concurring in the result.
I concur in the result, on the narrow ground that where the offense is capital, Article 2 (ll)1 cannot constitutionally be applied to the trial of civilian dependents of members of the armed forces overseas in times of peace.
Since I am the only one among today’s majority who joined in the Court’s opinions of June 11, 1956, which sustained the court-martial jurisdiction in these cases, 351 U. S. 470, 487, I think it appropriate to state the reasons which led to my voting, first, to rehear these cases, 352 U. S. 901, and, now, to strike down that jurisdiction.
I.
The petitions for rehearing which were filed last summer afforded an opportunity for a greater degree of reflection upon the difficult issues involved in these cases than, at least for me, was possible in the short interval between the argument and decision of the cases in the closing days of last Term.2 As a result I became satisfied that this court-martial jurisdiction could in any event not be sustained upon the reasoning of our prior opinion. In essence, that reasoning was this: (1) Under In re Ross, 140 U. S. 453, and the Insular Cases,3 the requirement of a trial by an Article III court and the other specific safeguards of Article III and the Fifth and Sixth Amendments are not applicable to the trial of American citizens outside the United States; (2) there is thus no express constitutional prohibition against the use of courts-*66martial for such trials abroad; (3) the choice of a court-martial in cases such as these was “reasonable,” because of these women’s connection with the military, and therefore satisfied due process; (4) the court-martial jurisdiction was thus constitutional. I have since concluded that this analysis was not sound, for two reasons:
(1) The underlying premise of the prior opinion, it seems to me, is that under the Constitution the mere absence of a prohibition against an asserted power, plus the abstract reasonableness of its use, is enough to establish the existence of the power. I think this is erroneous. The powers of Congress, unlike those of the English Parliament, are constitutionally circumscribed. Under the Constitution Congress has only such powers as are expressly granted or those that are implied as reasonably necessary and proper to carry out the granted powers. Hence the constitutionality of the statute here in question must be tested, not by abstract notions of what is reasonable “in the large,” so to speak, but by whether the statute, as applied in these instances, is a reasonably necessary and proper means of implementing a power granted to Congress by the Constitution. To say that the validity of the statute may be rested upon the inherent “sovereign powers” of this country in its dealings with foreign nations seems to me to be no more than begging the question. As I now see it, the validity of this court-martial jurisdiction must depend upon whether the statute, as applied to these women, can be justified as an exercise of the power, granted to Congress by Art. I, § 8, cl. 14 of the Constitution, “To make Rules for the Government and Regulation of the land and naval Forces.” I can find no other constitutional power to which this statute can properly be related. I therefore think that we were wrong last Term in considering that we need not decide *67the case in terms of the Article I power. In my opinion that question squarely confronts us.
(2) I also think that we were mistaken in interpreting Ross and the Insular Cases as standing for the sweeping proposition that the safeguards of Article III and the Fifth and Sixth Amendments automatically have no application to the trial of American citizens outside the United States, no matter what the circumstances. Aside from the questionable wisdom of mortgaging the future by such a broad pronouncement, I am satisfied that our prior holding swept too lightly over the historical context in which this Court upheld the jurisdiction of the old consular and territorial courts in those cases. I shall not repeat what my brother Frankfurter has written on this subject, with which I agree. But I do not go as far as my brother Black seems to go on this score. His opinion, if I understand it correctly, in effect discards Ross and the Insular Cases as historical anomalies. I believe that those cases, properly understood, still have vitality, and that, for reasons suggested later, which differ from those given in our prior opinions, they have an important bearing on the question now before us.
II.
I come then to the question whether this court-martial jurisdiction can be justified as an exercise of Congress’ Article I power to regulate the armed forces.
