Windmere, Inc. v. International Ins. Co.

105 N.J. 373 (1987)

522 A.2d 405

WINDMERE, INC., A CORPORATION OF THE STATE OF NEW JERSEY, PLAINTIFF-APPELLANT, v. INTERNATIONAL INSURANCE COMPANY, DEFENDANT-RESPONDENT.

The Supreme Court of New Jersey.

Argued November 3, 1986.

Decided March 19, 1987.

*375 Terence P. Corcoran argued the cause for appellant (Corcoran and Higgins, attorneys).

Edmund E. Lynch argued the cause for respondent (Lynch and Lynch, attorneys).

The opinion of the Court was delivered by O'HERN, J.

Once again we are presented with the question of the admissibility of voiceprints into evidence as reliable scientific tools for determining the identity of a human voice. In this case we have the benefit of hindsight to confirm that adherence to our rules for determining the reliability of scientific evidence precludes the admissibility of voiceprints on this record. In the unusual circumstances of this case, however, we affirm the judgment of the Appellate Division, 208 N.J. Super. 697 (1986), sustaining a jury verdict that denied plaintiff a fire loss recovery from the insurer. The verdict followed a jury trial in which voiceprints were used to assist the jury in comparing the voice *376 of plaintiff's maintenance man with the voice recorded by police when the fire was reported anonymously by telephone. But because the record was cumulative of other reliable evidence sufficient to sustain the verdict, and the verdict does not appear to have been a miscarriage of justice, we affirm the judgment of the Appellate Division.

I.

Few subjects have generated more literature than the debate about the scientific reliability of the voiceprint.[1] In New Jersey, the debate has remained unsettled since the trilogy of decisions emanating from State v. Cary 49 N.J. 343 (1967): State v. Andretta, 61 N.J. 544 (1972), State v. Perez, 150 N.J. Super. 166 (App.Div. 1977), and then, most recently, D'Arc v. D'Arc, 157 N.J. Super. 553 (Ch.Div. 1978).

The theory behind identification by the voice spectrography method is that people have different vocal cavities and articulators. Thus, one expert's theory of voice uniqueness was based on the way "the articulators interacted with one another and the vocal cavities. The articulators include the lips, teeth, tongue, *377 soft palate, and jaw muscles. The vocal cavities are the throat, nose, and the two oral cavities formed in the mouth by the positioning of the tongue." * * * Since every person has these unique anatomical characteristics, some experts feel that the sound produced through the interaction of these features should lead to a means of identifying people by their voices alone. [Cornett v. State, 450 N.E.2d 498, 500 (Ind. 1983) (citation omitted).]

In brief, the scientific procedure was described by Judge Imbriani in D'Arc v. D'Arc, supra, 157 N.J. Super. at 555-56.

A sound spectrograph is an instrument which produces a spectrogram that can be utilized by a trained examiner to purportedly identify the speaker of an unknown voice. The system * * * operates by an electrical process which produces on a sheet of paper a pictorial representation or graphic display of sounds, which is called a sound spectrogram or voiceprint. * * * [The graph represents a tri-dimensional plane with time shown on the horizontal axis, frequency shown on the vertical axis, and the relative amplitude of the voice's intensity displayed by the darkness of the lines. Cornett v. State, supra, 450 N.E.2d at 500.] Identification of an unknown speaker is made by an examiner who compares the unknown voice with the known voice exemplars of one or more persons. * * * The device does not operate automatically; an individual must operate the machine, adjust numerous controls to obtain readings and subjectively interpret the spectrograms to arrive at an opinion. (citations omitted).

The admissibility in evidence of the results of voiceprint analysis depends upon its scientific reliability. In a recent series of cases, this Court has restated the principles that it employs in determining whether scientific evidence is admissible to prove an underlying fact in dispute. See State v. Kelly, 97 N.J. 178 (1984) (general acceptance within the professional community can demonstrate scientific reliability of `battered woman's syndrome'); Romano v. Kimmelman, 96 N.J. 66 (1984) (general acceptance within scientific community demonstrates scientific reliability of breathalyzer test); State v. Cavallo, 88 N.J. 508 (1982) (diagnosis of tendency to commit rape on basis of rapist profile is not scientifically reliable when procedure is not shown to be accepted generally by a scientific community); State v. Hurd, 86 N.J. 525 (1981) (hypnotically refreshed testimony may be admissible if it can be demonstrated that use of hypnosis in a given case was a reasonably reliable means of restoring memory); see also State v. Matulewicz, 101 N.J. 27 (1985) (laboratory report of controlled dangerous *378 substances may be admissible without further proof if test is shown to be scientifically reliable).

