325 U.S. 226 (1945)

No. 84.
Supreme Court of United States.

Argued October 13, 1944.
Decided May 21, 1945.
Mr. W.H. Strickland for petitioners.

Hughes J. Rhodes, Assistant Attorney General of North Carolina, with whom Harry McMullan, Attorney General, and Mr. J.E. Tucker were on the brief, for respondent.

MR. JUSTICE FRANKFURTER delivered the opinion of the Court.

This case is here to review judgments of the Supreme Court of North Carolina, affirming convictions for bigamous cohabitation,[1] assailed on the ground that full faith and credit, as required by the Constitution of the United States, was not accorded divorces decreed by one of the courts of Nevada. Williams v. North Carolina, 317 U.S. 287, decided an earlier aspect of the controversy. It was there held that a divorce granted by Nevada, on a finding that one spouse was domiciled in Nevada, must be respected in North Carolina, where Nevada’s finding of domicil was not questioned, though the other spouse had neither appeared nor been served with process in Nevada and though recognition of such a divorce offended the policy of North Carolina. The record then before us did not present the question whether North Carolina had the power “to refuse full faith and credit to Nevada divorce decrees because, contrary to the findings of the Nevada court, North Carolina finds that no bona fide domicil was acquired in Nevada.” Williams v. North Carolina, supra, at 302. This is the precise issue which has emerged after retrial of the cause following our reversal. Its obvious importance brought the case here. 322 U.S. 725.

The implications of the Full Faith and Credit Clause, Article IV, ß 1 of the Constitution,[2] first received the sharp analysis of this Court in Thompson v. Whitman, 18 Wall. 457. Theretofore, uncritical notions about the scope of that Clause had been expressed in the early case of Mills v. Duryee, 7 Cranch 481. The “doctrine” of that case, as restated in another early case, was that “the judgment of a state court should have the same credit, validity, and effect, in every other court in the United States, which it had in the state where it was pronounced.” Hampton v. M’Connel, 3 Wheat. 234, 235. This utterance, when put to the test, as it was in Thompson v. Whitman, supra, was found to be too loose. Thompson v. Whitman made it clear that the doctrine of Mills v. Duryee comes into operation only when, in the language of Kent, “the jurisdiction of the court in another state is not impeached, either as to the subject matter or the person.” Only then is “the record of the judgment . . . entitled to full faith and credit.” 1 Kent, Commentaries (2d ed., 1832)[*] 261 n.b. The essence of the matter was thus put in what Thompson v. Whitman adopted from Story: “`The Constitution did not mean to confer [upon the States] a new power or jurisdiction, but simply to regulate the effect of the acknowledged jurisdiction over persons and things within their territory.'”[3] 18 Wall. 457, 462. In short, the Full Faith and Credit Clause puts the Constitution behind a judgment instead of the too fluid, ill-defined concept of “comity.”[4]

But the Clause does not make a sister-State judgment a judgment in another State. The proposal to do so was rejected by the Philadelphia Convention. 2 Farrand, The Records of the Federal Convention of 1787, 447-48.[5] “To give it the force of a judgment in another state, it must be made a judgment there.” M’Elmoyle v. Cohen, 13 Pet. 312, 325. It can be made a judgment there only if the court purporting to render the original judgment had power to render such a judgment. A judgment in one State is conclusive upon the merits in every other State, but only if the court of the first State had power to pass on the merits ó had jurisdiction, that is, to render the judgment.

“It is too late now to deny the right collaterally to impeach a decree of divorce made in another State, by proof that the court had no jurisdiction, even when the record purports to show jurisdiction . . .” It was “too late” more than forty years ago. German Savings Society v. Dormitzer, 192 U.S. 125, 128.

Under our system of law, judicial power to grant a divorce ó jurisdiction, strictly speaking ó is founded on domicil. Bell v. Bell, 181 U.S. 175; Andrews v. Andrews, 188 U.S. 14. The framers of the Constitution were familiar with this jurisdictional prerequisite, and since 1789 neither this Court nor any other court in the English-speaking world has questioned it. Domicil implies a nexus between person and place of such permanence as to control the creation of legal relations and responsibilities of the utmost significance. The domicil of one spouse within a State gives power to that State, we have held, to dissolve a marriage wheresoever contracted. In view of Williams v. North Carolina, supra, the jurisdictional requirement of domicil is freed from confusing refinements about “matrimonial domicil,” see Davis v. Davis, 305 U.S. 32, 41, and the like. Divorce, like marriage, is of concern not merely to the immediate parties. It affects personal rights of the deepest significance. It also touches basic interests of society. Since divorce, like marriage, creates a new status, every consideration of policy makes it desirable that the effect should be the same wherever the question arises.

It is one thing to reopen an issue that has been settled after appropriate opportunity to present their contentions has been afforded to all who had an interest in its adjudication. This applies also to jurisdictional questions. After a contest these cannot be relitigated as between the parties. Forsyth v. Hammond, 166 U.S. 506, 517; Chicago Life Ins. Co. v. Cherry, 244 U.S. 25, 30; Davis v. Davis, supra. But those not parties to a litigation ought not to be foreclosed by the interested actions of others; especially not a State which is concerned with the vindication of its own social policy and has no means, certainly no effective means, to protect that interest against the selfish action of those outside its borders. The State of domiciliary origin should not be bound by an unfounded, even if not collusive, recital in the record of a court of another State. As to the truth or existence of a fact, like that of domicil, upon which depends the power to exert judicial authority, a State not a party to the exertion of such judicial authority in another State but seriously affected by it has a right, when asserting its own unquestioned authority, to ascertain the truth or existence of that crucial fact.[6]

These considerations of policy are equally applicable whether power was assumed by the court of the first State or claimed after inquiry. This may lead, no doubt, to conflicting determinations of what judicial power is founded upon. Such conflict is inherent in the practical application of the concept of domicil in the context of our federal system.[7] See Worcester County Co. v. Riley, 302 U.S. 292; Texas v. Florida, 306 U.S. 398; District of Columbia v. Murphy, 314 U.S. 441. What was said in Worcester County Co. v. Riley, supra, is pertinent here. “Neither the Fourteenth Amendment nor the full faith and credit clause requires uniformity in the decisions of the courts of different states as to the place of domicil, where the exertion of state power is dependent upon domicil within its boundaries.” 302 U.S. 292, 299. If a finding by the court of one State that domicil in another State has been abandoned were conclusive upon the old domiciliary State, the policy of each State in matters of most intimate concern could be subverted by the policy of every other State. This Court has long ago denied the existence of such destructive power. The issue has a far reach. For domicil is the foundation of probate jurisdiction precisely as it is that of divorce. The ruling in Tilt v. Kelsey, 207 U.S. 43, regarding the probate of a will, is equally applicable to a sister-State divorce decree: “the full faith and credit due to the proceedings of the New Jersey court do not require that the courts of New York shall be bound by its adjudication on the question of domicil. On the contrary, it is open to the courts of any State in the trial of a collateral issue to determine upon the evidence produced the true domicil of the deceased.” 207 U.S. 43, 53.

Although it is now settled that a suit for divorce is not an ordinary adversary proceeding, it does not promote analysis, as was recently pointed out, to label divorce proceedings as actions in rem. Williams v. North Carolina, supra, at 297. But insofar as a divorce decree partakes of some of the characteristics of a decree in rem, it is misleading to say that all the world is party to a proceeding in rem. See Brigham v. Fayerweather, 140 Mass. 411, 413, 5 N.E. 265, quoted in Tilt v. Kelsey, supra, at 52. All the world is not party to a divorce proceeding. What is true is that all the world need not be present before a court granting the decree and yet it must be respected by the other forty-seven States provided ó and it is a big proviso ó the conditions for the exercise of power by the divorce-decreeing court are validly established whenever that judgment is elsewhere called into question. In short, the decree of divorce is a conclusive adjudication of everything except the jurisdictional facts upon which it is founded, and domicil is a jurisdictional fact. To permit the necessary finding of domicil by one State to foreclose all States in the protection of their social institutions would be intolerable.

But to endow each State with controlling authority to nullify the power of a sister State to grant a divorce based upon a finding that one spouse had acquired a new domicil within the divorcing State would, in the proper functioning of our federal system, be equally indefensible. No State court can assume comprehensive attention to the various and potentially conflicting interests that several States may have in the institutional aspects of marriage. The necessary accommodation between the right of one State to safeguard its interest in the family relation of its own people and the power of another State to grant divorces can be left to neither State.

The problem is to reconcile the reciprocal respect to be accorded by the members of the Union to their adjudications with due regard for another most important aspect of our federalism whereby “the domestic relations of husband and wife .. . were matters reserved to the States,” Popovici v. Agler, 280 U.S. 379, 383-84, and do not belong to the United States. In re Burrus, 136 U.S. 586, 593-94. The rights that belong to all the States and the obligations which membership in the Union imposes upon all, are made effective because this Court is open to consider claims, such as this case presents, that the courts of one State have not given the full faith and credit to the judgment of a sister State that is required by Art. IV, ß 1 of the Constitution.

But the discharge of this duty does not make of this Court a court of probate and divorce. Neither a rational system of law nor hard practicality calls for our independent determination, in reviewing the judgment of a State court, of that rather elusive relation between person and place which establishes domicil. “It is not for us to retry the facts,” as was held in a case in which, like the present, the jurisdiction underlying a sister-State judgment was dependent on domicil. Burbank v. Ernst, 232 U.S. 162, 164. The challenged judgment must, however, satisfy our scrutiny that the reciprocal duty of respect owed by the States to one another’s adjudications has been fairly discharged, and has not been evaded under the guise of finding an absence of domicil and therefore a want of power in the court rendering the judgment.

What is immediately before us is the judgment of the Supreme Court of North Carolina. We have authority to upset it only if there is want of foundation for the conclusion that that Court reached. The conclusion it reached turns on its finding that the spouses who obtained the Nevada decrees were not domiciled there. The fact that the Nevada court found that they were domiciled there is entitled to respect, and more. The burden of undermining the verity which the Nevada decrees import rests heavily upon the assailant. But simply because the Nevada court found that it had power to award a divorce decree cannot, we have seen, foreclose reexamination by another State. Otherwise, as was pointed out long ago, a court’s record would establish its power and the power would be proved by the record. Such circular reasoning would give one State a control over all the other States which the Full Faith and Credit Clause certainly did not confer. Thompson v. Whitman, supra. If this Court finds that proper weight was accorded to the claims of power by the court of one State in rendering a judgment the validity of which is pleaded in defense in another State, that the burden of overcoming such respect by disproof of the substratum of fact ó here domicil ó on which such power alone can rest was properly charged against the party challenging the legitimacy of the judgment, that such issue of fact was left for fair determination by appropriate procedure, and that a finding adverse to the necessary foundation for any valid sister-State judgment was amply supported in evidence, we cannot upset the judgment before us. And we cannot do so even if we also found in the record of the court of original judgment warrant for its finding that it had jurisdiction. If it is a matter turning on local law, great deference is owed by the courts of one State to what a court of another State has done. See Michigan Trust Co. v. Ferry, 228 U.S. 346. But when we are dealing as here with an historic notion common to all English-speaking courts, that of domicil, we should not find a want of deference to a sister State on the part of a court of another State which finds an absence of domicil where such a conclusion is warranted by the record.

