DAVIS, Judge.
A former naval disbursing officer, Lieutenant (j. g.) Royal Barry Shaw, brings this action to recover back pay and allowances under 28 U.S.C. § 1491 on the ground of wrongful dismissal, as well as to obtain “relief from responsibility for [the] loss” of Government funds under the special provisions of 28 U.S.C. §§ 1496 and 2512.
Before his dismissal, Lieutenant Shaw was in the Supply Corps, United States Naval Reserve, serving on active duty at the Naval Air Station, Willow Grove, Pennsylvania. On October 4, 1948, he was to be relieved as the disbursing officer by another naval officer. Prior to effecting this changeover, plaintiff discovered that his cash account was not in order, and that he was apparently short $5,614.01. This shortage was immediately brought to the attention of the Air Station’s Commanding Officer who suspended Lieut. Shaw, according to Navy regulations. An informal auditing board and, later, a formal inspection board investigated the deficiency and reaffirmed the earlier findings. Throughout this period of discovery and investigation, plaintiff acknowledged the existence of the apparent shortage, but often expressed his inability to understand how it might have developed.
On November 20, 1948, the Navy lodged four court-martial charges against him: — for “embezzling money of the United States intended for the naval service thereof”; “neglect of duty”; “culpable inefficiency in the performance of duty”; and “violation of a lawful regulation issued by the Secretary of the Navy.” On December 2nd, a General Court-Martial was convened. After a 21-day trial the court-martial found Lieut. Shaw guilty on all charges, and sentenced him to be dismissed from the naval service, and to be imprisoned at hard labor for three years. A series of appeals and reviews followed. On February 3, .1949, the convening authority approved the proceedings, findings, and sentence, with immaterial exceptions. The Navy Judge Advocate General recommended, in September 1949, that the convening authority’s findings and actions on the several specifications of the second, third, and fourth charges be set aside, and that Shaw’s embezzlement conviction (the first charge) be approved. The Under Secretary of the Navy accepted this recommendation, affirming plaintiff’s embezzlement conviction and the dismissal. Lieut. Shaw was dismissed on February 8, 1950. Attempts to have the conviction and dismissal reviewed by the Court of Military Appeals were unavailing.
Plaintiff urges a number of constitutional grounds for holding his conviction void. The Government denies any invalidity in the military proceedings, and also asserts that this court should not reweigh and reassess the evidence or in
quire into possible errors of law in the court-martial. Stated broadly, the two questions before us are (1) whether we can look into any of the challenges to the court-martial conviction, and (2) whether, if such an examination is warranted, we ean find adequate ground for holding that plaintiff’s dismissal from the service, as a result of his conviction, was void.
r
Although our jurisdiction to reexamine court-martial decisions has not been defined broadly, that it exists is beyond question. Almost two decades ago, we unequivocally declared that a court-martial’s denial of a plaintiff’s fundamental constitutional rights operated to deprive it of jurisdiction and vested us with the power to grant relief by way of a money judgment if the serviceman had been removed from the service by the sentence of the court-martial. Shapiro v. United States, 69 F.Supp. 205, 107 Ct.Cl. 650 (1947).
A conviction was held no bar to recovery “if the verdict of the court martial was absolutely void and, therefore, forms no foundation for plaintiff’s dismissal.” Id. 69 F.Supp. at 207, 107 Ct.Cl. at 654. Some years later, Burns v. Wilson, 346 U.S. 137, 73 S.Ct. 1045, 97 L.Ed. 1508 (1953), held that civil courts could “determine whether the military have given fair consideration” to claims of fundamental unfairness, but that they could not “reexamine and reweigh each item of evidence * * * ” Id. at 144, 73 S.Ct. at 1050. A majority of the Supreme Court agreed that the scope of collateral review by ha-beas corpus went beyond limited traditional notions of “jurisdiction”. Id. at 142, 148, 154, 73 S.Ct. 1045. We have never held that the scope of our “review” by way of. a proceeding for a money judgment was any more' restrictive. On the contrary, our opinions have consistently stated or assumed that denial of significant constitutional rights would render the military conviction invalid, and permit this court to award back-pay.
