1967]

FEDERAL COURTS AND STATE LAW

321

Finally, one dependable barometer of the harmony of diversity with state autonomy is the volume of complaints from the states sparked by Erie practice. Throughout American history state governments and their spokesmen have eagerly and vehemently objected to every felt or imagined federal trespass onto state preserves. If any state legislators, judges, lawyers, or litigants have issued public protests founded on states' rights principles against federal court assumption of diversity jurisdiction, those protests have not crossed my desk.

II

All of these elements lead me to believe that the ALI has judged diversity jurisdiction presumptively bad for reasons that are amazingly slight. Reasons warranting hostility to diversity may still exist, however, even though the ALI has failed to cast them under its spotlight. Surprisingly, Professor Moore's article on diversity defends it while only once mentioning the doctrine of Erie v. Tompkins.“ I express surprise because in my view serious deliberation of diversity cannot proceed unless Erie is squarely fastened in the forefront of our attention. Only in this context does the most serious objection to diversity become visible: diversity practice under Erie can be an awkward, ineffectual way of administering a system of judge-made law.

This awkwardness is easily demonstrated. First, the process of deciding cases under Erie is laborious, often onerously so, at least when state law on the point at issue is less than immaculately clear. Our mandate from Erie is to find out how the state high court would decide the issue today." In making that prediction, the federal judge exhaustively analyzes all the state court cases even remotely in point. While federal courts are not necessarily bound by lower state court rulings, or by the dicta and ancient holdings of the state high court, our practice, born of fear of federal-court law-making, is to defer to them far more generously than the state high court itself probably would."1

Nor is our inquiry ended simply by consulting the state cases. The Erie court must search out all the legal materials-law review articles, the treatises on the subject, and the cases from other jurisdictions— which might impress and influence the state high court in deciding the issue today. We should not be surprised, then, that Erie-opinions are

19. Moore & Weckstein, supra note 6, at 20.

20. Nolan v. Transocean Airlines, 365 U.S. 293, 295-96 (1961). See generally Harnett & Thornton, Precedent in the Erie-Tompkins Manner, 24 N.Y.UL. Rev. 770 (1949); Notes, 59 Harv. L. Rev. 1299 (1946), 40 Ind. L.J. 541 (1965).

21. Eg, Franklin Life Ins. Co. v. State Neon Sign Co., 329 F.2d 456 (5th Cir. 1964); Doucet v. Middleton, 328 F.2d 97 (5th Cir. 1964); McClourth Steel Corp. v. Mesta Mach. Co., 214 F.2d 608 (3d Cir. 1954).

322

WAYNE LAW REVIEW

[Vol. 13

laced with paragraph-long statements of the facts of conceivable precedents, and with long string citations representative of what one commentator calls the “legal-massive” style of legal writing.*

It has even been contended that Erie obliges the federal court somehow to intuit underlying themes or messages of the state's jurisprudence, and to elicit by induction from reported cases the decisionmaking techniques of the highest state court: for example, its implicit attitude toward stare decisis." Meeting these responsibilities would surely foster more accurate predictions. But they would also require the federal court to digest and master the entire corpus of state law; my fear is that we on the federal bench lack the acuity, leisure, and stamina for undertaking these intellectual ordeals.

My point, then, is that it is easier to make good law than successfully to predict how it will be made. The state-law opinion of a state high court can quite acceptably ride along a crest of common sense, avoiding the extensive citation of authority. If anything, most judges question the value of an overly scholastic, law review-like judicial method. If the orgy of Erie research yielded a payoff in terms of a wiser result, it of course would find justification. But, remember, the federal-court research is dedicated to the single mission of rendering its determination indistinguishable from the position the state court itself would take, be it wise or otherwise.

There are other serious diseconomies in the administration of justice Erie-style. Reason tells us, as experience confirms, that Erie retards the formation and development of state law. Let us imagine a lawsuit bringing into view an undecided or ambiguous state law question. If suit is brought in state court, and if the losing parties pursue their successive appeals upward to the highest court of the state, the issue of state law will be definitively laid to rest by that court's decision. But if, instead, a federal court expropriates the case under its diversity jurisdiction, the end result of the litigation, even if it works its way to the federal court of appeals or the Supreme Court itself, will only be a federal court estimation of the state law issue. Its conclusive resolution is thereby postponed until the later date, if ever, when another controversy raises the issue anew, and then, again, only if the parties to that controversy determine to litigate, and if circumstances conspire to lead the case to state court.