At the outset, I cannot accept the implication of my brother Black’s opinion that this Article I power was intended to be unmodified by the Necessary and Proper Clause of the Constitution,4 and that therefore this power *68is incapable of expansion under changing circumstances. The historical evidence, in fact, shows quite the opposite. True, the records of the time indicate that the Founders shared a deep fear of an unchecked military branch. But what they feared was a military branch unchecked by the legislature, and susceptible of use by an arbitrary executive power.5 So far as I know, there is no evidence at all that the Founders intended to limit the power of the people, as embodied in the legislature, to make such laws in the regulation of the land and naval forces as are necessary to the proper functioning of those forces. In other words, there is no indication that any special limitation on the power of Congress, as opposed to the power of the executive, was subsumed in the grant of power to govern the land and naval forces. Alexander Hamilton, indeed, stated exactly the opposite: 6
“The authorities essential to the common defense are these: to raise armies; to build and equip fleets; to prescribe rules for the government of both; to direct their operations; to provide for their support. These powers ought to exist without limitation, because it is impossible to foresee or define the extent and variety of national exigencies, or the correspondent extent and variety of the means which may be necessary to satisfy them. The circumstances that endanger the safety of nations are infinite, and for this reason no constitutional shackles can wisely be *69imposed on the power to which the care of it is committed. This power ought to be coextensive with all the possible combinations of such circumstances; and ought to be under the direction of the same councils which are appointed to preside over the common defense.
“. . . Shall the Union be constituted the guardian of the common safety? Are fleets and armies and revenues necessary to this purpose? The government of the Union must be empowered to pass all laws, and to make all regulations which have relation to them. ...
“Every view we may take of the subject, as candid inquirers after truth, will serve to convince us, that it is both unwise and dangerous to deny the federal government an unconfined authority, as to all those objects which are intrusted to its management. ... A government, the constitution of .which renders it unfit to be trusted with all the powers which a free people ought to delegate to any government, would be an unsafe and improper depositary of the national interests. Wherever these can with propriety be confided, the coincident powers may safely accompany them.”
No less an authority than Chief Justice Marshall, in McCulloch v. Maryland, 4 Wheat. 316, has taught us that the Necessary and Proper Clause is to be read with all the powers of Congress, so that “where the law is not prohibited, and is really calculated to effect any of the objects entrusted to the government, to undertake here to inquire into the degree of its necessity, would be to pass the line which circumscribes the judicial department, and to tread on legislative ground.” Id., at 423.
*70I think it no answer to say, as my brother Black does, that “having run up against the steadfast bulwark of the Bill of Rights, the Necessary and Proper Clause cannot extend the scope of [Art. I] Clause 14.” Eor that simply begs the question as to whether there is such a collision, an issue to which I address myself below.
For analytical purposes, I think it useful to break down the issue before us into two questions: First, is there a rational connection between the trial of these army wives by court-martial and the power of Congress to make rules for the governance of the land and naval forces; in other words, is there any initial power here at all? Second, if there is such a rational connection, to what extent does this statute, though reasonably calculated to subserve an enumerated power, collide with other express limitations on congressional power; in other words, can this statute, however appropriate to the Article I power looked at in isolation, survive against the requirements of Article III and the Fifth and Sixth Amendments? I recognize that these two questions are ultimately one and the same, since the scope of the Article I power is not separable from the limitations imposed by Article III and the Fifth and Sixth Amendments. Nevertheless I think it will make for clarity of analysis to consider them separately.
A.
I assume, for the moment, therefore, that we may disregard other limiting provisions of the Constitution, and examine the Article I power in isolation. So viewed, I do not think the courts-martial of these army wives can be said to be an arbitrary extension of congressional power.
