Choice Group

Suppose that a treatment T for some medical condition C is already available and is universally recognized to be effective; and suppose that some new treatment T´ for that same condition C is under investigation. From the scientific point of view (that is, laying aside pragmatic issues such as, perhaps, the relative cost of the two treatments) there would, it seems, be no interest in testing the experimental treatment T´ against placebo. What the physician wants to know is whether T´ is better than (or, at any rate, at least as good as) the already-known-to-be-effective T not whether it is merely better than placebo. Accordingly, trials on T´ should, one would have thought, be “active-controlled” (meaning that those in the control group receive the active treatment T) rather than “placebo-controlled”. (There also seem to be clear ethical objections to conducting placebo-controlled trials in such a situation since those in the control group in such trials are knowingly given a treatment that is less effective than another available treatment.)

Nicolas Cote (LSE): “Weakness of Will and the Measurement of Freedom” Abstract: Weakness of will often seems to get in the way of free choice. In a wide range of situations, ranging from serious cases of depression and drug addiction to more mundane cases of weakness of will, it is clear that deficiencies of willpower make certain courses of actions…

Abstract: Fundamental physics makes no clear use of causal notions; it uses laws that operate in relevant respects in both temporal directions and that relate whole systems across times. But by relating causation to evidence, we can explain how causation fits in to a physical picture of the world and explain its temporal asymmetry. This paper presents a deliberative account…

Abstract: This paper argues that while expertise in ill-defined economic problems benefits from some level of expertise in well-defined problems, the two kinds of expertise are not the same.  Expertise in ill-defined problems requires different skills.  It is important for applied economics to understand the relationship between ability in well-defined economic problems and ill-defined problems as many economic policy issues…

Abstract: Pensions involve transfers from those who are young, healthy, able-bodied, and productive to those who are elderly, infirm, and out of work. Are these justified as redistributive transfers between distinct individuals – from those who are lucky to others who are unlucky – in order to eliminate brute luck unfairness? Or are they justified on grounds of reciprocity involving…

In a projected book, "Most Counterfactuals Are False", Alan Hájek infers the truth of its title from the ubiquity of chance. I argue in this talk that he is wrong: the ubiquity of chance does not verify his title: chances do not falsify counterfactuals. Single-case chances are perfectly consistent with determinism, i.e. with hidden variables that make relevant counterfactuals safe (i.e. truth-preserving). Not even indeterminism enables chances to stop these counterfactuals being safe and, for some values of the chances, knowably so.

Silvia Milano: “Bayesian Beauty” Abstract: The Sleeping Beauty problem has attracted considerable attention in the literature as a paradigmatic example of how self-locating uncertainty `creates havoc' for standard Bayesian principles of Conditionalisation and Reflection. Furthermore, it is also thought to raise serious issues for diachronic Dutch Book arguments. I show that, contrary to the consensus view, it is possible to…

Our moral decisions and actions are guided by what we take to be morally permissible and impermissible. In this talk I consider another factor which may affect both our judgment of moral permissibility and our moral conduct, our moral self-image. In particular, I ask whether a positive view of our own moral character traits is conducive to making good moral decisions and acting well. I discuss arguments from self-consistency that support this hypothesis. I then turn to the bias known as the better than average effect, and argue that our need for a positive moral self-image can lead us to be insensitive to evidence that we are acting immorally. The belief that we are morally superior facilitates unwarranted complacency and can lead to warped moral judgment via mechanisms of self-justification. This danger is particularly high when moral self-descriptions and evaluations of behaviour are very abstract. Very concrete moral self-ascriptions on the other hand are likely to have a positive effect. I conclude that while a positive moral self-image can be of limited benefit under tightly circumscribed conditions, it will in many cases be detrimental to moral judgment and conduct.

The rough idea of what I call "the veil of ignorance principle" is to identify the moral or "social" point of view with the point of view of a self-interested individual who is uncertain of his or her own identity. In conjunction with expected utility theory (EUT), such a principle was suggested by Harsanyi and others as validating an additive "utilitarian" aggregation rule; more generally, it suggests a systematic correspondence between norms of individual rationality and those of social choice. But it has generally been obscure how to properly formulate this principle, and whether it has any plausible justification.

In this talk, I show how three normatively natural axioms are jointly equivalent to a precise evaluative version of the veil of ignorance principle. I will also explore the resulting correspondence between individual and social evaluation, focusing on two issues. First, it is puzzling that an argument for "utilitarianism" should rely on EUT, since none of the latter's axioms seems central to the utilitarian project. We show that one can go almost all the way using (ex ante) Pareto and (ex post) impartiality instead. Second, I will discuss how basic issues of population ethics correspond, behind the veil, to what I call the "risky existential question": how should one evaluate an individual's prospects when there is only a chance that he or she will exist? Some initially plausible answers to this question generate untenable views in population ethics; perhaps the most plausible package overall validates a general form of total utilitarianism.

PhD students Chloé de Canson and Bastian Steuwer present their work to the LSE Choice Group.

We conduct a lab-experimental study of bargaining over the distribution of monetary losses. Groups of four participants who are differently endowed in terms of real money must agree, as a group, on the contribution each participant will make to cover a financial loss imposed on this group. The study sheds light on burden sharing and what loss allocation rules groups adopt. The rule to share the loss in proportion to the initial endowment which was found in experiments dealing with bequest or bankruptcy receives only little support in our experiments. The constrained equal awards rule and a rule that we conceived, a rule that focuses on an equal share of the total loss and a weighted difference between the own initial endowment of a player and the average endowment over all players, get much more support. We introduced several variations of our basic game. One variant which stages a knowledge quiz at the beginning of the experiment, shows in comparison with the base game (a) that the most successful player tends to offer less and (b) that the other players seem to honour this player’s success.

This talk applies probabilistic knowledge to problems in legal and moral philosophy. I begin by arguing that legal standards of proof require knowledge of probabilistic contents. For instance, proof by a preponderance of the evidence requires the factfinder to have greater than .5 credence that a defendant is liable, and also requires this probabilistic belief to be knowledge. Proof of guilt beyond a reasonable doubt requires knowledge of a significantly stronger content. The fact that legal proof requires knowledge explains why merely statistical evidence is insufficient to license a legal verdict of liability or guilt. In addition to explaining the limited value of statistical evidence, probabilistic knowledge enables us to articulate epistemic norms that are violated by acts of racial and other profiling. According to these norms, it can be epistemically wrong to infer from statistics that a person of Mexican ancestry is likely undocumented, for instance, even when inferring parallel facts about ordinary objects is perfectly okay.

This paper explores approaches to comparative justice (Sen, 2009) by drawing on Social Choice Theory. We introduce a procedure to correct for the influence of unquestioned parochial values on individual justice rankings: individuals are put into the position of other members of society allowing them to question (and possibly change) their justice ordering. In a first step, it is shown under which conditions this procedure leads to a domain restriction such that majority rule yields a social justice ordering. In a second step, it is examined how the introduced procedure can be used to distinguish between "reasoned" and "unreasoned" agreement. The paper concludes with a discussion as to how the findings cast doubt on the unqualified acceptance of the (weak) Pareto condition.