Rather than adding to the thousands of pages that have already been written about the gatekeeping standard set out by the Supreme Court in Daubert v. Merrell Dow Pharmaceuticals, Inc.,¹ and its subsequent two opinions, General Electric v. Joiner² and Kumho Tire Co. v. Carmichael,³ I want to discuss the Judicial Oversight of Experts in the Courtroom from a vantage point other than that of evidence law. I'd like to seize the opportunity provided by an audience of tort scholars to discuss whether the courts' treatment of causation in toxic tort cases furthers the objectives of tort law. And because this is a joint session with the Civil Procedure section I also want to explore the impact of the Court's evidentiary rulings on the goals and values that our system of procedure seeks to ensure.

At times it appears as though the Supreme Court's opinions - all of which dealt with proof of causation - transformed the field of toxic torts into a subdivision of evidence law. It is in toxic tort cases, the subject of both Daubert and Joiner, that plaintiff's burden to prove causation is most crucial. Plaintiff is unlikely to have any direct explantion of the causal process that allegedly led to the adverse health effects for which plaintiff is suing.⁴ Furthermore, we currently lack a full understanding about the complex interactions between genetic and environmental factors that are thought to produce illnesses and birth defects. Plaintiff is therefore forced to rely on expert proof based on statistics and reasoning by analogy in seeking to prove causation.

Although much is as yet unknown about the impact of Daubert and its progeny,⁵ observers are agreed that in the decade since the Daubert opinion, it has become considerably more difficult for plaintiffs to convince a federal district judge that their experts should be allowed to testify.⁶ The trial judge is now the "gatekeeper" who must screen the experts' testimony to determine whether it is admissible. Numerous judicial opinions have refused to allow plaintiffs' experts to offer an opinion on causation on the ground that their opinions are not sufficiently reliable to satisfy the Daubert test. The result of an exclusion is likely to be outcome-determinative: once plaintiffs' experts on causation are excluded, the trial court will grant defendant's motion for summary judgment on the ground that plaintiffs have failed to produce evidence on an essential element of their case.

Markedly missing from almost all discussions of Daubert has been any consideration of how the courts' heightened requirements with regard to the admissibility of expert proof mesh with the tort law's goal of deterrence. The law of evidence does not concern itself with how risk ought to be allocated,⁷ and tort law has paid little attention to whether the outcome-determinative impact of Daubert has deprived the public of needed protections.

Neither evidence law nor tort law have focused on the consequences that ensue because a manufacturer putting a new drug or chemical on the market cannot know whether it is safe. Even if it has been tested, and many chemicals have not been investigated at all, it is not possible to rule out latent effects that may not surface until many years have passed, to examine the product's effect on all populations, or to consider the drug's interaction with other substances. Unlike the ordinary tortfeasor who knows, or ought to know, what conduct 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. At this point in time, as has become particularly evident post-Enron, the chief compensation of corporate executives consist of stock options. Obviously, intimations about problems with a corporation's product may lead to a decrease in the price of its stock. Whether corporate executives delude themselves into thinking that nothing is the matter, or whether they deliberately conceal troubling information that may affect their financial renumeration is beside the point. The result is the same: corporate silence, and an unwillingness to undertake needed research to determine if the product is capable of causing the harms about which complaints are surfacing. Taking an ostrich-like approach to intimations of harm means that not only may stock prices remain high, but also that little useful information about the product will be available to plaintiffs if litigation starts and discovery begins.

Of course, the corporation is aware that there are cases in which a failure to warn may result in bankruptcy or punitive damages. But the difficulty plaintiffs will have in proving causation means that potential liability may appear quite minimal when adverse reports about the product first surface. In interpreting Daubert, the courts have shown a marked preference for epidemiologic studies over other proof of causation. Such studies take considerable time. The corporate executives know that there will be a considerable time-lag before causation can be established. Consequently, any loss to the corporation is unlikely to occur on the watch of those individuals who put the product on the market, and/or failed to respond when complaints were first received. They may well have cashed their stock options and moved on. Another fact of American corporate life is the mobility of corporate executives. The harder it is for plaintiffs to prove causation the less incentive there is for corporate executives to deal candidly with information about current risks, and to undertake research needed to clarify the likelihood of future harm.

