Barrel of Fun, Inc. D/B/A the Music Factory v. State Farm Fire & Casualty Company

GARWOOD, Circuit Judge: This case presents the question of whether expert testimony based solely on the results of a psychological stress evaluation (PSE) is admissible evidence. Because we conclude that no relevant distinction can be drawn between such testimony and expert testimony based on polygraph evidence— long held inadmissible by this Court — we hold that PSE evidence is likewise inadmissible. On November 24, 1981, a fire occurred at The Music Factory, a retail store selling phonograph records, tapes, stereo equipment, and musical instruments, in Larose, Louisiana. The Music Factory was owned by Barrel of Fun, Inc., a corporation operated and wholly owned by Floyd Guilbeau and his wife, Mona. After the fire, Barrel of Fun submitted a claim to State Farm Fire & Casualty Company pursuant to the store’s $40,000 fire insurance policy with State Farm. Asserting that the fire was intentionally set by the insured, State Farm refused to pay the claim. Barrel of Fun then filed this suit against State Farm to force payment on the policy. State Farm asserted the affirmative defense of arson and filed a counterclaim against Barrel of Fun and a third-party claim in subrogation against the Guilbeaus for property damage sustained by other tenants of the building in which The Music Factory was located. Following a bench trial, the district court found that Floyd Guilbeau was involved in setting the fire 1 and rendered judgment for State Farm, rejecting Barrel of Fun’s claims and awarding State Farm damages on its counterclaim. On appeal, Barrel of Fun asserts that the district court erred in admitting testimony on the results of the PSE. We agree, and accordingly vacate and remand for reconsideration. 2 *1030 During the trial below, the district court refused to admit testimony from Howard L. Dey, Jr., an investigator for the Louisiana State Fire Marshal’s office, regarding the results of the PSE he administered to Floyd Guilbeau. 3 Nevertheless, the district court admitted over plaintiff’s objections testimony from William M. Roth, Jr., an arson investigator for the Fire Marshal’s office, 4 that, based essentially only on the results of the PSE given by Dey, it was Roth’s opinion that Guilbeau had prior knowledge of and authorized the setting of the fire. 5 The overwhelming majority of those courts that have considered the issue have held that PSE evidence is inadmissible. United States v. Traficant, 566 F.Supp. 1046, 1047 (N.D.Ohio 1983); Caldwell v. State, 267 Ark. 1053, 594 S.W.2d 24, 28 (1980); State v. Makerson, 52 N.C.App. 149, 277 S.E.2d 869, 872 (1981); People v. Tarsia, 67 A.D.2d 210, 415 N.Y.S.2d 120, 122 (1979), affd, 50 N.Y.2d 1, 427 N.Y.S.2d 944, 405 N.E.2d 188 (1980); State v. Ochalla, 285 N.W.2d 683, 684 (Minn. 1979); State v. Schouest, 351 So.2d 462, 468-69 (La. 1977); Smith v. State, 31 Md.App. 106, 355 A.2d 527, 535-36 (Md.1976). See Joubert v. Travelers Indemnity Co., 736 F.2d 191, 194 (5th Cir.1984) (“argument that the trial court erred in holding that a voice stress test was inadmissible is without merit”), citing United States v. Clark, 598 F.2d 994 (5th Cir.1979) (per curiam), vacated, 608 F.2d 238, reinstated, 622 F.2d 917 (1980), cert. denied, 449 U.S. 1128, 101 S.Ct. 949, 67 L.Ed.2d 116 (1981). 6 *1031 Our precedents “unequivocally hold” that polygraph evidence 7 is inadmissible. Clark, at 995. Accord, Smith v. Gonzales, 670 F.2d 522, 528 n. 3 (5th Cir.), cert. denied, 459 U.S. 1005, 103 S.Ct. 361, 74 L.Ed.2d 397 (1982); United States v. Masri, 547 F.2d 932, 936 (5th Cir.), cert. denied, 434 U.S. 907, 98 S.Ct. 309, 54 L.Ed.2d 195 (1977); United States v. Cochran, 499 F.2d 380, 393 (5th Cir.1974), cert. denied, 419 U.S. 1124, 95 S.Ct. 810, 42 L.Ed.2d 825 (1975); United States v. Gloria, 494 F.2d 477, 483 (5th Cir.), cert. denied, 419 U.S. 995, 95 S.Ct. 306, 42 L.Ed.2d 267 (1974); United States v. Frogge, 476 F.2d 969, 970 (5th Cir.) (per curiam), cert. denied, 414 U.S. 849, 94 S.Ct. 138, 38 L.Ed.2d 97 (1973). “We have made no exception to the rule for civil cases____” Smith, supra, at 528 n. 3. See California Insurance Company v. Allen, 235 F.2d 178, 180 (5th Cir.1956). 