Corporate social responsibility (CSR) programmes have emerged in response to public pressure for corporations to uphold ethical, environmental, health, and labour standards.¹ Given the negative public perception of the tobacco industry and its products, it is not surprising that two tobacco companies, British American Tobacco and Phillip Morris, have developed CSR initiatives.^2,3 British American Tobacco claims that its CSR programme works with community and governmental stakeholders to “establish dialogue with its critics”.² In 2004, Hirschhorn described the motivation behind Philip Morris’s CSR programme as a measure to regain public and investor confidence and, in examining tobacco company documents, found that internal motivations and statements were similar to those expressed publicly.⁴ Some of the objectives of the Philip Morris CSR programme relevant to this paper are as follows:


 “Help Reduce Youth Smoking…Market Our Products Responsibly …Communicate the Health Effects of Our Products…Support Reasonable Regulation…Comply with Legal and Regulatory Requirements…Provide Shareholder Return.”⁴

Many of these goals are congruent with those of the public health community; however, response to these programmes has not been positive. The World Health Organization in 2003 suggested that corporate responsibility for tobacco companies was an “inherent contradiction”.² While issues with the specifics of CSR policies such as promoting ineffective or counter-productive programmes have been noted, the WHO report argues that the selling and promotion of tobacco is irreconcilable with corporate responsibility. Similarly, in 2002 Warner suggested that an appropriate response for a social responsible tobacco company entailed relinquishing many of the decisions concerning distribution and promotion of the product to the government or other bodies.⁵ Essentially, adequate social responsibility appears to entail abandoning the right to sell tobacco and, in fact, Philip Morris considered at one point withdrawing from selling tobacco entirely.⁶ Nevertheless, Philip Morris continues in the tobacco business, and British American Tobacco still promotes its CSR programme.

Responsibility

Collin and Gilmore suggest that the general proposition underlying corporate social responsibility is the belief that corporations should address their environmental and social impacts, voluntarily developing practices that exceed merely legal compliance.⁷ CSR policies develop as an attempt to resolve social issues and concerns surrounding the effects of its business practices. The popularity of CSR among major companies has been attributed to its perceived ability to forestall legislation and its perceived financial advantage in appealing to “ethical investors”.

Zadek describes five stages of corporate responsibility: denial, compliance, managerial, strategic, and civil.¹ In this model, corporate responsibility is a process in which companies act in response to emerging social issues in a stepped process. Initially, corporations react by refusing to engage with the issue and questioning the legitimacy or existence of the issue itself.¹ For example, when Nike was originally criticised for its labour policies, the first reaction was to deny that a problem existed.¹ In the compliance and managerial stages, a corporation will strictly comply with the regulations, but will not attempt anything beyond the letter of what is legally required. When merely complying with old values becomes insufficient, corporations may then take an active interest in engaging in policy issues beyond the legal requirements, either to forestall legislation in the strategic stage, or to work actively with other businesses and social bodies in the civil stage.

In addition to the stages of responsibility are the stages of pressure (latent, emerging, consolidating, and institutionalised) on a company to take action based on the maturity of the issue. Zadek also described a scale developed to measure the maturity and public expectation of social issues.¹ In the latent stage, the issue is driven by activist pressure with weak scientific evidence and is largely ignored by the business community. As research evidence builds and political awareness develops, the emerging stage occurs, leading some businesses to develop approaches for handling the issue. During the consolidating stage, litigation takes place, and pressure builds for legislation and regulation. In response, companies develop voluntary standards. Finally, in the institutionalised stage, legislation occurs or business norms are established.

During the latent stage, companies are able to deflect responsibility, but as the maturity of the issue increases, companies must increase their responsibility to avoid risk (fig 1). Being ahead of an issue provides gains for the business in terms of reputation, while delay could entail serious risks, particularly if other companies have adopted new standards. If a company is unable to meet the operational standard, it might find itself in a position where it could no longer compete profitably.

