Opinion of the Court
Paul W. BROSMAN, Judge:
The accused was tried by a general court-martial, convened on the island of Guam by the Commander, Naval Forces Marianas. The charge was sodomy, in violation of the Uniform Code of Military Justice, Article 125, 50 USC § 719. The accused pleaded not guilty, but was convicted and sentenced to receive a bad-conduct discharge, as well as to total forfeitures and confinement at hard labor for eighteen months. The convening authority approved the findings but reduced the confinement adjudged to eleven months. Following affirmance by a board of review in the office of The Judge Advocate General, United States Navy, we granted the accused’s petition to determine whether the admission of certain expert testimony concerning a witness’ veracity amounted to prejudicial error.
II
To place the legal question in perspective it will be necessary to recite portions of the sordid testimony. Summarized, the prosecution evidence in chief established that one afternoon during March 1952 Adkins, together with several other sailors, had gone on liberty. At the time the accused was driving an automobile of which he and one of his companions were co-owners. At some point during the excursion, one Sandoval attached himself to the party by invitation of at least one of its number. Subsequently Adkins directed the vehicle from the main highway into a side road, and there came to a stop inside an abandoned quonset hut — described at the trial as an “elephant hut.” Part of the group went into another structure — also deserted — and began drinking from a bottle of whiskey. However, Sandoval remained in the car.
A prosecution witness by the name of Evans testified that thereafter he returned to the automobile, and there committed an unnatural sexual act with Sandoval. Sandoval himself asserted that, in addition to Evans, several others — among them the accused — came singly to the car to participate in homosexual acts with him. In fact, according to Sandoval, the accomplishment of this misconduct was a chief purpose in leaving the highway and seeking the privacy of the quonset hut. He added that various homosexual overtures toward him had been made in the vehicle prior to arrival at the party’s destination.
Evans himself did not intimate in his testimony that the accused had returned to the car or had at anytime [495]*495been alone with Sandoval. One Newport, another member of the group— and a prosecution witness — similarly did not testify to any matter suggesting that the accused had been alone in the automobile with Sandoval. Newport denied that he himself had engaged in an act of sodomy with Sandoval — although it appeared that a general court-martial had earlier convicted him of that offense.
The defense relied heavily on a searching cross-examination of Sandoval. During this interrogation it was adduced that, at the outset, Sandoval had misidentified one of the participants in the alleged orgy; that he had spoken inconsistently as to various details connected with the offenses; and that he was a confirmed homosexual. Indeed, relying on its cross-examination of this person — the central witness for the Government — the defense moved for findings of not guilty after the prosecution had rested its case. However, the motion was denied by the law officer without objection from any member of the court-martial.
The defense then called a seaman named Brown, who had also been accused by Sandoval as a figure in the same episode. Brown emphatically denied the truth of the accusation against him, and asserted that, when Sandoval offered a homosexual invitation after the former had initially alighted from the car in the hut, the proposal had been forcefully rejected. Brown, it developed, also had previously been tried by general court-martial on a charge of sodomy but — despite Sandoval’s testimony there — had been acquitted. It should be mentioned in passing that, since the main witnesses at the present trial were identical with those at the preceding courts-martial of Newport and Brown, the record is replete with instances of cross-examination based on testimony given at the earlier hearings.
Following the appearance of Brown, Adkins took the stand, and denied virtually all of the allegations of Sandoval. In substance the accused conceded only that he had driven the car on the day in question, and that he had gone with the group into the hut where the drinking party was later held. Although he had not partaken — for he rarely drank alcoholic beverages — he remained with the group until the bottle had been emptied and its members had agreed to return to the Naval Base for the evening meal. Both the examination and cross-examination of the accused covered a broad range. Among other items, defense counsel brought out that the accused had been treated medically for epididymitis shortly before the incident now under scrutiny. Trial counsel, on the other hand, went into Adkins’ heterosexual activities and into his family background. In the latter connection the accused testified that he had four sisters and five brothers, and that for an undetermined period during his youth he had shared a bed with other children of the family.
