with whom
Justice Brennan joins, and with whom Justice Blackmun joins in all but the last paragraph, dissenting.Less than 20 years ago, this Court held in O’Callahan v. Parker, 395 U. S. 258 (1969), that, to be subject to trial by court-martial, a criminal offense charged against a member of the Armed Forces had to be “service connected,” lest the phrase “cases arising in the land or naval forces” in the Fifth Amendment “be expanded to deprive every member of the armed services of the benefits of an indictment by a grand jury and a trial by a jury of his peers.” Id., at 273. Today the Court overrules O’Callahan. In doing so, it disregards constitutional language and principles of stare decisis in its singleminded determination to subject members of our Armed Forces to the unrestrained control of the military in the area of criminal justice. I dissent.
I
The majority begins by assuming that the limitation on court-martial jurisdiction enunciated in O’Callahan was based on the power of Congress, contained in Art. I, § 8, cl. 14, “[t]o make Rules for the Government and Regulation of the land and naval Forces.” It then rejects this asserted limitation of congressional power on the ground that the Framers intended to give Congress plenary authority over the *453government of the military. But the Court in O’Callahan did not simply address whether Art. I, §8, el. 14, granted Congress the authority to create court-martial jurisdiction over all crimes committed by members of the Armed Forces. Congress’ Article I power to regulate the Armed Forces is limited by the Fifth Amendment right to indictment or presentment by a grand jury and the Sixth Amendment right to trial by jury.1 “[T]he constitutional grant of power to Congress to regulate the armed forces,” this Court has previously stated, “itself does not empower Congress to deprive people of trials under Bill of Rights safeguards, and we are not willing to hold that power to circumvent those safeguards should be inferred through the Necessary and Proper Clause.” United States ex rel. Toth v. Quarles, 350 U. S. 11, 21-22 (1955). The majority simply disregards the limitations the Bill of Rights imposes on the reach of Art. I, §8, cl. 14.
The rights to grand jury process and to trial by jury are, of course, of restricted application in military cases. The Fifth Amendment excepts from the grand jury requirement “cases arising in the land or naval forces, or in the Militia, when in actual service in time of War or public danger,”2 and the *454Court has held this exception applicable to the Sixth Amendment right to trial by jury as well. Ex parte Milligan, 4 Wall. 2, 123 (1867). But the text of the exception is inconsistent with the majority’s conclusion that the only relevant factor in determining whether a court-martial has jurisdiction over a case is the status of the defendant as a member of the Armed Services.3
The Fifth Amendment’s exception covers only “cases arising in the land and naval forces” (emphasis added). It makes no reference to the status of the individual committing the crime. Had that been the Framers’ intent, it would have been easy to have said so, given that the grand jury provision of the Amendment, which states that “[n]o Person shall be held to answer for a capital, or otherwise infamous crime, unless on a presentment or indictment of a Grand Jury,” speaks not in terms of “crimes” or “cases,” but of individual defendants. Nonetheless, the exception contained in the Fifth Amendment is expressed — and applies by its terms — only to cases arising in the Armed Forces. O’Callahan addressed not whether Art. I, § 8, cl. 14, empowered Congress to create court-martial jurisdiction over all crimes committed by service members, but rather whether Congress, in exercising that power, had encroached upon the rights of members of Armed Forces whose cases did not “arise in” the Armed Forces. This is clear from the Court’s statement of its holding in O’Callahan:
“We have concluded that the crime to be under military jurisdiction must be service connected, lest ‘cases arising in the land or naval forces, or in the Militia, when in actual service in time of War or public danger,’ as used in the Fifth Amendment, be expanded to deprive every *455member of the armed services of the benefits of an indictment by a grand jury and a trial by a jury of his peers.” 395 U. S., at 272-273 (footnote omitted).4
*456The protections afforded individuals by the Fifth and Sixth Amendments are central to our constitutional scheme of justice. The right to trial by jury, in particular, “ranks very high in our catalogue of constitutional safeguards.” United States ex rel. Toth v. Quarles, 350 U. S., at 16. These protections should not be lightly abrogated. Consequently, the exception in the Fifth Amendment for cases arising in the Armed Forces must be strictly construed. This was the basis for the Court’s conclusion, in Toth, that the power to authorize trial by court-martial should be limited to “‘the least possible power adequate to the end proposed.’ ” Id., at 23 (emphasis omitted), quoting Anderson v. Dunn, 6 Wheat. 204, 231 (1821).