Other courts have also been willing to look beneath final court-martial convictions (and dismissals) to search for fundamental unfairness, and to order corrective action when the service has refused to do so. A recent example is Ashe v. McNamara, 355 F.2d 277 (C.A. 1, Dec. 14, 1965).
To be sure, where the plaintiff was “accorded a
full and fair
hearing” (emphasis supplied) within the military system, we have said, following Burns v.
Wilson, that we would refuse to cover the same ground by reconsidering the plaintiff’s objections anew. Begalke v. United States, supra, 286 F.2d at 608-610, 148 Ct.Cl. at 401, 403. In that case the constitutional issues (involving alleged illegal searches and seizures, police interrogation, criticism of the accused’s counsel during the trial, and claimed interference with the right to appeal) all turned on an evaluation of the particular facts; this court was unwilling to reassess those circumstances for itself, once the military had done so with care. Whether or not this rule of deference to the military findings has been modified or refined by Townsend v. Sain, 372 U.S. 293, 83 S.Ct. 745, 9 L.Ed.2d 770 (1963) and Sanders v. United States, 373 U.S. 1, 83 S.Ct. 1068, 10 L.Ed.2d 148 (1963), we think that such abstinence is not to be practiced where the serviceman presents pure issues of constitutional law, unentangled with an appraisal of a special set of facts. That type of unmixed legal question this court has always decided for itself. See the cases cited in fn. 4, supra. The problem we consider in Part II, infra, is of this order. We must face the claim that, on the Navy’s own factual findings, plaintiff could not constitutionally be held to have committed any crime.
Furthermore, our reluctance to consider Begalke’s contentions
de novo
was prompted, in part, by the fact that he had had the opportunity to bring his conviction to the Court of Military Appeals whose action was by statute “final and conclusive”. Id. 286 F.2d at 607-608, 148 Ct.Cl. at 399, 403.
See, also Moses v. United States, 137 Ct.Cl. 374, 379-380 (1957); Narum v. United States, supra, 287 F.2d at 907, 151 Ct.Cl. at 330 (1960) (dissenting opinion). In other cases, the right to review here has been clouded by doubt as to whether the plaintiff had timely raised his constitutional claims before the military tribunals (Narum v. United States, supra, 287 F.2d at 901-903, 151 Ct.Cl. at 319-322 (I960)), or had objected to the alleged unconstitutional treatment he received. Fly v. United States, supra, 100 F.Supp. at 442, 120 Ct.Cl. at 498 (1951); Krivoski v. United States, supra, 145 F.Supp. at 241, 243, 136 Ct.Cl. at 455, 458 (1956). These considerations play no part in Lieut. Shaw’s case, since the Court of Military Appeals ruled it lacked jurisdiction of his appeal (see fn. 2), and he has vigorously asserted his constitutional rights throughout.
II
At the end of the successive reviews within the military hierarchy, the only surviving charge against plaintiff was the first, “embezzling money of the United States intended for the naval service thereof.” All the other accusations were quashed. This count of “embezzlement” specified that Lieut. Shaw, while a disbursing officer, did
during the period of April 19, 1948 to October 4, 1948, fail safely to keep all monies received in his possession and under his control in the execution and under color of his office as aforesaid, namely, a sum of public money of about five thousand six hundred fourteen dollars and one cent ($5,614.01), property of the United States intended for the naval service thereof, and did, therein and thereby, then and there embezzle the said sum of money.