8 Upon reviewing the record before us, we cannot find a principled distinction between the inadmissible results of polygraph tests and the results of the PSE offered in this case. Both types of tests are identical in one essential respect: by measuring a person’s physiological responses to various questions, the tests produce data which, when interpreted by an expert, purportedly indicate whether the person was lying in his answers. Heisse v. Vermont, 519 F.Supp. 36, 40 (D.Vt.1980). Moreover, we are persuaded that those concerns promptiug this and other courts to bar polygraph evidence from the courtroom apply with at least equal force to the PSE. Significant among those concerns is that the polygraph does not command sufficient scientific acceptance and “is not generally believed to be sufficiently reliable in ascertaining truth and deception to justify its utilization in the trial process.” United States v. Alexander, 526 F.2d 161, 164 (8th Cir.1975). Accord, United States v. Gloria, 494 F.2d at 483; United States v. Bursten, 560 F.2d 779, 785 (7th Cir.1977) (per curiam); United States v. Marshall, 526 F.2d 1349, 1360 (9th Cir.), cert. denied, 426 U.S. 923, 96 S.Ct. 2631, 49 L.Ed.2d 376 (1976); Frye v. United States, 293 F. 1013, 1014 (D.C.Cir.1923). 9 See also Fed.R.Evid. 401. Of equal concern is the risk that the jury or judge, sitting as triers of fact, “will abdicate' their responsibility for determining credibility, and rely instead upon the assessment of a machine.” Bursten at 785. Accord, Alexander at 168-69; United States v. Wilson, 361 F.Supp. 510, 513 (D.Md.1973); United States v. Stromberg, 179 F.Supp. 278, 280 (S.D.N.Y.), rev’d in part on other grounds, 268 F.2d 256 (2d Cir.1959); E. Cleary, McCormick’s Handbook of the Law of Evidence § 207 at 507 (2d ed. 1972). 10 Because the PSE has not yet gained general judicial recognition, the proponent *1032 of such evidence has the burden of showing as a predicate to its admission that the proffered test has achieved scientific acceptability and that the test has a reasonable measure of trustworthiness. Alexander at 163; Marshall at 1360; United States v. Wainwright, 413 F.2d 796, 803 (10th Cir.1969), cert. denied, 396 U.S. 1009, 90 S.Ct. 566, 24 L.Ed.2d 501 (1970). Based on the record before us, we conclude that State Farm has failed to discharge this burden. The only evidence proffered below pertaining to the general acceptance and use of the PSE was Dey’s testimony that the test is utilized by the State Fire Marshal’s office in its arson investigations and that Dey had administered 300 to 500 of the tests. The evidence, however, does not establish the extent to which PSE results are actually relied upon in those investigations, nor whether the PSE has attained broader scientific acceptance among experts in fire investigation, psychology, psychiatry, physiology, and other related disciplines. See Alexander at 164 & n. 6. See also Heisse v. Vermont, 519 F.Supp. at 41 (noting “disagreement in the scientific community about the validity of PSE testing”); United States v. Traficant, 566 F.Supp. at 1047 (same); Horvath, Detecting Deception: The Promise and the Reality of Voice Stress Analysis, 27 J. Forensic Sci. 340 (1982) (hereinafter cited as Horvath). Furthermore, although Dey testified that during the nine years in which he had administered PSE tests for the Fire Marshal’s office, “we haven’t had one incident come back and prove to us factually that we had the wrong indication from the instrument,” that fact by itself, in light of the lack of proof of general scientific acceptance of the PSE, does not convince us of the reliability of the test. See Traficant at 1047; State v. Makerson, 277 S.E.2d at 872; People v. Tarsia, 415 N.Y.S. at 122. See also Horvath, supra, at 349 (“All of the reliable evidence now available shows that none of the voice stress devices is useful in detecting deception; ____ The fact that voice stress devices have apparently been accepted rather uncritically by some law enforcement agencies, and for some forensic science purposes, is a development which, judging from the available evidence, cannot now be justified.”). We further observe that the district court refused to qualify Dey as an expert in this field {see note 3, supra). Finally, when either PSE or polygraph evidence is offered at trial, “it is likely to be shrouded with an aura of near infallibility, akin to the ancient oracle of Delphi.” United States v. Alexander, 526 F.2d at 168. Accord, United States v. Wilson, 361 F.Supp. at 513; People v. Leone, 25 N.Y.2d 511, 518, 307 N.Y.S.2d 430, 435, 255 N.E.2d 696, 700 (1969). Both tests invite the trier of fact to substitute the test results for its own credibility determinations, Traficant, 566 F.Supp. at 1047, thus inducing the fact finder to surrender “the responsibility for determining the truth of the evidence.” United States v. Price, 722 F.2d 88, 90 (5th Cir.1983). 11 Here, the PSE evidence directly contradicted the trial testimony of Floyd Guilbeau that he was not involved in, nor had any prior knowledge of, the arson. 