Trial testimony

In the 1990s, litigation became a major concern for the tobacco industry, particularly in the United States.^8,9 In 1994 and 1995 four states—Mississippi, Minnesota, Florida, and West Virginia—filed lawsuits against the tobacco companies for reimbursement of medical expenses, and by 1997, 41 states and two jurisdictions had filed. Under this wave of litigation, a group of state attorney generals developed a Master Settlement Agreement to settle the cases brought by government bodies. In the civil court cases brought both by governments and other parties, tobacco company representatives provided testimony covering the basic issue of responsibility for tobacco and the harms associated with its use. This body of trial testimony provides a valuable resource to examine how tobacco companies respond to public tobacco policy concerns.

It is important for crafting future tobacco control legislation and programmes to understand the underlying arguments concerning changes in responsibility for tobacco control policy. Although CSR programmes, themselves, are a relatively recent phenomenon, CSR reflects the larger issue of how corporations respond and adapt to social and political pressure. This paper examines the arguments presented by the US tobacco industry at trial concerning responsibility for tobacco control policies. However, the purpose of this paper is not to look at legal culpability, either within the framework of individual trials or in the general sense of liability for the actions of the tobacco industry. Instead, we examine how their stated degree of responsibility is associated with the stages of corporate responsibility described by Zadek.¹

METHODS

Design

This study employed a qualitative iterative design to investigate tobacco control policy perspectives in trial testimony collected in the Deposition and Trial Testimony Archive (DATTA) dataset. DATTA contains transcripts from tobacco-related litigation collected from a variety of sources by the Center for Tobacco Use Prevention and Research in Okemos, Michigan and contains more than 800 000 pages of testimony.

Tobacco control policies

Definitions for tobacco control policy were developed by DATTA coding teams to refer to policies, activities and regulations enacted by governments or other bodies with the purpose of reducing the harm caused by tobacco use. The following policies were identified: in the testimony were health information (health warnings, tar and nicotine information, listing of cigarette ingredients, Food and Drug Administration regulation of cigarettes, advertising regulations, promotion of public awareness of smoking and health issues), second-hand smoke policies, and youth smoking policies (prevention programmes access legislation).

Hypothesis generation

The initial step used open coding with a broad selection of trial testimony concerning policy as determined independently by coders associated with the DATTA project who identified a selection of testimony related to tobacco control policy. A small random sample of 25 documents from this DATTA selection were extracted and reviewed by the authors to generate hypotheses associated with the discussion of tobacco control policies in US trial testimony. This process was repeated twice, refining and adapting the hypotheses found in the first reading. From this, a main hypothesis was generated concerning industry responsibility for tobacco control policies, specifically that the level of responsibility for a tobacco policy issue as declared by representatives of the tobacco industry was associated with the level of pressure on the company to manage the issue.

Selection criteria for subjects

Testimony was gathered from individuals selected for their ability to represent the policy stance of their tobacco company. In order to undertake a close reading of testimony and because of the amount of repetition and message consistency found in the previous reading, a restricted sample of testimony was selected by choosing an appropriate representative from each of the major American tobacco companies. A universal sample of all tobacco company executives was not chosen because specific individuals were there as representatives to present the views of their company, and senior representatives who were responsible for tobacco policy positions were of interest. By collecting the complete available testimony from an individual, a coherent narrative could be developed in a way that would not be possible from the analysis of random documents. As such, these key individuals chosen were thought to have the greatest potential for contribution to the theory.

Selection criteria required the inclusion of a senior staff person or CEO from each of four major US companies: Brown & Williamson, Philip Morris, RJ Reynolds, and Liggett. Less senior staff or staff involved in the scientific division might be less representative and less knowledgeable about the company’s corporate policy. The individuals must have had clear policy responsibilities and must not have been specifically associated with the scientific divisions of the industry. They must have testified in defence of the tobacco industry; been authorised to represent the views of the industry or specific company; and have testified multiple times in a variety of trials, including at least once before the Master Settlement Agreement announcement in June of 1997. A list of individuals who had testified was sorted by the number of testimony appearances. The individual with the greatest number of appearances who fulfilled all criteria was selected to represent each US tobacco company.

Subjects

Four witnesses, identified from the DATTA collection of trial testimony, fit the selection criteria: Ellen Merlo (Philip Morris), Nicholas Brookes (Brown & Williamson), Bennett LeBow (Liggett), and Andrew Schindler (RJ Reynolds). A total of 42 documents were collected, comprising over 4000 pages of testimony, from 1984 to 2003 (table 1).^10–80 Brookes, LeBow, and Schindler were chief executive officers of their companies and Merlo held various positions including vice-president responsible for policy. Each was able to speak authoritatively on the position of their respective tobacco companies, but on occasion spoke as individuals.

