Food Fights: Canadian regulators are under pressure to face the uncertainties of genetically modified food.
By Katherine Barrett
Reprinted courtesy of Alternatives Journal: Environmental Thought, policy and Action
Vol. 28 No1, Winter 2002.
In the fall of 2000, US newspapers reported that genetically modified (GM) corn not approved for human consumption had been found in taco shells, corn chips and other corn products  The corn, called "Starlink", was engineered with a toxin from bacteria that allows the plant to withstand attacks by insect pests. The US Environmental Protection Agency (EPA) had previously assessed Starlink for its potential to cause allergic reactions. They concluded that the GM protein involved was "tougher" than similar proteins already in the human food chain, that it did not readily break down in the gut, and therefore that it may pose a greater risk of allergic response. Consequently, US government regulators approved Starlink for animal
feed but not for human use, a so-called "split" approval .
Despite this regulatory restriction, Starlink corn found its way to grocery store shelves and restaurant tables. Dozens of people reported allergic or other adverse reactions. Distributors began recalling over 300 corn products, but the enormous difficulty of extracting Starlink from seed stocks, grain elevators, shipping routes, food processing plants and supermarkets soon became apparent.
The amount of Starlink corn in Canada's food supply is not known. In March 2001, Agriculture Minister Lyle Vanclief admitted that although Starlink has not been approved in Canada "some of it did get into the animal feed system" and there was a "slim chance" it had entered the human food chain .
The Starlink case is a wake-up call for those who believe the risks of GM food can be easily predicted and controlled. The complexity of tracing and testing Starlink corn highlights key questions in the ongoing debate over GM food: does genetic modification pose increased risks of toxic and allergic reactions? Are particular populations, such as children, more susceptible? What are the potential environmental effects of GM organisms? How can we test for potential harms, and what standards should be used to determine adequate safety?
As the world's third largest producer of GM crops, Canada has huge stakes in the answers to such questions. Serious doubts about the adequacy of our current approach to food biotechnology regulation have been raised, most notably in the recent report of an independent expert panel of the Royal Society of Canada . Until these doubts are addressed, Canada is unlikely to build confidence in the safety of GM foods, or our ability to identify, test and prevent potentially negative health and environmental effects from food biotechnology.
Promotion Without Precaution
Genetically modified food has been marketed in Canada since 1996. In 2000, over three million hectares - nearly five percent of Canadian farmlands - were planted in transgenic canola, soy, corn, flax and potatoes. Engineered tomatoes and squash have also been approved for import and human consumption, but are not yet grown in Canada. Additional varieties of GM crops, including sugar beet, grape vine, mustard, wheat, tobacco and alfalfa are currently under development and field testing .
The rapid expansion of Canadian agricultural biotechnology is due in large part to the concerted efforts of industry and the federal government over the past 20 years. The first National Biotechnology Strategy, launched in 1983, aimed to provide program and policy support to encourage the commercial growth of biotechnology. Financial incentives for business and funding initiatives for research fostered a globally competitive biotechnology sector in Canada. In 1998, this strategy was renewed and renamed the Canadian Biotechnology Strategy.
Whereas strategies for promoting biotechnology were devised in 1983, procedures for evaluating associated health and environmental hazards were not established until the early 1990s. The first regulatory framework, announced in 1993, set ground rules for the "science-based" assessment of GM organisms. Rather than create new legislation specific to GM organisms, policies would build on laws already in place for conventional agricultural practices. This reflects a broader philosophy of the federal government "that genetically engineered organisms are not fundamentally different from traditionally derived organisms and can be assessed using well-defined and understood principles of risk assessment"  The federal regulatory framework appears in stark contrast to justifications for government promotion and investment that stress the innovative character of genetic modification techniques.
