The Cigarette Century

by Allan Brandt

  By 1962, after nearly a decade of work, Hill & Knowlton was eager to be able to explicitly demonstrate the impact of their interventions on their client’s behalf.

  Now—can we, from this experience, answer this fundamental public relations question: Is such preparation and effort for simultaneous comment on attacks on your client worth the effort it requires?

  We say the answer is unequivocally yes!

  Proof? Well, how do you prove it?

  From time to time, man-on-the-street interviews ask about the smoking question. In almost every one of these, there will be a quotation that is almost an exact paraphrase of some statement issued for the tobacco accounts.111

  Hill & Knowlton had successfully produced uncertainty in the face of a powerful scientific consensus. So long as this uncertainty could be maintained, so long as the industry could claim “not proven,” it would be positioned to fight any attempts to assert the regulatory authority of public health.

  Even as the tobacco industry carried out its massive effort to sustain and amplify scientific skepticism during the 1950s, its own researchers were confirming and adding to the evidence showing the connection between lung cancer and cigarette smoking. Internal assessments of the potentially carcinogenic characteristics of cigarette smoke stand in sharp contrast to the companies’ public statements. In February 1953, Claude Teague, an R.J. Reynolds research scientist, closely reviewed epidemiologic studies and animal studies and concluded:The increased incidence of cancer of the lung in men which has occurred during the last half century is probably due to new or increased contact with carcinogenic stimuli. The closely parallel increase in cigarette smoking has led to the suspicion that tobacco smoking is an important etiologic factor in the induction of primary cancer of the lung. Studies of clinical data tend to confirm the relationship between heavy and prolonged tobacco smoking and incidence of cancer of the lung.

  . . . There appears to be a growing suspicion, or even acceptance, among medical men and cancer researchers that the parallel increase in cigarette consumption and incidence of cancer of the respiratory system is more than coincidence.112

  Despite such candid internal assessments, company executives continued to offer blanket reassurances to consumers and stockholders.113

  At the same time that the Frank Statement was asserting that “there is no proof that cigarette smoking is one of the causes [of cancer],” the industry documented a large number of known carcinogens in its product. Although the industry (except for Liggett & Myers) insisted that it did no in-house research on smoking and health, it conducted a number of studies on the constituents found in tobacco smoke. In-house researchers also tried to replicate Wynder’s experiments painting mice with tar condensates, and some were successful.114

  Alan Rodgman, another scientist at R.J. Reynolds, drew conclusions similar to Teague’s several years later. Rodgman, who had trained at the University of Toronto in the distinguished lab of Frederick Banting, wrote an extensive paper, “The Analysis of Cigarette Smoke Condensate,” based on experiments done at R.J. Reynolds using standard Camel cigarettes. He explained:In view of this data, it is logical to assume that the carcinogenic activity of cigarette smoke condensate is due to the presence of one or more carcinogenic polycyclic aromatic hydrocarbons.115

  Rodgman proposed that the company try to develop technologies to filter out these substances. When R.J. Reynolds President E. A. Darr learned of this research, he was “somewhat perturbed” that one of his scientists had found benzopyrene in cigarette smoke.116 Rodgman’s findings were clear:Since it is now well-established that cigarette smoke does contain several polycyclic aromatic hydrocarbons, and considering the potential and actual carcinogenic activity of a number of these compounds, a method of either complete removal or almost complete removal of these compounds from smoke is required.117

  But accomplishing this would prove no simple matter. Rodgman, eager to get on with the research, expressed frustration that so-called biological work was to be left to the TIRC, which did nothing. In 1959, in a summary of his work to date, he reported “some thirty-odd polycyclic hydrocarbons,” eight found to be carcinogenic in mice.118

  Some industry scientists not only acknowledged carcinogens in cigarettes but advocated more intensive research into decreasing the health risks of cigarettes.119 One strategy they considered was to identify all the constituents in tobacco smoke, single out any carcinogens, and develop the technical capacity to remove them. As one Philip Morris scientist explained in a memo to company executives:Moreover, because it is not only good business but is our duty to our customers and stockholders, we are dedicated to going beyond any efforts of TIRC, ACS or any other organization. Our objective is to determine all chemical constituents of smoke, and to develop means of removing any which are considered harmful.120

  This strategy, however, entailed risks the companies were unwilling to take. The removal of certain constituents from the smoke was an admission that tobacco might be harmful. This seemed too great a liability, especially since, despite the hype of filters, no clear technology for their removal had emerged.

