Tobacco Documents Online

BY JOHN C. LUIK THE Co1N°rEXr the Dangers of Politically Corrupted Science for Democratic Public Policy HE ASSUMPTIONS ABOUT THE NATURE OF persons and of thc legitimate role of the State (of necessity unargued for) which structure our argument are those of an unreconstructed liberal individualist, namely, that the individuals who make up democratic society are the best judges of the shape thcp wish their lives to take, and con- sequently they should be accorded the maximum liberty, compatible with similar liberty for everyone else, to think, hclicvc, and live as thcv choose. This means that the State's •,Ic is at least fourfold: first, to prevent or minimize harms by one individual to another individual; second, to minimize and where necessary adjudicate the inevitable conflicts that occur between indi%iduals and between individuals and the communin•; third, to defer, wherever possible, from moral judecmcnts about how its citizens choose to shape their lives; and finally, to create the minimal institutional condi- tions which allow its citizens' self-chosen lives the best chance of fulfillment. What this means is that the State will resist the impulse, ho«•cver u•cll-intentioncd, to undermine and intrude upon its citizens' capacities and inclinations for autonomy by defin- ing one vast and unassailable conception of the good life to which all must subscribe. What this means is that the State will sec its citizens as persons of intrinsic worth, fully equivalent in moral standing with itself, with lives not to be managed or saved, bur to be allowed to develop in ways of their own choosing. In order to understand the ways in which science and pub- lic policy intcrscct in the debate over smoking - as indeed so wI rtEt •1)_•1 4 tostor1 1 over many other "health" issues - it is necessary to place the smoking controversy in general and the Environmental Tobacco Smoke controversy in particular within the larger context of both governmental and nongovernmental ct`forts to discourage tobacco use. For most of the twentieth century the campaign to dclcgitimizc smoking has employed two major :veapons, science, particularly epidemiology, and moral- ity, within the general conceptual framework of what can be called health paternalism. Though the mix has varied, the con- junction of the two has been not only consistent, but also highly effective. For example, once it was established that smoking increased the risks of ill health in smokcrs, the groundwork was laid for a series of moral arguments that purported to show that subjecting oneself to these risks was both so irra- tional and immoral as to justifi, government efforts to prc- vent one from assuming the risks. The health paternalism at work here rests on a series of assumptions about reason, autonomy, and the nature of persons that include the fol- lowing: 1. autonomy is not the foundational democratic vaiuc inasmuch as considerations of happiness and welfarc frequently take precedence over it; 2. individuals arc frcqucntly irrational in that they a. often do not understand their interests; and b. even if they do understand their interests they do not know how best to realize those interests; 3. individuals need the State's help in a. discovering and realizing their "true" interests; and b. avoiding irrational courses of action that result in unhappy consequences.

W hat unites these assumptions is the belief that the State is justified in protecting competent adults from the allegedly harmful consequences of their actions through restricting their autonomy. Based on these assumptions, health paternalism advances the following claims: 1. health is the precmirtent value which outweighs, in most instances, all other values such that a rational pcrson would not normally place his health at risk in the interests of some other value; 2. there is but one hcalthy/rational way to live one's life and such a way does not include activities that carry with them significant risks to wcll-being or longevity; 3. individuals have a moral obligation to order their lives in this hcalthy/radonal way; and 4. the State is justified, indeed the State has a moral • obligation, to ensure that its citizens conform to this hcalth/radonal paradigm, even if they wish not to or are unable to through their own efI'orts. H caith paternalism is thus a subtle shift away from the generally uncontroversial right of the State to ensure that consumers are fullv informed about the risks of certain products or activitics to their health to the highly contentious claim that the State is justificd in attempting to manipulate and coerce. Despite its highly problematic char- actcr, health paternalism has been to some degree immune from the sorts of objections that arc routinely brought against other forms of paternalism. Health paternalism's immunitv from such criticism does not derive from the cogency of its arguments but from the fact that it rests not just on moral argument but on "unquestionable" scientific fact. And in a world in which science is increasingly the source of both truth and value the scientific character of i health paternalism is decisive. But, however dosely alimed to science, the abilitv of health paternalism to secure all of the public-policy objecti.•es of the and-smoking movement was always constrained by the fact that, at least within democratic societies, the justifications for government intervention to protect adults from thcm- seives - to coerce "healthy" lifestvles - would continue to have a totalitarian flavor about them that would ensure sig- nificant and widespread opposition. It is only bv demon- strating that the dangers from smoking transcend the smokcr and extend to innocent bystanders that the anti-smoking movement could move beyond obvious health paternalism and enlist unambiguous support for public-policy measures designed to restrict, ban, and criminalize public smoking. The movement away from the risks ofsmoking for smokers to the alleged dangers of secondhand smoke for nonsmokers docs not mean that the hcalth-paternalist arguments have suddenly disappeared from the public-policy agenda of the anti-smok- ing movement. Whether one is considering the policy impli- cations of tobacco advertising or the supposedly addictive properties of nicotine, the paternalist justifications