Reluctance to acknowledge that the process of genetic engineering may pose unique health and environmental hazards is, however, increasingly out of touch with world wide movements - from local organic food markets to international environmental agreements – that recognize the novel character and potential hazards of GM organisms. In current controversies, "genetic modification" usually refers to recombinant DNA techniques that are used to isolate one or more genes from a donor organism and transfer these genes into an unrelated organism. The recent Cartagena Protocol on Biosafety recognizes the distinctive qualities of GM organisms and establishes special rules for international trade in such organisms. This protocol clearly distinguishes the methods of traditional breeding from those that overcome natural reproductive barriers to create organisms with novel combinations of genetic material.7 Similarly, the Codex Alimentarius Commission, a UN body charged with establishing food safety standards, recognizes that products of recombinant DNA technology warrant specific regulatory oversight.
Public protests have also focused on the distinctive character and potential risks of GM foods and have resulted in wildly unstable markets. Canadian canola exports plummeted in the mid 1990s as many overseas buyers refused to import GM crops. The Canadian Wheat Board, a coalition of producers, currently opposes the commercialization of transgenic wheat until markets can be assured and processes for segregating GM and non-GM varieties are in place . Large manufacturers such as McCain Foods have recently adopted a no-GM policy . Recent polls and focus groups indicate that a majority of North Americans favour mandatory labelling of GM foods, and that many would use these labels to avoid consuming GM products . These
stories of public scepticism and producer hesitancy - bolstered by the recent Starlink fiasco - are further supported by an increasing number of scientific reports that suggest organisms developed through recombinant DNA technology pose uncertain benefits and uncertain risks .
The above examples clearly demonstrate that the scientific, economic, political and cultural dimensions of the GM debate are closely tied. All point to the distinctive character of GM technology, and to the need for cautious and comprehensive regulations. But how cautious is cautious enough?
The Expert Panel's Report
The Cartagena Protocol on Biosafety was adopted in January 2000. The protocol changed the tenor of international debate on GM organisms by explicitly invoking the precautionary approach in the preamble and including precautionary language in the binding text of the agreement.
The precautionary principle states that when a technology or activity threatens public health or the environment, measures to avoid adverse effects are warranted, even if there remains scientific uncertainty about the nature and extent of harm that may occur. In other words, the principle advises us to err on the side of caution when we are unsure if our actions will result in harm. Precaution, however, does not equal inaction. The term precautionary principle is translated from the German vorsorgeprinzip, a concept of environmental policy that means foresight, and emphasizes forward planning to anticipate and prevent harm before it occurs.
Although it has been incorporated into numerous international agreements as well as the national policies of several countries (including Canada), serious application of the precautionary principle remains a contentious idea. In negotiations leading to the adoption of the Cartagena Protocol, Canada and other major producers of GM crops lobbied against inclusion of the precautionary principle . In fact, the Canadian government has argued that current regulations are sufficiently sound and are "based on the latest and best scientific knowledge we have" .
Public resistance and wavering markets have prompted the Canadian government to test these assertions through more independent means. In January 2000, Health Canada, the Canadian Food Inspection Agency and Environment Canada commissioned an expert panel of the Royal
Society of Canada to examine the scientific and regulatory capacity of federal agencies to ensure the safety of GM food.
The final report, entitled Elements of Precaution was released in January 2001, and is the first comprehensive review of food biotechnology in Canada by an independent, (i.e., non-government, non-industry) committee. The report focuses on the scientific aspects of GM food, examining the current state of knowledge and questioning our ability to predict GM risks. How, for example, can we test new GM foods for potential toxic or allergenic effects? What do we know about the environmental and health impacts of transgenic fish? How can we predict if a GM crop will become a invasive weed?
Based on this review, Elements of Precaution makes over 50 recommendations, many of which may have dismayed the federal agencies that commissioned the study. The report clearly demonstrates that GM foods pose potential health and environmental risks, and that our scientific knowledge about them remains incomplete and uncertain. Overall, the report advises that Canada adopt a more transparent and precautionary regulatory process for approving GM food.
What would a precautionary regulatory system for GM foods look like? How different would it be, in process and outcomes, from the current approval system in Canada? In the following section, I draw on the detailed scientific findings of the Royal Society report and suggest a framework for applying the precautionary principle to GM food .
