Abstract
Although the tobacco industry helped fund the attack on “junk science,” it has created its own dubious scientific scholarship for its expert witnesses.
We suggest that plaintiffs’ counsel should be proactive in using Daubert hearings to exclude the tobacco industry defendants’ scientific expert witnesses by introducing documentation, such as we have found through researching previously privileged internal industry documents, to prove that much of their proposed testimony was developed by and for their lawyers.
DURING THE PAST THREE decades, litigation has become a tool for achieving public health goals, particularly within the tobacco control movement.1 Lawsuits by smokers suffering cigarette-related diseases and the estates of those who died from smoking continue to be particularly effective at holding cigarette manufacturers accountable for their wrongdoing. Product-liability and other private lawsuits against cigarette manufacturers ask the question whether their “corporate decisions that contributed to the plaintiff’s illness were reasonable or unreasonable, in good faith or fraudulent.”2
Cigarette manufacturers would prefer that juries never have the opportunity to answer such questions. In addition to supporting ultraconservative advocacy organizations and legislation aimed at dismantling many of the rights under tort law through so-called tort reform, cigarette manufacturers prolong pretrial litigation to make lawsuits as expensive as possible to discourage plaintiffs. They also spend large sums cultivating experts and manufacturing evidence to confuse jurors about the relevant scientific issues. It would seem that the evidentiary rule on expert testimony developed by the United States Supreme Court in the 1993 case Daubert v Merrell Dow Pharmaceuticals, Inc3 (Daubert I ) and expanded by the Appeals Court to which the case was remanded (Daubert II 4), would be widely used by plaintiffs’ attorneys to exclude these experts and this evidence. Daubert and the cases that follow it allow either party to require the judge in all federal courts and about one-third of state courts to determine whether an expert’s conclusions and underlying research are scientifically sound. Unfortunately, however, the costliness of these cases to plaintiffs’ attorneys and their unwillingness to further complicate the cases by filing their own evidentiary motions has made Daubert challenges of tobacco industry witnesses rare.
The Supreme Court in Daubert I suggested hallmarks of sound science for judges to use when determining whether to admit expert testimony. The court’s recommended considerations are (1) whether the evidence is based on a testable theory or technique; (2) whether the theory or technique has been peer reviewed; (3) in the case of a particular technique, whether it has a known error rate and standards controlling the technique’s operation, and (4) whether the underlying science is generally accepted.3 When the case was remanded back to the appeals court for a decision applying the newly enunciated standard, the presiding judge added more criteria for whether expert testimony should be admissible: (1) whether testimony was prepared in anticipation of litigation, (2) whether there is an evident bias toward a particular conclusion in a scientist’s testimony, and (3) whether the expert based testimony on preexisting or independent research.4
Such challenges are expensive to mount because plaintiffs already are overburdened financially by tobacco companies’ defense tactics of delaying trials with a blizzard of pretrial motions and challenges aimed at depleting the plaintiffs’ often meager funds. Indeed, the tobacco industry is well aware of the financial dilemma faced by plaintiffs’ attorneys. As an attorney for RJ Reynolds put it in an informal memo celebrating the voluntary dismissal of several plaintiffs’ cases:
The aggressive posture we have taken regarding depositions and discovery in general continues to make these cases extremely burdensome and expensive for plaintiffs’ lawyers, particularly sole practitioners. To paraphrase General Patton, the way we won these cases was not by spending all of [RJR]’s money, but by making that other son of a bitch spend all of his.5
Four years before Daubert I, Tobacco Institute attorney Lee Stanford cautioned “all tobacco attorneys” about the threat of plaintiff challenges to industry science in the courtroom:
A question arises whether admissibility of our general causation testimony could be challenged as having insufficient scientific support and/or acceptance by the scientific community. One can foresee a plaintiff’s motion to exclude [Arthur] Furst-type testimony.6
Arthur Furst testified for the industry for years based on his industry-funded work exposing animals to the tar found in cigarettes. Furst testified that because he was unable to induce lung cancer in the test animals, he concluded that smoking did not induce cancer in humans.7 Under the new Daubert I and II standards, plaintiffs would be able to challenge his testimony as to all of the new criteria, provided they were willing and able to allocate funds to raise the issue in a Daubert pretrial motion.
