Abel v. Eli Lilly & Co.

Williams, C.J. This case is but one of many filed *317 in state and federal courts by daughters of women who had taken DES during pregnancy and their spouses against the manufacturers of synthetic estrogen products. Very briefly, synthesis of estrogen was first reported by C. E. Dodds, a British researcher, in 1938. Dr.’ Dodds never patented the drug, known as diethylstilbestrol or DES, thus allowing any manufacturer to develop the drug who chose to do so. The Food and Drug Administration first granted several companies’ requests to market DES for non-pregnancy uses in 1941. In 1947, several companies filed supplemental requests to market DES to prevent complications in pregnancy. The FDA granted permission to market the drug for pregnancy uses the same year, and the drug was thereafter generically marketed 1 for pregnancy uses. In 1971, Dr. Herbst reported a statistical association between the use of DES by pregnant women and cancer and other cellular abnormalities in the *318 reproductive systems of their children. 2 As a result, the FDA banned the marketing of DES for use by pregnant women. It is the women who were exposed in útero to DES and their spouses who now seek to hold the drug manufacturers liable for their injuries in cases filed throughout the country. Although each case presents its own factual nuances, one common problem continually reappears: many of the plaintiffs are simply unable to identify the manufacturer of the estrogen product to which they were exposed. The plaintiffs therefore seek some way to circumvent the traditional tort element of causation in fact. In order to bypass the identification requirement, plaintiffs assert that defendants (who are purportedly all the manufacturers of synthetic estrogens for pregnancy use in Michigan) are jointly and severally liable. Several theories have been advanced to justify this joint liability including alternative liability, concert of action, and what is referred to as "collective” or "industry-wide” liability. Defendants jointly sought partial summary judgment pursuant to GCR 1963, 117.2(1) against all plaintiffs who were unable to identify the manufacturer of the prescription drug allegedly responsible for their injuries. The trial judge denied this motion, without prejudice, and ordered further *319 discovery on "the alleged joint or enterprise liability and conspiracy of all defendants”. After approximately two years of discovery, the defendants again moved jointly for partial summary judgment on the basis of GCR 1963, 117.2(1) and (3). Finding that the plaintiffs had indeed failed to state a cause of action, the trial judge granted summary judgment in favor of defendants and against those plaintiffs who were unable to identify the manufacturer of the drug which allegedly caused their injury. Also, with regard to those plaintiffs able to identify the manufacturer, the court dismissed all other defendants. On appeal, the Court of Appeals reversed the summary judgment, finding that the plaintiffs had made sufficient allegations to support the alternative liability and concert of action theories. Abel v Eli Lilly & Co, 94 Mich App 59, 77; 289 NW2d 20 (1979). The Court of Appeals reviewed only the defendants’ GCR 1963, 117.2(1) claims (failure to state a cause of action); the GCR 1963, 117.2(3) issues (no genuine issue of material fact) were not addressed. We are called upon today to review the correctness of these rulings. We hold that the plaintiffs have made sufficient allegations to support both the concert of action and alternative liability causes of action. Pleadings The female plaintiffs in this case allege that they are daughters whose mothers when pregnant with each took a synthetic estrogen product upon their doctors’ advice in order to prevent miscar *320 riage. The remaining plaintiffs are spouses of the daughters. 3 The complaint alleges that as a result of prenatal exposure to these drugs, the plaintiffs are now suffering from cancerous or pre-cancerous lesions of the vagina. According to the complaint, the defendant drug companies 4 actively promoted and sold these drug preparations to the medical community, representing that the drugs were safe and effective for use by pregnant women in the prevention of miscarriage. As a result of these representations the drugs were administered to pregnant women during a time period from 1947 to 1964. Plaintiffs maintain that the defendants knew or should have known that the drugs which they marketed were derived from stilbene, a known cancer-causing agent, and that scientific investigators had recommended against medical use of stilbene derivatives because of possible cancer risks. The complaint also contends that these synthetic estrogen drugs are ineffective in preventing miscarriage. Plaintiffs’ initial complaint stated that all the plaintiffs’ mothers had ingested one of seven synthetic estrogens. This number has since been reduced to three: diethylstilbestrol (DES), dienestrol, and diethylstilbestrol dipropionate (DSD). The complaint sounds in negligence, breach of express and implied warranty, fraud and deceit, strict liability, and conspiracy. 