While each of these cases dealt with the use of expert scientific evidence or a scientific device in the context of a criminal trial, the principles are generally applicable to civil proceedings and will assist us here. (In the criminal context, conditions of admissibility must be "clearly established." State v. Johnson, 42 N.J. 146, 171 (1964)). Moreover, even though these principles bear a close resemblance to the familiar Frye test,[2] there are differences. There is no need to review in detail each decision. We may summarize their principles by this passage from Romano v. Kimmelman:

In New Jersey, the results of scientific tests are admissible at a criminal trial only when they are shown to have "sufficient scientific basis to produce uniform and reasonably reliable results and will contribute materially to the ascertainment of the truth." State v. Hurd, 86 N.J. 525, 536 (1981) (quoting State v. Cary, 49 N.J. 343, 352 (1967)). Scientific acceptability need not be predicated upon a unanimous belief or universal agreement in the total or absolute infallibility of the techniques, methodology or procedures that underlie the scientific evidence. * * * Reliability of such evidence must be demonstrated by showing that the scientific technique has gained general acceptance within the scientific community. State v. Johnson, 42 N.J. 146, 170-71 (1964). The fact that a possibility of error exists does not preclude a conclusion that a scientific device is reliable. This Court in Johnson noted: "Practically every new scientific discovery has its detractors and unbelievers, but neither unanimity of opinion nor universal infallibility is required for judicial acceptance of generally recognized matters." Id. at 171. Once the showing of general acceptability has been made, courts will take judicial notice of the given instrument's reliability and will admit in evidence the results of tests from the instrument without requiring further proof. Id. [96 N.J. at 80 (citations omitted).]

*379 Obviously, it is the last feature the judicial validation of a scientific device for all future use that is the critical end product of the inquiry into reliability. Thus, for example, the results of the breathalyzer test may be admissible, subject to evidence of maintenance of the device and the other adjustments noted in Romano, supra, 96 N.J. at 72-73, in all future cases. Judicial notice shall be taken of the reliability of the instrument.

There are generally three ways in which a proponent of expert testimony or scientific results can prove the required reliability in terms of its general acceptance within the professional community: (1) the testimony of knowledgeable experts; (2) authoritative scientific literature; and (3) persuasive judicial decisions which acknowledge such general acceptance of expert testimony. State v. Cavallo, supra, 88 N.J. at 521. Each of these criteria for establishing the reliability of a scientific device has been seriously challenged or impugned with respect to voiceprints.

(1) Testimony of Knowledgeable Experts Did Not Establish Its General Acceptance Within The Professional Community.

General acceptance within a scientific community can be an elusive premise to prove. There will always be some detractors to any scientific theory. State v. Kelly, supra, 97 N.J. 178, illustrates the kind of inquiry that must be made concerning the definition of the community and the degree of acceptance within that scientific community. There, the majority and concurrence disagreed as to the evaluation of the relative degree of consensus within the defined psychological community concerning the reliability of evidence presented about the syndrome of characteristic responses to repeated physical and mental abuse. See id. at 210-11, 224. The majority, despite the expert's testimony and the Court's reference to "at least five books and almost seventy scientific articles and papers *380 about battered-woman's syndrome," refrained from ruling that New Jersey's standard of acceptability for scientific evidence had been satisfied. Id. at 211 (emphasis added).

An initial concern about the science of voiceprint analysis is that it appears to be a sole source industry, with the principal place of research located at the Michigan State University Speech and Hearing Research Laboratory. In this case, as in repeated other cases, Hodo v. Superior Court, Riverside County, 30 Cal. App. 3d 778, 106 Cal. Rptr. 547 (1973), Reed v. State, 283 Md. 374, 391 A.2d 364 (1978), Commonwealth v. Lykus, 367 Mass. 191, 327 N.E.2d 671 (1975), D'Arc v. D'Arc, supra, 157 N.J. Super. 553, and State v. Cary, 99 N.J. Super. 323 (Law Div. 1968), the only witnesses who testified were experts affiliated with the development of the device at the Michigan State University and Professor Louis J. Gerstman, a critic of the reliability of the voiceprint who frequently testifies for those opposing its use.