Sandoval was recalled by the Government, and supplied an even more elaborated account of the alleged act of fellatio performed with Adkins. Following this testimony, he was again cross-examined in great detail, and repeated his testimony to the effect that eight persons had been in the automobile on the day in question — whereas other witnesses had insisted that only five were present. He admitted once more that initially he had misidentified Newport, but sought to explain this error and referred to its prompt correction.
Subsequently Lieutenant Francis Yates, a medical officer, was called as a prosecution witness to explain the significance and cause of epididymitis. Apparently trial counsel sought to suggest by this witness that the accused’s ailment possessed no sort of probative weight in establishing that he had not committed the offense with which he was charged. A further prosecution witness was called for an inconsequential impeachment purpose, after which one Kinniry, an agent of the Office of Naval Intelligence, appeared as a rebuttal witness — and furnished the testimony which presents the legal issue in this case.
Kinniry — who had served with Naval Intelligence for some ten years — had [496]*496been assigned to Guam for a period of fourteen months preceding the trial. Approximately one-half of this time had been devoted to the investigation of homosexual activities on the island, and during his tenure there — he reported— he had investigated from sixty to seventy such cases, involving civilian as well as Naval personnel. Something more than one-half of these cases, Kinniry stated, had to do with active homosexuals — -by which term he meant “one who generally takes the part of the girl and who makes the passes.” He placed Sandoval within the “active” category, and thereafter was asked by trial counsel: “Mr. Kinniry, have you ever known a homosexual of the active type to knowingly, willfully, and falsely accuse a person?” Defense counsel immediately objected, stating, “I don’t think the witness has been qualified to express any opinion.” However, the Office of Naval Intelligence agent was allowed to respond as follows, “It has been my experience in the cases I have investigated that when an active homosexual makes a statement that he has engaged in sex with somebody that he is 100% true in so naming the individual he had sex with.
Free access — add to your briefcase to read the full text and ask questions with AI
Opinion of the Court
Paul W. BROSMAN, Judge:
The accused was tried by a general court-martial, convened on the island of Guam by the Commander, Naval Forces Marianas. The charge was sodomy, in violation of the Uniform Code of Military Justice, Article 125, 50 USC § 719. The accused pleaded not guilty, but was convicted and sentenced to receive a bad-conduct discharge, as well as to total forfeitures and confinement at hard labor for eighteen months. The convening authority approved the findings but reduced the confinement adjudged to eleven months. Following affirmance by a board of review in the office of The Judge Advocate General, United States Navy, we granted the accused’s petition to determine whether the admission of certain expert testimony concerning a witness’ veracity amounted to prejudicial error.
II
To place the legal question in perspective it will be necessary to recite portions of the sordid testimony. Summarized, the prosecution evidence in chief established that one afternoon during March 1952 Adkins, together with several other sailors, had gone on liberty. At the time the accused was driving an automobile of which he and one of his companions were co-owners. At some point during the excursion, one Sandoval attached himself to the party by invitation of at least one of its number. Subsequently Adkins directed the vehicle from the main highway into a side road, and there came to a stop inside an abandoned quonset hut — described at the trial as an “elephant hut.” Part of the group went into another structure — also deserted — and began drinking from a bottle of whiskey. However, Sandoval remained in the car.
A prosecution witness by the name of Evans testified that thereafter he returned to the automobile, and there committed an unnatural sexual act with Sandoval. Sandoval himself asserted that, in addition to Evans, several others — among them the accused — came singly to the car to participate in homosexual acts with him. In fact, according to Sandoval, the accomplishment of this misconduct was a chief purpose in leaving the highway and seeking the privacy of the quonset hut. He added that various homosexual overtures toward him had been made in the vehicle prior to arrival at the party’s destination.
Evans himself did not intimate in his testimony that the accused had returned to the car or had at anytime [495]*495been alone with Sandoval. One Newport, another member of the group— and a prosecution witness — similarly did not testify to any matter suggesting that the accused had been alone in the automobile with Sandoval. Newport denied that he himself had engaged in an act of sodomy with Sandoval — although it appeared that a general court-martial had earlier convicted him of that offense.
The defense relied heavily on a searching cross-examination of Sandoval. During this interrogation it was adduced that, at the outset, Sandoval had misidentified one of the participants in the alleged orgy; that he had spoken inconsistently as to various details connected with the offenses; and that he was a confirmed homosexual. Indeed, relying on its cross-examination of this person — the central witness for the Government — the defense moved for findings of not guilty after the prosecution had rested its case. However, the motion was denied by the law officer without objection from any member of the court-martial.