5 The plaintiffs seek *321 to impose joint and several liability upon all the defendants. Some plaintiffs have specifically named the product to which they were exposed in útero and its manufacturer. Other plaintiffs acknowledge that they are unable to identify the specific estrogen product and that they are unable to identify the manufacturer that produced or distributed the product. They allege that good-faith efforts have been made to determine the specific product and manufacturer. However, they cite Michigan law, MCL 338.1118(1); MSA 14.757(18X1), which required the preservation of the prescription records of pharmacies for only five years and the generic marketing scheme utilized by defendant drug companies to explain this lack of identification. Finally, the complaint states that "the defendants named herein constitute all of the known manufacturers or distributors of stilbene derivatives during the time period in controversy: to wit, 1947 to 1964.” 6 The answers of defendants generally deny the significant allegations of the complaint. Motion for Summary Judgment Defendants moved for a summary judgment under GCR 1963, 117.2(3) which states: "The motion for summary judgment shall state that the moving party is entitled to judgment in his favor because of any 1 of the following grounds: *322 "(3) That except as to the amount of damages there is no genuine issue as to any material fact, and the moving party is therefore entitled to judgment as a matter of law.” A motion under subsection 3 asks the court to test the factual foundation of the suit. The defendants have directed the court’s scrutiny to three "factual deficiencies” in plaintiffs’ case: 1) lack of proof that all defendants are before the court (as required by their alternative liability claim); 2) lack of evidence of concerted activity among the defendants in the promotion and manufacture of their estrogen products; and 3) failure to define the applicable "industry” for their "industry-wide” liability theory. While defendants’ subsection 3 motions were properly filed with the trial court, the trial judge never reached these factual questions, having disposed of the case by ruling that the plaintiffs’ pleadings failed to state causes of action under alternative liability and concert of action theories. Nor did the Court of Appeals address defendants’ subsection 3 motion. The Court of Appeals panel confined its review to the correctness of the ruling on the subsection 1 motion. Therefore, no lower court has yet evaluated the defendants’ contention that no genuine issue of material fact exists in the present record. Since the issue has not been discussed below, we decline to review these claims. Brill v Cherwin, 346 Mich 507, 514; 78 NW2d 122 (1956); Brewster v Martin Marietta Aluminum Sales, Inc, 107 Mich App 639, 646; 309 NW2d 687 (1981). However, it is clear that this motion has not been withdrawn or abandoned by the defendants. Therefore, unless it is, on remand and prior to this *323 cause being tried, the trial court is required to hear and decide defendants’ motion for summary judgment pursuant to GCR 1963, 117.2(3). Defendants also assert that plaintiffs’ affidavits in opposition to the defendants’ motion for summary judgment were not made on personal knowledge as required by GCR 1963, 117.3 and 116.4. Necessarily encompassed within the trial court’s decision as to the GCR 1963, 117.2(3) motion is the determination whether three affidavits of the parties are sufficient to meet the requirements of the court rule. Defendants also sought summary judgment under GCR 1963, 117.2(1) which states: "The motion for summary judgment shall state that the moving party is entitled to judgment in his favor because of any 1 of the following grounds: "(1) the opposing party has failed to state a claim upon which relief can be granted, "(3) * * * and the moving party is therefore entitled to judgment as a matter of law.” A subsection 1 motion for summary judgment seeks to test the genuineness of a claim by challenging the legal adequacy of the pleadings. 1 Honigman & Hawkins, Michigan Court Rules Annotated (2d ed), 1982 Supp, p 142. The standard used to assess the subsection 1 motion has been well stated: "The test which the court should apply in considering motions under GCR 1963, 117.2(1) is whether plaintiff’s claim, on the pleadings, is so clearly unenforceable as a matter of law that no factual development can possibly justify a right to recovery.” Crowther v Ross Chemical & Mfg Co, 42 Mich App 426, 431; 202 NW2d 577 (1972). *324 In applying GCR 1963, 117.2(1), the trial court does not act as a factfinder, nor does the court attempt to probe the parties’ ability to prove their allegations. Thus, the court accepts as true all well-pleaded 7 facts. Bielski v Wolverine Ins Co, 379 Mich 280, 283; 150 NW2d 788 (1967); Hiers v Detroit Superintendent of Schools, 376 Mich 225, 233; 136 NW2d 10 (1965). Defendants find two deficiencies in the plaintiffs’ allegations: first, plaintiffs have failed to establish the causation in fact element which is required in any products liability action, and, second, plaintiffs assert a collective, industry-wide liability cause of action which is not recognized under Michigan law. I. Alternative Liability Count (A) Asserted Deñciency in Pleading Defendants correctly assert that the threshold requirement of any products liability action is identification of the injury-causing product and its manufacturer. Piercefield v Remington Arms Co, Inc, 375 Mich 85, 98-99; 133 NW2d 129 (1965). The plaintiff must produce evidence of a defect which caused the accident and trace that defect into the hands of the defendant. 