Data analysis

All available testimony from each witness selected was collected and imported in text format into WinMax, a qualitative data program. As electronic search methods were found to miss a substantial number of references to tobacco control policies, a hand search of the documents for terms relating to tobacco control policies was conducted. All trial documents were read through and policies were coded where they appeared. When a discussion of tobacco control policy was encountered, our interpretation of the perceived stage of responsibility (denial, compliance, strategic, civil) and stage of issue maturity (latent, emerging, consolidating, institutionalised) for the tobacco control policies was coded. The coded sections were organised by policy area, and a narrative was generated relating the categories and their relationships.

RESULTS

Health information

Discussion on responsibility for health warnings centred on the imposition of the US Surgeon General’s warning on cigarette packages with the Public Health Cigarette Smoking Act of 1969. The position of the tobacco industry was that this act created a strict obligation to provide the health information required by the Surgeon General and only that information. This position was upheld to various degrees, particularly in the 2003 trial where the judge ruled that the tobacco companies could not be questioned on the warnings and could not be held liable for failure to warn.⁶⁹

The precedent of the congressionally mandated warning allowed the American tobacco companies to argue that these warnings pre-empted additional warnings or messages; that their responsibility for action had been removed. The witnesses claimed that they were not responsible for adding information or warnings because these were not legally required and, in fact, argued that they could not stray from the legal requirement. In 1997, Brookes stated this pre-emption explicitly:


 “Q…has Brown & Williamson voluntarily added anything in addition to those warnings that are required by the federal government?
 A: …We haven’t, no, but I’m not quite sure whether we’re legally entitled to. But I think there is a case that we are pre-empted in some way by federal legislation.”¹⁰

Brookes, Schindler, and Merlo argued that it was the role of government or a similar authority to determine the institution and content of the warnings. For example, when asked if RJ Reynolds would have voluntarily sought out the Surgeon General for input on warnings if it were not mandated, Schindler suggested that the institution of warnings was an inevitability:


 “I’m saying that the process that existed in our society, given the issues surrounding smoking, would work its way through the total society in terms of social policy and that that’s how you would end up with warning labels, which is what happened, you know, 30 some years old. I think that process would happen again.”⁷¹

Similarly, information on tar and nicotine deliveries, ingredient disclosures, and advertising regulations have been institutionalised since the US Federal Trade Commission (FTC) began posting regulations starting with the FTC cigarette advertising guides in 1955. Here, industry representatives argued that their responsibility was solely to abide by any regulations that might be in effect. They contended that because no regulations required disclosure beyond what was submitted to the government, there was no responsibility to inform consumers further. These government standards created an obligation to fulfil, whether or not the public health community felt that this information was useful, sufficient, or counter-productive. The tobacco companies claimed that they had no choice or responsibility for the issue beyond a passive response to the consumer, the government and other authorities and that claims of risk and safety could not be assigned to the tobacco industry. Brookes attributed the light and mild health claims to “the U.S. and the U.K. Public Health authorities”.¹⁴ Schindler suggested that both the determination of risk and the tobacco company response of offering lower tar and nicotine cigarettes were in response to consumer and institutional demands:


 “As to whether or not that product is less risky than a higher tar product, I don’t know. I know that it has less tar, which is an area of focus that the public health people have had for years.”⁷¹

Relying on the authority of the Surgeon General allowed the tobacco companies to avoid discussion of issues of responsibility. There was an official position, and these legislated norms were all that was necessary to comply with the settled issue. This suggests that industry understood the smoking and health issue as a being in the institutional stage. While industry personnel may have had their own beliefs—for example, on the health benefits of low tar cigarettes—the public position was to allow the Surgeon General and other health officials to be the only voices on this issue. Consequently, in their view, the Surgeon General was accountable for that information and for the truth of the situation. For example, Merlo stated, in court in 1997, that smoking and health information was the responsibility of “appropriate health authorities”.⁶³ Schindler, similarly, suggested that the tobacco company bore no responsibility for the truth of the health warnings. When asked if the warning was “true”, Schindler responded:


