The Genetically Modified (GM) Food Labelling ...

ABSTRACT In the debate surrounding genetically modified (GM) food, intense controversies pertain over whether, or how, GM food products ought to be labelled. This paper examines how the GM-supportive and GM-sceptical alliances use arguments regarding labelling so as to strengthen their respective positions. It is an examination of conflicting arguments across social coalitions, corporations and policy-makers, mainly in the USA, but with certain European comparisons. The empirical material consists of written statements by the different groups. The paper suggests that the ideological and epistemological tenets are radically transformed, or even ‘crossed over’, between GM proponents and opponents when the focus is moved from GM per se to labelling. Two types of crossovers are identified: (i) the crossover of ideologies, and (ii) the crossover of epistemologies. The paper concludes that, while implementing mandatory GM labelling may have several democratic advantages, it is more urgent that both alliances become more reflexive and communicative concerning inconsistent or eclectic crossovers – both ideological and epistemological. Keywords strategy

alliances, critical realism, consumerism, epistemic relativism, framing,

The Genetically Modified (GM) Food Labelling Controversy: Ideological and Epistemic Crossovers Mikael Klintman This paper examines one of the most intensely disputed controversies in the public debate over genetically modified (GM) food: namely, whether – or how – GM products ought to be labelled. The paper analyses conflicting arguments across social coalitions, corporations and policy-makers. Empirically, the paper uses statements written by the different groups. (For a methodological discussion, see the Appendix, pp. 85–86) The debate in the USA is given the main attention in this study, since there is an interesting discrepancy between policy and public opinion in the GM labelling debate in that country; by contrast, the public and policy-makers in, for example, the European Union (EU) are much more in agreement: according to them, GM foods ought to be labelled on a mandatory basis. Nevertheless, a few important international comparisons are made in the paper, particularly epistemic statements about labelling. The GM labelling debate illustrates what can be called ‘argumentative crossovers’. I define these as cases where an alliance uses a certain type of Social Studies of Science 32/1(February 2002) 71–91 © SSS and SAGE Publications (London, Thousand Oaks CA, New Delhi) [0306-3127(200202)32:1;71–91;024026]

72

Social Studies of Science 32/1

argumentation for one issue, but shifts into using an opposing alliance’s type of argumentation for a closely related issue with different practical implications. Concretely, the paper elucidates how the conflicting tenets surrounding GM food per se are radically transformed, and sometimes even reversed, between GM proponents and opponents when the focus of the debate is moved to labelling. Two types of argumentative crossovers are presented and analysed: (i) the crossover of ideologies, and (ii) the crossover of epistemologies.1 One should note that these crossovers could, but need not, entail logical inconsistencies. Instead, the crossovers that are examined here are better described as instances of ideological and epistemological eclecticism, where inconsistencies could be part of the picture. I hold that crossovers, whether logically inconsistent or ‘merely’ eclectic, are much more than theoretical curiosities. They may have crucial consequences for the degree of internal and external efficiency of alliances. Moreover, it is very important to analyse the implications of crossovers on decision-making in democracies, particularly in this food issue, where consumer choices may have dramatic social and ecological implications on a global scale. This paper concludes that, while the implementation of mandatory GM labelling may have several democratic advantages, it is more urgent – in terms of democracy – that both alliances become more reflexive and communicative regarding crossovers. The Social Context: Alliances and Crossovers When studying the GM labelling controversy, it is much more fruitful to analyse the camps on equal terms, as ‘the anti-GM alliance’ versus ‘the pro-GM alliance’, rather than as ‘the anti-GM movement’ versus ‘the establishment’. One reason for this is that the ‘opposite side’ endorses the prevailing GM legislation and its institution in the USA – as compared to, for instance, the EU. Treating both camps as alliances makes it easier to compare conflicts as they have taken place across countries with different GM policies.2 Moreover, the ‘pro-GM alliance’, although endorsing a legislative status quo in the USA, nevertheless has to struggle to change the broader cultural climate so that GM foods will become normalized (or even labelled) as ‘organic’ (suggesting ‘better products’). The cultural struggles among both alliances entail ‘boundary framing’, referring to attributional processes that delineate the boundaries between protagonists and antagonists, ‘good’ and ‘evil’.3 The two alliances are characterized by an internal heterogeneity, in terms of motives, social positions, and so forth. For example, some GM opponents include environmental activists, who stress that the use of GM causes diffusion of environmental risk throughout time and space. Other key opponents are leading chefs within the USA, who have criticized GM food production by emphasizing ‘the dignity of the food supply’,4 which goes beyond plain risk and safety issues. Diversity in motives and interests among the agents for or against GM food does not speak against their status as alliances or movements.5 The diversity may even be a powerful

Klintman: Crossovers in the GM Food Labelling Controversy

73

tool in this process: it ought to make it possible to formulate a wide range of arguments for and against GM, and to work strategically at different levels at the same time, since the alliances involve such diverse actors, and a wide range of motives. By acknowledging the strategic dimension of GM food labelling arguments, this paper examines how each alliance constructs different types of strategic framing regarding how to relate the GM issue to the specific tool of labelling.6 Although crossovers may partly be regarded as instruments in alliance tactics, they nevertheless ought to be scrutinized on the basis of their mixed implications for efficiency within the alliances, as well as for the democratic process and public trust in the alliances. These issues are elaborated at the end of the paper.

The Crossover of Ideologies The most well known rationale of GM food labelling is consumer empowerment. A vast majority of the radical anti-GM alliance subscribes to this rationale. Still, one does not need to twist the matter much to argue that a central ideology related to it is that of ‘Economic Man’, operating among ‘free’ market mechanisms (such as labelling). This ideal, although not always followed in market practice, holds that the consumer – not the state – ought to be sovereign, so that the consumer’s purchases reflect decisions of how society’s resources should be allocated. The libertarian economist Ludwig Edler von Mises states it eloquently (probably inspired by Adam Smith): The consumers are the masters, to whose whims the entrepreneurs and capitalists must adjust their investments and methods of production. The market chooses the entrepreneurs and the capitalists and removes them as soon as they prove failures. The market is a democracy in which every penny gives a right to vote and where voting is repeated every day.7

Yet scepticism surrounding societal consequences of the Economic Man ideal was already common, even in the early days of liberal capitalism. Conservative religious alliances were worried about the 19th-century trend of secularization. Its new individualist foundation was Economic Man, who acted (as the Encyclopædia Britannica puts it) ‘with no intuition of unseen realities, no sensitivity to art or nature, no humility, and no inbred morals or sanction for their dictates’.8 And from the left, Marx’s division of humankind under capitalism into (a) political citizen (hopefully concerned about broader societal consequences) and (b) economic man (motivated by an individualist rationality), implies radical scepticism regarding the consumer-role ˆ as the optimal one for making decisions with far-reaching socio-political consequences.9 Still, when it came to food purchasing and consumption, the Economic Man notion might have been reasonable in times when price, freshness and taste were the main food concerns. (For now, I allow myself to omit the enormous issue of economic consumer inequality.) However, the green revolution and advances in pesticide chemistry, as well as food technologies and chemicals in the latter half of