It is suggested that historically the Article I power was intended to embody a rigid and unchangeable self-limitation, namely, that it could apply only to those *71in the actual service of the armed forces.7 I cannot agree that this power has any such rigid content. First of all, the historical evidence presented by the Government convinces me that, at the time of the adoption of the Constitution, military jurisdiction was not thought to be rigidly limited to uniformed personnel. The fact is that it was traditional for “retainers to the camp” to be subjected to military discipline, that civilian dependents encamped with the armies were traditionally regarded as being in that class, and that the concept was not strictly limited to times of war.8 Indeed, the British, who are no less sensitive than we to maintaining the supremacy of civil justice, have recently enacted a law comparable to the statute involved here.9
Thinking, as I do, that Article I, still taking it in isolation, must be viewed as supplemented by the Necessary and Proper Clause, I cannot say that the court-martial jurisdiction here involved has no rational connection with the stated power. The Government, it seems to me, has *72made a strong showing that the court-martial of civilian dependents abroad has a close connection to the proper and effective functioning of our overseas military contingents. There is no need to detail here the various aspects of this connection, which have been well dealt with in the dissenting opinion of my brother Clark. Suffice it to say that to all intents and purposes these civilian dependents are part of the military community overseas,10 are so regarded by the host country, and must be subjected to the same discipline if the military commander is to have the power to prevent activities which would jeopardize the security and effectiveness of his command.11 The matter has been well summarized by General Palmer, Commander of the Eighth Army, stationed in Japan:
“Jurisdiction by courts-martial over all civilians accompanying the Army overseas is essential because of the manner in which U. S. Armed Forces personnel *73live in their overseas military communities. In this command, almost all personnel serving in or accompanying the U. S. Armed Forces live in or near separate, closely-knit U. S. military communities which are basically under the control, administration and supervision of the local U. S. Commander who is in turn responsive to the normal military chain of command. This responsibility which is vested in the military commander extends to the administration and supervision of the operation and use of all facilities and major activities of the community including the proper control of occupants and users which is inherent in such supervision overseas. In the absence of a supporting judicial system responsive to the same government as the military, such as is the case existing in the United States and overseas possessions, and as the law enforcement requirement stems primarily from the immediate unalterable responsibilities of the overseas commander and his subordinate commanders, it is essential that the commander be vested with the law enforcement authority commensurate with his responsibilities.”
It seems to me clear on such a basis that these dependents, when sent overseas by the Government, become pro tanto a part of the military community. I cannot say, therefore, that it is irrational or arbitrary for Congress to subject them to military discipline. I do not deal now, of course, with the problem of alternatives to court-martial jurisdiction; all that needs to be established at this stage is that, viewing Art. I, § 8, cl. 14 in isolation, subjection of civilian dependents overseas to court-martial jurisdiction can in no wise be deemed unrelated to the power of Congress to make all necessary and proper laws to insure the effective governance of our overseas land and naval forces.
*74B.
I turn now to the other side of the coin. For no matter how practical and how reasonable this jurisdiction might be, it still cannot be sustained if the Constitution guarantees to these army wives a trial in an Article III court, with indictment by grand jury and jury trial as provided by the Fifth and Sixth Amendments.
We return, therefore, to the Ross question: to what extent do these provisions of the Constitution apply outside the United States?
As I have already stated, I do not think that it can be said that these safeguards of the Constitution are never operative without the United States, regardless of the particular circumstances. On the other hand, I cannot agree with the suggestion that every provision of the Constitution must always be deemed automatically applicable to American citizens in every part of the world. For Ross and the Insular Cases do stand for an important proposition, one which seems to me a wise and necessary gloss on our Constitution. The proposition is, of course, not that the Constitution “does not apply” overseas, but that there are provisions in the Constitution which do not necessarily apply in all circumstances in every foreign place. In other words, it seems to me that the basic teaching of Ross and the Insular Cases is that there is no rigid and abstract rule that Congress, as a condition precedent to exercising power over Americans overseas, must exercise it subject to all the guarantees of the Constitution, no matter what the conditions and considerations are that would make adherence to a specific guarantee altogether impracticable and anomalous. To take but one example: Balzac v. Porto Rico, 258 U. S. 298, is not good authority for the proposition that jury trials need never be provided for American citizens tried by *75the United States abroad; but the case is good authority for the proposition that there is no rigid rule that jury trial must always be provided in the trial of an American overseas, if the circumstances are such that trial by jury would be impractical and anomalous. In other words, what Ross and the Insular Cases hold is that the particular local setting, the practical necessities, and the possible alternatives are relevant to a question of judgment, namely, whether jury trial should be deemed a necessary condition of the exercise of Congress’ power to provide for the trial of Americans overseas.
