Vol. 17 No. 2 (February, 2007) pp.104-108

TOXIC TORTS: SCIENCE, LAW, AND THE POSSIBILITY OF JUSTICE, by Carl F. Cranor.  New York and Cambridge: Cambridge University Press, 2006.  414pp.  Hardback. $99.00/£55.00.  ISBN: 0521861829.  ebook format. $79.00.  ISBN: 9780511243271.

Reviewed by Andrew T. Hayashi, Department of Economics and Boalt Hall School of Law, University of California, Berkeley. Email: ahayashi[at]econ.berkeley.edu.

TOXIC TORTS by Carl Cranor takes a remarkably broad look at the relationship between science and the law. The book’s central argument is that three recent Supreme Court decisions, which attempted to guide lower courts in utilizing the Federal Rules of Evidence to make decisions on the admissibility of scientific opinion testimony, have made it more difficult for plaintiffs to meet their evidentiary burden. Although the admissibility rules were supposed to be fairly liberal, the effect of the Court’s rulings has been to deny plaintiffs the use of essential expert testimony, a problem compounded by the deference shown to admissibility decisions by granting them heightened abuse of discretion review. The net result of this increased burden has been to limit the ability of the tort system to function as a deterrent to negligent conduct and as an arbiter of justice.

Plaintiffs in a chemical torts action must be able to demonstrate that the substance which is alleged to have caused their injury is both capable of causing their harm (general causation) and that it did, in fact, cause their harm (specific causation). Scientific testimony is necessary for the plaintiff to establish general causation. The Supreme Court’s rulings in DAUBERT, JOINER, and KUMHO TIRE require judges to assess whether the reasoning and methodology underpinning an expert’s testimony is scientifically valid. This is a tall order for judges without scientific training and who, Cranor argues, have responded by demanding more evidence, and of a more restricted kind, than they ought. Cranor argues that the new standard has particularly dire consequences in the area of toxic torts because the regulatory mechanisms designed to ensure the safety of chemicals introduced to the public are inadequate, firms have incentives to avoid conducting research on the toxicological properties of their products, and toxicological evidence is, in any event, sparse. Indeed, this sparseness is one of the more alarming facts motivating Cranor’s thesis: of the 3000 chemical substances produced in the highest volume, there were substantial gaps in our knowledge of the toxicity properties of 75% as recently as 1998 (p.12).

TOXIC TORTS is clearly written and well-organized. Each chapter is broken into numerous short sections; this allows Cranor to put the reader on notice when a new subtopic is being addressed or a short detour is being taken, but it fragments the argument somewhat. He writes to a wide audience, and legal scholars, practitioners, and interested scientists will all find something of interest. Cranor makes a compelling case that the practical limitations on [*105] collecting toxicity data warrant a more nuanced, sophisticated, and permissive standard for the admission of expert testimony in toxic torts cases. Judges simply cannot expect voluminous, flawless, evidence on the toxicological properties of the chemicals that they encounter in litigation. Cranor argues convincingly that “courts must learn to review a wider range of evidence with greater sensitivity and sophistication than many have to date in order to serve law, science, and justice much better” (p.29).

Chapter 1 sets the stage by vividly illustrating the importance of admissibility standards by recounting five torts suits, all of which highlight the fact that excluding scientific evidence may deny justice to worthy plaintiffs. In Chapter 2, Cranor provides an outline of a torts action. This chapter does an effective job of making the later discussion accessible to non-lawyers; however, Cranor leaves unexamined issues related to the relative competencies of judge and jury and the justification for admissibility review in the first place. There is a brief discussion in Chapter 3, but the inquiry focuses largely on the possibility that admissibility review infringes on the constitutional right to a jury trial, and Cranor mentions only in passing a social science literature which shows that juries have the ability to evaluate scientific evidence. At the preliminary hearing where admissibility decisions are made, the judge conducts a reliability review, which aims to ensure that the testimony is not too likely “to lead the factfinder to an erroneous conclusion” (p.287). Non-lawyers are bound to wonder whether judges are in any better position to evaluate scientific testimony than the jury and why expert testimony should be excluded rather than probed by opposing counsel during cross-examination.

Chapter 5 contains what may be some of the most important content in the book. It reveals that many toxins have features which limit the ability of scientists to detect the risks they pose in a timely and accurate manner. For instance, many toxins have long latency periods, do not generate signature effects which make it possible to rule out alternative causes, and operate through mechanisms which are hard to understand or difficult to uncover. These features contribute to the paucity of evidence on their toxicity. In light of the pragmatic barriers to the discovery of harmful effects emanating from these substances, courts ought to be liberal in their admissibility reviews and avoid excluding expert testimony on the grounds that it is not based on ideal evidence. This sort of review is not, however, typically done.

Chapter 6 draws attention to some of the peculiar practices which courts have adopted to try to separate well-supported testimony from groundless opinion. Courts have seized upon several rules-of-thumb for deciding what qualifies as good evidence, often arriving at different conclusions than scientists themselves about whether the evidence warrants a causal inference. In particular, courts fail to consider the evidence for toxicity as an integrated whole. Instead, they evaluate each piece of evidence individually, excluding it if does not meet a threshold such as statistical [*106] significance or doubling the risk of harm.

