Resolving the Problem of Identifying DES Manufacturers


In 1947, diethylstilbestrol (DES), a synthetic estrogen was approved by the Food and Drug Administration (FDA) for use as a miscarriage preventative to be marketed on an experimental basis. By 1952, when the need for FDA approval was dropped because DES was no longer considered a new drug, 123 drug companies had obtained FDA approval. The ultimate number of companies involved in the market, however, has been estimated to be as high as 300. The FDA banned the use of DES as a miscarriage preventative in 1971 following a finding that the daughters of women who used DES during pregnancy were more likely to contract vaginal or uterine cancer than the daughters of non-users.

Emerging Theories of Proof in Products Liability: Resolving the Problem of Identifying Des Manufacturers, Villanova University School of Law, Volume 26 | Issue 5 Article 2, 1981.

Numerous suits have been filed against DES manufacturers by these women with cancerous and pre-cancerous conditions. However, given the fact that the adverse effects of DES cannot be detected until after a latency period of ten to twelve years, victims are often unable to identify the manufacturer whose product was sold to their mothers. While this inability has caused the dismissal of many of the DES suits, two recent cases have permitted the plaintiff to proceed to trial using novel liability theories.

This comment will first review three rubrics which relieve negligence plaintiffs from identifying the specific tortfeasor – concert of action, alternative liability and enterprise liability. Next, the comment will examine three recent cases in which courts applied these rubrics to DES actions. Finally, the comment will analyze the relative merits of the approaches taken by the three courts.


In a cause of action for negligence, the plaintiff must prove that -the injury was caused by the defendant’s breach of duty. In a product-liability negligence action, this generally entails identifying an individual defendant as the manufacturer or seller of the injury-causing item. However, three theories have been recognized for excepting the plaintiff from this general rule: concert of action, alternative liability and enterprise liability.

A. Concert of Action

The concert of action doctrine developed to deter hazardous group conduct, and is based upon the premise that injury results from the dangerous situation created by the negligent conduct of all the defendants. Once concert has been established, each defendant is held jointly and :severally liable for the entire harm. Most commonly, the theory is applied to cases involving racing on public highways to hold all participants jointly and severally liable when a third party is struck by one of the racers. However, it has also been applied to cases involving pollution and shooting accidents.

The Restatement of Torts has defined concert as follows:

§ 876. Persons Acting in Concert For harm resulting to a third party from the tortious conduct of another, one is subject to liability if he

  • (a) does a tortious act in concert with the other or pursuant to a common design with him,
  • or (b) knows that the other’s conduct constitutes a breach of duty and gives substantial assistance or encouragement to the other so to conduct himself,
  • or (c) gives substantial assistance to the other in accomplishing a tortious result and his own conduct, separately considered, constitutes a breach of duty to the third person.

Mere knowledge of what another party is doing will not create concert since there is no affirmative duty to interfere with another per son’s activities. Furthermore, a defendant must be proceeding tortiously or negligently to be held liable under concert.

B. Alternative Liability

The alternative liability theory developed to relieve the plaintiff of the burden of establishing which of several defendants caused an injury. The basis for the theory was developed in Ybarra v. Spangard where the plaintiff awoke after an appendectomy partially paralyzed. Not knowing the person or the instrumentality which had caused his injury, the plaintiff sued six doctors and nurses, each of whom had been responsible for him at some point during his unconsciousness. Although the defendants claimed that res ipsa loquitur was inapplicable because of the number of defendants, the court applied the doctrine, stating: “Plaintiff was rendered unconscious for the purpose of undergoing surgical treatment by the defendants: it is manifestly unreasonable for them to insist that he identify any one of them as the person who did the alleged negligent act.”

The requirements of the alternative liability theory were set forth in Summers v. Tice. The plaintiff in Summers was injured when the two defendants independently fired in plaintiff’s direction while all three were hunting. Since the evidence adduced at trial was insufficient to determine which shot had actually caused the injury, the California Supreme Court held that, whereas the negligence of one of them had injured the plaintiff, “a requirement that the burden of proof on [causation] be shifted to defendants becomes manifest.”

Under the Summers approach, the plaintiff must establish an injury, a breach of duty (negligence) on the part of each defendant, and that one of the defendants caused the injury; which amounts to a primafacie case of negligence against the defendants as a group. Following this showing by the plaintiff, the burden of proof shifts to the defendants, all of whom are held liable if they cannot prove which one of them was responsible for causing the injury. Any particular defendant is released from liability upon a showing that it could not have been the cause of plaintiffs’ injury.

