Alleged refutations of utilitarianism are not uncommon, so it is unlikely that the title of the present essay will raise eye-brows. ‘Another paper about utility's failure to account for our duty to be just’ (thought with a yawn, as it were), is apt to be the prevailing reaction to the title's stated objective. This is understandable. For utilitarianism has been taken to task on just this score more than a score of times. And rightly so, I believe, though I shall not argue that point here. Here I intend to offer a refutation of utilitarianism which turns, not on the duty of justice, but on the value of friendship, a refutation which, so far as I am aware, has never previously been advanced in the not inconsiderable body of literature critical of that theory.

Functionalism, as it is currently understood, is the view that each type of mental state is identical with a state that is a causal consequent of certain kinds of inputs and other mental states and which, in turn, causally brings about certain kinds of outputs and other mental states.

There is trouble at the foundations of Grice's theory of conversational implicature, or so I shall argue. Grice's commentators seem to agree, and some of Grice's own remarks suggest, that every case of implicature is one in which ‘the speaker gets across more than he says…. ’ The problem is that there are cases in which nothing is said - in which case it is not clear that there is any vehicle by which the implicature might be carried, and consequently not clear that Grice's theory can account for the implicature which does appear to be carried.

I want to deal in this essay with the group of philosophers that G.E.M. Anscombe includes under the term ‘modern moral philosophy’ in her essay by that name. The stars of this group are Hobbes, Hume, Adam Smith, Mill, Sidgwick, Moore. I mean to include as well generally the last hundred years of emotivists, utilitarians, and those theorists who have emphasized universalizability in its various versions. For reasons which I hope will soon become clear, I will refer to this broad group as the ‘atomistic tradition.’ I will be assuming, though not often saying, that the most forthright and thorough spokesman of the tradition is David Hume who, in his purely skeptical moments, was so atomistic and extensionalist that he even denied personal identity over time. Anything less than that is not purely atomistic.

La théorie des descriptions de Bertrand Russell est sans aucun doute l'une des thèses philosophiques qui, au vingtième siècle, ont donné lieu au plus grand nombre de commentaires, de critiques, voire de querelles. Portée aux nues par certains-Ramsey l'a qualifiée de ‘paradigme de philosophie’-elle sera violemment contestée par d'autres, en particulier par Strawson qui s'avisera, quelque quarantecinq ans plus tard, qu'elle comporte des ‘erreurs fondamentales.’

Our states of belief and desire are no doubt supervenient on the overall pattern of our physical states. But can this minimal physicalist presumption be strengthened into a claim to the effect that our mental states are each identical with some specific corresponding physical state? A developed identity theory will need, in a sense to be made clear, a schema for specifying the physical state which is supposed to be identical with a given mental state. And there are problems in formulating such a schema. The special difficulties faced here by a type identity theory are well known, and will not be discussed further in this present paper. But equally, there are problems in constructing an identification schema for employment with a token identity theory.

In an ingenious argument, Ng and Singer claim to show that it is possible to derive the general principle of utility from another principle, Weak Majority Preference (WMP), which many who are not utilitarians would be prepared to accept. WMP does indeed sound acceptable: ‘For a community of N individuals choosing between two possibilities, X and Y, if no individual prefers Y to X, and at least individuals prefer X to Y, then X increases social welfare and is preferable’ (232). But from this seemingly innocuous rule, Ng and Singer maintain, quite substantial results follow if one takes account the well-established fact of intransitive indifference owing to the existence of finite sensibility.