Food and Agricultural Biotechnology: Ethical Issues Behind Research Policy Choices

Paul B. Thompson, Purdue University

Issue: Food and agricultural biotechnologies have demanded new interpretations of longstanding policies in food safety, environmental impact and animal health. The controversy over agricultural biotechnology reveals a number of ethical concerns in existing approaches to agricultural research evaluation.

Relevance: Ethical issues might be raised with respect to many types of technology. An overview of the controversies associated with agricultural biotechnology provides a microcosm for understanding a key set of ethical issues associated with any attempt to evaluate research and technology development, or to use such evaluations in research management policies.

Introduction

An important issue in science today is the tension between research policy, which stresses optimal trade-off of risk and benefit, and policy that aims to satisfy criteria of informed consent. A second major issue concerns the role of agency and intention in making research choices, and questions of responsibility for both intended and unforeseen consequences. A third concerns the appropriate role of uncertainty and precaution in research evaluation and planning. A fourth concerns the socio-economic impact of agricultural biotechnology on the viability of farms, rural communities and the process of globalization.

The agricultural sciences present a particularly interesting case study for examining ethical issues in technology evaluation, and especially so with respect to the products of gene transfer. Worldwide, agricultural research has been organized and conducted by public institutions such as national and international research centres and government funded university scientists for over 100 years. Agricultural research has, from the outset, been highly applied, aimed at discoveries and technology development that would lead to improvements both in food and fibre production, and in rural quality of life. Furthermore, agricultural technology and its socio-economic effects have been extensively studied by social scientists for many decades. Although this research has not been undertaken within the framework of officially mandated evaluation activities, it has been conducted in a variety of national settings, has deployed many social science research methods, and has examined a wide range of issues.

The agricultural sciences present a particularly interesting case study for examining ethical issues in technology evaluation, and especially so with respect to the products of gene transfer

Given this knowledge base, research administrators approached the development of new agricultural and food technologies using techniques for mapping and transferring genes from one organism to another with a high degree of confidence in their ability to prospectively select appropriate topics for research. Given the controversy and chaos that greeted the products of food and agricultural biotechnology, it is apparent that their confidence was somewhat misplaced. Neglect of ethical issues, both in the evaluation of agricultural research and in the policy decision making that was based on the social science literature extant in the 1980s when many key decisions were made, is one of the key elements in accounting for this failure.

Agricultural science and technology have been subjected to social impact evaluation for several decades mainly through the use of economic simulation models

As noted, agricultural science and technology have been subjected to social impact evaluation for several decades. This research is based on economic simulation models developed primarily for monitoring and predicting price and quantity performance of the agricultural sector for both developed and developing country economies. Many agricultural technologies lend themselves to estimation of impact on commodity yields, and data on farmer adoption rates and reported yields allow these estimates to be incorporated into agricultural sector models. Beginning in the 1960s, these basic models began to be augmented with techniques to evaluate negative externalities attributable to environmental impact and social dislocation. A number of the key papers in this literature are collected by Berardi and Geisler (1984).

Beyond this sketch, the evaluation literature on agricultural science is too broad to summarize in this context. Vernon Ruttan’s book Agricultural Research Policy (1985) provides an overview, and Rossmiller’s Agricultural Sector Planning (1978) describes an interdisciplinary methodology for identifying research needs and implementing a participatory system for continuous evaluation of research impact. More recent publications by Huffman and Evenson (1993) and the U.S. Department of Agriculture Economic Research Service (1995) update this literature. Johnson and Bonnen’s Social Science Agricultural Agendas and Strategies (1991) collects essays by many of the world’s leading agricultural researchers summarizing what is known and what needs have yet to be met regarding the evaluation of agricultural and food systems performance.