  Helmut Wakeham joined Philip Morris as a research chemist in 1958 and became director of research and development in 1960. Wakeham, who introduced the use of gas chromatography for assessing the constituents in tobacco smoke, would champion, within the company, the development of a “medically acceptable” cigarette that would be free of the now well-documented carcinogens. He wrote numerous internal memoranda that acknowledged the cancer-causing effects of cigarette smoke. In a memo dated September 22, 1959, for instance, he wrote: “One of the main reasons people smoke is to experience the physiological effects of nicotine on the human system. Nicotine, to the best of present knowledge, does not produce cancer. Hence, in theory one could achieve the major advantage of smoking without the hazard of cancer. But nicotine in tobacco smoke is present in the tar phase.”121 Wakeham became committed to solving this dilemma.

  In a 1961 proposal to investigate the possibilities of reducing carcinogens in smoke, Wakeham listed fifteen carcinogens and twenty-four cocarcinogens, or “tumor promoters,” in cigarette smoke. He also cited the belief that “cardiovascular ailments that may arise from smoking are due to the physiological effects of nicotine,” noting, in particular, nicotine’s “[s]pecific effects on the adrenal medulla, causing it to discharge epinephrine, a hormone which accelerates the heartbeat, contracts the peripheral blood vessels, and raises the blood pressure.” Wakeham also remarked that of the more than four hundred gas and particulate compounds in cigarette smoke, including those specifically recognized as carcinogens, some 84 percent could be found in sidestream, or secondhand, smoke. He concluded:Low irritation and low nicotine cigarettes for commercial exploitation will be developed in the course of our present R & D program during the next two to five years with an expenditure of not more than 25% of the R & D budgets during this period.

  A medically acceptable low-carcinogen cigarette may be possible. Its development would require:

  TIME

  MONEY

  UNFALTERING DETERMINATION122

  At the same time that industry researchers, such as Rodgman and Wakeham, were detailing carcinogenic substances in cigarettes and potential strategies for their removal, the TIRC put out a press release explaining: “Chemical tests have not found any substance in tobacco smoke known to cause human cancer or in concentrations sufficient to account for reported skin cancer in animals.”123 The industry would aggressively exploit public desire for a safe product while simultaneously denying any adverse health effects of smoking.

  In 1958, three scientists—Herbert Bentley from Imperial Tobacco, David Felton from British American Tobacco, and W. W. Reid from Carreras—representing the Tobacco Manufacturers Standing Committee, the British counterpart to the TIRC, visited the United States to assess the status of industry-related science. They met with research directors of major tobacco companies, the Scientific Advisory Board of the TIRC, and its
Industry Technical Committee, as well as other experts in tobacco and disease in the academy and government. They noted that there was virtual consensus among industry researchers that cigarettes played a role in human cancers.

  With one exception (H.S.N. Greene) the individuals whom we met believed that smoking causes lung cancer if by “causation” we mean any chain of events which leads finally to lung cancer and which involves smoking as an indispensable link. In the U.S.A. only Berkson, apparently, is now prepared to doubt the statistical evidence and his reasoning is nowhere thought to be sound.

  Having interviewed prominent scientists at universities, the NIH, and within the industry, the visitors went on to summarize these scientists’ comments regarding the TIRC:In their opinion, T.I.R.C. has done little if anything constructive, the constantly re-iterated “not proven” statements in the face of mounting contrary evidence has thoroughly discredited T.I.R.C., and the S.A.B. of T.I.R.C. is supporting almost without exception projects which are not related directly to smoking and lung cancer. Liggetts [sic] felt that the problem was sufficiently serious to justify large-scale investment by the Company directly in experimental research on smoke and cancer, accepting privately that a strong case against tobacco had been made out and avoiding any public comment until their own research had provided something concrete to offer.

  These British industry scientists believed that a clear consensus on causality had emerged:The majority of individuals whom we met accepted that beyond all reasonable doubt cigarette smoke most probably acts as a direct though very weak carcinogen on the human lung. The opinion was given that in view of its chemical composition it would indeed be surprising if cigarette smoke were not carcinogenic. This undoubtedly represents the majority but by no means the unanimous opinion of scientists in U.S.A. These individuals advised us that although it is not possible to predict unambiguously the effect of any substance on man from its effect on experimental animals the generally successful use of animals in other fields as a model for man fully justifies their use in our problem.124

  This document is important because it reflects the candid assessment of industry officials and scientists. It depicts a broad consensus about tobacco as a carcinogen even among industry researchers, a consensus the industry repeatedly denied publicly through its comprehensive public relations operations. Further, the document shows how marginal individual skeptics like Berkson and Greene had become by 1958.