12 To the *1033 extent the district court considered the test results unimpeachable, the defense case was won. See Alexander at 168. We therefore hold that our precedents ruling polygraph evidence inadmissible also preclude the admission of PSE results. See Joubert v. Travelers Indemnity Company, supra. 13 Although the district court properly excluded the testimony of Dey relating the results of the PSE, because Roth’s testimony implicating Guilbeau was, according to Roth's own words, in essence based only on the results of the PSE, his testimony was little more than a reiteration of the inadmissible PSE evidence. The fact that Roth’s knowledge of the test was second hand 14 does not enhance the value of his testimony or lessen its harm. To the extent Dey’s testimony was inadmissible, so was Roth’s. In admitting Roth’s testimony, the district court relied on Fed.R.Evid. 703. 15 We believe the court misapprehended the scope and effect of that Rule. It correctly noted that under Rule 703, expert testimony may not be excluded merely because it is based on facts or data that are inadmissible in evidence. However, the Rule does not guarantee the admissibility of all expert testimony that meets its criteria if such testimony runs afoul of other evidentiary requirements. For instance, expert testimony otherwise admissible under Rules 702 and 703 may still be excluded “if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury____” Fed.R.Evid. 403. Because we hold today that PSE evidence, whether in the form of raw data or expert opinion interpreting or extrapolating upon that data, is inherently suspect, Rule 703 cannot, standing alone, provide an avenue for its admission. Nor are we convinced that Roth’s testimony complied with Rule 703’s threshold requirement that facts or data forming a basis for an expert’s opinion be “of a type reasonably relied upon by experts in the [witness’] particular field in forming opinions or inferences upon the subject” testified on. In determining whether this requirement is met, “the court should concentrate on the reliability of the opinion____” Bauman v. Centex Corp., 611 F.2d 1115, 1120 n. 5 (5th Cir.1980), quoting 3 Weinstein & Berger, Weinstein’s Evidence ¶ 703[03] at 703-17 (1978). When making this assessment, “the trustworthiness of the underlying data is not irrelevant.” Emerging Problems Under the Federal Rules of Evidence, 1983 A.B.A. Sec. Lit. 208. As we have noted, the record does not sufficiently establish the trustworthiness of the PSE or its acceptance in the relevant scientific community. The principal purpose for adoption of Rule 703 was to codify the Rule adopted in this and other Circuits permitting an expert witness to express an opinion “based in part or solely upon hearsay sources.” United States v. Williams, 447 F.2d 1285, 1290-91 (5th Cir.1971) (en banc), cert. denied, 405 U.S. 954, 92 S.Ct. 1168, 31 L.Ed.2d 231 (1972). Our holding today is in *1034 no way predicated upon any hearsay objection to Roth’s testimony. Nor do we suggest that an expert’s opinion would necessarily become inadmissible under Rule 703 merely because the expert gave some consideration to a PSE test. Here, however, Roth’s opinion as to Guilbeau’s involvement is stated to be based essentially on Guilbeau’s PSE test results alone (see note 5, supra). The district court relied upon a variety of evidence in • finding that State Farm had proven the involvement of Floyd Guilbeau in the fire. We cannot determine from the district court’s opinion and findings of fact whether Roth’s challenged testimony respecting Guilbeau was critical to the district court’s ultimate finding that Guilbeau was involved in the arson. We note, however, that Roth’s opinion in this respect was elicited by the district court on its own initiative, and over the express objection of Barrel of Fun (see note 5, supra). Moreover, among the evidence cited by the court in its memorandum was Roth’s opinion implicating Guilbéau based upon 'the results of the PSE, evidence which that memorandum expressly states was properly admitted. The record affirmatively suggests the strong possibility that the district court may have crucially relied on this aspect of Roth’s testimony. Accordingly, we vacate the district court’s judgment and remand to allow the court to reconsider its determination without reliance upon Roth’s testimony insofar as it was based upon the results of Floyd Guilbeau’s PSE. 16 VACATED and REMANDED. Footnotes [1] Barrel of Fun did not dispute the fact, stipulated at trial, that the fire at The Music Factory was caused by arson. It did contest, however, that it, or either of the Guilbeaus, was responsible. [2] Barrel of Fun also claims that the evidence is insufficient to support the district court’s finding that Floyd Guilbeau was involved in setting the fire. We reject this contention. We hold that the evidence, though