OPINION
The state charged defendant with thirteen counts of criminal sexual penetration of a minor under the age of thirteen and three counts of criminal sexual penetration of a minor over thirteen but under sixteen years of age. Defendant moved to suppress the testimony of the complaining witness (the child) and her therapist on the ground that the child’s memory had been hypnotically enhanced without compliance with the safeguards of State v. Beachum, 97 N.M. 682, 643 P.2d 246 (Ct.App.1981). The district court granted the motion. We reverse because full compliance with the specific safeguards of Beachum was not required in the circumstances of this case. We remand for a determination of the reliability of the testimony — that is, whether the use of hypnosis here was reasonably likely to result in recall comparable in accuracy to normal human memory.
FACTS
There appears to be no dispute concerning the facts necessary for our disposition of the case. The child was referred to Mary Wales-North, a psychotherapist, because she suffered from nightmares, inappropriate anger, depression, and other similarly unexplained symptoms. When Ms. Wales-North took a case history from the child and members of her family, sexual abuse was not mentioned. Nor were there any allegations of sexual abuse during the first several sessions that Ms. Wales-North conducted with the child, her mother, and stepfather. The child’s emotional problems were thought to be post-traumatic stress syndrome caused by a recent automobile accident.
*540The child alleged the sexual abuse by defendant at a therapy session about six weeks after the initial session. At the time of the disclosure Ms. Wales-North was using “relaxation therapy” to make the child comfortable enough to discuss what was troubling her. After the first report of sexual abuse, the child discussed the matter in subsequent sessions without any relaxation exercises.
The child testified that the relaxation therapy did not bring back any memory she had lost and she had never forgotten the molestations. Yet prior to her allegations during therapy she had not told anyone that she had been the victim of sexual abuse or that defendant had done anything bad to her. She said that she finally told Ms. Wales-North because she felt safe and trusted her.
The district court heard testimony concerning the nature of relaxation therapy from Ms. Wales-North and Dr. Michael Dudelczyk, a psychiatrist under whom Ms. Wales-North was conducting an internship. Ms. Wales-North described the technique as telling the child to take a few deep breaths, to begin to notice the sounds outside, and to think about what she was feeling. Dr. Dudelczyk identified the technique as “Ericksonian hypnosis,” but stated that practitioners of classical hypnosis would not label Ms. Wales-North’s procedure as hypnosis. Ms. Wales-North testified that there is a possibility of suggestibility with the technique, although Dr. Dudelczyk seemed to minimize that risk.
THE BEACHUM SAFEGUARDS
In Beachum we set forth guidelines for the admissibility of testimony by a witness whose memory has been stimulated by hypnosis. The case involved the use of classical hypnosis for forensic purposes. The victim of a rape and robbery had identified the defendant’s voice at a line-up but was unable to make a positive visual identification. Several weeks later the prosecutor suggested that she undergo hypnosis to help her improve her identification of the assailant. She was hypnotized by the chief of the local police department in a session held at the police station. The prosecutor, a police sergeant, and three others attended. The session was tape recorded.
Following the leading case of State v. Hurd, 86 N.J. 525, 432 A.2d 86 (1981), we affirmed the district court’s decision to exclude the hypnotically enhanced testimony. We quoted Hurd’s description of the problems with such testimony:
First, a person undergoing hypnosis is extremely vulnerable to suggestions. * *
A second aspect of hypnosis that contributes to its unreliability is the loss of critical judgment. A person under hypnosis is more willing to speculate and will respond to questions with [a] confidence he would not have as a waking person. * * * The third * * * phenomenon is the tendency to confound memories evoked under hypnosis with prior recall. [Citations omitted].
State v. Beachum, 97 N.M. at 687, 643 P.2d at 251 (quoting State v. Hurd, 86 N.J. at 539-40, 432 A.2d at 93). See generally American Medical Association Council Report, Scientific Status of Refreshing Recollection by Use of Hypnosis, J. A.M.A. 253(13), 1918-23 (1985).