Although the charge was sustained, the Navy expressly found (i) (through a formal opinion of the Judge Advocate General) that “the evidence fails to show any neglect or culpable inefficiency on the part of the accused, but on the contrary contains considerable testimony of prosecution and defense witnesses indicating that the accused exercised due caution and care in his handling of public money in his custody”, and (ii) (through a dispositive memorandum of
the Under Secretary) that Shaw’s conviction was of “statutory embezzlement, which requires no proof of criminal intent” and that “the record does not disclose'evidence to the effect that he converted any of the funds to his own use.” The only issue we need reach is whether the Navy could constitutionally convict the plaintiff of “fail[ing] safely to keep” Navy funds when the Navy itself has affirmatively found that he had no criminal intent to convert, and did not convert, and was not negligent, inefficient, or careless. On the Navy’s own findings, we hold that it could not constitutionally make the shortage of money, in these circumstances, a crime punishable by dismissal from the service.
A. The prosecution of Lieut. Shaw for his alleged failure safely to keep naval funds rested on paragraph 8 of article 14 of the then Articles for the Government of the United States Navy (AGN) (34 U.S.C. § 1200 (1946), as amended, effective until May 31, 1951), and 18 U.S.C. § 650 (1948). Paragraph 8 enables courts-martial to punish “any person in the naval service” “who steals,
embezzles,
knowingly and willfully misappropriates, applies to his own use or benefit, or wrongfully and knowingly sells or disposes of” Government money or property “furnished or intended for the military or naval service * * * ” (Emphasis supplied). Since “embezzles” is employed in this article without definition, the Navy’s practice, when charging offenses under it, has been to define the term by reference to the statutes of the United States.
In this case, resort was had to 18 U.S.C. § 650 which establishes as “embezzlement” the act of “any public depositary [who] fails to keep safely all moneys” of the United States deposited with him by certain named persons “or other person having money of the United States * * *.” Section 649(b) of 18 U.S.C. makes § 650 applicable “to all persons charged with the * * * disbursement of the public money * * * ”, and in the Navy’s view it was open to it to turn to the latter for a definition of “embezzles” in paragraph 8. The Navy has long extended § 650’s coverage to disbursing officers like Lieut. Shaw in this way, and we do not quarrel, at this point in history, with that practice (although there is some question whether paragraph 8 was not originally designed to cover only actual fraud or actual conversion).
It is very clear that, in applying § 650 to Shaw, the Navy’s position was (1) that the statute initially required the prosecution to prove only a shortage in Shaw’s funds; (2) that Shaw then had
the burden of explaining that loss by showing that it was caused by someone or something else, without any fault or negligence on Shaw’s part; and (3) that proof of fraudulent intent was, in any event, not a prerequisite to conviction. The claimed irrelevance of actual fraudulent intent is demonstrated not only by the Navy’s past practice but also by reiterated comment in this very case. The Under Secretary (approving plaintiff’s conviction) wrote that “Shaw now stands convicted of statutory embezzlement, which requires no proof of criminal intent, and the record does not disclose evidence to the effect that he converted any of the funds to his own use.” When plaintiff’s counsel appealed, a Navy review officer replied (to the argument of lack of proof of fraud or willfulness) that the actions of one failing to keep safely public money “are made penal by 18 U.S.C. 650 (1948) * * *, and are denominated embezzlement without regard to [their] * * * being done fraudulently.”
The other two elements of the Navy’s conception of the crime — the prosecution need show only a shortage; the burden of proof then shifts to the accused who can escape only by proving that the loss was wholly attributable to another individual or circumstance — are equally plain. At the court-martial trial, the Judge Advocate, summarizing his case, insisted that, when the prosecution establishes an “unexplained shortage”, “the disbursing officer is
prima facie guilty and must show
what has become of the missing funds”, quoting the Naval Digest at 206 (1916.)
The details of this rule were spelled out in the Digest: — after proof of a shortage, “the accused in. order to rebut the prosecution’s evidence that the money is missing, which is prima facie evidence of embezzlement, must show not only that the funds were stolen or misappropriated by another, but, furthermore, must affirmatively show that such theft or misappropriation by another was not due to fault on the part of himself, the accused.” Naval Digest 209 (1916). (See also the Digest’s comments on 18 U.S.C. § 3487, discussed infra, in subpart II, C, of this opinion.)
