MR. JUSTICE FRANKFURTER delivered the opinion of the Court.
In 1952, petitioner, while a citizen and resident of Illinois, purchased from respondent in Illinois the contract of insurance upon which this suit is based. The respondent is a British company licensed to do business in Illinois, Florida, and nine other States.
The policy, which petitioner bought for a lump sum, ran for three years. Designated a "Personal Property Floater Policy (World Wide)," it provides world-wide coverage against "all risks" of loss or damage to the property covered, property generally classified as personal property having no fixed situs. A provision of the policy, which has given rise to this controversy, required that suit on any claim for loss must be brought within twelve months of the discovery of the loss.
Some months after purchasing the policy the petitioner moved to Florida, where he brought this suit for losses sustained in Florida in the winter of 1954-1955. Petitioner reported the losses to the respondent on February 1, 1955, and on April 1, 1955, respondent denied liability.
The action, resting on diversity of citizenship, was instituted in the United States District Court for the Southern District of Florida on May 20, 1957, more than two years after discovery of the losses. The respondent defended on two grounds: (1) that under the time limitation for bringing suit, a restriction concededly valid under Illinois law, the suit was barred; and (2) that the "all risks" coverage of the policy does not include the losses resulting from willful injury to or appropriation of the insured property by the insured's spouse.1 The jury was charged that if the losses were caused by the deliberate acts of petitioner's wife, they were not therefore excluded from coverage. The jury found for petitioner, and judgment in the amount of $6,800 was entered. The District Court, without opinion, denied a motion for judgment non obstante veredicto, which was based, inter alia, upon the suit clause, apparently believing that Florida Statutes (1957) § 95.03, which is set out in the margin,2 rendered the clause ineffective.
On appeal the Court of Appeals for the Fifth Circuit reversed (one judge dissenting), sustaining the defense based upon the suit clause on the ground that Florida could not apply its statute to this Illinois-made contract consistently with the requirements of due process. 265 F.2d 522. The court considered the preliminary question of state law—whether the Florida statute, § 95.03, in fact applies to a contract made in these circumstances. Strangely enough, it did not decide this threshold question because it apparently found it easier to decide the constitutional question that would be presented only if the statute did apply. Such disposition of a serious constitutional issue justified bringing the case here. 361 U.S. 874.
By the settled canons of constitutional adjudication the constitutional issue should have been reached only if, after decision of two non-constitutional questions, decision was compelled. The lower court should have first considered: (1) whether, under the law of Florida, § 95.03 is applicable to this contract; and (2) whether the losses sued upon were within the "all risks" coverage of the policy if in fact caused by petitioner's wife.
It would be a temerarious man who described the constitutional question decided below as frivolous. The seriousness of the question becomes manifest from a recital of the decisions of this Court relevant to the determination of the issue on which the court below passed.
In Home Insurance Co. v. Dick, 281 U.S. 397, the Court held that Texas could not constitutionally apply its own law to invalidate a suit clause in a contract of fire insurance covering a tugboat. The plaintiff was at all pertinent times both a Texas domiciliary and a resident of Mexico. The contract, of which he was an assignee, was made in Mexico between a Mexican insurer which had no contact whatever with Texas, and a Mexican resident. The premium was paid in Mexico, and the policy covered the tug only while it was in Mexican waters. In Hartford Accident & Indemnity Co. v. Delta & Pine Land Co., 292 U.S. 143, the Court held that Mississippi could not constitutionally apply its own law to invalidate a contract clause limiting the insurer's liability on a surety bond against defalcations by the insured's employees "in any position, anywhere," to losses of which notice was given within fifteen months after the termination of coverage. The contract was made in Tennessee where the insured had offices and the insurer was licensed to do business. Mississippi's action was struck down although the contract covered an ambulatory risk, the default giving rise to the claim actually occurred in Mississippi, the insurer was under license doing business there, and the insured was incorporated there.
