State paternalism and the neutrality-perfectionism debate

The purpose of this thesis is to compare the paternalistic implications of two opposing political theories: neutrality and perfectionism. Neutrality holds that the state must not appeal to conceptions of the good as the justification for its decisions. Perfectionism rejects this constraint. Applied to paternalism, perfectionism makes it permissible for the state to appeal to a conception of the good when it acts paternalistically, that is, interferes with a person for his own good. Neutrality requires that paternalism must instead be guided by considerations that are in some sense neutral between various conceptions of the good. Perfectionism supports forcing people into worthwhile conceptions of the good. It provides motivation for a wider range of paternalistic policies than neutrality does. A number of perfectionist strategies for resisting this conclusion are examined. The first is to appeal to the value of autonomy as a component of well-being. Paternalism, some perfectionists argue, undermines the value of living an autonomous life. I try to show how various arguments for the value of autonomy, even if accepted, fail to rule out a wide range of paternalistic policies. A second strategy is to appeal to the endorsement constraint. According to this, a person's well-being cannot be advanced by forcing her into activities that she does not endorse as valuable. I argue that the endorsement constraint is not plausible in its strong form, and that whilst a weaker form is plausible, it allows a wide range of paternalism. A number of other strategies, such as appealing to the value of activity, claiming that many worthwhile activities require people to approach them with the right intentions for those activities to be for their own good, and that paternalism undermines trust in the government are examined and criticised. These difficulties do not mean that perfectionism should be rejected. But they do support setting aside conceptions of the good when the state acts paternalistically, whilst not necessarily ruling out perfectionism in non-paternalistic state action. This conclusion is strengthened in two ways. First, by taking Rawlsian contractualism as a method of elucidating neutrality, it is shown that neutrality supports a plausible principle of paternalism. Second, a number of recent attempts to set out necessary conditions for justified paternalism, such as that liberty must be balanced against wellbeing, that the consent of the patemalised is needed, and that the conduct must be nonvoluntary, are examined. The arguments for these conditions all suggest that neutrality is a necessary condition for justified paternalism. The conclusion of the thesis is two-fold. A conclusion about the neutralityperfectionist debate is that neutrality is required for paternalistic state action whilst perfectionism may be acceptable in the non-paternalistic sphere. A conclusion about state paternalism is that it is justified only if guided by neutral considerations.