 “Well, Congress working through the Surgeon General, public health people and Congress said this is the law, this is the warning label to go on the packs, which says that the Surgeon General says that cigarettes cause these diseases.”⁷¹

Consequently, it was also not the responsibility of the company to research questions relative to the health effects of low tar cigarettes. Schindler contrasted the internal research of Reynolds, suggesting that the external community held responsibility for that question:


 “I don’t know of any research that Reynolds is doing relative to that question. I believe that there are people in the public health community or medical researchers that are, I imagine, are looking at that.”⁶⁹

In the late 1990s, a shift occurred in the stated arguments regarding the responsibilities to inform consumers. The catalyst came during the discussions leading up to the Master Settlement Agreement, when Liggett added an addiction warning to their cigarette packs.³⁹ In the terminology of Zadek,¹ Liggett acted as the “first mover,” the leading business that demonstrates that action on an issue is possible. In the years since warning and disclosure requirements were institutionalised, the issue had re-emerged with a re-evaluation of the evidence and the appropriateness of the wording to warn consumers adequately of the health effects of tobacco. The stage of pressure in terms of the maturity of the issue had moved from latent to emerging. When Liggett took action, that action required the other tobacco companies to recognise the risk involved in not addressing the issue of health information.

Consequently, at this point Brookes,²⁴ Schindler⁷⁵ and Merlo,⁵¹ while continuing to maintain pre-emption of the mandated warnings on the labels themselves, began introducing new voluntary measures. In 2000, for example, Merlo at trial demonstrated a website that linked directly to the health statements of the Surgeon General, without conceding any change in personal or corporate beliefs:


 “... [The] website encourages people to rely on the information that is contained in those links in making their decision about smoking… We’re not saying that it’s right or it’s wrong.”⁵¹

The voluntary activities included adding new information to their websites, stating explicitly that smokers should rely on the Surgeon General and other public authorities for information on smoking and health, and adding information to advertisements beyond the obligations of the FTC. For example, Brookes described a contemporaneous advertisement:


 “Well, by agreement with the Federal Trade Commission, we are required to, as I said earlier, put the tar and nicotine averages per cigarette on all our advertising. So that’s the first part of that. That’s as required by the Federal Trade Commission.
 The next part is our own voluntary information we’ve added to our advertising. I think you can see it says: Actual deliveries will vary based on how you hold and smoke your cigarette.... We thought it would be the responsible thing to do to actually alert smokers in our advertising that it didn’t necessarily mean safer”¹⁵

When adding voluntary information on ingredients to their websites, Brookes, Merlo and Schindler presented this disclosure as a voluntary act outside of any regulatory standard. It was specifically represented as an action taken that was beyond the legal requirement, unlike the direct compliance stance taken earlier.^13,69 In contrast to the stance taken in the compliance stage, the voluntary activities reflected a move to the strategic stage that levelled the playing field among US companies by attempting to create industry norms of acceptable behaviour rather than risk future legislation or litigation. While the initial issue of having warnings had been settled decades earlier, fresh concerns had emerged about the efficacy of those warnings, and the stance taken by the witnesses addressed this new set of concerns.

Second-hand smoke

Unlike health warnings, which had been regulated for decades, second-hand smoke was an issue where regulation was ongoing and incomplete. Here, the witnesses expressed opinions that reflected the denial stage of responsibility. There was no acceptance by the cigarette manufacturers of the authority of bodies like the Surgeon General and the 1986 “Health consequences of involuntary smoking”. Instead, Merlo and Schindler would refer repeatedly to their reliance on their internal scientists and would debate the scientific accuracy of claims from the independent scientific literature. Schindler of RJ Reynolds suggested that company scientists were responsible for developing his position, and the company’s position, on second-hand smoke and stated that “the epidemiology, as I understand it, and talk with our scientists, does not establish [second-hand smoke] as a risk factor”.⁷¹ Merlo of Philip Morris was also dismissive of research evidence from public health authorities:


 “There have been Surgeon General’s reports and there have been reports that have dealt with second-hand smoke in the past, yes. I don’t believe they were any more based on scientific evidence than the EPA [Environmental Protection Agency] report.”⁶²

However, Merlo denied that the reliance on internal scientists and the attempts to discredit the science on second-hand smoke were attempts to alter perceptions of the risks of second-hand smoke. As she states: “We do not try to convince [people who believe environmental tobacco smoke (ETS) is a health risk] to change their minds.”⁵⁰ However, the defensive posture of the witnesses was consistent with the denial stage—an attempt to defer responsibility for dealing with the health effects of second-hand smoke.

