Conceptual analysis and special-interest science: toxicology and the case of Edward Calabrese

One way to do socially relevant investigations of science is through conceptual analysis of scientific terms used in special-interest science (SIS). SIS is science having welfare-related consequences and funded by special interests, e.g., tobacco companies, in order to establish predetermined conclusions. For instance, because the chemical industry seeks deregulation of toxic emissions and avoiding costly cleanups, it funds SIS that supports the concept of “hormesis” (according to which low doses of toxins/carcinogens have beneficial effects). Analyzing the hormesis concept of its main defender, chemical-industry-funded Edward Calabrese, the paper shows Calabrese and others fail to distinguish three different hormesis concepts, H, HG, and HD. H requires toxin-induced, short-term beneficial effects for only one biological endpoint, while HG requires toxin-induced, net-beneficial effects for all endpoints/responses/subjects/ages/conditions. HD requires using the risk-assessment/ regulatory default rule that all low-dose toxic exposures are net-beneficial, thus allowable. Clarifying these concepts, the paper argues for five main claims. (1) Claims positing H are trivially true but irrelevant to regulations. (2) Claims positing HG are relevant to regulation but scientifically false. (3) Claims positing HD are relevant to regulation but ethically/scientifically questionable. (4) Although no hormesis concept (H, HG, or HD) has both scientific validity and regulatory relevance, Calabrese and others obscure this fact through repeated equivocation, begging the question, and data-trimming. Consequently (5) their errors provide some undeserved rhetorical plausibility for deregulating low-dose toxins. K. Shrader-Frechette (B) Department of Philosophy and Department of Biological Sciences, Center for Environmental Justice and Children’s Health, University of Notre Dame, 100 Malloy Hall, Notre Dame, IN 46556, USA URL: www.nd.edu/∼kshrader e-mail: [email protected]