The defense then called a seaman named Brown, who had also been accused by Sandoval as a figure in the same episode. Brown emphatically denied the truth of the accusation against him, and asserted that, when Sandoval offered a homosexual invitation after the former had initially alighted from the car in the hut, the proposal had been forcefully rejected. Brown, it developed, also had previously been tried by general court-martial on a charge of sodomy but — despite Sandoval’s testimony there — had been acquitted. It should be mentioned in passing that, since the main witnesses at the present trial were identical with those at the preceding courts-martial of Newport and Brown, the record is replete with instances of cross-examination based on testimony given at the earlier hearings.
Following the appearance of Brown, Adkins took the stand, and denied virtually all of the allegations of Sandoval. In substance the accused conceded only that he had driven the car on the day in question, and that he had gone with the group into the hut where the drinking party was later held. Although he had not partaken — for he rarely drank alcoholic beverages — he remained with the group until the bottle had been emptied and its members had agreed to return to the Naval Base for the evening meal. Both the examination and cross-examination of the accused covered a broad range. Among other items, defense counsel brought out that the accused had been treated medically for epididymitis shortly before the incident now under scrutiny. Trial counsel, on the other hand, went into Adkins’ heterosexual activities and into his family background. In the latter connection the accused testified that he had four sisters and five brothers, and that for an undetermined period during his youth he had shared a bed with other children of the family.
Sandoval was recalled by the Government, and supplied an even more elaborated account of the alleged act of fellatio performed with Adkins. Following this testimony, he was again cross-examined in great detail, and repeated his testimony to the effect that eight persons had been in the automobile on the day in question — whereas other witnesses had insisted that only five were present. He admitted once more that initially he had misidentified Newport, but sought to explain this error and referred to its prompt correction.
Subsequently Lieutenant Francis Yates, a medical officer, was called as a prosecution witness to explain the significance and cause of epididymitis. Apparently trial counsel sought to suggest by this witness that the accused’s ailment possessed no sort of probative weight in establishing that he had not committed the offense with which he was charged. A further prosecution witness was called for an inconsequential impeachment purpose, after which one Kinniry, an agent of the Office of Naval Intelligence, appeared as a rebuttal witness — and furnished the testimony which presents the legal issue in this case.
Kinniry — who had served with Naval Intelligence for some ten years — had [496]*496been assigned to Guam for a period of fourteen months preceding the trial. Approximately one-half of this time had been devoted to the investigation of homosexual activities on the island, and during his tenure there — he reported— he had investigated from sixty to seventy such cases, involving civilian as well as Naval personnel. Something more than one-half of these cases, Kinniry stated, had to do with active homosexuals — -by which term he meant “one who generally takes the part of the girl and who makes the passes.” He placed Sandoval within the “active” category, and thereafter was asked by trial counsel: “Mr. Kinniry, have you ever known a homosexual of the active type to knowingly, willfully, and falsely accuse a person?” Defense counsel immediately objected, stating, “I don’t think the witness has been qualified to express any opinion.” However, the Office of Naval Intelligence agent was allowed to respond as follows, “It has been my experience in the cases I have investigated that when an active homosexual makes a statement that he has engaged in sex with somebody that he is 100% true in so naming the individual he had sex with. I found that to be true in all the cases that I have worked on.”
Subsequently, Mr. Kinniry explained further that he had investigated from three hundred to four hundred cases of homosexuality during the ten years prior to trial; that in those cases he had taken occasion to observe the traits and family background of the persons involved; and that it had been his experience that both the active and the passive homosexual are “products óf broken homes.” He added, “In cases where there are large families and where there are three and four children sleeping in one bed, I find that homosexuals were developed in that way.” On the basis of this experience he also stated to the court that he could not “tell a passive member by looking at him,” and that “Passive homosexuals are usually bisexual and would just as soon have relations with women as with men.” On redirect examination, and under examination by the court, Mr. Kinniry testified that his experience with homosexuals indicated that “birds of a feather flock together.”
