I look forward to hearing your thoughts on the attached paper at the session on Friday. The paper is part of a larger project on how we should regulate conduct that is socially productive, but poses some risk of harm to others. The official technique for risk regulation in the modern administrative state is some form of cost/benefit analysis: we tote up the expected social benefits and expected social costs of alternative courses of conduct, and opt for that course that is expected to generate the largest aggregate benefits (net of costs). There is a vast and growing critical literature on the normative, conceptual, and administrative problems with cost/benefit analysis. I agree with many of those criticisms. But my particular interest in this project is the objection raised to any aggregative procedure, on the familiar deontological ground that it fails to respect the distinct rights and interests of individuals. Accepting the moral impulse behind that objection as understandable, maybe even compelling, the question I want to address is whether nonconsequentialists have offered—or could offer‐‐ any coherent alternative to aggregation in this context.

The short answer is, I don’t think so. I want to underscore that this is not a broadside attack on nonconsequentialist principles or a categorical defense of consequentialism. It is a domain‐specific concern. For reasons I touch on briefly in this chapter, I think that deontological principles by their nature are the wrong tools to solve the fundamental problems raised by accidental (unintended) harms, and that this reality has been obscured by the limited and somewhat peculiar focus of the immense philosophical literature on harm to others.

While these larger concerns are in the background, this chapter mostly focuses on the central deontological objection to aggregative solutions to regulating potentially harmful conduct: that it violates our duty not to act in a fashion that will result in harm to others.

Please do not cite to without permission of the author.