It is incontestable that the Navy considered Lieut. Shaw’s case in just this light. Not only did the trial Judge Advocate urge the Naval Digest rule on the court-martial, supra,
but naval personnel analyzing plaintiff’s trial at various levels of review also relied, expressly, on this rule to support the legality of the conviction. For example, in rebutting plaintiff’s post-trial contention that the Navy had not carried its burden, a JAG officer concluded that “once the shortage has been established as it was in the instant case, conversion is presumed, a prima facie case has been made out, and the burden shifts to the accused to account for or explain the shortage.” Similarly, when plaintiff appealed to the Secretary of Navy, the official response was that the specification under the charge of “embezzlement” did “not allege a conversion of the money but states as the gravamen of the offense, that the accused did ‘fail safely to keep all monies received in his possession and control in the execution and under the color of his office’ * * *. There was ample evi
dence to show the shortage of money alleged and, in fact, the accused prior to trial admitted such shortage.”
The Navy went beyond a mere shifting of the burden of justification to the accused. It did not consider that affirmative proof by Shaw that he was free from negligence or carelessness could absolve him. As we have pointed out, the service acknowledged that “there is no evidence in the record which would support a finding that the accused was negligent or culpably inefficient in the handling of public monies in his custody, * * * ”, or that he had converted any of the funds to his own use. The appellate authorities within the Navy agreed, too, that there was affirmative proof supported by “considerable testimony of prosecution and defense witnesses indicating that the accused exercised due caution and care in his handling of public money in his custody.”
We think it indisputable that the evidence demonstrated, to the Navy’s own satisfaction, that the cash deficit was not attributable to any default by Shaw; in fact, this lack of proof warranted dismissal of charges against him of “Neglect of Duty” and “Culpable Inefficiency in the Performance of Duty”. Nevertheless, he was convicted of a greater offense — embezzlement — since he was unable to undercut the presumption of guilt by showing affirmatively “that the [absent] funds were stolen or misappropriated by another”, or lost through other
proved
circumstances, as the Naval Digest, supra, required.
B. In judging the constitutional validity of such a conviction, we assume that the standards for proving guilt under 18 U.S.C. § 650 are less exacting than those generally applicable in cases of conventional embezzlement. Normally, proof of embezzlement requires a showing a fraudulent intent. Cf. Morissette v. United States, 342 U.S. 246, 263-273, 72 S.Ct. 240 (1952). The crime is closely akin to larceny, the chief difference being that in larceny the taking follows an unlawful trespass, while embezzled property is, at the outset, in the lawful possession of the wrongdoer. See Moore v. United States, 160 U.S. 268, 269-270, 16 S.Ct. 294, 40 L.Ed. 422 (1895); Hall, Theft, Law and Society 6-10 (1935). Section 650 and its predecessors, however, have been read to eliminate the need for proof of true fraudulent intent. Originally, the provision was part of “An Act to provide for the better Organization of the Treasury, and for the Collection, Safe-Keeping, Transfer, and Disbursement of the public Revenue.” Sec. 16, 9 Stat. 59, 63 (1846). Its manifest purpose was to promote care and to require a higher level of conduct in a particular class of federal officers— those “charged * * * with the safekeeping, transfer, and disbursement, of the public moneys * * 9 Stat. 63. See 28 Ops.Att’y Gen. 286, 297 (1910). The 1846 statute’s proscription against various defaults by Government officers was specifically extended eleven years later when Congress decreed that “public depositaries shall safely keep all moneys deposited” with them; the “failure so to do” was made “embezzlement”. Sec. 2, 11 Stat. 249 (1857). At other times, related sections were enacted which singled out and defined specific defaults by disbursing officers to be characterized as embezzlement.
