In this diversity action, appellees brought suit to recover damages for the wrongful deaths of their decedents as a result of the alleged failure of Volkswagen of America, Inc., and Volkswagenwerk A.G. (VW) to design a crashworthy vehicle. VW moved for dismissal and summary judgment on the grounds that the collision was not caused by any design defect. The district court concluded that this Court's decision in Wilson v. Ford Motor Company, 656 F.2d 960 (4th Cir.1981), was not controlling and, therefore, denied VW's motions. VW appeals pursuant to 28 U.S.C. Sec. 1292(b). We reverse.

In Wilson, a van operated by plaintiff's intestate was involved in a collision with another vehicle. The allegations were that the dash of the van was defective, the collision caused it to impact and fracture, and plaintiff's intestate was thrown into the defective dash and sustained severe injuries to his head and other parts of his body resulting in his death. The allegedly defective dash neither caused nor contributed to the collision. Under such circumstances, we held that the North Carolina Supreme Court would not hold a manufacturer liable for injuries arising from the defect. Id.

In the instant case, the 1964 Volkswagen occupied by appellees' decedents and designed, manufactured and distributed by VW, burst into flames due to an allegedly defective gas tank when it was struck by a 1972 Dodge operated by Michael Smith on the wrong side of the highway. It is not alleged that the defective gas tank caused the collision. Because we are bound by our decision in Wilson and because we find the two cases to be indistinguishable, we conclude that the district court erred in denying VW's motions for dismissal and for summary judgment.

Because of our disposition of this appeal, we further conclude that there is no need to decide the other issues raised by appellees. Accordingly, the judgment of the district court is reversed.

REVERSED.

JAMES DICKSON PHILLIPS, Circuit Judge, concurring specially:

Of course I concur in the result announced in the per curiam opinion. I write separately only to address some aspects of Judge Murnaghan's criticism--using the medium of a dissent to denial of rehearing en banc--of our prior panel decision in Wilson.

* I observe initially that there is a certain awkwardness in responding to a criticism first formally expressed by this indirect means when the criticized panel decision is being applied as precedent. Under our procedures designed to insure collegial decisions there exists an opportunity to express criticism of proposed panel decisions before they are filed, and indeed to seek en banc reconsideration of the appeal to forestall the proposed panel decision's becoming precedent. Disagreements within the court are much better expressed and published in that context than in this belated one.

But, conceding the awkwardness, I am constrained to respond to this belated criticism of the panel decision in Wilson. The criticism contains perceptions of our proper role in finding state law in diversity cases with which I strongly disagree.

We of course can only predict with great trepidation the likely course of state decisional law in a context such as this one. Particularly where, as here, the relevant doctrine is in a state of flux with a growing "majority view" apparently waxing apace, and with the composition of the affected state appellate courts in a similar state of flux, the enterprise is doubly difficult.

But it seems to me a particular disservice to the state courts involved for federal diversity courts in such situations to engage in what might be considered the gentle pressure tactics of "assuming" that the state court will necessarily follow a view proclaimed by the federal court to be "enlightened." "Enlightenment" is quite likely, in this as in other matters, to be in the eye of the beholder. Just as there are many reasons other than social or legal enlightenment that may explain a contemporary trend in decisional law, so there are many reasons other than unenlightenment that may explain a particular state's reluctance or failure to fall in with the trend.

Admittedly, when better indicators are not available the trend factor is one that must be considered. But Judge Murnaghan's preference for the Third Circuit's prediction in Seese1--which is rested largely upon the enlightened-trend factor--over that of the Wilson panel2 of this court demeans both the Wilson panel's basis for prediction and the legitimacy of the reasons that North Carolina might have for not following the trend.3

The primary legal indicators for diversity courts should be what the courts of the state have most recently said and what basic doctrinal premises they have seemed most consistently to hold to rather than the way the general law may be trending. Cf. Seese, 648 F.2d at 836-37 (noting error in district court's recent prediction that North Carolina would adopt theory of strict liability in tort). It is on this basis that the Wilson court's prediction--which of course may turn out to have been a faulty one--was based. And it is only to the extent that the indicators relied upon in Wilson can be shown to be irrational that I think the Wilson decision is subject to fair criticism from within or without this court whose law it remains.

To emphasize that the Wilson prediction is not an irrational one, whatever the general trend of the law may be and, more importantly, to emphasize that there are better guides at hand for "finding" present North Carolina law than making substantive judgments about the merits of the rule of decision in question, I offer the following thoughts.