The most recent case in the series is Watson v. Employers Liability Assurance Corp., Ltd., 348 U.S. 66. Without questioning either Dick, or Delta & Pine, the Court sustained Louisiana's application, in a suit by a Louisiana citizen, of its own "direct action" statute although thereby it invalidated an express provision against direct liability of the insurer in a contract negotiated and paid for within Illinois and Massachusetts, in both of which the clause was valid. The contract insured Toni, an Illinois corporation distributing its product nationally, against liabilities arising from the use of the product. The insurer was a British corporation licensed to do business in several States, including Massachusetts, Illinois and Louisiana. Toni had no contact with Louisiana and could not be served there. The Louisiana plaintiff had sustained her injury in Louisiana. The Court found Louisiana's contact with the subject justified its application of the statute to make an insurer doing business in Louisiana amenable to suit by a locally injured citizen.
The relevant factors of the present case are not identic either with Dick, or Delta & Pine, or Watson, and not one of them can fairly be deemed controlling here. The bearing of all three on the immediate situation would have to be considered and appropriately evaluated in adjudicating the precise constitutional issue presented by it, were that issue inescapably before us. The disposition of either of two unresolved state law questions may settle this litigation. The Court of Appeals was therefore not called upon initially to reach this constitutional question; nor is this Court. The doctrine that the Court will not "anticipate a question of constitutional law in advance of the necessity of deciding it," Liverpool, N. Y. & P. S. S. Co. v. Emigration Commissioners, 113 U.S. 33, 39, relied on by Mr. Justice Brandeis in his well-known concurring opinion in Ashwander v. T. V. A., 297 U.S. 288, 347-348, is a well-settled doctrine of this Court which, because it carries a special weight in maintaining proper harmony in federal-state relations, must not yield to the claim of the relatively minor inconvenience of postponement of decision. Of course we do not remotely hint at an answer to a question that is prematurely put.
While both questions not disposed of by the Court of Appeals are questions of local law, the question whether under Florida law § 95.03 is applicable to this contract is one on which the state court's determination is controlling. But, as the Court of Appeals indicated, it could not, on the available materials, make a confident guess how the Florida Supreme Court would construe the statute. See, e. g., Hoagland v. Railway Express Agency, 75 So.2d 822; Equitable Life Assurance Society v. McRee, 75 Fla. 257. The Florida Legislature, with rare foresight, has dealt with the problem of authoritatively determining unresolved state law involved in federal litigation by a statute which permits a federal court to certify such a doubtful question of state law to the Supreme Court of Florida for its decision. Fla. Stat. Ann., 1957, § 25.031.3 Even without such a facilitating statute we have frequently deemed it appropriate, where a federal constitutional question might be mooted thereby, to secure an authoritative state court's determination of an unresolved question of its local law. See Allegheny County v. Mashuda Co., 360 U.S. 185, 189, and cases cited; see also Meredith v. Winter Haven, 320 U.S. 228, 236.
Vacated and remanded.
MR. JUSTICE BLACK, whom THE CHIEF JUSTICE and MR. JUSTICE DOUGLAS join, dissenting.
The Court today holds that this Court and the federal courts below must refrain from exercising their jurisdiction to decide this lawsuit properly brought. It remands the case to the Court of Appeals and implies that a state court should be the one to determine two questions of state law to avoid a federal constitutional question which is also presented. In so doing, I believe this Court is carrying the doctrine of avoiding constitutional questions to a wholly unjustifiable extreme. I agree that it is frequently better not to decide constitutional questions when decision of nonconstitutional questions also presented will dispose of a case. But I do not agree that this is such an occasion. The state law questions do not call for first interpretation of a broad, many-pronged, state regulatory scheme.1 They do not involve peculiarly local questions such as the eminent domain power a State has allowed a city to exercise,2 or the local land law of a State.3 Nor are the state questions here difficult ones depending on ambiguous or vague state law,4 but instead they border on the frivolous. Since I think the answer to the constitutional question also is clear, I believe we should decide all the questions in the case. The Court's refusal to do so, together with the language it uses, seems to me to be an automatic application of "canons of constitutional adjudication" so absolute that a federal court can never under any circumstances or conditions decide a constitutional question if there is any possibility of turning a case away on other grounds. I believe that there are times when a constitutional question is so important that it should be decided even though judicial ingenuity would find a way to escape it. I would decide this case here and now.