When this case was first here, North Carolina did not challenge the finding of the Nevada court that petitioners had acquired domicils in Nevada. For her challenge of the Nevada decrees, North Carolina rested on Haddock v. Haddock, 201 U.S. 562. Upon retrial, however, the existence of domicil in Nevada became the decisive issue. The judgments of conviction now under review bring before us a record which may be fairly summarized by saying that the petitioners left North Carolina for the purpose of getting divorces from their respective spouses in Nevada and as soon as each had done so and married one another they left Nevada and returned to North Carolina to live there together as man and wife. Against the charge of bigamous cohabitation under ß 14-183 of the North Carolina General Statutes, petitioners stood on their Nevada divorces and offered exemplified copies of the Nevada proceedings.[8] The trial judge charged that the State had the burden of proving beyond a reasonable doubt that (1) each petitioner was lawfully married to one person; (2) thereafter each petitioner contracted a second marriage with another person outside North Carolina; (3) the spouses of petitioners were living at the time of this second marriage; (4) petitioners cohabited with one another in North Carolina after the second marriage. The burden, it was charged, then devolved upon petitioners “to satisfy the trial jury, not beyond a reasonable doubt nor by the greater weight of the evidence, but simply to satisfy” the jury from all the evidence, that petitioners were domiciled in Nevada at the time they obtained their divorces. The court further charged that “the recitation” of bona fide domicil in the Nevada decree was “prima facie evidence” sufficient to warrant a finding of domicil in Nevada but not compelling “such an inference.” If the jury found, as they were told, that petitioners had domicils in North Carolina and went to Nevada “simply and solely for the purpose of obtaining” divorces, intending to return to North Carolina on obtaining them, they never lost their North Carolina domicils nor acquired new domicils in Nevada. Domicil, the jury was instructed, was that place where a person “has voluntarily fixed his abode . . . not for a mere special or temporary purpose, but with a present intention of making it his home, either permanently or for an indefinite or unlimited length of time.”

The scales of justice must not be unfairly weighted by a State when full faith and credit is claimed for a sister-State judgment. But North Carolina has not so dealt with the Nevada decrees. She has not raised unfair barriers to their recognition. North Carolina did not fail in appreciation or application of federal standards of full faith and credit. Appropriate weight was given to the finding of domicil in the Nevada decrees, and that finding was allowed to be overturned only by relevant standards of proof. There is nothing to suggest that the issue was not fairly submitted to the jury and that it was not fairly assessed on cogent evidence.

State courts cannot avoid review by this Court of their disposition of a constitutional claim by casting it in the form of an unreviewable finding of fact. Norris v. Alabama, 294 U.S. 587, 590. This record is barren of such attempted evasion. What it shows is that petitioners, long-time residents of North Carolina, came to Nevada, where they stayed in an auto-court for transients, filed suits for divorce as soon as the Nevada law permitted, married one another as soon as the divorces were obtained, and promptly returned to North Carolina to live. It cannot reasonably be claimed that one set of inferences rather than another regarding the acquisition by petitioners of new domicils in Nevada could not be drawn from the circumstances attending their Nevada divorces. It would be highly unreasonable to assert that a jury could not reasonably find that the evidence demonstrated that petitioners went to Nevada solely for the purpose of obtaining a divorce and intended all along to return to North Carolina. Such an intention, the trial court properly charged, would preclude acquisition of domicils in Nevada. See Williamson v. Osenton, 232 U.S. 619. And so we cannot say that North Carolina was not entitled to draw the inference that petitioners never abandoned their domicils in North Carolina, particularly since we could not conscientiously prefer, were it our business to do so, the contrary finding of the Nevada court.

If a State cannot foreclose, on review here, all the other States by its finding that one spouse is domiciled within its bounds, persons may, no doubt, place themselves in situations that create unhappy consequences for them. This is merely one of those untoward results inevitable in a federal system in which regulation of domestic relations has been left with the States and not given to the national authority. But the occasional disregard by any one State of the reciprocal obligations of the forty-eight States to respect the constitutional power of each to deal with domestic relations of those domiciled within its borders is hardly an argument for allowing one State to deprive the other forty-seven States of their constitutional rights. Relevant statistics happily do not justify lurid forebodings that parents without number will disregard the fate of their offspring by being unmindful of the status of dignity to which they are entitled. But, in any event, to the extent that some one State may, for considerations of its own, improperly intrude into domestic relations subject to the authority of the other States, it suffices to suggest that any such indifference by a State to the bond of the Union should be discouraged, not encouraged.

In seeking a decree of divorce outside the State in which he has theretofore maintained his marriage, a person is necessarily involved in the legal situation created by our federal system whereby one State can grant a divorce of validity in other States only if the applicant has a bona fide domicil in the State of the court purporting to dissolve a prior legal marriage. The petitioners therefore assumed the risk that this Court would find that North Carolina justifiably concluded that they had not been domiciled in Nevada. Since the divorces which they sought and received in Nevada had no legal validity in North Carolina and their North Carolina spouses were still alive, they subjected themselves to prosecution for bigamous cohabitation under North Carolina law. The legitimate finding of the North Carolina Supreme Court that the petitioners were not in truth domiciled in Nevada was not a contingency against which the petitioners were protected by anything in the Constitution of the United States. A man’s fate often depends, as for instance in the enforcement of the Sherman Law, on far greater risks that he will estimate “rightly, that is, as the jury subsequently estimates it, some matter of degree. If his judgment is wrong, not only may he incur a fine or a short imprisonment, as here; he may incur the penalty of death.” Nash v. United States, 229 U.S. 373, 377. The objection that punishment of a person for an act as a crime when ignorant of the facts making it so, involves a denial of due process of law has more than once been overruled. In vindicating its public policy and particularly one so important as that bearing upon the integrity of family life, a State in punishing particular acts may provide that “he who shall do them shall do them at his peril and will not be heard to plead in defense good faith or ignorance.” United States v. Balint, 258 U.S. 250, 252, quoting Shevlin-Carpenter Co. v. Minnesota, 218 U.S. 57, 69-70. Mistaken notions about one’s legal rights are not sufficient to bar prosecution for crime.

We conclude that North Carolina was not required to yield her State policy because a Nevada court found that petitioners were domiciled in Nevada when it granted them decrees of divorce. North Carolina was entitled to find, as she did, that they did not acquire domicils in Nevada and that the Nevada court was therefore without power to liberate the petitioners from amenability to the laws of North Carolina governing domestic relations. And, as was said in connection with another aspect of the Full Faith and Credit Clause, our conclusion “is not a matter to arouse the susceptibilities of the States, all of which are equally concerned in the question and equally on both sides.” Fauntleroy v. Lum, 210 U.S. 230, 238.

As for the suggestion that Williams v. North Carolina, supra, foreclosed the Supreme Court of North Carolina from ordering a second trial upon the issue of domicil, it suffices to refer to our opinion in the earlier case.


MR. JUSTICE MURPHY, concurring.

While I join in the opinion of the Court, certain considerations compel me to state more fully my views on the important issues presented by this case.

The State of Nevada has unquestioned authority, consistent with procedural due process, to grant divorces on whatever basis it sees fit to all who meet its statutory requirements. It is entitled, moreover, to give to its divorce decrees absolute and binding finality within the confines of its borders.

But if Nevada’s divorce decrees are to be accorded full faith and credit in the courts of her sister states it is essential that Nevada have proper jurisdiction over the divorce proceedings. This means that at least one of the parties to each ex parte proceeding must have a bona fide domicil within Nevada for whatever length of time Nevada may prescribe.

This elementary principle has been reiterated by this Court many times. In Bell v. Bell, 181 U.S. 175, this Court held that “because neither party had a domicil in Pennsylvania” the Pennsylvania court had no jurisdiction to grant a divorce and its decree “was entitled to no faith and credit in New York or in any other state.” The same rule was applied in the companion case of Streitwolf v. Streitwolf, 181 U.S. 179. Referring to these two prior cases as holding that “domicil was in any event the inherent element upon which the jurisdiction must rest,” the Court in Andrews v. Andrews, 188 U.S. 14, repeated that bona fide domicil in a state is “essential to give jurisdiction to the courts of such state to render a decree of divorce which would have extra-territorial effect.” The Andrews case made it clear, moreover, that this requirement of domicil is not merely a matter of state law. It was stated specifically that “without reference to the statute of South Dakota and in any event” domicil in South Dakota was necessary. 188 U.S. at 41. All of the opinions in Haddock v. Haddock, 201 U.S. 562, recognized this principle, with Mr. Justice Brown’s dissenting opinion stating that “the courts of one state may not grant a divorce against an absent defendant to any person who has not acquired a bona fide domicil in that state.” Finally, in Williams v. North Carolina, 317 U.S. 287, the Court acknowledged that the plaintiff’s domicil in a state “is recognized in the Haddock case and elsewhere (Beale, Conflict of Laws, ß 110.1) as essential in order to give the court jurisdiction which will entitle the divorce decree to extraterritorial effect, at least when the defendant has neither been personally served nor entered an appearance.” See also Atherton v. Atherton, 181 U.S. 155.

The jury has here found that the petitioner’s alleged domicil in Nevada was not a bona fide one, which in common and legal parlance means that it was acquired fraudulently, deceitfully or in bad faith. This means, in other words, that the jury found that the petitioners’ residence in Nevada for six weeks was not accompanied by a bona fide intention to make Nevada their home and to remain there permanently or at least for an indefinite time, as required even by Nevada law. Lamb v. Lamb, 57 Nev. 421, 430, 65 P.2d 872. This conclusion is supported by overwhelming evidence satisfying whatever standard of proof may be propounded. Under these circumstances there is no reason to doubt the efficacy of jury trials in relation to the question of domicil or to speculate as to whether another jury might have reached a different verdict on the same set of facts.