Existing agricultural sector models provide a number of ways for analysts to disaggregate the various influences on system performance, and to make defensible judgements about the relative influence of various contributing causes

Existing agricultural sector models provide a number of ways for analysts to disaggregate the various influences on system performance, and to make defensible judgements about the relative influence of various contributing causes, including scientific and technological innovation. Given some plausible assumptions, models designed to predict agricultural sector performance with respect to commodity prices and environmental externalities can serve as a basis for technology evaluation. First, one must assume that under the conditions of competitive markets, gains to agricultural producers or to food consumers (in the form of lower food prices) represent beneficial outcomes. Second, one must assume that it is possible to compare such beneficial outcomes with negative outcomes in a manner that adequately reflects all the parameters that are relevant to the evaluation process. Despite the initial plausibility of these assumptions the experience with agricultural biotechnology reveals several areas of weakness.

First, economic and sociological methods in use prior to 1980 were almost exclusively focused on a utilitarian or consequentialist model of technology evaluation, and provided decision makers with little insight into the way that science and technology could raise ethical issues relating to rights and consent. This problem became particularly evident with respect to the way that consumer interests were neglected in evaluating the use of rDNA techniques to produce genetically modified (GM) foods. Second, these methods were so strongly focused on outcomes that they neglected the way that the motives and moral character of actors would be used to evaluate the acceptability of products and the credibility of information regarding biotechnology.

Despite their intuitive plausibility, models designed to predict agricultural sector performance suffer from a number of weaknesses arising, in particular, from their exclusive focus on a utilitarian or consequentialist model of technology evaluation

A third weakness concerns the environmental risks of genetically engineered crops, and a fourth involves the effect of genetic engineering on the socio-economic structure of farming and rural communities. These two weaknesses are surprising in that social science research on agricultural innovation provided ample reason to expect that environmental impact and social consequences would be sources of controversy. One might conclude that decision-makers simply ignored the lessons from previous evaluations of agricultural technology. On the contrary, however, with respect to both environmental and social consequences, the way in which evaluation information is conceptualized can presuppose a particular ethical orientation. The current social controversy over agricultural biotechnology can be traced to a somewhat narrow way of understanding the ethical significance of the evaluation process (Thompson, 1997a).

Prior to 1980, virtually all of the social science research on the evaluation of agricultural research had an implicit ethical orientation toward what philosophers would characterize as utilitarian or consequentialist thinking. In this model, scientific research and technology development is implicitly understood as an initiating event that causes a variety of effects on human, animal and environmental health, the economic well-being of individuals, and on aggregate socio-economic indicators such as GDP, distribution of income, national budgets, balance of payments and the like (Thompson, Ellis and Stout, 1991). As Barré (1999) points out, the methodological challenges in modelling the complex social mechanisms that mediate this causality, and in measuring these impacts, can be overwhelming. However, in this context, it is more important to highlight the normative presumptions involved in this conceptualization of evaluation.

Utilitarian or consequentialist thinking sees scientific research and technology development as an initiating event that causes a variety of more or less quantifiable effects

The key idea is that each impact of research, as characterized above, can be understood as having beneficial or detrimental effects on social or individual welfare. Investments in science and technology should promote beneficial impacts. Given the likelihood that there will be few instances of unalloyed benefits, research administrators should be sure that benefits outweigh detrimental outcomes, and should regard research investment as a process of seeking the best possible ratio of beneficial to detrimental outcomes. This norm is commonly expressed in terms of seeking the best ratio of benefit to cost. It derives from John Stuart Mill’s statement of the utilitarian maxim, which states that the action that is ethically justified is the one that achieves "the greatest good for the greatest number."

The premise of utilitarianism is to act so as to achieve the greatest good for the greatest number. From this point of view investments in science and technology should promote beneficial impacts whilst minimizing negative ones

Agricultural biotechnology was subjected to a great deal of prospective evaluation research using a decision model suggested by the utilitarian maxim. Milton Hallberg (1992) produced a volume of essays reviewing the extensive evaluation literature on just one product of biotechnology, recombinant bovine somatotropin. Since risks to food consumers were judged to be very small, and benefits in the form of lower food prices were estimated to be significant, the acceptability of biotechnology to food consumers was never taken to be an important criterion in ex ante evaluations. As is now well documented, however, even consumers who expressed interest in trying GM foods were deeply resentful of a marketing approach that denied them the opportunity to give or withhold consent. Even consumers who thought of themselves as potentially benefiting from GM foods nevertheless insisted upon the right to decide for themselves whether to eat them or not (see Durant, Bauer and Gaskell, 1998).