The historical evidence considered by the Court in O’Callahan is therefore relevant, not to what the Framers intended to include in the scope of the congressional power to regulate the Armed Forces in Art. I, § 8, cl. 14, but to what the Framers, wary of military jurisdiction and familiar with strong restrictions on the scope of that jurisdiction, considered “cases arising in the armed forces.” Even assuming that they intended to assign control over the scope of the Article I power to the Legislature, this does not imply that the meaning of the Fifth Amendment’s “arising in” exception can be interpreted without reference to the practices of that time.
In that respect it is significant that the British political and legal writing of the 17th and 18th centuries demonstrates a longstanding suspicion of broad court-martial jurisdiction. This suspicion was well known in colonial America, and was based on familiar history.5 British writers and legislators *457took a narrow view of the appropriate scope of court-martial jurisdiction, which manifested itself in a very limited grant of authority to try offenses by court-martial during the period of which the Framers would have been most acutely aware. See, e. g., M. Hale, The History of the Common Law of England 42-43 (6th ed. 1820). Not only was. that jurisdiction narrow, it was expressly limited to eases having some connection with the military. The test was not one of status, but one of military relationship. See S. Adye, A Treatise on Courts Martial 60 (1786) (“The crimes that are cognizable by a court martial, as repugnant to military discipline, are pointed out by the mutiny act and articles of war . . . and as to other crimes which officers and soldiers being guilty of, are to be tried for by the ordinary course of law, in like manner with other subjects”); see also 1 C. Clode, Military Forces of the Crown; Their Administration and Government 158 *458(1869) (“It has been a subject of controversy to distinguish the offences that are purely Military (and therefore properly within the cognizance of a Court-martial), from others that are Civil or Political (and therefore properly within the cognizance of the civil tribunals of the community)”); Grant v. Gould, 2 H. Bl. 69, 99-100, 126 Eng. Rep. 434, 450 (C. P. 1792) (“In this country, all the delinquencies of soldiers are not triable, as in most countries in Europe, by martial law; but where they are ordinary offences against the civil peace they are tried by the common law courts. . . . The object of the mutiny act. . . is to create a court invested with authority to try those who are a part of the army . . . and the object of the trial is limited to breaches of military duty”) (emphasis omitted). The reach of military law in Britain at the time of the Revolution thus permitted courts-martial only for offenses committed by members of the Armed Forces that had some connection with their military service.
The majority disputes the O’Callahan Court’s suggestion that the British Articles of War forbade the trial of civil offenses by court-martial. The Court points to Section XIV, Article XVI, of the British Articles of War of 1774, reprinted in G. Davis, Military Law of the United States 581, 593 (3d rev. ed. 1915), which provided:
“All Officers and Soldiers are to behave themselves orderly in Quarters, and on their March; and whosoever shall commit any Waste or Spoil either in Walks of Trees, Parks, Warrens, Fish Ponds, Houses or Gardens, Corn Fields, Inclosures or Meadows, or shall maliciously destroy any Property whatsoever belonging to any of Our Subjects, unless by order of the then Commander in Chief of Our Forces, to annoy Rebels or other Enemies in Arms against Us, he or they that shall be found guilty of offending herein shall (besides such Penalties as they are liable to by law) be punished according to the Nature and Degree of the Offence, by the Judgment of a Regimental or General Court Martial.”