Thus the court below properly concluded that Nevada was without jurisdiction so as to give extraterritorial validity to the divorce decrees and that North Carolina was not compelled by the Constitution to give full faith and credit to the Nevada decrees. North Carolina was free to consider the original marriages still in effect, the Nevada divorces to be invalid, and the Nevada marriage to be bigamous, thus giving the Nevada marriage the same force and effect that Nevada presumably would have given it had Nevada considered the original marriages still outstanding. Cf. State v. Zichfeld, 23 Nev. 304, 46 P. 802.

By being domiciled and living in North Carolina, petitioners secured all the benefits and advantages of its government and participated in its social and economic life. As long as petitioners and their respective spouses lived there and retained that domicil, North Carolina had the exclusive right to regulate the dissolution of their marriage relationships. However harsh and unjust North Carolina’s divorce laws may be thought to be, petitioners were bound to obey them while retaining residential and domiciliary ties in that state.

No justifiable purpose is served by imparting constitutional sanctity to the efforts of petitioners to establish a false and fictitious domicil in Nevada. Such a result would only tend to promote wholesale disregard of North Carolina’s divorce laws by its citizens, thus putting an end to “the existence of all efficacious power on the subject of divorce.” Andrews v. Andrews, supra, 32. Certainly no policy of Nevada dictates lending the full faith and credit clause to protect actions grounded in deceit. Nevada has a recognizable interest in granting only two types of ex parte divorces: (a) those effective solely within the borders of Nevada, and (b) those effective everywhere on the ground that at least one of the parties had a bona fide domicil in the state at the time the decree was granted. Neither type of divorce is involved here. And Nevada has no interest that we can respect in issuing divorce decrees with extraterritorial effect to those who are domiciled elsewhere and who secure sham domicils in Nevada solely for divorce purposes.

There are no startling or dangerous implications in the judgment reached by the Court in this case. All of the uncontested divorces that have ever been granted in the forty-eight states are as secure today as they were yesterday or as they were before our previous decision in this case. Those based upon fraudulent domicils are now and always have been subject to later reexamination with possible serious consequences.

Whatever embarrassment or inconvenience resulting to those who have made property settlements, contracted new marriages or otherwise acted in reliance upon divorce decrees obtained under conditions found to exist in this case is not insurmountable. The states have adequate power, if they desire to exercise it, to enact legislation providing for means of validating any such property settlements or marriages or of relieving persons from other unfortunate consequences.

Nor are any issues of civil liberties at stake here. It is unfortunate that the petitioners must be imprisoned for acts which they probably committed in reliance upon advice of counsel and without intent to violate the North Carolina statute. But there are many instances of punishment for acts whose criminality was unsuspected at the time of their occurrence. Indeed, for nearly three-quarters of a century or more individuals have been punished under bigamy statutes for doing exactly what petitioners have done. People v. Dawell, 25 Mich. 247; State v. Armington, 25 Minn. 29, 56 S.E. 673; People v. Baker, 76 N.Y. 78; State v. Westmoreland, 76 S.C. 145. Petitioners especially must be deemed to have been aware of the possible criminal consequences of their actions in view of the previously settled North Carolina law on the matter. State v. Herron, 175 N.C. 754, 94 S.E. 698. This case, then, adds no new uncertainty and comes as no surprise for those who act fraudulently in establishing a domicil and who disregard the laws of their true domiciliary states.

As Mr. Justice Holmes said in his dissenting opinion in the Haddock case, 201 U.S. at 628, “I do not suppose that civilization will come to an end whichever way this case is decided.” Difficult problems inevitably arise from the fact that people move about freely among the forty-eight states, each of which has its own policies and laws. Until the federal government is empowered by the Constitution to deal uniformly with the divorce problem or until uniform state laws are adopted, it is essential that definite lines of demarcation be made as regards the scope and extent of the varying state practices. See 91 Cong. Rec. 4238-4241 (May 3, 1945). This case illustrates the drawing of one such line, a line that has been drawn many times before without too unfortunate dislocations resulting among those citizens of a divorced status. There is no reason to believe that any different or more serious consequences will result from retracing that line today.

The CHIEF JUSTICE and MR. JUSTICE JACKSON join in these views.


Once again the ghost of “unitary domicil” returns on its perpetual round, in the guise of “jurisdictional fact,” to upset judgments, marriages, divorces, undermine the relations founded upon them, and make this Court the unwilling and uncertain arbiter between the concededly valid laws and decrees of sister states. From Bell and Andrews to Davis to Haddock to Williams and now back to Haddock and Davis through Williams again[1] ó is the maze the Court has travelled in a domiciliary wilderness, only to come out with no settled constitutional policy where one is needed most.

Nevada’s judgment has not been voided. It could not be, if the same test applies to sustain it as upholds the North Carolina convictions.[2] It stands, with the marriages founded upon it, unimpeached. For all that has been determined or could be, unless another change is in the making, petitioners are lawful husband and wife in Nevada. Williams v. North Carolina I, 317 U.S. 287; Williams v. North Carolina II, ante, p. 226. They may be such everywhere outside North Carolina. Lawfully wedded also, in North Carolina, are the divorced spouse of one and his wife, taken for all we know in reliance upon the Nevada decree.[3] That is, unless another jury shall find they too are bigamists for their reliance. No such jury has been impanelled. But were one called, it could pronounce the Nevada decree valid upon the identical evidence from which the jury in this case drew the contrary conclusion. That jury or it and another, if petitioners had been tried separately, could have found one guilty, the other innocent, upon that evidence unvaried by a hair. And, by the Court’s test, we could do nothing but sustain the contradictory findings in all these cases.

I do not believe the Constitution has thus confided to the caprice of juries the faith and credit due the laws and judgments of sister states. Nor has it thus made that question a local matter for the states themselves to decide. Were all judgments given the same infirmity, the full faith and credit clause would be only a dead constitutional letter.

I agree it is not the Court’s business to determine policies of divorce. But precisely its function is to lay the jurisdictional foundations upon which the states’ determinations can be made effective, within and without their borders. For in the one case due process, in the other full faith and credit, commands of equal compulsion upon the states and upon us, impose that duty.

I do not think we perform it, we rather abdicate, when we confide the ultimate decision to the states or to their juries. This we do when, for every case that matters, we make their judgment conclusive. It is so in effect when the crucial concept is as variable and amorphous as “domicil,” is always a conclusion of “ultimate fact,” and can be established only by proof from which, as experience shows, contradictory inferences may be made as strikes the local trier’s fancy. The abdication only becomes more obviously explicit when we avowedly confess that the faith and credit due may be determined either way, wherever “it cannot reasonably be claimed that one set of inferences rather than another” could not be drawn concerning the very matter determined by the judgment; and the final choice upon such a balance is left with the local jury.

No more unstable foundation, for state policies or marital relations, could be formulated or applied. In no region of adjudication or legislation is stability more essential for jurisdictional foundations. Beyond abnegating our function, we make instability itself the constitutional policy when the crux is so conceived and pivoted.


What, exactly, are the effects of the decision? The Court is careful not to say that Nevada’s judgment is not valid in Nevada. To repeat, the Court could not so declare it, unless a different test applies to sustain that judgment than supports North Carolina’s. Presumably the same standard applies to both; and each state accordingly is free to follow its own policy, wherever the evidence, whether the same or different, permits conflicting inferences of domicil, as it always does when the question becomes important.[4]

This must be true unless, contrary to the disclaimer, this Court itself is “to retry the facts.” The Court no more could say that the Nevada evidence permitted no conclusion of domicil there than it now can say the North Carolina evidence would not allow a finding either way. This apparently is conceded. The proof was not identical. But it was not so one-sided in either case that only one conclusion was compelled. The evidence in Nevada was neither that strong nor that weak.[5] Seldom, if ever, is it so.

The necessary conclusion follows that the Nevada decree was valid and remains valid within her borders. So the marriage is good in Nevada, but void in North Carolina, just as it was before “the jurisdictional requirement of domicil [was] freed from confusing refinements about `matrimonial domicil,’ see Davis v. Davis, 305 U.S. 32, 41, and the like.” See also Haddock v. Haddock, 201 U.S. 562.

The characterization “in rem” has been dropped. But it is clear from the result and from the opinion that the more “confusing refinements” and consequences, including the anomalous status Haddock approved, have not completely disappeared. We are not told definitely whether Nevada’s adjudication or North Carolina’s must be respected, when the question is raised in some one of the other forty-six states. But one thing we do know. “The State of domiciliary origin should not be bound by an unfounded, even if not collusive, recital in the record of a court of another State.” The opinion goes on to repeat: “If a finding by the court of one State that domicil in another State has been abandoned were conclusive upon the old domiciliary State, the policy of each State in matters of most intimate concern could be subverted by the policy of every other State.” (Emphasis added.)

The question is not simply pertinent, it is imperative, whether “matrimonial domicil” has not merely been recast and returned to the play under the common law’s more ancient name of “domicil of origin.” For North Carolina is the only state which, upon the facts, conceivably could qualify either as “matrimonial domicil” or as “domicil of origin,” whether or not they differ. Under the former conception it was at least doubtful whether sheer reexamination of “the jurisdictional fact” previously determined could be made outside the state granting the divorce and the state of “matrimonial domicil.”[6] Now we are told the decree “must be respected by the other forty-seven States provided ó and it is a big proviso ó the conditions for the exercise of power by the divorce-decreeing court are validly established whenever that judgment is elsewhere called into question.” (Emphasis added.)

If this means what it says, the proviso is big. It swallows the provision. Unless “matrimonial domicil,” banished in Williams I, has returned renamed in Williams II, every decree becomes vulnerable in every state. Every divorce, wherever granted, whether upon a residence of six weeks, six months or six years, may now be reexamined by every other state, upon the same or different evidence, to redetermine the “jurisdictional fact,” always the ultimate conclusion of “domicil.” For the grounds of the decision wholly negate that its effect can be limited to decrees of states having so-called “liberal” divorce policies; or to decrees recently granted; or to cases where different evidence is presented. It is implicit and inherent in the “unitary-domicil, jurisdictional-fact, permissible-inference” rule that any decree, granted after any length of time, upon any ground for divorce, and however solid the proof, may be reexamined either by “the state of domiciliary origin” or by any other state, as the case uncertainly may be. And all that is needed, to disregard it, is some evidence from which a jury reasonably may conclude there was no domiciliary intent when the decree was rendered. That is, unless the Court means to reserve decision upon the weight of the evidence and thus “to retry the facts,” contrary to its declared intention, in some case or cases not defined or indicated.