I think the above thought is crucial in approaching the cases before us. Decision is easy if one adopts the constricting view that these constitutional guarantees as a totality do or do not “apply” overseas. But, for me, the question is which guarantees of the Constitution should apply in view of the particular circumstances, the practical necessities, and the possible alternatives which Congress had before it. The question is one of judgment, not of compulsion. And so I agree with my brother ErankfuRter that, in view of Ross and the Insular Cases, we have before us a question analogous, ultimately, to issues of due process; one can say, in fact, that-the question of which specific safeguards of the Constitution are appropriately to be applied in a particular context overseas can be reduced to the issue of what process is “due” a defendant in the particular circumstances of a particular case.
On this basis, I cannot agree with the sweeping proposition that a full Article III trial, with indictment and trial by jury, is required in every case for the trial of a civilian dependent of a serviceman overseas. The Government, it seems to me, has made an impressive showing that at least for the run-of-the-mill offenses committed by dependents overseas, such a requirement would *76be as impractical and as anomalous as it would have been to require jury trial for Balzac in Porto Rico.12 Again, I need not go into details, beyond stating that except for capital offenses, such as we have here, to which, in my opinion, special considerations apply, I am by no means ready to say that Congress’ power to provide for trial by court-martial of civilian dependents overseas is limited by Article III and the Fifth and Sixth Amendments. *77Where, if at all, the dividing line should be drawn among cases not capital, need not now be decided. We are confronted here with capital offenses alone; and it seems to me particularly unwise now to decide more than we have to. Our far-flung foreign military establishments are a new phenomenon in our national life, and I think it would be unfortunate were we unnecessarily to foreclose, as my four brothers would do, our future consideration of the broad questions involved in maintaining the effectiveness of these national outposts, in the light of continuing experience with these problems.
So far as capital cases are concerned, I think they stand on quite a different footing than other offenses. In such cases the law is especially sensitive to demands for that procedural fairness which inheres in a civilian trial where the judge and trier of fact are not responsive to the command of the convening authority. I do not concede that whatever process is “due” an offender faced with a fine or a prison sentence necessarily satisfies the requirements of the Constitution in a capital case. The distinction is by no means novel, compare Powell v. Alabama, 287 U. S. 45, with Betts v. Brady, 316 U. S. 455; nor is it negligible, being literally that between life and death. And, under what I deem to be the correct view of Ross and the Insular Cases, it is precisely the kind of distinction which plays a large role in the process of weighing the competing considerations which lead to sound judgment upon the question whether certain safeguards of the Constitution should be given effect in the trial of an American citizen abroad. In fact, the Government itself has conceded that one grave offense, treason, presents a special case: “The gravity of this offense is such that we can well assume that, whatever difficulties may be involved in trial far from the scene of the offense . . . the trial should be in our courts.” I see no reason for not applying the same principle to any case where a civilian *78dependent stands trial on pain of life itself. The number of such cases would appear to be so negligible that the practical problems of affording the defendant a civilian trial would not present insuperable problems.
On this narrow ground I concur in the result in these cases.
50 U. S. C. § 552 (11).
The cases were argued on May 3, 1956, and decided on June 11, 1956.
Downes v. Bidwell, 182 U. S. 244; Hawaii v. Mankichi, 190 U. S. 197; Dorr v. United States, 195 U. S. 138; Balzac v. Porto Rico, 258 U. S. 298.
