The advent of biotechnology in the global agri-food system has challenged our conventional conception of science, the social choices underlying the current risk analysis system and, ultimately, our confidence in the scientifically based risk assessments being undertaken by our regulators. The food safety evaluation systems operating in most OECD countries are primarily scientifically based processes that combine the identification and characterization of hazards with assessments of exposure to characterize risk. In essence, they purport to objectively assess the probabilistic outcomes of discrete adverse events, for the most part abstracting from issues related to risk management or risk communications. The practice is that governments establish a risk threshold for products or classes of products that would reject new products with unacceptable risks but would allow those with acceptable impacts to enter the market. In a rational world where scientific judgments reigned, one would expect that the risk threshold would be comparable across all socially mediated activities and that governments would thereby standardize risks for citizens and consumers. Reality is much different. Public policies implicitly weigh some risks higher than others, with the result that the implicit public valuation of life varies widely across different categories. Voluntary, familiar and predictable risks often generate no public concern, which works to mute the perceived risk and can lead to under-investment in assessing or managing that risk. In contrast, involuntary, exotic and random risk can generate outrage, amplifying the perceived risk and leading to over-investment in assessing or managing that risk. This chapter suggests that the risk assessment formula should be expanded to include a socializing factor (called the outrage factor) to account for the divergences in public choice related to risk. The application of scientifically based assessment processes to biotechnology raises three interrelated issues about risk assessment for new technologies, which have three corresponding debates involving different groups of actors. First, the introduction of a new transformative technology has raised questions about whether the use of biotechnology and the resulting products can be properly accommodated within what Kuhn called 'normal' or conventional science. For the most part, this debate involves the global scientific community, traditionally led by tenured academics but increasingly involving industry and regulatory scientists. Second, normal or conventional science is incorporated in the risk analysis system through a political process. The translation of scientifically based perspectives into operational rules, policies and procedures involves both big P and small p politics. Most of the big decisions on what, how, and when to assess risks related to a new technology or a related new product are undertaken in thoughtful and transparent political processes, involving a wide range of non-scientific actors. Third, at the product assessment level, the other two debates are largely ignored, and the focus is on applying the objective standards derived from the policy discussion in the context of specific technologies and products in actual market situations. This process relies on contextual knowledge, which can sometimes amplify the measurement of the risk and in other circumstances can mute the assessment. The objective at this level of engagement is for the system to deliver objective, transparent, accountable judgments on specific products. To this end, there should be only limited engagement in the scientific or policy debates raised above. The focus should be on administrative delivery in the context of normal science as put into practice by the policy system. The difficulty is that while the rules may be clear, their application to specific products and markets may necessitate many incremental decisions be taken on the fly. This assessment of the risk analysis framework, and risk assessment in particular, suggests that the artificial divide between risk assessment, risk management and risk communications is not fruitful and when addressed to transformative technologies, can be counterproductive to understanding of the challenges, issues and opportunities related to transforming technological change.
