(dissenting):
I dissent.
While at first blush this may appear to be a case which requires drastic action to penalize illegal practices of agents of the Government, a proper interpretation of the facts establishes that such is not the situation. Contrary to my associates, I do not believe we are confronted with a failure to warn in contravention of Article 31. That Article requires warning to one suspected of an offense and part of the warning consists - of informing the offender of the offense of which he is suspected. In the case at bar, until the accused related his diabolical scheme, military authorities and the Central Intelligence Agency operated under the belief he was a credit to the service. He was being considered for advancement to a critical and sensitive job, and it was solely to determine his qualifications for promotion that the examination was given. I dare say that Federal authorities were completely taken by surprise at his disclosures for there would be no purpose in choosing him for advancement if he was suspected of an offense. Only a serviceman who is not under a cloud of suspicion is selected for this type of assignment and when that situation prevails I fail to see why a warning is required. And I further do not comprehend what crime the operator of a lie detector machine would use as the basis for a warning. To interpret Article 31 to require a warning under the facts of this case is to read into the Article requirements which are neither expressed nor implied.
I am willing, however to go one step further and assume arguendo that the confession was obtained illegally. But before advancing my reasons for holding the derivative evidence admissible, I prefer to make several general observations.
I am not so willing as my associates to hold that the Chief Judge’s language in United States v Fair, 2 USCMA 521, 10 CMR 19, does not announce a good principle of law. I concurred in that decision, and I do not believe the principle should be buried without a few words said in its behalf by at least one of its sponsors. Mr. Justice Owen Roberts, in Smith v Allwright, 321 US 649, 88 L ed 987, 64 S Ct 757 (1944), noting a tendency on the part of the Supreme Court to reverse previous decisions, made the following sage comment which I believe can be applied to our present trend. He said the rulings were tending “to bring adjudications of this tribunal into the same class as a restricted railroad ticket, good for this day and train only.” Certainly, the boards of review and those in the field who seek guidance in our pronouncement must wonder at the cavalier manner in which we overturn our previous holding. We write for their benefit, and there is *797some validity to the contention that law must have some degree of stability. It should not vacillate on a case to case basis and, unless a principle once established is found unsuitable, unworkable, or unsupportable, it should be permitted to crystallize. The rule overthrown in this instance has been supported by the great weight of authority from the early common law until the present time. As a matter of interest, my attention has not been called to a single jurisdiction which rejects the rule and, if the reader is interested in researching the problem, I am convinced he will look in vain for any authority which goes as far as the present decision. That is not to say that cases such as Silverthorne Lumber Co. v United States, 251 US 385, 64 L ed 319, 40 S Ct 182 (1920); Nardone v United States, 308 US 338, 84 L ed 307, 60 S Ct 266 (1939); and Coplon v United States, 191 F 2d 749 (CA DC Cir) (1951), cert den 342 US 926, 96 L ed 690, 72 S Ct 363 (1952), cannot be analogized and considered as authority for the proposition that derivative evidence which is tainted by a prior illegal search or wiretapping is inadmissible. However, there is involved in those cases a sort of disciplinary concept which has not been extended into the area of confessions and, while it can be argued that the rule should be broadened, there is no valid reason for considering that possibility in this instance. Here, the agents of the Government did not violate any statutory or constitutional right of this accused in originally obtaining his statement and, so, to apply the “fruit of the poisoned tree” doctrine is to overlook the simple fact that the tree was not poisoned.
While it must be conceded that I am losing the battle on many fronts, I have not yet been driven from the concept that “where a Manual provision does not lie outside the scope of the authority of the President, offend against the Uniform Code, conflict with another well-recognized principle of military law, or clash with other Manual provisions, we are duty-bound to accord it full weight.” United States v Villasenor, 6 USCMA 3, 19 CMR 129, and cases therein cited. Congress gave the President the power to prescribe rules of evidence not inconsistent with the Code, and he promulgated paragraph 140a of the Manual for Courts-Martial, United States, 1951 which provides as follows:
“Although a confession or admission may be inadmissible because it was not voluntarily made, nevertheless, the circumstance that it furnished information which led to the discovery of pertinent facts will not be a reason for excluding evidence of such pertinent facts. For example, the fact that an accused charged with larceny made an involuntary and therefore inadmissible statement to the effect that he stole the missing articles and hid them in his footlocker would not require the exclusion of evidence that the articles were discovered in his footlocker, even though the discovery was made solely because of the information contained in the statement of the accused.”