Let me offer you a few examples of this retardation of state law

22. E.g., Union Bank v. First National Bank, 362 F.2d 311 (5th Cir. 1966); Putnam v. Erie City Mfg. Co., 338 F.2d 911 (5th Cir. 1964); Cullborg v. Rizzo, 331 F.2d 557 (3d Cir. 1964); Ohio Cas. Ins. Co. v. Smith, 297 F.2d 265 (7th Cir. 1962). The author is, again, John Frank, Marble Palace 297 (1958).

23. Note, 40 Ind. L.J. 541, 553-54 (1965),

1967]

FEDERAL COURTS AND STATE LAW

323

growth. Were there no diversity jurisdiction, the citizenry and bar of Tennessee, Texas, and Mississippi would now be advised as to whether “privity" is still a bar to the suit of an ultimate user against a manufacturer for negligence. Instead, they have only the guesswork, educated though it may be, of the Sixth and Fifth Circuits on these state questions." But for diversity, the returns would now be in as to whether under Kansas law a wife may sue for loss of consortium, and whether the death statutes in South Carolina and Pennsylvania recognize a right of action on behalf of a stillborn fetus for injuries suffered during viability."

To prolong legal uncertainty, other things equal, is a misfortune; and there is no reason to suspect that stalling the resolution of these legal questions has reaped any compensatory benefits. Rather, the student must join in the sentiments Judge Sobeloff expressed in a recent suit between insurance companies concerning a Virginia statute. “It is regrettable,” he chided the parties, “that two insurance companies operating in Virginia should avoid resort to the only court that is empowered to give a final authoritative answer to this question of statutory construction.””

The final and by far the gravest objection to Erie is that it can badly squander the resources of the federal judiciary. Ordinarily, a legal decision has a dual achievement. First, it decides the case before it, settling the parties' rights and liabilities. From a higher perspective its more profound function, carried out through its holding and opinion, is to establish a precedent and organize a body of law. It is axiomatic that were there an absence of precedent the uncertainties of the law would unnerve human behavior and overwhelm the courts with lawsuits resulting from lawyers' inability reliably to predict the courts' doctrinal positions.

A glaring characteristic of Erie litigation is that it is competent to pursue only the former of these objectives. An Erie decision fixes no binding precedent, it stills the controversy brewing in no region of the law. A state court can come along after an Erie decision has been handed down and ignore it altogether; or, citing it, can simply say it was decided wrongly.” That state courts can so behave exposes the wasteful use of the federal judiciary which Erie often entails.

24. Berry v. American Cyanamid Co., 341 F.2d 14 (6th Cir. 1965); Putman v. Erie City Mfg. Co., 338 F.2d 911 (5th Cir. 1964); Necaise v. Chrysler Corp., 335 F.2d 562 (5th Cir. 1964).

25. Criqui v. Blaw-Knox Corp., 318 F.2d 811 (10th Cir. 1963); Todd v. Sandridge Constr. Co., 341 F.2d 75 (4th Cir. 1964); Cullborg v. Rizzo, 331 F.2d 557 (3d Cir. 1964). 26. Maryland Cas. Co. v. Burley, 345 F.2d 138, 139 (4th Cir. 1965).

27. Kelley v. City of Austin, 268 S.W.2d 773 (Tex. Civ. App. 1954), negating City of Lubbock v. Green, 201 F.2d 146 (5th Cir. 1953); Rader v. Pennsylvania Turnpike

324

WAYNE LAW REVIEW

[Vol. 13

This waste can be graphically illustrated. Four years ago, in the Merritt-Chapman & Scott Co. case," the Second Circuit in an Erie appeal was confronted with the question whether under New York law sovereign immunity permitted attaching a levy on the New York funds of a State of Washington public utility district. The court's eminently sensible conclusion was that sovereign immunity would forbid the levy if, but only if, its imposition would substantially interfere with the performance by the district of its key public functions. To reach this result, however, the court had to work its way through a forest of state cases and other authorities. By my rough count, Judge Medina's illuminating opinion contains 71 cites to New York decisions, many more to New York statutes and public documents, 38 to cases from other jurisdictions, 23 to treatises or law review articles; and the court alludes to its additional acquaintanceship with “a vast body of decisional law" throughout the nation on the sovereign immunity issue. Before making its opinion official, the court, borrowing from the practice of administrative agencies, turned it over to the parties, soliciting criticism; after submisson of another round of briefs, the court added several new paragraphs of text.

