Robert Kirk

How Physicalists Can Avoid Reductionism

ABSTRACT

Kim maintains that "a physicalist has only two genuine options, eliminativism and reductionism". But physicalists can reject both by using the Strict Implication thesis (SI). Discussing his arguments will help to show what useful work SI can do.
(1) His discussion of anomalous monism depends on an unexamined assumption to the effect that SI is false.
(2) His conclusion on multiple realizability is much stronger than his reasoning warrants.
(3) In discussing supervenience, he is wrong to assume that the only approach to explaining why "physical truths determine all the truths" must be via psychophysical laws.
(4) His general argument rests on a mistaken assumption that the only alternative to psychophysical identity theses is to accept "supervenient causal relations".



How Physicalists Can Avoid Reductionism

Robert Kirk

1 Must physicalists choose between eliminativism and reductionism?
Jaegwon Kim has recently argued that there is no room for a "substantial form of physicalism" that is combined with rejection of psychophysical reduction. He asserts that "the choices we face concerning the mind-body problem are rather stark: there are three – antiphysicalist dualism, reductionism, and eliminativism". So "a physicalist has only two genuine options, eliminativism and reductionism" (Kim, 1993, p. 267). I think he has overlooked something. There is a substantial form of physicalism which can consistently reject both eliminativism and reductionism. It is based on a thesis of strict implication: from the physical to the mental. Discussing Kim's claim provides a good opportunity to see what useful work a thesis of strict implication can do. But first, a word about reduction.
Kim remarks that in the philosophy of mind the word `reductionism" "seems ... to have acquired a negative, faintly disreputable flavour. ... Being a reductionist", he aptly comments, "is a bit like being a logical positivist or member of the Old Left – an aura of doctrinaire naivetΘ hangs over him" (op. cit., p. 266). He goes on to say that many physicalists today think "that we can assuage our physicalist qualms by embracing `ontological physicalism', the claim that all that exists in spacetime is physical, but, at the same time, accept `property dualism', a dualism about psychological and physical attributes, insisting that psychological concepts or properties form an irreducible, autonomous domain" (loc. cit.). But he thinks that position is unstable – and I agree there is not much to be said for it: the physicalism I defend does not involve property dualism. He asserts that "if you have already made your commitment to a version of physicalism worthy of the name, you must accept the reducibility of the psychological to the physical, or, failing that, you must consider the psychological as falling outside your physicalistically respectable ontology" (op. cit., p. 267). I think that is a mistake.
At any rate I think it is a mistake if maintaining the reducibility of the mental to the physical commits you to anything like one of the following three claims:
(a) Every true mental statement is translatable by a physical statement.
(b) Every true mental statement is logically equivalent to a physical statement.
(c) Every primitive mental predicate is correlatable by a bridge law (an identity statement or a biconditional) to a physical predicate.
Of course it is open to Kim to say that the view I favour, which is that each true mental statement is strictly implied by the totality of true physical statements, is itself a form of reduction. But it is not usually counted as such; nor does Kim do so. He seems to assume that reduction requires bridge laws connecting mental and physical predicates. And (although he does not spell this out) it is obvious that such bridge laws would permit the reducing science to take over the work of the reduced science. That was the whole point of the original idea of reduction. It aimed at intellectual economy. It promised to show how a science which had appeared to be distinct and independent was merely a disguised way of carrying on the reducing science: psychology would turn out to be part of physics. The Strict Implication thesis, in deliberate contrast, holds out no such promise. It is neutral as to whether reduction in that important sense is possible. Indeed it avoids any commitment to unacceptably tight links between physical and mental predicates. (If you think it implies such links, that needs argument: see below.) So even if it counts as a kind of reductionism, it is an extremely loose kind, which, as I will try to make clear, leaves plenty of scope for some kinds of autonomy for other sciences than physics, including psychology.

2 The Strict Implication thesis
We start by pretending we have an idealized version of today's physics, including a set of laws and a special vocabulary. Let P be a conjunction of all truths statable in terms of that theory. P includes statements of all physical laws as well as all truths about the distributions and states of elementary particles throughout space and time. If we now assume that our idealized physics is itself true, P is a specification of the whole physical universe, covering past, present, and future. The special physical vocabulary is defined so as not to include psychological expressions. So P does not include any statements of whatever psychophysical laws or true generalizations there may be. Now consider truths in psychological language about the mental states of the organisms whose existence is provided for by P, and let Q be a conjunction of all actually true statements ascribing mental states to those organisms. The Strict Implication thesis is:

P strictly implies Q. That is, it is impossible that P should be true and Q false.*