The one-sided application of Daubert due to the tobacco industry’s access to virtually unlimited funds allows their experts to escape scrutiny and consequently to continue to make questionable scientific conclusions in their testimony. A search of the industry’s internal documents, which are posted online as part of the 1998 settlement of state lawsuits against the industry, demonstrates that much of the “science” on smoking and health developed by cigarette manufacturers and offered into evidence should fail under a Daubert analysis.
Curiously, although the tobacco industry has relied on “junk science” to defend its cases, the industry actually helped develop the movement to criticize much of plaintiffs’ expert testimony as junk science. De-privileged documents of the tobacco industry reveal that it gave substantial monetary and public relations support8 to Peter Huber’s book, Galileo’s Revenge: Junk Science in the Courtroom, through his sponsoring organization, the Manhattan Institute. Huber’s book has been thoroughly criticized by one scholar as containing its own fair share of junk science.9 Perhaps the tobacco industry assumed that the resource imbalance between plaintiffs’ attorneys and its own powerfully funded legal team would keep concerns about junk science from boomeranging against the industry.
The tobacco industry even used the attack on junk science as a vehicle to propagate its own junk science. Philip Morris, with the aid of public relations firms and its outside counsel, created a group in 1993 called The Advancement for Sound Science Coalition (TASSC), whose stated mission was to advocate the use of “sound science” in public policy decisionmaking.10 The group’s true purpose, however, was to influence perception in the scientific and regulatory arenas on the subject of environmental tobacco smoke (ETS) in such a way as to imply that cigarette smoke is not a harmful carcinogen, as the Environmental Protection Agency had designated in 1992.11 Philip Morris tried to conceal its primary sponsorship and the group’s main goal of undermining the belief that ETS was harmful by including in TASSC’s purview other junk-science issues such as regulation of food additives and automobile emissions.12 In Daubert II the judge stated that this sort of manufactured scientific testimony would be excluded from evidence.
Daubert II cast particular doubt on science that was developed by and for lawyers.4 In fact, much of the science used by the tobacco industry witnesses was commissioned by their lawyers. The tobacco industry has been well aware that much of the scientific research it commissioned was doubtful and usually geared toward thwarting liability claims and developing a strong public relations front. In the minutes of a meeting dated September 10, 1981, at which general counsel from several tobacco companies were present, the discussion included frank admissions that the continued use of some of their expert witnesses “will start to lose credibility for themselves and for us.”13 These attorneys talked about their concerns not only for the quality of their scientific witnesses (with one lawyer stating that “[a B&W lawyer] is concerned if the science is not worth a damn”), but also for the fact that they were using these witnesses to “make advocacy primary and science secondary.”
Four examples of the tobacco industry’s improper creation of expert testimony illustrate its potential vulnerability to Daubert challenges.