Many courts in other jurisdictions have adopted the Summers rationale; it has even been extended to situations where not all defendants were proved negligent. In a New Jersey case, Anderson v. Somberg, alternative liability was applied against a physician, a hospital, a medical supply distributor, and a manufacturer of surgical instruments, after the plaintiff was injured when an instrument broke in his spinal canal during an operation. The Anderson court, relying upon Summers, not only shifted the burden of proof on the causation issue to the defendants, but also stated that the jury must find at least one defendant liable.

C. Enterprise Liability

Enterprise liability provides a third theory of recovery for a plaintiff who is unable to identify the defendant responsible for her injuries. The purpose of the theory is to alleviate the necessity of proving which of many defendants was the cause of injury, and the effect is to hold all defendants jointly liable. The theory originated in Hall v. E.I. DuPont de Nemours & Co.,  where several children were injured by blasting caps in a series of separate accidents. In separateactions which were consolidated for trial, representatives of the injured children brought suit against six blasting cap manufacturers  who comprised virtually the entire blasting cap industry of the United States. The plaintiffs’ contended that the defendants had created an unreasonable risk of injury by failing to place warnings on the caps and by failing to take other safety precautions. The plaintiffs’ maintained that the industry practice of delegating safety functions to a trade association, which collected statistics on accidents and had considered and rejected the idea of labelling individual caps, demonstrated conscious agreement among the defendants sufficient to impose jointliability.

The Hall court, citing as precedent the principles inherent in work. men’s compensation statutes and the doctrine of respondeat superior, where employers are held vicariously liable because they are “the most strategically placed participants in a risk-creating process .. “… held that the entire blasting cap industry could be held jointly liable. Under this novel enterprise liability theory, plaintiffs would have to prove defendants’ joint awareness of the risks and their joint capacity to reduce or affect those risks, and the plaintiffs would then be relieved of the burden of identifying the manufacturer of each individual blasting cap. The entire industry would thus be held liable for harm caused by its operations.


Recent cases have suggested three approaches to resolving the identification problem, drawn from the three theories previously discussed.

A. Sindell v. Abbott Laboratories

Plaintiff Judith Sindell brought a class action against ten drug manufacturers, alleging that the manufacturers were jointly liable for her injuries which resulted from her mother’s ingestion of DES. While Sindell was unable to identify the manufacturer of the drug taken by her mother and thus responsible for her injuries, she alleged that the defendants were jointly liable because of their express and implied agreements to collaborate in, rely upon, and adopt each other’s testing and marketing methods.

The trial court dismissed the action on the ground that no cause of action was stated since there was no identification of the defendant whose DES had harmed the plaintiff. The California Court of Appeals reinstated the action, finding that the complaint stated a cause of action under the concert of action and alternative liability theories. On review, the Supreme Court of California disagreed with the lower court’s determination regarding the applicability of the concert of action and alternative liability theories, and also concluded that the enterprise liability theory was inapplicable. Nevertheless, the court held that the plaintiff’s right of action was not precluded by her inability to identify the specific manufacturer of the DES which her mother had ingested. Rather, the court found that she had stated a cause of action under a new theory – market-share liability.

The Sindell court, in considering the alternative liability theory, accepted the defendants’ argument that, since only five of the two hundred companies which manufactured DES had been joined as defendants, there was no rational basis on which to infer that any named defendant had caused the plaintiff’s injuries. The court, noting that in Summers all responsible parties had been joined, concluded that the Summers rule of alternative liability was not applicable to a situation where there was a possibility that none of the joined defendants had caused the injury.

The California court next considered and rejected use of the concert of action theory, stating that the defendants’ alleged failure to test the drug adequately or to give sufficient warning of its dangers did not constitute concert of action. Moreover, the court rejected the plaintiff’s assertion that, since the defendants had engaged in parallel or imitative practices and had relied on each others’ promotional and marketing techniques, the defendants had acted in concert. Instead, the court described these actions as a common industry practice and stated that to apply the concert of action theory to the case at bar would so expand the doctrine as to “render virtually any manufacturer liable for the defective products of an entire industry.”

The Sindell court declined to apply the enterprise liability theory for three reasons. First, there were at least two hundred manufacturers of DES, and the Hall court had cautioned against use of the theory in a case involving a large number of producers. Second, the plaintiff could not show that the defendants had jointly controlled the risk. In Hall, joint control of the risk had been based upon the fact that responsibility for the formulation of safety standards had been delegated to a trade association; here, no such trade association of the DES manufacturers existed. Moreover, the plaintiff had failed to meet the standard for joint control of the risk under a concert of action theory. Finally, the court acknowledged the federal government’s role in setting standards for the testing and manufacture of drugs, and stated that it would be unfair to impose liability simply for following government-imposed standards.