The first state question is whether, under state interpretation, the clause of this insurance policy which insures the petitioner against "all risks," protects him against destruction and loss of the property caused by his wife.5 The policy does not intimate any exception to its coverage for such a risk although it has pages of small printed type stating its extensions, limitations, exclusions and general conditions. The United States District Judge who tried this case, experienced in Florida law, not surprisingly paid scant attention to this contention. No case in which we have ever "abstained" from passing on difficult state questions offers the faintest support for the holding that a contention so unlikely to be sustained anywhere can be used as a reason to avoid passing on a constitutional question, even one much more serious than I see the one here to be.
The second state question that the Court is sending back, with the suggestion that the Court of Appeals should refer it to the Florida Supreme Court for decision, is almost equally devoid of plausibility. A Florida state statute provides that all contractual provisions fixing a period of time in which suits may be brought under such contract at a period of time less than that provided by the statute of limitations of Florida are illegal and void. The statute also forbids any court in Florida to "give effect to any provision or stipulation of the character mentioned in this section."6 Since the contract of insurance here provided for a period of limitation shorter than the State's five-year period for unsealed, written contracts,7 this contractual provision would be void under the Florida statute if it applies.8 The only way to get ambiguity into this section is to import it. Statutes of a similar nature exist in 31 States and the District of Columbia.9 They are in line with the protective safeguards that States have felt it necessary to create so as to preserve a fair opportunity for people who have bought and paid for insurance to go to court and collect it. And state courts in the main have interpreted and applied such statutes so as to carry out the legislative policy adopted.10 Florida's particular interest in this very statute is shown by the fact that the Attorney General of the State filed briefs and participated in oral argument to support both the full meaning petitioner claimed for the statute and its constitutionality when so interpreted. I see no reason to send this particular question back to the Court of Appeals, much less, ultimately, to the state court. The statute's plain language, its interpretation by the experienced trial judge who sat on the case and its interpretation by the Attorney General of the State should be sufficient to show to even the most doubtful that this state law applies to this printed provision of the contract and requires the company to try this lawsuit on its merits (unless, of course, the statute is unconstitutional when so applied). I think no cloud should be cast on the statute's clear meaning and I certainly do not think it is necessary to point out to the Florida court that it also could, if it wished, avoid the constitutional question the Court makes so much of by limiting the meaning the Florida legislature obviously intended to give this statute.11 If "maintaining proper harmony in federal-state relations" is the objective of the Court, I would think it best to give this statute its plain meaning and to settle the constitutionality of this statute Florida passed (according to its Attorney General) to protect its people.
I now come to the constitutional question which is avoided and which I would decide. This insurance contract was made in the State of Illinois. There are Illinois cases indicating that the contractual provision shortening the Illinois state statute of limitations might be treated as valid in a court of that State.12 There are no cases, however, indicating that Illinois wanted to project its law into the State of Florida so as to nullify a Florida law invalidating such contractual provisions in Florida courts.13 The constitutional question raised is this. Since the policy's restrictive provision would probably be upheld in Illinois courts in a suit on an Illinois contract, does either the Due Process Clause or the Full Faith and Credit Clause require Florida to pay it homage?
The Florida statute is, in my judgment, constitutional as applied by the District Court in this case. I believe it violates neither the Due Process Clause nor the Full Faith and Credit Clause of the Constitution. There was a time in the evolution of conflict of laws theories when the idea was championed that every detail and element of a contract, every action taken under it, was governed by the law of the place where the contract was made. This concept ran into many difficulties. Was the contract made at the home office of an insurance company or at the place where an agent dropped it in the mail to send it to a man in another State? Exceptions sprang up such as the rule applying the law of the place where the contract was to be performed to issues of performance. Soon it was discovered that it was almost as puzzling to tell where a contract was intended to be performed or what part of activities under a contract could be considered performance as it had been to determine where a contract was made. These and other such academic problems dissipated the dream of a fixed rule or rules for deciding which law governed contract cases. As the concepts developed, there came an emphasis upon having a contract governed by the law which the parties intended to be applied. But it was not always possible to tell which law the parties had agreed upon, and there was resistance on the part of some jurisdictions having close interests in the events leading to litigation to applying foreign law, against their deeply felt policies, solely because the parties at one time preferred it.