for gov- erttmcnt interventions in the lives of smokers still constitute a significant strand of the anti-smoking argument. What has changed is that arguments about harms to self have assumed a secondary place to arguments about harms to others. What has remained the same is the fact that the new arguments about harms to others, to innocent bystanders, strongly resemble the old "paternalistic" (often religious) arguments. What is new is the consideration of science and moralitv to justif,v public policy. We wish to argue that: 1. the "science" supporting the claims about tobacco's risks to non-smokers is corrupted science, science that has been politically laundered, science that because of its corrupted status actually ceases to be science; 2. the use of such science by the government and the anti-smoking movement reveals not only the illc- gitimacy of their public-policy agenda but the flawed character of the advocacy as wcll; and 3. the existence and use of such corrupted scieiacc poses a significant threat to legitimate democratic public polic}c CORRUPTED SCIENCE I f the government and the anti-smoking lobbV arc to jus- dfy their claim that the debate in socict,v about smoking in the last decade of the twentieth centurv is really a debate about the dangers that smoking poses to nonsmokers then it is imperative that those dangers be demonstrable, compelling, unequivocal, and significant. And it is only sci- encc that can deliver dangers with the requisite pedigree. Should the scientific evidence be less than decisive, the debate about smoking will return to a debate about the legitimacy of health paternalism, which is a debate that nci- thcr the government nor the anti-smoking movement can easily win. Everything, therefore, depends on science. And .vith so much at stake, the pressure to adjust, shave, create, ignore, rcc.•aluate, even manipulate, is cnormous. The pressure comes from at least four sources: from the dispositions of scientists themselves, from the rational char- acter of science itself, from the structure of the scientific enter- prise, and from society's expectations of science. • Scientists, like everyone else, have their own personal perspective, values, and ideological agenda, whatever their commitment as scientists to the objective nature of science. Indeed, a good many scientists, either apart from their sci- endfic training or because of it, believe that tobacco use is dan- gerous and immoral. Most scientists and most hcalthcare pro- fessionals have come of age in a professional atmosphere taStouI A wI rEEC 'r!-'9 4 SE

that for the last twenty-five yeats has been increasingly hostile to tobacco use. It is not unreasonable to assume that many scientists have a strong profcssional interest in establishing their personal bcliefs and the beliefs of their milicu as scien- tific fact. As William Broad and Nicholas Wade observe in Batraycrs of Truth: Fraud and Deceit in Science, "science is not a perfectly objective process. Dogma and prejudice, when suitably garbed, creep into science just as easily as into any other human enterprise, and maybe more easily since their entry is unexpected." • The ideology of science, which focuses almost exclusively on the objective and rational character of science, tends to blind both scientists and those who rely on science to the fact that there are other elements involved in science besides strict logic and rationality. "The presence of a strong ratio- nal element in science has been taken to mean that that is the only significant clement of scientific thought. But cre- ativit}•, imagination, intuition, persistence, and many other • onrarional elements are also essential parts of the scientific process, and other less vital qualities such as ambition, envv and the propensity to deception also play a role" (Broad and Wade, p. 218). These other aspects of the scientific process, which arc common to othcr areas of creative endeavor, tend to work against objectivity, particularly when they are not acknowledgcd as part of the scientific paradigm. • Scientific careers are advanced on the basis of pub- lished, pccr-rcvicwcd findings - findings that for the most part grow out of funded research. The decisions about which research project to pursue and which research results to publish are often determined by a subtle interplay between scientific orthodoxy, funding procedures, quite genuine career considerations, and the search for scientific truth. • Science exists within a society that has certain expecta- tions about what science ought to do, one ofwhich is that sci- ence exists to make the case, as it were, against whatever it is •`iat society considers to be dangerous. The fact that much of society believes tobacco to be dangerous creates a strong series of incentives to establish and indeed enlarge the range of smoking-induced harms, while at the same rimc ignoring or suppressing research that questions these reccived ortho- doxics. But what, it might be asked, is corrupted science? And more importantly, «•hat is the evidence that thc ofiicial Emi- ronmental Tobacco Smoke (better known as ETA or "passive smoke") story is indeed an instance of such a corrupted sci- ence? Could it not be the story is indeed an instance of such science? Could it not be the case, for instance, that the Envi- ronmental Protection Agency's condusions about ETS arc in fact simply incompetent science, science that has failed to do its.vork properly but not science that has deliberately decided to tell the wrong ston•? It is certainly true that not everv instance of weak or flawed science is an instance of corrupted science. Corrupted S2 wI NtEt 'I I-'r4 IoSiorI I science has at least three characteristics that mark it off from simply incompetent science. First, corrupt science is science that moves not from hypothesis and data to conclusion but from mandated or acceptable conclusion back to selected data in order to reach the mandated or acceptable condusion. That is to say, it is sci- encc that uses selected data to reach the "right" conclusion, a conclusion that by the very nature of the data necessarily misrepresents rcalit}: Second, corrupt science is science that misrepresents not just reality, but its own process in arriving at its conclusions. Rather than acknowledging the selectivity of its