Article I, § 8, cl. 18 of the Constitution provides that Congress shall have the power “to make all Laws which shall be necessary and proper for carrying into Execution the foregoing Powers, and all other Powers vested by this Constitution in the Government of the United States, or in any Department or Officer thereof.”
Thus, proposals to limit the size of the standing army in times of peace to a specific number of men in the Constitution were defeated at the Constitutional Convention. See 5 Elliot’s Debates 442-443 (“no room for . . . distrust of the representatives of the people”). See also The Federalist, No. 24: “[T]he whole power of raising armies was lodged in the Legislature, not in the Executive; . . . this legislature was to be a popular body, consisting of the representatives of the people periodically elected . . . ."
The Federalist, No. 23.
To be sure, the opinion does “recognize that there might be circumstances where a person could be ‘in’ the armed services for purposes of [Art. I, § 8] Clause 14 even though he had not formally been inducted into the military or did not wear a uniform.” It continues, however, to state categorically that “wives, children and other dependents of servicemen cannot be placed in that category . . . .”
The essential element was thought to be, not so much that there be war, in the technical sense, but rather that the forces and their retainers be “in the field.” The latter concept, in turn, would seem to have extended to any area where the nature of the military position and the absence of civil authority made military control over the whole camp appropriate. See, in general, Blumenthal, Women Camp Followers of the American Revolution. The British history is the same. See, in particular, Samuel, Historical Account of the British Army and of the Law Military, pp. 691-692.
Army Act, 1955, 3 & 4 Eliz. II, c. 18, §209; and see Fifth Schedule, id., at 219.
These dependents are taken abroad only because their presence is deemed necessary to the morale and proper functioning of our armies overseas. They are transported at government expense, carry passports identifying them as service dependents, are admitted to the host country without visas, use military payment certificates, and receive the benefit of army postal facilities and privileges. They enjoy the tax exemptions and customs benefits of the military. They are treated at service hospitals, their children go to schools maintained by the Government, and they share with the military the recreational facilities provided by the Government. They are housed and furnished heat, light, fuel, water, and telephone service by the military, as well as receiving transportation, food, and clothing from military sources.
This necessity is particularly acute with regard to peculiarly “military” and “local” offenses which must be dealt with swiftly and effectively. Thus security regulations at these military installations must be enforced against civilian dependents as well as servicemen; the same is true of base traffic violations, black marketeering, and misuse of military customs and post-exchange privileges.
The practical circumstances requiring some sort of disciplinary jurisdiction have already been adverted to, supra, pp. 71-73. These circumstances take on weight when viewed in light of the alternatives available to Congress — certainly a crucial question in weighing the need for dispensing with particular constitutional guarantees abroad. What are these alternatives? (1) One is to try all offenses committed by civilian dependents abroad in the United States. But the practical problems in the way of such a choice are obvious and overwhelming. To require the transportation home for trial of every petty black marketeer or violator of security regulations would be a ridiculous burden on the Government, quite aside from the problems of persuading foreign witnesses to make the' trip and of preserving evidence. It can further be deemed doubtful in the extreme whether’foreign governments would permit crimes punishable under local law to be tried thousands of miles away in the United States. (2) Civilian trial overseas by the United States also presents considerable difficulties. If juries are required, the problem of jury recruitment would be difficult. Furthermore, it is indeed doubtful whether some foreign governments would accede to the creation of extraterritorial United States civil courts within their territories— courts which by implication would reflect on the fairness of their own tribunals and which would smack unpleasantly of consular courts set up under colonial “capitulations." (3) The alternative of trial in foreign courts, in at least some instances, is no more palatable. Quite aside from the fact that in some countries where we station troops the protections granted to criminal defendants compare unfavorably with our own minimum standards, the fact would remain that many of the crimes involved — particularly breaches of security — are not offenses under foreign law at all, and thus would go completely unpunished. Add to this the undesirability of foreign police carrying out investigations in our military installations abroad, and it seems to me clear that this alternative does not commend itself.