8 Caldwell v Fox, 394 Mich *325 401, 410; 231 NW2d 46 (1975). See also 1 Hursh & Bailey, American Law of Products Liability 2d, § 1.41, p 125; Anno: 51 ALR3d 1344, 1349. It is undisputed that all the plaintiffs in this action will not be able to meet the identification requirement. Paragraph 16 of the Thirteenth Amended Complaint reads: "some of the plaintiffs will be unable to identify the specific defendant that manufactured the injury-producing drug that was sold to their mothers in spite of all good-faith efforts to determine same”. Despite this hiatus in the pleadings, defendants recognize that plaintiffs would still have set forth a cause of action if their factual situation meets the requirements of the theory of alternative liability. Also called "clearly established double fault and alternative liability”, 9 this procedural device shifts the burden of proof on the element of causation in fact to the defendants once an innocent plaintiff demonstrates that all defendants acted tortiously, but only one unidentifiable defendant caused plaintiffs injury. Summers v Tice, 33 Cal 2d 80; 199 P2d 1 (1948). If the defendants cannot meet this burden and exculpate themselves, joint and several liability will be imposed. Defendants assert that plaintiffs’ reliance on alternative liability is totally unsupportable by the pleaded facts of the present case. In order to evaluate this claim, we must explore the theory of alternative liability as it has developed in the jurisprudence of this state. (B) Alternative Liability The doctrine of alternative liability first received *326 formal recognition 10 in the case of Summers v Tice, 33 Cal 2d 80; 199 P2d 1 (1948). In that case, plaintiff was injured by one shot during a hunting expedition. Plaintiff sued both of his hunting companions, claiming that both negligently shot at him, although he was unable to determine which one fired the specific shot that injured him. The Summers court agreed, as a preliminary matter, that both defendants were at fault in having acted negligently toward the plaintiff. The court also recognized that plaintiff had failed to meet his traditional burden regarding cause in fact. 11 Only one shot had injured plaintiff; both defendants could not have shot it. Since the most the plaintiff could prove was a 50% probability that either of the defendants had caused the injury, plaintiff had failed to establish that either one of the defendants was more likely than the other to have caused the injury. 33 Cal 2d 84. The court then decided — as a matter of policy— that it was preferable that the two wrongdoers, both of whom had acted negligently toward the plaintiff and had created the situation wherein plaintiff was injured, should bear the burden of absolving themselves rather than leaving the innocent plaintiff remediless. Therefore, the court *327 placed the burden of proof on the issue of causation in fact upon the defendants. 12 This rule is now embodied in 2 Restatement Torts, 2d, §433B(3), pp 441-442. 13 Commentary to the Restatement agrees that the reason for the exception to traditional rules is to prevent the injustice of allowing proved wrongdoers to escape liability for an injury inflicted upon an innocent plaintiff "merely because the nature of their conduct and the resulting harm has made it difficult or impossible to prove which of them has caused the harm”. Comment f, p 446. Accord, Bowman v Redding & Co, 145 US App DC 294, 305; 449 F2d 956 (1971). This policy has found expression in the opinions of this state’s courts. For example, Benson v Ross, 143 Mich 452; 106 NW 1120 (1906), presented a fact situation quite similar to Summers. Three defendants were taking turns shooting a single rifle when one of the shots struck the plaintiff. Although the plaintiff was unable to identify which one of the three negligent defendants fired the injury-causing shot, the Court found that the plaintiff had alleged sufficient facts to withstand a motion for directed verdict, relying on a concert of action theory. This principle was also developed in Holloway v *328 General Motors Corp (On Rehearing), 403 Mich 614, 626, 628; 271 NW2d 777 (1978), in which the question arose whether a plaintiff in a products liability action must establish the specific cause of the defect among alternative possible causes. In that case, testimony showed that the break in a ball-joint assembly may have been due to a defect in design, material, assembly, or a combination of design and material. The Court found that greater specificity in identifying the defect was not required for two reasons: 1) the purpose of tort law is to compensate injured persons (not merely to deter the manufacture of a specific product defect, a goal which would place a premium on defect identification), and 2) plaintiff was not at fault in her inability to produce more specific evidence of the defect and had not been deficient in her duty to prove the particular defect " 'by the most accurate evidence reasonably available’ ”. The Court recognized that the situation was "analogous” to the theory of alternative liability expressed in Summers v Tice. 403 Mich 625, fn 15. Finally, the same equitable adjustment of burden of proof appears in a line of apportionment of damages cases. In Maddux v Donaldson, 362 Mich 425; 108 NW2d 33 (1961), the plaintiff was involved in a multiple collision of automobiles, suffering two separate collisions, each of which contributed to her injuries. Initially, the trial court dismissed the case because the plaintiff was unable to present any evidence that the damages could be apportioned. This Court disagreed, recognizing the injustice of the plaintiff’s burden: "When we impose upon an injured plaintiff the necessity of proving which impact did which harm in a chain collision situation, what we are actually expressing is a judicial policy that it is better that a plaintiff, injured *329 through no fault of his own, take nothing, than that a tort-feasor pay more than his theoretical share of the damages accruing out of a confused situation which his wrong has helped to create. The mere statement of the policy exposes its aberrations. It is at war with at least the last hundred years of judicial progress.” 