Opinion of the Court
GEORGE W. LATIMER, Judge:I
Despite his plea to the contrary, the accused was found guilty by a general court-martial of unpremeditated murder, in violation of Article 118, Uniform Code of Military Justice, SO USC § 712; robbery, in violation of Article 122, Uniform Code of Military Justice, 50 USC § 716; and three offenses of aggravated assault, in violation of Article 128, Uniform Code of Military Justice, 50 USC § 722. He was sentenced to dishonorable discharge, total forfeitures, and confinement at hard labor for life. The convening authority approved but the board of review set aside the proceedings, holding that the court was appointed by an officer having no legal authority to convene a general court-martial. The Judge Advocate General of the Navy certified that question to this Court, and we reversed the board of review. United States v. Bunting, 4 USCMA 84, 15 CMR 84. In our disposition of that certificate, we returned the record to the board of review and directed that appellate tribunal to consider the appeal on its merits. Pursuant to this mandate, a majority of the board concluded that as a matter of fact, and after consideration of all the evidence in the record, they had a reasonable doubt as to the sanity of the accused at the time of the commission of the offenses. Accordingly, the board set aside the findings and sentence, and dismissed the charges. The Judge Advocate General of the Navy then certified to this Court the following questions:
“(a) Did the Board of Review err, as a matter of law, in its analysis of the testimony and thereby abuse its discretion in reversing the findings of the court-martial?
“(b) When the accused, upon the advice of counsel, refuses to talk to doctors appointed by the convening authority to inquire into his mental condition, pursuant to 121,122, MCM, 1951, and these doctors subsequently testify to their opinion that the accused was legally sane at the time of the offenses, whereas the accused fully *173cooperates over an extended period of time with psychiatrists who subsequently testify that in their opinion he was legally insane when he committed the offenses, what standard shall guide a Board of Review as to the weight to assign their respective conflicting opinions?”
II
The facts of the tragedy itself need not long detain us, for they were stipulated at trial. The sole issue there— and it was bitterly contested — was the mental responsibility of the accused. However, certain of the bizarre acts of the accused cast light on the question to be decided, and so a recitation of some of them is in order. At about 1:00 p.m. on January 12, 1952, the accused and several companions took liberty leave from their ship, which was docked at Yokosuka, Japan. During the course of the next several hours, the accused consumed five or six glasses of beer at several local bars. A good-natured dispute arose between him and a Navy companion, but it soon became rather heated. To prevent further difficulty, the accused withdrew, and went on his way alone. At about 6:00 p.m. on that same day, he appeared at a Japanese police box on a metropolitan street in Yokosuka, and mumbled something inaudible to the patrolman on duty there. He appeared confused, but not intoxicated, and he scribbled down the name of his ship for the policeman. While the officer was attempting to decipher this written communication, the accused suddenly struck him several times, overpowered him, and appropriated his revolver and ammunition. The accused, moving in a very deliberate manner, then walked away from the police box. As he proceeded down the street, he broke out the headlights of oncoming bicycles, struck a person with the revolver he had taken, shot and seriously wounded two others, and terminated his activities by killing a second policeman whom he met on the street. This last fatal shot was fired at close range, and struck the deceased full in the face. No provocation or motive was shown to account for any of these acts, and the accused did not attempt to hide or escape. He offered no resistance when he was apprehended a few moments after firing the last shot mentioned above. A medical examination made several hours after the shootings revealed that he was not intoxicated at that time.
The accused testified that after leaving his friends he proceeded to a serviceman’s club and — -save for an “island” of memory involving a fight with another Japanese man — remembered nothing thereafter until he was searched by the military police following the offense.
The remainder of the evidence concerned itself with the sanity of the accused at the time of the offense, and, so far as it is pertinent, will be related later in our discussion.
Ill
We have previously considered the power granted to us by the Uniform Code of Military Justice to interfere with a decision of a board of review, and we have suggested certain limits beyond which our authority does not run. In United States v. Zimmerman, 2 USCMA 12, 6 CMR 12, we recognized our power to review questions of law resolved by a board of review by saying:
. . Article 67(d), supra, [of the Code] explicitly gives to the Court power to act with respect to findings and sentence which have been ‘set aside as incorrect in law’ by a board of review. A board of review decision clearly based on matter of law, therefore, does not possess such finality that it may be assimilated to court-martial findings of not guilty.”