Obviously more is involved than full faith and credit for judgments of other states. Beneath the judgment of Nevada lie her statutory law and policy. These too are denied recognition. This is not a case in which the denial extends, or could extend, to the judgment alone. For the North Carolina verdict and judgment do not purport to rest on any finding of fraud or other similar ground, whereby the petitioners procured judgments from the Nevada courts which the manner of their procurement vitiates.[7]

No such issue, impeaching the Nevada decree, has been made. The state asked no instructions on such a theory and none were given.[8] The verdict and judgment therefore have not determined and do not rest upon any such ground.

In view of this fact I am completely at loss to understand what is meant, in the context of this case, by “an unfounded, even if not collusive, recital” which the state of domiciliary origin, perhaps others too, is free to disregard. The statement itself negates collusion as a ground for the decision. And, as I read the remainder of the opinion, it concedes and must concede, if the two judgments are to be tested alike, that the Nevada decree was not unfounded. The shape the issues have taken compels this conclusion.

Accordingly the case must be considered as shorn of any element of fraud, deceit or evasion of Nevada’s law, of showing that the Nevada court was imposed upon in any way or did other than apply the Nevada law according to its true intent and purpose. It must be taken also as devoid of any showing that Nevada failed in any way to comply with every requirement this Court has made respecting jurisdiction or due process of law, for rendering a valid divorce decree. Williams v. North Carolina, 317 U.S. 287.

The case therefore stands stripped of every difference, presently material, from the Nevada proceedings save two. There was none, jurisdictionally, in the issues. There was only different evidence upon which the same issue was determined in opposite fashions. And the states had different policies concerning divorce.

The difference in the evidence affected solely events taking place after the Nevada decree, the return to North Carolina and the cohabitation there. Ordinarily, valid judgments are not overturned, Schneiderman v. United States, 320 U.S. 118, or disregarded upon such retroactive proof.[9] But here this proof was not tendered in attack upon the Nevada decree. It was offered and admitted exclusively to relitigate the same issue that decree had determined, upon adequate evidence and in full compliance with Nevada law and the federal law giving Nevada jurisdiction to determine it. Williams I; Williams II. Its sole function was to show that petitioners did not have the very intent the Nevada court, with eyes not blinded,[10] had found they possessed.

Moreover, the character of the Court’s ruling makes the difference in the evidence, as it bore upon the controlling issue, of no materiality. It is not held that denial of credit will be allowed, only if the evidence is different or depending in any way upon the character or the weight of the difference. The test is not different evidence. It is evidence, whether the same or different and, if different, without regard to the quality of the difference, from which an opposing set of inferences can be drawn by the trier of fact “not unreasonably.” Presumably the Court will not “retry the facts” in either case.

But it does not define “not unreasonably.” It vaguely suggests a supervisory function, to be exercised when the denial strikes its sensibilities as wrong, by some not stated standard. So to suspend the matter is not law. It is only added uncertainty.

If the Court means not “to retry the facts,” the suggestion is wholly out of place. Then the test will be as it is in other cases where the question is whether a jury’s verdict will be sustained, upon an issue alleging want of supporting evidence. There will be no “weighing.” There will be only examination for sufficiency, with the limits marked by “scintillas” and the like.[11]

If this is the test, for all practical purposes the Court might as well declare outright that states of domiciliary origin are free to deny faith and credit to divorces granted elsewhere. For the case will be rare indeed where, by this standard, “domicil” can be determined as a matter of law, when divorce has been secured after departure from such a state. These are the only cases that matter. The issue does not arise with stay-at-homes. With others, it always can be raised and nearly always with “some” evidence, more than a “scintilla,” to sustain both contentions.

But if the test is different, “weighing” necessarily becomes involved and implicitly is what has been done in this case, notwithstanding the disclaimer. In that event, the crux of jurisdiction becomes the difference in the evidence; in this case, the return to North Carolina and cohabitation there.

If this is the decision’s intended effect, it should be squarely so declared. Too much hangs for too many people and for the states themselves upon beclouding it with a “different set of inferences ó refusal to retry the facts” gloss or otherwise. It cannot be assumed that the matter will affect only a few. For this has become a nation of transient people. Lawyers everywhere advise for or against divorce and courts grant or deny it, depending not on the probability that the case will come here, but on what is done here with the few cases which do come. The matter is altogether too serious, for too many, for glossing over the crucial basis of decision.

Whether the one test or the other is intended, or perhaps still another not suggested, North Carolina’s action comes down to sheer denial of faith and credit to Nevada’s law and policy, not merely to her judgment; and the decision here, to approval of this denial. The real difference, in my opinion the only material one, as the issues and the decision have been made on this record, is that one suit and judgment took place in Nevada, the other in North Carolina, and the two states have different policies relating to divorce. Nor does the degree or quality of the difference in policies matter. It also is not weighed.[12] The difference may be small for anything that is said, yet there is freedom to withhold credit.

If this is the test, every divorce granted a person who has come from another state is vulnerable wherever state policies differ, as they do universally if no account is taken of the weight of difference.

It is always a serious matter for us to say that one state is bound to give effect to another’s decision, founded in its different policy. That mandate I would not join in any case if not compelled by the only authority binding both the states and ourselves. Conceivably it might have been held that the full faith and credit clause has no application to the matters of marriage and divorce. But the Constitution has not left open that choice. And such has not been the course of decision. The clause applies, but from today it would seem only to compel “respect” or something less than faith and credit, whenever a jury concludes “not unreasonably,” by ultimate inference from the always conflicting circumstantial evidence, that it should not apply. Wherever that situation exists, the finding that there was no “bona fide” domiciliary intent comes in every practical effect to this and nothing more.

Permitting the denial is justified, it is said, because we must have regard also for North Carolina’s laws, policies and judgments. And so we must. But thus to state the question is to beg the controlling issue. By every test remaining effective, and not disputed, Nevada had power to alter the petitioner’s marital status. She made the alteration. If it is valid, neither North Carolina nor we are free to qualify it by saying it shall not be effective there, while it is effective in Nevada, and stands without impeachment for ineffectiveness there.

Just that denial is what the terms of the Constitution and the Act of Congress implementing them forbid. It is exactly for the situation where state policies differ that the clause and the legislation were intended. Without such differences, the need for constitutional limitation was hardly one of magnitude. The apparent exceptions for fraud and want of jurisdiction were never intended to enable the states to disregard the provision and each other’s policies, crystallized in judgment, when every requisite for jurisdiction has been satisfied and no showing of fraud has been presented. They have a different purpose, one consistent with the constitutional mandate, not destructive of its effect. That purpose is to make sure that the state’s policy has been applied in the judgment, not to permit discrediting it or the judgment when the one validly crystallizes the other. Such an exception, grafted upon the clause, but nullifies it. It does so totally when the weight and quality of the difference in policies has no bearing on the issue.

Lately this fact has been recognized increasingly in relation to other matters than divorce.[13] The very function of the clause is to compel the states to give effect to the contrary policies of other states when these have been validly embodied in judgment. To this extent the Constitution has foreclosed the freedom of the states to apply their own local policies. The foreclosure was not intended only for slight differences or for unimportant matters. It was also for the most important ones. The Constitution was not dealing with puny matters or inconsequential limitations. If the impairment of the power of the states is large, it is one the Constitution itself has made. Neither the states nor we are free to disregard it. The “local public policy” exception is not an exception, properly speaking. It is a nullifying compromise of the provision’s terms and purpose.

The effort at such compromise, in matters of divorce and remarriage, has not been successful. Together with the instrument by which the various attempts have been made, i.e., the notion of “unitary domicil” constitutionalized as “jurisdictional fact,” this effort has been the source of the long confusion in the circle of decision here. To it may be attributed the reification of the marital status, now discarded in name if not in substance, and the splitting of the res to make two people husband and wife in one state, divorced in another. Haddock v. Haddock, supra; cf. Williams II. Now it leads to practical abandonment of the effort, of this Court’s function, and of the obligation placed upon the states, by committing to their juries for all practical effects the final choice to disregard it.


I do not concur in the abdication. I think a major operation is required to prevent it. The Constitution does not mention domicil. Nowhere does it posit the powers of the states or the nation upon that amorphous, highly variable common-law conception. Judges have imported it. The importation, it should be clear by now, has failed in creating a workable constitutional criterion for this delicate region. In its origin the idea of domicil was stranger to the federal system and the problem of allocating power within it. The principal result of transplanting it to constitutional soil has been to make more complex, variable and confusing than need be inherently the allocation of authority in the federal scheme. The corollary consequence for individuals has been more and more to infuse with uncertainty, confusion, and caprice those human relations which most require stability and depend for it upon how the distribution of power is made.

In my opinion these consequences are inevitable as long as “unitary domicil” usurps the role of “jurisdictional fact” and is applied under the “permissible inference” rule to turn questions of power first for creating jurisdiction, then for nullifying the effects of its exercise, to settle and then unsettle the human relations resting upon the power’s exertion. The conception has outlived its jurisdictional usefulness unless caprice, confusion and contradiction are the desirable criteria and consequences of jurisdictional conceptions.

Stripped of its common-law gloss, the basic constitutional issue inherent in the problem is whether the states shall have power to adopt so-called “liberal” divorce policies and grant divorces to persons coming from other states while there transiently or for only short periods not sufficient in themselves, absent other objective criteria, to establish more than casual relations with the community. One could understand and apply, without decades of confusion, a ruling that transient divorces, founded on fly-by-night “residence,” are invalid where rendered as well as elsewhere; in other words, that a decent respect for sister states and their interests requires that each, to validly decree divorce, do so only after the person seeking it has established connections which give evidence substantially and objectively that he has become more than casually affiliated with the community. Until then the newcomer would be treated as retaining his roots, for this purpose, as so often happens for others, at his former place of residence. One equally could understand and apply with fair certainty an opposite policy frankly conceding state power to grant transient or short-term divorces, provided due process requirements for giving notice to the other spouse were complied with.

Either solution would entail some attenuation of state power. But that would be true of any other, which would not altogether leave the matter to the states and thus nullify the constitutional command. Strong considerations could be stated for either choice. The one would give emphasis to the interests of the states in maintaining locally prevailing sentiment concerning familial and social institutions. The other would regard the matter as more important from the standpoint of individual than of institutional relations and significance. But either choice would be preferable to the prevailing attempt at compromise founded upon the “unitary domicil-jurisdictional fact-permissible inference” rule.

That compromise gives effect to neither policy. It vitiates both; and does so in a manner wholly capricious alike for the institutional and the individual aspects of the problem. The element of caprice lies in the substantive domiciliary concept itself and also in the mode of its application.