362 Mich 430. Therefore, rather than deny an innocent plaintiff recovery, the Court held that a plaintiff who proves an indivisible injury is relieved of the burden of apportioning damages; instead, the defendants are held jointly and severally liable. This equitable policy has frequently been reaffirmed. See, e.g., Brownell v Brown, 407 Mich 128; 283 NW2d 502 (1979); Michie v Great Lakes Steel Division, 495 F2d 213 (CA 6, 1974), cert den 419 US 997 (1974); Oakwood Homeowners Ass'n, Inc v Ford Motor Co, 77 Mich App 197; 258 NW2d 475 (1977). Clearly then, the policy underlying the theory of alternative liability has been adopted in this state. 1 We now formally express our approval of the] theory of alternative liability. We observe, however, that the reasoning of Summers, the polestar case for alternative liability, does not define a theory of recovery that provides a neatly fitting analytical template for application to this case. The situation in Summers is substantially and significantly distinguishable from this DES litigation, although it is not sufficiently so to make the theory of alternative liability entirely inapplicable. The requirements of the alternative liability theory in Summers must be modified, however, to accommodate the unique facts of this unusual litigation. Summers involved one plaintiff and two defendants. This case involves 180 plaintiffs, some of *330 whom were injured directly and some derivatively, and at least 16 defendants. In Summers, all the parties who could have caused harm to the single plaintiff were before the court. Here, whether the plaintiffs have brought all defendants before the court is a contested issue. The plaintiffs claim they have sued all known manufacturers of stilbene derivatives who promoted the drugs to the medical profession in Michigan for use in pregnancy during the period between 1947 and 1964. Defendants assert, however, that there are several hundred defendants that should be before the court if our requirement of the presence of all those who could have caused the plaintiffs’ injuries is to be met. In Summers, the asserted negligence and the injury resulting therefrom, and the activity of all the parties, occurred at essentially the same time and at one place, providing so-called simultaneity. Here, the alleged tortious activity occurred over a span of almost two decades and conceivably throughout the United States. Perhaps the most fundamental, and arguably the most important, factual difference between Summers and this case is that in Summers each defendant was negligent toward the sole plaintiff; each could have caused the injury to the plaintiff although only one in fact did so. Here, the plaintiffs do not even claim that each of the defendants was negligent toward each of the plaintiffs. Therefore, each of the defendants in this case could not have caused injury to each of the plaintiffs. Stated differently, in Summers, each defendant was negligent toward the plaintiff; here, each defendant was negligent toward a plaintiff, but each defendant was not negligent toward each plaintiff. Thus, all defendants were not negligent toward each plaintiff, and each defendant could not have caused each plaintiff’s injury. *331 Although the rationale of the alternative liability theory in Summers is not squarely applicable to this DES litigation, partly because the facts of the two cases are so distinctly different, the theory as first detailed in Summers can nevertheless be tailored to accommodate the unique facts of this case, and in fairness ought to be. It should be understood, however, that in approving application of alternative liability to this case, we are not only extending the policy of traditional alternative liability as espoused in Summers to accommodate the facts; we are actually fashioning and approving a new DES-unique version of alternative liability. The requirements which the plaintiffs must meet before availing themselves of the eased burden of proof may be garnered from the cases above. First, it must be shown that all the defendants have acted tortiously, cf. Shunk v Bosworth, 334 F2d 309 (CA 6, 1964); second, that the plaintiffs have been harmed by the conduct of one of the defendants (in order to support this second requirement, the plaintiffs must bring before the court all the actors who may have caused the injury in fact); 14 third, that the plaintiffs, through *332 no fault of their own, are unable to identify which actor caused the injury. In that connection, plaintiffs must make a genuine attempt to locate and identify the tortfeasor responsible for the individual injury. The genuineness of the attempt to do so is to be measured by the traditional due diligence standard, and it is the trial court’s duty to ascertain whether a duly diligent effort on the part of a plaintiff has been made. A trial court finding of a lack of diligence will preclude plaintiffs’ resort to DES-modified alternative liability. We also restrict, for the time being, the use of this theory of recovery to those allegations sounding in negligence. Plaintiffs may not resort to the theory for warranty allegations or for strict liability allegations. 