The “safely keep”
provision adopted in 1857 was reenacted several times, until it appeared in 1948 as § 650. A central purpose of these legislative efforts seems to have been to alter the standards for proving “embezzlement” by those responsible for public money. We assume, therefore, that proof that plaintiff negligently lost the cash, or negligently allowed it to be lost or taken, would have been sufficient.
C. We come now to consider whether the Navy applied a fair and reasonable rule of prima facie proof or of presumption in convicting plaintiff. Could the Navy justifiably conclude that Shaw had carelessly failed to “keep safely” the naval funds entrusted to him unless he affirmatively explained the loss or proved that it was due to another’s misappropriation or fault?
The Navy’s answer seems to draw its support from two principal .roots. One is 18 U.S.C. § 3487 (1948), 62 Stat. 833, providing that “the refusal of any person * * * charged with the * * * disbursement of the public money * * * to transfer or disburse any such money, promptly, upon the legal requirement of any authorized officer, shall be deemed, upon the trial of any indictment against such person for embezzlement, prima facie evidence of such embezzlement.” Long ago, the Navy interpreted an earlier version of § 3487 as authorizing the use of a prima facie rule in prosecutions under the then formulation of 18 U.S.C. § 650. This interpretation is unequivocally set forth in the Naval Digest: “For more than half a century it has been held * * * that
the mere failure
of a disbursing officer * * * to produce or account for the public moneys in his hands when required to do so
constitutes embezzlement,
unless such officer is able satisfactorily to explain such failure
or
to show that the funds which he could not produce were, without fault on his part, lost or stolen.” Naval Digest 207 (1916) (emphasis supplied)
Whatever its current validity,
this interpretation does not sanction plaintiff’s conviction since he has proved, in the Navy’s own eyes, that he was without fault. Such proof satisfies the requirements of the Naval Digest’s comment on § 3487; that rule does not compel the accused, who shows his own lack of fault, to prove affirmatively the actual cause of the loss.
To define the defendant’s burden under § 650, the Navy has also drawn on a 1910 opinion of the Attorney General. 28 Ops.Att’y Gen. 286. The situation there was similar to the one before us.
An assistant Navy paymaster was charged with embezzlement (in violation of Article 14, AGN, paragraph 8) and culpable negligence. The sole act alleged in the embezzlement specification was that the accused “did fail to safely keep and account for” a sum of money. The court-martial refused to find that the accused “actually appropriated” the missing money, “or that he made any particular disposition of it”, but did find that the defendant had been negligent. It thus acquitted the defendant of embezzlement, and found him guilty of “culpable negligence in the performance of duty.” In effect two questions were put to the Attorney General: (i) Did the specification allege facts sufficient to constitute the offense of embezzlement, or was it necessary
to allege
the existence of fraudulent intent?; (ii) did the findings of the court “render the accused, as a matter of law, guilty of that offense”, or was it necessary for the Navy
to prove
an evil intent? The answer to the first question was that, if fraudulent “intent or knowledge were necessary, their absence, under the positive provisions of this statute [§ 650’s predecessor], would be a matter of defense, and it would not be necessary to specifically declare their presence in the specification.” Id. at 292. The response to the second inquiry was lengthier. The Attorney General reviewed the authorities for and against the proposition that proof of fraudulent intent is an element of embezzlement, concluding only that the state of the law was clouded. He decided, however, “that there may be such character of negligence as will take the place of criminal intent”, and, since the court-martial found that the defendant had been
negligent,
his conduct “was equivalent to a criminal intent”. Id. at 296, 298. The Navy could thus convict him under the then version of § 650.