process and the official necessity of demonstrating the right conclusion, and rather than admitting the complexity of the issue and the limits of its evidence, it invests both process and its con- clusions with a mantlc on indubitability. Third, and perhaps most important, whereas normal sci- ence deals with disscnt on the basis of the quality of its c%i- dence and argument and considers ad homincm argument as inappropriate in science, corrupt science seeks to create for- midable institutional barriers to dissent through excluding dis- senters from the process of revie.v and contriving to silence dissent not by challenging its quality but by questioning its character and motivation. In effect then, corrupt science is science that is flawed in both its substance and its process and that seeks to conceal these esscr.cial flaws. It is essentially science that wishes to claim the policy advantages of genuine science without doing the work of real science. The evidence that the EPA's science on ETS is corrupt sci- ence falls into two categories: evidence about the substance ofthe science and evidence about the process involved in cre- ating and using the science. THE SUBSTAh'TIV'E ISSUE T he EPA's report Rcspiratory Hcalth Efftcts of Passive Smoking: LungCancurand OthcrDisordcrsclaims that "based on the weight of the available scientific cvi- dcncc, the U.S. Environmcntal Protection Agency has con- cluded that the widespread exposure to environmental tobacco smoke in the United States presents a serious and substantial public health impact." The hedging is in the swerve from "hazard" to "impact." Is this the case? In order to answer this question one must first know something about the data on which the EPA's decision is based. The EPA's report refers to the thirty epidemiologic studies on spousal smoking and lung cancer that have been published betnvicen 1982 and 1990. It is important to note that in reterring to the report then-EPA administrator William

Reilly spoke about ETS in the workplace and, though the Report has been used as a basis for demanding smoking bans both in public places and in workplaces, the EPA did not cxaminc those studies that look at workplace ETS cxposurn. Thc over- whciming majority of these do not find a statistically significant association between exposure to ETS and lung can- cer in nonsmokers: a fact that by itself destroys the legitimacy of any harm- based demand for public or workplace smoking bans. Thus, to begin with, the EPA's case is based not on workplace or public-place There is a strong likelihood that even the weak relative risk is a reflection not of some real-world risk, but of problems with confounding variables or interpretative bias. ETS exposure, but on the risks of non-smoking spouses onrracting lung cancer from their smoking spouse. But what Of the thirty studies? The thirty studies come from different countries and vary substantially in size. Some studied fewer than twentry subjects, others are based on larger populations, with the largest study involving 189 cancer cases. Of the thirty studies, twenty-four reported no statistically significant asso- ciation; only six reported a statistically significant associa- tion, that is, a statistically significant increased risk for those nonsmoking spouses. Relative risks are furnccr classified into strong risks or weak risks depending on the magnitude of the risk ratio. Within the thirty studies on ETS and lung cancer none reported a strong relative risk. Moreover, when- ever the assessment of relative risk is weak, there is a sub- stantial possibility that the finding, the assessment, is artifi- cial rather than real. That is to say, there is a strong likelihood that even the wcak relative risk is a reflection not of some real- world risk, but of problems with confounding variables or interprerative bias. There are, for instance, at least twenty con- 0junding factors ranging from nuttition to sociocconomic sta- tus that have been identified as associated with the develop- ment of lung cancer. Yet none of the thirty studies attempts to control for all of these factors. So in assessing the global scientific evidence about ETS and lung cancer, the crucial conclusion is that none of the studics rcport a strong relative risk for nonsmokcn married to smokcrs The EPA Report discusses all thirty studies but limits its statistical analysis to only eleven U.S. studies of spouses of smokers. Of the eleven studies, the EPA claims that ten reported no statistically significant association between ETS exposure and lung cancer, and only one reported a statisti- cally significant association. But this claim about the one U.S. study is in fact true only by using the EPA's unique 90 pcr- cent confidence level. Using the accepted 95 percent confi- dence level none of the eleven studies reported a statistically significant risk. The EPA analysis of these eleven studies claims that together they show a statistically significant difference in the number of lung cancers occurring in the nonsmoking spouses of smokers, such that they suffer 119 such cancers com- pared with 100 such cancers in nonsmoking spouse of non- smokers. It is this finding of statistical significance, a finding based on only eleven U.S. studies, none of which demonstrate a statistically significant increased risk unless they are "reana- lyzed" using the EPA's 90 percent confidence interval (and even such "coaxing" of the figures could produce only one study that purported to show statistically signiftcant increased risk), that provides the only basis for the EPA's decision to das- sify ETS as a "Group A" carcinogen. In order to arrive at its "condusion," the EPA pooled the data from the eleven studies into a combined data assess- ment called a meta-analysis. Mcta-analysis is governed by its own rules: not every study is a candidate for such combined analysis. In general, meta-analysis is appropriate only when the studies being analyzed together have the same structure. The difficulty with the EPA's use of ineta-analysis of the eleven ETS studies is that it has failed to provide the requi- site information about the structure of those studies, infor- mation crucial for an independent assessment of whether the studies arc indeed candidates for mcta-analysis. Thus, the EPA conclusion is based on a meta-analvsis that is diffi- cult, if not impossible, to verifi: Adjusted confidence levels are not, however, the only prob- lem with the EPA analysis. Equally disturbing is the EPA's use of a one-tailed test as opposed to a two-tailcd test. Two-tailed tests (see sidebar) are generally used in statistical analysis because it is rare for one to know a priori that a null hypoth- esis can be discredited in only one direction. By using a one- tailed test the EPA assumes that ETS exposure can only increase the lung cancer risk, despite the fact that a substan- tial number ofstudiashom a dccrrascd risk. (One recent largc study even showed a statistically significant decreased risk.) While the EPA claims that a onc-tailed analysis actually com- pensates for a 90 percent confidence interval, what the use of such an analysis actually does is reduce the confidence level even further. 11ostoU11• wInrtR •11-•14 53