15 Plaintiffs’ ability to meet these conditions precedent will allow them to avail themselves of DES-modified alternative liability, thus relieving them of the traditional burden of proof of causation in fact. We emphasize that although plaintiffs are absolved of identifying specifically the tortfeasor who caused each specific harm, their burden of proof as to other aspects of causation is not lessened. In that connection, as necessary corollaries of shifting the burden of proving causation in fact *333 and the requirement that plaintiffs bring all potential tortfeasors before the Court, the plaintiffs are required to prove the following: 1) that all defendants distributed or manufactured one or more of the three drugs involved: DES, DSD, or dienestrol; 2) that the female plaintiffs’ mothers ingested DES, DSD, or dienestrol (it is not sufficient that plaintiffs prove merely that their mothers ingested a synthetic estrogen as there may be other synthetic estrogens prescribed for the same reason which are not at issue here. Conceivably plaintiffs could satisfy this requirement by showing that no other synthetic estrogens prescribed for the same purposes as DES, DSD, or dienestrol were prescribed for during the relevant period); 3) that the female plaintiffs’ mothers ingested DES, DSD, or dienestrol manufactured or distributed in Michigan; 16 and 4) that DES, . DSD, and dienestrol each caused the type of injury of which the plaintiffs complain; that is, plaintiffs must prove that these three drugs are essentially identical in their injury-producing results. We note in passing that defendants’ access to evidence of causation is not a relevant factor. The California court in Summers did mention that "[ojrdinarily defendants are in a far better position to offer evidence to determine which one caused the injury”. 33 Cal 2d 86. However, the requirements of better access to evidence and its counterpart, exclusive control, are more correctly facets of the doctrine of res ipsa loquitur. See Ybarra v Spangard, 25 Cal 2d 486; 154 P2d 687 (1944); *334 Anderson v Somberg, 67 NJ 291, 305; 338 A2d 1 (1975). Even as a factor within that doctrine, the defendant’s access to evidence as a controlling consideration has been criticized. See, e.g., Prosser, Torts (4th ed), § 39, p 225. The burden of proof that is shifted to each defendant is to prove by a preponderance of the evidence that it neither produced nor marketed the DES, DSD, or dienestrol ingested by female plaintiffs’ mothers. A defendant may, of course, meet that burden by proving that it did not market or produce DES, DSD, or dienestrol during the period in which plaintiffs’ mothers were exposed to the drug or by showing that its products were not available in the relevant geographical area in which plaintiffs’ mothers acquired the drug. In sum, alternative liability will be applied in cases in which all defendants have acted tortiously, but only one unidentifiable defendant caused plaintiff’s injury. If a plaintiff brings all the possible defendants into court and establishes the other elements of the underlying cause of action, the court should equitably shift an onerous burden of causation in fact to the defendants. If the defendants are unable to exonerate themselves, joint and several liability results. (C) Plaintiffs’ Allegations In light of this examination of the requirements of alternative liability, it is apparent that some plaintiffs have alleged sufficient facts to withstand summary judgment on the causation issue. Plaintiffs have alleged that all defendants acted negligently in manufacturing and promoting preparations of dienestrol, DES, and DSD, despite scientific research indicating that such drugs were likely to cause cancer. Plaintiffs also state that all *335 the manufacturers who promoted the drugs for use during pregnancy in Michigan during the time period 1947 to 1964 are before the Court. Difficulties arise, however, in that not all plaintiffs have alleged that they are unable to identify the manufacturer of the product that their mothers ingested. As of the fourteenth amended complaint, approximately 70 plaintiffs allege that they are able to identify the manufacturer of the drug which injured them. As we have seen, the raison d’etre of alternative liability is to shift an onerous proof requirement where to do otherwise would leave an innocent plaintiff remediless. Where plaintiffs are able to identify the causation in fact of their injury, traditional tort remedies must be used to secure relief. Accord, Lyons v Premo Pharmaceutical Labs, Inc, 170 NJ Super 183, 192; 406 A2d 185 (1979). Of course, pleading practices in Michigan permit the assertion of inconsistent claims. GCR 1963, 111.9. Even in this situation, where proof of one claim must defeat the existence of another, the plaintiff is allowed to present both claims. The winnowing of issues and scrutiny of claims is accomplished by discovery procedures, pretrial conferences, and summary judgment motions, not through pleading technicalities. Therefore, plaintiffs’ antithetical pleadings alone do not warrant summary judgment relief. 