It is implicit in the grant of authority found in Article 67 of the Code that a board of review may not permissibly defeat review in this Court by labeling a matter of law, or a mixed holding of law and fact, as a question of fact. To avoid that impasse, we look to the substance of the holding, and its rationale, not to the characterization by the board of review. United States v. Benson, 3 USCMA 351, 12 CMR 107. Further*174more, we have consistently held that where a board of review makes a truly factual determination based upon the evidence of record, we may not overturn it. United States v. Thompson, 2 USCMA 460, 462, 9 CMR 90. When faced with such a determination, we are free only to insist that the holding, if it affirms a finding, be based upon substantial evidence, United States v. Hernandez, 4 USCMA 465, 16 CMR 39, and that the board of review express clearly the exercise of its fact-finding powers. United States v. Sell, 3 USCMA 202, 11 CMR 202; United States v. Moreno, 5 USCMA 500, 18 CMR 124.
As an offshoot to the foregoing principle is the precise concept which is presented by this certificate. Here the board of review found that the evidence in the record was insufficient to establish the sanity of the accused beyond a reasonable doubt. The nature of this finding does not permit us to say accurately that the evidence is sufficient to support the finding, for to do so would suggest that we are placing a burden on the accused to establish his insanity. The test we view as more appropriate in this instance may be found in this reasoning: On the one hand, if all reasonable men would conclude that the Government had established sanity beyond a reasonable doubt, then as a matter of law the board of review erred. On the other hand, if all reasonable men would conclude that the accused was insane, a holding to that effect by a board of review would be untouchable. Between the two extremes there exists an area where reasonable minds would differ as to whether the Government had established its burden beyond a reasonable doubt. If the facts place the issue in that area, then the board of review may resolve the conflict either way — in the exercise of its fact-finding powers — without abusing its discretion. Obviously, intertwined within the board’s power to resolve disputed questions of fact is its authority to believe or disbelieve witnesses.
As we read the board of review decision which is before us, we find nothing more than a factual determination of an evidential dispute. The issue of the sanity of this accused was one of fact, not law, and as noted by a majority of the board, the medical testimony concerning his mental responsibility was conflicting. The prosecution relied primarily upon the testimony of Colonel Donald Peterson, United States Army, the neuropsychiatric consultant for the Far East Command. The principal medical expert who testified for the defense was Commander Lebensohn, United States Naval Reserve, who had been recalled to active duty solely to undertake a psychiatric evaluation of the accused. Both possessed formidable qualifications as expert witnesses. Each expressed a great deal of respect for the other’s opinion. They agreed that the accused suffered from a genuine amnesia during the period of his berserk behavior. They agreed that he was mentally competent to cooperate intelligently in his own defense at trial. They both believed that he probably suffered from a dissociative reaction precipitated by alcohol at the time of the shooting. Dr. Lebensohn was of the opinion that this dissociative reaction had reached psychotic proportions at the time of importance here, and that the accused was legally insane. Dr. Peterson believed that the condition had not progressed so far as to become a state of psychosis, and that the accused was mentally responsible for his acts. Numerous other medical experts testified at the trial, but in the end, the court-martial chose to rely upon the testimony of the prosecution witnesses. Honest men could have arrived at a different result and that in fact happened. A member of the court obj'ected to the law officer’s ruling denying a motion for a finding of not guilty based on insanity and a vote was taken by the court on the ruling. In addition, the members of the board of review announced irreconcilable views on the issue. Two members of the reviewing board concluded the Government had failed to prove sanity beyond a reasonable doubt, while one member believed otherwise.
We have been faced with some cases *175in which we were doubtful of the power a board of review was seeking to exercise when it reversed a court-martial on a finding of guilt, but we are morally certain in this instance that the board was attempting to do the very thing Congress authorized it to do. The evidence for both the Government and the defense was paired off, compared and evaluated. The experience and training of the experts were considered and their qualifications measured. A conscious effort was made to balance the evidence in the pans of scales weighted by reasonable doubt. When the scales showed the evidence was wanting in that quality and quantity which would require all reasonable men to conclude the accused was sane, a majority of the board refused to find the accused sane. The reasoning indulged in by the majority to support their conclusion can be extracted from the following excerpts quoted from their decision:
“It is clearly seen that the medical testimony consisting in part of a number of outstanding psychiatrists is in conflict as to the issue of insanity at the time of the commission of the offenses. The prosecution experts were of the opinion that the accused was mentally responsible as his condition was simple amnesia or a dissociated reaction — that is, a retrograde amnesia in that during the performance of an anti-social act, the individual has memory continuity but at a certain time after the period in question, he develops a protective amnesia for that period. The defense experts were of the opinion that the accused was not mentally responsible as he was suffering from a dissociative reaction which had reached psychotic proportions — responding to unconscious forces beyond his control at the time of the offense.