74


the 19th century, made matters more complicated for consumers to survey. Currently, with the development of GM food production, the US legislators hold that certain food decisions should be kept away from consumers. A Brief Summary of the US GM Food Labelling Regulations Labelling requirements are established by the US Department of Agriculture (USDA) for meat and poultry, and by the Food and Drug Agency (FDA) for all other food products. According to US law, the quality and characteristics of the final product should be the only basis for mandatory labelling, whereas food processes have long been considered irrelevant to US legislation. And, since the FDA has rarely found essential differences between biotech foods and conventional foods, no label is currently required on GM foods, nor is any standardized safety testing by an independent agency required: The agency is not aware of any information showing that foods derived by these new methods differ from other foods in any meaningful or uniform way, or that, as a class, foods developed by the new techniques present any different or greater safety concern than foods developed by traditional plant breeding. For this reason, the agency does not believe that the method of development of a new plant variety (including the use of new techniques including recombinant DNA techniques) is normally material information within the meaning of 21 USC paragraph 321 (n) and would not usually be required to be disclosed in labelling for the food.10

One of the few exceptions of this non-requirement is if the FDA perceives risks of allergic reactions from products which have not usually been associated with such reactions – for instance, tomatoes produced with genes derived from nuts. Ideological Arguments Against GM Labelling Requirements A broad range of actors – from biotech companies, dependent and independent scientists, and certain NGOs 11 – fully support the principles of the US regulation. One of the arguments concerns freedom of choice. Nobody appears openly to challenge the virtues of free choice. Although there has often been a gap between ideal and practice as regards consumer sovereignty on the ‘free market’,12 it would be odd if ideological criticism of free choice came from market-friendly biotech proponents. Instead, two arguments frequently emerge which turn around the issue of free choice. The first is the irrelevance argument. This argument claims that food labelling would be confusing, misleading and irrelevant for consumers. Some scientists and physicians hold this view – for instance, the strong GM proponent Henry Miller: Even a message that is accurate, in the narrowest sense, can mislead and confuse consumers if it is irrelevant, unintelligible, or so craftily selected that it provides inadequate or biased information.

In the same article, biotechnology labelling is also said to be as irrelevant as. . .


75

. . . a label that assures consumers [that] ‘no uranium’ or ‘no rattlesnake venom’ is contained in the product.13

The confusing and irrelevant character of these GM labels would thereby not only be unproductive in terms of providing consumer choice, but even counterproductive: a senior research scientist at the University of Saskatchewan claims that GM labels ‘inform no one and they diminish consumer choice’.14 Furthermore, the argument by a member of the editorial board of World Food Regulation touches on the ‘guilt by association’: Such a label statement, particularly in the current climate, falsely implies the food is less safe than ‘conventional’ foods. Conversely, a label claim of ‘GMO-free’ falsely implies such a food is safer and better than GMOcontaining foods.15

Similarly, the Biotechnology Industry Organization (BIO) holds that it ‘consistently has supported the public’s right to know what we all eat. BIO also has consistently supported making sure the public gets accurate information’. Still, this position leads them to oppose mandatory GM food labelling, since they claim that it would mislead consumers.16 Another way of supporting the prevailing legislation points to an economic irrationality of consumers. This argument contends that a requirement for labelling would significantly increase costs, both for producers and consumers. Product information as a consumer service, therefore, is neither necessarily benign nor beneficial. The real world also must deal with costs. The cost of labelling is far more than just the ink and sticker. Thus consumers wishing to avoid GM entirely will pay more for the privilege. But they’re being misled and misinformed.17

This argument expands from rationality to morality when the poor are brought into the picture. Jeremy Foltz is one among many researchers who elaborate on issues of GM labelling in the developing world.18 Another research scientist, Alan McHughen, also discusses poverty issues related to labelling: And the poor, who must buy at the bottom of the market regardless of their personal opinions, pay a disproportionately higher share of the increased cost to the benefit of no one, especially themselves. No matter what your position, GM labels fail to provide their intended raison d’être – informed choice.19

Ideological Arguments in Favour of GM Labelling Requirements Regardless of the legislation and the arguments presented above, everyone involved admits that the general public supports mandatory GM food labelling, be it in the USA, Europe, Japan, or any other country. For instance, an extensive survey carried out in the year 2000 by Penn, Schoen & Berland Associates indicates that 85% of Americans were in favour of

76


mandatory labelling of GM food. However, it should immediately be noted that this support of labelling in the USA does not necessarily have any clear connection to opinions about GM food, positive or negative: a much lower percentage, approximately 51%, were opposed to GM food as such.20 In Europe, public opposition to GM food is stronger. According to the Eurobarometer 2000, 69% held that genetic engineering should not be encouraged in food production.21 In its demand for GM labelling requirements, the general public, not least in the USA, is supported by a very large number of NGOs. For instance, the Campaign to Label Genetically Engineered Foods was launched in March 1999, supported by numerous NGOs and citizens. In addition, the Genetically Engineered Food Right to Know Act, HR 3377, was introduced into the House of Representatives in November 1999.22 What are the arguments in favour of labelling, if we leave aside arguments which underscore scientific uncertainty and explicit political strategy? First, the argument about democratic rationality refers to the intrinsic value of transparency. The disembeddedness (or alienation) of modern industrial food production from ordinary people’s supervision and monitoring appears to frustrate people. Yet it seems to take a technological revolution of the magnitude of the genetic engineering of food before the disembeddedness becomes perceived as democratically unsatisfactory. Along these lines, various actors present arguments emphasizing the right to information, even without understanding or opposing the new technology. For example, Julian Edwards, the Director of Consumers International, maintains that although consumers have not entirely rejected GM products, people need information about ‘what scientists were doing to their food and had a right to know what they were eating’.23 The intrinsic value of transparency is echoed in a statement by Congressman Bernie Sanders. He notes that science rarely claims that artificial colouring and flavouring are dangerous, although these additives still require labelling. The same requirements should cover GM food: Someone who, for whatever reason, wants to know whether GMO ingredients are in their food, should be able to expect a clear label that indicates whether their food has been genetically altered.24

A second, closely related set of arguments in favour of labelling is based on the diversification of consumer rationalities: ethical or religious concerns. Up until quite recently, individuals were still usually able to see, or ask, whether food products involved ingredients or processes which were in conflict with their ethical or religious concerns. New GM processes, if not labelled, make things much less visible. Michael Darby and Edi Karni conceptualize the visibility aspect by distinguishing between (a) search goods, for which price and direct looks reveal everything; (b) experience goods, for which the consumption, and eating, will reveal their characteristics; and (c) credence goods, which are impossible to evaluate even when using, or eating, them.25 The ethical or religious aspects are


77

obviously connected to credence aspects, not least when concerns relate to food processing. Friends of the Earth, for instance, stresses the new ‘credence problems’ created by genetic engineering of food. One example is raised by vegetarians, who have always been able to select vegetables at the ‘search’ or ‘experience’ levels. With the introduction of GM food – if not separated by a mandatory label – all foods have ended up at the level of uncertainty: Many consumers believe that GE food crops are unethical and violate religious dietary laws, including kosher rules against hybridization. A coalition of groups including representatives of Jewish, Buddhist, Moslem and Christian denominations is presently suing the US FDA for failing to label GE foods. Similarly, vegetarians are concerned about ingesting animal DNA by eating for example fish genes that have been inserted into tomatoes.26

