The Blackman Paper: Rewriting Scientific Documents
The extent to which lawyers for B&W were concerned about having damaging scientific statements linked to the company is conveyed in a letter dated October 25, 1984, from J. Kendrick Wells, III, B&W's corporate counsel, to H. A. Morini, a lawyer at BAT {1833.01}. The letter contains Wells's comments on a draft paper titled "The Controversy on Smoking and Health—Some Facts and Anomalies," which had been written by Dr. L. C. F. Blackman, executive director of research and development for BAT (figure 9.1) {1833.02}. The paper presents the industry's view on the "controversy" over the health effects of smoking. As it was originally written, the Blackman paper contained a reasonably complete presentation of the evidence that smoking causes disease, and then used quotes from various scientists and scientific reports to support the claim that a causal link between smoking and disease had not been proven. However, as Wells's letter shows, even merely acknowledging the existence of some evidence pointing to such a link was unacceptable.
The title page of the Blackman paper indicates that it contains "Notes on talks given at the BAT Management Centre, Chelwood, by Dr. L. C. F. Blackman" {1833.02}, suggesting that the paper was intended to be a primer for BAT executives on issues related to smoking and health. The introduction states:
This booklet is not a comprehensive review of all the research on the issues, but it sets out some of the reasons for stating that a controversy exists, and gives examples of some of the research reports which are inconsistent with the view that smoking has been proven to be a cause of disease. {1833.02, p. 3}
Wells made comments on virtually every page of Blackman's thirty-three-page paper, some of which are written in the margins of the draft. He also made forth-five detailed comments, including line-by-line recommendations regarding items that he thought should be deleted or included in the final report. These comments were too long to be written in the margins and are instead enumerated in his letter to Morini. Many of the suggested changes involve deletion or reconsideration of scientific authors cited in the paper, usually because the scientists cited had
published work (sometimes not even cited in the paper) that did not support the tobacco industry's position. In other instances Wells suggests that material be rewritten so that it will conform to the company's stated positions, rather than the published scientific evidence.
Wells's comments on the draft paper are prefaced with a cautionary note:
Recent developments have reaffirmed the need for the attention we customarily have given to proposed BAT publications. The smoking and health litigation in the U.S. has demonstrated that plaintiffs' lawyers are aggressive in questioning tobacco CEOs about published company statements, as we had predicted they would be. Peter Taylor's Smoke Ring [1] demonstrated that BAT publications which may be intended for limited distribution can be obtained and scrutinized by our most articulate adversaries. {1833.01, p. 1}
This comment suggests that attorneys for B&W routinely reviewed BAT documents prior to publication, even if they were only intended for "limited distribution" within BAT.
Wells began by changing the titles of some of the chapters (figure 9.1). Some of his revisions appear to have been designed to reinforce the idea that there is a "controversy" over the dangers of smoking. For example, Wells recommended changing the chapter on "Background to the Medical Concern" to "Background to the Scientific Dilemma." Interestingly, Wells also recommended deleting the word "facts" from two of the chapter titles: "Some Facts and Anomalies in the Literature Regarding Cancer" was changed to "Some Anomalies in the Literature Regarding Cancer." A similar change was made in the chapter on heart disease.
Wells seemed particularly interested in removing references to researchers who had published any findings that did not agree with the tobacco industry's position. For example, the original document contained the following quote from Sir Richard Doll and Richard Peto, who conducted one of the first epidemiological studies showing that smoking is associated with lung cancer and heart disease:
To say that these conditions were related to smoking does not necessarily imply that smoking caused (or prevented) them.
The relation may have been secondary in that smoking was associated with some other factor, such as alcohol consumption or a feature of the personality, that caused the disease. {1833.02, p. 8}
In his letter to Morini, Wells suggests that this reference be deleted:
8. Delete Doll and Peto reference. Doll and Peto have published a table which shows "cancer of the lung" is "caused by cigarette smoking" and have
FIGURE 9.1. Cover and table of contents for the Blackman report {1833.02}, which was based on talks
given at the BAT Management Centre by Dr. L. C. F. Blackman. Note that the editing by the lawyer
J.K. Wells {1833.01} changed "medical concern" to "scientific dilemma," deleted the term "facts,"
and emphasized the use of the term "causality."