Domicil, as a substantive concept, steadily reflects neither a policy of permanence nor one of transiency. It rather reflects both inconstantly. The very name gives forth the idea of home with all its ancient associations of permanence. But “home” in the modern world is often a trailer or a tourist camp. Automobiles, nation-wide business and multiple family dwelling units have deprived the institution, though not the idea, of its former general fixation to soil and locality. But, beyond this, “home” in the domiciliary sense can be changed in the twinkling of an eye, the time it takes a man to make up his mind to remain where he is when he is away from home. He need do no more than decide, by a flash of thought, to stay “either permanently or for an indefinite or unlimited length of time.”[14] No other connection of permanence is required. All of his belongings, his business, his family, his established interests and intimate relations may remain where they have always been. Yet if he is but physically present elsewhere, without even bag or baggage, and undergoes the mental flash, in a moment he has created a new domicil though hardly a new home.

Domicil thus combines the essentially contradictory elements of permanence and instantaneous change. No legal conception, save possibly “jurisdiction,” of which it is an elusive substratum, affords such possibilities for uncertain application. The only thing certain about it, beyond its uncertainty, is that one must travel to change his domicil. But he may travel without changing it, even remain for a lifetime in his new place of abode without doing so. Apart from the necessity for travel, hardly evidentiary of stabilized relationship in a transient age, the criterion comes down to a purely subjective mental state, related to remaining for a length of time never yet defined with clarity.

With the crux of power fixed in such a variable, small wonder that the states vacillate in applying it and this Court ceaselessly seeks without finding a solution for its quandary. But not all the vice lies in the substantive conception. Only lawyers know, unless now it is taxpayers[15] and persons divorced, how rambling is the scope of facts from which proof is ever drawn to show and negate the ultimate conclusion of subjective “fact.” They know, as do the courts and other tribunals which wrestle with the problem, how easily facts procreative of conflicting inferences may be marshalled and how conjectural is the outcome. There is no greater legal gamble. Rare is the situation, where much is at stake, in which conflicting circumstances cannot be shown and where accordingly conflicting ultimate inferences cannot be drawn.

The essentially variable nature of the test lies therefore as much in the proof and the mode of making the conclusion as in the substantive conception itself. When what must be proved is a variable, the proof and the conclusion which follows upon it inevitably take on that character. The “unitary domicil-jurisdictional fact-permissible inference” variable not only is an inconstant, vacillating pivot for allocating power. It is inherently a surrender of the power to make the allocation.

That effect is not nullified by vague reservation of supervisory intent. For supervision in any case that matters, that is, wherever the issue is crucial, nullifies the test. I think escape should be forthright and direct. It can be so only if the attempt to compromise what will not yield to compromise is forsworn, with the ancient gloss that serves only to conceal in familiar formula its essentially capricious and therefore nullifying character. This discarded, choice then would be forced between the ideas of transiency with due process safeguards and some minimal establishment of more than casual or transitory relations in the new community, giving the newcomer something of objective substance identifying him with its life.

With this choice made, objective standards of proof could apply, for the thing to be proved would be neither subjective nor so highly variable as inference of state of mind in ambiguous situation always must be. Neither domicil’s sharp subjective exclusions between the old and the new nor its effort to probe the unprovable workings of thought at some past moment, as in relation to the length of time one purposed remaining or whether there was vestigial and contingent intent to return, would be material.

With the subjective substratum removed, the largest source of variable and inconstant decision would disappear. This would be true, whether transiency guarded by due process or some more established but objectively determinable relation with the community were chosen for the standard to turn the existence of power. Either choice would be preferable to the variable which can give only inconstant and capricious effects, nullifying both policies.

If by one choice states of origin were forced to modify their local policies by giving effect to the different policies of other states when crystallized in valid judgments, that would be no more than the Constitution in terms purports to require. And it may be doubted their surrender would be much greater in practical effects than the present capricious and therefore deceptive system brings about.[16] If by some more restrictive choice states now free to give essentially transient divorce were required to modify that policy for locally valid effects, within the limits of any objective standard that conceivably would be acceptable for constitutional purposes, the obligations they owe to the nation and to sister states would seem amply to justify that modest curtailment of their power. It is hard to see what legitimate substantial interest a state may have in providing divorces for persons only transiently there or for newcomers before they have created, by reasonable length of stay or other objective standards, more than fly-by-night connections.

I therefore dissent from the judgment which, in my opinion, has permitted North Carolina at her substantially unfettered will to deny all faith and credit to the Nevada decree, without in any way impeaching or attempting to impeach that judgment’s constitutional validity. But if she is not to be required thus to give the faith and credit due, in my opinion she should not be allowed to deny it by any standard of proof which is less than generally is required to overturn or disregard a judgment upon direct attack. Cf. Schneiderman v. United States, 320 U.S. 118. The solemnity of the judicial act and the very minimum of “respect” due the action of a sister state should compel adherence to this standard, though doing so would not give the full faith and credit which the Constitution commands. To approximate the constitutional policy would be better than to nullify it.

MR. JUSTICE BLACK, dissenting.

Anglo-American law has, until today, steadfastly maintained the principle that before an accused can be convicted of crime, he must be proven guilty beyond a reasonable doubt. These petitioners have been sentenced to prison because they were unable to prove their innocence to the satisfaction of the State of North Carolina. They have been convicted under a statute so uncertain in its application that not even the most learned member of the bar could have advised them in advance as to whether their conduct would violate the law. In reality the petitioners are being deprived of their freedom because the State of Nevada, through its legislature and courts, follows a liberal policy in granting divorces. They had Nevada divorce decrees which authorized them to remarry. Without charge or proof of fraud in obtaining these decrees,[1] and without holding the decrees invalid under Nevada law, this Court affirms a conviction of petitioners, for living together as husband and wife. I cannot reconcile this with the Full Faith and Credit Clause and with Congressional legislation passed pursuant to it.

It is my firm conviction that these convictions cannot be harmonized with vital constitutional safeguards designed to safeguard individual liberty and to unite all the states of this whole country into one nation. The fact that two people will be deprived of their constitutional rights impels me to protest as vigorously as I can against affirmance of these convictions. Even more, the Court’s opinion today will cast a cloud over the lives of countless numbers of the multitude of divorced persons in the United States. The importance of the issues prompts me to set out my views in some detail.

Statistics indicate that approximately five million divorced persons are scattered throughout the forty-eight states.[2] More than 85% of these divorces were granted in uncontested proceedings.[3] Not one of this latter group can now retain any feeling of security in his divorce decree. Ever present will be the danger of criminal prosecution and harassment.

All these decrees were granted by state courts. Erie R. Co. v. Tompkins, 304 U.S. 64, and cases following it, recognized the obvious truth, that rules of law laid down by state courts are binding. These judicial “laws” are represented by decrees, judgments and court opinions. Today’s opinion, however, undermines and makes uncertain the validity of every uncontested divorce decree. It wipes out every semblance of their finality and decisiveness. It achieves what the Court terms the “desirable effect” of providing the “same” quality to every divorce decree, “wherever the question arises” ó it endows them all alike with the “same” instability and precariousness. The result is to classify divorced persons in a distinctive and invidious category. A year ago, a majority of this Court in a workmen’s compensation case declared that the Full Faith and Credit Clause of the Constitution was a “nationally unifying force”;[4] today, as to divorce decrees, that clause, coupled with a new content recently added to the Due Process Clause, has become a nationally disruptive force. Uncontested divorce decrees are thus so degraded that a person who marries in reliance upon them can be sent to jail. With much language the Court has in effect adopted the previously announced hypothesis upon which the North Carolina Supreme Court permitted another person to be sent to prison, namely, that “the full faith and credit clause does not apply to actions for divorce, and that the states alone have the right to determine what effect shall be given to the decrees of other states in this class of cases.” State v. Herron, 175 N.C. 754, 758, 94 S.E. 698; cf. Matter of Holmes, 291 N.Y. 261, 273, 52 N.E.2d 424.

The petitioners were married in Nevada. North Carolina has sentenced them to prison for living together as husband and wife in North Carolina. This Court today affirms those sentences without a determination that the Nevada marriage was invalid under that state’s laws. This holding can be supported, if at all, only on one of two grounds: (1) North Carolina has extra-territorial power to regulate marriages within Nevada’s territorial boundaries, or, (2) North Carolina can punish people who live together in that state as husband and wife even though they have been validly married in Nevada. A holding based on either of these two grounds encroaches upon the general principle recognized by this Court that a marriage validly consummated under one state’s laws is valid in every other state.[5] If the Court is today abandoning that principle, it takes away from the states a large part of their hitherto plenary control over the institution of marriage. A further consequence is to subject people to criminal prosecutions for adultery and bigamy merely because they exercise their constitutional right to pass from a state in which they were validly married into another state which refuses to recognize their marriage. Such a consequence runs counter to the basic guarantees of our federal union. Edwards v. California, 314 U.S. 160. It is true that persons validly married under the laws of one state have been convicted of crime for living together in other states.[6] But those state convictions were not approved by this Court. And never before today has this Court decided a case upon the assumption that men and women validly married under the laws of one state could be sent to jail by another state for conduct which involved nothing more than living together as husband and wife.

The Court’s opinion may have passed over the marriage question on the unspoken premise that the petitioners were without legal capacity to marry. If so, the primary question still would be whether that capacity, and other issues subsidiary to it, are to be determined under Nevada, North Carolina, or federal law. Answers to these questions require a discussion of the divorce decrees awarded to the petitioners in a Nevada court prior to their marriage there.

When the Nevada decrees were granted, the petitioners’ former spouses lived in North Carolina. When petitioners were tried and convicted, one of their former spouses was dead and the other had remarried. Under the legal doctrine prevailing in Nevada and in most of the states, these facts would make both the decrees immune from attack unless, perhaps, by persons other than the North Carolina spouses, whose property rights might be adversely affected by the decrees.[7] So far as appears from the record no person’s property rights were adversely affected by the dissolution decrees. None of the parties to the marriage, although formally notified of the Nevada divorce proceedings, made any protest before or after the decrees were rendered. The state did not sue here to protect any North Carolinian’s property rights or to obtain support for the families which had been deserted. The result of all this is that the right of the state to attack the validity of these decrees in a criminal proceeding is today sustained, although the state’s citizens, on whose behalf it purports to act, could not have done so at the time of the conviction in a civil proceeding. Furthermore, all of the parties to the first two marriages were apparently satisfied that their happiness did not lie in continued marital cohabitation. North Carolina claims no interest in abridging their individual freedom by forcing them to live together against their own desires. The state’s interest at the time these petitioners were convicted thus comes down to its concern in preserving a bare marital status for a spouse who had already married again. If the state’s interest before that time be considered, it was to preserve a bare marital status as to two persons who had sought a divorce and two others who had not objected to it. It is an extraordinary thing for a state to procure a retroactive invalidation of a divorce decree, and then punish one of its citizens for conduct authorized by that decree, when it had never been challenged by either of the people most immediately interested in it. I would not permit such an attenuated state interest to override the Full Faith and Credit Clause of the Constitution and an Act of Congress pursuant to it.[8] Here again, North Carolina’s right to attack this judgment, despite the Full Faith and Credit Clause and the Congressional enactment, is not based on Nevada law; nor could it be. For in Nevada, even the Attorney General could not have obtained a cancellation of the decree on the ground that it was rendered without jurisdiction. State v. Moore, 46 Nev. 65, 207 P. 75. This makes it clear beyond all doubt that North Carolina has not given these decrees the same effect that they would be given in the courts of Nevada.