In People v. Kelly, the California Supreme Court expressed its reservations about this feature of the evidentiary record before it. Speaking of the witness, that court said:

[His] background thus discloses that he is one of the leading proponents of voiceprint analysis; he has virtually built his career on the reliability of the technique. This situation is closely akin to that in People v. King, * * * 266 Cal. App. 2d 437, 72 Cal. Rptr. 478 [1968], in which Kersta, a pioneer in the field, was the chief prosecution witness supporting the admissibility of voiceprint evidence. The court in King rejected Kersta's testimony regarding the scientific basis of voiceprint analysis and warned that "[b]efore a technique or process is generally accepted in the scientific community, self-serving opinions should not be received which invade the province of the trier of fact." * * * Likewise, [the witness], a strong advocate of the voiceprint technique, may be too closely identified with the endorsement of voiceprint analysis to assess fairly and impartially the nature and extent of any opposing scientific views. A more detached and neutral observer might more fairly do so. In the absence of additional and impartial evidence regarding general acceptance, the trial court was in a similar position to that presented in King, in which "a court could only receive Kersta's opinion on faith." [17 Cal. 3d 24, 38, 549 P.2d 1240, 1249, 130 Cal. Rptr. 144, 153 (1976) (citations omitted).]

*381 In this case two witnesses testified about the reliability of the device at the Rule 8 hearing,[3] Lieutenant Lonnie Smrkovski and Ernst F.V. Alexanderson. Smrkovski was commanding officer of the Voice Identification Unit of the Michigan Department of State Police and had worked with Dr. Oscar Tosi at Michigan State. Alexanderson was the operator of Voice Identification, Inc., which made the voiceprints in this case. From 1969 to 1973, Alexanderson was employed by Voiceprint Laboratories, Somerville, New Jersey. He worked as an "apprentice" under Laurence G. Kersta, who was president of the company and a leading proponent of voiceprint identification, learning about the process and training law enforcement officers. In 1973, Alexanderson formed his own company, Voiceprint Identification, Inc. Despite his experience in the field, Alexanderson had little formal training in the area of voiceprint technology or speech pathology: he had only several credit hours and had attended approximately eight seminars sponsored by either Michigan State University or the Michigan State Police. At these seminars, lectures were conducted by Dr. Tosi and Lieutenant Smrkovski. Alexanderson was certified as an expert by the International Association for Voice Identification and the International Association for Identification in 1978 and 1983 respectively. He was also on the five-member Board of the International Association for Identification, whose other members included Doctors Tosi and Kersta.

Our concern is not with the qualifications of the experts but with the limited breadth of their experience and their potential bias. Almost all the experts have been associated with Michigan State University, either directly or indirectly. The evidence presented in this case falls considerably short of the kind of record that we have found necessary to establish the admissibility of test results as in Romano v. Kimmelman, supra, 96 N.J. *382 66. There, several days of trial were held before the trial court, which entered findings with respect to reliability. The State presented seven expert witnesses who were subjected to "extensive and effective cross-examination" by defense counsel. Id. at 77 n. 1. In the appendix to the trial court's opinion, the technical backgrounds of the various experts who testified were discussed. At least four of those experts had no connection with the device or the State Police. A detailed and thorough analysis of the scientific evidence supporting uniform and reasonably reliable results from use of the breathalyzer was undertaken and filed with this Court. In this case, the trial court could consider Alexanderson an expert, but as in Kelly, supra, "could only receive [his] opinion on faith." 17 Cal. 3d at 38, 129 Cal. Rptr. at 153, 549 P.2d at 1249. Moreover, in the report Smirkowski submitted to the court, he was candid to admit that the device lacked scientific consensus and in his testimony conceded that the scientific community was a "divided camp." The trial court recognized this fact, but nevertheless believed the device "could be reliable." In contrast, in Romano, supra, the trial court concluded that there was no "residue of legal concern that the breathalyzer, given proper functioning and operated by a competent person, is `a scientifically reliable and accurate device for determining the alcoholic content of the blood.'" 96 N.J. at 82 (citations omitted). Romano, supra, therefore requires that the court find the technique to be "reasonably reliable."[4]Id. at 80. Consequently, unlike the breathalyzer tests in Romano and Johnson, it simply cannot be said that there is general acceptance within the relevant scientific community.

*383 (2) Authoritative Scientific Literature Does Not Demonstrate That There Is Any Measure Of Universal Acceptance Of The Device's Reliability.

