Jamie Terence Kelly

In a number of recent publications, Cass Sunstein and Richard Thaler have argued for a novel approach to the design of public policy. Their proposal has received a great deal of attention, both within academic circles and the public at large. Drawing upon evidence from behavioral economics and empirical psychology, the authors attempt to demonstrate that the conventional antagonism between libertarians and paternalists within political theory dissolves in conditions that obtain widely in public decision-making. Where free choice and the promotion of individual welfare can coexist, the authors believe that designers of public policy ought to be libertarian paternalists. In this paper I criticize their proposal on grounds that the authors are unable to sufficiently motivate the paternalistic element of their approach. I argue that the empirical evidence cited by the authors is capable of supporting a number of competing approaches, including what I call libertarian utilitarianism and libertarian justice. Since the evidence that the authors draw upon does not provide us any grounds for selecting between these rival approaches, I conclude that Sunstein and Thaler are unable to provide us with a convincing guide for the design of public policy. In order to show that this is the case, I consider three arguments in favor of libertarian paternalism, and find each lacking. I end with some comments about what we can properly conclude on the basis of Sunstein and Thaler’s arguments.

In a number of recent publications, Cass Sunstein and Richard Thaler have argued for a novel approach to the design of public policy. Their proposal has received a great deal of attention, both within academic circles and the public at large. Drawing upon evidence from behavioral economics and empirical psychology, the authors attempt to demonstrate that the conventional antagonism between libertarians and paternalists within political theory dissolves in conditions that obtain widely in public decision-making. Where free choice and the promotion of individual welfare can coexist, the authors believe that designers of public policy ought to be libertarian paternalists. In this paper I criticize their proposal on grounds that the authors are unable to sufficiently motivate the paternalistic element of their approach. I argue that the empirical evidence cited by the authors is capable of supporting a number of competing approaches, including what I call libertarian utilitarianism and libertarian justice. Since the evidence that the authors draw upon does not provide us any grounds for selecting between these rival approaches, I conclude that Sunstein and Thaler are unable to provide us with a convincing guide for the design of public policy. In order to show that this is the case, I consider three arguments in favor of libertarian paternalism, and find each lacking. I end with some comments about what we can properly conclude on the basis of Sunstein and Thaler’s arguments.

Robert Talisse’s recent attempt to justify liberal democracy in epistemic terms is in many ways a breath of fresh air. However, in the present paper we argue that his defense faces two inter-related problems. The first problem pertains to his defense of liberalism, and owes to the fact that a commitment to the folk-epistemological norms in terms of which he makes his case does not commit one to partaking in liberal institutions. Consequently, our (alleged) commitment to the relevant epistemic norms does not justify liberal democracy. The second problem pertains to his defense of democracy. The problem is that, if Talisse provides what we take to be the most plausible response to the first problem, framed in terms of his acceptance of a form of epistemic perfectionism, he is able to maintain his commitment to liberal institutions, but at the price of leaving democracy behind in favor of what we will refer to as a liberal epistocracy.

The past thirty years have seen a surge of empirical research into political decision making and the influence of framing effects — the phenomenon that occurs when different but equivalent presentations of a decision problem elicit different judgments or preferences. During the same period, political philosophers have become increasingly interested in democratic theory, particularly in deliberative theories of democracy. Unfortunately, the empirical and philosophical studies of democracy have largely proceeded in isolation from each other. As a result, philosophical treatments of democracy have overlooked recent developments in psychology, while the empirical study of framing effects has ignored much contemporary work in political philosophy. In Framing Democracy, Jamie Terence Kelly bridges this divide by explaining the relevance of framing effects for normative theories of democracy. Employing a behavioral approach, Kelly argues for rejecting the rational actor model of decision making and replacing it with an understanding of choice imported from psychology and social science. After surveying the wide array of theories that go under the name of democratic theory, he argues that a behavioral approach enables a focus on three important concerns: moral reasons for endorsing democracy, feasibility considerations governing particular theories, and implications for institutional design. Finally, Kelly assesses a number of methods for addressing framing effects, including proposals to increase the amount of political speech, mechanisms designed to insulate democratic outcomes from flawed decision making, and programs of public education. The first book to develop a behavioral theory of democracy, Framing Democracy has important insights for democratic theory, the social scientific understanding of political decision making, economics, and legal theory.

This paper points out an ambiguity in Cass Sunstein’s recent work concerning whose lives and interests are to be promoted by libertarian paternalism, and argues that this ambiguity stems from a lack of clarity regarding how we should understand the relevance of the heuristics and biases literature for democratic theory. The paper attempts to extract from Sunstein’s work an account of how we should go about identifying biases in social choices. It argues that Sunstein’s view on this issue has changed over the years, and identifies a trend in his thinking moving away from libertarian paternalism and toward libertarian welfarism.

What is the optimal size of a democratic society? While not taking an explicit stand on this issue, Hélène Landemore's model of democracy in Democratic Reason suggests that democracies ought to be small, certainly smaller than many existing states. If, as Landemore argues, we must rely on the random selection of representatives, then we should be concerned about both the size of the population and the way cognitive diversity is distributed within it. Given the realities of party politics and media framing, this means that smaller political societies will yield wiser decisions than very large ones.

Robert Talisse’s recent attempt to justify liberal democracy in epistemic terms is in many ways a breath of fresh air. However, in the present paper we argue that his defense faces two inter-related problems. The first problem pertains to his defense of liberalism, and owes to the fact that a commitment to the folk-epistemological norms in terms of which he makes his case does not commit one to partaking in liberal institutions. Consequently, our (alleged) commitment to the relevant epistemic norms does not justify liberal democracy. The second problem pertains to his defense of democracy. The problem is that, if Talisse provides what we take to be the most plausible response to the first problem, framed in terms of his acceptance of a form of epistemic perfectionism, he is able to maintain his commitment to liberal institutions, but at the price of leaving democracy behind in favor of what we will refer to as a liberal epistocracy.