First and foremost is the fact that North Carolina has at this late date not yet joined the crashworthiness "trend."

To dismiss this with the suggestion that it has not been possible to join because an appropriate case has not yet been presented and to forecast that--because of diversity's refuge--it will not likely be presented in the future denigrates the wit both of the North Carolina courts and of counsel practicing in those courts. Courts minded to join "enlightened trends" in decisional law have no difficulty reaching out in "near" cases to join up. This could easily have been done, for example, in Miller v. Miller, 273 N.C. 228, 160 S.E.2d 65 (1968), a case discussed later, where the closely related question of contributory negligence as a proximate cause of "second crash" injury was presented. Cf. Miller v. Premier Corp., 608 F.2d 973, 985-86 (4th Cir.1979) (relying on general trends in choice-of-law doctrine to determine prevailing law where there was a long lack of opportunity in the state courts to apply earlier rule of two isolated cases).

And to break any shackles imposed on state law development in diversity cases by Wilson 's application of assumed North Carolina law, all that is needed is to break up complete diversity by appropriate joinder of defendants in a state court action. Factual patterns presenting that opportunity are surely bound to arise--and almost certainly already have arisen.4

Next is the fact that North Carolina courts have not either been minded to join what may be considered comparable trends in liberalizing tort recovery by such doctrinal expansions as strict liability and comparative negligence.5 While to some this may appear unenlightened, it may to others reflect a completely respectable and deep-seated attitude of judicial restraint and deference to legislative primacy in making significant changes in long-established common law tort doctrine. In any event, judicial restraint in these related areas is a fact that must be taken into account by a federal diversity court in assessing the probable view of North Carolina's appellate courts on the propriety of judicially adopting the crashworthiness doctrine. So the Wilson panel properly did.

Finally, there is the guidance to be had from recent doctrinal expressions by North Carolina's highest court. Judge Murnaghan rightly points out that the crashworthiness principle most directly involves, in conceptual terms, the proximate causation element in tort law. And he correctly points out that, along with other courts, North Carolina's have treated the issue of proximate causation, where it is substantively in play, as one so heavily fact-laden that resolving it as a matter of law on the pleadings or by directed verdict is seldom appropriate. The North Carolina cases cited by Judge Murnaghan certainly bear that out. But they are beside the point.

The underlying conceptual problem in substantive crashworthiness doctrine precisely concerns identification of the accident-occurrence upon which the proximate causation inquiry is to be focused. Is it the initial impact of vehicle with some external object--another vehicle, a tree, a ditchbank--that sets in train a series of traumatic "crashes"? Or is it the specific physical trauma traceable to second (and third, etc.) "crashes" that are in turn arguably traceable in causal terms to design defects that concededly have no causal relation to the "first crash"? Courts that reject crashworthiness doctrine are likely to do so by a purely conceptual analysis that identifies the first impact as the sole accident-occurrence upon which proximate causation injury is rightly focused, with liability for all direct and consequential damages flowing from that impact (including all ensuing "crashes") then being imposed solely upon the actor whose negligence proximately caused that impact.

In these purely conceptual terms, there is strong indication that the North Carolina Supreme Court presently identifies the "first impact" as the critical and sole one for proximate causation--hence tort liability--analysis. In Miller v. Miller, supra, Justice Susie Sharp, writing for the court, employed exactly this analysis in rejecting as a matter of law a vehicle accident defendant's defense of contributory negligence based upon the plaintiff's failure to wear a seat belt. The defense was not available, either in complete bar or for purposes of mitigation of damages by apportionment, wrote Justice Sharp. The plaintiff's failure--even if negligent--"did not contribute to the occurrence of the accident," 273 N.C. at 231, 160 S.E.2d at 68 (emphasis added). Furthermore, damages were not rationally apportionable on a theory of mitigation so that theory must also be rejected as a basis for the defense. Cf. Seese, 648 F.2d at 851-52 (Adams, J., dissenting) (must distinguish injuries attributable to accident from injuries due solely to defect).

Miller is not a pure "crashworthiness"--doctrine case. But its conceptual analysis of the critical causation issue must be the starting point for any honest reappraisal leading to adoption by North Carolina courts of the crashworthiness doctrine. A federal diversity court, seeking to divine state law in this realm, must also take this analysis as a starting point for its own conceptual analysis of where state law presently stands. If, in the event, the federal court's finding of state law based upon that analysis turns out to be faulty, it will not be because the court failed to look first to that primary indicator. On the other hand, any prediction that fails to reckon with this doctrinal expression as the benchmark for conceptual analysis could not be one faithful to our obligation to find state law as it exists at the time for decision.

