Causation has always been the crucial issue in toxic tort litigation. If plaintiffs cannot prove causation they lose. Observers agree that it has become more difficult for plaintiffs to discharge this burden in the decade since the Supreme Court decided Daubert v. Merrell Dow Pharmaceuticals, Inc. and began its exegesis on expert proof. The trial judge is now the gatekeeper who rules on defendant's in limine motion to exclude plaintiff's experts; if defendant succeeds, the court then grants summary judgment.

Little attention has been paid to how Daubert has affected the objectives of tort law or procedural justice. Do plaintiffs' difficulties in proving causation undermine the goals of deterrence and compensation? A manufacturer putting out a new drug or chemical cannot know if it is completely safe even if it has been tested. It is impossible to rule out latent effects, or to anticipate how the substance will interract with other products, or other populations. Unlike the ordinary tortfeasor who knows, or ought to know, about conduct that will cause harm, the manufacturer of a drug or chemical may be unaware of any increased risk of harm until it starts to receive adverse reports about its product.

It is at this point that the realities of corporate life intersect with the Daubert test. Any problems with a product may affect the price of the corporation's stock, and most executives' pay is now tied to stock options. Whether executives delude themselves into thinking nothing can be the matter, or whether they deliberately ignore warning signals is beside the point. The difficulty plaintiffs will have in proving causation means that potential liability may appear quite minimal when adverse results about a product first surface, particularly as the courts interpreting Daubert have favored epidemiologic proof of causation. Such studies take a long time. Even if plaintiffs ultimately succeed in establishing causation, this may not happen until after the executives on whose watch the problems first appeared have cashed in their options and moved on. We need to consider whether the current approach to causation offers too great an incentive for corporations to adopt an ostrich-like approach that fails to protect the public adequately against harm. Perhaps new causes of action are needed, such as the Second Circuit recently endorsed in Desiano v. Warner-Lambert Co. when it permitted a lawsuit by insurers to proceed against a drug company whose failure to disclose problems may have kept the drug's price artificially high. Such a cause of action does not, however, provide compensation for the users.

Compensation may also be problematic because the greatly increased cost of expert testimony since Daubert means that persons with valid claims, but lesser damages, may be unable to find representation. We should consider the effect of Daubert jurisprudence on access to the courts. Finally, the outcome-determinative effect of in limine motions excluding plaintiff's experts on causation eliminates a trial at which plaintiffs may present evidence of defendant's breach of duty. The consequence is ignorance about irresponsible corporate behavior that is highly relevant in deciding how the law needs to be structured to provide adequate protection to the public at a time when we as yet have imperfect knowledge about the mechanisms of disease.