The Court permits North Carolina to disregard the decrees on the following line of reasoning. No state need give full faith and credit to a “void” decree. A decree rendered by a court without “jurisdiction” is “void.” No state court has “jurisdiction” to grant a divorce unless one of the parties is “domiciled” in the state. The North Carolina court has decided that these petitioners had no “domicile” in Nevada. Therefore, the Nevada court had no “jurisdiction,” the decrees are “void,” and North Carolina need not give them faith or credit. The solution to all these problems depends in turn upon the question common to all of them ó does state law or federal law apply?

The Constitution provides that “Full Faith and Credit shall be given in each State to the public Acts, Records, and judicial Proceedings of every other State. And the Congress may by general Laws prescribe the Manner in which such Acts, Records and Proceedings shall be proved, and the Effect thereof.” (Emphasis added.) Acting pursuant to this constitutional authority, Congress in 1790 declared what law should govern and what “Effect” should be given the judgments of state courts. That statute is still the law. Its command is that they “shall have such faith and credit given to them . . . as they have by law or usage in the Courts of the state from which they are taken.” 28 U.S.C. 687. If, as the Court today implies, divorce decrees should be given less effect than other court judgments, Congress alone has the constitutional power to say so. We should not attempt to solve the “divorce problem” by constitutional interpretation. At least, until Congress has commanded a different “Effect” for divorces granted on a short sojourn within a state, we should stay our hands. A proper respect for the Constitution and the Congress would seem to me to require that we leave this problem where the Constitution did. If we follow that course, North Carolina cannot be permitted to disregard the Nevada decrees without passing upon the “faith and credit” which Nevada itself would give to them under its own “law or usage.” The Court has decided the matter as though it were a purely federal question; Congress and the Constitution declared it to be a state question. The logic of the Court does not persuade me that we should ignore these mandates of the Congress and the Constitution. Nevada’s decrees purported to grant petitioners an absolute divorce with a right to remarry. No “law or usage” of Nevada has been pointed out to us which would indicate that Nevada would, under any circumstances, consider its decrees so “void” as to warrant imprisoning those who have remarried in reliance upon such existing and unannulled decrees.

A judgment may be “void” in the general sense, and yet give rise to rights and obligations. While on the books its existence is a fact, not a theory. And it may be said of decrees, later invalidated, as of statutes held unconstitutional, that “The past cannot always be erased by a new judicial declaration. The effect of the subsequent ruling as to invalidity may have to be considered in various aspects, ó with respect to particular relations, individual and corporate, and particular conduct, private and official . . . an all-inclusive statement of a principle of absolute retroactive invalidity cannot be justified.” Chicot County Drainage District v. Baxter State Bank, 308 U.S. 371, 374. Despite the conclusion that a judgment is “void,” courts have in the interest of substantial justice and fairness declined to attribute a meaning to that word which would make such judgments, for all purposes, worthless scraps of paper.[9] After a judgment has been declared “void” it still remains to decide as to the consequences attached to good faith conduct between its rendition and its nullification. That determination, I think, must, in this case, under the Full Faith and Credit Clause, be made in accordance with the “law or usage” of Nevada ó not of North Carolina or the federal government.

This brings me to the Court’s holding that Nevada decrees were “void.” That conclusion rests on the premise that the Nevada court was without jurisdiction because the North Carolina Court found that the petitioners had no “domicile” in Nevada. The Nevada court had based its decree on a finding that “domicile” had been established by evidence before it. As I read that evidence, it would have been sufficient to support the findings, had the case been reviewed by us. Thus, this question of fact has now been adjudicated in two state courts with different results. It should be noted now that this Court very recently has said as to the Full Faith and Credit Clause and the 1790 Congressional enactment, that “From the beginning this Court has held that these provisions have made that which has been adjudicated in one state res judicata to the same extent in every other.” Magnolia Petroleum Co. v. Hunt, supra, at 438.[10] That it was appropriate for the Nevada court to pass upon the question of domicile can hardly be doubted, since the concurring opinion in our first consideration of this case correctly said that the “Nevada decrees do satisfy the requirements of the Due Process Clause and are binding in Nevada on the absent spouses . . .” 317 U.S. 287, 306. The Court today, however, seems to place its holding that the Nevada decrees are void on the basis that the Due Process Clause makes domicile an indispensable prerequisite to a state court’s “jurisdiction” to grant divorce. It further holds that this newly created federal restriction of state courts projects fact issues which the state courts cannot finally determine for themselves. Davis v. Davis, 305 U.S. 32, provides a possible exception to this holding. It decided that where both spouses appeared, a state court could finally determine the question of domicile. Whether the Court today overrules that case I cannot be sure. Certainly, if a state court cannot finally determine the question of domicile because it is a federal question, each divorce controversy involving domicile must be subject to review here whether both parties appear or not.

I cannot agree to this latest expansion of federal power and the consequent diminution of state power over marriage and marriage dissolution which the Court derives from adding a new content to the Due Process Clause. The elasticity of that clause necessary to justify this holding is found, I suppose, in the notion that it was intended to give this Court unlimited authority to supervise all assertions of state and federal power to see that they comport with our ideas of what are “civilized standards of law.” See Malinski v. New York, 324 U.S. 401. I have not agreed that the Due Process Clause gives us any such unlimited power, but unless it does, I am unable to understand from what source our authority to strip Nevada of its power over marriage and divorce can be thought to derive. Certainly, there is no language in the Constitution which even remotely suggests that the federal government can fix the limits of a state court’s jurisdiction over divorces. In doing so, the Court today exalts “domicile,” dependent upon a mental state, to a position of constitutional dignity. State jurisdiction in divorce cases now depends upon a state of mind as to future intent. Thus “a hair perhaps divides” the constitutional jurisdiction or lack of jurisdiction of state courts to grant divorces. Cf. Pollock v. Williams, 322 U.S. 4, 21. And this “hair-line” division involves a federal question, apparently open to repeated adjudications at the instance of as many different parties as can be found to raise it. Moreover, since it is a federal question, each new litigant has a statutory right to ask us to pass on it.

The two cases cited by the Court do not support this novel constitutional doctrine. Bell v. Bell, 181 U.S. 175, held a Pennsylvania decree invalid on the ground that there was no domicile shown. It specifically stated, however, that Pennsylvania law required one year’s domicile. Neither the decision in that case, nor any of the others on which it relied, rested on an interpretation of the Due Process Clause as requiring “domicile.”[11] Nor did the decision in Andrews v. Andrews, 188 U.S. 14, support today’s Due Process Clause extension, for there it was said that “. . . it is certain that the Constitution of the United States confers no power whatever upon the government of the United States to regulate marriage . ..”[12]

It is a drastic departure from former constitutional doctrine to hold that the Federal Constitution measures the power of state courts to pass upon petitions for divorce. The jurisdiction of state courts over persons and things within their boundaries has been uniformly acknowledged through the years, without regard to the length of their sojourn or their intention to remain. And that jurisdiction has not been thought to be limited by the Federal Constitution. Legislative dissolution of marriage was common in the colonies and the states up to the middle of the Nineteenth Century. A legislative dissolution of marriage, granted without notice or hearing of any kind, was sustained by this Court long after the Fourteenth Amendment was adopted. Maynard v. Hill, 125 U.S. 190; cf. Pennoyer v. Neff, 95 U.S. 714, 734-735. The provision that made “due process of law” a prerequisite to deprivation of “life, liberty, or property” was not considered applicable to proceedings to sever the marital status. It was only when legislatures attempted to create or destroy financial obligations incident to marriage that courts began to conclude that their Acts encroached upon the right to a judicial trial in accordance with due process.[13] The Court’s holding now appears to overrule Maynard v. Hill, sub silentio. This perhaps is in keeping with the idea that the Due Process Clause is a blank sheet of paper provided for courts to make changes in the Constitution and the Bill of Rights in accordance with their ideas of civilization’s demands. I should leave the power over divorces in the states. And in the absence of further federal legislation under the Full Faith and Credit Clause. I should leave the effect of divorce decrees to be determined as Congress commanded ó according to the laws and usages of the state where the decrees are entered.[14]

Implicit in the majority of the opinions rendered by this and other courts, which, whether designedly or not, have set up obstacles to the procurement of divorces, is the assumption that divorces are an unmitigated evil, and that the law can and should force unwilling persons to live with each other. Others approach the problem as one which can best be met by moral, ethical and religious teachings. Which viewpoint is correct is not our concern. I am confident, however, that today’s decision will no more aid in the solution of the problem than the Dred Scott decision aided in settling controversies over slavery. This decision, I think, takes the wrong road. Federal courts should have less, not more, to do with divorces. Only when one state refuses to give that faith and credit to a divorce decree which Congress and the Constitution command, should we enter this field.

The Court has not only permitted North Carolina to invalidate a Nevada decree contrary to the law and usage of that state. It has actually placed the burden of establishing the validity of that decree on a defendant charged with crime. The only contested question was the validity of the decree, since the petitioners openly lived together as man and wife. And the only issue involved concerning that validity was domicile. The burden of proving that single issue upon which petitioners’ liberty depended was cast upon them. Cf. State v. Herron, 175 N.C. 754, 759, 94 S.E. 698. The jury was not charged that the state must prove the defendants guilty; they were required to prove their innocence. The result is that a state court divorce decree is no protection from being sent to prison in another state unless a defendant charged with acting as it authorized can prove the state court rendering the decree made no error in resolving facts as to domicile. State court judgments exalted by the Constitution and by Congress are thus degraded to a lowly status by today’s decision. State courts, no less than federal courts, were recognized by the founding fathers as instruments of justice. I would continue to recognize them as such. At the very minimum we should not permit holders of these decrees to be convicted of crime unless another state sustained the burden of invalidating them. In a case involving nothing but property, this Court has declined to permit a second marriage to be impugned through an alleged prior marriage “save upon proof so clear, strong and unequivocal as to produce a moral conviction of the existence of that impediment.” Sy Joc Lieng v. Sy Quia, 228 U.S. 335, 339. And we declined to permit a naturalization decree to be set aside because of an absence of “clear, unequivocal and convincing evidence.” Schneiderman v. United States, 320 U.S. 118, 125, 159, 164, 166-170. It is no justification for requiring a less burdensome requirement here to say that in these former cases we were dealing with federal questions. That is exactly what is done here. For the basic question in this case revolves around the Full Faith and Credit constitutional provision and the 1790 Congressional Act. The standard of proof sustained is a federal, not a state, standard. To require a defendant in a criminal case to carry the burden of proof in sustaining his decree to prove his innocence deprives him of all but the last shred of protection that the Full Faith and Credit Clause and the 1790 Act of Congress sought to give him. Cf. Tot v. United States, 319 U.S. 463, 473. It makes of human liberty a very cheap thing ó too cheap to be consistent with the principles of a free government.

