Functional events tokens

I will speak of NRP theorists as taking functional properties to be constitutive properties of events. Alternatively, we could say that they take functional properties to be essential properties of events in either Lewis s or Yablo s (1992b) sense of essential properties of events, for Yablo s property exemplification account of events also allows for events with functional essences. By a functional event token, I will henceforth mean a second-order (or higher-order) event token, an event with a functional property as a constitutive property (or an event with a functional essence). Notice that although this role-functionalist view of event tokens does not treat quantification as an object-forming operation, it treats it as a particularforming operation, for events are particulars. 

The NRP theorists in question hold that functional properties can be constitutive properties of events. Indeed, on their view, it is because functional properties can be constitutive properties of events (or event essences) that no functional event token is identical with its physical realization. If this view is correct, then token physicalism is false. This event dualism, however, is compatible with the thesis of supervenience on the physical. Proponents of the view can hold that any minimal physical duplicate of... 

Thus, even on a property exemplification conception of events, type role-functionalism can be combined with token physicalism (for events). Type role-functionalism, you will recall, is the view that a mental event type M is the event type of undergoing an event of some type or other tokens of which would play a certain role R. It is open to a type role-functionalist to maintain that every instance of an Af type event is a physical event. The property of being an M type event (an event tokens of which would play R) would be a characterizing property of an event rather than a constitutive property. A physical event (an event with a constitutive physical property) would be an instance of M in virtue of the fact that it plays role R. This combination of type role-functionalism and token physicalism is compatible with mental events being causes even given physical closure and the physical effects principle, for it entails that every mental event is a physical event. 

I won t pursue these matters here. The reason is that NRP theorists must reject this combination of type role-functionalism and token physicalism, for they deny that mental event tokens are identical with physical event tokens. Mental event tokens, they hold, are exemplifications of functional properties and are not identical with exemplifications of physical properties that realize the functional properties. Let us see how to spell out their idea using Kim s theory of events. The idea is that functional properties will be constitutive properties of events rather than characterizing properties of events. Thus, let T be a functional property and P be one of its physical realizers. An exemplification of T by x at t will be the event [x, F, t] the property of which the event is an exemplification will be functional property F. If property T is realized on the occasion in question by P, then X will have Fat t in virtue of having Pat t. It follows that x, P, t] occurs and has role R, and indeed x, P, t will realize x, F, t by virtue of [x, P, t] having R But although [x, P, t R, it is nevertheless the case that x, P, t + x, F, t]. The reason is that P F. Thus, if there are functional properties and they are constitutive properties of events, then exemplifications of functional properties are not identical with exemplifications of their physical realizers. 

Also as concerns a mental event token and its narrow realizer (what we earlier called its core physical realizer ), he says If an Mi token is realized by a Pj-token, then the Pj-token just is part of the Mi-token (Melnyk, 2003, p. 137). He says nothing in defense of these claims. In fact, it is not the case that if an Mj token is realized by a Pj-token, then the Pj-token just is part of the Mj-token (p. 137). And it is not the case that the narrow realizer of a C-type event is surely a part or constituent of that C-type event (p. 160). The relation that the narrower physical realizer bears to the functional event — namely, the realization relation — is not a part-whole relation. It is, rather, the relation of role-occupancy. Narrow or core realizers are not parts or constituents of functional events. 

At the very least, M and P must be related by supervenience. However, it may be that mere supervenience is not enough to give us an adequate accoimt of mental causation within a physicalist framework. We might want to hold a view like Davidson s, where not only does M supervene on P, but P realizes M, and M and P are properties of a single event. So as properties, M and P are distinct, but as event, M and P are token identical. Whether physicalism requires mere supervenience, a picture like Davidson s with realization and event identity, a functional reduction of M to P as Kim advocates, or a fullblown type identity of M and P will be discussed shortly. 

The basic idea offiller-functionalism (for mental events) is that a mental event type Af is the event type, whatever it is, such that tokens of it would occupy a certain role R, which includes a causal roled The causal role will consist of conditional roles both as a cause and as an effect it will consist of the role of being such that its tokens would have certain kinds of effects in certain conditions and would be effects of certain kinds of events in certain conditions. On this view, an event will realize Af in virtue of being the event type tokens of which would occupy or fill role R. 

Now, if physical event type P is the event type tokens of which would occupy R, then P is Af it is Af in virtue of filling R, for being Af just is filling that role. On the assumption that it is contingent that tokens of P would occupy R, the type identity statement will be a contingent statement of identity since the description the event type, whatever it is, tokens of which occupy R is a nonrigid designator. This brand of filler-functionalism is compatible with type physicalism — the view that every mental event type is identical with a physical event type. Indeed, it entails type physicalism when combined with the thesis that whenever a mental... 

Let us turn, then, to the role-functionalist implementation of the notion of realization. NRP theorists typically appeal to it, and so are typically role-functionalists. According to type role-functionalism, a mental event type M is a second-order event type it is the event type of undergoing an event of some type or other tokens of which would play a certain role R, which includes a causal role. ° (The account of R is as before.) If an event type is such that tokens of it would play R, then it is a realizer of M. If there is more than one event type tokens of which would play R, then Af is multiply realizable. Type role-functionalists can maintain, without commitment to type physicalism, that whenever a mental event type is realized, it is realized by some physical event type or other. On this role-functionalist view, the... 

These categories are not mutually exclusive. Any individual horse, for example, is a token of the natural kind horse we might conceive of compositions as a certain kind of wholes, for example as structured wholes, and of wholes and compositions as individuals we could conceive of events as properties, and of particularized events as particularized properties. For what follows, property , type and kind will be used interchangeably. The term relation refers to what can also be referred to by relational property , relational type and relational kind , or, briefly, by kind , property and type (thus, there are n-ary kinds, types and properties with n > 1 as well as with n = 1). If you prefer a nominalist ontology, take properties to be sums of individuals (and replace the idea of properties not being located in space and time by some other idea that serves your purpose), if you are more liberal, take them to be sets of entities, if you are even more liberal, take them to be functions from worlds to extensions. For the discussion to come, it will be useful to think of properties in one of the intensional senses or, for most of the properties the bearers of which are individuals occupying a position in space and time, in functional terms. 

---