This opinion, says the Naval Digest, sustains the rule that a defendant can overcome the legal presumption of embezzlement under § 650, raised by proof of a shortage, only by introducing evidence to support findings (a) “that the funds were stolen or misappropriated by another”,
and
(b) “that such theft or misappropriation * * * was not due to fault on the part of” the accused. Naval Digest at 209. This overextends the reach of the Attorney General’s opinion which said nothing to suggest that the accused had to prove, in addition to his own freedom from fault, that the loss was in fact due to another’s fault or to extraneous circumstances unconnected with the defendant. In the case put, the Navy had already proven negligence. The Attorney General assumed that a fraudulent “intent or knowledge [was] * * * necessary”, and held only that proof of negligence was adequate fulfillment of a requirement of criminal intent or
mens rea
(if that was an element of the crime). The Attorney General did not say or intimate that a defendant who disproved his own negligence would have to go further, to clear himself, and show that another person or event was actually responsible for the loss.
We find, therefore, that the specific authorities on which the rule applied in Lieut. Shaw’s case is said to rest furnish no support. On the other hand, there are very good reasons for holding that that rule, as used here, would deprive plaintiff (a criminal accused) of constitutional rights by creating an invalid presumption of wrong-doing.
It goes without saying that the prosecution has the burden in a criminal proceeding of proving guilt beyond a reasonable doubt. See Speiser v. Randall, 357 U.S. 513, 523-524, 78 S.Ct. 1332, 2 L.Ed.2d 1460 (1958). Evidentiary rules creating presumptions (i. e., prima facie rules)
to assist the prosecution in meeting that burden cannot be sustained where there is "no rational connection between the fact proved and the ultimate fact presumed, if the inference of the one from proof of the other is arbitrary because of lack of connection between the two in common experience.” Tot v. United States, 319 U.S. 463, 467-468, 63 S.Ct. 1241, 1245, 87 L.Ed. 1519 (1943). Incriminating presumptions are not ordinarily to be constructed, except by legislation. Morissette v. United States, supra, 342 U.S. at 275, 72 S.Ct. 240, 96 L.Ed. 288. But even when done by legislation the extent to which the prosecution’s burden may be shifted to the defendant is limited by the Due Process Clause of the Constitution. Tot v. United States, supra, 319 U.S. at 466-467, 63 S.Ct. 1241; United States v. Romano, 382 U.S. 136, 86 S.Ct. 279, 15 L.Ed.2d 210 (1965).
Measured by these standards, the Navy’s rule requiring defendants charged with “embezzlement” under § 650 to prove both their own lack of fault
and
another’s culpability (or some other explanation of the loss) goes too far. The “ultimate fact presumed” is that the accused was guilty of negligence or fault. Even though he has adequately proved, by specific evidence, that he was
not
negligent or at fault, the Navy would still infer that fact from his failure to go further and prove that another person was responsible or that there was some other explanation (such as fire). In the face of an accused’s affirmative showing of good character, due caution and care in handling funds (see fdg. 13(c), 14(a)), this additional demand cannot be justified in the circumstances present here. “Common experience” tells us that there will be enough occasions when a custodian or depositary who has exercised the greatest of care will be unable to explain a shortage, or definitively to pin it on another. Sometimes such a showing may be impossible or beyond human compass; often it will be extremely difficult.
It is unreasonable to vitiate satisfactory affirmative proof of due care simply because the accused has not been able to perform the impossible or the overly difficult. To insist, in this situation, on basing the conviction on the prosecution’s mere proof of a shortage shown primarily by office records kept over a six-months period creates an “inference [of guilt beyond a reasonable doubt] * * * so strained as not to have a reasonable relation to the circumstances of life as we know them * * Tot v. United States, supra, 319 U.S. at 468, 63 S.Ct. at 1245. See Manley v. State of Georgia, 279 U.S. 1, 5-7, 49 S.Ct. 215, 73 L.Ed. 575 (1929).
