(dissenting):
Because this opinion is controlled by *215the principles announced in United States v. Wilmer Keith (No. 503), 4 CMR 85, decided July 30, 1952, and United States v. McConnell (No. 596), 4 CMR 100, decided July 31, 1952, and other allied cases, and because I did not participate in those cases, I consider it appropriate to express my views on the concepts announced therein. In doing so, I shall first voice my reasons for being critical of the rule announced by the Court and then- proceed to set out generally the principles which I believe necessary for a proper solution of the issues involved.
I have watched the doctrine of “general prejudice” develop and it has followed a pattern of first being a hint, then being the subject matter of dicta and finally becoming the rationale of a series of decisions — decisions which I believe are calculated, however unwittingly, to confuse and muddle the administration of military justice. Not only are those decisions fraught with the likelihood of undesirable consequences, they ignore the plain wording of the Uniform Code of Military Justice, 50 USC §§ 551-736, and the experience gained by the Federal civilian courts over their full period of existence.
The foregoing observations are not intended to convey the impression that military justice will come to an end because of the concepts announced in the particular line of cases, but they are intended as a caveat to the effect that it will be extremely difficult for courts-martial, convening authorities, and boards of review to administer the Act, and, that unless opinions of this Court announce rules of law which can be applied practically by tribunals and other persons who are required to follow our decisions, its usefulness will be impaired. I suggest that persons administering the Code will always encounter some difficulty in testing whether an accused’s “substantial rights have been materially prejudiced” in a particular case, but, when this rule is discarded and another announced which is so incomprehensible and so lacking in identifiable standards that it can be measured only by the individual whims of appellate judges, then a practical application of the principle is most difficult, if not impossible.
I appreciate the fact that merely stating conclusions proves nothing, but to substantiate my charge that the rule laid down lacks discernible standards by which it may be measured, I quote from a decision by Judge Brosman in the case of United States v. Lee (No. 200), 2 CMR 118, decided March 13, 1952, as later approved in United States v. Berry (No. 69), 2 CMR 141, decided March 18, 1952, which were the decisions giving birth to “general prejudice.” In attempting to define his concept Judge Brosman stated:
“. . . . We have in mind here a situation in which the error consists not in a violation of constitutional or legislative provisions, but involves instead an overt departure from some ‘creative and indwelling principle’ — some critical and basic norm operative in the area under consideration.”
I call attention to the fact that the Court in defining general prejudice in the quoted decision disclaims any intent to be dealing with violations of a constitutional or legislative provision. This disclaimer should be considered as somewhat modified by what the Court said in United States v. Lee, supra. I quote from that decision:
“It has been suggested, however, that departure from a specific requirement of the Manual should, of itself, constitute ‘general prejudice’ of the type we are here discussing. We have no doubt that this may be true in a proper case if, but only if, the policy underlying the Manual requirement is so overwhelmingly important in the scheme of military justice as to elevate it to the level of a ‘creative and indwelling principle.’
Violations of Congressional mandates would be easy to identify and relatively simple to deal with, but to escape the necessity of erecting well-defined guideposts to fix the boundaries of the rule with some degree of certainty the language of the opinions seems to contemplate an imaginary area of unlim*216ited breadth in which this Court may operate to reverse convictions regardless of Congressional limitations. This suggests an assumption of power by this Court, undisclosed, undefined and uncontrolled, to deal with any case in which we desire as a matter of policy or administration, to reach a desired result, even to the extent of reversing a ease for purely punitive purposes. If this interpretation is warranted by the decisions, and this I leave to the readers of the opinions for determination, then I believe the Court has stepped out of character, in that our duty is not to administer the lash but rather to determine whether in á particular case an accused has been justly tried, convicted and sentenced. If he has, we should affirm. If he has not, we should take corrective action. Surely a guilty person should not be rewarded with a new trial because we desire to reprimand a law officer for an infraction of the Code when the accused is the beneficiary of the error. Law may not be logic but it ought not to be that illogical.
