When my family discusses how we should spend a summer holiday, we start from certain common understandings about our preferences. We prefer self-catering accommodation to hotels, and hotels to campsites. We prefer walking and looking at scenery and wildlife to big-city sightseeing and shopping. When it comes to walks, we prefer walks of six miles or so to ones which are much shorter or much longer, and prefer well-marked but uncrowded paths to ones which are either more rugged or more popular. And so on. These common understandings greatly simplify the task of choosing between holiday destinations and activities, by allowing us quickly to eliminate many options. But what does it mean to say that we prefer one thing to another?

Trygve Haavelmo's 1944 article ‘The Probability Approach in Econometrics’ is considered by most to have provided the foundations for present day econometrics (Morgan, 1990, Chapters 8 and 9). Since Haavelmo (1944), extraordinary advances have been made in econometrics. However, over the last two decades the efficacy and scientific status of econometrics has become questionable. Not surprisingly, the growing discontent with econometrics has been accompanied by a growing interest in econometric methodology.

The most influential egalitarian perspective is undoubtedly Rawls's (1971, 1993), which assigns absolute priority to the least advantaged in society (the difference principle). However, many have claimed that even though an egalitarian perspective should imply some priority to the worst off, the Rawlsian perspective is too demanding. One response to this criticism is to argue in favour of an egalitarian perspective that never assigns absolute priority to the worse off, but which still includes limited priority to those members of society in distributive conflicts. A different response to the demandingness criticism is to agree that the worse off should not always be given absolute priority, but to argue that there are some cases where they should be. In this paper, we elaborate on this view, and look at the possibility of deviating from the leximin approach within this category of egalitarian principles.

A search is under way for a theory that can accommodate our intuitions in population axiology. The object of this search has proved elusive. This is not surprising since, as we shall see, any welfarist axiology that satisfies three reasonable conditions implies at least one of three counter-intuitive conclusions. I shall start by pointing out the failures in three recent attempts to construct an acceptable population axiology. I shall then present an impossibility theorem and conclude with a short discussion of how it might be extended to pluralist axiologies, that is, axiologies that take more values than welfare into account.

Backwards induction is an intriguing form of argument. It is used in a number of different contexts. One of these is the surprise exam paradox. Another is game theory. But its use is problematic, at least sometimes. The purpose of this paper is to determine what, exactly, backwards induction is, and hence to evaluate it. Let us start by rehearsing informally some of its problematic applications.

It seems plausible, even truistic, that when an agent is faced with the choice of giving up one belief or another, the decision should be based on the relative strengths of these beliefs along some dimension of doxastic merit. This said, however, two non-trivial questions arise:

(1) Which dimension?

(2) How should the contraction outcome be affected by the distribution of beliefs along this dimension?

Individuals often seem to misjudge their own interests. One reason is inadequate information. Other reasons are failures of reasoning and volition. These reasons have all been construed as paternalist motives for the state to intervene. But in a recent article in this journal, New (1999), criticizes earlier accounts of paternalism. He argues that imperfect information constitutes a standard form of market failure, and consequently policies that respond to it do not require a paternalist motivation. The purpose of this note is to evaluate New's claim.

Bill New's (1999) thoughtful paper has performed the valuable service of clarifying the meaning and the policy implications of paternalism. His careful formulation delimits the domain of justified state paternalism. Having argued successfully, in our view, for a narrow ambit, New proceeds to identify situations that justify paternalism. This comment is written in the spirit of a friendly reformulation that refines and improves the specification of when paternalism is justified. Our argument is two-fold. First, we argue that New's formulation, properly understood, will not readily permit the paternalistic interventions he argues are justified. Second, we identify a class of potentially justified interventions that have paternalistic aspects, but which are neither strictly paternalistic nor market-failure remedies.