17 On the other hand, once plaintiffs prove the identity of the manufacturer of the DES product *336 ingested, the option of alternative liability is no longer available to them. Presently, 113 plaintiffs have alleged that they are unable to identify the manufacturer of the product which harmed them. They assert that they are blameless in their inability to meet the identification requirement. These plaintiffs explain their failure to name a specific manufacturer by citing legislation requiring pharmacists to maintain prescription drug records for only five years, MCL 338.1119; MSA 14.757(19), as amended by MCL 333.17751; MSA 14.15(17751), and the defendants’ use of a generic marketing scheme to promote the product. These pleadings constitute sufficient allegations to withstand summary judgment under GCR 1963, 117.2(1). Finally, the 70 plaintiffs who allege that they are able to identify the manufacturer of the product which harmed them may bring forth their proofs at trial. If they are unable to establish the required identification, they have the option of pursuing their alternative liability theory. II. Concert of Action (A) Allegations of Novel Theories Defendants also request summary judgment claiming that plaintiffs improperly rely on theories of liability not recognized in Michigan. Plaintiffs’ complaint alleges that defendants are "jointly and severally responsible” for the alleged injuries. Clearly, the concept of joint and several liability is not foreign to this state. See, e.g, Cuddy v Horn, 46 Mich 596; 10 NW 32 (1881); King v Herfurth, 306 Mich 444; 11 NW2d 198 (1943). Various rationales have been argued to justify its imposition in *337 the present case: concert of action, enterprise liability, as enunciated in Hall v El DuPont de Nemours & Co, 345 F Supp 353 (ED NY, 1972), market share liability, as set forth in Sindell v Abbott Laboratories, Inc, 26 Cal 3d 588; 163 Cal Rptr 132; 607 P2d 924 (1980), cert den 449 US 912 (1980), or some other modification of a collective, industry-wide liability. 18 Of these varied theories, only concert of action is recognized in Michigan. While it is not this Court’s intention to be unreceptive to developing theories in the ever-changing matrix of the law, neither should the Court adopt new theories where no need exists. Only when traditional concepts fail to meet the demands which are placed upon them must novel responses develop to fill the void and answer society’s need for equitable loss distribution. In the present case, plaintiffs have not presented this Court with any innovative legal claims to evaluate. They have not urged us to adopt "enterprise”, "industry-wide”, or "market share” liability. When the parties thus feel that traditional theories are sufficient, this Court will not assume the task of evaluating relief not requested. (B) Concert of Action Plaintiffs, however, do seek to proceed on the traditional theory of concert of action. This theory, *338 although not developed to ease plaintiffs’ traditional burden of proof of causation, 19 may have that effect. If plaintiffs can establish that all defendants acted tortiously pursuant to a common design, they will all be held liable for the entire result. The concept is perhaps most clearly illustrated in the racing context. If three drivers join in a drag race, as a result of which one pedestrian is injured, all three may be held liable. Thus a legal fiction is created: all three drivers are found to be the cause in fact, although only one driver may have actually struck the pedestrian. In order to withstand a motion for summary judgment based on a failure to state a cause of action, a plaintiff need only allege that the defendants were jointly engaged in tortious activity as a result of which the plaintiff was harmed. Walters v Sargent, 390 Mich 775; 210 NW2d 315 (1973); McCoy v DeLiefde, 376 Mich 198; 135 NW2d 916 (1965) (opinion of Souris, J.). Unlike the procedural device of alternative liability, a concert of action case does not require that the plaintiff be unable to identify the specific defendant who caused his injury in fact. The plaintiff is usually able to make that determination. See, eg., Gaufin v Valind, 268 Mich 269; 256 NW 335 (1934); Mahnke v Freer, 126 Mich 572; 85 NW 1099 (1901); King v Herfurth, 306 Mich 444; 11 NW2d 198 (1943). 20 Such identification does not preclude liability on a concert of action theory. Here plaintiffs have alleged that defendants *339 acted together in negligently manufacturing and promoting drugs which were ineffective and dangerous, were inadequately tested, and were distributed without sufficient warnings. These allegations are sufficient to withstand summary judgment under GCR 1963, 117.2(1). Those plaintiffs who have alleged that a specific manufacturer caused their injuries are not precluded from this cause of action. Conclusion Plaintiffs have made sufficient allegations to support their concert of action claim. Additionally, those plaintiffs who, on the standard we have enunciated, demonstrate inability to identify the manufacturer of the product which harmed them and who sustain the remaining burdens of proof as we have