These concerns have led some jurisdictions to ban the use of hypnotically refreshed testimony altogether. See Rock v. Arkansas, 483 U.S. 44, 57, 107 S.Ct. 2704, 2712, 97 L.Ed.2d 37 (1987). We rejected that view, stating:
A rule of per se inadmissibility, we conclude, is unnecessarily broad and may result in the exclusion of evidence that may be valuable and accurate. The better rule is that testimony of pre-hypnotic recollections is admissible in the sound discretion of the trial court, but posthypnotic recollections, revived by the hypnosis procedure, are only admissible in a trial where a proper foundation has also first established the expertise of the hypnotist and that the techniques employed were correctly performed, free from bias or improper suggestibility.
State v. Beachum, 97 N.M. at 688, 643 P.2d at 252.
In furtherance of this rule we adopted the requirement set forth in Hurd *541that six safeguards be satisfied before hypnotically refreshed testimony is admissible: (1) “ ‘a psychiatrist or psychologist experienced in the use of hypnosis must conduct the session’ (2) “ ‘the professional conducting the hypnotic session should be independent of and not regularly employed by the prosecutor, investigator or defense’ (3) “ ‘any information given to the hypnotist by law enforcement personnel or the defense prior to the hypnotic session must be recorded’ (4) “ ‘before inducing hypnosis the hypnotist should obtain from the subject a detailed description of the facts as the subject remembers them’ (5) “ ‘all contacts between the hypnotist and the subject must be recorded’ and (6) “ ‘only the hypnotist and the subject should be present during any phase of the hypnotic session, including the pre-hypnotic testing and the post-hypnotic interview.’ ” State v. Beachum, 97 N.M. at 689-90, 643 P.2d at 253-54 (quoting State v. Hurd, 86 N.J. at 545-46, 432 A.2d at 96-97 (emphasis deleted)).
We reaffirm our holding in Beachum. The present case, however, differs from Beachum in two significant respects. First, Beachum involved the use of classical hypnosis; here Ericksonian hypnosis was used. There is evidence in the record that Ericksonian hypnosis does not create the same reliability problems as classical hypnosis. Nevertheless, we believe that the record in this case is inadequate for us to state definitively that the Beachum safeguards are unnecessary when Ericksonian hypnosis is used. We therefore do not rest our decision on that ground. We leave that issue to a future case.
The second distinction is that here the subject matter of the disclosure by the hypnotized witness was totally unanticipated by the hypnotist. This was not a case of hypnosis conducted for forensic purposes. It is undisputed that the purpose of the hypnosis of the victim was therapy and no one present at the session, with the possible exception of the victim herself, had any reason to believe that the victim would disclose any allegations of sexual abuse. Ms. Wales-North had not been working with law enforcement agencies in any way. The surprise was not a matter of detail — such as the identity of the perpetrator of a crime — but of the very existence of an offense. In Beachum and Hurd, on the other hand, the purpose of hypnosis was specifically to enhance the memory of a witness to an offense in order to obtain testimony for use at a trial.
It is apparent from the language in Beachum and Hurd that the opinions did not have in mind the situation in this case. For example, in agreeing with the proposition that hypnotically refreshed testimony should on occasion be admissible, we wrote, “To hold otherwise presents the State with the dilemma of choosing to use a particular witness at trial or to use hypnosis on the witness as an investigatory tool.” State v. Beachum, 97 N.M. at 688, 643 P.2d at 252. The Hurd safeguards likewise were expressed in terms that on their face apply only to the use by the prosecution or defense of hypnosis to enhance the testimony of a previously identified witness. For example, the second safeguard stated, “[T]he professional conducting the hypnotic session should be independent of and not regularly employed by the prosecutor, investigator or defense.” State v. Beachum, 97 N.M. at 689, 643 P.2d at 253 (quoting State v. Hurd, 86 N.J. at 545, 432 A.2d at 96 (emphasis added) (footnote omitted)). Similarly, the third safeguard stated, “[A]ny information given to the hypnotist by law enforcement personnel or the defense pri- or to the hypnotic session must be recorded, either in writing or another suitable form.” Id. at 689, 643 P.2d at 253 (emphasis added).
This is not to say that the analysis in Beachum and Hurd has no application to the situation here. In particular, we are not saying that “therapeutic” hypnosis— hypnosis for treatment, undertaken when no litigation is anticipated — presents no reliability problems. It is, however, appropriate to examine the Beachum safeguards to see if they should be required in a context that is so different from the one in that case.