In 1988, Philip Morris met with United Kingdom industry counterparts to discuss a strategy, nicknamed the Whitecoat Project, for recruiting scientists to do research that would attempt to counter and obfuscate recent findings that ETS was harmful to nonsmokers.14 The effort would be overseen by American tobacco industry lawyers from Covington & Burling and the Tobacco Institute. A remarkable document summarizing this meeting concedes that the “Philip Morris strategy is perhaps questionable in some respects, e.g., involvement of lawyers at such a fundamental scientific level; disadvantages in perception of what will only be perceived as a ‘pro-industry’ group of scientists.”15 The industry took care to screen the scientists and to eliminate “obvious ‘anti-smokers’ or those with ‘unsuitable backgrounds’.” This strategy, plus the involvement of lawyers, however, worried some of the project’s planners, and they expressed concern that “the rather oblique initial approach may appear to be somewhat less than honest to many scientists.” The industry recruited Whitecoat Project scientists with an eye toward making them potential pro-industry witnesses or spokespeople. Another memo regarding the project listed as one of its primary objectives: “Preserve product liability defences.”16
This practice of conducting scientific studies in anticipation of litigation contravenes the directives of both Daubert I and II. Tobacco industry scientists disseminated their results through industry-sponsored, non–peer-reviewed publications that simply summarized industry-sponsored and-organized conferences, something that Daubert I specifically rejects as an acceptable basis for expert testimony.17 These conferences or symposia, which were biased and clearly pro-tobacco (factors discussed in Daubert II ), were held in order to give the scientists less critical forums for the publication of their theories that ETS was not harmful.18
Associates for Research into the Science of Enjoyment (ARISE) is a nonprofit organization primarily funded by tobacco industry companies and led by psychologist David Warburton of the University of Reading in the United Kingdom. The group’s stated purpose is to “discuss the ways in which ‘everyday pleasures’ such as eating chocolate, smoking, drinking tea, coffee, and alcohol, contribute to the quality of life.”19 Warburton is frequently cited by the tobacco companies as the primary support for their claim that nicotine is not addictive.20
Industry documents reveal that ARISE is primarily focused on advocating on behalf of the tobacco industry, and uses the support of other industries such as soft-drink manufacturers as a front for its real purpose. For example, in fiscal 1993–1994, American and European cigarette manufacturers contributed 99.35% of its operating budget. For 1994–1995, a memo suggests funding in the amount of $200,000 each for itself, RJ Reynolds, and Rothman’s Europe.21 The tobacco industry’s heavy financial support of this group also makes the objectivity of their research questionable. For example, a member of ARISE, Dale Atrens, testified in a smoking case on behalf of British American Tobacco’s Australian subsidiary, WD&HO Wills, that research did not persuade him that smoking causes cancer. During cross-examination, however, he admitted that not only had he not read all the research he relied on, but that he didn’t know that many of the studies he quoted had a notation on them stating: “Funded by Philip Morris.”22 Again, this would be unacceptable under the two factors of Daubert I that require scientific testimony to be based on a theory or technique that has been peer reviewed and, in the case of a particular technique, that it have a known error rate and standards controlling its operation. Further, this contravenes Daubert II ’s requirement that testimony not be “prepared in anticipation of litigation” or show “an evident bias towards a particular conclusion in a scientist’s testimony.”
In 1985, the Tobacco Institute joined forces with an indoor air quality (IAQ) assessment company called ACVA Atlantic, which later changed its name to Healthy Buildings International (HBI).23 The Tobacco Institute thought HBI’s work, which consisted of measuring IAQ in buildings and then recommending improved ventilation (as opposed to embracing indoor smoking bans), would be a useful diversionary tactic in case its efforts to cast doubt on the harmfulness of ETS were ineffective.24 In an industry document, the “HBI Concept” is described as follows: “The key objective of the HBI concept is to broaden the debate on indoor air quality to deflect the ETS challenges and to gain acceptance of a systems approach to maintaining indoor air quality.”25
With Orwellian irony, the Tobacco Institute referred to HBI’s public relations tours as “the truth squad” in its internal memos26 even though HBI’s consistent message (dictated by the Tobacco Institute) that banning smoking would not improve IAQ and that proper ventilation systems were the key to improved IAQ apparently was based on falsified test results. One of the true purposes of the Tobacco Institute’s truth squad, as stated in the HBI Concept memo, was to influence the tone of scholarship and the scientific community’s attitude toward ETS in ways that would favor the tobacco industry’s position. Thence the institute’s point of view could be introduced in future litigation as legitimate science, without being detected for the junk science it truly was.