The “comparative convenience of producing evidence” (Tot v. United States, supra, 319 U.S. at 467, 63 S.Ct. 1241) will not warrant imposition on the defendant, at his peril, of the task of satisfactorily explaining the loss. Convenience, fairness, and perhaps necessity, sanctioned the casting on Lieut. Shaw of the burden of producing evidence of his own due care (and he did so). There
is, however, a critical distinction between requiring a defendant, on the one hand, to come forward with evidence where he “has the better means of information”, the “more convenient access to the proof, and where requiring him to go forward with proof will not subject him to unfairness or hardship” (Tot v. United States, supra, 319 U.S. at 469-470, 63 S.Ct. at 1246), and forcing him, on the other hand, to prove the actions of third parties, known and unknown, or to show the effects of outside circumstances, known and unknown. Defendants have no special access to knowledge of such facts. Indeed, in Shaw’s case, the reallocated burden was weightier still. The Navy charged him with the failure safely to keep public monies “during the period April 19,1948 to October 4,1948”, apparently because it was unable to pinpoint with any greater accuracy the time at which the loss occurred (see fdgs. 5, 19). As a result, the accused was given not only the task of determining how the loss occurred; he was also, in effect, asked (as a necessary preliminary) to determine when within a six-months period it occurred. There were, moreover, several other persons who had access to the monies during this time and there was the possibility of undiscovered mistakes in the office records. We think it plain that “in such circumstances the promotion of convenience from the point of view of the prosecution .[is] * * * outweighed by the probability of injustice to the accused.” Morrison v. People of State of California, 291 U.S. 82, 94, 96-97, 54 S.Ct. 281, 286, 78 L.Ed. 664 (1934).
We must hold, for these reasons, that Congress could not have meant 18 U.S.C. § 650 or 18 U.S.C. § 3487, on “embezzlement”, to embody the rule contended for by the Navy, and that It was not constitutionally open to the Navy to read that rule into these sections.
D. Our faith in this conclusion is the stronger because our decision accords with the Court of Military Appeals’ view of the proper function of presumptions in embezzlement prosecutions under the Uniform Code of Military Justice.
In United States v. Crowell, 9 U.S.C.M.A. 43, 25 C.M.R. 305 (1958), that court set forth an exacting standard of proof in embezzlement cases. The accused (an Army Lieutenant) was entrusted with certain funds and responsible for their disbursement. An audit of the funds revealed a shortage for which the accused was unable to account. There was no direct evidence of a conversion. “The theory of prosecution was based on the' inference of larceny [i. e., embezzlement] that may be drawn from a failure to account by one who has assumed the custody of another’s property.” This theory was set out in the following instruction:
“There is a well-established legal presumption that one who has assumed the stewardship of another’s property has embezzled such property if he does not or cannot account for or deliver it at the time an accounting or delivery is required of him. The burden of going forward with the proof of exculpatory circumstances then falls upon the steward, and his explanatory evidence, when balanced against the presumption of guilt arising from his failure or refusal to render a proper accounting of or to deliver the property entrusted to him, creates a controverted issue of fact which is to be determined by you members of the court.”
Passing on the validity of this instruction, the Court of Military Appeals first stated that the term “presumption” “refers only to justifiable inferences which the members of a court-martial may draw from the facts and accept or reject according to their experience,” Id. at 308. It then held the quoted instruction “glaringly defective” and prejudicial, since it clothed a “presumption” with “an aura of unquestioned legal sanctity which is neither deserving nor correct”; it left the accused with the burden of producing an undue amount of evidence; and the phrase “presumption of guilt” “runs counter to the well-known presumption of innocence * * *.” Id. at 308-309. These observations are applicable with equal force to the rule urged on the court-martial at plaintiff’s trial. That rule, which was central to Shaw’s conviction, had the proscribed effect of creating a legal presumption of guilt on the evidence introduced at the trial.
