Another factor which might be thought to count against epiphenomenalism is that the supposition of efficacy plays a causal role in the explanation of human behaviour .sx Such behaviour exhibits certain complex regularities which call for explanation and which , at present , we explain ( at least partly ) in psychological terms .sx These psychological explanations , though typically of a rational rather than a mechanistic kind , attribute a causal efficacy to the mental :sx they represent the subject's behaviour as falling under the control of his beliefs and desires , or under the control of his decisions and intentions , which are responsive to his beliefs and desires .sx The claim might then be that such explanations gain credibility from the fact that , as well as being in their own terms successful , they cannot at present be replaced by non - psychological explanations which cover the same ground .sx These common-sense points against the epiphenomenalist are not , to my mind , decisive .sx There is no denying the conflict between epiphenomenalism and our ordinary conception of ourselves as agents .sx But the fact that this is our ordinary view of the situation does not mean that it is correct .sx Just what grounds do we have for supposing that human behaviour is intentional , and does qualify as action , in the relevant sense ?sx Again , there is no denying that , for ordinary purposes , we need to make use of psychological explanations .sx But it could still be claimed that the ultimately correct explanations ( maybe only discernible from a God's-eye view ) are purely physical .sx And indeed this is something which the defender of the science-efficacy argument himself accepts .sx The most that follows from these lines of reasoning is that the onus is on the epiphenomenalist to try to justify his unusual position .sx But presumably he is happy to take on this challenge .sx For he can say that what supplies the justification is the scientific evidence in favour of the view that the physical world is a closed system , together with the already established case against token-identity .sx However , it seems to me that epiphenomenalism is open to a much more powerful objection than either of these .sx For I cannot see how , if epiphenomenalism were true , the mind could form a topic for overt discussion .sx Certainly , if mental items have no causal access to our speech centres , the notion of an introspective report collapses :sx even if the subject retains an introspective knowledge of his mental states , his utterances do not count as expressing that knowledge if it contributes nothing to their production .sx But I cannot even see how , on the epiphenomenalistic view , our language , as a medium for our utterances , makes semantic contact with the mind at all .sx In what sense , for example , could the word 'pain' , as overtly used , be said to signify a certain type of sensation , if neither the occurrence of the sensations nor our introspective conception of their type affects its overt use ?sx Quite generally , it seems to me that if the mental contributes nothing to the way in which the linguistic practices involving 'psychological' terms are developed and sustained in the speech-community , and in no other way affects the production of utterances employing these terms , then , in respect of their overt use , the terms should be analysed in a purely behaviourist or functionalist fashion - which would deprive the epiphenomenalist of the linguistic resources to enunciate his thesis .sx It is true , of course , that each language-user may mentally interpret each term as signifying a certain kind of ( dualistically conceived ) mental item .sx But I cannot see how such private interpretations could have any bearing on the objective meaning of the terms , as employed in speech and writing , if , with respect to this employment , they are causally idle .sx ( This is not , of course , to endorse Wittgenstein's private language argument .sx The sort of private interpretations which Wittgenstein was trying to exclude would not , on my interactionist-dualist view , be causally idle .sx ) .sx It does not follow from this that the state of affairs which epiphenomenalism postulates is one which could not obtain - that there could not be a world in which the mental had no causal influence on the physical .sx Nor does it even follow that we can know a priori that the actual world is not of this kind .sx For it cannot be established a priori that the mental actually is a topic for overt discussion .sx None the less , if the point is correct , there is a sense in which , as overtly expressed , epiphenomenalism becomes self-refuting .sx For if it is only possible to provide a publicly audible or visible formulation of the thesis if the mental is causally efficacious , then any attempt to provide such a formulation can only succeed if the thesis is false .sx In effect , then , we either have to accept that the thesis is false , or abandon the attempt to make its truth or falsity an issue for public debate , or even for private but vocal soliloquy .sx If this does not quite refute epiphenomenalism , it at least renders it the sort of position which cannot be seriously entertained - the sort of position whose falsity ( to echo Hume on the existence of body ) we are forced to take for granted in all our reasonings , even of a philosophical kind .sx Having pressed this objection against epiphenomenalism , I must now add a qualification .sx In claiming that if epiphenomenalism were true , the mind could not form a topic for overt discussion , I have been assuming that if mental events have no causal influence on the physical world , then their occurrence will be , in every way , irrelevant to any explanation of physical phenomena .sx This assumption is a very natural one - and something which orthodox epiphenomenalists are unlikely to dispute .sx However , so long as we grant the coherence of theism , I think we can envisage two situations in which the assumption would be false .sx The first situation is that in which , while mental items have no causal influence on physical phenomena , they serve as 'occasions' for God to bring about certain physical events .sx For example , it might be that whenever a human subject decides to act in a certain way , God , taking note of the decision , causes