Perhaps a Court in its infancy and when forming a legal structure out of a newly enacted Code should be extraordinarily vigilant in not countenancing erroneous practices but in doing so it should not overlook the cardinal principle announced by Chief Justice John W. Marshall in Osborn v. Bank of United States, 9 Wheat 738, 22 US 738, 6 L ed 204. In that case he stated at page 866:
. . . Courts are the mere instruments of the law and can will nothing. . . . Judicial power is never exercised for the purpose of giving effect to the will of the judge; always for the purpose of giving effect to the will of the legislature; or, in other words, to the will of law.”
The rule announced in that case seems to me to be sound for us to follow, but let me illustrate how it has been disregarded. In doing so I shall columnize Article 59(a) of the Uniform Code of Military Justice, 50 USC § 646, and a paragraph taken from United States v. Wilmer Keith, supra:
Uniform Code of Military Justice
“ART. 59. Error of law; lesser included offense.
(a) A finding or sentence of a court-martial shall not be held incorrect on the ground of an error of law unless the error materially prejudices the substantial rights of the accused.” [Emphasis supplied]
United States v. Keith
“. . . . However, since we have based our resolution of the limitation imposed by Article 59a, supra, upon the concept of general prejudice, it is unnecessary to assess in detail the possibility or probability of specific and individual prejudice in this case. . . . .” [Emphasis supplied]
My interpretation of Article 59 (a) of the Uniform Code is that Congress willed that we must search the record in the particular case to determine whether any substantial right of the accused has been materially prejudiced and that the finding and sentence shall stand unless we detect prejudicial error. The opinion announces that regardless of the Congressional expression, it is the will of the Court that we will not perform any such task. It would appear to me to be the height of inconsistency for us to criticize a law officer for his failure to comply strictly with the provisions of the Code and in the same opinion announce boldly that we refuse to follow another provision which is so positive and clear as to be readily understandable by anyone subject to the Code.
I concede the Court attempts to predicate its holding on good administrative practice but, as I interpret the Act, if there is any authority in the Uniform Code of Military Justice for us to act in an administrative capacity, it can only be found in Article 67 (g), 50 USC § 654, which places the judges on a committee with the Judge Advocates General for the express purpose of recommending to Congress any corrective measures deemed appropriate. It is difficult for me to stretch that provision, any other one written in the Code or any principle of law far enough to find authority for us to lay down *217rules, even though they be characterized as administrative, contrary to the law as spelled out by Congress. While I might prefer to have Congress delegate additional powers to the Court, I do not think we can assume authority because it may appear that the Supreme Court of the United States has held it has authority to control the activities of Federal officers by excluding from evidence the fruits of their improper activities. See United States v. McNabb, 318 US 332, 87 L ed 819, 63 S Ct 608. Over the years the Supreme Court has carved out the rules for admission of evidence in the Federal courts but we are not dealing with the admissibility of evidence. However, assuming the cases were comparable, that Court was, by the Constitution of the United States, created as a separate and independent department of the Government, and in certain spheres of activity cannot be controlled by Congressional enactments. Operating within the sphere of Congressional control that Court may not ignore the expressed intent of Congress and neither should this Court. We were brought into existence by Congress and the powers we possess are only those granted by that branch of the Government. Accordingly, we must comply with its mandates. It announced in clear and unmistakable language that we should not reverse a conviction unless we could find in the record of that individual case some prejudice to the substantial rights of an accused. Laying aside a comparison of our powers with those of the Supreme Court, I believe we inflate our authority by arrogating unto ourselves the right to reverse cases without testing them by the only measuring rod given to us.
I have previously stated that the rule announced ignores the experience gained by the Federal civilian courts during their existence. We have repeatedly announced in our decisions that Congress intended to pattern the military-judicial system after the Federal criminal system. Moreover, we have continually cited Federal cases to fortify those pronouncements. In addition, the Uniform Code of Military Justice, supra, delegated to The President the right to promulgate rules and regulations governing military justice patterned after the Federal criminal procedure in so far as practicable and not contrary to the Code. In view of all this, I would be inclined to consider the holding in comparable Federal cases before summarily disposing of' the issue. For some reason the Court abandons that approach to the problem, refuses to follow its own precedents, refuses to be persuaded by the rationale of previous cases, and assigns as a reason that we must forcibly deal with violations because the Uniform Code revolutionized the former procedure.