John Fagan develops this claim further by indicating that there is no ethical difference between (a) genes that are actually brought from animals to vegetables, and (b) gene information that has been brought from animals to vegetables. He holds the separation of the two processes to be not only ethically insensitive but also ‘scientifically untenable’: Thus, it is scientifically untenable to claim, as the biotechnology industry and regulatory agencies wish to do, that the rabbit hexokinase gene is no longer a rabbit gene once it is introduced into a tomato plant. No matter how many generations the rabbit hexokinase gene is propagated in tomatoes, the gene still corresponds in information content to the rabbit gene, and the catalytic and kinetic properties of the enzyme still correspond to those of rabbit hexokinase, and not tomato hexokinase. It may not be rabbit material, but it is still rabbit information.27

Fagan claims that groups with such concerns are growing, and that these diversified consumer rationalities should not be discounted. GM food labelling should therefore be required.28 Comments on Ideological Crossovers These arguments will be further analysed below. For now, it is interesting to note that those who are sceptical of GM food – and present mandatory labelling as a sufficient means towards democracy and rational choices – subscribe in practice to the Economic Man principles (albeit extended) to a far higher degree than do the GM proponents who are against labelling. The former alliance, although partly represented by radical environmentalists and market sceptics, embraces in its rhetoric one of the main positions among libertarians and market liberals: namely, labelling as a tool for the political sovereignty of consumers. As is discussed in sections below, this crossover (apart from reflecting a lack of ideological awareness) may be based on the position that the state is distancing itself from important social issues which it should address; it may also be part of opportunistic tactics used by the alliance against GM. The GM-supportive alliance, vice

78


versa, with its partly agro-industrial orientation, ought in principle to be in favour of an open, consumer-oriented market. On the free market, new advanced food technologies can be innovated and distributed to rational consumers – whose demands give fuel to market innovation. However, when the issue is moved from GM food per se to its labelling, the GM advocates emphasize the shortcomings of the labelling tool, basing their argument on the fragile character of Economic Man. It is important to note, however, that by no means all criticism of mandatory labelling is inconsistent with the free market ideal: parts of the concerns with ‘misleading’ information are consistent with the ideal of a market where consumers ‘govern’ by using valid information. Still, ideological crossovers become clear when the labelling critics (in a paternalistic way) claim that consumers have to be protected against making ‘irrational’ choices, even if consumers find these choices worthwhile (for instance, paying more for conventional or organic types of processes of food production distinguished by a mandatory GM label). In sum, the examples brought up reveal a crossover of applied ideological positions between the two alliances. Certain crossovers may be perceived as frame inconsistencies.29 Involved in the conflicting arguments are struggles over the meaning of ‘free consumer choice’. These struggles strike what William Gamson calls the process of ‘cultural resonance’,30 where alliances use widely held beliefs (for example, free consumer choice), and formulate these beliefs in accordance with the interests of the alliance. As we have seen, these formulations may lead to diametrically different practical propositions. The key missing link in this discussion of choice and rationality is of course epistemology – that is, preconditions for knowledge and informed choices. I argue that this is another area where pragmatic flexibility can be identified across the two alliances.

The Crossover of Epistemologies A large part of the conflict surrounding GM food labelling concerns its role ˆ as a precondition for stimulating a broadening of the ‘pecuniary sagacity’ that Thorstein Veblen discussed a century ago.31 Can labelling accurately reflect valid knowledge and real risks? The social potentials and limits to knowledge have long been a main theme within sociology.32 In the sociology of nature and the environment this theme is core, not least due to the complex relationships between science, policy-making and daily life.33 Critical realism provides a fruitful bridge between the philosophy of science and sociology. With regard to the perceived preconditions for valid knowledge, critical realism holds two standpoints as fundamental: (a) the endorsement of the naturalist, epistemological postulate, which holds that certain true correspondence is possible between assertions about the environment and health and environment/health conditions as such. In critical realist terms, this means a rejection of judgemental relativism,34 namely the position that ‘all beliefs are equally valid [or invalid] in the sense that there are no rational grounds for preferring one to another’.35


79

(b) the endorsement of epistemic relativism,36 that is, the claim that, while it in principle is possible to obtain (conceptually filtered) knowledge about the environment and health, all knowledge is socially dependent and fallible.

The question is how proponents and opponents of GM food labelling position themselves vis-à-vis these tenets. Below, I identify interesting crossovers regarding how arguments can combine a strong belief in the certainty of knowledge claims in science with an equally strong sense of the uncertainty of scientifically-based labelling, and vice versa. Against Labelling: The Judgemental Relativism of Labelling There are at least two types of knowledge-related arguments against labelling which frequently overlap. The first type concerns ecological knowledge uncertainty, and the second refers to socio-political knowledge uncertainty. A common way of illuminating the ecological knowledge uncertainties of labelling is to move from discussing mandatory GM labelling to commenting on its supposedly negative counterpart. This argument suggests (in an unreasonably perfectionist way, in my opinion) that in order for mandatory GM labelling to be meaningful, its counterpart must be clearly, or even perfectly, separable. As Alan McHughen states: Negative labels such as ‘GM-free’ or ‘This product contains no GM ingredients’ could help such people identify desired foods. However, the retailers or manufacturers printing such labels could never prove their claims. In theory as well as in practice, negatives can’t be proven. In many jurisdictions, the onus is on the ‘labeler’ to prove the claim before the label is allowed.37

Accordingly, since ‘GM-free’ can never be verified in an absolute sense (due to the risk of adventitious or accidental presence of material of GMO origin), a mandatory label stating that certain foods are genetically modified can never reflect perfectly separated products; the argument contends that such a label therefore lacks meaning, and hence also authority. A closely related argument in this perfectionist vein follows the notion that if society has already created a certain risk, there is no use in trying to avoid a larger risk, since that risk is already reality. Elise Golan and her fellow economists at the US Department of Agriculture formulate it as follows: If anyone accepts the risks of GM food, the risks will already be there, and there will not be any reason to have labels that make others avoid GM food.38

Another type of knowledge-related argument criticizing labelling has to do with trust and social knowledge uncertainty. Social arguments among GM proponents against labelling do not usually point to whether society can trust the motives of the regulators, once labelling is mandatory. Instead, one type of argument pinpoints an inherent subjectivity problem of labelling: it states that attempts to introduce mandatory labelling are

80


themselves inherently biased – politically and socially – and are hence ecologically irrelevant. Kelly Johnston (Executive Vice-President, Government Affairs and Communications of the US National Food Processors Association), for example, claims: She [Senator Barbara Boxer (D-Calif), who introduced a bill to require mandatory labelling] is responding to political pressure from activists, rather than to any real food safety concerns. This is bad public policy; the laws and regulations governing food safety must be based on sound, modern science, not politics.39