Moreover, the Court’s unjustifiable devitalization of the Full Faith and Credit Clause and the Act passed pursuant to it creates a situation which makes the North Carolina statute an inescapable trap for any person who places the slightest reliance on another state’s divorce decree ó a situation which a proper interpretation of the federal question would avoid. The North Carolina statute excludes from its coverage those who “have been lawfully divorced.” Who after today’s decision can know or guess what “right” he can safely exercise under a divorce decree in the intervening period between the day of its entry and the day of its invalidation by a jury?[15] This Court has said that “a statute which either forbids or requires the doing of an act in terms so vague that men of common intelligence must necessarily guess at its meaning and differ as to its application, violates the first essential of due process of law” (italics added).[16] The North Carolina statute, as applied to condemn these two petitioners to serve prison sentences, falls precisely within this description. It does so, because the sole essential contested issue in this case was the validity of the divorce decrees. Involved in this issue are questions of law mixed with questions of fact which perplex lawyers and judges little less than they baffle “men of common intelligence.” Today’s decision adds new intricacies to the whole problem for lawyers to argue about. It provides a new constitutional concept of “jurisdiction,” which itself rests on a newly announced federal “concept of domicile.” No final determination as to its own “jurisdiction” can hereafter be made by a state court in an uncontested divorce case. And so far as I can tell, no other court can ever finally determine this question. It might do so as between any two litigants, but I suppose the question of domicile would still be left open for others to challenge. A man might be tried for bigamy in two or more states. He might be convicted in one or both or all, I suppose. The affirmance of these convictions shows that a divorced person’s liberty, so far as this North Carolina statute is concerned, hinges on his ability to “guess” at what may ultimately be the legal and factual conclusion resulting from a consideration of two of the most uncertain word symbols in all the judicial lexicon, “jurisdiction” and “domicile.” While the doctrine that “Ignorance of the law excuses no man” has sometimes been applied with harsh consequences, American courts have not been in the habit of making ignorance of law the crucial and controlling element in a penitentiary offense. Men have from time to time been sent to prison for violating court commands which were later held invalid.[17] It is quite a different thing, however, to send people to prison for lacking the clairvoyant gift of prophesying when one judge or jury will upset the findings of fact made by another.

In earlier times, some Rulers placed their criminal laws where the common man could not see them, in order that he might be entrapped into their violation. Others imposed standards of conduct impossible of achievement to the end that those obnoxious to the ruling powers might be convicted under the forms of law. No one of them ever provided a more certain entrapment, than a statute which prescribes a penitentiary punishment for nothing more than a layman’s failure to prophesy what a judge or jury will do. This Court’s decision of a federal question today does just that.

MR. JUSTICE DOUGLAS joins in this dissent.

[1] The prosecution was under ß 14-183 of the General Statutes of North Carolina (1943): “If any person, being married, shall contract a marriage with any other person outside of this state, which marriage would be punishable as bigamous if contracted within this state, and shall thereafter cohabit with such person in this state, he shall be guilty of a felony and shall be punished as in cases of bigamy. Nothing contained in this section shall extend . . . to any person who at the time of such second marriage shall have been lawfully divorced from the bond of the first marriage . . .”

[2] “Full Faith and Credit shall be given in each State to the public Acts, Records, and judicial Proceedings of every other State.”

[3] It is interesting to note that this more critical analysis by Mr. Justice Story of the nature of the Full Faith and Credit Clause first appeared in 1833, twenty years after his loose characterization in Mills v. Duryee, supra. 3 Story, Commentaries on the Constitution (1st ed., 1833) p. 183.

[4] “There is scarcely any doctrine of the law which, so far as respects formal and exact statement, is in a more unreduced and uncertain condition than that which relates to the question what force and effect should be given by the courts of one nation to the judgments rendered by the courts of another nation.” James C. Carter and Elihu Root, Appellants’ brief, p. 49, in Hilton v. Guyot, 159 U.S. 113. See, as to “comity,” Loucks v. Standard Oil Co., 224 N.Y. 99, 120 N.E. 198.

[5] The reach of Congressional power given by Art. IV, ß 1 is not before us. See Jackson, Full Faith and Credit ó the Lawyer’s Clause of the Constitution (1945) 45 Col. L. Rev. 1, 21-24; Cook, Logical and Legal Bases of Conflict of Laws (1942) 98 et seq.

[6] We have not here a situation where a State disregards the adjudication of another State on the issue of domicil squarely litigated in a truly adversary proceeding.

[7] Since an appeal to the Full Faith and Credit Clause raises questions arising under the Constitution of the United States, the proper criteria for ascertaining domicil, should these be in dispute, become matters for federal determination. See Hinderlider v. La Plata Co., 304 U.S. 92, 110.

[8] As to petitioner Hendrix these included the pleadings, evidence and decree. As to petitioner Williams essentially the same evidence with respect to his domicil is in the record from witnesses in this case. It shows when Williams left North Carolina, when he arrived in Nevada, the prompt filing of his divorce suit (Nevada requires six weeks’ residence prior to filing a suit for divorce), marriage to petitioner Hendrix immediately after petitioners were divorced, and his prompt return to North Carolina. All of this bears on abandonment of the North Carolina domicil and the intent to remain indefinitely elsewhere.

[1] Cf. text infra Part I.

[2] Presumably it would be our function “to retry the facts” no more if the Nevada decree were immediately under challenge here than it is to do so when the North Carolina judgment is in issue. It would seem therefore that we owe the same deference to Nevada’s finding of domicil as we do to North Carolina’s. Cf. text at note 4 et seq.

[3] The record indicates that Mr. Hendrix “had brought no divorce proceeding against the feme defendant prior to the first trial of this cause, . . . but that he has since and remarried.” Although the evidence shows institution of this proceeding, it does not show a decree was entered prior to his remarriage. Whether or not he actually relied upon the Nevada decree, thousands of spouses so divorced do so rely, thus founding new relations which are equally subject to invalidation by jury finding and are always beclouded by a judgment like that rendered in this case.

[4] Cf. text at notes 2, 5, 7, 9, 11 et seq.

[5] The Nevada court knew that petitioners recently had come from North Carolina, resided in tourist quarters, an auto court, and by inference at least that they had come together. There was in the facts sufficient basis for conclusion that they had no “bona fide” intention of remaining permanently or indefinitely, after the decrees were rendered, if the court had wished to draw that conclusion. Credibility in such circumstances is always for the trier of fact. Worcester County Trust Co. v. Riley, 302 U.S. 292, 299; Burbank v. Ernst, 232 U.S. 162, 164.

[6] Haddock v. Haddock, 201 U.S. 562, 572.

[7] The case was not tried on any theory that Nevada’s court was defrauded or her law evaded. No effort was made to bring it within that well recognized exception to the binding effect of judgments generally. United States v. Throckmorton, 98 U.S. 61; Toledo Scale Co. v. Computing Scale Co., 261 U.S. 399. Nor is that ground asserted here to support the denial of credit. It was not suggested, and is not now, that Nevada either would, or could be required to, set aside her judgment or reach a different result, upon the evidence this record presents; or that she now is bound to give full faith and credit to North Carolina’s decision. Nor has it been contended that the Nevada evidence was not adequate to support her finding.

[8] Petitioners’ motion for judgment by nonsuit, which the court denied, was grounded in part upon the absence of evidence of fraud upon the Nevada court or law and alleged incompetence of such evidence if tendered. They also objected to the portions of the charge which submitted the issue of “bona fide domicil” without reference to the effect of the evidence as tending to vitiate the Nevada decree. “Bona fides” is inherently an element in domiciliary intent. Merely adding the phrase as qualifying adjective does not raise an issue of fraud. For this reason, founded in the state of the record, the Court eschews grounding the decision upon fraud or collusion.

[9] Cf. Cochrane v. Deener, 95 U.S. 355; United States v. Maxwell Land-Grant Co., 121 U.S. 325, 381; United States v. San Jacinto Tin Co., 125 U.S. 273, 300; Lalone v. United States, 164 U.S. 255; United States v. American Bell Tel. Co., 167 U.S. 224. See 9 Wigmore, Evidence (3rd ed.) ß 2498.

[10] Cf. note 5.

[11] Cf. Commissioners of Marion County v. Clark, 94 U.S. 278, 284; Jones v. East Tennessee, V. & G.R. Co., 128 U.S. 443, 445; Tiller v. Atlantic Coast Line R. Co., 318 U.S. 54, 68; Bailey v. Central Vermont R. Co., 319 U.S. 350, 353, 354; Tennant v. Peoria & P.U.R. Co., 321 U.S. 29, 35; 9 Wigmore, ß 2494.

[12] Cf. note 16.

[13] Cf. Milwaukee County v. M.E. White Co., 296 U.S. 268; Titus v. Wallick, 306 U.S. 282; Texas v. Florida, 306 U.S. 398, 410.

[14] Citation of authority is hardly needed for reference to the difficulties courts have encountered in the effort to define this intent. “Animus manendi” is often a Latin refuge which succeeds only in evading, not in resolving, the question with which Job wrestled in his suffering.

[15] Cf. Tilt v. Kelsey, 207 U.S. 43; Iowa v. Slimmer, 248 U.S. 115; Worcester County Trust Co. v. Riley, 302 U.S. 292; Texas v. Florida, 306 U.S. 398; Sweeney v. District of Columbia, 113 F.2d 25, cert. denied, 310 U.S. 631. Compare District of Columbia v. Murphy, 314 U.S. 441, with District of Columbia v. Pace, 320 U.S. 698. See 121 A.L.R. 1200; Tweed and Sargent, Death and Taxes Are Certain ó But What of Domicile? (1939) 53 Harv. L. Rev. 68.