As previously noted, there is no want of literature that deals with the reliability of voiceprint analysis. What appears from the record is that the trial court had the benefit of an exhibit prepared by Lieutenant Smrkovski. The trial court, while it reviewed many other cases concerning the admissibility of the voiceprint, did not undertake to resolve the conflict among the relevant and more recent scientific articles concerning the reliability and acceptance of the technology.

We have made no effort to resolve this battle of experts except to note the varied views. Those proposing the use of the voiceprint for identification purposes claim it is accurate; those opposing it argue that it has an excessive error rate and creates a "false air of scientism"; but even its opponents admit that it is at least a reliable method to eliminate a person as the unidentified speaker.[5] 208 N.J. Super. at 710-13.

Reservations about the acceptance of the device in scientific literature have been expressed in a number of jurisdictions. In Reed v. State, supra, 283 Md. 374, 391 A.2d 364, the court held that authoritative scientific literature is not in accord as to acceptance of the device by experts. As summarized by Justice Kaplan in his dissent in Commonwealth v. Lykas, supra, after canvassing the literature on the subject:

[O]pinion is divided on the [Michigan State voiceprint] method; the journal material shows turbulence and discord rather than that "general acceptance" which the Frye case lays down as a precondition of admissibility. Nor can it be plausibly said that those with adverse views are either unqualified to have opinions worthy of respect or are strangers to the relevant scientific "field." [367 Mass. at 206, 327 N.E.2d at 682 (footnote omitted).]

*384 As noted, the breadth of literature on this subject is extensive and is set forth in a variety of decisions. E.g., Reed v. State, supra, 283 Md. at 377 n. 3, 391 A.2d at 366 n. 3; Commonwealth v. Lykas, supra, 367 Mass. at 199 n. 3, 327 N.E.2d at 676 n. 3. Members of the Speech Communications Section of the Acoustical Society of America voted 42-0 against the reliability of the procedure because a majority of articles on this subject were negative in their characterization of the process. Reed v. State, supra, 283 Md. at 394, 391 A.2d at 374. Consequently, it is not possible for us, without purporting to scale the relative authorities, to resolve the issue. We can agree, however, that the journals are in disarray.

(3) There Is No General Judicial Acceptance of Voiceprint Analysis.

Some courts confronted with a similar battery of witnesses and divergent scientific reports have been reluctant categorically to validate the scientific reliability of the device. See, e.g., State v. Gortarez, 141 Ariz. 254, 686 P.2d 1224 (1984); People v. Kelly, supra, 17 Cal. 3d 24, 549 P.2d 1240, 130 Cal. Rptr. 144; Reed v. State, supra, 283 Md. 374, 391 A.2d 364; People v. Tobey, 401 Mich. 141, 257 N.W.2d 537 (1977); D'Arc v. D'Arc, supra, 157 N.J. Super. 553; Commonwealth v. Topa, 471 Pa. 223, 369 A.2d 1277 (1977).

On the other hand, many jurisdictions have sustained the validity of the device. See, e.g., United States v. Williams, 583 F.2d 1194 (2d Cir.1978); United States v. Jenkins, 525 F.2d 819 (6th Cir.1975); United States v. Baller, 519 F.2d 463 (4th Cir.), cert. denied, 423 U.S. 1019, 96 S. Ct. 456, 46 L. Ed. 2d 391 (1975); Alea v. State, 265 So. 2d 96 (Fla. Dist. Ct. App. 1972); State v. Williams, 388 A.2d 500 (Me. 1978); Commonwealth v. Lykus, supra, 367 Mass. 191, 327 N.E.2d 671; State ex rel. Trimble v. Hedman, 291 Minn. 442, 192 N.W.2d 432 (1971); People v. Bein, 114 Misc.2d 1021, 453 N.Y.S.2d 343 (Sup.Ct. 1982); State v. Williams, 4 Ohio St.3d 53, 446 N.E.2d 444 (1983); State v. Wheeler, 496 A.2d 1382 (R.I. 1985). We are thus left with a *385 split among the jurisdictions as to the scientific reliability of the instrument. Obviously, we cannot wait for the concept to develop universal acceptance. On the other hand, we continue to await the more conclusive evidence of scientific reliability sought by Justice Proctor in State v. Cary, supra, 49 N.J. 343.

II.

