This Louisiana diversity appeal requires us to determine whether the law of that state would dispense with proof of causation against some of multiple defendants in an asbestosis case.

Plaintiff Thompson, employed as an insulation worker from 1952 to 1978, handled products containing asbestos during that period. As a result, he contends, he inhaled dust and fibers that produced the disease or condition of asbestosis. He was able to recall the brand names of some of the products to which he was exposed. Because he did not recall those of today's four appellees and was unable to point to any other evidence connecting their products with his alleged injuries, each was granted summary judgment on the ground that as to it the element of causation was lacking as a matter of law. Thompson appeals, suggesting the applicability to his case of various legal theories that he asserts dispense with causation as requisite to a recovery of damages. Partly on the basis of an Erie guess, we affirm.

We are met at the threshold of our analysis with the contention of several appellees that since Mr. Thompson filed no response to their motions for summary judgment, we need look no further and must affirm on that basis.

Appellees point to no case law supporting their proposition but do cite FRCP 56(e), which reads in relevant part:

When a motion for summary judgment is made and supported as provided in this rule, an adverse party may not rest upon the mere allegations or denials of his pleading, but his response, by affidavits or as otherwise provided in this rule, must set forth specific facts showing that there is a genuine issue for trial. If he does not so respond, summary judgment, if appropriate, shall be entered against him.

The Notes accompanying the 1963 Amendment to this rule explain that these two sentences were added to overcome a line of cases which existed, chiefly in the Third Circuit, permitting parties defending against motions for summary judgment to rest on their pleadings.

The quoted language generally supports the argument advanced; however, recent precedent in our circuit, by which we are bound, squarely defeats it. In White v. Thomas, 660 F.2d 680, 682-83 (5th Cir.1980), cert. denied, 455 U.S. 1027, 102 S. Ct. 1731, 72 L. Ed. 2d 148, we held that the court must look to the allegations of facts made in the pleadings even where plaintiff,

filed nothing in opposition to the motion save a memorandum asserting that some of the facts were disputed. Even in cases where the party against whom summary judgment is sought fails to comply with Fed.R.Civ.P. 56(e), it is inappropriate to grant summary judgment on the merits on those issues whose decision depends on resolution of evident factual disputes not put beyond dispute by the affidavit, especially if a limited evidentiary hearing did not afford the party an opportunity to adduce his full case.

Various theories are advanced by Mr. Thompson for dispensing with proof that the products of these defendants caused or contributed to cause his condition. Some clearly will not serve to do so.

Clearly inapposite to his case are such authorities as Hilburn v. Johnson, 240 So. 2d 767 (La.Ct.App.1970), and Landers v. East Texas Salt Water Disposal Co., 151 Tex. 251, 248 S.W.2d 731 (1952). These permit the imposition of joint and several liability on persons whose separate wrongful actions, not done in concert, contribute in unknown proportions to cause an indivisible injury. The first concerns two persons who collided with a motorcyclist, causing his death. The second treats of two pipeline carriers whose separate salt water lines ruptured, polluting a lake. In neither was there doubt that each of the two defendants in fact contributed to the overall injury in some degree; and neither authority does more than decline to exonerate a defendant merely because the amount by which he contributed to cause it cannot be ascertained. Even less in point are concert of action theories, where persons acting in knowing collaboration cause an injury and are held liable in solido for the effect of their common scheme.

Also to be distinguished are theories of alternative liability such as those of Summers v. Tice, 33 Cal. 2d 80, 199 P.2d 1 (1948) (liability imposed on both of two hunters who fired toward plaintiff, burden on each defendant to show he did not cause injury), and of group res ipsa loquitur. Ray v. Ameri-Care Hospital, 400 So. 2d 1127 (La.Ct.App.1981). All of these concern defendants who were proved to have some factual connection with the plaintiff's injury; to apply them to defendants as to which there is no proof of any such connection would beg the question of causation entirely. Like the indivisible injury cases, these shift the burden of proof of causation from the injured plaintiff to those defendants who were connected with it and one or more of whom undoubtedly caused it although, unlike those cases, the result may be that--because of his inability to prove himself innocent--one who in fact contributed nothing to the injury may still be held liable for it. The policy choice underlying such theories seems to be that in the extreme case where one or more of a defined group unquestionably caused an injury and the choice is between exonerating all because the plaintiff cannot show which did so and holding all liable who cannot prove their innocence, the latter result is preferable. Such is not the case here, where a large group of defendants whose products Thompson did recall using remain before the court. These defendants are not members of it.

Mr. Thompson's final set of theories do, in fact, dispense with particular proof of causation. These are ones of "enterprise" and "market share" liability, exemplified by Hall v. E.I. DuPont de Nemours & Co., 345 F. Supp. 353 (E.D.N.Y.1972), suggesting the possibility of joint industry-wide liability for particular blasting cap injuries, and Sindell v. Abbott Laboratories, 26 Cal. 3d 588, 163 Cal. Rptr. 132, 607 P.2d 924 (1980), imposing and allocating several liability by market share. We see little purpose in discussing in detail the potential applicability of these theories to Mr. Thompson's case; writing in diversity, we write on the wind. Louisiana will or will not adopt either or both at some future time. In no reported case has it done so at present, nor have any substantial number of other jurisdictions. Both theories represent radical departures from traditional theories of tort liability. All that Thompson can advance in support of his claim that Louisiana would adopt either if presented his case is a supposed general tendency or trend on the part of Louisiana courts to expand the liability of manufacturers.