largely circumstantial, was nevertheless sufficient, that the district court’s findings are not clearly erroneous, and that the Louisiana requirements as to the sufficiency of the evidence in this type of case, see Joubert v. Travelers Indemnity Co., 736 F.2d 191, 193 (5th Cir.1984); Rist v. Commercial Union Insurance Company, 376 So.2d 113 (La. 1979), were also at least minimally met. No useful purpose would be served by detailing all the evidence. It included proof that the fire was an inside job; preparations for the fire were relatively extensive; entry was gained by someone having a key to the premises; the perpetrator attempted to make it appear that entry had been gained by force from the outside; there were only three keys, one held by each Guilbeau and one by Peggy Guidroc, who was employed by plaintiff until shortly after the fire; Guidroc testified th.at a few weeks before the fire Floyd Guilbeau stated in her presence to Kathleen Molaison that "if the business [at The Music Factory] didn’t pick up ... you will see the biggest bonfire you ever saw”; Molaison testified she could not remember whether or not Guilbeau made such a statement, but admitted she had given a sworn statement saying that he did and that subsequently she had been rehired by him and was so employed at the time of trial; The Music Factory business had not been good and was getting worse; it was losing money and was a cash drain on the Guilbeaus, who were in tight financial circumstances; Floyd Guilbeau had engaged in highly irregular and deceptive practices concerning the business’ financial records and cash; he denied having anything to do with the fire or having made the statement attributed to him by Guidroc, and stated he knew of no one who might want to set the fire because of animosity toward him; Mona Guilbeau was at the store alone the entire day before the fire (which took place in the evening or early morning), and that was the *1030 only occasion she had ever been there alone all day; the $75 left in the cash register when she closed that day was still there after the fire. We do not suggest that the district court was bound to find as it did. For example, the district court would surely have found otherwise had it credited all the testimony of Floyd Guilbeau, and likely also had it failed to credit Guidroc’s testimony. [3] The PSE, a type of "voice stress analysis,”' purportedly indicates whether a person is lying by measuring stress in his voice. The stress is assumed to accompany conscious deception. See, e.g., Heisse v. Vermont, 519 F.Supp. 36, 40 (D.Vt.1980); United States v. Traficant, 566 F.Supp. 1046 (N.D.Ohio 1983); Horvath, Detecting Deception: The Promise and the Reality of Voice Stress Analysis, 27 J. Forensic Sci. 340 (1982). In Dey’s testimony proffered by State Farm, he stated that the PSE he administered to Guilbeau indicated that Guilbeau had prior knowledge of who set the fire at The Music Factory, and lied when he denied such knowledge. The district court refused to accept the proffer, apparently on the ground that Dey was not shown to be adequately qualified as an expert in interpreting the test results. The court stated, "I will not recognize him [Dey] as an expert in the psychological test examination” and "[since] ... the [test] results are not like an adding machine tape, but are a matter of interpretation by a person, I am going to have to disallow him [Dey].” [4] Roth was qualified by the district court as an expert in fire investigation. [5] During direct examination by defense counsel, Roth testified that he believed the fire was set from the inside of The Music Factory store by someone who had access to the building with a key. Roth did not implicate Guilbeau, however, except in response to questioning by the district court. Roth initially testified, "The fire was without a doubt an incendiary fire, an arson case.” The court then asked, "Do you have any idea who set it?” to which Roth replied, "Now, I can give you an opinion.” The court asked Roth to do so, and Roth then proceeded to explain why he concluded it was done "from inside" by someone with a key, at the end of which he stated: "We did go further in this investigation and it left me with an opinion, if you want me to . state it? "THE COURT: Yes. I would like you to. "THE WITNESS: It had to do with the psychological stress evaluation test of Mr. Guilbeau’s voluntarily admitted to twice, not once, twice — •” There then followed objections by counsel for Barrel of Fun to testimony based on the PSE, which the court overruled. Roth then testified: "My honest opinion sitting here in Court today, Your Honor, is that the investigation I conducted, the scene that I viewed, the evidence that I