The opinions concede that the United States Court of Appeals and the United States Supreme Court have passed on a similar question, i.e., a judge discussing a case with a jury or jurors in the absence of a defendant. Those authorities have generally held that when such a transgression has occurred the case would not be reversed if it appears the discussions did not result to the prejudice of the defendant. The various opinions written by this Court on this particular problem acknowledge the civilian rule but announce the reasoning of the opinions of other courts would not be evaluated because Congress emphatically decreed that the law officer should be prohibited from communicating with the court in the absence of the accused. I do not find the language of the Act to require such a sweeping prohibition. True, the Code provides that certain things are to be done by the law officer only in the presence of the accused, but it likewise provides for him to advise with the court, in the absence of the accused, to assist in the preparation of the findings. The Act does not spell out the same complete separation as is required between the judge and the jury in civilian practice. On the contrary, it has the tendency to throw the law officer and the court together in the absence of the accused, and, in practically every instance, the discussions between them upon which this Court predicates reversal arose out of a misapplication of *218that provision. Had Congress commanded the positive separation pictured by the Court, it failed to so state in the Act. I venture to suggest there would be few, if any, violations had the Code not authorized some secret consultations. It should be obvious that in some instances incidental matters may creep into a discussion on the form of finding without any intent on the part of either the Court or law officer to violate the prescriptions of the Code. Under such circumstances rather than refuse to follow concepts which have been considered and accepted by other courts with years of experience, I would take the principles announced and apply them to the peculiarities of the military system, realizing full well that there are requirements in the military law which make it mandatory that law officers discuss certain matters with the court.
1 will not belabor this opinion with a detailed analysis of the many Federal cases dealing with this subject. Suffice it to say they hold that unless a defendant is prejudiced by a communication between the judge and jury no reversal will result. I subscribe to the reasoning and quote from the language of Judge Learned Hand in the case of United States v. Compagna, 146 F2d 524 (1944), certiorari denied, 324 US 867, 89 L ed 1422, 1423, 65 S Ct 912, 913. In that case the trial judge conversed with the jury in the absence of the defendant, and the opinion states the rule as follows:
“. . . . As to the visit of the judge, it is true that courts are extremely jealous of anything of the kind, once the jury has been locked up; and we do not wish to abate that jealousy in the least; it is most undesirable that anything should reach a jury which does not do so in the court room. This is, indeed, too well settled for debate. Mattox v. United States, 146 US 140, 150, 13 S Ct 50, 36 L ed 917; Fillippon v. Albion Vein Slate Co., 250 US 76, 81, 39 S Ct 435, 63 L ed 853; Dodge v. United States, 2 Cir., 258 F 300, 304, 7 ALR 1510; Little v. United States, 10 Cir., 73 F2d 861, 864, 865, 96 ALR 889. But, like other rules for the conduct of trials, it is not an end in itself; and, while lapses should be closely scrutinized, when it appears with certainty that no harm has been done, it would be the merest pedantry to insist upon procedural regularity. Dodge v. United States, supra, 258 F 300, 7 ALR 1510; Rice v. United States, 2 Cir., 35 F2d 689, 696; United States v. Graham, 2 Cir., 102 F2d 436, 444. There cannot be the slightest doubt here that the informality— for, at most, it was no more — did not prejudice the accused.”
Another concept which is announced in the Keith and McConnell cases, supra, and which must be peculiar to this Court is' that the condemned acts of the law officer and the court are presently considered horrendous but in the not too distant future they may be viewed as only minor errors. I am at a loss to understand how, and in what way, such acts could, at this time, be such a devastating violation of some substantial right, privilege or administrative regulation and of sufficient gravity to require a reversal when military personnel are unfamiliar with administering a newly enacted piece of legislation, and yet when time has permitted them to become fully acquainted with the law, the same error will be of lesser importance. The Court seems to anticipate it may cast aside gross and flagrant errors in the future. If the rationale of the opinion, that some overt departure from an indwelling principle has occurred, is founded in substance, the Court should never lessen its vigilance in detecting similar errors nor cast aside the present reasons for the reversals. When a system is new and personnel are not entirely familiar with its operations, we should be inclined to be more charitable than we ought to be after experience has been gained. Certainly, if I believed I had the power to administer the lash I would not apply it until I could reasonably believe the violation was not occasioned by something more than lapses which are bound to occur in a newly enacted criminal code. However, I have a concept that the public has a definite interest in the *219preservation of law and order in the military service, that there are two or more parties who have interests to be protected, and that we should limit our review of a case to see that the substantial rights of all parties are considered. It is singular to note, and I will later deal with this specifically, that in most of the decisions reasons are not advanced that any accused was denied a fair and impartial trial or that he was unjustly convicted. To the contrary, in practically every cause, the evidence of guilt was overwhelming and the error was post-findings. This causes me to stop and ponder whether the Court has not overlooked the fact that criminal prosecutions are not unilateral contests.