After Kinniry’s testimony, the defense called a number of character witnesses, who related that they considered the accused to be a truthful person and one of “good morals.” The defense then entered once more into the matter of Sandoval’s confusion concerning Newport’s name, which was apparent at the time he had originally identified the homosexual participants on this occasion.
Both the trial counsel and the two defense attorneys offered extensive closing arguments. The Government’s representative emphasized that Mr. Kin-niry had afforded the court the “benefit of his experience and told us that in all the cases he has investigated he has never known a confirmed homosexual to intentionally name and falsely accuse the wrong person and stick to it.” Next he adverted to the comment with respect to “birds of a feather,” and inquired rhetorically “If that is of any probative value, how about Adkins running around with NEWPORT?” Newport — it will be recalled — was the prosecution witness who had been convicted of sodomy following the incident in suit, and was the co-owner, together with Adkins, of the automobile used at that time. At a later point trial counsel remarked that he had developed in evidence the accused’s family background with a particular object in mind, “and you can tie that in with the testimony of Mr. Kinniry that where children sleep together they are likely to get this tendency of homosexuality.”
Ill
The relationship of homosexuality to credibility is presented here within a novel context. Previously, in United States v. Long, 2 USCMA 60, 6 CMR 60, we had before us the problem of whether a law officer had erred in forbidding to defense counsel the examination of a prosecution witness with respect to suspicion of homosexuality on her part. We held there that the law officer’s ruling placed a reasonable and fair limit on the testing of credibility, and pointed [497]*497out the absence of a showing that the offered acts would in any degree reveal characteristics which might serve to minimize the witness’ credit. The Court of Appeals for the Second Circuit reached a kindred result in United States v. Provoo, 215 F2d 531. There the accused had been cross-examined extensively concerning hospitalization and confinement for alleged homosexuality. The court held that this cross-examination constituted reversible error —since its sole effect had been to degrade the defendant. It emphasized that “No authority has been cited which suggests that homosexuality indicates a propensity to disregard the obligation of an oath.”
The present case involves the converse situation — an effort to establish, by allegedly expert testimony, that Sandoval’s homosexuality enhanced the probability that he was telling the truth at the time he accused Adkins of engaging in an act of sodomy with him. Our initial difficulty lies in an acceptance of the proposition that Mr. Kin-niry in fact qualified as an expert for the purpose of announcing some of the propositions concerning homosexuals and homosexuality with respect to which he testified at the trial. According to the Manual for Courts-Martial, an expert Witness is “one who is skilled in some art, trade, profession or science or who has knowledge and experience in relation to matters which are not generally within the knowledge of men of common education and experience.” Paragraph 138e. We have earlier pointed out that a law officer is to be granted wide discretion in applying this standard. United States v. Hagelberger, 3 USCMA 259, 12 CMR 15. Yet we are sure that the present law officer exceeded sound limits in ruling that Mr. Kinniry was qualified to offer certain of the opinions he was permitted to express at the trial.
Concededly the latter had enjoyed greater experience with homosexuals than is possessed by “men of common education and experience.” So far as the record shows, however, he lacked any sort of medical or other scientific training in psychiatric disorders. It is common knowledge that homosexuality remains an enigma to numerous scientific students who have devoted impressive periods of time to the consideration of its causes and effects. At their disposal — either through professional observation or in medical literature— are innumerable case histories. We doubt that, in our present state of knowledge, anyone could qualify as sufficiently expert to express a judicially admissible opinion on the veracity of the homosexual. Certainly we find no reason to construct an exception for Mr. Kinniry — despite his greater-than-average acquaintance with these unfortunates.