As business boomed throughout our growing country giving more States than one an interest in what a contract meant and how it should be enforced for the benefit of the citizens who made it or for whose benefit it was made, practical men began to see that there could not be one single rule of law to govern a contract in which the citizens of many States were interested. One of the many opinions of this Court recognizing that fact was Pacific Employers Ins. Co. v. Industrial Accident Comm'n, 306 U.S. 493, in which Mr. Justice Stone, later Chief Justice, stated that:
"[T]he conclusion is unavoidable that the full faith and credit clause does not require one state to substitute for its own statute, applicable to persons and events within it, the conflicting statute of another state, even though that statute is of controlling force in the courts of the state of its enactment with respect to the same persons and events." Id., at 502.
Later, in Klaxon Co. v. Stentor Electric Mfg. Co., 313 U.S. 487, and in Griffin v. McCoach, 313 U.S. 498, this Court recognized that the courts of a State are not compelled to enforce all provisions of all contracts, but have much freedom to exercise their own state policy in their own courts. See also Pink v. AAA Highway Express, 314 U.S. 201; Hoopeston Canning Co. v. Cullen, 318 U.S. 313.14
After these and a host of other cases recognizing the constitutional power of States to apply their own laws in many ways to contracts made outside the State, we decided Watson v. Employers Liability Assurance Corp., 348 U.S. 66. That case involved a law of Louisiana which provided that injured persons could bring direct actions against liability insurance companies that had issued policies contracting to pay judgments imposed against persons who had inflicted the injuries. The insurance contract in that case, however, contained a clause, binding and enforceable under the law of the places where the contract was made and delivered, that prohibited direct action against the insurance company until after final determination of the insured's obligation to pay damages. A person injured in Louisiana by an insured company sued the insurance company there directly. Application of the Louisiana law was challenged as an unconstitutional denial of equal protection, due process, full faith and credit, and an unconstitutional impairment of contract. We rejected all these contentions. The policy of insurance there, like the one here, was to be given nation-wide effect. We held there, MR. JUSTICE FRANKFURTER disagreeing with the grounds of the Court's opinion, that none of the provisions of the Constitution relied on requires States automatically to subordinate their own contract laws to the laws of other States in which contracts happened to have been executed. We said:
"Where, as here, a contract affects the people of several states, each may have interests that leave it free to enforce its own contract policies." Id., at 73.
In the Watson case we also rejected a contention that the cases relied on by the Court here as throwing a cloud upon the Florida statute, Hartford Accident & Indemnity Co. v. Delta & Pine Land Co., 292 U.S. 143, and Home Ins. Co. v. Dick, 281 U.S. 397, required that we hold Louisiana's law unconstitutional. The reasons we gave for rejecting the contention about those cases there apply equally to the Florida statute here. In the Dick case the Court's opinion carefully pointed out that the decision in that case might have been different had activities relating to the contract there held binding in Texas been carried on in that State. And in the Delta & Pine Land Co. case, we pointed out that the Court had considered that the Mississippi activities in connection with the policy sued on there were found to be so "slight" and so "casual" that Mississippi could not apply its own law. I, myself, have grave doubts that the Delta & Pine Land Co. case would be treated the same way today on its facts. But however that may be, as it stands, it does not require a holding that Florida's law is unconstitutional. If thought to suggest such a holding, it only means that we should decide this case to remove any such suggestion once and for all. The only philosophy on which the Dick and Delta & Pine Land Co. cases could be made to apply here would be on the old idea that the law of the place where the contract is made always governs every activity under it, a rule that had been repudiated by courts and commentators everywhere, especially as a constitutional rule.15
Our later cases previously discussed express the only workable rule for this country today.16 Insurance companies, like other contractors, do not confine their contractual activities and obligations within state boundaries. They sell to customers who are promised protection in States far away from the place where the contract is made. In this very case the policy was sold to Clay with knowledge that he could take his property anywhere in the world he saw fit without losing the protection of his insurance. In fact, his contract was described on its face as a "Personal Property Floater Policy (World Wide)." The contract did not even attempt to provide that the law of Illinois would govern when suits were filed anywhere else in the country. Shortly after the contract was made, Clay moved to Florida and there he lived for several years. His insured property was there all that time. The company knew this fact. Particularly since the company was licensed to do business in Florida, it must have known it might be sued there, and that Florida courts would feel bound by Florida law.