The social trust issue here is implied in Boxer’s apparent trust in Congress rather than the FDA, the scientific community, or the food industry. Johnston therefore feels that mandatory labelling would be somewhat socially suspect. Extrinsic and practical social issues pertain to the difficulty of implementing standardized tests to detect GM food components.40 Ford Runge and Lee Ann Jackson present another criticism: it concerns the social arbitrariness of permitted thresholds of GM ingredients that would not require labelling, whether 1% or 0.1%.41 One has to give them their point, in the sense that the thresholds used are not yet founded on any evidence that, for instance, less than 1% GM will not cross the allergenic threshold or will not lead to genetic pollution; instead, the thresholds are only based on estimates of the GM levels which scientists assume they can trace with a high level of certainty. Risks to health or the environment have not been part of decisions surrounding thresholds. Moreover, the ‘social arbitrariness of interpretation of scientific data’ has led to concerns about misleading labelling. The legality of labelling is sometimes held to be based on socially determined interpretations of scientific data. Karen Goldman maintains that the social impact is a challenge to the provision of material and non-misleading information conveyed in a label.42 Certain sociologists share versions of this scepticism. Reidar Almas, for example, claims that ‘even if these new foods are labelled, the consumer has to choose which experts to believe before to buy and eat [sic]’.43 Finally, an argument that encompasses both the ecological and social problems is the practical concern with the potential slippery slope of labelling. Per Pinstrup-Andersen, Director of the International Food Policy Research Institute, asks: What do you label? If you feed a cow on GM soy beans, do you label the milk? If a roll has GM soy beans do you label the roll? If you have oil made from GM corn, which has none of the protein in it, should that be labelled? 44

This points to the problem that a socially constructed and (as he views it) ecologically ungrounded line has to be drawn between products that ought to be subject to labelling, and those that ought not.


81

In Favour of Labelling: The Epistemic Absolutism of Labelling It soon becomes apparent that the epistemic arguments in the USA in favour of labelling are less developed than the arguments (often made by the same actors and organizations) against the technology itself. Although there may be a few exceptions, the GM sceptical alliance in general clearly gives priority to elaborating on complexities of knowledge uncertainties concerning GM risks, while they rarely confront their opponents’ epistemic arguments against labelling at the level of reasoning where these arguments have been presented. Even in sophisticated studies of GM risks which conclude that labelling is desirable, the following type of claims is common: This is a distinct class of risk which is directly associated with the process by which genetically engineered foods are produced. Thus, foods carrying this class of risk can be easily identified, based on the process by which they were developed—genetic engineering. In light of this, it is only fair that consumers be informed of this class of risks and thereby be allowed to exercise their own judgment as to whether or not to accept that risk. In short, genetically engineered foods should be labeled as such.45 [emphasis added]

The molecular biologist making this claim also holds that ‘Labeling provides consumers with knowledge upon which to base rational choices regarding the foods they eat’.46 Statements by other spokespersons illustrate the belief that labelling will provide consumers with ‘the truth about what they eat’, or ‘with the assurance that their products are not harmful’. For instance, the Union of Concerned Scientists, which is well known for its advanced work on knowledge uncertainties, nevertheless states: Not only do consumers have the right to know what they are eating, but with labeling they can use purchasing decisions to influence the extent to which producers rely on the technology. Until much more is known about the risks and benefits, this is clearly a prudent course.47

Although GM sceptics often stress the imperfection and epistemic relativity of science and knowledge, the reasoning concerning labelling reveals a completely different epistemology. The knowledge associated with the use of a label is regarded as objective and unproblematic, at least when discussed in open documents. A crossover has taken place from epistemic relativism and towards an uncritical scientism, or ‘epistemic absolutism’ of labelling. The epistemic arguments become more nuanced when applied to the EU legislation of mandatory GM labelling.48 The EU legislation itself endorses the ‘weak version’ of the ‘Precautionary Principle’ (PP). While the strong PP states that no technology should be accepted unless it can be proven completely harmless, the core of the weak PP is a rejection of full scientific certainty as a necessary criterion for taking regulatory measures.49 The most elucidating application of this Principle is a key amendment, which was made by Commission Regulation 49/2000 (criticized above by the labelling opponents): it provides a labelling threshold of 1%

82


of accidental presence of GM-derived material (during cultivation, harvest, transport or processing); for products that contain less than 1% GMderived substances, labelling is not mandatory. Furthermore, Regulation 50/2000 requires no particular labelling requirements for GM additives or flavourings. Pro-labelling comments on the EU legislation (comments which sometimes also are directed as advice to the USA) are typically formulated as prescribed modifications. Among consumer groups, a desire to reduce the 1% threshold has been expressed.50 Another suggestion is that the 1% threshold should be extended to cover any ingredients and additives in the production chain. Moreover, dubious expressions on GM labels should, according to Enrico Suardi, be replaced by affirmative ones: ‘this food may contain’ should be replaced by ‘this food contains’.51 Yet labelling advocates have only indirectly commented on the labelling critics’ argument about the social dependency of a threshold. The European advocates have done this by referring to the Precautionary Principle, and to the scientific acceptance of knowledge imperfection and uncertainty. At the same time, it is important to note that subscription to the EU legislation – with or without suggested modifications – still implies a profound belief in science. One has to trust science in order to assure, or be assured, that a 1% GM threshold will never be exceeded. This may be a departure from epistemic relativism, at least if one ignores the risk that the threshold may not be maintained. Also, to assume that the threshold would be based on any established knowledge about risks to health or environment is simply erroneous. To change the threshold to a lower level would neither eliminate the principal issue of epistemology, nor enable us to say much about reduced risk levels. Furthermore, both the suggestion of expanding the coverage of the threshold, and of using more affirmative formulations, reveal an even stronger belief in what the scientific and regulatory institutions can achieve. Unless the social dependence and the inherent knowledge uncertainty of labelling are fully acknowledged, such suggestions may be regarded as a further departure from epistemic relativism – from a critical realism to a much more credulous realism. Comments on Epistemological Crossovers In summary, the analysis has indicated clear crossovers of epistemologies between the two camps. The statements made by GM proponents imply a belief in the potential of science to produce safe technology, and knowledge which corresponds closely to reality (as compared with the ‘naturalist, epistemological postulate’ I cited above). In cases where science admits probabilities of risk (that is, of external costs), these could still be weighed rationally against potential benefits – economic, social, and ecological.52 However, despite their strong beliefs in the possibilities for valid knowledge, GM proponents question whether labelling could provide any valid information at all. They may accept scientific ‘probabilism’ (a euphemism for ‘uncertainty’) of the GM technology itself, but require (impossible and


83

unreasonable) scientific ‘absolutism’ of labelling – which implies a judgemental relativistic view of the results of labelling, since their requirements cannot be met. The reverse is true for the GM-opposing statements that I have studied, which all highlight the imperfection and subjectivity of knowledge,53 while treating labelling as if it were the objective mirror of truth. Thus, in their rhetoric, both camps move between epistemic absolutism and judgemental relativism, depending on the issue. This may be perceived as crossovers of epistemological framings.54 Although not necessarily ‘inconsistencies’, in a logical sense, these crossovers reflect at least an eclectic treatment of the principal preconditions for knowledge, something which may itself be misleading. Although sometimes leading to unreasonable judgemental relativism, knowledge aspects of labelling are more interestingly argued by its opponents. They have made efforts to discuss epistemic and methodological aspects of labelling, while proponents have dealt with labelling as an unproblematic source of valid knowledge. In my research data, I could not, for instance, find actors who held themselves to be against GM while at the same time discussing fundamental knowledge problems of labelling. Such a position would be natural for more radical, anti-science camps that prefer the banning of GM foods.55 Still, the lack of such nuanced variations is itself interesting. This clear polarization raises the issue of common pragmatic strategies within ideologically diversified alliances.