*459The majority contends that this provision counters any argument that court-martial jurisdiction in Britain at the time of the American Revolution was in any respect limited to offenses not punishable by civil law. Ante, at 443. The latter provision, however, appears in a section of the Articles of War captioned “Of Duties in Quarters, in Garrison, or in the Field,” and its text suggests that the activities it forbade were considered derelictions of military duty, and were punishable by court-martial on that basis.6
American colonists shared the British suspicion of broad military authority in courts-martial. One of the grievances stated in the Declaration of Independence was King George Ill’s assent to “pretended Legislation: For quartering large bodies of armed troops among us: For protecting them, by a mock Trial, from punishment for any Murders which they should commit on the Inhabitants of these States.” The Framers thus were concerned both with protecting the rights of those subjected to courts-martial, and with preventing courts-martial from permitting soldiers to get away with murder — literally—in the civilian community. This “known hostility of the American people to any interference by the military with the regular administration of justice in the civil courts,” Coleman v. Tennessee, 97 U. S. 509, 514 (1879), makes it unlikely that the Framers considered any crime committed by a member of the Armed Forces, regardless of *460its lack of connection to military service, to give rise to a “case arising in” the Armed Forces of the new Nation.7
This is borne out by provisions in the American Articles of 1776 that are comparable to those in the British Articles of War of 1774. See Section X, Article I, reprinted in 2 W. Winthrop, Military Law and Precedents 1494 (1896); Section XIII, Article 16, reprinted in 2 Winthrop, supra, at 1497; Section XVIII, Article 5, reprinted in 2 Winthrop, supra, at 1503. The provisions created military offenses where the crimes involved were service connected. This tradition continued after the adoption of the Constitution. With respect to the 1874 Articles of War, for example, Davis wrote:
“As to whether an act which is a civil crime is also a military offense no rule can be laid down which will cover all cases, for the reason that what may be a military offense under certain circumstances may lose that character under others. . . . But if the act be committed on a military reservation, or other ground occupied by the army, or in its neighborhood, so as to be in the constructive presence of the army; or if committed while on duty, particularly if the injury be to a member of the community whom it is the offender’s duty to protect; or if com*461mitted in the presence of other soldiers, or while in uniform; or if the offender use his military position, or that of another, for the purpose of intimidation or other unlawful influence or object — such facts would be sufficient to make it prejudicial to military discipline . . . Davis, supra, at 476.
Viewed historically, then, O’Callahan’s recognition of the service connection requirement did not signify a meaningful change in what could be tried in courts-martial. Quite the reverse: not until the enactment of the Uniform Code of Military Justice in 1950 did Congress attempt to give courts-martial the authority to try the crimes of murder and rape committed in peacetime within the United States. See Duke & Vogel, The Constitution and the Standing Army: Another Problem of Court-Martial Jurisdiction, 12 Vand. L. Rev. 435, 452-453 (1960). Common-law felonies in peacetime were only brought within the court-martial jurisdiction in 1916. Wiener, Courts-Martial and the Bill of Rights: The Original Practice I, 72 Harv. L. Rev. 1, 10-12 (1958). The Framers’ conception of what could properly be tried in a court-martial must have informed their understanding of what cases arise in the Armed Forces, thus permitting what would otherwise be unconstitutional infringements of Fifth and Sixth Amendment rights. The relatively recent expansion of the authority of military tribunals appears to disregard the Framers’ understanding.
Instead of acknowledging the Fifth Amendment limits on the crimes triable in a court-martial, the Court simply ignores them. But “[t]he concept that the Bill of Rights and other constitutional protections against arbitrary government are inoperative when they become inconvenient or when expediency dictates otherwise is a very dangerous doctrine and if allowed to flourish would destroy the benefit of a written Constitution and undermine the basis of our Government.” Reid v. Covert, 354 U. S. 1, 14 (1957) (plurality opin*462ion). The limitations may not, in the view of the majority, be desirable, but that does not mean they do not exist.
The requirement of service connection recognized in O’Callahan has a legitimate basis in constitutional language and a solid historical foundation. It should be applied in this case.