confiscated and in the PSE, I honestly say that my opinion is that Mr. Guilbeau did not set the fire, but the test showed that he had strong knowledge that it was to be done. Meaning that he was involved to a point of having the fire set. He had knowledge of it.” [6] The New Mexico Court of Appeals in Simon Neustadt Family Center v. Bludworth, 97 N.M. 500, 641 P.2d 531 (1982), departed from the predominant view in holding PSE evidence conditionally admissible. The court noted that, unlike other courts that had ruled PSE evidence *1031 inadmissible, it already allowed the admission of polygraph evidence. Id. 641 P.2d at 536. [7] A polygraph, commonly called "lie detector,” measures changes in blood pressure, respiration, pulse rate, and galvanic skin response. The operator presumably is able to determine when a test subject is lying by interpreting the tracings of the polygraph. [8] Although considerable controversy has arisen regarding the continued advisability of the rule against admission of polygraph evidence, see, e.g., 22 C. Wright & K. Graham, Federal Practice and Procedure § 5169 (1978), “[ajny suggested deviation from the clear law of this Circuit would have to be addressed to the en banc court." United States v. Clark, 598 F.2d 994, 995 (5th Cir.1979) (per curiam), vacated, 608 F.2d 238, reinstated, 622 F.2d 917 (1980), cert. denied, 449 U.S. 1128, 101 S.Ct. 949, 67 L.Ed.2d 116 (1981). [9] The Frye court first enunciated the now familiar rule that in order for expert opinion to be admitted into evidence, the scientific principles on which the opinion is based "must be sufficiently established to have gained general acceptance in the particular field in which it belongs.” Id. at 1014. Although this Court has noted that it is an "unresolved question whether the Federal Rules of Evidence silently abolished or adopted the Frye test,” United States v. Valdez, 722 F.2d 1196, 1201 (5th Cir.1984), we have continued to utilize Frye’s "general scientific acceptability” criteria. See, e.g., United States v. Arias, 639 F.2d 1183, 1185 (5th Cir.1981); United States v. Ortiz, 610 F.2d 280, 282 (5th Cir.) (per curiam), cert. denied, 445 U.S. 930, 100 S.Ct. 1320, 63 L.Ed.2d 764 (1980). [10] These objections do not invariably vanish in a bench trial. See, e.g., People v. Sinclair, 21 Mich.App. 255, 175 N.W.2d 893 (1970). [11] In Price, this Court reversed a judgment of conviction due in part to the erroneous admission of testimony by a prosecution witness that he believed the statements of other government witnesses. The Court stated: "When bolstering testimony suggests to the jury that it may shift to a witness the responsibility for determining the truth of the evidence, its admission may constitute reversible error. [The government witness'] ... statements were tantamount to 'expression^] of his general belief as to how the case should be decided,' which are not only prejudicial but also ‘wholly without value to the trier of fact in reaching a decision.’ ” Id. at 90 (citations omitted). [12] This is not a case where the subject of the PSE was unavailable for trial, and we need not speculate whether our decision might be different in such a case. We do not suggest that the use of such tests is improper in the private sector or in administrative determinations, or that proof of reliance on such a test may not be made in judicial proceedings for the purpose of showing the good faith or reasonableness of the party so relying, where that is in issue, or for the purpose suggested in California Insurance *1033 Company v. Allen, supra, 235 F.2d at 180. Nor do we speak to whether, or in what circumstances or to what extent, if any, the Constitution may restrict the use of such evidence in judicial trials. [13] As the Maryland high court aptly stated in Smith v. State, 31 Md.App. 106, 355 A.2d 527, 536 (Md.1976), ‘‘The difference, if any, between the psychological stress evaluation test and a lie detector test is too minor and shadowy to justify a departure from our prior decisions], A lie detector test by any other name is still a lie detector test.” [14] The record gives no indication that Roth was either involved in administering the test or interpreting its results, or that he had any expertise regarding the test. [15] Rule 703 states: "The facts or data in the particular case upon which an expert bases an opinion or inference may be those perceived by or made known to him at or before the hearing. If of a type reasonably relied upon by experts in the particular field in forming opinions or inferences upon the subject, the facts or data need not be admissible in evidence.” [16] If the district court deems it appropriate to also reopen the evidence, it has discretion to do so.