Scott Soames's What is Meaning? is an excellent book on an important foundational topic in the philosophy of language. The central question with which the book is concerned is, of course, that of the title. More specifically, Soames largely 'take[s] it for granted' (1) that linguistic expressions have meanings, and that in the case of sentences, those meanings are propositions.1 The issue is what propositions are. Although I am sympathetic to Soames's basic position, which departs from the orthodoxy in allowing a role for psychology in constituting propositions, I will argue that he ought to have been more radical. Nevertheless, the position which Soames articulates marks an important departure from the standard view — and one which, to my mind, constitutes an improvement.
II The Basic Position
Propositions, according to Soames, are expressed by sentences, and yet are distinct from the linguistic items which express them; they are also [End Page 575] the objects of the propositional attitudes (i.e., things believed, known, and asserted), and the bearers of (necessary and contingent) truth and falsity (2-3). So far this is little more than definitional for Soames.2 But theories of the nature of propositions must also explain (i) what it is to bear an attitude to them; and (ii) why they are representational, and accordingly bear truth conditions (6). On 'traditional' accounts, advocated by e.g. Frege and the early Russell, propositions are 'denizens of a 'third realm' (beyond mind and matter), which are 'grasped' by a mysterious intellectual faculty of platonic extrasensory perception' (7); moreover, they are 'intrinsically representational, and that from which all other representational bearers of truth conditions — sentences, utterances, and mental states — inherit their representationality' (7). But Soames argues that such platonic propositions don't exist; instead he advocates a 'cognitive realist' theory of propositions which reverses the traditional order of explanation.
According to the cognitive realist theory, propositions are cognitive event types. There is, of course, much more detail in Soames's theory, and more will be said in due course about what types of cognitive events propositions are for Soames; but it is worth first noting how plausible this basic position is. As Soames himself argues, the view provides a simple explanation of what it is to entertain a proposition: it is simply to be the agent of a cognitive event (token) of the relevant type. No mysterious faculty of extrasensory perception here! Moreover, the theory accounts nicely for the fact that propositions are representational, and have truth-conditions: on this view 'propositions are representational because of the relations they bear to inherently representational mental states and cognitive acts of agents' (7).
To begin to get a richer feel for the view (or at least one precisification of it3) we may think of it as follows. Many words of philosophical interest exhibit an act/object ambiguity; and the central terms at the foundations of semantics and the philosophy of language — 'assertion,' 'thought,' 'proposition' — are not exceptions. Yet while the objects of some acts are distinct from the acts themselves, those of others are not. Consider, for instance, the ontology of art. Confronted with the difficulty of giving necessary and sufficient conditions for objects to be works of art in terms of the intrinsic features of those objects, some [End Page 576] philosophers4 have been led to the view that artworks are action types. This has the decided advantage of locating artistry in the creative activity rather than its material product (when there is one); that is, art is not characterized by the making of a certain sort of object — rather, objects are artworks because of the nature of the activity which produces them. However, while it is plausible in some cases — say, jazz improvisations or ballet performances — to identify an artwork with the action type which produces it, in other cases this is plainly mistaken: paintings and sculptures, for example, are clearly objects distinct from, but produced by, certain characteristic activities. Soames's proposal regarding the propositional attitudes is that they are like performances, not paintings: their objects...