II
Application of the service connection requirement of O’Callahan, as further elaborated in Relford, v. Commandant, U. S. Disciplinary Barracks, 401 U. S. 355 (1971), demonstrates that petitioner’s Alaska crimes do not have an adequate service connection to support the exercise of court-martial jurisdiction. Petitioner’s offenses did not detract from the performance of his military duties. He committed these crimes while properly absent from his unit, and there was no connection between his assigned duties and his crimes. Nor did petitioner’s crimes threaten people or areas under military control. The crimes were committed in petitioner’s private home in the civilian community in Juneau, where there is not even a base for Coast Guard personnel. Petitioner’s acts were not likely to go unpunished; the court-martial judge determined that the offenses were of a type traditionally prosecuted by civilian courts, that such courts were available, and that, while the Alaska courts had deferred prosecution in light of the court-martial proceeding, the State had not declined to prosecute the offenses. Nor did the crimes implicate any authority stemming from the war power; they were committed within the territorial United States while the Nation was at peace.
Moreover, the crimes caused no measurable interference with military relationships. Though the victims were dependents of Coast Guard members, the court-martial judge found that there was only de minimis military interaction between petitioner and the fathers of the victims, and that the relationships between petitioner and the families of the vie*463tims “were founded primarily upon the ages and activities of the children and additionally upon common sporting interests, common spousal interest and employment and neighborly relationships,” App. to Pet. for Cert. 58a, rather than the connection of petitioner and the families through the Coast Guard.8 Because the crimes did not take place in an area within military control or have any effect on petitioner’s military duties, their commission posed no challenge to the maintenance of order in the local command. The military judge found that the Government had not demonstrated any impact of the offenses on “morale, discipline, [or] the reputation or the integrity of the Coast Guard in Juneau.” The only connection between the military and the offenses at issue was the fact that the victims were military dependents. But the military judge found explicitly that the military association of petitioner and the victims’ fathers did not facilitate petitioner’s crimes,9 and that “[t]he impact apparent in this case, that is, on the parents and the victims themselves is no different than that which would be produced by [a] civilian perpetrator.” Id., at 57a.
The military judge, after properly reviewing the Retford, factors, concluded correctly that they did not render petition*464er’s offenses service connected and dismissed the charges. Engaging in what can only be describéd as impermissible appellate factfinding,10 the Coast Guard Court of Military *465Review reversed the dismissal. 21 M. J. 512 (1985). It concluded that the military judge’s finding that the offenses had had no impact on morale or discipline was erroneous because the judge should have considered the effect the offenses would have had on the community in Juneau had they come to light while the victims and their families were still in Alaska, and the impact of the offenses on morale and discipline at Governors Island. Without remanding for further factfinding, the court held that the Alaska offenses had a direct impact upon the good order, discipline, morale, and welfare of Coast Guard personnel at Governors Island. Id., at 519. It further asserted, again without basis in the facts found by the military judge, that the Coast Guard’s interest in deterring the offenses was greater than that of the civilian authorities, and that the concerns of the victims’ parents would have been different had the offender been a civilian. Id., at 519-520. On the basis of these newly found facts, the Court of Military Review held petitioner’s crime sufficiently service connected to justify the exercise of court-martial jurisdiction. Id., at 522.
The Court of Military Appeals affirmed. 21 M. J. 251 (1986). While conceding that its “precedents involving off-base sex offenses against civilian dependents of military personnel would point to a different conclusion,” id., at 254, it concluded that a “recent development in our society” — specifically, an increase in concern for the victims of crimes — meant that sex offenses committed against young children of members of the military, which would have “a continuing effect on the victims and their families,” id., at 256, sufficed to establish service connection.
The military judge’s straightforward application of O’Callahan and Relford was plainly correct given the facts as he found them, facts that the reviewing courts have not demonstrated to have been clearly erroneous. The Court of Mili*466tary Appeals’ apparent conclusion that serious or disturbing crimes committed upon military dependents sufficed to create court-martial jurisdiction ignored this Court’s prior decisions.
The majority asserts that “the service connection approach, even as elucidated in Relford, has proved confusing and difficult for military courts to apply.” Ante, at 448. It is true that the test requires a careful, case-specific factual inquiry. But this is not beyond the capacity of the military courts. Indeed, the military judge in this case engaged in a thorough and thoughtful application of the Relford factors. It should not be surprising that such determinations may at times be difficult or time consuming or require the drawing of narrow distinctions. The trial of any person before a court-martial encompasses a deliberate decision to withhold procedural protections guaranteed by the Constitution. Denial of these protections is a very serious matter. The Framers declined to draw an easy line, like that established by the Court today, which would sweep an entire class of Americans beyond the reach of the Bill of Rights. Instead, they required that the protections of the Fifth and Sixth Amendments be applied in any case not “arising in” the Armed Forces. This requirement must not be discarded simply because it may be less expeditious than the majority deems appropriate.