In the 2003 Allen v. Philip Morris case, in which the judge had restrained discussion concerning health warnings and knowledge of health effects due to the congressionally mandated warnings, the difference between the responsibility taken for health information and that for second-hand smoke was a source of frustration for the plaintiff’s lawyer. Whereas the companies acknowledged the Surgeon General’s findings on the health effects of cigarette smoking, they would debate the scientific basis for the health effects of second-hand smoke. As the lawyer stated:


 “We get into this issue of second-hand smoke, where it’s our position that the CEO’s have said they will not debate public health issues, but on the issue of second-hand smoke they do.”⁷⁶

With respect to second-hand smoke, the tobacco industry remained in the denial stage of responsibility. For example, voluntary programmes of accommodation, which usually mark the strategic stage, were not characterised as health issues but rather as issues of etiquette. The industry claimed they were not responsible for harm to passive smokers, but they did feel responsible for the “rights” of smokers, attempting to “ensure that adults who ch[o]se to smoke could be accommodated”.⁶⁰

However, the goal of the accommodation programme did have strategic implications in avoiding government regulation, as the voluntary programmes instituted by the tobacco companies may have been meant to pre-empt further action by government. Accommodation legislation was promoted in jurisdictions where regulation was threatened as an attempt to fight more “restrictive legislation”.⁶¹ Each of the major tobacco companies supported the accommodation programme, which was described by Schindler as involving good manners and social graces and not health issues:


 “Let’s have dialogue, let non-smokers and smokers try to get along without government intervention…that’s what I mean when I call it one of Reynolds’s accommodation ads.”⁶⁹

Youth smoking

The strategic and civil stages of responsibility are seen in the industry position on youth access legislation. Brookes of Brown and Williamson touted their wide-ranging support for both legislation and their own corporate responsibility to prevent youth smoking: hiring a youth responsibility vice-president, funding youth smoking programmes, and so on. Brookes recognised that “getting involved in trying to keep kids from smoking is obviously full of pitfalls, and we are looking to avoid them”.¹⁵

Denial stage aspects of responsibility were evident, however. Merlo from Philip Morris argued that youth access was not entirely their responsibility, because tobacco companies did not sell cigarettes to kids. Rather, the focus lay on the retailers who actually did the selling because “What goes on at retail between a consumer and a retailer is not an issue in the chain of responsibility”.⁶⁹ The concept of a diffuse network of responsibility was consistent among witnesses, even when discussing action on preventing youth smoking. Apart from retailers, Merlo also cited parents and older siblings as directly responsible for youth gaining access to cigarettes:


 “…Work with parents and older siblings to make sure that kids can’t get access to cigarettes…. so we are trying to ask adults to take some responsibility in the way they deal with this issue.”⁶⁸

This argument regarding the responsibility of the family or the retailer was less prominent in Merlo’s later testimony. After 2000, the stance of Philip Morris, as stated by Merlo, was to actively seek responsibility for youth—in other words, to “Do our part to proactively identify and capitalize on opportunities to discourage underage use of our products”.⁵⁰ This position was uncategorical in declaring youth smoking a problem and an issue that had to be dealt with appropriately: “We do not want children to smoke, not only because it is illegal to sell cigarettes to minors in every state, but also because of the health risks of this product.”⁶⁹ As with Merlo and Brookes, Schindler of RJ Reynolds was also clear that the youth smoking programmes were intended to prevent children from smoking. Beyond preventing children from obtaining cigarettes, programmes were meant to “actively convince people, convince kids, not to smoke”.⁶⁸

If companies could effectively address the youth smoking issue, it might be possible to move past the other areas in which they felt they were being attacked. As Merlo states: “In order to be a responsible manufacturer and marketer of an adult product that contained risk, we had to deal with the issue of youth smoking prevention.”⁵⁰ For her, addressing youth, defined a way of selling tobacco. There was an unacceptable way—that was selling to youth—and there was an acceptable way—selling to adults. This division created a space for a “responsible company” to make and market a dangerous product.