defined them, may take advantage of the burden-shifting feature of the alternative liability theory to withstand summary judgment on the causation issue of the negligence claims. The number and posture of the parties and the novelty and complexity of the issues incidental to the theory of recovery we approve today suggest a major challenge to the trial management skills of the trial judge and the advocacy skills of trial counsel. In addition, unanticipated jurisprudential procedural and substantive issues will inevitably arise. We do not presume the prescience to anticipate all of them at this remove, let alone address and resolve them. We are deciding the issues before us today in a virtual factual vacuum. The fairness of the application at trial of the theory of alternative liability remains to be seen. A factual record may reveal a number of unanticipated inequities affecting any of the parties in trying to a *340 verdict the new cause of action we approve today. One defendant asserts, for example, that it sold approximately $75 worth of DES in Michigan during 1953. Whether that can be proved remains to be seen. If it is, difficult questions of allocation of damages may be presented. We are not unmindful of the possibility of such difficulties, but neither do we attempt to anticipate and reconcile them in advance of trial. Thus, we explicitly reserve judgment concerning the validity of any verdict that may result from a trial of the cause of action we have approved. The judgment of the Court of Appeals, as modified, is affirmed and the case is remanded to the Third Judicial Circuit Court for proceedings consistent with this opinion. Kavanagh, Levin, Ryan, Brickley, and Cavanagh, JJ., concurred with Williams, C.J. Boyle, J., took no part in the decision of this case. Footnotes [1] An individual drug product acquires various names. First, there is the chemical name which attempts to spell out the structural makeup of the drug. Next comes the generic name which may or may not represent an abbreviation of the more complex chemical name. Ordinarily a drug has one generic name, but there are cases where two or three are employed. Finally a drug may acquire multiple trade names used by the various companies engaged in the promotion of the product. In this case, diethylstilbestrol (or DES), dienestrol, and diethylstilbestrol dipropionate (DSD) are the generic names of the products. Dierone, for example, was a Kremers-Urban trade name for dienestrol. Generic drugs are usually marketed at lower prices because their manufacturers avoid the research and development costs normally associated with the introduction of an original product. Comment, Products Liability for Prescription Drugs, 23 Syracuse L Rev 887 (1972); United States v Generix Drug Corp, 460 US 453; 103 S Ct 1298; 75 L Ed 2d 198 (1983). According to defendants’ responses to interrogatories, several defendants, for example, Cole Pharmacal Co., Rexall Drug Co., Korer, Inc., Tutag & Co., and Vale Chemical Co., used no trade name at all. Plaintiffs’ First Set of Interrogatories to Defendants, Question 2. [2] The link between DES and cancer was discovered through the appearance of a rare form of cancer. Researchers took detailed histories of the women who developed the cancer, including information on maternal ingestion of drugs. From this data, the statistical association was derived. Comment, DES and a Proposed Theory of Enterprise Liability, 46 Fordham L Rev 963, 964, fn 5. Since the cancer and cellular abnormalities did not develop in the users of the estrogen products, but rather in their offspring, a significant time delay occurred between the ingestion of the drug and the injury which it allegedly caused. [3] Initially, 16 women and two spouses were named in the complaint. Amendments adding plaintiffs were allowed until May, 1976. The complaint now includes 184 plaintiffs. (Only 182 of these plaintiffs have chosen to appeal.) [4] Originally, 30 corporations were named as defendants. Several summary judgments were granted dismissing individual defendants. The complaint now names 16 companies as defendants. [5] Plaintiffs have apparently abandoned the conspiracy count. [6] Plaintiffs’ initial pleading also included a complaint for mandamus requesting that defendants affirmatively be required to notify women who may be afflicted by DES-related injuries so that they may join as plaintiffs, to establish a comprehensive screening program to identify women who may be injured, and to provide treatment for these women. This part of the complaint was dismissed without prejudice. [7] Pleading requirements are set forth in GCR 1963, 111. [8] This identification requirement is, of course, a facet of the factual causation element of tort law. See, generally, Prosser, Torts (4th ed), § 41, pp 237, 239. The function of the causation element is to limit the scope of potential liability. If an actor were held responsible for all possible harms which might flow from his acts, he could be deterred from performing useful and desirable activity. By adjusting the causation requirement, the court is able to strike a balance between deterring harmful behavior and encouraging useful activity. Causation requirements also reflect common notions of moral responsibility or blame. Hart & Honoré, Causation in the Law (Oxford: Clarendon Press, 1959), p 59; Malone, Ruminations on Cause-In-Fact, 9 Stan L Rev 60, 66 (1956); Becht & Muller, The Test of Factual *325 Causation in Negligence and Strict Liability Cases (St Louis: Washington Univ, 1961). [9] Prosser, Torts (2d ed), § 41, p 243. [10] The Summers court recognized that the same result had been reached in an earlier case through a contorted use of the concept of concert of action. 