The impossibility involved here has to be the strongest kind there is: `absolute', not just nomic. The reason, briefly, is this. Any (non-eliminative) physicalist will agree that when we are talking about the mental states of actual organisms, we are talking only about the physical – not as physical, of course: the point is not that it is immediately obvious that our mental statements are ways of talking about physical things and processes. It is the M. Jourdain point: we're doing it all the same, whether we realize it or not. So any physicalist ought to agree that there is no possible world at all where P is true and Q is false. Physically, such a world would be exactly like the actual world, yet it would differ in respect of the mental states of whatever organisms were provided for by P. But in that case there would be more to the mental than is involved in the physical. That would contradict the assumption that when we are talking about the mental (in the actual world), we are talking only about the physical. (To be sure, not all those who call themselves physicalists have recognised that commitment; but according to me they are mistaken.)
Of course today's physics may turn out to be false – even when idealized. P might be empty, in which case it could not rule out the contradictory of any member of Q. And even if P were not empty, the fact that the statements in it were expressed in terms of a false theory would probably prevent it from strictly implying Q. But I do not think such considerations affect the philosophical relevance of the Strict Implication thesis. Physicalism is essentially linked to physics. Instead of appealing to such exploded notions as those of `matter and `solidity', as old-fashioned materialism did, it appeals to the theoretical concepts of physics. I suggest this implies that there is no need to go into the question of just how to define `the physical', nor into the question of what any possible future sciences that may be called `physics' may be like, nor relativize the doctrine to the true physics, whatever that may turn out to be. Physicalism is interesting only if the true physics bears a sufficiently close family resemblance to today's, in which case a suitably modified version of the Strict Implication thesis will be available. If, on the other hand, the true physics (to great surprise) introduces dualistic concepts, then physicalism will have to be abandoned. I see no reason to think that physicalism loses anything of interest if we stick by the version of the Strict Implication thesis stated above.
The Strict Implication thesis is not by itself sufficient to yield even a minimal physicalism. For that purpose it has to be supplemented with two further theses: that Q does not also strictly imply P, and that nothing exists other than what is strictly implied by P. Even with those supplementary theses, the resulting minimal physicalism cannot purport to be a physicalistic solution to the mind-body problem. What it does is provide a reasonably clear statement of what a physicalistic solution must do. A solution must explain how it is that the Strict Implication thesis holds. Clearly that calls for philosophical work as well as empirical work in the neurosciences and psychology.
Now, my contention is that a physicalism based on the Strict Implication thesis does not commit you to the view that the mental can be reduced to the physical – unless, of course, you count the Strict Implication thesis itself as a kind of reductionism; but I have explained why that does not bother me.

3 Anomalous monism
Kim discusses and rejects three main routes by which he supposes physicalists might hope to escape his claim: Davidson's anomalous monism; multiple realizability; and supervenience. He ends with a fourth, general, argument. I think all four lines of argument fail because he has overlooked a physicalism based on a thesis not of supervenience, but of strict implication. I will examine each line of argument in turn.
First, Davidson's anomalous monism, according to which there are no strict laws governing either relations among mental events or relations among mental and physical events. Given that nevertheless there are causal relations between physical and mental events, and given Davidson's doctrine that events can stand in causal relations only if they are covered by causal laws, it follows that each individual mental event (at any rate each one that enters into causal relations) is identical with a physical event. Now Kim claims that this doctrine entails that "mentality does no causal work". He supports this by urging that "on anomalous monism, events are causes or effects only as they instantiate physical laws, and this means that an event's mental properties make no causal difference" (Kim, 1993, p. 269f.). And he backs up this claim in turn by the assertion that: "... to suppose that altering an event's mental properties would also alter its physical properties and thereby affect its causal relations is to suppose that Psychophysical Anomalism [his name for the doctrine that there are no strict laws governing psychophysical relations], a cardinal tenet of anomalous monism, is false" (op. cit., p. 270). But here, it seems to me, he has overlooked a vital consideration.
This is that if the Strict Implication thesis is true, then the physical properties of things in the world fix their mental properties even if there are no strict psychophysical laws. If that is right, there is a clear sense in which a thing's having the mental properties it does have involves (in fact) nothing over and above its having whatever physical properties it has. So for its mental properties to do `causal work' requires nothing more than that its physical properties do whatever causal work they do. There is no possibility whatever that the thing should have had exactly the same physical properties and different mental ones (still less that it should have had no mental properties at all). So Kim's objections to anomalous monism are without foundation. (There are other objections too, but they will not concern us here.)
Those remarks will become clearer, I think, in the course of examining Kim's later arguments.