Desiano v. Warner-Lambert Co. ,⁸ a case involving Rezulin, the diabetes drug, may test the proposition that plaintiffs' difficulties in proving causation play a role in inducing corporations to ignore the risks their products pose. Desiano is not a toxic tort case. The plaintiffs in Desiano are health insurers who sought to bring a class action on behalf of all health benefit providers that paid $1.4 billion for the diabetes drug Rezulin between February 1997 and April 2001. They claim that defendant marketed a defective product, that it failed to disclose information relevant to the safety of the product, and that consequently the price of the drug was far higher than it would have been absent these alleged torts. The trial court dismissed the complaint for failure to state a claim. The Second Circuit reversed.

Relevant to the discussion about the impact of Daubert are the facts stated in the opinion about what Warner-Lambert knew at the time it was claiming that Rezulin was a drug with "Side Effects Comparable to Placebo."⁹ Of course these facts must be accepted as true for the purposes of a motion to dismiss but defendant will have an opportunity to contest them. But it certainly appears that defendant continued to claim its drug was safe even though it was aware that the risk of severe liver damage associated with Rezulin was rising, and that there had been a number of fatalities in persons taking the drug.

The trial judge who dismissed the Desiano complaint is also handling multi-districted toxic tort suits by plaintiffs seeking damages after using Rezulin. A motion to certify a class in those cases was denied, so plaintiffs will have to prove causation in order to recover. In Desiano, however, the Second Circuit held that the health insurers' injury "is unaffected by whether any given patient who ingested Rezulin became ill."¹⁰ All they need prove is that misrepresentations about the drug's safety led them to pay excessively for defendant's product, a far easier burden than proving causation in each individual case brought on a toxic tort theory. If the Desiano plaintiffs succeed, it will be interesting to see what effect this will have on corporate behavior, and on the substantive law governing toxic torts. It may suggest other innovative theories that would allow plaintiffs to recover without having to shoulder the enormous burden of proving causation.¹¹ Such a shift may be in the public interest, particularly at a time when administrative agencies' resources for enforcement are strained.

Daubert may also have an impact on procedure that needs to be explored. The study on Daubert by the Rand Institute of Justice concluded that the rate at which expert testimony is being excluded has begun to taper off, and suggested that this may be due to plaintiffs offering better expert proof.¹² But there is an alternative explanation. Plaintiffs may be avoiding exclusion of their experts by offering more expert testimony than they formerly did, and accordingly, are litigating only those cases in which the potential award is large enough to cover these additional expenses, and the cost of the additional time that they must spend on preparation. The Supreme Court's opinion in Weisgram v. Marley,¹³ which is sometimes referred to as the Court's fourth opinion on expert proof, pushes plaintiffs' attorneys into making this choice. In Weisgram, the trial court allowed plaintiffs' experts to testify, and plaintiffs were successful at trial. The appellate court, however, found that the experts should have been excluded, reversed and entered judgment for defendant. The plaintiffs argued that they were entitled to a new trial at which they could produce other experts.

The Supreme Court granted certiorari but only on plaintiffs' claim that they were entitled to a new trial. The Court did not review the exclusion of the plaintiffs' experts by the appellate court. The Court affirmed and explained:

Since Daubert, moreover, parties relying on expert evidence have had notice of the exacting standards of reliability such evidence must meet. . . It is implausible to suggest, post-Daubert, that parties will initially present less than their best expert evidence in the expectation of a second chance should their first try fail.

An attorney in a contingent fee case may well have chosen not to go to the expense of additional experts when it saw that the district court would allow its experts to testify. This is a gamble that an attorney can no longer afford to take. Further adding to the expense of expert testimony are the expenses attributable to the 1993 amendments to the Federal Rules of Civil Procedure that sought to improve the quality of expert testimony by requiring written reports from experts. Additional experts, reports and preparing experts for Daubert hearings mean that attorneys in contingent fee cases now need to spend considerably more time than they used to, time for which they of course do not bill.