The foregoing provision meets the tests set forth above and it might be pertinent at this point to note it does not announce a principle unknown and unsupported in civilian circles. Dean Wigmore, in his work on Evidence, 3d ed, § 859, has this to say:
“Discovered Facts themselves always admissible. It was once contended that the impropriety of the inducement to the confession tainted the facts discovered in consequence of it, and that they also, as well as the confession, should remain inadmissible. Such a doctrine needs only to be stated to expose its equal lack of logic, principle, and expediency. It was fortunately repudiated at the outset in an opinion which leaves nothing to be said:
“1783,Warickshall’s Case, 1 Leach Cr. L., 3d ed., 298; a confession of stealing had been made, and in consequence of it the property was found concealed in • the lodgings of the accused; but the confession itself was otherwise inadmissible; ‘it was contended by her counsel that as the fact of finding the stolen property *798in her custody had been obtained through the means of an inadmissible confession, the proof of that fact ought also to be rejected’, as obtained by a breach of faith; the Court, NARES, L, and EYRE, B. (after the passage quoted ante, § 823) : ‘This principle respecting confessions has no application whatever as to the admission or rejection of facts, whether the knowledge of them be obtained in consequence of an extorted confession or whether it arises from any other source; for a fact, if it exists at all, must exist invariably in the same manner, whether the confession from which it is derived be in other respects true or false. Facts thus obtained, however, must be fully and satisfactorily proved without calling-in the aid of any part of the confession from which they may have been derived; and the impossibility of admitting any part of the confession as a proof of the fact clearly shows that the fact may be admitted on other evidence; for as no part of an improper confession can be heard, it can never be legally known whether the fact was derived through the means of such confession or not.’ ”
Wharton, Criminal Evidence, 11th ed, § 600, states substantially the same rule, where it provides:
“Admissibility of evidence of in-culpatory facts. The rule is settled that, notwithstanding the inadmissibility of the confession, all facts discovered in consequence of the information given by the accused, and which go to prove the existence of the crime of which he is suspected are admissible as testimony. Thus, where the accused, in confessing, points out or tells where the stolen property is; or, in case of homicide, states where the body can be found; or where the deceased was shot, which is verified by blood stained earth at that spot; or with what weapon, and in what part of the body, the deceased was shot, which is verified by exhuming and examining the body; or where he gives a clue to other evidence which proves the case, all such facts are admissible. But few courts have questioned this rule.
“It is obvious that a search made as a consequence of information given by the accused must result in the discovery of the inculpatory facts, as otherwise no testimony, either as to the confession, or as to the search instituted in consequence of it, is admissible. In connection with the discovery of the alleged in-culpatory facts, there should be proof, beyond a reasonable doubt, of the identity of the property, the body, or other fact. This is the rule with regard to larceny, and in other crimes, identification should be complete before admission of the incul-patory facts. But when the search reveals the inculpatory facts, and there is conclusive identification of such facts, this necessarily brings with it the reception in evidence of the accused’s statements in giving the information. It has also been held that evidence of conduct in retrieving property from a hiding place after a confession is also admissible.”
If there should be any question about the unanimity of the authorities on this principle, I suggest a reading of the following additional authorities: United States v Nardone, 106 F 2d 41 (CA 2d Cir) (1939), reversed on other grounds, 308 US 338, 60 S Ct 266, 84 L ed 307 (1939); United States v Hunter, 26 Fed Cas 436, No. 15, 424 (CC D DC) (1806); United States v Richard, 27 Fed Cas 798, No. 16,154 (CC D DC) (1823); Annotation, 53 LRA 402; 22 CJS, Criminal Law, § 831; 20 Am Jur, Evidence, § 402; Underhill, Criminal Evidence, 5th ed, § 404; Wharton, Criminal Evidence, 12th ed, § 358.
Apparently my brothers are not willing to accept the principle of law that the foregoing authorities espouse. Unquestionably, it is less favorable to an accused than the rule of this case, but that factor should not cause us to announce bad law. The public is entitled to have its rights considered, and a widespread application of the present rule would very effectively hamper the prosecution of an admitted *799offender. This case offers a concrete example of how justice might be defeated. The evidence which convicted this accused was supplied by witnesses, including a victim of his extortions, whose only connection with the alleged confession was that their identity was thereby established. If, because the Government first learned their names through the accused, their testimony is not usable, then so far as I am able presently to visualize a rehearing, the victim of a most atrocious extortion scheme must forever remain silent. It thus appears to me the Court goes too far for, unless it can now be shown that the Government was aware of the plan being operated by the accused and the identity of his victims and confederates before he made his statements, there is no way to escape the taint. That concept necessarily follows because, regardless of any subsequent developments, it can always be asserted that when the principal actors were identified by the accused, the knowledge acquired by the Government made independent identification an impossibility.
I would affirm the the decision of the board of review.