The German government's Council of Economic Advisors soberly concluded in 1991 that "The dynamic growth of chlorine chemistry during the 50s and 60s represents a decisive mistake in twentieth century industrial development." In Pandora's Poison, Joe Thornton traces the mundane roots and astonishing consequences of that error, dissects Western environmental authorities' blindness to it, and argues systematically that society should -- and can -- phase out major uses of the problematic element.

Thornton, an evolutionary biologist at Columbia University's Earth Institute, previously was a chemical-pollution researcher for Greenpeace (where one of us, Jeff Howard, was his colleague). There he learned to understand environmental issues as complex knots of scientific and sociopolitical forces, and Pandora's Poison is a technically and politically sophisticated analysis of one such knot. This vividly written, deftly argued, and extensively documented account draws on a wide range of disciplines: biology, toxicology, epidemiology, environmental chemistry, industrial chemistry, industrial history, economics, environmental policy, science and technology policy, political theory, sociology of science.

Thornton's accomplishment is turning these disparate strands into a coherent whole, a rich and cohesive account of chlorine chemistry and the dilemmas it poses for individual organisms, for ecosystems, and for democratic societies. An introductory essay neatly summarizes the overall argument (and would make a worthy addition to undergraduate syllabi in most of the fields listed above). The text's major points are:

**Organochlorines are hazardous as a class. Adding chlorine to carbon-based molecules makes them more fat soluble, allowing them to accumulate in organisms; and it almost always makes them more toxic. "The fundamental act of the chlorine industry -- the chlorination of organic compounds -- is biologically and ecologically problematic" (p. 71).

**Organochlorines represent a major threat. Organochlorine contamination "poses a long-term, global hazard to human health and the environment. [A] growing body of evidence suggests that global toxic pollution is already contributing to a slow, worldwide erosion of the health of humans and other species" (p. 6).

**The current regulatory paradigm is defective. The Risk Paradigm (RP), as Thornton calls it, attempts to scientifically micromanage and technologically control individual pollutants emitted from individual sources, so as to prevent well-defined health impacts in discrete local populations. It is ill equipped to respond to the global, multigenerational threat posed by chlorine -- and congenitally blind to the need to do so.

**Data gaps and illusions abound. The RP labors vainly at the Sisyphean task of filling enormous gaps in toxicological, epidemiological, and ecological data. More than 11,000 organochlorines are produced commercially, and thousands of others occur as byproducts. Because there is no way to study all these chemicals and their interactions, the Risk Paradigm's concept of "acceptable exposure" levels is bankrupt. So-called no-effect levels deduced by exposing laboratory animals to individual chemicals "are artifacts, reflections of the limits, the conventions, and the politics of toxicology" (p.75).

**A new regulatory paradigm is needed. Thornton proposes an Ecological Paradigm (EP) based on the precautionary principle, under which protective action is required even in the absence of scientific consensus (Freestone and Hey 1996; Raffensperger and Tickner 1999). Thornton's EP would supplement this principle with "zero discharge" of chemicals that are persistent or bioaccumulative; "clean production," the systematic reduction of the use of toxic substances; and "reverse onus," which (as in U.S. regulation of pharmaceuticals) shifts the burden of proof from public to manufacturer.

**Chlorine should be phased out. Organochlorines should be treated as a class, and policy should move toward a "chlorine sunset" -- a "carefully planned process of technological conversion, a transformation of our industrial infrastructure that will take several decades" (pp. 13-14). "[S]afe and effective alternatives exist now for the vast majority of chlorine uses," Thornton argues (p.364).

**The new paradigm opens space for debate on values. "By shielding ethical and political issues of extraordinary consequence from public debate," the RP "undermines democratic rights and processes" (pp. 417-18). The EP, in contrast, opens the values dimension of chemical policy making to public view and control.

Beyond chlorine, the concept of chemical class may be trickier than Thornton acknowledges. Moreover, it is not clear that reverse onus, clean production, and zero discharge can "become the general method for making environmental decisions" (pp.427-8). For an elaboration of the precautionary principle that is coherent across a wide range of issues, it may be necessary to develop an additional tier of subprinciples that generalize the theoretical foundation of zero discharge and clean production and that act as bridges between chemical regulation and other regulatory domains.

The final chapter draws substantively on the work of Gieryn, Jasanoff, Epstein, Sclove, Winner, and other science and technology studies scholars, but Thornton might have gone farther, for example, by drawing connections to the literature on social steering of technology. His concern about persistent pollutants finds support in Morone and Woodhouse's (1986) concept of intelligent trial and error, which requires that technologies be designed and monitored so as to permit rapid error correction. Collingridge's (1992) work on preserving flexibility is equally pertinent.

The chemical industry, engaged in a vigorous public relations and lobbying campaign to defend chlorine, is portraying Pandora's Poison as anti-science extremism (e.g., www.pandoraspoison.org). However, most scholars in the policy sciences, environmental studies, and technology policy will find the book a reasoned and intellectually honest fusion of activist instinct, scientific rigor, and social-scientific insight. Thornton provides a provocative and disturbing look at one of the most important technological trajectories of the last century, at the industrial mindset that created and extended it, and at the regulatory mindset that has allowed it to escape close scrutiny.

References
David Collingridge (1992), The Management of Scale (Routledge, London).
David Freestone and Ellen Hey (editors) (1996), The Precautionary Principle and International Law
(Kluwer Law International, The Hague).
Joseph G. Morone and Edward J. Woodhouse (1986), Averting Catastrophe: Strategies for Regulating
Risky Technologies (University of California, Berkeley).
Carolyn Raffensperger and Joel Tickner (editors) (1999), Protecting Public Health and the Environment: Implementing the Precautionary Principle (Island, Washington DC).
*******************************************************
Jeff Howard wrote this as a PhD student (Tel: +1 817 536 1855; E-mail: howarj@rpi.edu).
Edward J. Woodhouse is Associate Professor Dept. of Science and Technology Studies, Rensselaer Polytechnic Institute, Troy, NY 12180-3590, USA. Tel: +1 516 276 8506; E-mail: woodhouse@rpi.edu