The quality of the EPA`s ETS science is the issue of "confidence intervals." Even by limiting its analysis to only eleven studies, and even by lumping these studies together through a mcta-analysis, the EPA could not have achieved the "right" result if it had not engaged in a creative use of.vhat epidemiologists call confidence intervals. Essentially, confidence intervals express the likelihood that a reported association could have occurrcd by chance. The gcnerally accepted confidence interval is 95 percent, which means that there is a 95 percent confidence that the association did not occur by chance. Inasmuch as most epidemiologists use the 95 percent confidence interval, the EPA itsclf, until the ETS report, always used this interval. Curiously, the EPA decided that in this instance it would use a 90 percent confidence interval, something that effectively doubles the chance of bcing.vrong. Even more curious is the fact that when asked to justify this departure from • cccpted scientific procedure, EPA administrator Rcilly simply replied that the 90 percent confidence interval " was recommended to us bv the scientific community as appropriate to this data." What Mr. Reilly rcally mcans by "appropriate to this data" is that mirhout using this 90 ptrccnt szandard, the EPA could not havc found that tlic tlcvcn U.S. studirs were asratisricallv si,cJnifscant." Without employing a novel standard, without in effect changing the accepted rules ofcpidcmiological report- ing, the EPA result, already painfully coaxed into exis- tcncc, would not have existed, and ETS could not TWO-TAILED TESTS A null hypothesis is a precisely A stated assertion associated with a tatistical test; results of that test ••e intended to deten•nine whether the null hypothesis should be accepted (regarded as truc) or rejected (regarded as untrue). Because we are more comfortable accepting demonstrations that state- ments arc false than otherwise, statisti- cians usually amnge their experiments so that the null hypothesis is contrary to the underiying thesis. Thus, rcjeccon of the null hypothesis corresponds to con- firmation of the thesis. Suppose that like the EPA wc want to demonstrate that exposure to ETS increases the risk of lung cancer. Since .ve cannot examine everyone exposed to ETS we dcsign a statistical cxperi- 54 WIn1El 'f3-•14 10STONII have been labelled a "Group A" carcinogen. Thus, despite all ofits careful selection of the right data, its meta-analysis and finally its relaxed confidence intervals, the conclusive point remains, as Huber, Brockic, and Mahajan note in Consumcrs Rr.rcarch in the United Stares (199I ), that © no matter how the data from all of the cpidemiological stud- ics arc manipulated, rccalculatcd, 'cooked,' or 'massaged,' the risk from exposure to spousal smoking and lung cancer remains weak.... No matter how these data arc analyzcd, no one has reported a strong risk relationship for exposure to spousal smoking and lung cancer." THE PROCESS ISSUE W hile a careful look at the substance of the EP A's ETS claims clearly shows why this science can be called nothing less than corrupt science, an examination of the process underlying this science demonstrates even more clearly its wholly corrupted character. There arc at least ten specific process issues .vorth noting, each of which highlights a slightly different dimension of the corrupted character of the EPA's ETS science. , First, EPA science issues from a perspective that can be 1. traced back to the Lalonde Doctrine propounded by for- mer Canadian Minister of National Health and Wc1farc, Marc Lalonde. Lalonde argued that health messages must bc vig- orously promoted even if the scientific evidencc was incorim- pletc, ambiguous, and divided. Health messages must be "loud, dear and unequivocal" even if thc evidence did not sup- mcnt to determine whether our thesis seems to be true. Our null hypothesis is: Exposure to ETS does not increase the risk of lung cancer. Ncxt we select random samples of individuals exposed to I:TS and random samples ofindivid- uals not exposed to ETS. If equality holds betrnvecn the two samples, that is if the rates of lung cancer are not differ- ent. we have failed to demonstrate our thesis. If, on the other hand, individuals exposed to ETS have significantly higher rates of lung cancer we can reject the null h,vpothesis. In posing a null hypothesis for statis- tical testing one always states an altema- tive hypothesis which is to be accepted if the null h}pothesis is rejected. The alternative h)•pothesis must encompass the entire range of alternatives to the null hypothesis. In this case the correct alternadve hypothesis is that the risk of lung cancer in populations exposed to ETS and populations not exposed to ETS arc diffirenr, that is populations exposed to ETS might have incrcaud risks of lung cancer or they might have reduced risks of lung cancer. This is an example of a two-taiIed anah•sis since exposure to ETS can either increase or decreasc the risk of lung cancer. In using a onc-tailcd test, the EPA failcd to state the correct alter- native hypothesis to its null hypothesis. The EPA in cffect a.aumad that ETS exposure could only increase the risk (one tail) of lung cancer. Since a sub- stantial number of studies have shown a decreased risk with ETS exposure - including a large recent one which was statistically significant -= t.vo-taiIcd tests are required.