As noted, once reliability is established, further inquiry into the reliability of the device is normally foreclosed. In this way, for example, radar devices were held to be scientifically acceptable, State v. Dantonio, 18 N.J. 570 (1955), State v. Cardone, 146 N.J. Super. 23 (App.Div. 1976); the breathalyzer has been held to be scientifically acceptable, Romano v. Kimmelman, supra, 96 N.J. 66; and fingerprints were determined to be scientifically acceptable. Davis v. Mississippi, 394 U.S. 721, 89 S. Ct. 1394, 22 L. Ed. 2d 676 (1969) (cited by State v. Harris, 143 N.J. Super. 314, 318 (Law Div. 1976)). The rationale supporting admission of such evidence is that each method relies primarily upon objective factors for reaching a conclusion, with subjective factors playing only a minimal role in the analysis. Consequently, under similar review, we have rejected as insufficient the evidence that hypnotic procedures produce uniform and reasonably reliable results. State v. Hurd, supra, 86 N.J. 525. The use of this type of evidence was therefore limited to a case-by-case determination of its admissibility under New Jersey Rule of Evidence 4.[6]

We prefer not to take that approach here, although that was essentially the route taken by the trial court and the Appellate Division. The latter court recognized that "[w]hile the admissibility of spectrographic evidence in New Jersey is not entirely *386 clear, * * * in this civil case[] the trial judge did not abuse his discretion in admitting the evidence." 208 N.J. Super. at 713 (footnote omitted).

We note that our standard that a scientific technique have "sufficient scientific basis to produce uniform and reasonably reliable results and contribute materially to the ascertainment of truth," State v. Cary, supra, 49 N.J. at 352, Romano v. Kimmelman, supra, 96 N.J. at 80, State v. Hurd, supra, 86 N.J. at 536 (emphasis added), is not a rejection of the rule announced in Frye v. United States, 293 F. 1013 (D.C. Cir.1923). Rather, "general acceptance in the field to which [the technique] belongs," id. at 1014, is a critical factor in finding that there is "sufficient scientific basis to produce uniform and reasonably reliable results."

Leading courts and commentators have suggested revisions to the Frye test. In United States v. Downing, 753 F.2d 1224, 1237 (3rd Cir.1985), the Frye test was rejected, and the particular degree of scientific acceptance within the scientific community was deemed but one factor to consider when deciding whether to admit scientific evidence. In "The Admissibility of Novel Scientific Evidence: Frye v. United States, A Half-Century Later," 80 Colum.L.Rev. 1197 (1980), Professor Giannelli focuses upon three factors: "(1) the validity of the underlying principle, (2) the validity of the technique applying the principle, and (3) the proper application of the technique on a particular occasion," id. at 1201-02 (footnotes omitted), and suggests that "the admission of scientific evidence should be controlled by adjusting the burden of proof." Id. at 1247. While we recognize the concerns raised, we decline to adopt the proposed standards. Our test has been applied too often, and with satisfactory results, to depart from it now.

While in this case the voiceprint evidence should not have been admitted, its future use as a reasonably reliable scientific method may not be precluded forever if more thorough proofs as to reliability are introduced in other litigation. The science *387 of voiceprint analysis was originally developed at Bell Laboratories in New Jersey. We should not want for experts independent of Michigan State University or the Michigan State Police who could testify about the reliability of voiceprint technology. Technological advancements in computer modeling of the human voice are taking place continuously. Those devices will convert the wave lengths of energy from human speech into print. A science capable of that refinement should be able to aid a jury in its determination that voices are similar.

In addition, unlike the breathalyzer or radar tests, the instrument does not resolve the ultimate issue. Its capacity for error can always be weighed by the human experience of the jury in considering the evidentiary value of the voiceprints. Undoubtedly the human brain goes through an electronic sorting of the impulses that have struck the ear to reach the conclusion that two voices heard may be from the same person. One would think that science can aid the jury in sorting the similarities of voice. But post-trial developments in this case seriously undermine the reliability of the voiceprint for identifying a human voice and lead us not to reverse the judgment below despite the erroneous admission of the voiceprints into evidence.

III.

The reason that we leave the judgment undisturbed stems from a review of the entire record, facts, and procedure. On March 12, 1982, in the early morning hours, fire destroyed the Crosswinds Restaurant. It is without dispute that the fire was deliberately set for the purpose of destroying the premises. The evidence showed that an accelerant, gasoline, had been spread around the building prior to the fire; a five-gallon plastic container with gasoline residue was found on the premises. Moreover, the anonymous caller who reported the fire, and whose voice was recorded, also stated, "There's a bomb in the building * * * warn the fireman." Hence, the only issue was who set the fire.