Discussion and Conclusions All this leads to three main points. First, my initial assumption was that the late modern diversification of rationalities so often analysed in sociology would entail conflicts within the two alliances in the GM-labelling debate. Therefore, it was striking to see how few internal conflicts there are, at least in the written documents studied.56 There appears to be a rather solid polarization between the two alliances regarding the pros and cons of mandatory labelling.57 This polarization, combined with crossovers of ideologies and epistemologies, calls for a theoretical framework which takes unifying alliance strategies and consensus mobilization into account. Second, from the viewpoint of the labelling advocates, it is obvious that the US legislation concerning GM food labelling is out of touch with the public. The legislation completely ignores the demand for ‘transparency’ (whether fully achievable or not), both as an intrinsic end, and as a means for rational actors to perform in a reflexive and diversified sense. By tradition, the legislation treats all invisible, credence-dependent processes and externalities as irrelevant. This is particularly striking in the case of governmental agencies for which social externalities are beyond their area of responsibility: ‘it is not the FDA’s jurisdiction to start dealing with social questions related to food supply’.58 This leaves social aspects largely outside the scope of policy-makers. As a result, some claim that so-called

84


‘free consumer choice’ is moving towards complete consumer responsibility, while food corporations and policy-makers are relieving themselves from several global responsibilities. Others, also represented by several social coalitions, would hold that this is exactly what a modern democracy of active public participation should look like – especially in a society with all of its late modern uncertainties. From the perspective of labelling opponents, however, these aspects represent consumer irrationalities, since they lie outside the sphere of the traditional Economic Man: according to this group, people should not be allowed to base their purchase selections on whatever principles they want; they should be free to choose, but only within the narrow framework of economic individualism upon which the market mechanisms rest. A development of broadened rationalities would only confuse the market system. Among labelling opponents, then, one can trace an interesting disappointment: parts of the public are threatening to let down the loyal community of narrow individualism upon which capitalist society is kept in harmony and balance. Finally, it becomes relevant to ask: What are the problems, if any, with crossovers? When do the strategic dimensions of cultural resonance and boundary framing run the risk of making crossovers counterproductive? To elaborate on this question one needs to distinguish between internal efficiency for the movements themselves, and the external efficiency in the alliances’ relations to the public, as well as the democratic issue of transparency. At one level the extensive polarization (including the crossovers) between the GM-supportive and GM-sceptical alliances in the labelling issue reflects a high degree of internal efficiency. Problems emerge, however, for groups within the alliances who do not share the combination of positions; the US National Farmers Union, for instance, is in favour of GM, while working in favour of mandatory labelling for liability reasons. At another level, parts of the alliances may accept some of the crossovers which merely reflect eclecticism, while pure inconsistencies run the risk of reducing the internal loyalty. This point may apply even more to external efficiency. We have seen, for example, how the paternalistic argument of consumer irrationalities combined with an open market position already troubles people, whereas the argument about misleading labelling is more challenging to confront. This assumption is echoed by Robert Benford and David Snow who, when referring to framings in general, suggest that the more apparent the contradictions are, the less resonant the alliances’ preferred framings may be. They also find certain support for this suggestion in the literature.59 It is plausible that coherence makes the influence on political insiders more efficient, and that it facilitates the outsider strategies of public trust development.60 Still, it is not evident that insider and outsider strategies go hand in hand in all cases and stages. For instance, GM-sceptics’ treatment of mandatory labelling as reflections of an absolute truth may be effective during an early stage of creating a public opinion in favour of labelling. Given the results of the opinion polls, it is fair to claim that this crossover has been successful so far. On the other hand, to convince policy-makers in


85

the USA, and to have the public maintain its main position, would probably require an open and epistemologically conscious debate surrounding how to deal with knowledge uncertainties of labelling. From a democratic perspective, however, the conclusions have to be broader than merely addressing strategies. Here it becomes necessary for the public to demand that the alliances reflect openly on the fundamental values and goals behind their struggles in the labelling issue. In resource mobilization terms, this is a call for a clearer consistency between prognostic framing (about whether labelling is or is not a remedy) and motivational framing (referring to the goals for engaging in the alliances’ struggles). Are the struggles towards mandatory labelling really struggles towards completely banning GM food altogether? 61 This question is relevant to the large part of the public which is in favour of GM labelling, while not opposed to GM technology. It is particularly important to those who, based on a concern for global equity and Third World development, believe that the thousands of grassroot GM projects in developing countries will improve the situation for the poor. Another crucial question is: What are the plans and visions among the GM-supporters in terms of corporate power, in light of the massively criticized, increased global food dependency on a few large corporations? Who should have the responsibility for these issues if GM food, as credence goods, is not labelled on a mandatory basis? From a democratic viewpoint, it is critical that these bottom lines be reflected on and clarified, both among the alliances themselves and, ideally, for the public – so that the public can decide whom, if any, they choose to stand behind. Perhaps this ought to be regarded as an even more important part of the ‘right to know’ principle: the democratic right to know about political and economic strategies towards less visible goals with vast social and environmental consequences. The respective visions of the groups regarding how social and ecological problems should be alleviated also ought to be made explicit, so that the public could choose sides or create further alternatives. One can wish that the public will increasingly make such choices in their roles ˆ as political actors, basing their choices on a broader reflexivity and morality, by participating in existing coalitions or by creating new ones. Further reflexivity concerning food technology and its relation to societal power becomes crucial here. The words of Paul Goodman in the late 1960s keep gaining relevance: ‘whether or not it draws on new scientific research, technology is a branch of moral philosophy, not of science’.62

Appendix About my methodological choice of document analysis. In studies of conflicting arguments a few methodological options are available, such as interviews with various stakeholders and coalitions, discourse analyses of interactions between stakeholders, as well as analyses of written documents and of information distributed by the coalitions. I have chosen the latter method of analysing documents and written information. An important aim of this