port such clarity u.u definition. What we have in the EPA is simply the Lalonde Doctrine as an institutionalized process. Clearly the substance of the ETS data does not support its "Group A" status, nor does it support public and workplace smoking bans (desirable as some might find them on other grounds ) on the grounds that ETS threatens the health of nonsmokers. But the substance ofthc ETS dam is to be ignored because the Lalonde Doctrinc requires that the substance be portraycd The Lalonde Doctrine requires that the substance be portrayed as something that it is not in order to further the health agenda. as something that it is not in order to furthcr the health agenda. What this does is to build into the heart of the scientific enterprise an institutionalized motivation and justification for allowing ends extrinsic to science to determine the findings nf science, for allowing science to be subject to an agenda not es own, for allowing science to lie with a clear conscience. Once one has come to see science as something that of ncces- sity happens within the context of health promotion, then the process corruptions of the EPA follow quite "naturally." This explains why at one level those involved with the EPA decision on ETS arc quite frank about the process. For instance, the EPA official responsible for the revised ETS risk assessment was quoted in Science (July 31, 1992) as admitting that "shc and her colleagues engaged in some fancy statistical foorwork' to come up with an"indictmcnt" of ETS. (The footwork to which she refers is the novel 90 per- cent confidence interval and the one-tailcd test. ) Or to take another process example, the Science Advisory Board which rcvicwed the initial draft risk assessment on ETS, and found the case against ETS based on its association with lung can- ccr unconvincing, actually urged the EPA staff to attempt to "make the case" against ETS on the basis of the similarities `awccn ETS and mainstream smoke. ' To be fair, the consequences of the Lalondc- Doctrine are not confined to the EPA's anti-smoking agenda. For instancc, an article in the Journal of tha American Medical Association for July 29, 1989, reported a study that claimed to show a link benvccn ETS exposure and an increased risk of cervical cancer. In response to critics who noted that such a link was biologically implausible and that the study had ignored con- founding factors, the authors replied that the study was jus- tificd simply on the ground that it might reinforce the "dan- gcrs of smoking" message. "While we do not know of a biologic mechanism for either active ... smoking or ETS to be related to cervical cancer, we do know that cigarette smok- ing is harmful to health. The message to the public, as a result of this study, is one that reinforces the message that smoking is detrimental to health." It would be difficult to find a more succinct example of the Lalonde Doctrine at work. There is no compelling cvidcnce to support our claim, the authors all but admit, but it is important, in the interests of health pro- motion, that the public be made to think that there is scicn- tific evidence of harm. 2 But second, while those involved in the EPA process arc e at one level open about the process, at another levcl thcy are profoundly dissembling. For insrancc, the EPA fails to men- tion that the "Group A" status for ETS was arrived at using a process that violatcs its own Guidalincs for Carcinogenic Risk Asrcssmant. Rather than acknowledging that this suggested that both the substance ofits findings and the process wcre corrupt, the Science Advisory Board reviewing the ETS issue argued that this suggested a need: not that ETS posed no threat to the health of nonsmokers, but rather that the Guidclines for Car- cinogenic Risk Assessment be changed. Given that the right conclusion must be reached and the data do not support that condusion, one must manipulate the data and revise the guidc- lincs governing the process and•the condusion. 3 Third, the ETS risk assessment process has been cor- * rupted from the outset by the fact that it has repeat- cdlv violated the standards of objectivity that prevail in legit- imatc science by utilizing individuals with anti-smoking biases. One member of the group working on the ETS issue at the EPA is an active member of U.S. and-smoking orga- nizations, while the Science Advisory Board that examined the EPA's ETS work included not only a leading anti-smok- ing activist, but several others strongly opposed to tobacco usc. Finally, the EPA contracted some of the work on cer- tain documents related to the ETS risk assessment to one of the founders of a leading anti-smoking group. ~ Fourth, the EPA changed the accepted scientific stan- d dard with respect to confidence intervals, %vithout offer- ing any compelling justification, in order to make its sub- stantive findings statistically significant. Fifth, the EPA's Workplace Policy Guide which as a pol- 5. icy document would, in the coursc of normal scientific process, be dcvclopcd only after the scientific evidcncc was in, .vas actualhv written before the scientific risk assessment was even completed, let alone reviewed and finalized. Quite obviously, science was to be made to fit with policy, rather than policy with science. a osEOria wirfER '11- m

