Abstract

There is a widespread view in philosophy of language and in philosophy of mind according to which the «quinean» rejection of analyticity can be made compatible with some sort of realism about meaning. Against such compatibilist claim, Paul Boghossian has recently held the thesis that one cannot coherently reject the analytical/synthetical distinction maintaining at the same time a meaning realism. His arguments are very pervasive, but they can be replied. The main objective of this paper is to show that in fact it is possible to reject analyticity being at the same time a meaning realist, even a meaning realist of a non-holist kind. The prevailing view is basically right. Moreover, it is possible to go on maintaining the compatibilist claim in its most radical form. In short, even if we adopt a non-holist meaning realism, we must reject analyticity because language is always conceptually motivated and engaged with reality. There is no linguistic arbitrariness. That forces us to go far from classical conceptions of meaning and to have a much more pluralistic one. With respect to it, for instance, to say that some statements are true once their meanings are fixed would not entail that they are true by virtue of meanings. The problem to get such a conception of meaning remains open. However, the reasons against analyticity do not force us to any irreductible meaning holism