“It is thus seen that the experts, no less than ordinary mortals are not infallible, even in their chosen fields, and where, as here, the opinions of experts are in hopeless, irreconcilable conflict, the absence of infallibility becomes manifest. We, like the trial court, are not required to accept the opinions of experts in preference to others but should judge the weight of the testimony of each witness, including that of the experts, and determine for ourselves the issue of mental responsibility at the time of the commission of the offenses.
“While we readily acknowledge that the reasons and the conclusions of some of the expert witnesses have influenced our views upon the issue of mental responsibility, our determination that the prosecution did not prove, beyond a reasonable doubt, that the accused was mentally responsible at the time of the offenses, does not rest conclusively upon the opinions expressed by these experts, but is arrived at on the basis of the whole evidence. This includes the opinions and reasons therefore of the experts, the presumption of sanity, the burden of proof of sanity in issue, and the evidence of the witnesses who testified on the issue of sanity and on the commission of the offenses.”
The dissenting member used the same mental processes but he arrived at a different conclusion. His views are supportable from the record but his position only establishes that the board of review stayed within its bounds. The area of dispute was one of fact and our views as to which is the more tenable position are unimportant. In that connection it should be noted that the facts of the substantive offenses were admitted. There was credible evidence that the accused was sane and there was believable evidence that he was insane. The board of review, by express Congressional mandate, may weigh evidence, judge the credibility of witnesses and determine controverted questions of fact. Had it expressly found the accused sane, the record would sustain the finding. On the contrary, had it found the accused insane, the same situation would exist. The majority made a conscious and informed choice in that area, and surely if the evidence would sustain a finding of insanity, a fortiori, it supports a holding that the Government had not established sanity beyond a reasonable doubt.
Government counsel contend, how*176ever, that the board of review had a duty to request that the accused be given another psychiatric examination, before taking action on the findings and sentence. Had the board done so, we are told, the members would have learned that the accused is presently mentally competent in all respects. We have previously held that it is proper for a board of review to take additional testimony when the sanity of an accused is in issue and the board believes further inquiry would be helpful. United States v. Burns, 2 USCMA 400, 9 CMR 30. But we there expressed the caveat that a board of review need not launch into an independent investigation if the issue was fully litigated at the trial level. Within the confines of this record may be found the testimony of eleven psychiatrists and psychologists, all of whom furnished relevant testimony. Without some showing to the contrary, it must be concluded that the issue was fully developed at trial and that another medical board would be able to offer nothing but cumulative testimony. The degree of normality exhibited by the accused at the present time is of small consequence, for it was agreed at trial that the accused was then competent, and Commander Leben-sohn alone believed the possibility was strong that his temporary loss of mental stability would recur.
It is our conclusion, then, that the board of review did not err, as a matter of law, in concluding that the evidence did not establish mental responsibility at the time of this occurrence beyond a reasonable doubt, and in not requiring a further mental examination.
IY
We are not sure of just what is requested of us by the second certified question. It should be accepted as a basic principle of military law that when board members undei’take to make findings of fact they should give full and fair consideration to all of the evidence properly before them. They should consider the inherent probability or improbability of the evidence presented; the credibility of witnesses; their — the witnesses’ — interest or bias; and the opportunity open to the witnesses to know or determine the true facts. So far as we know, the board members considered those factors.
Following the convening of the court-martial, and the reception of evidence relative to the legal sanity of the accused, a continuance was taken to permit further inquiry into the matter. When the court-martial reconvened, it developed that during the interval the accused was brought before a medical board. Upon the advice of individual military counsel, who was at the time physically unavailable, the accused refused to talk to the medical board members. He, however, was a little more generous with those favorable to his contention as he cooperated to the best of his ability with the psychiatrists relied upon by him at trial. Nevertheless, those experts who could not question the accused examined the material available to them and concluded that he was legally competent at the time of the offense.