Moreover, an expert opinion should bear some measurable relation to empirical observation. As the scientist would put it, the opinion should be based on results substantially verifiable. Mr. Kinniry spoke of the credibility of an active homosexual’s accusations. How in the past had he verified the truthfulness of such allegations? It is indisputable that sodomy is not customarily performed under conditions of publicity. Generally only the participants in the act are present during its performance, and each might well recognize an interest in furnishing a false account of the transaction. Thus, the situation is quite unlike that in which a typical scientific experiment is conducted — one in which the expert ordinarily is present to test his theories, and where observers are without motive to falsify results. Presumably there would have been but a single source of verification of the allegations of the active homosexual in most of the cases Kinniry had handled — the statement of the accused person, himself. This statement would necessarily assume the form of either (1) an admission, or (2) a denial. Insofar as the alleged participant denied the accusation, it is at once apparent that its truthfulness cannot be said to have been verified- — -at least in any scientific usage of the term. Certainly the factual determination of any subsequent board or court-martial functioning in the premises cannot be permitted to serve this purpose. And we cannot suppose that the totality of Mr. Kin-niry’s investigations had resulted in [498]*498a perfect score with respect to confessions. Indeed, the sampling furnished by the quonset hut interlude — with which we are concerned here — would seem to point in the direction of a conclusion wholly contrary to that expressed by him from the witness stand, for it will be remembered that three of the persons accused by Sandoval as homosexual participants had recorded sworn denials of the charge. It must be apparent that the Kinniry testimony as to Sandoval’s veracity should not have been admitted.1
The challenged statements regarding the family background of the homosexual share in part the infirmity of an insufficient basis in observation. This deficiency appears to be both quantitative and qualitative. In light of the inconclusive scientific results based on far wider cross-sections, it strikes us that the relatively few cases observed by Agent Kinniry should not suffice to permit in court his sweeping conclusions concerning the domestic milieu of homosexuals. Moreover, it seems clear that, in his work as a busy investigator, he would have enjoyed little or no opportunity for an intensive study of the family life of the alleged homosexuals with whom he was concerned. And to form valid conclusions based on domestic background would probably have taxed Mr. Kinniry’s training even more than to arrive at a reliable assessment of the credibility of persons suspected as active perverts.
As to the admissibility of an opinion from Kinniry relating to the truthfulness of the accusatory assertions of active homosexuals, we suggest that an analogy is to be found in the numerous cases excluding evidence of the results obtained by the operators of so-called “lie detectors.” State v. Bohner, 210 Wis 651, 246 NW 314; People v. Forte, 279 NY 204, 18 NE2d 31; People v. Becker, 300 Mich 562, 2 NW2d 503; State v. Lowry, 163 Kan 622, 185 P2d 147; Boeche v. State, 151 Neb 368, 37 NW2d 593; People v. Wochnick, 98 Cal App2d 124, 219 P2d 70; State v. Pusch, 77 ND 860, 46 NW2d 508; Peterson v. State, 157 Tex Cr 255, 247 SW2d 110; Kaminski v. State, — Fla —, 63 So2d 339; United States v. Pryor [CM 348625], 2 CMR 365; Inbau and Reid, Lie Detection and Criminal Interrogation, 3d ed, 1953, pages 122-141 (collecting the cases). Whether this exclusion is based on the lack of expertise of such operators, the scientific unreliability of the device involved, or the invariable inadmissibility of expert opinion with respect to credibility, it constitutes an a fortiori case for challenging the opinion ventured in the instant case. Many polygraph operators have interviewed large numbers of witnesses, and there is available a substantial body of literature relating to the techniques for and the results of polygraphic interviews. The percentage of accuracy here is relatively high. See Inbau and Reid, supra, pages 110-114. To some extent results can be verified objectively for accuracy — for instance, through the discovery of real evidence resulting from a lie detector “lead.” Moreover, the success of the lie detector is explicable in terms of a general scientific theory. Inbau and Reid, supra, pages 1 — 121; Wigmore, Evidence, 3d ed, 1940, § 999.
We have been referred, however, to no respectable scientific theory which would suggest even remotely that an active homosexual may not lodge a false accusation. Why may he not wish to enhance his feeling of importance by means of a false charge? Why may not his allegations be tinged with either malice or fantasy? We do not know, and neither does scientific learning. Neither — we venture to suggest — does Mr. Kinniry. In short, the expert opinion of the witness here respecting the invariable truthfulness of an active homosexual — an opinion which reduces to a statement that, as a matter of scientific fact, Sandoval must have been tell[499]*499ing the truth in the ease at bar — is so much more suspect from a scientific, standpoint than the results of poly-graphic interviews, that we are unable to justify accepting the former while rejecting the latter.