The utilitarian model used to conceptualize the evaluation process failed to foresee public resistance to agricultural biotechnology because of its blind spots with respect to issues of coercion, consent and rights

This should be seen as an ethical flaw not in the sense that anyone in agricultural research planning or evaluation wilfully acted unethically. Instead, the utilitarian orientation of the model being used to conceptualize the evaluation process had blind spots with respect to issues of coercion, consent and rights. The weakness of utilitarian thinking with respect to such issues has been recognized for many years. Although it may be possible to adapt consequence-evaluating social science research models to more accurately reflect the importance of consent and rights, it may be easier to conceptualize the issues of rights and consent as distinct elements of the evaluation process. That is, in addition to evaluating the causal effects of a research programme, evaluators should ask how the research (or its expected products) could affect the rights held by affected parties prior to the implementation of a new technology. If rights will be affected, the burden on research administrators is to make a good faith effort to secure the participation and consent of affected parties, rather than simply to convince themselves that the expected benefits of the research outweigh its costs (see Thompson 1997b; 2000).

A second weakness of standard utilitarian research evaluation is that in looking so intently at the future consequences of a research activity, it tends to neglect the past. However, in the ethical evaluation of human action, the things that lead up to an action can be much more important than the consequences that follow it. In particular, the intentions, interests and motives of the agents who perform actions make all the difference in deciding whether certain acts are right or wrong. One act of pulling the trigger may be heroic or forgivable while another is murderous, and an analysis of the consequences involved in either case may not be particularly sensitive to the difference between the two.

A second weakness of standard utilitarian research evaluation is that it fails to take into account intentions, interests and motives. Thus it failed to foresee the negative impact of the public’s perception of the primary actors as seeking to control and dominate research on seeds and agricultural technology

With respect to agricultural biotechnology in the 1980s and 1990s, the primary actors appear to have been positioning themselves to control and dominate research on seeds and agricultural technology well into the future. Certainly the commercial firms such as Monsanto, Novartis and Dupont have expected to profit from their research activities, but there have been arguments that their activities extend well beyond simple profit seeking into more extensive attempts to control the food and agricultural system. Furthermore, while one expects commercially funded research to meet the test of profit, the new agricultural biotechnologies have been accompanied by a growing recognition of the way that profit-seeking motivations may affect research policy in the publicly governed research sector, too. These developments may have undermined the credibility of key actors in the agricultural biotechnology arena.

Barré (1999) calls for more descriptive attention to the networks of actors involved in research and development, and agricultural biotechnology may provide a case in point. With a few exceptions, agricultural research evaluation conducted prior to 1980 was largely inattentive to the interests, intentions and motives of actors. Even when it was attentive, it was infrequently integrated into prospective evaluations of agricultural biotechnology in a manner that would have helped research administrators appreciate the way in which evolving industry-public sector research relationships might compromise the acceptability of the technology in the future. As with respect to consent, the reason for this is at least arguably to be found in the limitations owing to the utilitarian bias of the primary social science instruments for conducting and communicating the evaluation of agricultural research.

The environmental consequences of agricultural biotechnology have been hotly contested ever since the first field tests for genetically engineered ice-nucleating bacteria were proposed in the early 1980s. As such, research administrators, planners and evaluators could not have overlooked the need to assess environmental impact. Furthermore, the utilitarian approach to evaluation of research is certainly capable of incorporating evaluation of environmental impact. Although the technical specification of environmental risk can become quite complex, risk is treated as an expected value in most utilitarian or consequential approaches to evaluation. Under the expected value approach, detrimental outcomes are evaluated probabilistically, and the risk of a technology is rendered as a function of probability and degree of harm.