Having discarded all of the extant theories which would have allowed the plaintiff to recover, but persuaded by strong policy arguments that the plaintiff ought to be compensated, the Sindell court developed the theory of market-share liability. The supreme court shifted the burden of proof of causation to the defendants, upon the plaintiff’s showing that those manufacturers whose sales of DES accounted for a “substantial share” of the market were joined as defendants. Failing to prove that its drug was not the one which had injured each individual plaintiff, each defendant manufacturer would be held liable for the proportion of the judgment represented by its share of the DES market. Anticipating that some discrepancy between market share and liability will occur under the new doctrine, the California court reasoned that the theory would treat each defendant fairly because “each manufacturer’s liability would approximate its responsibility for the injuries caused by its own products.” Thus, the court stated that the difficulty in apportioning damages and in defining the relevant market should not militate against adoption of the market-share liability theory.

In a dissenting opinion, Justice Richardson opined that the majority’s holding allowed the plaintiff to recover damages without proof that any defendant had either in fact caused or probably caused the plaintiff’s injuries. Moreover, Justice Richardson observed that market-share liability would fall disproportionately on those manufacturers amenable to suit in California, would treat DES-plaintiffs more favorably than other tort plaintiffs, and would be harmful from a social policy viewpoint.

B. Ferrigno v. Eli Lilly and Co.

In an action by eight daughters of DES users against twenty-two drug manufacturers, the New Jersey Superior Court stated that the identification problem posed two separate issues: first, the inability to identify the precise causative agent, and second, the possibility that the causative agent was not among the defendants before the. court. While it has been subsequently overruled, the Ferrigno, decision shows an interesting approach taken by one court to resolving the DES identification dilemma. The court viewed Anderson v. Somberg, as resolving the first issue. The court read that case as holding that, where a plaintiff, through no fault of her own, is unable to identify which of several defendants harmed her, it becomes the defendants’ burden for each to establish that it was not the cause of harm.

In addressing what it considered the more difficult second issue, the Ferrigno v. Eli Lilly and Co. court, again relying upon Anderson, decided that the burden of proof could be shifted to the defendants, even in the absence of joinder of all potential defendants. While the court interpreted the plurality opinion in the Anderson case as assuming that all potential defendants were before the court, the Ferrigno court concluded that even the dissenters in the Anderson case would uphold the shifting of the burden of proof.”

The Ferrigno court further concluded that the Anderson rationale of alternative liability should be extended beyond the particular facts of that case and applied to the case at bar. The court reasoned that, in both Ferrigno and Anderson, the plaintiffs had been innocent victims of an unforeseeable injury.” Also, in both cases, all of the defendants were potentially at fault and all of the defendants owed a special duty to the plaintiff.” The court considered the Ferrigno defendants to be more culpable than those in Anderson, however, in that they caused the identification problem by making a fungible good with long-delayed effects. And finally, since all of the defendants in Ferrigno were alleged tortfeasors, whereas it was presumed in Anderson that only one defendant was at fault, the shift in the burden of proof was thought to be more compelling.

Acknowledging the Sindell court’s rejection of the alternative liability theory, the Ferrigno court nevertheless was satisfied that, under New Jersey law, the theory could and should be applied. Having found the alternative liability theory a satisfactory method for providing recovery to the plaintiff, the New Jersey court declined to consider either the enterprise liability or the market-share liability theories and also rejected the concert of action theory on these facts. The court stated that any defendant could exculpate itself by identifying the actual manufacturer, by showing that it did not manufacture DES until after plaintiff was born, by showing that its drug could not have reached the location where the plaintiff-mother purchased it, or by showing that its drug was not the same shape, size, or color as that ingested by the mother.

However, the court decided that, for the purposes of allocating damages among the defendants, it would adopt the Sindell rationale and hold each defendant liable for the proportion of the judgment represented by its share of the market.

C. Abel v. Eli Lilly & Co.

Reviewing the grant of summary judgments in favor of defendants, the Michigan Court of Appeals, having noted that the concert of action theory was “well-established,”  found that the plaintiff had stated a cause of action under that theory  and held that the summary judgments which had been granted by the trial court were improper.

Moreover, in discussing the theory of alternative liability, the Abel v. Eli Lilly & Co. court noted that, although there were no Michigan cases directly on point, the burden of proof on causation had been shifted to the defendants in other contexts under Michigan law. In adopting the alternative liability theory, the Michigan court cautioned that each plaintiff must prove by a preponderance of the evidence that each defendant had breached its duty of care, and that the harm which she suffered had been caused by her mother’s ingestion of DES which had been manufactured by one of the named defendants. According to the Abel court, the defendants would then be required to apportion the damages among themselves.


