Instance-identity thesis

This is in fact the view Jaegwon Kim has advanced in several places about instances of second-order properties and instances of their first-order realizers.As Kim has noted, such an identification requires a revision of his property-exemplification account of events assuming that mental properties are second-order properties, it requires the exclusion of mental properties as constitutive properties of events. This instance-identity thesis is supposed to support reductionism about the mental. But there is a tension between this thesis and Kim s formulation in several places of his causal inheritance principle, which says that the causal powers of an instance of a higher-order property are identical with or are a subset of [emphasis mine] the causal powers of the instance of its realizer. Clearly, if the causal powers of the realized property instance were... 

I think that the subset version of the causal inheritance principle is clearly preferable to the version that says that the causal powers of the realized property instance are identical with the causal powers of the realizer instance, so I think that the instance-identity thesis is false. We can make sense of the idea of an instance of a higher-order property having a realizer different from the realizer of the instance of its determinate or property realizer if we can make sense of the idea of the higher-order property instances having cores that are distinct from the cores of the realizers of the instances of their determinates or physical property realizers. And 1 think we can do this. 

An advantage of this account is that it holds instances of mental properties to be physically realized in a way that allows them to be causally efficacious, and it removes the threat that their causal efficacy is preempted by the instantiation of their physical property realizers. In the case where pain is property-realized by C-fiber stimulation, it is not true to say that the causal work we would like to ascribe to the pain instance is really done insteadhy the C-fiber-stimulation instance. Nor is it true, as it is on Kim s instance-identity thesis, that, although this work is done by the pain instance, this is only because it is identical with the C-fiber-stimulation instance. The work is indeed done by the C-fiber-stimulation instance, but this is because the core of the C-fiber-stimulation instance realizer has the core of the pain instance realizer as a part. It is only because the C-fiber-stimulation instance realizer contains the pain instance realizer that it has the relevant effects. Recall that the core of a property instance realizer consists of the states of affairs that contribute directly to the implementation of the causal profile of the property. The core of the C-fiber-stimulation instance realizer contains states of affairs that contribute to the implementation of the causal profile of pain, but these make this contribution only because they realize states of affairs contained in the state of affairs that is the core of the pain instance realizer. 

---