Judge Medina's synthesis and rationalization of this confusing appendage of the law is a masterful job, framed in an opinion which is an awesome legal document. It is just appalling that such sharp research and legal reasoning, undertaken here by perhaps the country's finest court, the Supreme Court aside, should issue in a judgment that fails to establish even for the immediate future what the law is. It is woeful to contemplate that the New York Court of Appeals or, alas, even a trial judge on the state supreme court, trial term, can effectively repudiate Judge Medina's holding overnight. Ever since I was kicked upstairs to the Court of Appeals for the District of Columbia Circuit, I have been virtually relieved of Erie duties, but if I somehow were again presiding over an Erie suit implicating a legal issue as confounding as the one Judge Medina dealt with, I would enter into my research with a dangerously low morale.

In a sense, however, I do concede that Meritt-Chapman is a biased illustration, stretching the Erie-originating disadvantages to an extreme. Recently I read through all the Court of Appeals cases applying Erie reported during the last few years; Merritt-Chapman by a clear margin was the most awful instance I found of this misallocation of

Comm'n, 407 Pa. 609, 621, 182 A.2d 199, 205 (1962) (dissent); See also Albin v. National Bank, 60 Wash. 2d 745, 755, 375 P.2d 487, 493 (1962) (dissent).

28. Merritt-Chapman & Scott Corp. v. Public Utility Dist. No. 2, 319 F.2d 94 (2d Cir. 1963), cert. denied, 375 US. 968 (1964).

1967]

FEDERAL COURTS and STATE LAW

325

judicial resources. Nevertheless, the evils it vividly expresses are unique only in their degree; those evils are endemic to the Erie situation. Unless, that is, a way can be found, within the Erie framework, for transferring the law-declaring aspect of a diversity suit back to the state courts. Two interesting practices have developed along these lines, and they should be considered here.

The first of them, of course, is federal court abstention. Although the Supreme Court on the heels of Erie pronounced that the obscurity of state law furnishes no excuse for the federal court in a diversity case to abstain from decision," one of its 1959 opinions can plausibly be read as modifying this principle;" and sporadically federal courts have self-effacingly ruled that abstention was in order." This abstention, however, effectively deprives the litigants of trial in a federal court. In my view that fact alone demonstrates its wrongheadedness. Congress having determined that parties are entitled to invoke diversity jurisdiction, it is no business of the federal court to withdraw that permission.

A far more ingenious escape from Erie's troubles is the certification procedure." This happens when a federal appellate court "certifies" or remands a troubling question of state law involved in an Erie suit to the state high court for its resolution. Since certification does not forfeit the parties' rights to all the advantages of a federal trial, it conceivably encompasses the best of all possible worlds. So far, however, only one or two states have passed legislation enabling their courts to receive these federally-certified questions. Congress may have authority to require state courts to accept and decide certified questions. Such legislation would be rationally connected to valid federal objectives, and I perceive no constitutional prohibition cancelling federal authority in this area." But doubtlessly it would be highhanded for federal law to pose such decisional obligations on the unconsenting highest organ of the state's judiciary.

Certification is not without other notable drawbacks. Answering legal questions abstracted from the gestalt of a lawsuit is invariably a dubious way for courts to make law. To shuttle back and forth be

29. Meredith v. Winter Haven, 320 US. 228, 234 (1943).

30. Louisiana Power & Light Co. v. City of Thibodaux, 360 US. 25 (1959). 31. Eg, United Services Life Ins. Co. v. Delaney, 328 F.2d 483 (5th Cir. 1964). Compare Commerce Oil Ref. Co. v. Miner, 303 F.2d 125 (1st Cir. 1962).

32. See Liebenthal, A Dialogue on England: The England Case, Its Effect on The Abstention Doctrine, and Some Suggested Solutions, 18 W. Res. L. Rev. 157, 201-09 (1966); Note, 73 Yale L.J. 850 (1964).

33. See Fla. Stat. § 25.031 (1961); Fla. App. R. 4.61.

34. But cf. England v. Louisiana State Bd. of Medical Examiners, 375 US. 411, 434 (1964) (Douglas, J., concurring) (Supreme Court lacks power to compel states to allow certification).