The general doctrine of the
Crowell
case, supra, has been applied where the prosecution’s theory was that the “mere proof of an existing shortage in a trust fund”, together with the presumption arising from the failure to account, established a “prima facie” case of embezzlement without additional evidence. United States v. Pettiford, supra, 27 C.M.R. 617, 621, 623-624 (1958) (Army Board of Review). See United States v. Sellers, 12 U.S.C.M.A. 262, 30 C.M.R. 262, 269-270 (1961); United States v. Troutt, 8 U.S.C.M.A. 436, 24 C.M.R. 246, 249 (1957); see, also, United States v. Powell, 29 C.M.R. 688, 700-705 (1959) (Navy Board of Review). There are, of course, limits circumscribing the rule that the failure to account does not alone constitute embezzlement. Where other factors cast suspicion on the accused, the Court of Military Appeals has approved an instruction that the “failure to account when demand is made or when an accounting is due” will
permit
an inference of actual conversion. United States v. Keleher, 14 U.S.C.M.A. 125, 33 C.M.R. 337, 343 (1963). The court has also allowed the law officer to instruct that proof of both a deficiency and the alteration of records would
permit
an inference by the court-martial of an actual, wrongful conversion. United States v. Lyons, 14 U.S.C.M.A. 67, 33 C.M.R. 279 (1963). In such cases, however, the court has been careful to caution that the inference is permissible, not mandatory; that the court-martial must consider it against the backdrop of the whole case; that the court-martial is “wholly unfettered” as to whether such an inference should be drawn; and, in
Lyons',
that proof of both elements (a deficiency and alteration of the records) is essential.
Crowell,
supra, at 309;
Keleher,
supra, at 344;
Lyons,
supra, at 282. No such saving words of caution restricted the effect to be accorded the presumption of embezzlement in Lieut. Shaw’s case.
E. We end by noting, briefly, that plaintiff’s conviction cannot be upheld on the ground of absolute responsibility for the loss, irrespective of fault. Neither paragraph 8 of Article 14 of the Articles for the Government of the Navy nor 18 U.S.C. § 650 — each of which refers to embezzlement — seeks to establish such a crime-without-fault, and the Navy did not review Shaw’s conviction on that basis. We take the proceedings as we find them. Cf. United States v. Romano, supra, 382 U.S. at 143-144, 86 S.Ct. 279, 15 L.Ed.2d 210 (1965). There is, moreover, the gravest doubt whether Congress could validly make it a crime, punishable by imprisonment and dishonorable dismissal from the service, for a careful and honest custodian of naval funds to have a shortage not preventable by due care. Cf. Morissette v. United States, supra,
342 U.S. at 254-263, 265, 72 S.Ct. 240, 96 L.Ed. 288.
Nor is there any substance to the defendant’s contention that, under the “theory that the greater includes the lesser”, the Navy’s intention to dismiss Lieut. Shaw embraced the intention to release him from active duty under § 5 of the Naval Reserve Act of 1938 (52 Stat. 1175-76, as amended, 56 Stat, 739 (1942), 34 U.S.C. § 853c (1946 ed.)), empowering the Secretary of the Navy to “release any member from active duty either in time of war or in time of peace.” First, it is apparent that Shaw’s dismissal was not predicated on that section, but on his embezzlement conviction. Second, since the expressed ground for the discharge was void, so too is the discharge (and, in defendant’s terms, all its lesser included parts), and it cannot be saved by any general but unasserted power resting with the Secretary of the Navy. Cf. Clackum v. United States, 296 F.2d 226, 229, 148 Ct.Cl. 404, 410 (1960).
Accordingly, plaintiff is entitled to recover back pay and allowances, less appropriate offsets, from the date on which his pay was improperly withheld to the date of judgment. See Motto v. United States, 348 F.2d 523, 525, 527, 172 Ct.Cl. - (July 1965). The court also finds, under 28 U.S.C. §§ 1496 and 2512 (see fn. 1, supra), that the loss of $5,-614.01 in the funds entrusted to plaintiff (which the Navy apparently set off against his pay) was without his fault or negligence. Judgment is entered to that effect. The amount of recovery will be determined under Rule 47(c).