The Manual for Courts-Martial, United States, 1951, paragraph 151c (2), page 287, provides that in military law:
“(2) Communications to medical officers and civilian physicians. — It is the duty of medical officers to attend sick members of the armed forces, to make periodical physical examinations as required by regulations and to examine persons for enlistment, and medical officers may be specially directed to observe, examine, or attend a member of the armed forces. Such observation, examination, or attendance would be official and the information thereby acquired would be official. Although the ethics of the medical profession forbid medical officers and civilian physicians to disclose without authority information acquired when acting in a professional capacity, no privilege attaches to such information or to statements made to them by patients.”
Thus, as we observed in United States v. Biesak, 3 USCMA 714, 14 CMR 132, *177the question of “Whether he [the accused] may be compelled to answer questions, and to give information necessary for a psychiatric inquiry, is veiled in uncertainty.” Certain civilian eases have held that a doctor who does not treat a prisoner, but only examines him in order to testify about his condition, is not barred from testifying by any sort of privilege. Catoe v. United States, 131 F2d 16, (CA DC Cir) (1942); People v. Sliney, 137 NY 570, 33 NE 150, 154 (1893); Commonwealth v. Di Stasio, 294 Mass 273, 1 NE2d 189, 195 (1936). But in the Catoe case, supra, the defendant was warned about making statements against interest, and it, therefore, is not authority for the proposition that the accused there could have been compelled to speak.
Article 31 of the Uniform Code provides :
“Art. 31. Compulsory self-incrimination prohibited.
“(a) No person subject to this code shall compel any perspn to incriminate himself or to answer any question the answer to which may tend to incriminate him.
“(b) No person subject to this code shall interrogate, or request any statement from, an accused or a person suspected of an offense without first informing him of the nature of the accusation and advising him that he does not have to make any statement regarding the offense of which he is accused or suspected and that any statement made by him may be used as evidence against him in a trial by court-martial.
“(d) No statement obtained from any person in violation of this article, or through the use of coercion, unlawful influence, or unlawful inducement shall be received in evidence against him in a trial by court-martial.”
The Army and the Air Force have interpreted that provision, and I believe rightly so, to require that the accused be warned if he is suspected of an offense and the evidence is to be used by a court-martial. In TM 8-240, AFM 160-42, Psychiatry in Military Law, May 1953, I find the following:
“Before starting his examination, the medical officer conducting the psychiatric examination should make clear to the accused the scope and purpose of the examination. The patient should be informed that the examiner is impartial and is neither ‘for him’ nor ‘against him’; that the accused need not answer any questions ; that anything he does say may be used as evidence against him (UCMJ Art 31); and that the examiner is taking notes which may be used as a basis for an opinion during a court martial.” [Paragraph 20/.]
While the Naval Service may not have adopted the provisions of that Manual, Article 31 of the Code denies the use of evidence obtained without a warning, or that obtained by compulsion.
My brothers prefer not to pass upon this problem of warning, believing that it is not squarely before us. For the purposes of this case, it is sufficient to say that had the Government considered that its experts were insufficiently informed, after the accused had taken the stand and testified to his amnesia, any information relevant to that condition might have been obtained on cross-examination.
Often, it is true, a personal interview with the accused is essential to the formulation of an opinion as to his sanity, and the lack of such an interview is a factor which may militate against the conclusion reached by the medical examiner. But the record here discloses accused’s refusal and we have no reason to assume that the majority members of the board of review did not consider that factor. Furthermore, the Government does not complain that its experts could not form an opinion because of insufficient information, as they testified the accused was sane. Because the accused claimed a loss of memory for the episode itself, Dr. Peterson testified he was not handicapped by the lack of a personal interview. In light of the fact that the expert witnesses had before *178them a complete personal history of the accused obtained from his parents and friends; the statements of the witnesses to the assaults and killing; information received from service friends of the accused; the results of psychiatric tests willingly taken by the accused; and the statements which he had previously made to other psychiatrists, it is unlikely that the board of review discounted their testimony. Even if that were not so, we cannot direct the weight that a board of review should give to the testimony of witnesses taken under different conditions. Too many imponderables clog that path.
For the foregoing reasons, the first certified question is answered in the negative. The second question is answered as indicated in the opinion. The decision of the board of review is affirmed.