Although the utilitarian approach is able to incorporate an evaluation of environmental impact, the fact that the benefits of reducing environmental risks are public goods means standard markets do not provide incentives for reducing them

The expected value interpretation of risk suggests two related strategies for coping with the unintended environmental consequences of technology. One is to treat risks conceptually as forms of cost, which are then subjected to cost-benefit style optimizing. In one sense there is nothing unique about environmental consequences, since in standard cost-benefit analysis virtually all costs and benefits are probabilistic in nature and are modelled for evaluation as expected values. However, environmental costs are thought to exhibit characteristics that make them differ from straightforward costs borne directly by producers and consumers in that benefits of reducing environmental risks and costs are public goods. As such, standard markets do not provide incentives for reducing environmental risks and costs.

The upshot is that when risks are interpreted as expected values, unintended environmental consequences are understood as regulatory problems. Government regulatory agencies evaluate environmental impact and regulate markets so that public benefits of reduced pollution and environmental damage are realized. This general approach was accepted by both public and private research entities undertaking product development of agricultural biotechnologies in the 1980s and into the early 1990s (Bosin, 1990; Townsend, 1993). However, the evaluation of environmental risk was also conceptualized as an essentially technical activity, rather than one in which ethics would play a role. In this respect, the utilitarian presumptions of the prevailing approach to agricultural research and technology assessment may have led decision makers astray. By 1999, public confidence in regulators’ ability to address environmental risks of agricultural biotechnology had fallen considerably (Durant, Bauer and Gaskell, 1998).

In moving immediately to a characterization of risk in terms of probability and degree of environmental harm, the expected value orientation of the utilitarian evaluation model militates against a precautionary mode of decision making

The underlying ethical issue concerns the appropriate standard to apply in evaluating possible environmental impact. In moving immediately to a characterization of risk in terms of probability and degree of environmental harm, the expected value orientation of the utilitarian evaluation model militates against a precautionary mode of decision making. In the precautionary model, uncertainties are given precedence over what is known about possible benefits and costs; environmental risk is not characterized in terms of known potential outcomes and their measured probability of occurrence. In comparison with a predilection for choosing the optimal trade-off between cost and benefit, the precautionary stance is considerably more open-ended. As an approach to evaluation, it takes pains to present evaluative information in a way that does not lead the audience to presume that known risks should be the basis for decision, rather than unknown risks. In doing so, it leaves open the possibility that a non-comparative, non-optimizing criterion for addressing environmental impact may be the appropriate one for decision-makers to use in a given situation (Ticknor, 1999).

Based on robust studies of previous agricultural technologies, social scientists were quick to predict that agricultural biotechnology would be a contributing factor in the restructuring of agriculture, the continuing decline of small-scale family-operated farms, and the concentration of economic power among a few agribusiness firms (Kalter, 1985; Kloppenburg, 1985; Kenny, 1986). These predictions precipitated several years of heated debate in the United States, mostly focused on the acceptability of social consequences associated with recombinant bovine somatotropin. Some argued that benefits to consumers counterbalance social costs incurred by farmers and rural communities (Tweeten, 1991). Others argued that these social consequences should be regulated in much the same ways as public health and environmental impact (Lacy and Busch, 1991). The issue was effectively decided in the United States when a 1994 executive report concluded: "At no time in the past has the Federal Government prevented a technology from being adopted on the basis of socio-economic factors," (U.S. Executive Office of the President, 1994, pp. 35-6).