Given that the right conclu- sion must be reached and the data do not support that 8 Eighth, the EPA represents its process . as a comorehensive and objective I 1W_ conclusion, one must manipulate ( ~ the data and revise the guideiines. 4 Sixth, the EPA fails to note that, if the two most recent (f. U.S. ETS studics were to be included along with its eleven other studies, it would have resulted in a risk asscss- mcnt that .vas not statistically significant, even using the 90 percent confidence interval. With its entire "conclusion" at risk, there are exceedingly compelling process reasons for the EPA to have excluded these two later studies from their • nalysis. ~Sevcnth, exclusion, however, was apparently insufficient, /. for the EPA does more than simply not use the studies, it actually refers to them in an appendix and misrepresents one of them by claiming that it supports the EPA's ETS conclu- sions. The study, by Brownson, et. al, which appeared in the November, 1992 Asncrican Journal of Public Hcalth, reported no statistically significant increase in risk between lung cancer and ETS exposure. In order to get around this politically unacceptable conclusion, the EPA quotes Brown- son as concluding: "Ours and other recent studies suggest a small but consistent increased risk oflung cancer from passive smoking." But this is not the issue, as the EPA well knows. The question is not whether there is a small increased risk, but whether there is a statistically significant risk, which Brown- son concludes there is not. In effect, the EPA misrepresents a scicndfic finding by changing the terms of reference from taristical significance to just plain risk. • This penchant for misrepresentation is nor, however, con- fined to recent studies. For instance, the EPA anah•sis consis- tcntly makes reference to the Garfinkel, et. al. studti•. At Chap- ter 5.48 the EPA claims that the Garfinkel studv presents "at least su¢s;estive evidence of an association bet,%•cen ETS and lung cancer. ..." But a carcful rcading of Garfinkcl does not confirm this at all. Garfinkel actually says that -%.-c tbund an elevated risk of lung cancer, ranging from 13-31 percent, in women exposed to smoke ofonccrs, althouch the increase %vas not statistically significant." (L. Garfinkcl er al, "involuntarv Smoking and Lung Cancer: A Case-Control Studv," Journal ofrbe National Cancer Instiruti, 75, 1985.) Thc entire ques- tion of suggestivc cvidcncc is bogus: the rcic.•ant question is whether Garfinkcl found a risk that was statistically significant. He did nor, and the EPA misrepresents his findings. S6 wi.EEt '1)-'r4 t0SS0NIt (i 7 rcvicwed for the report. I'S data. In the usual Eis would imply a care- the criticisms that have studies used to reach its ver, a careful cxamina- Ephy accompanying the at this is not the case. with the bibliography )t a "comprehensive list ilabic on the topic," it references cited and Yet, to take but one example, one would never know from the report or its bibliography that the work ofTrichopoulous had been subjected to significant crit- icism by both Burch and Heller, since neither is mentioned in the bibliograph}: Nor would one know that Trichopoulous acknowledged such criticism and even criticized his study himsclf. (See Trichopoulous ct. al, "Lung Cancer and Passive Smoking" Int. J. Cancer, 27:1-4.) Now the possible explanations for such selectivity are that: 1. The authors of the study are not familiar with such criticisms, which would suggest incompetence, or 2. they arc familiar with the criticisms but have mis- understood, ignored, or discounted them. But even if one were to discount or ignore them, it is still odd, if one is committcd to objectivity and openness, not to cite them. Not to cite them suggests that one wishes to act as if they didn't exist, and to do this is to give rise to more than the suspicion that the EPA's ETS work is really an instance of a closed-loop process abuse. In a closed loop the circle is never opened up to divergent, dissenting views that challenge the orthodox conclusion. It is not simply that such divergent vicn•s arc discounted, it is rather that, as the EPA discussion and bibliography indicate, they simply arc never heard - indeed judging from the bibliography they don't exist. When one considers this doscd loop process in the context not merely of what the EPA cxduded in terms of dissenting voices, but in the context of what it sought to include in terms of determining voices - the anti-smoking movement - then it is hard to assign any degree of objec- tivity to the process. 9 Ninth, there is significant evidence that the EPA ignored the misgivings of its own scientists about its ETS asscss- ment process and conclusion. Two internal EPA documents (April 27, 1990, and March 23, 1992), both by the EPA's ~ Environmental Criteria and Assessment Office and both ~ ~ recently released by Congressman Tim Valentine, suggest that the EPA process and report was badly conceived and argued, C4 that the allegcd "causal" connection between lung cancer and t•r- ~ ETS was overstated, and that the evidencc does nor support a Group A carcinogen classification for ETS. ~ C7t

] Tenth, despite the significant difficulties that have been 10l . raised about the quality of EPA science, the EPA process is incapable of correcting itself. This was made partic- ulariy clear by the Expert Panel in its report SaftBuarding tht Fitturc: Crcdibk Science, Crcdibk Dauions, which noted that: 1. EPA "science is of uneven quality"; 2. the "EPA has not clearly conveyed to those outside 3. 4. 5. • or even inside the Agency its desire and commitment to make high-qualicy science a priority"; "the science advice function - that is the process of ensuring that policy decisions are informed by clear understanding of relevant science - is not.vell defincd or cohcrcnth, organized within EPA"; thc "Agency docs not have a uniform process to ensure a minimum level of quality assurance and pecr review for all the science developed in support of Agency decision making"; the "Agency lacks the critical mass of externally recognized scicntisu needed to make EPA science generally credible to the wider scientific commu- nity"; 6. "science should never be adjusted to fit policy." This is perhaps the most significant process corruption of all, namehy a process that is quite conscious of its problems but is umvilling and unable to address them. Of course even this characterization is perhaps too kind given that what the Expert Panel describes as problems are really, for the anti- smoking movement, just the normal way that science must proceed if it is to make the anti-smoking case. If this is the case, then there is no conscious sense of process problems. What the Expert Panel's Report actually provides, of course, is another description of corrupted science - science cor- rupted in its substance and its process; science driven by a pre- dctcrmincd policy agenda; science based on inadequate data; science of uneven quality and inadequately pecr-re%iewed; sci- ence lacking critical validation by outside