One fundamental consideration involved in an appellate court rendering written decisions is to announce principles of law which can be used by inferior tribunals to determine subsequent cases. If any court-martial, staff judge advocate, convening authority or board of review can obtain a working rule out of these decisions, it would be that until such time as this Court gives the all-clear signal, every case in which a law member converses with the court, out of the presence of the accused, must be reversed regardless of how innocuous the conversation. Maybe this can be characterized as “temporary general prejudice.” But consider how illogical and unworkable this rule is. Boards of review and convening authorities are ordered by Congress not to reverse a case unless they find substantial prejudice. Is it intended that they, too, adopt the theory of “general prejudice” and reverse a case when the error benefits the accused? Just how we can expect any tribunal to accept and follow such a doctrine is beyond my comprehension. That is asking a military tribunal to disregard the express provisions of the Code.
Maybe a reference to some of our cases will be helpful in illustrating how the inconsistent rulings of this Court have a tendency to confuse the other tribunals. I do not see how we can expect others to anticipate the error the Court will consider - of sufficient gravity to be covered by the cloak of general prejudice when I, myself, am unable to limit its scope. As a preface to this, it is well to again mention that the error consisted of violating a provision of the Code. In this particular instance the law officer offended. He likewise did that in United States v. Lucas, (No. 7) 1 CMR 19, decided November 8, 1951, and allied cases, in that there he failed to instruct the Court on the essential elements of the offense, after a plea of guilty, and further failed to require a secret ballot on findings as required by the Code. We affirmed those findings of guilty because of no prejudice. In United States v. May (No. 241) 2 CMR 80, decided February 13, 1952, and United States v. Marcy (No. 260) 2 CMR 82, decided February 13, 1952, we held that the trial of the accused on charges which have not been sworn to was contrary to the Manual but the violation was not prejudicial. In United States v. Hutchison (No. 425) 3 CMR 25, decided April 9, 1952, we were faced with the issue of a Code violation on failure to appoint a qualified assistant defense counsel. Again, we called attention to the error but searched the record and found no prejudice. In United States v. Bartholomew (No. 166) 3 CMR 41, decided April 16, 1952, we found a violation of the spirit of the Code in not appointing defense counsel with qualifications equal to those of trial counsel. In that decision we held there was no prejudice. In United States v. Benjamin Jones (No. 79) 3 CMR 36, decided April 14, 1952, the court-martial usurped the powers of the law officer but we reaffirmed the finding of guilt. Chief Judge Quinn held there was no prejudice and the writer held that assuming there was error and prejudice, the latter was cured by the action of the convening authority and the board of review.
In United States v. Bound (No. 201) 2 CMR 130, decided March 13, 1952, an officer who sat as a member of the court was expressly disqualified by the Code and yet was permitted to participate in the findings of guilty and the imposition of the sentence. Conceding a guilty plea, the board of review believed the *220violation of the Code required a reversal. We affirmed the board of review but went to some length to establish, prejudice. In speaking of this, the opinion states:
“It seems to us that the mandate of this statutory directive is clear. We are.not to reverse for error of law unless we are of opinion on the basis of the proceedings in their entirety that the substantial rights of the accused have been prejudicially affected. It also is clear to us that the policy of Article 59 (a) is highly salutary and as applicable to military as to civilian criminal law administration — perhaps in essence more so.
I wonder why in these cases that directive is so unimportant?