86


paper is to acknowledge how the GM-supportive and GM-sceptical alliances adopt arguments regarding labelling opportunistically and strategically. I hold that the written documents that I have analysed all have a certain strategic dimension, and my perspective is to look partly at arguments as strategies to impact public opinion and policy. Although some actors may lack awareness of the crossover character of their arguments, it is beyond the scope of this paper to discuss to what extent these arguments also reflect people’s inner ideological and epistemic beliefs. As a methodological alternative, in-depth interviews would perhaps provide us with more detailed expressions of strategies, but to take us closer to the inner beliefs (although it is beyond the scope of this paper) would hardly be possible with conventional in-depth interviews. Still, it would be very interesting to have the purveyors subsequently reflect on my findings in this paper in interviews, although such reflections would not be essential to this paper. About representativeness of the analysed arguments. The sampling has been strategic, and has been based on the nature of the labelling arguments among groups who are sceptical or in favour of GM food. I have studied information from well over 100 organizations and actors, and have only used arguments that have appeared frequently and consistently within the respective alliances. The initial objective was to map out the divergent and conflicting arguments within NGOs, as reflected in the information they release to the public. After analysing this comprehensive material, it was surprising to see the high level of consensus that prevails within the respective camps in the labelling issue, despite the heterogeneous characters of the coalitions. This led me to acknowledge the strategic dimensions of these arguments and claims, where crossovers turned out to be one tendency. In the paper, only the frequently repeated arguments in each coalition have been selected, and I have avoided bringing up ‘unrepresentative’ claims without clarifying that they are rare. The selection of arguments has been based on literature and web searches in a very inclusive way. I have been open to instances that speak against the claims of this paper, and would have seen such instances as positive challenges. Yet one should note that in this paper representativeness does not have to be conceived as an issue of proportion in the strict ‘democratic’ sense. Even if one found that less than half the organizations in one of the alliances use a certain type of crossover argument, it could still remain an interesting phenomenon. And vice versa: if, for instance, one were to find that a few anti-GM actors use sophisticated epistemic arguments in favour of labelling, this would not change the overall picture of the commonly simplistic pro-labelling claims made in information to the public. Nevertheless, proportions should, if possible, be explained to the reader.

Notes This work was supported by the Swedish Knut and Alice Wallenberg Foundation, which is gratefully acknowledged. I also wish to thank William Gamson, Ken Goldsmith, Sheila


87

Jasanoff, Kelly Joyce, Calestous Juma and Dara O’Rourke for useful comments on earlier versions of this paper. 1.

I have left aside other types of arguments for and against GM food labelling – for instance, the economic argument that American GM food may not be accepted on the European market. 2. Furthermore, the complexity of the picture makes it insufficient to categorize the pro-GM alliance as (self-)interest-based, and the anti-GM alliance as value- and ethics-based. Throughout this paper, I note that both camps have interest- and value-based components: cf. Joyce Tait, ‘More Faust than Frankenstein: The European Debate about the Precautionary Principle and Risk Regulation for Genetically Modified Crops’, Journal of Risk Research, Vol. 4, No. 2 (2001), 175–89. For a useful summary of the European debates and factions, see Nigel Williams, ‘Agricultural Biotech Faces Backlash in Europe’, Science, Vol. 281 (7 August 1998), 768–71. 3. See Scott A. Hunt and Robert D. Benford, ‘Identity Talk in the Peace and Justice Movement’, Journal of Contemporary Ethnography, Vol. 22 (1994), 488–517. See also Doug McAdam, ‘Culture and Social Movement’, in Enrique Larana, Hank Johnson and Jozeph R. Gusfield (eds), New Social Movements (Philadelphia, PA: Temple University Press, 1994), Chapter 2, 44–63, at 49. McAdam argues that social movements (or alliances) are often more likely to have a cultural impact than a political or economic impact. To be sure, he is referring to opposition movements. Still, I hold that the situation may be just as sociologically interesting when an alliance is supported by legislation, although the cultural climate is lagging behind (as in the labelling issue). 4. Sheldon Krimsky & Roger P. Wrubel, Agricultural Biotechnology and the Environment: Science, Policy, and Social Issues (Urbana: University of Illinois Press, 1996), 235. 5. Mario Diani, ‘Analysing Social Movement Networks’, in M. Diani and Ron Eyerman (eds), Studying Collective Action (London: Sage, 1992), 107–35. Heterogeneity in terms of motives is often said to be characteristic of ‘new social movements’ and alliances (in contrast to the ‘old’ working class movement). The newer type of alliances tends to be more pragmatic and less ideological than the old one. See Claus Offe, ‘New Social Movements: Challenging the Boundaries of Institutional Politics’, Social Research, Vol. 52 (1985), 817–68. 6. Three types of framing are distinguished: (a) diagnostic framing, which refers to claims about blame and responsibility behind the situation defined as problematic; (b) prognostic framing, which has to do with the proposed solutions and strategies; and (c) motivational framing, which can be regarded as the stated rationale for the alliance activity, aimed at mobilizing organizations and actors towards a goal. See Robert D. Benford, ‘Frame Disputes within the Nuclear Disarmament Movement’, Social Forces, Vol. 71 (1997), 677–701; also R.D. Benford and David A. Snow, ‘Framing Processes and Social Movements: An Overview and Assessment’, Annual Review of Sociology, Vol. 26 (2000), 611–39. 7. Ludwig E. von Mises, Omnipotent Government: The Rise of the Total State and Total War (New Haven, CT: Yale University Press, 1944), 17. 8. < http://members.eb.com/bol/topic?eu = 108608&sctn = 7 > (accessed 27 February 2001). 9. For extended discussion of these matters, see J. Gay Tulip Meeks (ed.), Thoughtful Economic Man: Essays on Rationality, Moral Rules and Benevolence (Cambridge: Cambridge University Press, 1991). 10. US Food and Drug Administration, Statement of Policy: Foods Derived from New Plant Varieties, Federal Register (29 May 1992), 57: 22984–23005, at 22991. For the latest regulatory developments in food biotechnology, see < http://www.aphis.usda.gov/biotech > (accessed 5 December 2001). For further discussion, see C. Ford Runge and Lee Ann Jackson, ‘Labeling, Trade and Genetically Modified Organisms’, Journal of World Trade, Vol. 34, No. 1 (2000), 111–22. 11. One of the stronger NGOs in favour of GM food and opposed to GM food labelling is The Alliance for Better Foods: < http://www.betterfoods.org/ > (accessed 27 February 2001). In practice, this NGO has a certain pro-industry orientation, but this does not override the social and ecological value basis of its work.