Precaution In Practice
The precautionary principle is just that - a principle for guiding decisions when the consequences of our actions may be harmful but remain uncertain. There is no uniform, global recipe for turning this general principle into a codified set of rules or specific actions. On the contrary, precaution must remain rooted in and responsive to diverse social and ecological contexts. Nonetheless, it is possible and important to set broad procedural guidelines, such as those below, to ensure that implementation is not arbitrary.
These practices are symptomatic of a larger problem. Although the federal government has actively supported agricultural biotechnology over two decades, it has only recently shown interest in the views of Canadian citizens. The 1998 Biotechnology Strategy attempted to remedy this oversight by organizing a series of "stakeholder consultations" in major cities across Canada in 1998 and again in 2001. Unfortunately for democratic process, however, these have been "by-invitation" meetings, not open to the general public. The latest round was boycotted by all major non-government organizations in the country. According the Brewster Kneen, a well-known commentator on food issues who spearheaded the boycott, "We saw the consultations as
a way to manage a very limited public discussion rather than putting such significant public policy issues before the House of Commons where they belong. It has been our goal to democratize the discussion of genetic engineering, rather than treat it as a matter of special interests."
Examine the Scope of Discussion
A first step under the precautionary principle is to examine - and then to challenge - the scope of questions on the table. The mandate of the Royal Society was (as the panel clearly acknowledged) narrowly defined by the government agencies that commissioned the work: to
provide advice on the scientific and regulatory capacity that the federal government will require to ensure the safety of new food products developed through biotechnology. This mandate implies that the development and commercialisation of GM food will continue - that we simply need to refine the science, tweak the regulatory system, and do biotechnology better.
The precautionary principle invites us to ask much broader, "upstream" questions. For example, how can we achieve a healthful and sustainable agricultural system for current and future generations? Addressing this question will require setting long-term policy goals, and devising the most appropriate and effective means of meeting those goals. This process of working backwards from a desired end - or "backcasting" has been used successfully in countries such as
Sweden and the Netherlands to set far-reaching targets for environmental protection . Broadening the scope of questions will likely inspire alternative approaches to policy and technology development.
For two decades, the Canadian government has enthusiastically supported the development of GM crops through the Biotechnology Strategy. They have claimed that biotechnology will boost the economic competitiveness of our agricultural markets, that it will reduce the use of pesticides and herbicides, and that it will create a greater diversity of consumer products. Government promotion continues despite lack of clear evidence that GM crops are providing such benefits or that they are more promising than other agricultural initiatives, such as low-input or organic approaches.
Comparatively little government support has been provided for low-input or organic agriculture, the products of which are now so much in demand. In the summer of 2001, Agriculture and Agri-Food Canada announced financial support for a rapidly expanding organic Canadian agriculture industry and to establish an Organic Agriculture Centre in Nova Scotia. According to Minister Vanclief, the funding will help the agricultural sector "diversify and pursue new growth opportunities". This initiative suggests a growing recognition by the federal government that biotechnology will not solve our economic, environmental and agricultural problems.
Recognition of potential harm is a necessary step in a precautionary approach. It is critical, however, that "harm" is explicitly, openly and broadly defined because such definitions will bear on conclusions about safety. In its report, the Royal Society panel clearly addressed this problem. Although only mandated to assess the scientific aspects of health and environmental risks, the panel argued that scientific issues cannot be separated from social, economic and philosophical issues - that there are no exclusively scientific risks. Public demands for a safe as well as democratic and ethically sound food system appear to acknowledge the multiple dimensions of potential harm. In contrast, Canada's "science-based" biotechnology policies tend to restrict assessment of potential harm in several ways.
The current Canadian approach is ill-equipped to deal with cases where the level of harm is influenced by local circumstances or values that cannot be captured easily by a single scientifically defined standard. For example, the current approach discourages attention to situations where gene flow from GM crops to neighbouring fields will result in different levels of harm to conventional and organic farmers because, while both may experience ecological damage, only the organic farmers will also lose their certification and therefore their entire market. The science-based approach is particularly limited where differences in values affect the level of harm involved. This arises, for example, with the production of transgenic animals, which may be acceptable to some people but experienced as harmful to other people and to the animals themselves . Such values are considered to be non-scientific and are placed outside the boundaries of regulatory authority and responsibility.