Moreover, taking responsibility for youth issues allowed the witnesses of the tobacco companies to assert that they could start to move beyond the conflict that was limiting their stability in the marketplace. The pressure of public opinion was felt, and the focus on youth was an attempt, Merlo hoped, to isolate the negative connotations of youth smoking from that of smoking in general:


 “The public did not believe us on many issues, that they considered that we were out of step with society expectations of what a responsible company should be and where they should stand…we felt that if we could effectively address the issue of youth smoking prevention and come to some kind of agreement, we could move past that.”⁵⁰

By becoming, or presenting themselves as becoming, equal partners in handling the issue of youth smoking, Merlo’s statement reflected the civil stage of responsibility. The maturity of the health consequences of smoking for both adults and youth required that substantial action be taken, and embracing the issue of youth smoking reinforced the idea of smoking as an recognised adult behaviour.

DISCUSSION

Corporate responsibility reflects attempts to address political and public concerns about the effects of corporate policy while protecting core business activities. In a courtroom setting, corporate responsibility is evident in the responses of Brookes, Merlo and Schindler to tobacco control policy issues. The opinions expressed concerning health information suggested that the compliance stage shifted to the strategic stage of responsibility in response to the movement of political pressure. With respect to second-hand smoke, the relative immaturity of the issue allowed the companies to pursue denial stage responsibilities, while the maturity of the youth smoking issue required the tobacco companies to pursue strategic and civil responses.

The witnesses argued that the responsibility for tobacco and tobacco policy rests with various actors, including the manufacturers, government, public health authorities, and individuals. In the 1990s, with increasing and increasingly successful litigation, the issue of responsibility rose to the forefront as the tobacco industry attempted to define and limit what those responsibilities were. This response is consistent with other examinations of the tobacco industry to public policy: critics contend that in some cases voluntary programmes exist to prevent the implementation of more restrictive, and more effective, legislation^81–83 while in other cases, the industry has argued in favour of certain types of legislation, particularly legislation that would “pre-empt” future restrictions.^84–86

Legal status

By examining changes in the stages of responsibility for tobacco control policies, we can look at the levels of responsibility as a way of protecting the core business of the industry, that of selling tobacco. When the industry is threatened, they shift to protect their legitimacy as manufacturer and marketer of tobacco. While improving corporate image has other benefits, such as improved share prices and respectability,^2,87 both regulation and voluntary action that does not completely remove the agency of industry serves to re-enforce the ability to continue to stay in business. For example, Nike, when under pressure for using developing world labour, moved voluntarily to improve the labour conditions of the workers that it used. Once accepted, this move protected Nike’s ability to continue to use developing world labour and protected its ability to continue selling shoes at a profit. Unlike Nike, however, the core social issue for tobacco companies relates to the product itself and not the manufacturing process.

However, the larger issue is about who has authority—the tobacco companies, government, or public health officials—to control tobacco. The dance between regulation and voluntary action relating to tobacco control policies hides the “elephant in the room”: the authority of the government to allow tobacco to be sold as any other commercial product. The status of tobacco as an acceptable commercial product gives implicit permission to the tobacco companies to engage in the distribution, sale, and promotion of tobacco, and provides important principled standing for the industry.

Industry youth access programmes have been characterised as ineffective and counter-productive. Whether or not the programmes are successful, banning youth smoking while allowing unrestricted adult smoking implicitly condones adult smoking. Reducing youth smoking does not inherently mean that tobacco use would die out. Even if youth smoking prevention programmes were 100% successful, the industry would be able to trade a loss in the overall size of the market in return for increased stability in the acceptability of adult smoking. When Liggett made its deal with the government, the dynamic of protecting the business was at work. Essentially, Liggett traded all responsibility for tobacco control policy in exchange for the continued ability to sell tobacco. Just as Nike was willing to make accommodations in its labour policies to protect its core business, similar accommodations made by the tobacco companies are also aimed at protecting the core business. Tobacco control policies were sometimes characterised as threatening the core business. See how Schindler characterises a number of tobacco control strategies:


 “I think all these things that are described in here, extremely high levels of taxation, the total banning of smoking, the correlations or attempting to correlate cigarette smoking with heroin...represent the attempt or inclination towards a backdoor prohibition...directed toward some form of criminalization of the selling of the product.”⁶⁹

As tobacco control policies are seen as attempts to undercut the ability to sell tobacco, policy responses are directed toward preserving a role for industry in the commercial distribution of tobacco to users. Voluntary action makes this explicit, but even governmental regulations and law provide standing for the industry. Formal and informal CSR programmes create an area in which the industry is protected, even if this area is increasingly smaller. Schindler felt safe in making statements such as “If the day comes when society doesn’t want cigarettes to be sold anymore, then so be it”⁶⁹ because the legal status of tobacco is engrained within the current regulatory and rhetorical framework.

Limitations of interpretations of results

Although we have limited our testimony to four individuals, the major tobacco companies are represented, and there was strong consistency of argument among Merlo, Schindler, and Brookes. This suggests that additional testimony would not have altered the general findings. There was little to distinguish the positions of the representatives of each company with respect to tobacco control policies. Nevertheless, generalisation of specific company policy beyond the interpretation of the individuals included is limited.

Trial testimony, itself, is limited as a source of information concerning policy and practice. Testimony is gathered for a specific purpose, and this paper attempts to use this testimony to make larger generalisations. Furthermore, the adversarial nature of a trial limits the actors to statements that may be more oppositional than otherwise might be expressed. The value of examining testimony can be more apparent when examining the arguments around a topic, rather than as a source of factual information, as testimony may not match statements outside the courtroom. Other resources and documents would be better suited to developing a historical narrative of actual events and activities of the tobacco industry. Research is needed to compare the accuracy of testimony in a trial to those positions represented by internal company documents and those positions presented publicly. Statements in a trial are also more constrained than they otherwise might be and may be more defensively oriented to limit liability.

All testimony is drawn from the perspective of the US tobacco industry. The split between US tobacco companies and their international counterparts suggests that the United States is seen as a unique market with distinctive risks and modes of operation.⁶ Consequently, conclusions drawn from the American industry may not be applicable to the international tobacco market. However, the principles of corporate responsibility reflecting public pressure should be applicable internationally, although the stage of maturity of issues will vary among different countries.^88,89

Implications for regulation and legislation

The model of stages of responsibility suggests that we can predict future directions for tobacco company arguments on tobacco policy. As issues mature, industry positions will potentially shift to reflect public understanding and, in doing so, protect their ability to sell cigarettes profitably. In the second-hand smoke debate, for example, with further regulation, the debate will likely shift to a passive acceptance of the dangers of second-hand smoke, with a promoted policy that protects the “rights” of the individual to keep buying the product, thus protecting the industry’s interest. It is possible that we are seeing the beginnings of this already in the smokeless tobacco debate.⁹⁰ Here, the smokeless tobacco companies, as prime players, accept the public health community consensus on the dangers of second-hand smoke, and position smokeless as an appropriate response by the tobacco companies for the issue of second-hand smoke, without compromising on their core business.

The statements of responsibility for tobacco control policy and issues made by the tobacco industry representatives appear to be consistent with the conceptual framework developed by Zadek¹ to describe the changes in responsibility with changes in the level of maturity of an issue. Corporate responsibility appears to be a method of handling particular social and political issues in a way that protects the role companies have in the distribution, sale and promotion of the products they make. The corporate model that is so effective in creating and sustaining a market for other goods like shoes is also effective in creating and sustaining a market in tobacco. As the tobacco industry is responsible for the sale of tobacco, the industry is the actor with the greatest interest in the continuing use of tobacco. The status of tobacco as a commercial product then is a stumbling block in instituting rational tobacco control policies as it creates perverse incentives to increase or prevent decreases in use. This model is starkly at odds with public health and does little to reduce the burden of tobacco disease that affects all population groups. Other options for the distribution of tobacco that do not create conflict between that distribution and public health goals should be investigated such as the models suggested by Borland.⁹¹