the subject's motor-neurons to fire in the appropriate way .sx The decision itself does not cause the neuronal event ; it does not even indirectly cause it , by causally influencing God's decision , since God is here thought of as entirely active and free , and hence as not subject to any kind of external pressure .sx It is just that God chooses to control the physical on-goings in such a way as to match the subject's state of mind .sx The second situation is one in which , instead of controlling the physical on-goings in a piecemeal fashion , God deliberately arranges things , in advance and globally , so that the biological creatures which evolve are constituted in a way which secures a match between the functional roles of their neural states and the psychological character of the mental items which these states causally generate .sx Everything which occurs in the physical world has an efficient cause in the preceding physical conditions , and the mental items caused by neural states and events are themselves causally idle .sx But since God has selected the physical and psychophysical laws and the initial physical conditions of the universe with a view to ensuring that the mental and the physical realms harmonize in the appropriate way , there is still a sense in which the fact that neural states and events have certain psychological effects features in the ultimate explanation of the physical phenomena .sx Thus when a subject makes a decision and his muscles contract appropriately , we can say that , although it does not contribute anything causally to the muscular movement , this decision does have an ultimate explanatory bearing on the movement ; for it is only because the preceding central-neural event is empowered to produce that sort of decision that it is also empowered , by the structure of the organism and the physical laws , to produce that kind of physical effect .sx Now of these two situations , it is only the second which has any real relevance to the issue which concerns us .sx For the first , in which God takes human mental items as occasions for causing physical events , is not really an epiphenomenalist situation in the intended sense .sx It does not deny the causal efficacy of the mental in deference to the claims of physical science , and , if anything , it looks more like an occasionalistic version of interactionism than a form of epiphenomenalism .sx In any case , it would be hard to find any rationale for it except as part of a quite general occasionalist theory , such as that of Malebranche , which denies the causal efficacy of mental and physical events alike .sx The second situation , however , in which the physical world is a closed system and God is the transcendent architect of the psychophysical harmony , does yield a form of epiphenomenalism in the intended sense .sx And it might be that , if such a situation obtained , the explanatory link between mentality and behaviour would be enough to render the mind a topic for overt discussion - and , indeed , enough to allow us to retain some conception of ourselves as agents .sx I shall return , briefly , to this point presently .sx But , for the time being , let us continue with the assumption that epiphenomenalism represents the mind , not merely as having no causal influence on physical events , but as having no explanatory role either .sx On this assumption , as we have seen , epiphenomenalism is a position which , even if logically coherent , cannot be seriously entertained .sx And here the crucial point is not that it conflicts with common sense ( i.e. in respect of our conception of ourselves as agents and our employment of psychophysical explanations ) , but that the supposition of its truth would oblige us to suppose that the mind was not a topic for overt discussion .sx In this respect , then , things are looking good for the identity-theorist .sx All he needs to do , it seems , is to vindicate premise ( a ) of the science-efficacy argument , and then validate the move from ( a ) to ( b ) , and the identity of mental with physical events will automatically follow .sx However , while it is certainly true that claim ( b ) , together with the falsity of epiphenomenalism , would be enough to establish the token-identity thesis , it would be wrong to conclude that our argument against epiphenomenalism has made the prospects for this thesis any better .sx Quite the reverse .sx For a little reflection shows that the argument in question should be taken as an argument against the acceptance of claim ( b ) rather than as a step in the direction of token-identity .sx Let me explain .sx Claim ( b ) asserts that the physical world is a closed system .sx Taken in the context of premise ( a ) , from which it is inferred , this implies not just that the only events which cause ( or contribute to the causation of ) physical events are physical , but also that the only qualitative factors which are ultimately ( i.e. metaphysically-fundamentally ) operative in the causation of physical events are physical .sx But , thus interpreted , it is easy to show that , combined with what we have already established , claim ( b ) itself generates a sort of epiphenomenalism .sx For although it leaves room for the causal efficacy of mental items , it does not leave room for the causal efficacy of their psychological properties .sx Thus it allows us to say that a person's decision to cross the road caused the subsequent firing of his motor-neurons ; but it does not allow us to say that its being a decision , or its being a decision of that specific psychological type , played any causal role in bringing about this effect .sx In short , it obliges us to conclude that , with respect to the causation of the physical by the mental , psychological properties are causally idle .sx Now I am not suggesting that this conclusion immediately follows from claim ( b ) taken on its own .sx After all , someone might accept that it is only physical factors which are causally operative , but secure the causal efficacy of psychological properties by identifying them with physical properties .sx Or again , he might accept that it is only physical factors which are ultimately operative , but , by pressing some form of metaphysical reduction of mental facts to physical facts , allow psychological properties to enjoy a derivative efficacy .sx