FIGURE 9.1. (continued)
concluded that "much of the excess mortality in cigarette smoking can be attributed with certainty to the habit ..." {1833.01, p. 2}
Later in his comments, Wells notes that any mention of Doll should be handled with care.
35. ... Parenthetically, any reference to Doll must be crafted carefully because he is a dedicated advocate of the causal hypothesis. {1833.01, p. 6}
Wells also recommends deleting a reference to research conducted by Drs. Friberg, Cederloff, and Lundman, who had compared disease rates in smoking and nonsmoking twins. Interestingly, this research had been funded by the tobacco industry for years in an attempt to prove the "constitutional hypothesis." The constitutional hypothesis, which had originally been proposed by Sir Ronald Fisher in 1957, essentially argues that some people are genetically predisposed both to being the type of person who smokes and to developing diseases such as lung cancer and heart disease. Friberg and his colleagues all but disproved the constitutional hypothesis when they showed that a person who smokes is more likely to develop smoking-related diseases than his or her nonsmoking identical twin, even though they have the same genetic makeup and therefore the same genetic predisposition to disease. The Friberg group's work was referred to in the original draft of the Blackman paper in a quote taken from the 1979 Bibliography on Smoking and Health (published by the federal government):
The results from the twins study clearly demonstrate the importance of genetic, behavioural and psychosocial factors which have not been considered in conventional epidemiologic studies. {1833.02, p. 18}
Wells states in his letter that the reference to Friberg, as well as the reference to the bibliography, should be deleted.
21. Recommend delete references to Friberg and the Bibliography on Smoking and Health. Unfortunately, Friberg, Cederloff and Lundman published a monograph in 1977 which stated that "lung cancer is closely related to the amount smoked ...," that associations were confirmed between smoking and respiratory, cardiovascular and other symptoms of disease or ill-health, that there is no doubt about a causal link between smoking and lung cancer, and that the results from the Swedish monozygotic twin studies are contrary to the constitutional hypothesis advanced by Fisher.
The problem carries over into the quote from the Bibliography on Smoking and Health. ... The same article states that "the well documented evidence of a causal association between smoking and lung cancer found in other studies has been further supported." {1833.01, p. 4}
Ironically, the monograph of the Friberg group's work mentioned by Wells was actually funded as a CTR special project (table 8.1).
A quote from Dr. Lawrence Garfinkel, vice president for epidemiology of the American Cancer Society, and an author of one of the early papers on environmental tobacco smoke (ETS) and lung cancer that had produced equivocal results (2), was included in the draft:
Passive smoking may be a political matter, but it is not a main issue in terms of health policy. {1833.02, p. 31}
The tobacco industry had widely used this quote in its efforts to create controversy about the early evidence that ETS causes lung cancer (see chapter 10, particularly figure 10.1 and the associated discussion). Wells, however, notes in his comment that Garfinkel later retracted this statement after his second study on ETS and lung cancer showed an increased risk (3) and criticized the tobacco industry for continuing to use Garfinkel's first study in industry advertising.
40. Delete. Dr. Garfinkel has published a letter in the New York Times (June 5, 1984) which protests R. J. Reynolds' use of the passage quoted in the Draft and states:
It is irresponsible of Reynolds to attempt to create a false sense of security about the potential dangers of passive smoking, especially at a time when incriminating evidence continues to accumulate. {1833.01, p. 7}
Wells left unchanged quotations from an international symposium in Vienna in 1984, which was organized by an individual affiliated with the tobacco industry (4) and exonerated ETS as a cause of disease.
In his letter to Morini, Wells even recommends deleting a short subsection of the paper dealing with chronic obstructive lung disease (COLD). The subsection presented a table from the fourth report of the Royal College of Physicians. The data in the table suggested that COLD might be associated with social class. However, Wells notes,
26. ... The same paragraph in the fourth Report which presents the table reprinted in the Draft states "it must be emphasized that at present only the effects of tobacco are reliably known to be of substantial importance." {1833.01, p. 4}
Table 9.1 (on p. 377) summarizes some of the changes suggested to the Blackman paper. These editorial changes would have effectively removed any reasonable presentation of the then-current state of scientific information from the paper and turned it into a purely public relations piece. Given that the paper was apparently prepared for "limited
distribution" within BAT to review the "smoking and health controversy," it is remarkable that B&W evidently did not want its employees to be presented with arguments contrary to the company position, even when those arguments appeared in a document designed to discredit mainstream scientific opinion.