88


12. One should note that parts of industry have also opposed other types of labels – for example, risk labels on cigarettes, nutritional labels, and labels on irradiated foods. These practices can be regarded as further signs of ideological crossovers between market ideology and practice, from an ‘empowered consumer’ ideology to a ‘captive consumer’ practice. 13. Henry I. Miller and Suzanne L. Huttner, ‘Food Produced with New Biotechnology: Can Labeling Be Anti-Consumer?’, Journal of Public Policy & Marketing, Vol. 14, No. 2 (Fall 1995), 330–33, quotes at 330. 14. Alan McHughen, ‘Uninformation and the Choice Paradox’, Nature Biotechnology, Vol. 18, No. 10 (October 2000), 1018–19, at 1018; < http://biotech.nature.com > (accessed 25 February 2001). 15. John S. Eldred, ‘Labelling of GMO-Derived Food Ingredients: A Recipe for Misinformation’, paper presented to the 27th Session on Food Labelling (Ottawa, Canada, 27–30 April 1999). 16. BIO, ‘BIO Opposes Mandatory Food Labeling Bill Introduced by Sen. Boxer’, Biotechnology Industry Organization press release (23 February 2000). 17. McHughen, op. cit. note 14, 1019. 18. Jeremy Foltz, of the Department of Agriculture & Resource Economics at the University of Connecticut, refers to social equity with regard to developing countries. Opponents of labelling GM foods claim that labelling in the developed parts of the world will drastically reduce the export market of developing countries that use GM technology. Moreover, there is a risk of developing countries becoming ‘the dumping ground for unwanted and risky products produced in developed countries’: Jeremy Foltz, Labelling of Biotechnology Products in Developing Countries (Cambridge, MA: Center for International Development, Harvard University, 2000), 1. 19. McHughen, op. cit. note 14, 1019. 20. For this poll, Penn, Schoen & Berland Associates conducted a nationally representative telephone survey of 2,248 adult men and women between 20 April and 10 May 2000. The sampling error for adult women, adult men and teenagers is 3.1%, 4.4% and 4.4% respectively, at 95% confidence level. Additional results from the Oxygen/Markle Pulse (a joint project of Oxygen Media and the Markle Foundation) are available at < www.pulse.org > . For a comparison between a dozen opinion polls concerning mandatory GM labelling, see < http://www.centerforfoodsafety.org/facts&issues/polls.html > (accessed 30 September 2001). Despite the special interests of this organization, the readers can note a general tendency across the various polls. 21. The survey is part of the Eurobarometer 2000, 52:1 (15 March 2000), conducted by the European Union in 1999. For methodological details, see < http://europa.eu.int/comm/research/pdf/eurobarometer-en.pdf > (accessed 30 September 2001). 22. Companion legislation was introduced into the Senate, S 2080, by California Senator Barbara Boxer in February 2000. The bill has not yet been passed. 23. Julian Edwards is quoted by Patricia Reaney in ‘Consumer Group Wants Labels on all GM Products’ (Reuters: 29 February 2000). 24. Statement of Congressman Bernie Sanders on Genetically Engineered Food Right to Know Act (statement dated 10 November 1999): < http://www.house.gov/bernie/statements/1999–11–10-genetic.html > (accessed 25 February 2001). 25. Michael R. Darby and Edi Karni, ‘Free Competition and the Optimal Amount of Fraud’, Journal of Law and Economics, Vol. 16 (1973), 67–88. 26. < http://www.foe.org/safefood/factsheetge.html > (accessed 27 February 2001). 27. John A. Fagan, Science-Based, Precautionary Engineered Foods (2000); available online: < http://www.geocities.com/luizmeira/label.html > (accessed 3 March 2001). John Fagan is a Professor of Molecular Biology at Maharishi University of Management in Iowa. 28. Some ethicists have criticized such intrinsic arguments. Gary Comstock, for instance, maintains that vague GM-averse arguments such as ‘man should not play God’, or ‘food production should be “natural”’, need to be specified and preferably developed


29.

30. 31.

32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42.

43.

44.

45. 46. 47.

48.

89

into extrinsic arguments in order to gain plausibility and value: Gary L. Comstock, Vexing Nature? On the Ethical Case Against Agricultural Biotechnology (Boston, MA: Kluwer, 2000), esp. Chapter 5. The frame inconsistencies would here be based on a discrepancy between (a) diagnostic framing and (b) prognostic framing (see note 6). The framing of (a), regarding whom to blame – corporations, the free market versus the lack of individual consumer choice – leads to (b), prognostic framings which imply endorsements of principles that the respective movement itself criticizes. William Gamson, ‘Political Discourse and Collective Action’, International Social Movement Research, Vol. 1 (1988), 219–44. See, for example, Thorstein Veblen, ‘The Preconceptions of Economic Science’, The Quarterly Journal of Economics, Vol. 13 (1899), Part II. Veblen also uses the concept when discussing other roles ˆ than the citizen role: ˆ he uses it to criticize Adam Smith’s conception of ‘normal human nature’. See, for example, Karl Mannheim, Ideology and Utopia: An Introduction to the Sociology of Knowledge (San Diego, CA: Harcourt Brace Jovanovich, 1985 [1936]). Ulrich Beck, Risk Society (London: Sage, 1986; 2nd edn, 1992); Peter Dickens, Reconstructing Nature (London: Routledge, 1996). Andrew Collier, Critical Realism: An Introduction to Roy Bhaskar’s Philosophy (London: Verso, 1994), 90–91. Roy Bhaskar, Scientific Realism and Human Emancipation (London: Verso, 1986), 72. Roy Bhaskar, Reclaiming Reality: A Critical Introduction to Contemporary Philosophy (London: Verso, 1989), 23. McHughen, op. cit. note 14, 1018. Elise Golan, Fred Kuchler and Lorraine Mitchell, Economics of Food Labelling (Washington, DC: US Department of Agriculture, No. 793, 2000), 12–14. Kelly Johnston, EVP, Government Affairs & Communications, National Food Processors Association press release (23 February 2000). See, for example, Nigel Williams, ‘Can Regulations Requiring Labeling of Genetically Modified Foods Work?’, Science, Vol. 281 (7 August 1998), 769. Runge & Jackson, op. cit. note 10, 115. These authors study applied economics and law at the University of Minnesota. Karen A. Goldman, ‘Labelling of Genetically Modified Foods: Legal and Scientific Issues’, The Georgetown International Environmental Law Review, Vol. 12 (2000), 717–60. The author is an Adjunct Professor of Law at the Georgetown University Law Center. Reidar Almas, ‘Food Trust, Ethics and Safety in Risk Society’, Sociological Research Online, Vol. 4, No. 3 (1999), U282–U291, at U282; abstract at < http://socresonline.org.uk/4/3/almas.html > (accessed 5 December 2001). Per Pinstrup-Andersen, quoted by Con Bernama (of the Malaysian National News Agency) in ‘Labeling of GM Foods Impossible, Say Experts’ (press release, Melbourne, 27 January 2001); available online: < http://www.plant.uoguelph.ca/safefood/archives/agnet/ 2001/2-2001/ag-02-01-01-02.txt > (accessed 5 December 2001). Fagan, op. cit. note 27, 10; available online: < http://www.geocities.com/luizmeira/label.html > (accessed 3 March 2001). Ibid., 1. Union of Concerned Scientists, ‘What is UCS’s Stance on Biotechnology and Genetic Engineering?’, < http://www.ucsusa.org/index.html > (accessed 27 September 2001). A further example of a simplified epistemic treatment of labelling is found in Marc Lappé & Britt Beley, Against the Grain: The Genetic Transformation of Global Agriculture (London: Earthscan, 1999), 3: ‘Industry has decided to silently invade food market shelves by denying any visible identifiers of genetic engineering. . . . The net effect is to subvert the normal process of consumer choice by suppressing the knowledge needed to freely choose. The cornerstore of such a privilege is labeling’. Whereas the US legislation is based on the principle of ‘substantial equivalence’, the EU legislation endorses the ‘precautionary principle’. In the EU, for GM plants authorized after the second amendment of Directive 90/220/EEC (in 1997), positive

90

49.

50. 51.

52.

53.

54.

55.

56.

57.