Ill
O’Callahan v. Parker remains correct and workable today. The Court nonetheless insists on reopening a question which was finally and properly resolved in 1969. In doing so, it shows a blatant disregard for principles of stare decisis, and makes more dubious the presumption “that bedrock principles are founded in the law rather than in the proclivities of individuals.” Vasquez v. Hillery, 474 U. S. 254, 265 (1986). This in turn undermines “the integrity of our constitutional system of government, both in appearance and in fact.” *467Ibid.; see also Pollock v. Farmers’ Loan & Trust Co., 158 U. S. 601, 663 (1895) (Harlan, J., dissenting).
The Court’s willingness to overturn precedent may reflect in part its conviction, frequently expressed this Term, that members of the Armed Forces may be subjected virtually without limit to the vagaries of military control. See United States v. Stanley, post, p. 669; United States v. Johnson, 481 U. S. 681 (1987). But the Court’s decision today has, potentially, the broadest reach of any of these cases. Unless Congress acts to avoid the consequences of this case, every member of our Armed Forces, whose active duty members number in the millions, can now be subjected to court-martial jurisdiction — without grand jury indictment or trial by jury — for any offense, from tax fraud to passing a bad check, regardless of its lack of relation to “military discipline, morale and fitness.” Schlesinger v. Councilman, 420 U. S. 738, 761, n. 34 (1975). Today’s decision deprives our military personnel of procedural protections that are constitutionally mandated in trials for purely civilian offenses. The Court’s action today reflects contempt, both for the members of our Armed Forces and for the constitutional safeguards intended to protect us all. I dissent.
In any criminal proceeding brought against petitioner by the State of Alaska, the federal grand jury right would not attach; the Sixth Amendment right would apply by virtue of the Fourteenth Amendment. Whether petitioner would have these rights in any prosecution by a dual sovereign is not at issue here, however. The sole question is whether the Federal Government, when it proceeded against petitioner, was obliged to provide those safeguards guaranteed by the Fifth and Sixth Amendments. See Grafton v. United States, 206 U. S. 333, 352-354 (1907).
The language of this exception could be understood to mean that “cases arising in the land or naval forces,” as well as in the militia, are only excepted from the requirement of grand jury indictment or presentment “in actual service in time of War or public danger.” This Court, however, has interpreted the “time of war” provision as referring only to cases arising in the militia, not the land or naval forces. Johnson v. Sayre, 158 U. S. 109, 114 (1895). I am not convinced this reading of the Fifth Amendment is correct, but need not rely on a different interpretation here.
“This Court has constantly reiterated that the language of the Constitution where clear and unambiguous must be given its plain evident meaning.” Reid v. Covert, 354 U. S. 1, 8, n. 7 (1957) (plurality opinion).
See also Relford v. Commandant, U. S. Disciplinary Barracks, 401 U. S. 355, 362-363 (1971); Gosa v. Mayden, 413 U. S. 665, 672-673 (1973).
The majority contends that, before O’Callahan, this Court had held consistently that status as a member of the Armed Forces was an adequate basis for the assertion of court-martial jurisdiction. Ante, at 439. But a number of the precedents cited dealt with the assertion of court-martial jurisdiction over individuals who were not members of the Armed Forces and therefore, this Court ruled, did not come within the reach of Art. I, § 8, el. 14. See Kinsella v. United States ex rel. Singleton, 361 U. S. 234, 246-248 (1960) (military dependent in noncapital case); Reid v. Covert, supra, at 19-23 (plurality opinion) (military dependent in capital case); United States ex rel. Toth v. Quarles, 350 U. S. 11, 14-15 (1955) (discharged veteran); see also Grisham v. Hagan, 361 U. S. 278 (1960) (civilian military employee in capital case); McElroy v. United States ex rel. Guagliardo, 361 U. S. 281, 286 (1960) (civilian military employee in noncapital case). Having disposed of these cases on the threshold issue of the reach of the congressional power created by Art. I, § 8, cl. 14, the Court did not consider the limits imposed on the Article I power by the Fifth Amendment.