[16] The residence requirements of the states for absolute divorce vary depending at times on the ground for divorce relied on, the place where the cause of action arose, or other factors. Speaking generally, approximately 33 states require one year’s residence in most divorce actions. Nine states are more severe, 7 of these requiring 2 years’ residence and two a longer period. Six states are less severe. Of these North Carolina at present requires a 6 months’ residence and the others six weeks to three months. See Warren, Schouler Divorce Manual (1944) 705-720. Thus, practically speaking, 39 states require one year or less, only 9 longer.

It seems questionable, at any rate, that the grounds for divorce as such have “jurisdictional” significance. Presumably, if length of residence is the controlling factor, all of the states would be required to give effect to divorces granted by the 42 requiring one year or longer, unless the greatly preponderant legislative judgment is to be disregarded. The permissible denial accordingly would extend at the most to decrees granted by the six states requiring less than one year. It is difficult to see how greatly disruptive effects would be created for them or for the other states by requiring them to approximate the generally prevailing judgment as to the length of the period appropriate for granting impeccable divorce.

[1] Previous decisions of this Court have asserted that a state cannot justify its refusal to give another state’s judgment full faith and credit, at least, in the absence of a showing that fraud is an adequate ground for setting the judgment aside in the state where it was rendered. See Christmas v. Russell, 5 Wall. 290, 302-304; Maxwell v. Stewart, 22 Wall. 77, 81; Bigelow v. Old Dominion Copper Co., 225 U.S. 111, 134.

[2] According to the best available statistics more than five million divorces were granted in the last twenty years and the annual rate is steadily increasing. See Marriage and Divorce Statistics, Bureau of the Census, 1942, and the same reports for different years; Divorce, Depression and War, Social Forces, University of North Carolina Press, Dec. 1943, 191, 192; Social and Statistical Analysis, Law and Contemporary Problems, Duke University, Summer 1944; 1940 Census, Bureau of the Census, Vol. 4, Tables 29 and 48; Ogburn, Marriages, Births and Divorces, Annals, American Academy, Sept. 1943, 20.

[3] This percentage is shown by the various “Marriage and Divorce” publications of the Bureau of the Census, Department of Commerce. Careful studies in particular localities have indicated that the percentage of uncontested divorces is substantially above the 85% shown in Census Reports. In Maryland, for instance, 3,306 petitions for divorce were filed in 1929. 1,847 defendants failed to answer and the complainant had decrees in all but six cases. “A total of 1,459 defendants, however, filed answers to the plaintiff’s allegations and thus staged a technical contest. This does not necessarily mean that a given defendant was opposed to a decree being granted. Of these 1,459 technically contested actions, 442 dropped out without coming to hearing, thus leaving 1,017 technical contests in the field. . . . If we accompany the plaintiffs in the 1,017 remaining technical contests to the hearing, we find little in the way of substantial contest. There is a positive record of no contest in 808 cases; of a contest in 81 cases; and data are not available with respect to contest in 128 cases. . . . It seems likely that in less than 100 cases was there at the hearing a contest concerning whether a decree should be granted.” Marshall and May, The Divorce Court, 226-227.

[4] Magnolia Petroleum Co. v. Hunt, 320 U.S. 430, 439.

[5] Loughran v. Loughran, 292 U.S. 216, 223-225; Dudley v. Dudley, 151 Iowa 142, 130 N.W. 785; Ex parte Crane, 170 Mich. 651, 136 N.W. 587; see Radin, Authenticated Full Faith and Credit Clause, 39 Ill. L.R. 1, 32. See also Annotations, 60 Am. St. Rep. 942; 28 L.R.A.N.S. 754; 127 A.L.R. 437.

[6] This question has arisen most frequently in the application of state laws making it a criminal offense for persons of different races to live together as husband and wife. See e.g., State v. Bell, 7 Baxt. (Tenn.) 9. That case has been explained as a holding that “Without denying the validity of a marriage in another state, the privileges flowing from marriage may be subject to the local law.” Yarborough v. Yarborough, 290 U.S. 202, 218. See also Greenhow v. James, 80 Va. 636. Cf. Pearson v. Pearson, 51 Cal. 120; State v. Ross, 76 N.C. 242; Whittington v. McCaskill, 65 Fla. 162, 61 So. 236.

[7] See e.g., Foy v. Smith’s Estate, 58 Nev. 371, 81 P.2d 1065; Dwyer v. Nolan, 40 Wash. 459, 82 P. 746, 1 L.R.A.N.S. 551; Chapman v. Chapman, 224 Mass. 427, 113 N.E. 359; Matter of Bingham, 265 App. Div. 463, 39 N.Y.S.2d 756; Moyer v. Koontz, 103 Wis. 22, 79 N.W. 50; Leathers v. Stewart, 108 Me. 96, 79 A. 16, Ann. Cas. 1913B, 366, 369-372; Kirschner v. Dietrich, 110 Cal. 502, 42 P. 1064; Schouler Divorce Manual, 588-590.

[8] Here too we approach the domain where the line may be shadowy between the individual rights of people to choose and keep their own associates and the power of the state to prescribe who shall be their most intimate associates. People in this country do not “belong” to the state. Le Mesurier v. Le Mesurier, [1895] A.C. 517. Our Constitution preserves an area of individual freedom which the state has no right to abridge. The flavor of the Court’s opinion is that a state has supreme power to control its domiciliaries’ conduct wherever they go and that the state may prohibit them from getting a divorce in another state. In this aspect the decision is not confined to a holding which relates to state as opposed to federal rights. It contains a restriction of individual as opposed to state rights. See Radin, supra, 28-32.

[9] Gray v. Brignardello, 1 Wall. 627, 634; Colvin v. Colvin, 2 Paige (N.Y.) 385; Harper, The Myth of the Void Divorce, 2 Law and Contemporary Problems, 335; The Validity of Void Divorces, 79 U. Pa. L. Rev. 158; Tainter, Restitution of Property Transferred Under Void or Later Reversed Judgments, 9 Miss. L.J. 157.

[10] The Nevada court had general jurisdiction to grant divorces, and the complaint was required to allege domicile along with the other requisite allegations. Domicile is as much an integral element in the litigation as the proof of cruelty or any of the other statutory grounds for divorce in Nevada. Labeling domicile as “jurisdictional” does not make it different from what it was before. Since the Nevada court had no power to render a divorce without proof of facts other than domicile, there is nothing to prevent this Court, under its expansive interpretation of the Due Process Clause, from labeling these other facts as “jurisdictional” and taking more state powers into the federal judicial orbit. Both these types of facts, however labeled, were part of the controversy which the Nevada legislature gave its courts power to resolve. The state could label them “jurisdictional” and, having the exclusive power to grant divorces, could attach such consequences to them as it sees fit. But, while Congress might, under the Full Faith and Credit Clause, prescribe the “effect” in other states, of decrees based on the finding, I do not think the federal courts can, by their mere label, attach jurisdictional consequences to the state’s requirement of domicile. Hence, I think the quoted statement from the Magnolia Petroleum case should control this case.

[11] Streitwolf v. Streitwolf, 181 U.S. 179, decided the same day as Bell v. Bell, held a North Dakota divorce decree invalid. That holding did not rest on any “federal concept of domicile,” but on the fact that North Dakota law required “a domicile in good faith . . . for ninety days as a prerequisite to jurisdiction of a case of divorce.”

[12] Andrews v. Andrews did not assert that any particular federal constitutional provision made domicile a state jurisdictional requirement in divorce cases. It emphasized state and common law concepts of domicile and a state’s power over its “inhabitants.” This emphasis led the Court to permit Massachusetts to invalidate a South Dakota divorce decree, even though both husband and wife had appeared in the South Dakota Court. Cf. Davis v. Davis, supra. Massachusetts had a statute which prohibited its “inhabitants” from going into another state to get a divorce on account of conduct which occurred in Massachusetts, or for conduct which would not have authorized a divorce under Massachusetts law. This statute obviously rested on a hypothesis that each state possesses these sweeping powers over individuals: (1) power to make it a crime for its inhabitants to go to another state to engage in conduct which might be lawful there; (2) power to punish an inhabitant who went into another state and engaged in conduct in harmony with that state’s laws. If North Carolina has attempted to impose such sweeping statutory prohibition upon its inhabitants, it has not been called to our attention. In its absence the Andrews decision gives no support to the opinion and judgment in this case.

[13] Wright v. Wright’s Lessee, 2 Md. 429, 453; Crane v. Meginnis, 1 G. & J. Rep. (Md.) 463; Dwyer v. Nolan, supra. See also Owens v. Claytor, 56 Md. 129; 2 Schouler, Marriage, Divorce and Separation, Sixth Edition, Pars. 1471-1473; Validity of Legislative Divorce, 18 L.R.A. 95.

[14] For an interesting discussion of the consequences of shifting divorces from the legislatures to the courts, to be worked out in the pattern of adversary controversies, see Marshall and May, supra, Chap. VI, The Mirage of Judicial Controversy. For bibliography of pertinent discussions see same, 338-341.

[15] The answer is that, by reason of today’s decision, no person can exercise any right whatever under an uncontested divorce decree without subjecting himself to possible penitentiary punishment. “To make the enjoyment of a right dependent upon an impossible condition is equivalent to an absolute denial of the right under any condition, and such denial, enforced for a past act, is nothing less than punishment imposed for that act.” Cummings v. Missouri, 4 Wall. 277, 327. The “condition” here, that a divorced person cannot remarry without the possibility of being subjected to repeated prosecutions in all the states where he lives as a married person, would seem to rank as “an impossible condition.” If, therefore, the Court’s object is to make divorces dangerous, its object has been accomplished. I think divorce policy is the business of the people and their legislatures ó not that of this Court.

[16] Connally v. General Construction Co., 269 U.S. 385, 391. See also Yu Cong Eng v. Trinidad, 271 U.S. 500, 518; United States v. Cohen Grocery Co., 255 U.S. 81, 89-93; Lanzetta v. New Jersey, 306 U.S. 451; Smith v. Cahoon, 283 U.S. 553; Screws v. United States, ante, p. 91; cf. Nash v. United States, 229 U.S. 373, 377 with Cline v. Frink Dairy Co., 274 U.S. 445, 457, 463-464.

[17] See e.g., People v. Morley, 72 Colo. 421, 211 P. 643; Holbrook v. Prichard Motor Co., 27 Ga. App. 480, 109 S.E. 164; St. George’s Society v. Sawyer, 204 Iowa 103, 214 N.W. 877; State v. La Follette, 100 Ore. 1, 196 P. 412.