Even within a narrower scientific framework, however, recognized harms are often limited to short-term, direct impacts, such as the toxic effect of a single isolated protein on a laboratory test
animal. The Royal Society report identified many longer-term and indirect impacts that could also result in serious harm. For example, engineering fish with growth hormone genes has resulted in a number of other unrelated and unpredictable physiological changes such as altered life span, feeding rates, disease resistance and swimming ability. Assessing the hazards of raising transgenic fish in open aquaculture pens must include consideration of cumulative effects of all of these indirect changes on GM fish and on potential wild-GM fish hybrids over several generations.
The precautionary principle "has its roots in a sense of scepticism about the ability of science, or any system of knowledge, to understand and predict fully the function of complex biological and ecological systems." This statement by the Royal Society highlights the need to analyze the types and sources of uncertainty inherent in all regulatory decisions and in the scientific studies on which they are based. GM food is no exception, and at least three types of uncertainty were identified in each chapter of the Royal Society report.
Scientific uncertainty often arises from lack of data. We know which questions to ask, we have the means to address those questions, but there remain large gaps in our knowledge because the research has not yet been conducted. This kind of "technical uncertainty" is pervasive in research on GM food. For example, we can test GM proteins for potential allergenic characteristics in the lab and can compare these proteins to databases of known allergens. Yet gaps in our current knowledge about allergencity as well as gene flow, biodiversity and other potential impacts of GM organisms make conclusive predictions about long-term consequences very difficult. The Scientific Advisory Panel on Starlink corn observed, "It is amazing how little we know about many aspects and facets of allergen issues" .
These kinds of uncertainty have long been an expected part of science and of regulatory decisions based on scientific research. It is therefore remarkable that Canadian regulatory decisions about the safety of GM organisms have consistently failed to provide any significant discussion of uncertainty. Some promise of a shift is offered in the recently drafted regulations for transgenic fish include uncertainty analysis at each step . Nevertheless, the continuing general failure to acknowledge that we don't know all the consequences of releasing GM organisms has been one reason for eroding public trust in the biotechnology industry and supportive government agencies.
A further type of uncertainty arises when we know which questions to ask, but may not have the appropriate methods or models to address those questions. This "model uncertainty" was also clearly illustrated by the Starlink case. Government regulators, scientists, assessment teams, product developers and non-government organization held differing views on the type of laboratory testing required to determine if Starlink poses a health risk. Such evaluations are further complicated by the fact that particular populations, such as children, may be more susceptible to allergenic effects and/or may be exposed to higher levels of GM products . Addressing this uncertainty would require extensive epidemiological analyses not currently performed under Canadian or US biotechnology regulations.
Often we do not know which questions to ask, much less how to address them. This deeper level of uncertainty is inevitable when dealing with the complexities of "real world" situations. For example, even if methods for laboratory tests, field trials and epidemiological analyses were clearly established, how could we then account for the diverse ecological, social and political contexts in which GM organisms are released, planted, multiplied, consumed and traded?
Take Precautionary Action
There is no single form of precautionary action. Implementing the precautionary principle does not, as some critics maintain, lead to an automatic ban on all new technologies. Rather, appropriate actions will depend on a balance of evidence regarding potential harms and benefits, as well as the level of uncertainty associated with this evidence and the availability of alternatives to the proposed technology.
The Royal Society panel recommended that government agencies assume a more precautionary stance toward GM food overall, and that they adopt a number of specific precautionary measures. In general, this would entail delaying approval of GM products that may raise serious health or environmental risks until scientific uncertainties are reduced, increasing government support for research on the health and environmental impacts of GM technology, and adopting procedures to monitor long-term effects of GM organisms.
More specific recommendations included restricting some applications (e.g., the use of antibiotic resistance genes) and placing a moratorium on others (e.g., rearing transgenic fish in open pens). The panel also recommended that Canada avoid the kind of "split" regulations that lead to the Starlink debacle in the US. Approving a crop for a restricted purpose such as animal feed is not justifiable given the lack of feasible mechanisms for segregating human and non-human food sources and tracking the effects. It is curious, in this respect, that the panel found no scientific basis for mandatory labelling of GM foods given that this could greatly facilitate the tracing of GM organisms and therefore the identification of GM-related health and environmental impacts.