By leaving the market to decide the acceptance of a technology, U.S. Government policy permits the productive resources to be allocated according to market forces governing capital investments. Ethical and political debate about the distribution of wealth in society is then focused on redistributive tax policy and entitlement programmes

There is an implicit ethical orientation to the U.S. Government’s approach to social consequences. It is that social consequence debates revolve around the issue of distributing the costs and benefits of biotechnology. By not intervening in market mechanisms that influence producers’ decisions to adopt or reject technology, U.S. Government policy permits the productive resources to be allocated according to market forces governing capital investments. Ethical and political debate about the distribution of wealth in society is then focused on redistributive tax policy and entitlement programmes. It is difficult for government agencies to manipulate complex capital investment and producer decision making in a predictable fashion, and since private firms make these decisions, the public interest in the outcome of producer decision making may not be immediately obvious. In contrast, taxation and entitlements are debated in a political forum that is more obviously related to distributive justice. Furthermore, the U. S. consensus predilection for capitalism over centralized planning in the production sectors of the economy militates strongly in favour of this approach.

This argument would be persuasive were it not for the fact that technological innovations can result not only in restructuring that reflects production efficiencies and the distribution of wealth, but also in a transformation of capabilities that is equivalent to a redefinition of basic constitutional rights, (Winner, 1983). The system of property rights in effect at any given time reflects a social and legal consensus on rules for access, control, exchange and the right to profit from use of goods, but it also reflects basic physical and biological capabilities with respect to the goods in question. Legal systems of property rights include stipulations about the right to transport portable goods, for example, but not for parcels of land or bodies of water because these are, by their very nature, permanent fixtures of a given landscape. However, technological innovations in biotechnology have effectively disaggregated traits (such as specific plant or animal characteristics or in the case of the so-called Terminator gene, biological reproductive capacity itself) so dramatically as to have vitiated property rights that farmers would have once held in seeds or animal breeding stock. As such, it is doubtful that the old social consensus on rules for property exchange and capital investment in agricultural production can simply be carried over into a post-biotechnology world without opportunity for renegotiations and political debate (see Kloppenburg, 1988).

It is doubtful that the old social consensus on rules for property exchange and capital investment in agricultural production can simply be carried over into a post-biotechnology world without opportunity for renegotiations and political debate

Clearly, the economic, political and sociological issues that would bear on such a debate are complex, and any adequate discussion of them is far beyond the scope of the present article. The relevant point is simply that evaluation of socio-economic consequences of science and technology will be incomplete if it fails to address the possibility of restructuring at the constitutional level. This is, arguably, a point that has either been overlooked or repressed in the last decade of debate over the social consequences of agricultural biotechnology.

Perhaps the most obvious need for ethics in the evaluation of socio-economic impact from scientific research and technological innovation is simply that scientific integrity demands a fair and truthful evaluation process. To casual observers, it may seem as if ethical evaluation demands little more than rigour and honesty on the part of evaluators. However, the case of agricultural biotechnology indicates that there are a number of ways in which ethical bias can be hidden within evaluation methodology itself. Even those analysts who strive assiduously for objectivity can produce evaluation studies that fail to identify issues where the end products of scientific innovation create problems with respect to individual consent, or the intentions and conduct of key actors. Methodologies that presume a particular approach to the quantification of risk may neglect or conceal ethical issues that emerge when a precautionary approach is seen as the alternative to risk-benefit trade-off optimization. Assessment methodologies may also conceal the deeper significance of socio-economic consequences when they are represented solely as issues in distributive justice.

food and agriculture biotechnology, research evaluation, research policy choices, risk, social impact