In United States v. Doyle (No. 265) 4 CMR 174, decided May 20, 1952, we stated:
“. . . In several instances, we think the president unduly, restricted the scope of defense cross-examination. We have weighed these errors carefully against other and proper evidence of guilt. The compelling nature of the prosecution case,, aside from that testimony affected by the errors noted above, persuades us that it is neither necessary nor desirable to set aside this conviction. We are reminded that an appellate court is not and should not be concerned with emphasizing procedural errors to the point where an otherwise' guilty defendant is permitted to escape .the penalties properly .exacted by society for his offenses. While not condoning the departures from proper rules of evidence which this record discloses, it is our. considered judgment that they were not substantially prejudicial to the accused.”
In United States v. Valencia (No. 308) 4 CMR 7, decided June 3, 1952, the Court considered the misconduct of trial counsel. We reviewed the facts and determined that his acts were not deliberate and did not amount to misconduct. We did, however, announce this concept:
“It follows from what we have said above that there is substantial evidence to support the conviction. Further, the evidence is clear • and convincing. Therefore, even if the statements of trial counsel in relation to other crime constituted misconduct, an over-all view of the case indicates that it did not result in substantial prejudice to petitioner. . . .”
In United States v. Nash (No. 447) 4 CMR 130, decided August 7, 1952, the trial judge advocate under the previous provisions of Naval Courts and Boards stepped out of character, misinformed the Court and argued erroneous rules of law. We specifically searched the record and found prejudice. Based on this we reversed.
In eases involving inadequate, misleading instructions, errors on admission of evidence and other substantial errors encountered in the trial of criminal cases, we have considered the prejudicial impact the error had on the findings or sentence.
The foregoing cases all deal with substantial rights of an accused and yet we isolate them from the doctrine of general prejudice. This poses the question of when should a case be tested for specific prejudice and when should the visionary rule of general prejudice be the compelling motive for reversal. I know not why we abandon what I choose to label a common sense approach to the problem but in the recent cases general prejudice is the line of attack used to set aside the findings and sentence. The first case to be reversed for error in a conference between the law officer and the court was United States v. Wilmer Keith, supra. There, the law officer, after the findings of guilty had been announced, in open court, was called into a closed session with the court-martial while it was deliberating upon the sentence. The discussion concerned the difference between a bad-conduct discharge and a dishonorable discharge. The board of review noted the infraction of the provisions of the Code and Manual, but held, one member dissenting, that it was not prejudicial to the substantial rights of the accused, since the findings of guilty had been announced and since it was evident *221from the type of information requested by the court-martial that it had already determined to impose one of the types of discharge, and in view of the fact that it had imposed a bad-conduct rather than a dishonorable discharge, the result of the discussion was apparently favorable to the accused. The Court reversed not only the sentence but the findings of guilt which were not touched by the discussion.
United States v. McConnell, supra, also involved a post-finding conference and the discussion concerned only the appropriate forfeiture of pay. The question of whether this conference constituted prejudicial error either as to findings or sentence was not raised at the trial level, before the board of review nor by civilian counsel for the accused on appeal so it must not have been such a flagrant violation as to shock members of the bar. There again we not only set aside the sentence, we reversed the findings.
In United States v. Wingert, (No. 785) 4 CMR 166, decided August 8, 1952, the accused had been tried in common trial with McConnell and the same situation was involved. This Court, by per curiam opinion, again reversed, basing its decision upon the principle enunciated in the Keith and McConnell cases. A post-finding error concerning only forfeiture of pay was considered so important that a prior legal finding of guilt was reversed.
In United States v. Henry Smith (No. 512), 4 CMR 123, decided August 6, 1952, the Court reversed the board of review by per curiam opinion. There, the consultation between the law officer and the court-martial occurred prior to the finding, but it concerned the law officer’s former ruling excluding a dying declaration of the homicide victim which the prosecution had sought to have admitted in evidence during the trial. The entire proceedings were reported and reveal that the action of the law officer was most favorable to the accused. The discussion involved one specification of murder and the accused was acquitted of that charge. He was, however, convicted on another. I believe I can fairly say that the error of the law officer saved the accused from a murder conviction. Nevertheless, this Court, in disposing of the case on review, reversed.