Chapters 7 and 8 offer suggestions about how the current test for admissibility might be improved. Cranor suggests that courts examine the patterns of reasoning used by scientific committees to determine whether the evidence warrants the claims made by expert witnesses. Ultimately, courts should only ask whether the opinion offered falls within the boundaries of acceptable scientific disagreement.

The overarching theme of TOXIC TORTS is that courts fail to take proper account of the various kinds of evidence which scientists routinely rely on to make causal inferences, so the remainder of this review is devoted to Cranor’s analysis of two kinds of evidence.

Chapter 4 discusses scientific inference and the types of evidence that form the basis of causal arguments: clinical trials, epidemiological studies, animal studies, and case studies. An appraisal of each type of study can be usefully guided by the concepts of internal and external validity. The empirical and experimental economics literature uses the term internal validity to describe the conditions under which the design of a particular study admits a causal interpretation of the statistical results, and the term external validity to describe the plausibility of extrapolating a study’s findings to new contexts. Cranor essentially argues that courts are uncritical of the external validity of epidemiological studies, which often include only healthy male workers, while being unduly suspicious of the external validity of animal and case studies.

Courts have demonstrated a tendency to view epidemiological evidence as the only reliable foundation for causal claims about the harmful effects of substances on human health and have discounted other types of evidence. This is a very unusual position for courts to take, because epidemiological studies struggle with omitted variable biases, and randomized clinical trials are typically thought of as the gold standard for causal inference (clinical trials receive very light treatment in this book). Furthermore, in the absence of randomized trials, scientists themselves consider other evidence when drawing conclusions about the causal efficacy of a substance.

In addition to the external validity concern, epidemiological studies are typically afflicted by questions about their internal validity, i.e. whether the measured relationship between treatment and outcome represent the effect of the treatment or some omitted variable correlated with both. Cranor identifies several further concerns, such as the fact that studies which do not follow subjects for a sufficient period of time after exposure may not detect negative effects if the substance has a long latency period. On the other hand, long studies may also underestimate risks created by substances if they do not properly account for the non-random exit of participants (e.g., death). Epidemiological studies are useful, but courts should not require them as a basis for expert testimony on general causation and should be wary of the [*107] results that they provide. On these points, Cranor is quite right.

On the other hand, Cranor presents an optimistic view of the value of case studies in making causal inferences, something he does because he believes courts currently assign them too little weight in admissibility decisions. However, the same features of toxic substances which make them hard to research generally make case studies especially unhelpful. Cranor suggests that case studies work best when the time interval between treatment and observed effect is short. However, many carcinogens and toxic substances have long latency periods and so case studies are unlikely to reveal much about their effects. More importantly, the most compelling case studies Cranor describes are those in which the biological relationship between cause and effect are already understood. The case study cannot establish causation on its own. For instance, Cranor cites a study which rules out aspirin as a cause of renal dysfunction precisely because the effect followed too closely after taking the drug, implying an implausibly short latency period (p.121). Elsewhere, Cranor cites approvingly a study which found that a patient contracted Guillain-Barre Syndrome (GBS) on three separate occasions after receiving the tetanus vaccine. The patient contracted GBS 21, 14 and 10 days after receiving the vaccine, “time periods well within biological plausibility” (p.119). In these and other cases it is an understanding of the biological mechanism which allows researchers to relate observed effects back to putative causes, and, for many substances, knowledge of these mechanisms is a long time coming.

Cranor also claims that “a good case study rules out alternative explanations of the events” (p.123). This, of course, is a feature of any good study; however, while ruling out alternative explanations makes it more likely that the remaining putative cause is the right one, it is impossible to use this ‘ruling out’ method conclusively unless the universe of potential causes is known beforehand. Finally, the question of external validity is especially acute for case studies. A case study performed on someone who is especially vulnerable to negative effects of a drug cannot be reliably extrapolated to the larger population without some knowledge of how that person’s vulnerability affected the drug’s efficacy.

Of course, this is not a book about clinical trials. Cranor wants courts to better understand the strengths and weaknesses of each type of evidence, and where he presents a somewhat one-sided view, it is only because he is responding to undue skepticism or naïveté on the part of the courts. His emphasis on the value of evidence not utilized by courts, and the weaknesses of evidence privileged by courts, is not a shortcoming of the book.

TOXIC TORTS is an excellent book, filled with keen observations about the science/law interaction, the epistemic structure of scientific inquiry, the norms and conventions which regulate the community of researchers, and the special difficulties faced by the torts [*108] system in obtaining justice and deterring malfeasance in toxic tort cases.

CASE REFERENCES:

DAUBERT v. MERRELL DOW PHARM., INC, 509 U.S. 579 (1993).

GENERAL ELEC. CO. v. JOINER, 522 U.S. 136 (1997).

KUMHO TIRE CO. v CARMICHAEL, 526 U.S. 137 (1999).

*********************

© Copyright 2007 by the author, Andrew T. Hayashi.