33 Cal 2d 84. [11] Prosser, fn 8 supra, § 41, p 241, defines the burden of proof as follows: "[A plaintiff] must introduce evidence which affords a reasonable basis for the conclusion that it is more likely than not that the conduct of the defendant was a substantial factor in bringing about the result. A mere possibility of such causation is not enough; and when the matter remains one of pure speculation or conjecture, or the probabilities are at best evenly balanced, it becomes the duty of the court to direct a verdict for the defendant.” See also Jordan v Whiting Corp, 396 Mich 145, 151; 240 NW2d 468 (1976). [12] The court also reassigned the burden of proof regarding apportionment of damages. 33 Cal 2d 88. See 2 Restatement Torts, 2d, § 433B(2), p 441. Although the two issues were carefully distinguished by the Summers court, other courts have subsequently confused them. See, e.g., Shunk v Bosworth, 334 F2d 309, 312 (CA 6, 1964) ("In the Summers case, the real difficulty was in apportioning the damages between the negligent defendants”). [13] Section 433B(3) provides: "(3) Where the conduct of two or more actors is tortious, and it is proved that harm has been caused to the plaintiff by only one of them, but there is uncertainty as to which one has caused it, the burden is upon each such actor to prove that he has not caused the harm.” [14] Failure to meet this requirement has resulted in summary judgments in the majority of DES cases which attempted to utilize this theory. See, e.g., Ryan v Eli Lilly & Co, 514 F Supp 1004, 1016-1017 (D SC, 1981); Namm v Charles E Frosst & Co, 178 NJ Super 19, 32-33; 427 A2d 1121 (1981); Sindell v Abbott Laboratories, 26 Cal 3d 588, 602-603; 163 Cal Rptr 132; 607 P2d 924 (1980); Gray v United States, 445 F Supp 337, 338 (SD Tex, 1978). As explained by the Sindell court, p 603: "[T]he possibility that any of the five defendants [chosen from among a possible 200 companies] supplied the DES to plaintiffs mother is so remote that it would be unfair to require each defendant to exonerate itself. Thére may be a substantial likelihood that none of the five defendants joined in the action made the DES which caused the injury, and that the offending producer not named would escape liability altogether.” We also note that Comment A to 2 Restatement Torts, 2d, § 433B(3), p 446 suggests possible modification of the requirement that all defendants must be joined. *332 “It is possible that cases may arise in which some modification of the rule stated may be necessary because of complications arising from the fact that one of the actors involved is not or cannot be joined as a defendant, or because of the effect of lapse of time, or because of substantial differences in the character of the conduct of the actors or the risks which they have created.” On the basis of the pleadings of this case, it is not necessary to consider the appropriateness of modifying this requirement. [15] We read the briefs and arguments of the plaintiffs as asserting the theory of alternative liability in respect to the negligence but not the other theories of recovery, and accordingly we reserve the question whether the alternative liability theory is applicable on another theory of recovery. [16] The complaint alleges that "all of the known manufacturers of stilbene derivatives who promoted, in Michigan, said drugs to the medical profession for use in pregnancy” during the relevant period are before the Court as defendants. [17] On this point, we must disagree with the Court of Appeals opinion which seems to indicate that plaintiffs must opt for one theory or another, but may not allege both. The Court of Appeals stated: "Those [plaintiffs] who have already amended their complaint to allege a sole tortfeasor may either allege joint and alternative liability or stand on the most recent amendment.” 94 Mich App 77, fn 6. [18] Throughout the country, commentators are wrestling with the problems presented in the DES litigation. Their analyses are thoughtful and helpful. See, e.g., Comment, DES and a Proposed Theory of Enterprise Liability, 46 Fordham L Rev 963; Comment, Market Share Liability: An Answer to the DES Causation Problem, 94 Harv L Rev 668 (1981); Coggins, Industry-wide Liability, 13 Suffolk L Rev 980 (1979); Fischer, Products Liability — An Analysis of Market Share Liability, 34 Vand L Rev 1623 (1981); Comment, DES:,Alternative Theories of Liability, 59 J Urban L 387 (1982); Endress & Sozio, Market Share Liability: A One Theory Approach Beyond DES, 1983 Det C L Rev 1; Comment, Market Share Liability for Defective Products: An Ill-Advised Remedy for the Problem of Identiñcation, 76 Northwestern U L Rev 300 (1981). [19] The theory seems to have developed to deter hazardous group behavior, rather than in response to plaintiffs causation burden. See Comment, DES and a Proposed Theory of Enterprise Liability, 46 Fordham L Rev 963, 979; Lyons v Premo Pharmaceutical Labs, Inc, 170 NJ Super 183, 192; 406 A2d 185 (1979). [20] But cf. Fisher v Rumler, 239 Mich 224; 214 NW 310 (1927); Walters v Sargent, 390 Mich 775; 210 NW2d 315 (1973).