Several of the remaining cases cited involved crimes committed in the course of the performance of military duties that therefore clearly arose in the Armed Forces. See Grafton v. United States, supra (murder by Army private serving sentry duty on post); Johnson v. Sayre, supra (embezzlement of United States funds intended for the Naval service); Smith v. Whitney, 116 U. S. 167 (1886) (fraud on Navy contracts). In Smith, the Court concluded that “such conduct of a naval officer is a case arising in the naval forces, and therefore punishable by court martial under the articles and regulations made or approved by Congress in the exercise of the powers conferred upon it by the Constitution, to provide and maintain a navy, and to make rules for the government and regulation of the land and naval forces, without indictment or trial by jury.” Id., at 186.
The remaining cases cited by the majority are similarly inapposite. Coleman v. Tennessee, 97 U. S. 509 (1879), dealt with a murder committed by a soldier in time of war. In Ex parte Milligan, 4 Wall. 2 (1867), any reference to the reach of court-martial jurisdiction over persons in the Armed Services was dictum, since the holding of that case was that a civilian was improperly subjected to military jurisdiction during the Civil War in a State which had “upheld the authority of the government, and where the courts are open and their process unobstructed.” Id., at 121.
This attitude is evident in the Petition of Right in 1627, in which the two Houses of Parliament joined in a petition to the Crown to redress four major grievances, the last of which was the trial of soldiers by military commissions. See J. Tanner, English Constitutional Conflicts of the Seventeenth Century 61-62 (1983 reprint). The pertinent portion of the Petition stated:
“VII. [WJhereas no offender of what kind soever is exempted from the proceedings to be used, and punishments to be inflicted by the laws and *457statutes of this your realm: nevertheless of late time divers commissions . . . have issued forth . . . according to the justice of martial law, against such soldiers or mariners, or other dissolute persons joining with them, as should commit any murder, robbery, felony, mutiny or other outrage or misdemeanor whatsoever; and by such summary course and order as is agreeable to martial law, and as is used in armies in time of war, to proceed to the trial and condemnation of such offenders, and them to cause to be executed and put to death according to the law martial:
“VIII. By pretext whereof some of your Majesty’s subjects have been by some of the said Commissioners put to death, when and where, if by the laws and statutes of the land they had deserved death, by the same laws and statutes also they might, and by no other ought to have been judged and executed.” 3 Car. I, ch. 1.
The petition, which prayed revocation of the military commissions, ultimately received royal assent. Tanner, supra, at 64.
The Mutiny Act of 1689, 1 Wm. & Mary, ch. 5, went further, providing that “no man may be forejudged of Life or Limb, or subjected to any kind of Punishment by Martial Law, or in any other manner then by the Judgment of his Peers, and according to the known and established Laws of this Realm,” limiting this provision only with respect to “Soldiers who shall Mutiny or Stir up Sedition, or shall Defect Their Majesties Service,” who might “be brought to a more exemplary and speedy punishment than the usual forms of Law will allow.”
See also G. Davis, Military Law of the United States 437 (3d rev. ed. 1915) (“The acts of trespass, etc., indicated in this Article are made punishable as special breaches of discipline, and less for the protection of citizens than for the maintenance of the orderly behavior and morale of the military force”) (emphasis omitted); 2 W. Winthrop, Military Law and Precedents 1022 (1896) (“This Article, which, dating from an early period of the British law, first appeared in our code in the Articles of 1776, is designed, by making severely punishable trespasses committed by soldiers on the march or otherwise, to prevent straggling and maintain order and discipline in military commands, while at the same time availing to secure from intrusion and injury the premises and property of the inhabitants”) (footnotes omitted).
But cf. Cong. Globe, 37th Cong., 3d Sess., 953 (1863).