IV
The Manual provides that — after impeachment in specified ways — a witness may be rehabilitated by means of character or reputation testimony designed to show that he possesses a good character for truth and veracity. Paragraph 1536 (2) (a). Additionally, this passage authorizes an inquiry of the character witness with respect to whether he would believe the impeached witness on oath. However, we find neither Manual authority nor case precedent for permitting a character witness to testify that he believes particular testimony supplied by a witness during the course of a trial. One rationale for the exclusion of such testimony would be that the members of the court may be misled into believing that the character witness has himself made inquiry into the merits of the case, and has reached his own admissible conclusion. If the character witness is sufficiently distinguished — or responsibly placed — the conclusion assumed to have been reached might well serve to distract the court’s attention from other evidence presented to it. Similarly, Kinniry’s-testimony would have had the inevitable tendency to direct the court’s attention away from the other evidence before it— which included a searching cross-examination of Sandoval, whose veracity was a chief point in issue.
While there are categories of persons viewed by the law as especially wanting in testimonial credit — for instance, those previously convicted of offenses involving moral turpitude — military law insists for the most part that the triers of fact scrutinize the witness closely as an individual and carefully weigh his testimony in the particular case. Generalization concerning the veracity of classes of witnesses are and should be suspect. Cf. Quercia v. United States, 289 US 466, 77 L ed 1321, 53 S Ct 698. Kinniry’s testimony on the subject of veracity distinctly infringed on this policy — and its reception constituted error, regardless of the presence of real or fancied expertise.
V
We certainly have no desire to man-aéle a court-martial in its reception of evidence which logically might assist in resolving the problem before its members. With this in mind, we must point out that we find no error in permitting Mr. Kinniry to testify that he knew of no unerring test for the detection of homosexuals by appearance or public conduct. Such a conclusion would fall safely within the scope of his expertise. If, however, he had sought to propose to the court a test for detection, we are sure that he would have strayed beyond his area of competence. It seems fairly clear that science has found no ready reagent for the isolation of the sex pervert. Indeed, were the converse true, the Armed Forces presumably would entirely avoid the homosexual offender through preinduetion screening.
We have experienced somewhat greater doubt concerning the admissibility of the testimonial statement by Mr. Kinniry to the effect that “birds of a feather flock together.” Of course, Mr. Kinniry was in a better position to describe the homosexual’s associations than to testify regarding more recondite phases of homosexuality — such as family background and veracity. His opportunity to observe the gathering of known perverts would have been infinitely more favorable than the likelihood of his knowing with assurance what transpired in secret between two persons — one claiming and the other denying that an act of sodomy had taken place.
However, although we assume that Mr. Kinniry was qualified to furnish expert testimony on this score, further difficulty is present. It goes without saying that we do not accept the principle of “guilt by association.” Yet there is much in human experience consistent with the probability that homosexuals are characterized by a stronger [500]*500tendency to congregate than is possessed by other criminals. Conceivably some special rationale exists in the area of the sexual offense for erecting an exception to the usual prohibition against evidence of suspicious associations. Indeed, certain courts have seen fit to modify other evidentiary rules in sex trials. Thus, evidence of prior acts involving the same persons is admissible in many jurisdictions to establish sexual misconduct — although a different principle would govern prosecutions lacking the sexual element. Crawford v. United States, 198 F2d 976 (CA DC Cir); Hodge v. United States, 126 F2d 849 (CA DC Cir); Wigmore, Evidence, 3d ed, §§ 398-402. See also Bracey v. United States, 142 F2d 85 (CA DC Cir) (intimating willingness to extend the principle to prior sexual misconduct involving the accused and other persons), cert den, 322 US 762, 88 L ed 1589, 64 S Ct 1274. However, we are not required in this opinion to determine the status of Mr. Kinniry’s testimony with respect to homosexual associations.
VI
When we weigh the error implicit in the reception of segments of Mr. Kin-niry’s testimony, we find Headnote 5 that the balance falls heavily on the side of prejudice. In such a hotly contested case — one in which the Government relied chiefly on a single witness, who was contradicted specifically by several persons called by the defense — we must conclude that the accused was harmed by evidence importing that, as a matter of scientific necessity, the prosecution’s key witness was telling the truth. The only remedy lies in a rehearing — and it is so ordered.
Chief Judge Quinn and Judge Lati-mee concur.