That is not enough to support our adoption for Louisiana of a particular and radical mode of its expansion. Such departures are for the Louisiana courts, not for us. See Rhynes v. Branick Mfg. Corp., 629 F.2d 409 (5th Cir.1980) ("product line" liability).

AFFIRMED.

Today the majority has determined to take an Erie guess and "write upon the wind." The majority predicts that Louisiana, if faced with the novel and distinct problems of asbestos related injury, would do nothing more than reflexively apply traditional tort doctrines. I am unwilling to arrogate to myself the decisionmaking duties and competence of the Louisiana Supreme Court. Rather than writing upon the wind, I would consult the chief meteorologists; this case of first impression is an ideal specimen for certification, and because the majority declines to do so, I must dissent.

Disease caused by exposure to asbestos presents monumental problems for tort law relating to causation. First, it is nigh impossible to point to a single fiber of asbestos as the cause of a cancerous lesion in the lungs of an asbestosis victim; asbestos related diseases usually develop after a period of exposure, perhaps from the products of many manufacturers. Second, even if the period of exposure were known, the long time between exposure and onset of symptoms may make it impossible for the victim to remember the sources of exposure. Third, even if the victim has a photographic memory, that might be of no help--much exposure is the result of previously installed asbestos-containing material, so the victim had no chance to learn the source.

My point here is not to dramatize excessively the legal plight of asbestos victims, but to illustrate that the factual predicate giving rise to potential liability from asbestos exposure is simply different from those that generated most tort doctrines. Without enumerating all of the occasions, suffice it for me to note that tort law has been an area quite open to evolutionary, if not revolutionary, forms of liability in response to unanticipated factual paradigms. This has been particularly true in the area of causation, with a steady evolution, as circumstances demanded, from simple one-party negligence, to joint liability, to alternate liability, to burden-shifting presumptions such as res ipsa loquitur, to theories of market share and enterprise liability. Whether or not asbestos related injury is a circumstance calling for a doctrinal response is a weighty and difficult question; given the increasing inundation of the courts with asbestos related litigation, it is also a question of great consequence.

As I have briefly discussed, asbestos related injuries are different in legally important aspects from those types of injuries that present tort doctrines were designed to accommodate. Whether or not those doctrines will be extended to cover asbestos cases is a different sort of question from whether or not those doctrines will be applied to an automobile accident. In one instance a court writes upon a blank slate, in the other a court is closely hemmed in by precedent upon precedent.

In the case before us now, the majority affirmatively decides to apply traditional tort doctrines to a new and different species of injury. The majority notes that there are no Louisiana cases relying on market share or enterprise liability and justifies its result with a reluctance to impose a "radical" change in tort law upon Louisiana. What the majority chooses to ignore, however, is that there are no Louisiana cases refusing to apply market share or enterprise liability to similar factual situations. This deliberate blindness enables the majority to sub silentio impose upon Louisiana an equally radical extension of traditional tort doctrines to a case for which those doctrines are ill adapted. Asbestos related injuries are different and this is an important case of first impression.

One of the stellar virtues of Our Federalism is that it provides a hothouse in which fifty-one flowers may blossom. The federal courts, moreover, are not cast in the role of head gardener. Under Erie Railroad v. Tompkins, 304 U.S. 64, 58 S. Ct. 817, 82 L. Ed. 1188 (1938), federal courts sitting in diversity must follow state law. Occasionally, when the law of the state is not crystal-clear, a federal court may have to make an "Erie guess." See, e.g., Meredith v. City of Winter Haven, 320 U.S. 228, 64 S. Ct. 7, 88 L. Ed. 9 (1943). However, federal courts ought not necessarily assume the responsibility of deciding in the first instance important, unsettled questions of state law. E.g., Kaiser Steel Corp. v. W.S. Ranch Co., 391 U.S. 593, 88 S. Ct. 1753, 20 L. Ed. 2d 835 (1968). Cf. Clay v. Sun Insurance Office Ltd., 363 U.S. 207, 80 S. Ct. 1222, 4 L. Ed. 2d 1170 (1960).

We have a powerful mechanism for avoiding unnecessary Erie guesses--certification. See generally Brown, Certification --Federalism in Action, 7 Cum.L.Rev. 455 (1977) (one of certification's most ardent advocates, Circuit Judge John R. Brown, stating his case). When appropriate we may turn to the font of wisdom of Louisiana law for an authoritative statement. I can think of few cases more appropriate for certification than this one. The legal principles involved are unsettled and far-reaching. The rule established will have great impact on the well-being of the citizens of Louisiana. I am aghast at the possibility that after we deny this plaintiff relief, Louisiana will acknowledge enterprise or market share liability as a viable theory, but that is a risk the majority is willing to take.1CONCLUSION

As the majority opinion clickety-clacks down the Erie tracks, I fear I hear the sound of the cross-ties splintering. Were I the switchman I would sidetrack this case to the Louisiana switching yards where a locomotive of sufficient power to pull the freight of the majority opinion might be attached. As it is, I remain alone in the caboose, dissenting.