The consequence is that clients with smaller claims may be unable to find representation. One lawyer, quoted in the New York Times, explained: "I can no longer afford to spend $300,000 trying a case that is only worth $500,000, and that's ridiculous."¹⁵ Daubert may therefore be affecting access to the courts. It is certainly affecting access to the jury; a successful Daubert challenge means that jurors will not hear plaintiff's story. Case studies suggest that defendants in some cases acted reprehensibly even if plaintiffs could not ultimately prove causation.¹⁶ Daubert cuts off knowledge of behavior that ought to be the subject of public debate.

The stringent standard of proof imposed by Daubert has had consequences beyond that seen in individual cases. It interferes with tort law's objective of deterring unacceptable risk and it has made it more difficult for plaintiffs to have a day in court. Even those plaintiffs who have meritorious claims may find that they cannot obtain adequate representation. We need to consider not only whether the Supreme Court trilogy has resulted in better proof, but also whether this jurisprudence furthers substantive and procedural objectives.

¹ 509 U.S. 579 (1993).

² 522 U.S. 136 (1997).

³ 526 U.S. 137 (1999).

⁴ There are some cases, such as those involving the Dalkon Shield, in which plaintiff is able to provide an explanation for how the product caused her disease. And in the case of so-called "signature" diseases, the association of the substance with the disease is so strongly associated statistically that courts and scientists agree that there is a causal connection.

⁵ Whether the end result is better decisions in individual cases is the subject of debate. The most extensive study to date on the impact of Daubert found it impossible as yet to reach a conclusion.

While judges may be more actively evaluating reliability, we do not know if they are doing it in ways that produce better outcomes. Judges may feel compelled to evaluate reliability and yet not be knowledgeable enough in the relevant field to make accurate determinations. Evidence could thus be excluded because it is difficult to understand rather than because it is unreliable. And evidence could be admitted because the judge does not understand a flaw in an argument rather than because it is reliable.

⁶ See Molly Treadway Johnson et al., EXPERT TESTIMONY IN FEDERAL CIVIL TRIALS: A PRELIMINARY ANALYSIS (Federal Judicial Center ed., 2000); Rand Study, note 5, supra at 64 ("standards for admitting testimony have tightened").

⁷ Although evidence law does at times rest on substantive policy concerns. See, e.g., Rule 407 of the Federal Rules of Evidence which excludes evidence of subsequent remedial measures to prove liability in part because it seeks to promote repairs.

⁸ 2003 WL 1903447 (2d Cir. 2003).

⁹ Id. at *2.

¹⁰ Id. at *8.

¹¹ I had suggested in an earlier article that we need to develop a new tort that would impose liability for failure to provide substantial information relating to risk without requiring proof that the failure caused plaintiff's alleged injury. See Margaret A. Berger, Eliminating General Causation: Notes Towards a New Theory of Justice and Toxic Torts, 97 COLUM. L. REV. 2117 (1997). The Desiano cause of action achieves most of these objectives, but provides no compensation for the patients injured by the drug other than 10% of the excess charge.

¹² Dixon & Gill, note 5 supra at 62 ("This suggests that parties proposing expert evidence either did not propose or withdrew evidence that did not meet the standards, or better tailored the evidence that they did propose to the new standards.).

¹³ 528 U.S. 440 (2000).

¹⁴ 528 U.S. at 455.

¹⁵ Greg Winter, "Jury Awards Soar as Lawsuits Decline on Defective Goods, NY Times, sec. A, p. 1, col. 5 (Jan. 30, 2001) ("To better their prospects under the new rules [about expert testimony], many plaintiffs' lawyers are buttressing their arguments with platoons of experts, and improving their chances of winning by choosing only egregious cases involving the most seriously injuried parties.").

¹⁶ See Michael D. Green, Bendectin and Birth Defects 334 (1996) (raises questions about actions by Merrell that reflect on its good faith).