scientists represen- • lrive of "wider scientific community;" and science, finally, fullv aware of its corruption, but unable to heal itseif. THE USES OF CORRUPTED SCIENCE I t c is clear from the way in which the EPA has handled the ETS issue that the anti-smoking movement is awarc of, if not directly involved in, using corrupted science in the pursuit of its public-policy agenda. Indeed, as Alvan Fcin- stein, a Yale University epidemiologist writing in Taricologi- cal Patbolqp- noted, a prominent epidemiologist comment- ing on the EPA's work on ETS admitted that, "Yes, it's rotten science, but it's in a worthy cause. It will help us to get rid of cigarettes and to become a smoke-free society." But what sorts of moral questions are raiscd by the anti-smoking movemcnt, a movement that has always claimed the moral high ground for itself alone, using and pressuring govern- ments to use corrupted science as a basis for public policy? One moral question is obviously the question of the legit- imacy of misrepresentation, for corrupted science is at bot- tom science that misrepresents the state of reality. And what a careful analysis of the scientific claims of the EPA and the anti-smoking movement reveal is a profound and systematic disregard for the truth about the dangers from ETS. Not only are data manipulated to produce the desired results and sup- pressed or dismissed when they do not fit the standards of political correcmess, but accepted standards about confidence intervals are changed without justification. In cfl-cct, one has an ethic that legitimizes misrepresentation in the service of a good cause - "a smoke-free society." • But is a smoke-free society a sufficient justification for a public health movement founded on unreliable science and blatant misrepresentation? We would suggest that it isn't. First, structuring a public health campaign on deceit is to place it upon tertain that is both notoriously slippery and crevice- laden. The frightening thing about institutionalized deceit, even in the allegedly righteous cause of eliminating smok- ing, is that like any moral corrosive it is both so easy to jus- dfv and so difficult to restrict its use to the ends that originally justified its employment. Second, the entire project of cor- rupted science, like all projects of deception, is designed to manipulate individuals and society to do things that they would not normally do, and to do so based on a false pic- ture of rcalitu The liar's game, is after all, morally deviant pre- cisel,v because it subverts our autonomy by misinforming us. The liar distorts the truth in order to obtain our consent not through argument but through coercion. And the great enemy of frecdom is not so much overt coercion but the coer- cion brought about by biased information. The corruption of a science that misrepresents is moral corruption of the most foundational sense, for it corrupts a centerpiece of both morality and democracy, namely our ability to act freely. But there is a second moral question here that goes beyond the moraliny of misrepresentation into what might be called the morality of suppressing dissent. Both the process of producing corrupted science and of utilizing it as the basis for public policy demand a fundamental intolerance of disscnt, both scientific and otherwise. The imperatives of health pro- motion are such that both the ambiguities and uncertaintics that form a legitimate part of science and more importantly, serious qucstions about the quality of the evidence and whether it justifies the proposed public-policy rneasures, cannot be tolerated. This means that scientific and public-pol- icy dissent must be suppressed by portraying dissenters as in the pay of the tobacco industry or marginal to thc scientific establishment. This strategy raises a host of subsidiarv moral questions. Whatever the cost, "science" must be scen to provide a condusivc and united answer to the question of tobacco and its harms to the innocent. Thus, dcspitc the vital role ofqucstions, argument, and dissent in science as %% rll tOSTOMIt WIMTII 'tl-'9 4 sy

as in democratic Iifc, the anti-smoking movement seeks to silence dissent in the interests of protecting not the truth but its misrepresentation of the truth. The third moral question centers on what the manufacture and use of such science does both to science and to legit- imatc democratic public policy. Cor- rupted science is rather like an intellec- tual acid rain that cats away at evcrvthing that it touches. For instance, it gnaws No one who genuinely cares about good public policy can countenance the corruption of science. away at the distinguishing characteristic ofsciencc - its objec- dvity - and threatens to rcnder science cssentially.vorthless for public-policy purposes. Though science is never com- pletely objective, if indeed complete objectivity is possible, it at least, in distinction from much of the political process, pro- ses a fundamental interest in reason, evidence, and bias-free judgement. In fact, much ofscicnce's standing in contem- porary society derives from its objective character, as does muchvf its usefulness in the public-policy process. In effect, we have a high degree of confidence in the scf- entific process as providing a careful, evidcnced, and to some degree, value-frec, assessment of certain questions relating to public policy, and it is precisely this utility that the use of corrupted science threatens. If science ceases to work out- side of the political and policy process, if it ceases to be a tool available to all sides of an issue, if it becomes politicized and ideologically sensitive, then it ceases to be valuable in the policy process. Rathcr than acting on the voice of rea- son, it becomes nothing more than another special pleading. In this sense, to use corrupted science, for however allegedly worthy an end, is inevitably and irretrievably to corrupt science itsclf. No one who genuinely cares about ~,od public policy, policy crafted on the basis ofcarcful argu- ~lfient, cogent reasoning, and compelling data, policy that can stand the test of careful probing and consistent dissent, can countenance the corruption of science. But