Our examination of Beachum and Hurd indicates that the six safeguards consti*542tute, in essence, a prophylactic rule. The reasoning of the opinions would not support a contention that hypnotically refreshed testimony is always unreliable when one or more of the safeguards has not been observed. The Hurd opinion was explicit in that regard. The court wrote, “If our only concern was the failure to observe these procedures, we might not require suppression of the testimony, because the State could not have anticipated the details of the required procedures.” Id. at 549, 432 A.2d at 98. The reason the court suppressed the testimony in that case was the suggestiveness of the conduct of the hypnotist and of a detective present at the session. What the safeguards accomplish is to enhance the probability of reliability and to produce a record so that the opposing party can effectively pursue any reliability problems with the testimony. These virtues of the safeguards make it appropriate to require compliance with them when a party to litigation intends to use hypnosis to enhance the memory of a previously identified witness. In that context the only cost of the safeguards is a financial one and that cost is a small price to pay “[t]o provide an adequate record for evaluating the reliability of the hypnotic procedure, and to ensure a minimum level of reliability[.]’’ Id. at 545, 432 A.2d at 96.
As a prophylactic rule, however, the Beachum safeguards make no sense in the context of this case, when it is totally unanticipated that the hypnosis session will produce a disclosure relevant to litigation. For example, the fifth safeguard requires that all contacts between the hypnotist and the subject be recorded, preferably by videotape. Unless practitioners of hypnosis are to adopt a routine practice of recording all hypnosis sessions, requiring this safeguard in this context would be tantamount to always excluding testimony by witnesses hypnotized solely for therapeutic purposes. Similarly, the fourth safeguard requires the hypnotist to “obtain from the subject a detailed description of the facts as the subject remembers them.” State v. Beachum, 97 N.M. at 689, 643 P.2d at 253. When the subject matter of the disclosure by the subject was unanticipated, how could the hypnotist have known to make such an inquiry?
We would require compliance with the Beachum safeguards in all circumstances, including this case, if we were convinced that in every instance in which a witness’s recollection was stimulated by a hypnosis session not in compliance with the safeguards, either the refreshed testimony would be too unreliable to be admissible or it would be impossible to evaluate adequately the reliability. The Hurd court itself was not so convinced, as shown by its basing its suppression of the victim’s testimony in that case not simply on violation of the safeguards but on the suggestiveness of the specific hypnosis session at issue. We also are not so convinced. Rather, we have confidence that the district courts, giving due weight to any failure to conform to the safeguards, can make proper determinations of whether to admit testimony on matters unexpectedly elicited in a therapeutic hypnosis session. In that context the district court’s task is solely to determine whether the state has shown— by clear and convincing evidence — that “the use of hypnosis in the particular case was reasonably likely to result in recall comparable in accuracy to normal human memory.” State v. Hurd, 86 N.J. at 543, 432 A.2d at 95; see State v. Beachum, 97 N.M. at 690, 643 P.2d at 254 (admissibility to be established by clear and convincing evidence).
In this regard we note that the hypnosis session here failed to comply with Beachum’s first safeguard (that the hypnotist be a psychiatrist or psychologist) and perhaps the sixth safeguard (that the only persons present during the session should be the hypnotist and the subject). The district court on remand must give due weight to these features of the hypnosis session. Nevertheless, given that the disclosure by the victim at the hypnosis session was a surprise to all the other persons present, we believe that the district court could properly determine that the possibility of improper suggestion at the session was sufficiently small that the victim’s testimony should be admitted at trial. Indeed, the district court could find that the vie*543tim’s recollection predated the session and was not refreshed by hypnosis.
Because the district court’s ruling in this case was predicated on violation of the presumed absolute requirement of compliance with the six Beachum safeguards, the district court did not have occasion to apply the standard we have expressed in this opinion for admissibility of the child’s testimony. We remand for application of that standard.
CONCLUSION
We reverse the district court’s order suppressing the testimony of the child and of Ms. Wales-North and remand for further proceedings consistent with this opinion.
IT IS SO ORDERED.
ALARID, C.J., and APODACA, J„ concur.