The reasoning of the Abel court is well-supported by precedent. By slightly extending Michigan law with the adoption of the alternative liability theory, the Abel decision merely brings Michigan law into line with the law of many other jurisdictions. In contrast, the Ferrigno court based its decision to expand the scope of the alternative liability theory upon the New Jersey Supreme Court’s decision in Anderson. In particular, the Ferrigno court relied upon the Anderson court’s assumption that all of the defendants were before the court. However, where the Anderson court made this assumption because of the particular factual situation of the case, one in which the existence of other defendants was merely speculative, in DES cases it is more than likely that not all of the defendants can be joined. Moreover, it is doubtful whether the Anderson court in tended to allow a presumption of joinder of all defendants in any case where alternative liability is applied.

Unlike the New Jersey court, which molded the Anderson rationale to fit the facts of a DES case, the California Supreme Court in Sindell clearly acknowledged that it was taking a novel step beyond the Summers holding by extending alternative liability to a situation where not all of the defendants were joined, and in using market-share data to apportion damages.  This step was justified by strong public policy considerations which support the recovery of an innocent plaintiff over the protection of allegedly negligent defendants.

Although the alternative liability theory was the common device used by the Sindell court as well as the Abel 144 and Ferrigno courts to provide the plaintiff with a cause of action, the courts differed in their resolution of the problem caused by the inability to join all of the potential defendants. These contrasts among the courts can be seen, for example, in the language used by the Abel court indicating that it would require joinder of all of the potential defendants. The Sindell and Ferrigno courts recognized that this was a virtual impossibility. However, the Sindell court required that those manufacturers whose sales of DES accounted for a “substantial share” of the market be joined, while the Ferrigno court imposed no floor on the number of manufacturers who must be joined.

While the requirements regarding the joinder of defendants imposed by the Sindell and Ferrigno courts go far toward resolving the DES plaintiff’s dilemma of being unable to identify the manufacturer of the drug ingested by her mother, it is submitted that the Ferrigno result is inferior to the result reached in Sindell. This is because the joinder of only a small number of manufacturers would destroy the underlying presumption of the alternative liability theory, i.e., that with all manufacturers joined there is a high probability of causation collectively. In the interests of fairness to the defendants and the preservation of traditional tort principles, the courts should encourage joinder of the maximum number of defendants.

Addressing the issue of the apportionment of damages among defendants, the Abel court stated only that the defendants would be required to apportion damages among themselves, while the Sindell and Ferrigno courts stated that they would hold each defendant liable for damages in proportion to its market share. However, an unresolved issue in both Sindell and Ferrigno is whether the damages awarded by the jury will be paid in full by the defendants joined, or whether only that proportion of the damages will be paid which corresponds to those defendants’ percentage-share of the market. It is suggested that the latter result would better achieve the courts’ policy of balancing the interests of plaintiffs and defendants and achieving a result which is fair to each. This method of allocation would allow the Sindell court more flexibility in defining what constitutes a “substantial share” of the market, since the court would not have to be concerned that by setting too low a figure it would put an undue burden on the joined defendants. Moreover, this would protect the Ferrigno defendants against the imposition of an unfair burden which would result from the joinder of only a small number of manufacturers.

A question left unanswered in each of these three cases is whether a time-lag between the injury and its manifestation is necessary for the application of the doctrines developed by each court. It is submitted that the answer to this should be in the negative since, where the responsible manufacturer cannot be identified, the presence or absence of lapsed time in a case in no way changes the equities between the parties.


Unlike the Sindell and Ferrigno courts, the Abel court did not break new ground; and while the court in Abel did express concern for fairness to the plaintiffs, the decision provides little hope for future DES plaintiffs in light of the fact that the standards of proof set by the court are, it is submitted, simply too stringent for plaintiffs to meet.

In contrast, the Sindell and Ferrigno decisions provide great hope for recovery by DES plaintiffs. Moreover, the same policy considerations which led the California and New Jersey courts to their respective holdings should also extend the opportunity for recovery to plaintiffs injured in other circumstances, such as by pollution or industrial waste, where particular tortfeasors are seemingly untraceable. Finally, it is hoped that the Sindell and Ferrigno decisions will force manufacturers to show more concern for product safety, or, at the very least, provide drug manufacturers with an incentive to differentiate their products and, thus, eliminate the identification dilemma faced by DES plaintiffs.

Linda Mogul Madwayr, 1981.

Click to download the full paper.

More DES DiEthylStilbestrol Resources

Have your say! Share your views