the use of bogus ETS science to manipulate the pub- lic-policv debate on smoking threatens not just science, but also the standards of rationality that distinguish legitimate public poiin: Adherence to the norms of rationality require that the identification of problems, causes, and solutions be based on empirical evidcnce of the most rigorous sort, evi- dence that is specific, strong, consistent, or coherent and demonstratcs the appropriate causal connections, that rests on rational arguments which arc clear and logically com- the EPA and the anti-smoking movement's ETS science cannot meet any of the tests ofrationality that determine Icgit- imate public policy problems and solutions. The ETS "evi- dence" is not specific, strong, consistent, coherent, nor does it demonstrate the appropriate causal connections. If it fails these tests, it cannot provide compelling rational reasons - as opposed to rhetorical and emotional reasons - for its pub- lic-policy recommendations. The use of corrupted ETS science is, howevcr, more than simply an abandonment ofrcason in the public policy process; it is also frightcning, an attempt to institutionalize a particu- lar irrational vicw of the world as the only legitimate perspec- tivc: to replace rationality with dogma as the legitimate basis of public policy. If the use of corrupted ETS science by the EPA and the anti-smoking movement represented simply the abandonment of reason, then such actions would be simply nonrational. But the EPS's efforrs go beyond the nonrational to the irrational, to an assault on reason itself By rcfvsing to include aidcnce ofsdcntific disscnt from the officially detcr- mined "truth" about ETS, as evidenced in the omission from key bibliograhies of any references to criticisms of key find- ings and studies, by manipulating and mis-reporting data, and by portraying those who disagree as being "mouthpieces" for the tobacco industry, the proponents of the and-smoking policy agenda reveal themselves as enemies of the open and self-correcting process of reason. In a very real sense the "truth" about ETS ceases to be open to rational assessment and assumes instead the status of revealed dogma. And only those who ultimatcly fear, if not loathc, reason are comfort- able with dogma as the basis of public policy. By far the most morally objectionable aspcct of the anti- smoking movement's usc of the ETS issue is its readiness to use corrupted science to deprive smokers not only of their right to pursue their pleasure in public, but quite possibly to gain or retain their employment, or advance their prospects. Put O ~ pclling. Problems and solutions that cannot meet this stan- more bluntly, it is the question of whether it is morally justifi- ~ dard of argument are not allowed a place in the public-pol- icy process since to do so is to abandon the commitmcnt to reason that is a fundamental dcmocritic valuc. Yet the use of corruptcd ETS science as a basis of public able to use bad science to hurt people? What should never be lost sight of in this debate is that without the alleged scicn- tific justification of harm to innocent parties, there is no com- pelling public-policy rationale for banning or restricting smok- W ~ policy is nothing less than an abandonment of ntionalin, as ing in public piaccs or workplaces. Once the corrupted science ~~ a measure of icgitimatc public policy. As we noted above, is stripped away, there simply are no harms, and without St - wi.iEt •f]--t4 • tostoSIi

those harms, smoking becomes a scif-regarding behavior, interventions against which can only be advanced on patently paternalistic grounds. The and-smoking movement might still argue that public and workplace smoking should be banned in order to discourage smokcrs from smoking, but this argument loscs its compciling harm-to-othcrs' character and becomes • instcad nothing more than an argument about the statc intcr- vcning in the private lives ofcompctcnt adults. What is so morally offensive here is that truly morally blameless people - not the allcgcd victims of smokers - but smokcrs themselves, arc to be harmed in significant ways on the basis of bogus science and for no good reason. What makes the morality of the anti-smoking movement as cor- rupt as its science is that it is prepared to exploit for its own ends our readiness to deprive individuals of certain rights if the exercise of those rights appcars to harm others by cxplic- itly manufacturing harms to othcrs. In doing so, the anti- smoking movement simultaneously violates perhaps the two most fundamental moral principles, first by treating per- sons, in this case smokcrs and their alleged harms to others, as merely means to the end of a smokc-frce society and not as cnds in their own right, and second by inflicting substan- But the question of the moral justifiability of using cor- rupted science to hurt people goes beyond the question of depriving individuals of their right to a significant pleasure, or even of a job, to something fu more crucial, namely the justifiability of depriving individuals of their moral standing through stigmatizing them as moral outcasts. In the end, this is, of course, the logical outcome of ETS science, to make smokcrs a class of moral miscreants who see themselves and are seen by others as so ruthlessly intent on pursuing their own interests that they are blind to the harm they inflict on others. It is indeed but a short way from the claim, "Smok- ing kills" to the conclusion that "Smokcrs kill." But then, such a conclusion is the public-policy justification for bans on public smoking. The debate about Environmental Tobacco Smoke, though ostensibly a debate about smoking, is really a debate about much more than smoking. It is a debate at bottom about the lcgiti- macy ofpcrverting science and public policy founded on science in the interests of a particular health ideology. It is a dcbate at bottom about the worth ofa health paternalism that guarantees to leave all of us substantially less free but no less ill. tiai pain on an entire class of people without their consent Dr. John C. Luik, a nonsmoker, is a Senior Associate of the • and for no compelling reason. Niagara Inszitutc.