The Court contends that “American military records reflect trials by court-martial during the late 18th century for offenses against civilians and punishable under the civil law, such as theft and assault.” Ante, at 444. It apparently bases this conclusion on materials provided to the O’Callahan Court by the United States. See Brief for United States in O’Callahan v. Parker, O. T. 1968, No. 646, pp. 35-52, summarizing courts-martial during the period from 1775 to 1815 involving “apparently non-military criminal offenses committed by military personnel.” Id., at 35. I agree with the O’Callahan Court that, to the extent the courts-martial described there did not appear to deal with crimes that were committed during wartime, were committed by officers, or involved special military interests, the descriptions of the crimes “simply recite the offender and the offense and give no basis for judging the relationship of the offense to military discipline.” O’Callahan v. Parker, 395 U. S. 258, 270, n. 14 (1969).
See also 21 M. J. 512, 514 (C. G. C. M. R. 1985) (“A friendship had grown between the accused and both of the other families, grounded in one case, on the common sporting interests of bowling and basketball, and, in the other, on the proximity of living next door. The alleged victims came to the accused’s home on a regular basis to visit with his two sons. Both girls at one time played on a soccer team coached by the accused and they also bowled in a league in which the accused was active”).
See the military judge’s Supplemental Essential Findings of Fact, App. to Pet. for Cert. 62a (“To the extent that trust had a bearing on the opportunity for the alleged offenses, that trust arose out of friendships between the Solorio and Johnson and Solorio and Grantz families and not out of the respective fathers [sic] common association as members of the U. S. Coast Guard. The trust placed in a servicemember in general, and in the accused in particular, by virtue of status as a member of the Coast Guard was minimal and had no direct relationship to the offenses alleged”).
The appeal to the Court of Military Review was brought under Article 62, Uniform Code of Military Justice, 10 U. S. C. § 862 (1982 ed., Supp. III). Section 862(b) provides that “[i]n ruling on an appeal under this section, the Court of Military Review may act only with respect to matters of law, notwithstanding section 866(c) of this title (article 66(c)).” Title 10 U. S. C. 866(c), Article 66(c) of the Uniform Code of Military Justice, authorizes the Court of Military Review, in acting on findings of guilty and sentences, to “weigh the evidence, judge the credibility of witnesses, and determine controverted questions of fact.” See also United States v. Burris, 21 M. J. 140, 143-144 (Ct. Mil. App. 1985).
While the Court of Military Review acknowledged that it was bound by facts found at the trial level unless those findings were incorrect as a matter of law, 21 M. J., at 515, 517, it nonetheless proceeded to assume the facts necessary to its conclusion that the impact on the Coast Guard community at Governors Island created the requisite service connection to justify the exercise of court-martial jurisdiction. One judge on the Court of Military Review, dissenting in part from the court’s ruling, rejected the majority’s approach:
‘Where I depart from the majority is the holding that there was ‘service connection’ and therefore jurisdiction, in this case, as a matter of law. . . . [T]he [military] judge made no specific findings with respect to the possible effect of the offenses at Governors Island or on personnel under the authority and responsibility of the convening authority. Even if this case were before us for review under Article 66(c), U. C. M. J., 10 U. S. C. § 666(c), I would hesitate to determine that jurisdiction exists in light of this omission. . . . Since this case is before us for review under Article 62(b) U. C. M. J., 10 U. S. C. §862(b), I do not believe we are empowered to cure an omission from the essential findings of the trial judge.” Id., at 523 (Bridgman, J., concurring in part and dissenting in part).
Judge Bridgman would have remanded without prejudice to the accused’s right to renew his attack on the jurisdiction of the court-martial. Ibid.
The Court of Military Appeals suggested broadly that the Court of Military Review had violated its obligations under Article 62 in this case, but concluded that the violation was immaterial. See 21 M. J. 251, 254 (1986) (“A military judge’s factfinding power under Article 62 cannot be superseded by a Court of Military Review in an appeal under Article 62 ... . To some extent the Court of Military Review may have erred in this direc*465tion; but any such error is immaterial, because on the basis of indisputed facts, we conclude that the offenses in Alaska were service-connected”).