A similar situation is presented by the case now before the Court. Here, petitioners, together with another accused, were tried in common and convicted of rape. The evidence clearly established the commission of the offense by all three accused. The victim, a Korean national, testified that she was dragged into a cave, beaten, and forcibly raped by each of them. Petitioners and the other accused, Simmons, testified at the trial, and pre-trial statements signed by each of them were received in evidence. While their versions as to the degree of force employed by them differed to some extent from the story told by the victim, each of them admitted having sexual intercourse with her and that some force was used. After all evidence was received, and after proper and adequate instructions duly given to the court-martial members in the presence of the petitioners and-their counsel, the court-martial returned a finding that the accused were guilty as charged. Their findings were announced in open court and they were then instructed that the maximum punishment for the offenses was the death penalty. Thereafter they retired and the law officer was called in. The question asked him concerned only whether a discharge carried an- automatic forfeiture. A short time later the court was reopened and the sentences were pronounced. Each of the petitioners was sentenced to receive a dishonorable discharge, to forfeit all pay and allowances, and to be confined at hard labor for fifteen years. Because of the forfeiture discussion, the findings of guilt are reversed, a result which ought not to be reached.
I hold no brief for the offending law officers but I do not believe it is necessary to reverse valid convictions to eliminate any probability of command control if that is a consideration in these holdings. There are other legal and less shocking ways. If there is any showing that the transcript does not *222contain the proceedings in toto, or does not record all of the discussions between the law officer and members of the court, then the findings or the sentence should be reversed. But when the reporter is present and transcribes the complete proceedings, I do not condemn merely because of the possibility of prejudice based on the suspicion that the record does not fairly show all. I would prefer to assume contrariwise, that the reporter, members of the court and law officer are consciously and fairly trying to preserve a true record of the proceedings. I would grant to them the same good faith I would give to corresponding members of the civilian criminal system. Were they inclined to prejudice an accused, the conference on findings permitted by the Code offers them adequate opportunity.
Coming at this late date the only useful purpose I can accomplish by this dissenting opinion is to mark the path I would follow in disposing of similar errors in subsequent cases. I cannot know when, if ever, the rule will be relaxed and I cannot assure others who administer the Code that the Court will ultimately adopt these concepts. I do, however, propose the following as my recommended procedure. First, a law officer should religiously follow the Code mandate. Second, if perchance he errs, then I would divide the cases into two general categories, namely, those where the error crept into the case before a finding of guilt and those where the error affected only the sentence. I do not believe findings and sentence are so inexorably interwoven that error in respect to the latter necessarily impairs the former. In the first category I would adopt the following rule: if the record indicated the communication was prejudicial to the accused, I would reverse. If, on the contrary, it established that it was either to his benefit or did not prejudice materially his substantial rights, I would affirm. If the record was silent as to what transpired, I would presume it to be prejudicial to the accused and place the burden on the Government to establish that it was not to his detriment.
Substantially the same rule would be adopted if the communication took place after a finding of guilty. First, I should make it clear that I would not reverse the findings of guilt as the error could not possibly influence the court-martial in its deliberations on the guilt or innocence of the accused. The findings must be announced in open court before the court deliberates on sentence. At best, a post-finding error could only add to or detract from the severity of the sentence. In those instances, if the communication prejudiced the accused I would return the cause to the board of review for reconsideration of the sentence. If it did not, or if it was beneficial to him, I would affirm. Again, if the record was silent, I would return the record to the board of review to reconsider the sentence in the light of our holding. We have adopted a comparable practice when we find errors affecting a sentence. See United States v. Carter (No. 159), 2 CMR 14, decided January 18, 1952; United States v. Zimmerman (No. 261), 6 CMR 12, decided October 6, 1952; and United States v. Bobby L. Keith (No. 226), 4 CMR 34, decided July 3, 1952.
In treating the after-finding error I attempt to parallel the civilian practice. There a cause can be returned to a judge for reconsideration of the sentence! This cannot be done satisfactorily in the military system but Congress has furnished an adequate substitute. A boax-d of review has power - to consider the entire record and approve so much of the sentence as it believes correct in law and fact. Any prejudice which the accused might suffer by improper consultation on the sentence could be cured by the board of review assessing properly' the prejudicial effect of the error.