  In 1961, Hill & Knowlton celebrated its successes on behalf of its tobacco client. The total number of cigarettes sold annually had risen from 369 billion in 1954, the company’s first full year of service to the industry, to 488 billion. Per capita consumption had risen from 3,344 a year in 1954 to 4,025 in 1961, the highest ever. “From a business standpoint,” Hill & Knowlton crowed, “the tobacco industry has weathered this latest spate of health attacks on its products.”125 In less than a decade, the industry had been stabilized and was thriving. As Joseph Lelyveld concluded in the New York Times, “Surprisingly, the furor over smoking and health failed to send the industry into a slump. Instead, it sent it into an upheaval that has resulted in unforeseen growth and profits.” He went on to quote an unnamed American Cancer Society official who claimed, “When the tobacco companies say they’re eager to find out the truth, they want you to think the truth isn’t known. . . . They want to be able to call it a controversy.”126 TIRC, under Hill & Knowlton’s guidance, had turned tobacco science into yet one more political controversy on which people of good will could differ. So long as it could maintain this “liberal” notion of scientific knowledge, the industry remained free to aggressively promote tobacco without regulation or liability. This explains, in part, why the industry would tenaciously cling to the notion of controversy.

  By the early 1960s—in spite of categorical research findings indicating the harms of smoking—a significant “controversy” had arisen over the validity and meaning of these findings. Indeed, given the widespread acceptance of the conclusion, especially among those who had analyzed and evaluated the research most closely, the persistence of debate about the harms of smoking is a striking demonstration of the powerful impact of the tobacco industry’s public relations campaign. The industry insistence, at the direction of Hill & Knowlton, on the notion of “no proof ” and the need for “more research” was an inspired manipulation of the natural tendencies within science to encourage skepticism and seek more complete answers to important questions.

  Hill & Knowlton had served its tobacco clients with commitment and fidelity, and with great success. But the firm had also taken its clients across a critical moral barrier that would have two important impacts on American society. Trust in science, confidence in the media, and the responsibility of the corporate enterprise were all substantially harmed by Hill & Knowlton’s efforts on behalf of the tobacco industry. By making science fair game in the battle of public relations, the tobacco industry set a destructive precedent that would affect future debate on subjects ranging from global warming to intelligent design. And by insinuating itself so significantly in the practice of journalism, Hill & Knowlton would compromise the legitimacy and authority of the very instruments upon which they depended. The tobacco industry’s PR campaign permanently changed both science and public culture. It changed the character of public knowledge in dangerous new ways. So long as Hill & Knowlton and the tobacco industry could cling to their particular forms of denial and reshape them into a proactive public relations program, the industry had bought not only critical time, but a new generation of smokers.

  Hill & Knowlton could point to many successes on behalf of the industry. In June 1961, the New England Journal of Medicine published an exchange between Ernst Wynder and C. C. Little. Regardless of what was written, the very existence of such an exchange was a victory for the industry. The authors in their respective articles laid out their well-worn positions. Wynder reviewed, yet again, the substantive evidence demonstrating conclusively that smoking posed a serious and lethal risk to human health. He explained:I know of no other chronic disease that has been studied epidemiologically and statistically in such detail and with such uniform results as smoking and lung cancer. The tobacco industry would like to give the impression that scientific evidence is divided on this issue. I do not consider this to be the fact. . . . A number of reputable scientists have testified that they do not believe smoking to be a cause of cancer of the lung, but few of them have done research on the subject under discussion. It is true in virtually any field of human endeavor that one finds people who will take an opposing view to any proposed evidence. This was true in the days of Semmelweiss, as well as in the days of Pasteur, and in previous centuries just as it is true today.127

  The persistence of a few skeptics without direct research experience in the field, Wynder argued, should not detract from the overwhelming scientific consensus concerning the relationship of smoking and disease. In his conclusion, he let his frustration with such critics show:Of those who will not accept existing evidence, I should like to know what evidence would be acceptable. If one criticizes epidemiology for being statistical, if one criticizes animal research for being unrelated to the human problem and if one criticizes chemical identification of carcinogens as not have any bearing to human disease, I should like to ask if there is a form of evidence that would be accepted as being conclusive.

  If it were humanly possible we would at once set up a study that could yield such evidence. If it is humanly impossible, it is not a constructive kind of suggestion that would advance scientific knowledge.128

  In his response, not surprisingly, Little refused to answer Wynder’s question.

  To expect, as Dr. Wynder and some others do, that those still unconvinced should state the exact and specific evidence that would “convince” them is being completely unrealistic. If one could define such specific evidence the problem would be already solved.129

  As he had so often done, Little reiterated “that much more research in depth is nee
ded before definitive answers will be available.” He concluded:Lung cancer, indeed all cancer, is a challenge, an unsolved problem. Its etiology will probably long be an open question. As such, greater co-operation and exchange of ideas among experimenters, statisticians and clinicians are the goals to be focused on and to be attained.130

  Along with these two articles, the New England Journal ran an editorial written by Joseph Garland entitled “The Great Debate.” “Both authors,” Garland wrote, “are dedicated, sincere proponents of their points of view, each upholding what he believes to be the truth and nothing but the truth, each ready to admit that the whole truth has not yet been revealed to aspiring man.”131 In the end, Garland left the question to the individual judgment of physicians and the public:Each individual must choose his own course, whether to woo the lady nicotine or abjure the filthy weed, while the search for truth continues.132