(dissenting):
There is much in the principal opinion with which I fully agree. For example, I agree with the general de-^ *121marcation of the respective ■ powers of Congress and the President in the substantive and procedural areas of military justice. Unfortunately, however, I cannot accept the majority’s construction of the President’s exercise of one of his granted powers.
My first point of disagreement is with the meaning ascribed to the provision in the Manual for Courts-Martial, United States, 1951, on the extent of independent evidence required to corroborate a confession. Rather than begin with obeisance toward the rule of stare decisis, there is need for re-examination of fundamentals. As Judge Kilday ably points out, we appropriately start with the Manual. It recites the rule of independent corroboration in two sentences:
“An accused cannot legally be convicted upon his uncorroborated confession or admission. A court may not consider the confession or admission of an accused as evidence against him unless there is in the record other evidence, either direct or circumstantial, that the offense charged had probably been committed by someone.” [Paragraph 140a.]
Most citations to the meaning of this provision go back to United States v Isenberg, 2 USCMA 349, 8 CMR 149. There Judge Latimer comprehensively reviewed earlier military and civilian applications of the rule of corroboration. He concluded that a short absence without authority constituted insufficient independent evidence to corroborate a confession of desertion. The Isenberg decision was interpreted as “adopting, at least in substance, the rule announced in Forte v United States, 94 F2d 236 (CA DC Cir)” which requires “some evidence . . . [on] each element of the offense charged.” United States v Petty, 3 USCMA 87, 11 CMR 87. But neither Isenberg, nor its offspring United States v Villasenor, 6 USCMA 3, 19 CMR 129, purported to answer the question, “Why does the Manual require corroborative evidence as to each element of the offense charged?” As a matter of fact Isen-berg expressly disclaimed any need to inquire into the “intent behind the deletion” from the 1951 Manual of language appearing in the 1949 Manual to the effect that the corroborative evidence need not cover every element of the offense. It is, I believe, at this point that the present opinion and its predecessors, like Villasenor and United States v Mims, 8 USCMA 316, 24 CMR 126, go astray. They assume the Manual prescribes an inflexible rule of evidence, separate and distinct from the corroborative rule followed in the Federal courts. That assumption is, in my opinion, unfounded and erroneous.
The first case in which this Court considered the corroborative rule was United States v Brooks, 1 USCMA 88, 1 CMR 88. The case was tried before the effective date of the 1951 Manual, and therefore the trial was governed by the procedural rules in the 1949 Manual. It will be recalled that the discussion of the corroboration requirements in- the 1949 Manual indicated that the independent evidence did not need to extend to each element of the offense. Nevertheless, in a unanimous opinion, in United States v Brooks, we said there must be “substantial evidence of the corpus delicti other than the confession”; and we cited the Forte case to support the statement of the principle. How could we cite Forte in face of the 1949 provision that the evidence need not extend to each element of the offense? Was there a conflict between the 1949 Manual and Forte which we, sub silentio, resolved in favor of the latter? If so, were we wrong in applying the 1951 Manual to a trial begun under the 1949 procedures? Cf. United States v Sonnenschein, 1 USCMA 64, 1 CMR 64. Until the assumption of Villasenor, all these questions had consistent answers.
Underlying the decision in United States v Brooks, supra, and others to which I shall refer presently, is the idea that the Manual does not define a separate and distinct rule of evidence for military courts. Rather, it merely incorporates into military procedure the general rule prevailing in the regular Federal courts. This is the un-articulated but manifest premise in Brooks. It is also the basic premise in United States v Dolliole, 3 USCMA *122101, 11 CMR 101, which was decided after Isenberg. There, we said, the Manual provision merely follows “the prevailing American law on the subject”; and we stated the rule in terms of the “slightly different wording” we used in United States v Evans, 1 USCMA 207, 2 CMR 113, another case decided before Isenberg. Quoting from our Evans opinion we said:
. A general statement of the standard to be applied to a given crime is difficult to formulate and legally unnecessary. We shall be satisfied if, for any crime, there is substantial evidence which makes it probable that the accused did not confess to an offense which never occurred.”
Our reliance upon United States v Evans, supra, is significant because Evans, like Brooks before it, was tried while the 1949 Manual was still in effect. In its central purpose, therefore, Dolliole clearly contemplated continuity between the rule of corroboration described in the 1949 Manual and that set out in the 1951 Manual, despite differences in language between the two provisions. What was the nature of this continuity? The obvious answer is continuity of the intention that the Manual rule conform to the regular Federal rule.
Two circumstances convince me that the corroboration evidence discussion in the Manual does not promulgate a separate and inflexible rule of procedure for military courts, but merely intends, as we said in the Petty case, to adopt the prevailing Federal rule, whatever that rule happens to be. First, is the statement of intention by the draftsmen of the Manual. They say that the part of the Manual provision in issue “contains various rules of common application in the Federal courts.” Legal and Legislative Basis, Manual for Courts-Martial, United States, 1951, page 216. Next, is the text of the provision itself. It consists, as I noted earlier, of only two sentences. The first sentence says an accused “cannot legally be convicted upon his uncorroborated confession or admission.” That sentence is not the declaration of a procedural rule, but a statement of a fundamental rule of American substantive law. The second sentence says that the court-martial “may not consider the confession” unless there is independent evidence “that the offense charged had probably been committed by someone.” This provision has been construed as adopting, for military courts, the rule of corroboration then prevailing in the regular Federal courts. United States v Petty, supra; see also my dissent in United States v Uchihara, 1 USCMA 123, 2 CMR 29.