Social Studies of Science 32/1 GMO labelling is mandatory. However, genetically modified plant material authorized before 1997 under Directive 90/220/EEC (one soy, one maize and two rapeseed) is not subject to labelling requirements. Yet the revised Directive 90/220/EEC makes labelling provisions more comprehensive, as it requires labelling at all market stages. Moreover, Regulation (EC) No. 1139/98, currently the model for labelling in the EU, provides that the presence of DNA or protein resulting from genetic modification is used as the criterion for labelling of GM food or food ingredients. For detailed discussion of the PP and its application, see Edward Soule, ‘Assessing the Precautionary Principle’, Public Affairs Quarterly, Vol. 14, No. 4 (October 2000), 309–27, and Les Levidow, ‘Precautionary Uncertainty: Regulating GM Crops in Europe’, Social Studies of Science, Vol. 31, No. 6 (December 2001), 842–75. See Julian Morris, Defining the Precautionary Principle (Cambridge, MA: Center for International Development, Harvard University, 2000), 1–2. Enrico Suardi, Problems Implementing the EU Regulation on Labelling of Genetically Modified Foods (Cambridge, MA: Center for International Development, Harvard University, 2000), 1–2. This author works in the UK, at the London School of Hygiene and Tropical Medicine. Such weighing normally acknowledges potential ecological risks, although it stresses potential ecological benefits: reduced environmental impacts from pesticides, increased yield, and phytoremediation are a few of such assumed benefits. A ‘balanced’ overview of potential risks and benefits is found in Laressa L. Wolfenbarger and Paul R. Phifer, ‘The Ecological Risks and Benefits of Genetically Engineered Plants’, Science, Vol. 290 (15 December 2000), 2088–93. For a critical realist discussion of the world as ‘a stratified open system’, see Peter Dickens, ‘Linking the Social and Natural Sciences: Is Capital Modifying Human Biology in Its Own Image?’, Sociology, Vol. 35, No. 1 (February 2001), 93–110. The environmental concerns connected to limits of science also have repercussions within science itself. To discuss specific risks is beyond the scope of this paper. Among the most commonly mentioned ecological risks can briefly be mentioned crop-to-wild hybridization, evolution of resistant pests, ecological effects of insect-resistant plants, and effects on soil biota and fertility. More generally, there is the ‘scientific fallacy’ of treating results from isolated, small-scale, and short-term research projects as sufficient evidence of ecological product and process safety in the real world. For an extensive exploration of these risks, see Allison A. Snow & Pedro Moran-Palma, ‘Commercialization of Transgenic Plants: Potential Ecological Risks’, BioScience, Vol. 47, No. 2 (February 1997), 86–96. The frame inconsistencies would here consist of a discrepancy between (a) diagnostic framing and (b) prognostic framing (see note 6). The framings of (a), regarding how to perceive scientific knowledge claims of GM externalities, lead to reversed claims in (b), pertaining to the epistemologies of labelling. The anarchist NGO, ‘Bioengineering Action Network’, for instance, calls for a ban on GM food, based on social class arguments. They hold that labelling would only favour the upper class, who can afford to avoid GM food. Yet they do not see any knowledge uncertainty involved in labelling: see online < http://www.tao.ca/ ~ ban/1099broaderlook.htm > (accessed 2 November 2001). Perhaps (as one referee suggested) in-depth interviews with actors in the movements would reveal internal controversies. However, since open statements in written documents have been the focus of this paper, examination of internal conflicts through in-depth studies would require another study. One of the few internal conflicts to be found was between the general pro-GM/anti-GM labelling movement and the US National Farmers Union, who were in favour of GM food technology but also in favour of mandatory GM labelling. The reason for the farmers’ position had to do with insurance and liability issues, in the event that it was discovered that GM food would cause environmental or health problems: National Farmers Union press release, ‘US National Farmers Union supports GMO labeling’ (23 February 2000). This clear polarization, with a con-GM movement consensus mobilization regarding


58.

59. 60.

61.

62.

91

labelling, differs from the prognostic framing of certain other movements, which do not have the same degree of consensus. The US anti-death penalty movement, for instance, has been shown to be divided between abolitionists and litigators, the latter advocating saving the lives of the prisoners one by one, instead of fighting for an immediate abolition of capital punishment: see Herbert H. Haines, The Anti-Death Penalty Movement in America, 1972–1994 (New York: Oxford University Press, 1996); see also Benford & Snow, op. cit. note 6. David Schmidt (Senior Vice-President for Food Safety, International Food Information Council, Washington, DC), speaking on US National Public Radio, ‘Talk of the Nation’ (10 October 2000); the audiofile is available online: < http://www.npr.org/ramfiles/totn/20001020.totn.01.rmm > (accessed 4 December 2001). Benford & Snow, op. cit. note 6, 620. For a more extensive discussion of insider and outsider strategies of social movements, see Norman Ornstein and Shirley Elder, Interest Groups, Lobbying and Policy Making (Washington, DC: Congressional Quarterly Press, 1978). For instance, people in the Campaign to Label Genetically Engineered Foods, mentioned above, make no secret that mandatory labelling is not their final goal. They admit that they support a moratorium or ban on GM crops, and that they are planning to lobby the Environmental Protection Agency to get the moratorium passed. Yet their strategic focus is placed on mandatory labelling. The reason is that they have learned from the consequences in the EU of its mandatory labelling requirement. GM food resistance among environmental groups and consumers in the EU, with mandatory labelling as a tool, has triggered a de facto moratorium as of June 1999, in terms of novel releases of GMOs: Chantal Nielsen and Kym Anderson, GMOs, Trade Policy, and Welfare in Rich and Poor Countries (Adelaide, South Australia: Centre for Economic Policy Research, University of Adelaide, Policy Discussion Paper No. 0021, 2000), 6. Paul Goodman, ‘Can Technology be Humane?’, originally published in his The New Reformation: Notes of a Neolithic Conservative (1969), reprinted in Albert H. Teich (ed.), Technology and the Future (Boston, MA: Benford/St Martin’s Press, 8th edn, 2000), 90–102, at 93.

Mikael Klintman is a Wallenberg Fellow in Environment and Sustainability at the Department of Political Science at Massachusetts Institute of Technology. He received his PhD at the Department of Sociology at the University of Lund in Sweden, where he will return after his temporary fellowship at MIT; his dissertation was entitled Nature and the Social Sciences: Examples from the Electricity and Waste Sectors. He conducts research within the diverse areas of renewable energy, sustainable transportation and food labelling controversies, all three areas involving cross-Atlantic comparisons. The common objective of these studies is to scrutinize the pragmatic ideology of ‘green consumerism and green citizenship’ in light of late modern institutional and economic contexts. He participates in the international programme on ‘Alliance for Global Sustainability’ (through MIT), and in research seminars at Kennedy School of Government at Harvard University on ‘Global Governance of Biotechnology’ (led by Calestous Juma) and ‘Science Studies Research’ (led by Sheila Jasanoff). Address [before 15 June 2002]: Department of Political Science, Massachusetts Institute of Technology, Room E53:479, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139, USA; fax: +1 617 258 6164; email: [email protected] [after 15 June 2002]: Department of Sociology, University of Lund, Box 114, S-221 00 Lund, Sweden; fax: +46 46 222 41 00; email: [email protected]