In addition to the reasons already given for my view that Florida law constitutionally may govern this case— that Florida, the forum State, has sufficient contacts with the parties, the property insured and the lawsuit—I would add that when a contractual provision is one dealing with limitations on actions, it is particularly inappropriate to compel the forum State, as a constitutional matter, to apply the law of the place where the contract was "made." This Court has long recognized that the States where law-suits are tried are free to apply their own statutes of limitations. This has been the constitutional rule since the decision in 1839 of M'Elmoyle v. Cohen, 13 Pet. 312. The continued vitality of this principle was recognized by the Court in Wells v. Simonds Abrasive Co., 345 U.S. 514, 516-517. The only deviation from it appears to have been Order of United Commercial Travelers v. Wolfe, 331 U.S. 586, which applied a special rule freeing fraternal insurance companies because of their "indivisible unity," a distinction to which I registered my dissent. It is true that this case is not identical with one in which the forum seeks to apply an ordinary statute of limitations to a suit on a contract having no limitation clause. Here, Florida, seeking to be sure that its own limitation rules and no others apply to cases in its courts, has legislated that contractual limitations of too short duration are invalid. The Court of Appeals called it error to assume "that the issue presented concerned the choice of the applicable statute of limitations rather than the choice of the substantive law governing the validity of the contract itself." But the same reasons for the view that the forum may refuse to apply a foreign statute of limitations impel me to the view that the forum may refuse to apply a foreign contract of limitations. See Order of United Commercial Travelers v. Wolfe, 331 U.S. 586, 627-630 (dissenting opinion). And cf. Metropolitan Cas. Ins. Co. v. Brownell, 294 U.S. 580.
The Court, however, says that there is a serious constitutional question whether Florida can apply its own law here. Therefore, the Court refuses to decide the question (and the related state questions) on the ground, as I read the opinion, that there exists an unbending, unyielding, automatic canon of constitutional adjudication that if a constitutional question is not "frivolous," the Court must avoid it, unless decision is "compelled" after disposition of all nonconstitutional questions. In fact, the Court indicates that when a constitutional question lurks in the case, not even the lower federal courts sitting in diversity jurisdiction should decide the nonconstitutional questions.17 Of course, this view is not unprecedented altogether; it is in my opinion, however, wholly unprecedented in a case such as this. I agree that there is a judicial practice, wise perhaps in most instances, under which courts do not ordinarily decide constitutional questions unless essential to a decision of the case.18 This practice extends back to the early days of the country. But even the greatest of our judges have not always followed it as a rigid rule. Perhaps had they done so the great opinion of Chief Justice Marshall in Marbury v. Madison would never have been written.19 Only if the practice of occasionally avoiding decision of a constitutional question is first made into a rule and then elevated to a position of absoluteness denied by some even to constitutional commands themselves, are we wise in avoiding decision here. On the other hand, if the power to avoid deciding constitutional issues is discretionary, as I think it undoubtedly is, I believe that this is not a proper case for its exercise.