Until Opper v United States, 348 US 84, 99 L ed 101, 75 S Ct 158, it was generally agreed that the prevailing rule was that set out in the Forte case, which was decided by the United States Court of Appeals for the District of Columbia Circuit. In Opper, however, the Supreme Court held that the “better rule” does not require “independent evidence of the corpus delicti.” Consequently, since the Opper decision, the prevailing rule in the Federal courts does not require corroborative evidence of every element of the offense charged. Inasmuch as the Manual provision intends merely to incorporate into military practice the corroboration rule followed by the regular Federal courts, necessarily the Opper statement of the rule, and not that of the Forte ease, must be the one to which military courts should look.
As early as my dissent in United States v Uchihara, supra, I acknowledged that the Manual’s procedural statement in this specific area “is binding upon us.”1 But, the particular command of the Manual is a command *123only to follow the rule “of common application in the Federal courts.” Legal and Legislative Basis, Manual for Courts-Martial, 1951, page 216. That rule was settled by the United States Supreme Court in the Opper case. In conformity with the Manual provision, and with the dictate of Article 36, 10 USC § 836, that as far as practicable, the President shall prescribe “rules of evidence generally recognized” in the regular Federal courts, I would follow the Federal rule. See United States v Knudson, 4 USCMA 587, 16 CMR 161.
Although I disagree with the rule followed by the majority, even using that rule, the ineluctable conclusion is that the evidence is sufficient'to show the offense charged was probably committed.
We begin with the fact that a small child, six or seven years of age, was brought by a woman, who purported to be her mother, to the emergency room of the base hospital about 1:30 or 2:00 in the morning. This circumstance reasonably indicates that something of a sufficiently alarming nature had happened to the child to require immediate medical examination and treatment. We are not told directly what it was that happened to the child, but, in my opinion, the circumstantial evidence points unmistakably to the fact she was the victim of a sexual assault. Inasmuch as circumstantial evidence is sufficient to support a confession, it is patently sufficient to meet the rule requiring independent evidence of the probable commission of the offense charged. United States v Manausa, 12 USCMA 37, 30 CMR 37; United States v Young, 12 USCMA 211, 30 CMR 211.
The child was examined by three doctors. All testified she had bruises on her left cheek which resembled the shape of a human hand. In addition, there was a tear in the perineal area; there “was clotted blood in and around” the genital area; and there was a bruise around the anus. All the doctors agreed the injuries in the perineal area were probably caused by an external force, which one doctor said was “probably blunt in nature.” Dr. Burnell testified that the tear in the perineal area was “fresh” and still “moist” at the time of the initial examination. Similarly, Dr. Sweigard described the marks on the child’s cheek as “fresh bruise marks.” These descriptions of the several bruises strongly indicate they were probably inflicted at the same time. What then of the manner in which these injuries were incurred? Could the tear in the perineal area have come from a fall or some other innocent cause, as defense counsel argued at trial? In my opinion, the evidence shows they probably resulted from a criminal act.
If the cause of the injuries was innocent, is it reasonable to conclude this young child would be rushed to a hospital emergency room in the middle of a winter’s night? I think .not. The surface injuries were not so serious as to require emergency treatment. Common experience and common sense, therefore, indicate that the hospital visit was probably prompted by the knowledge on the part of the person who brought the child to the emergency room that she had been injured in an incident that was not innocent and not commonplace. What could this incident be?
I referred earlier to the imprint on the child’s cheek in the shape of a human hand. The imprint was still clearly observable several hours after the child’s initial examination. A blow sufficient to produce such a sharply defined mark for so long a period of time can reasonably be attributed to a conscious and deliberate battery, rather than to accident or some other innocent cause. The “fresh” nature of this mark of violence reasonably connects it to the “fresh” bruises in the genital area and indicates that all injuries were probably inflicted at the same time. And this connection between the injuries explains the emergency nature of the visit to the hospital. All the medical witnesses agreed that the perineal area injuries were compatible with the force*124ful impact of the male organ. The emergency visit to the hospital, the nature of the child’s injuries, and the apparent human agency that caused them are, in my opinion, sufficient to support the inference that the child was the victim of a criminal assault by a person intent upon gratifying his lust or desire. Probability of the commission of the offense charged is all the law requires. This probability is amply met by evidence independent of the accused’s confession. See United States v Fioco, 10 USCMA 198, 27 CMR 272; United States v Young, supra.
I would answer the certified question in the negative and return the record of trial to the board of review for further proceedings on the merits.
In United States v Villasenor, 6 USCMA 3, 19 CMR 129, I said I doubt “the rule requiring corroboration of a confession is a mere mode of proof.” This statement had reference to the basic rule of substantive law that a conviction cannot be had upon a con-fession alone, and was intended to point up my disagreement with the principal opinión which seemed to indicate that the Manual’s statement encompassed only a rule of procedure. As noted in the text above, the Manual’s statement on corroboration actually embraces the *123substantive principle and the procedural rule on the quantum of corroboration required. The distinction between the two is ably discussed in the principal opinion in this case. I agree with that discussion.