*388 The Crosswinds Restaurant was owned by plaintiff, Windmere, Inc. Although the Crosswinds was operated by Jerry Ciambrone, the owner and only officer of Windmere, Inc. was his brother, Thomas Ciambrone, Jr. At trial Thomas Ciambrone claimed the Crosswinds had been financially successful. The defendant introduced evidence to the contrary. An accountant for the insurance company testified that, based on an analysis of Windmere's tax returns, the Crosswinds had failed to show a profit in any year of its operation. For the fiscal years ending April 30, 1976, through April 30, 1982, Windmere had suffered a cumulative loss in excess of $194,000.00. If the loss from the fiscal year 1974 is also included, the cumulative loss totaled approximately $235,000.00. Furthermore, Thomas Ciambrone had personally loaned $89,960.00 to Windmere in an effort to keep the Crosswinds Restaurant operating. Notably, the amount claimed by Windmere as a result of the fire loss was over $460,000.00.

Additional evidence indicated that plaintiff had been continually attempting to sell the restaurant. Numerous classified advertisements were placed in various papers, but all negotiations failed to produce a sale of the business. In fact, the last failure occurred only 10 days before the fire destroyed the Crosswinds.

Beyond the financial evidence indicating a motive for setting the fire, there was evidence that, if believed by the jury, indicated that Thomas Ciambrone had knowledge that the fire would be set. It is undisputed that both Jerry and Thomas Ciambrone were in Florida when the fire was set. Nevertheless, there was no evidence of forced entry into the Crosswinds; in fact, a back door key was found in the remnants of the fire and the burglar alarm was not working. The plaintiff's maintenance man admitted that several days before the fire he had purchased two gallons of gasoline for use at the Crosswinds. The purchase was charged to the account of an individual *389 identified as "Top Cat," a name by which Thomas Ciambrone was known. The maintenance man, who had no financial stake in either Windmere or the Crosswinds, was at the scene of the fire soon after the 3:02 a.m. fire began. Moreover, Thomas Ciambrone, after being informed by his own uncle that the Crosswinds had burned down, then called a member of the fire department and asked the question, to which he already knew the answer, whether there had been a fire at the restaurant. The trial court viewed this as merely a cover. The jury apparently agreed.

After oral argument, we requested the tapes and voiceprints in order to examine the exhibits. We learned, however, that the Passaic County prosecutor had subpoenaed the evidence. We have since been informed that the Ciambrone brothers have been indicted for arson in connection with the Crosswinds fire after an unexploded bomb was recently discovered in the ruins of the restaurant. Moreover, a person other than plaintiff's maintenance man has pleaded guilty to third-degree arson, and through his attorney has admitted making the anonymous phone call. Consequently, the conclusion of Alexanderson that by use of the voiceprints, the anonymous voice "positively" was determined to be that of plaintiff's maintenance man seems erroneous.

Of course these are but accusations and admissions at this time and do not resolve the civil litigation. But based upon the cumulative effect of all the evidence, and excluding the voiceprints, we find that there was sufficient evidence to sustain the jury's verdict, by a preponderance of credible evidence, that the fire was set by means within the control or knowledge of the plaintiff. In the context of criminal trials where voiceprints had been involved, several courts have sustained guilty verdicts, although disagreeing with the admission of such evidence, when the courts were convinced that the evidence did not have the capacity to produce an unjust result. See United States v. *390 McDaniel, 538 F.2d 408 (D.C. Cir.1976) (admission of voiceprint evidence was harmless error since other evidence of organized gambling was overwhelming); Cornett v. State, supra, 450 N.E.2d 498 (admission of voiceprint evidence in murder trial was harmless error since it was merely cumulative and not decisive of guilt).

We are similarly satisfied that this civil verdict was not the unjust result of the admission of the voiceprints. In the approximately forty-one pages of closing argument by defense counsel, there was but one brief reference to the tapes. Counsel asked the jury to compare the tape of the anonymous caller with the tape of plaintiff's maintenance man; no specific reference was made to the testimony of the voiceprint expert. Moreover, plaintiff's counsel emphasized that even if the anonymous caller was the maintenance man, there was nevertheless insufficient evidence to prove that Thomas Ciambrone, Jr. was involved. Hence, the voiceprints played only a partial role in the case and in most probability were not the decisive factor in the jury's deliberations.