References

• Barré, R., Public research programmes: Socio-economic impact assessment and user needs. The IPTS Report 40 (Dec.):5-9, 1999.
• Berardi, G. M., and Geisler, C.C., Eds. The Social Consequences and Challenges of New Agricultural Technologies. Boulder, CO: Westview Press, 1985.
• Bosin, M., Policymakers address biotechnology: Issues and responsibilities. In D. J. Webber, Ed. Biotechnology: Assessing Social Impacts and Policy Implications. Westport, CT: Greenwood Press, Pp 167-180, 1990.
• Durant, J., Bauer, M.W., and Gaskell, G., Eds. Biotechnology in the Public Sphere: A European Sourcebook. London: The Science Museum, 1998.
• Hallberg, M., Ed. Bovine Somatotropin and Emerging Issues. Boulder, CO: Westview Press, 1992.
• Huffman, W. E., and Evenson, R., Science for Agriculture: A Long Term Perspective. Ames, IA: Iowa State University Press, 1993.
• Johnson, Glenn L. and Bonnen, J.T. Social Science Agricultural Agendas and Strategies. East Lansing, MI: Michigan State University Press, 1991.
• Kalter, R., The new biotech agriculture: unforeseen economic consequences, Issues in Science and Technology 13:125-133, 1985.
• Kenny, M., Biotechnology: The University-Industrial Complex. New Haven, CT: Yale University Press, 1986.
• Kloppenburg, J. Jr., The social impacts of biogenic technology in agriculture: Past and future, In G. M., Berardi and C. C., Geisler, Eds. The Social Consequences and Challenges of New Agricultural Technologies. Boulder, CO: Westview Press, Pp. 291-321, 1985.
• Kloppenburg, J. Jr., First the Seed: The Political Economy of Plant Technology: 1492-2000. Cambridge: Cambridge University Press, 1988.
• Lacy, W., and Busch, L., The fourth criterion: Social and economic impacts of agricultural biotechnology, In J.F. MacDonald, Ed. Agricultural Biotechnology at the Crossroads: Biological, Social and Institutional Concerns. Ithaca, NY: National Agricultural Biotechnology Council, 1991.
• Rossmiller, G. E., Agricultural Sector Planning — A General System Simulation Approach. East Lansing, MI: Michigan State University Press, 1978.
• Ruttan, Vernon., Agricultural Research Policy. Minneapolis: U. Minnesota Press, 1985.
• Thompson, Paul B., (1997a) Food Biotechnology in Ethical Perspective. London and New York: Chapman and Hall Aspen Publishing, 1997.
• Thompson, Paul B., (1997b). Food Biotechnology’s Challenge to Cultural Integrity and Individual Consent, Hastings Center Report, 27(4): 1997, pp 34-38.
• Thompson Paul B., Agricultural biotechnology, ethics, risk and individual consent, Encyclopedia of Ethical, Legal, and Policy Issues in Biotechnology New York: Wiley, 2000.
• Thompson, P. B., Lain, Ellis G., and Stout, B.A., Values in the agricultural laboratory, In P. B. Thompson and B. A. Stout, Eds. Beyond the Large Farm: Ethics and Research Goals for Agriculture Boulder, CO: Westview Press, 1991, pp. 3-32.
• Ticknor, Joel A., A map toward precautionary decision making, In C. Raffensparger and J. Ticknor, Eds. Protecting Public Health and the Environment: Implementing the Precautionary Principle. Washington, DC: Island Press, 1999, pp 162-186.
• Townsend, R., APHIS and agricultural research, In R. D. Weaver, Ed. U.S. Agricultural Research: Strategic Challenges and Opportunities. Bethesda, MD: Agricultural Research Institute, 1993, pp. 141-154.
• Tweeten, L., The costs and benefits of bGH will be distributed fairly, The Journal of Agricultural and Environmental Ethics 4: 1991, pp108-120.
• United States Department of Agriculture Economic Research Service. Agricultural Research and development: Public and private investments under alternative markets and institutions. Staff Paper 9517. Washington, DC: USDA, 1995.
• United States Executive Office of the President. Use of Bovine Somatotropin (BST) in the United States: Its Potential Effects – A Study Conducted by the Executive Branch of the Federal Government. Washington, DC: Government Printing Office, 1994.
• Winner, L., Techné and politea: the technological constitution of society. In P. T. Durbin and F. Rapp, Eds. Philosophy and Technology Dordrecht: D. Reidel, 1983, pp. 97-111.

Contact

Paul B. Thompson, Joyce and Edward E. Brewer Distinguished Professor, Department of Philosophy, Purdue University, West Lafayette, USA

E-mail: Contact Form