Adopt a Transparent and Inclusive Process
The steps outlined above require transparent and inclusive decision making. This is a fundamental element of the precautionary principle and perhaps the strongest message of Royal Society report.
Lack of transparency has been a major failing of Canadian biotechnology policies, and a constant source of public distrust and resistance. The approval process for GM foods is essentially a closed negotiation between government regulators and the product developers. The scientific data used to approve GM foods are withheld as confidential business information. In contrast to accepted scientific practice, the data are neither publicly available nor independently reviewed.
World wide protests over globalization and related issues are showing how citizens' views will eventually be heard - if not quietly through stakeholder consultations, then loudly and clearly through rapid exchange of information, persistent demands and direct actions. The Canadian government and other proponents of biotechnology would do well to listen, learn and respond.
Katherine Barrett has a PhD in botany from the University of British Columbia, and is currently project director with the Science and Environmental Health Network, and a research associate with the Polis Project on Ecological Governance at the University of Victoria.
 See "Firm looks for bushels of unapproved grain", Washington Post (October 19, 2000), p. A1; "Biotech critics cite unapproved corn in taco shells", Washington Post (September 18, 2000), p. A2.
 For an overview of the Starlink case see: Final Report, FIFRA, Scientific Advisory Panel Meeting; and W. Lin, G.K. Price, E. Allen, Starlink: Impacts on the US Corn Market and World Trade (2001)
 Reuters, Canada says banned GM corn fed to animals" March 17, 2001
 Royal Society of Canada, Elements of Precaution. Recommendations for the Regulation of Food Biotechnology in Canada" 2001
 Statistics from: C. James,. "Global Status of Commercialized Transgenic Crops in 2000," ISAAA, Briefs, 21 (2000). For a complete list of GM crops approved for release into the environment see "Status of Regulated Plants with Novel Traits in Canada"; For list of product in field testing see Detailed table for 2000 confined field trials
 Canadian Food Inspection Agency, Concerns and Issues about Biotechnology" (2001).
 The Cartagena Protocol on Biosafety uses the term "living modified organisms" rather than "genetically modified organisms".
 Canadian Wheat Board, "CWB Biotechnology Statement" (2001)
[9 ] McCain Food Limited, Frequently Asked Questions" (2001)
 For examples see US FDA, Report on Consumer Focus Groups on Biotechnology" (2000); <a href="http://http://strategis.ic.gc.ca/SSG/bh00258e.html">Public Opinion Research into Biotechnology Issues, Second Wave
 For example see L.L. Wolfenbarger, P.R. Phifer, "The Ecological Risks and Benefits of Genetically Engineered Plants", Science, 290 (2000), pp. 2088-93; J.J. Obrycki, et al., "Beyond Insecticidal Toxicity to Ecological Complexity", BioScience, 51:5 (2001), pp. 353-61; The EU-US Consultative Forum, Final Report (2000).
 United Nations Environment Program, Draft Report of Extraordinary Meeting of the COP for the Adoption of the Protocol on Biosafety to the CBD" (1999).
 L. Vanclief, "Food Safety Will Not be Compromised", London Free Press (June 25, 1999)
 For more information on the precautionary principle see C. Raffensperger and J. Ticker, "Protecting Public Health and the Environment: Implementing the Precautionary Principle", Washington DC, Island Press, 1999
 J. Tickner, "A Map Toward Precautionary Decision-Making", in Raffensperger and Ticker, ibid, pp162-86
 Agriculture and Agri-Food Canada, Centre Set to Bolster Canada's Organic Expertise", July12, 2001 press release; see also Agriculture and Agri-Food Canada, "Vanclief Announces Funding to Help Organic Growers Seize New Market Opportunities", June 8, 2001 press release.
 Elements of Precaution, note 4, p,95
 Final Report, note 2
 Elements of Precaution, note 4, p.59; and Final Report, note 2.
 Elements of Precaution, note 4, pp.164-65.