In this case every relevant consideration suggests that we should simply have reversed on the authority of Wilson, and left the matter there. I wish it had been possible to do so.

MURNAGHAN, Circuit Judge, dissenting from the denial of rehearing en banc:

In view of the Fourth Circuit's established practice of treating one of its panel decisions as binding precedent until there has been an overruling en banc, e.g., Doe v. Charleston Area Medical Center, Inc., 529 F.2d 638, 642 (4th Cir.1975), we were compelled to the result outlined in our per curiam opinion. However, I regard as simply wrong the estimate made in Wilson v. Ford Motor Co., 656 F.2d 960 (4th Cir.1981), of what rule of law the North Carolina Supreme Court would lay down if confronted by the facts in Wilson or the facts in the present case.

The rule set out in Wilson is one of proximate cause. The case holds that, however foreseeable in fact the likelihood of injury might be, an automobile manufacturer, as a matter of law, is free to ignore the inevitable consequences of its negligence when the chain of events triggering a plaintiff's injury (including injury due to the negligent construction of the vehicle) originates with the negligence of a third party. North Carolina law makes clear, though, that "it is only in exceptional cases, in which reasonable minds cannot differ as to foreseeability of injury, that a court should decide proximate cause as a matter of law." Williams v. Carolina Power & Light Co., 296 N.C. 400, 403, 250 S.E.2d 255, 258 (1979). And even where, apparently for policy reasons, North Carolina courts have relieved a class of defendants from excessive liability through proximate cause rules, North Carolina has been willing to draw reasonable lines that will permit cases that cry out for the imposition of tort liability to go to a jury. Compare Williams, supra, with Pugh v. Tidewater Power Co., 237 N.C. 693, 75 S.E.2d 766 (1953). See also Dupree v. Batts, 276 N.C. 68, 170 S.E.2d 918 (1969).

There is to be considered the by no means negligible authority of Seese v. Volkswagenwerk A.G., 648 F.2d 833 (3d Cir.1981), which addressed the very "crashworthiness" question presented by Wilson and the instant case. Seese came to the contrary conclusion that North Carolina tort law did include an actionable tort based on uncrashworthiness. The court stated, with persuasive authority to back it up:

In the absence of any expression by North Carolina and a split of authority by federal courts in that state, a prediction as to the law North Carolina would adopt can only be based on the greater persuasiveness of one of the conflicting theories, with an eye to the nationwide trend in judicial and legislative law-making.... [W]e are not disposed to base our prediction of North Carolina law on an outdated and repudiated doctrine. Like the district court here, we too are convinced that North Carolina is more likely to follow the enlightened rule of Larsen [v. General Motors Corp., 391 F.2d 495 (8th Cir.1968) ] rather than the rule of Evans [v. General Motors Corp., 359 F.2d 822 (7th Cir.1966) ].

648 F.2d at 841.

It is of no little importance that Judge Adams, the dissenter in Seese, did not reason that there was no uncrashworthiness tort in North Carolina. To the contrary, he accepted that there was such a North Carolina tort and found fault in the manner of computing damages. He voted for "remand for a reassessment of damages, ... to retry the crashworthiness claim ...." Seese, supra, 648 F.2d at 856.

Hence, for the purposes of our decision, made at a point in a case's development before damages have even been addressed, much less assessed, all members of the Seese court agree that North Carolina courts would permit recovery, upon an adequate showing of the car manufacturer's contribution to the injuries suffered as a consequence of uncrashworthiness.

Bearing in mind the likelihood that North Carolina will be unlikely to have an opportunity to consider a case raising the issue,1 we owe a duty not to shrug off an erroneous decision on the grounds that, if incorrect, it will all in due course be set straight by a North Carolina court.

It would be possible, on the basis of North Carolina authority, to construct narrow distinctions stringently circumscribing Wilson and in effect restricting it to the very facts of the particular case. However, that would be overruling in all but name, and the procedure is not a congenial one so far as I am concerned, given our rule developed in a sense of, and with the purpose of maintaining, collegiality.

Without in any way intimating what would happen if the case were permitted to be submitted to a jury, I am satisfied that, under North Carolina law, the allegations pleaded by the plaintiff, if made out at trial, are sufficient to permit recovery. I therefore dissent from our refusal to reconsider the question en banc which has the regrettable consequence of perpetuating an incorrect assessment of North Carolina law.