We have carefully considered the claim of improper comment by defense counsel on Windmere's failure to produce two key witnesses. The trial court agreed that this comment was both unfair and prejudicial, and instructed the jury to disregard it. The prejudice was not, however, of sufficient magnitude to warrant reversal. We further leave undisturbed the dispositions of the other issues raised on appeal.

The judgment of the Appellate Division is affirmed.

For affirmance Chief Justice WILENTZ and Justices CLIFFORD, HANDLER, POLLOCK, O'HERN, GARIBALDI and STEIN-7.

Opposed None.

NOTES

[1] E.g., Cederbaums, "Voiceprint Identification: A Scientific and Legal Dilemma," 5 Crim.L.Bull. 323 (1969) (Undecided); Decker and Handler, "Voiceprint Identification Evidence Out of the Frye Pan and into Admissibility," 26 Am.U.L.Rev. 314 (1977) (Admissible); Greene, "Voiceprint Identification: The Case in Favor of Admissibility," 13 Am.Crim.L.Rev. 171 (1975) (Admissible); Hollien and McGlone, "The Effect of Disguise on `Voiceprint' Identification," 2 J. of Crim. Def. 117 (1976); Jones, "Danger Voiceprints Ahead," 11 Am.Crim.L. Rev. 549 (1973) (Inadmissible); Jones, "Evidence Vel Non The Non Sense of Voiceprint Identification," 62 Ky.L.J. 301 (1974) (Undecided); Kamine, "The Voiceprint Technique: Its Structure and Reliability," 6 San Diego L.Rev. 213 (1969) (Undecided); Comment, "Mr. Kusta's Magic Box: The Admissibility of Voiceprint Evidence in Criminal Cases," 10 Houston L.Rev. 85 (1972) (Admissible); Comment, "The Evidentiary Value of Spectrographic Voice Identification," 63 J.Crim.L.C. & P.S. 343 (1972) (Admissible); Comment, "The Voiceprint Dilemma: Should Voices Be Seen and Not Heard?," 35 Md.L.Rev. 267 (1975) (Inadmissible); Comment, "The Admissibility of Voiceprint Evidence," 14 S.D. L.Rev. 129 (1969) (Admissible); Comment, "The Status of Voiceprints as Admissible Evidence," 24 Syracuse L.Rev. 1261 (1973) (Inadmissible); Comment, "Voiceprints: The End of the Yellow Brick Road," 8 U. San Fran.L.Rev. 702 (1974) (Inadmissible); Comment, "The Voiceprint Technique: A Problem in Scientific Evidence," 18 Wayne L.Rev. 1365 (1972) (Admissible).

[2] The standard adopted by most courts for determining the admissibility of scientific evidence was first proposed in Frye v. United States, 293 F. 1013 (D.C. Cir.1923). The test is relatively straightforward. When a scientific principle or discovery has "gained general acceptance in the particular field in which it belongs," expert testimony on the subject is admissible. Id. at 1014. Thus, a technique must pass from the "experimental" stage and reach a "demonstrable" stage before a court will recognize the approach. Relevancy alone is insufficient; general acceptance by the relevant scientific community is required first.

[3] Under Evidence Rule 8, a trial court may conduct a hearing out of the presence of the jury to determine the qualification of a person to be a witness.

[4] Although our holding that more proofs will be required before the spectrograph device can be found "reasonably reliable" convinces us that a thorough Rule 8 hearing should be held prior to trial, no grave injustice occurred in this case. The plaintiff conducted a thorough cross-examination of the expert witnesses and through exceptional diligence was able to obtain a qualified expert who testified against the reliability of the device.

[5] Similarly, in State v. Prudden, 212 N.J. Super. 608, 617 (App.Div. 1986), the court, while rejecting the use of footprint techniques to identify a defendant, recognized that the techniques were nevertheless sufficiently reliable to exclude a person as the guilty party.

[6] The use of hypnotically induced evidence is not inadmissible in the first instance. Nothing would suggest that a live witness should not be permitted to testify to his or her recollection of events. The science of hypnotism is relevant to attacking the credibility of the hypnotically induced memory recall after the evidence is offered. State v. Hurd, 86 N.J. 525 (1981).