, 432 A.2d at 96-97. The Hurd standards provide ample room for the police to employ hypnosis during an investigation. On the other hand, the accused is assured of a hypnosis session that is fair and neutral, and the complete record allows both the accused and the trial court to assess the extent to which the witness has been influenced, if at all, by the hypnosis itself. See Sprynczynatyk v. General Motors Corp., 771 F.2d 1112, 1122-23 (8th Cir.1985).
The present case is troubling because of the faulty police procedures during the hypnotic session. The police attempted to videotape and audiotape the proceedings, though the videotape failed because the lens cap was not removed, and the audiotape is of poor quality and has some gaps. A police lieutenant of the Montgomery County Police Department conducted the session. Although the record on appeal contains no positive evidence of suggestive identification procedures, and only some indication of “hardening” of Thompson’s recollection as a product of hypnosis,4 my objection is that the record of the hypnotic session was so flawed that a trial court simply could not make an adequate judgment about whether the circumstances of the session were in fact suggestive.
Harker does not merit reversal, however, because the record shows that the identification by Thompson was sufficiently reliable apart from the possible influence of the hypnosis. Prior to hypnosis, Thompson was able to assist in the preparation of a composite sketch which bore a strong similarity to Harker. Harker was first suspected when he was spotted driving a car which matched the description given by the victim; the car turned out to be owned by Harker. Thompson initially identified Harker in a photo array, ten days after the hypnosis session, and there is no indication that Thompson was exposed to Harker’s likeness any time prior to or during the hypnosis. Thus, while the police procedure in administering the hypnosis was flawed, the consistency of Thompson’s description before, during and after hypnosis show that the error was harmless beyond a reasonable doubt, and the petition should be denied. Chapman v. California, 386 U.S. 18, 87 S.Ct. 824, 17 L.Ed.2d 705 (1967).
The police should not be free to administer hypnosis, however, to witnesses in an uncontrolled and potentially suggestive environment and without adequately memorializing the proceedings. Use of testimony elicited under such circumstances poses too great a threat to the right of confrontation secured to the accused by the Sixth and Fourteenth Amendments, and to the right to due process secured by the Fourteenth Amendment.
. State ex rel. Collins v. Superior Court, 132 Ariz. 180, 644 P.2d 1266 (1982) (en banc); People v. Shirley, 31 Cal.3d 18, 181 Cal.Rptr. 243, 641 P.2d 775 (1982), cert. denied, 459 U.S. 860, 103 S.Ct. 133, 74 L.Ed.2d 114 (1982); People v. Quintanar, 659 P.2d 710 (Colo.App.1982); Bundy v. State, 471 So.2d 9 (Fla.1985); State v. Collins, 296 Md. 670, 464 A.2d 1028 (1983); People v. Gonzales, 415 Mich. 615, 329 N.W.2d 743 (1982); State v. Mack, 292 N.W.2d 764 (Minn.1980); State v. Palmer, 210 Neb. 206, 313 N.W.2d 648 (1981); People v. Hughes, 59 N.Y.2d 523, 466 N.Y.S.2d 255, 453 N.E.2d 484 (1983); State v. Peoples, 311 N.C. 515, 319 S.E.2d 177 (1984); Com. v. Naza-rovitch, 496 Pa. 97, 436 A.2d 170 (1981).
. The suggestive identification problem is not limited to testimony derived through hypnosis, but appears in a variety of circumstances where the government has offered a witness the opportunity to identify a suspect, under circumstances which suggest a particular individual. See, e.g., Stovall v. Denno, 388 U.S. 293, 302, 87 S.Ct. 1967, 1972, 18 L.Ed.2d 1199 (1967) (condemning, in general terms, the practice of showing suspects singly, rather than in groups).
. The victim first identified Harker in a photo array ten days after hypnosis. Thompson testified that at the photo show-up, he said "I think that looks very much like the man who shot me ... I think it’s definitely an older picture, but it looks very similar to the man and I would like to see that man in a line-up or something.” Later, the victim was taken to a courthouse where he spotted and identified Harker as the assailant out of a crowded hallway. At trial, Thompson identified Harker positively, saying: "Believing that on the final day God will be my judge and all the saints my jury, there stands the man who tried to murder me.”