Through a whistleblower’s confession on NBC Nightly News in 199227 and a subsequent congressional inquiry,28 the public learned that HBI had a lucrative financial and intimate ideological relationship with the Tobacco Institute that motivated it to falsify data from its IAQ measurements (buildings that the Tobacco Institute itself had chosen for evaluation).29 The congressional investigation showed rampant falsification of data to lessen the reported impact of ETS in buildings HBI assessed (which included many prominent federal buildings as well as the Tobacco Institute’s own headquarters), citing HBI employees’ affidavits that their field notes had been falsified and changed by the president of the company, Gray Robertson, or other HBI representatives. Moreover, Alfred H. Lowery, PhD, hired by the congressional committee to reevaluate HBI’s work, concluded that
HBI’s conclusions are not supported by its own data. In addition, in my opinion, the data is so marred by unsubstantiated data entries, discrepancies, and misclassifications that it raises serious questions of scientific fraud with resulting implications for worker health.28
Although HBI’s main testimonial value as expert witnesses for the tobacco industry seems to have been in the areas of regulatory hearings and public relations appearances, its work also served to influence the scientific community’s thinking on ETS issues through a scholarly article.30 Moreover, HBI’s Web site advertises, as one of its services, testimony by its expert witnesses “to assist Clients [to] avoid potentially damaging litigation.”31
The fact that HBI’s study, which was thoroughly debunked by the congressional investigation, appeared in a peer-reviewed journal—one of the main criteria in Daubert I—shows that shoddy and purposely falsified data can slip through even this higher level of academic scrutiny.32 Moreover, there is ample evidence that “the known or potential rate of error for the technique or theory when applied” (another Daubert I criterion) of HBI’s experiments is woefully inadequate to be considered as admissible expert testimony. In any Daubert hearing, courts should take into account the tobacco industry’s reputation and its aggregate record for using specious science as well as diversionary tactics disguised as sound science in making the decision as to whether to admit their scientists’ and other expert witnesses’ testimony.33
In Daubert II, Judge Kozinski disparages the type of “professional witness” whose testimony appears to be formulated and paid for by the party who hired the witness, and who spends more time in the courtroom than in the laboratory. Ironically, Judge Kozinski cites Peter Huber’s book Galileo’s Revenge, a book now recognized as being rife with errors and amounting to junk science itself,5 to support this statement: “That an expert testifies based on research he has conducted independent of the litigation provides important, objective proof that the research comports with the dictates of good science.”3 Yet one of the tobacco industry’s most relied-on witnesses is Kip Viscusi, a professor at Harvard University.34 Viscusi’s testimony appears to be based on research done mostly or entirely at the behest of the defendants, which in a Daubert hearing might not withstand scrutiny.
In the case of State of Minnesota and Blue Cross and Blue Shield of Minnesota v Philip Morris, Inc, et al., the plaintiffs were able to show that Viscusi’s testimony was based on surveys done in anticipation of litigation (a reason to exclude such testimony under Daubert II ), commissioned by Jones Day Reavis and Pogue, the law firm representing RJ Reynolds Tobacco Company.35 The defendants tried to argue that an independent study Viscusi performed to assess the surveys the tobacco industry conducted had undergone peer-review analysis, but the judge rejected it since Viscusi’s raw data was based on the other tainted surveys as well as under Daubert I factors that exclude testimony that is not “based on a testable theory or technique . . . and . . . in the case of a particular technique, does [not] have a known error rate and standards controlling the techniques operation.” It may well be that the reason a prolific expert witness like Viscusi was excluded from testifying in the Minnesota case is that the state’s attorneys had sufficient funds to file for a Daubert hearing. The fact that there is now a foundation for drafting an effective Daubert motion, based on the groundwork already laid in the Minnesota case, should encourage plaintiffs with more meager funding to follow suit.
The tobacco industry may not be alone among product liability and toxic tort defendants in attacking junk science while simultaneously creating and relying on it. Although it is expensive for plaintiffs’ attorneys to bring Daubert challenges against defense experts, it is even more costly for society to allow such ruses to insulate these companies from legal accountability for their dangerous products and outrageous misconduct. Perhaps public health–oriented organizations and foundations could fund Daubert challenges that are beyond the capabilities of most plaintiffs’ attorneys.
This work has been supported by the National Institutes of Health (grant RO1 CA 87571) and the Project on Scientific Knowledge and Public Policy.