Such a rigid, ironclad, all-encompassing rule as I understand the Court to promulgate here is, in my judgment, bad for the litigants, bad for the courts, and bad for the country. Litigants have a right to have their law-suits decided without unreasonable and unnecessary delay or expense.20 There come times, in my judgment, when a constitutional question is so ripe for decision, when its resolution is so much needed, that it would be proper to decide the constitutional question even though there might be a possibility or even a probability that by sending a case back some nonconstitutional question might be decided in a way that would remove the constitutional controversy from that particular case. Cf. Peters v. Hobby, 349 U.S. 331, 349 (dissenting opinion). The fact that one case presenting the constitutional issue in some clear form has survived the jurisdictional and practical obstacles to adjudication, the fact that such an issue has been tossed up from the maelstrom of trials and private disputes to the height of our appellate courts, is one sign that the issue needs deciding. However this particular case is or may be decided, the pressing need for deciding this constitutional question will remain the same. Our expanding commerce among the States guarantees that. The constitutional question is squarely presented and the way it is decided will have an important effect on the laws of many States in addition to Florida. It is here now. Why not decide it? Sometimes a conflict of view among the circuits and among the States on a constitutional question, like such conflicts on statutes or common-law questions, reaches such proportions that they cry out for an authoritative decision of our Court. At least in such instances I am not willing to tie myself down by a judicially created rule that would bar deciding constitutional questions when they get here.21 Subscribing as I have to the belief that there is virtue in the policy of not unnecessarily deciding constitutional issues, I think it would be better to abandon that policy entirely than to carry it to the extremes of the Court's opinion today. In my judgment, the rule in the rigid and sweeping form announced has not been the rule heretofore. It is true that some dissents might possibly have gone so far, but I do not think it can fairly be said that the whole Court has done so. That this Court has not heretofore followed the dogmatic rule announced today is very clear from our case of Chicago v. Atchison, T. & S. F. R. Co., 357 U.S. 77. Cf. United States v. Sullivan, 332 U.S. 689, 692-694. In the Chicago case, over a strong dissent, the majority of the Court refused to avoid the constitutional question on the ground that we should first wait to have a city ordinance interpreted by Illinois courts. We said:
"We see no ambiguity in the section which calls for interpretation by the state courts. Cf. Toomer v. Witsell, 334 U.S. 385. Remission to those courts would involve substantial delay and expense, and the chance of a result different from that reached below, on the issue of applicability, would appear to be slight." Id., at 84.
This was a fair application of the constitutional avoidance practice.22
The Court assumes that there is in Florida a method which will enable the Court of Appeals for the Fifth Circuit to obtain a decision of the Supreme Court of Florida by certifying to them the two questions of state law here involved. Florida does have such a law on paper, but evidently does not have one in fact. The state statute, first passed in 1945 and now appearing as Fla. Stat. Ann. (1959 Supp.) § 25.031, authorizes the Supreme Court of Florida to provide rules for obtaining such certifications from federal appellate courts, but the best information obtainable is that the Supreme Court of Florida has never promulgated any such rules, and evidently has never accepted such a certificate.23 This is not difficult to understand. Perhaps state courts take no more pleasure than do federal courts in deciding cases piecemeal on certificates. State courts probably prefer to determine their questions of law with complete records of cases in which they can enter final judgments before them. It seems rather unfortunate for this petitioner that he is to be made the guinea pig in the Court's effort to get the Supreme Court of Florida to put into effect a law that it has deliberately left unused for a period of 15 years.24 This suit was filed three years ago and, borrowing an expression, it would be a "temerarious man" who would forecast that it is sure to get back to us again before three more years. That would be all right if such an exasperating delay were necessary in order to achieve fair and just consideration of this case. I do not think it is necessary or justified in this case, and I think the Court's handling of the case sets up a precedent of such an extreme nature that the rule of avoiding constitutional questions might begin to produce more evil consequences than good.
I would affirm the judgment of the District Court.
MR. JUSTICE DOUGLAS, dissenting.
While I join the dissent of my Brother BLACK, I desire to give renewed protest to our practice of making litigants travel a long, expensive road in order to obtain justice. Congress has created federal courts with power to adjudicate controversies between citizens of different States. They are manned by judges drawn from the local Bars and fairly conversant with the laws of their respective areas. They are equipped to decide questions of local law as well as federal questions. As we stated in Meredith v. Winter Haven, 320 U.S. 228, 236:
"Congress having adopted the policy of opening the federal courts to suitors in all diversity cases involving the jurisdictional amount, we can discern in its action no recognition of a policy which would exclude cases from the jurisdiction merely because they involve state law or because the law is uncertain or difficult to determine."
The situations where a federal court might await decision in a state court or even remand the parties to it should be the exception not the rule. Only prejudice against diversity jurisdiction can explain the avoidance of the simple constitutional question that is presented here and the remittance of the parties to state courts to begin the litigation anew. Some litigants have long purses. Many, however, can hardly afford one lawsuit, let alone two. Shuttling the parties between state and federal tribunals is a sure way of defeating the ends of justice. The pursuit of justice is not an academic exercise. There are no foundations to finance the resolution of nice state law questions involved in federal court litigation. The parties are entitled—absent unique and rare situations— to adjudication of their rights in the tribunals which Congress has empowered to act.