4 Multiple realizability
The idea of multiple realizability is that for any type of mental event there are "endlessly diverse ways" (in Kim's phrase) in which it can be "physically realized" or "instantiated" or "implemented". If that is right, it seems at first that the biconditionals necessary for reduction will not be available. For in that case, as Kim puts it, "For any psychological property, there is in principle an endless sequence of nomologically possible physical states such that, though each of them `realizes' or `implements' it, none of them will by itself be coextensive with it" (op. cit., p. 272). So it seems that if mental states are multiply realizable by physical ones, the former are not reducible to the latter. (Kim notes, but agrees to disregard, the suggestion that there will still be disjunctive properties coextensive with the psychological ones.)
But now he asserts that the exponent of multiple realizability is committed to the assumption that "a physical state that realizes a mental event is at least nomologically sufficient for it" (op. cit., p. 273). He goes on to suggest that Putnam is actually committed to more than mere sufficiency, indeed that he is committed to "species-specific bridge laws" of the form `Si – (M – Pi)' which, relative to a species or structure Si, specifies a physical state, Pi, as both necessary and sufficient for the occurrence of mental state M. However, he concedes that "biological species may turn out to be too wide" to generate laws of this kind, given individual differences. He also concedes that the same mental state may be realized by different physical states at different times in the life of the same individual. So he ends up with the formula: "for each psychological state there are physical-biological structure-types, at a certain level of description or specification, that generate laws of this form" (loc. cit.). He does not, however, insist that the physicalist exponent of multiple realizability is logically committed to this position; only that it is consistent with multiple realizability. And he accepts that "global reduction is not in the offing". But he insists that "local reductions are reduction enough, by any reasonable scientific standards and in their philosophical implications" (loc. cit., p. 275).
I have two main objections to this position. The first arises from the fact that he offers no argument for his assumption that if a certain physical state is sufficient for the individual with the given structure to be in the given mental state, it is also necessary. The objection is that we have no good reason to accept that assumption, and some reason to reject it. For we know the brain is an extremely complicated organ, with many millions of components and vastly more connections among them. It has the potential for a staggeringly large number of different states (recall that many neurons have something like ten thousand connections with others). By comparison, our characterizations of mental states are limited in kind and number. So far as I can see, we have no good reason to assume that nevertheless the relation between possible mental states and brain states is one-one. On the contrary, reflection on connectionist models suggests that for each possible state characterizable in mental language, there is likely to be a huge number of different ways in which it could be realized or instantiated in the brain. The relation between mental and physical states seems likely to be one-many – even for an individual at a time. But unless Kim's assumption to the contrary is correct, there is no reduction of the mental to the physical even at the ground level of individual structures at times. So at the very least he has failed to justify his rejection of the claim that antireductionist physicalists can escape his stark choice by endorsing multiple realizability.
But suppose we disregard that objection, and for argument's sake concede his assumption that, for each individual at a time, there is a one-one correspondence between that individual's possible mental states and their possible (relevant) physical states. Still, his assertion that this is "reduction enough" is at best puzzling. For we noticed at the start that a main virtue of reduction, and a main motive for attempting to establish it, is that the work of the reduced science can then be taken over – at any rate in principle – by the reducing science. If each type of mental state is correlated with a single physical state, then a scientific treatment of mental states is much more straightforward than it otherwise would have been. (Psychology becomes part of physics.) But the narrower the class within which the correlations hold, the less feasible is a scientific treatment at the level of the supposedly reducing science, and the more autonomy the supposedly reduced science acquires. (For one thing, it is hardly a task for physics to consider and investigate psychologically relevant and interesting similarities and differences among the various possible individual structures.) What Kim calls `local reduction' – reduction for the structures of individual species-members at times – would rule out just the kinds of scientific simplification that made reduction an attractive goal in the first place. So if it is nevertheless a kind of reduction, it is not one the physicalist need be bothered by (and it hardly incurs a charge of `doctrinaire naivetΘ').

5 Supervenience
Kim next turns to consider the suggestion that physicalists can avoid reductionism by endorsing a doctrine of supervenience. He puts what he sees as the difficulty in these terms: "if a relation is weak enough to be nonreductive, it tends to be too weak to serve as a dependence relation; conversely, when a relation is strong enough to give us dependence, it tends to be too strong – strong enough to imply reducibility" (loc. cit., p. 276). He focuses attention on that particular variety of supervenience which he calls `global', and which he says "has seemed to many philosophers to hold the most promise as a nonreductive dependency relation". The thesis as it applies to the psychophysical case is that

Worlds that are indiscernible in all physical respects are indiscernible in mental respects (loc. cit.).

Now this thesis seems to be stronger than the Strict Implication thesis. It clearly implies that for every possible world, not just the actual world, the thesis which corresponds to the Strict Implication thesis (the thesis you get when `P' and `Q' respectively refer to the physical and mental truths holding there) holds for that world. And it rules out possible dualist worlds a priori: worlds where some mental activity is not strictly implied by physical activity. There may be a priori arguments for ruling out such possible worlds; but if so, I do not know what they are. Minimal physicalism based on the Strict Implication thesis is neutral on that question; which is one reason why I prefer it to versions based on global supervenience. Still, the argument Kim deploys against global supervenience would also apply against a physicalism based on the Strict Implication thesis. He starts by asserting that:
the global supervenience of the mental permits the following: Imagine a world that differs from the actual world in some minute physical detail (we may suppose that in that world one lone hydrogen atom somewhere in deep space is slightly displaced relative to its position in this world). This world with one wayward hydrogen atom could, consistently with the global supervenience of the mental, be as different as you please from the actual world in any mental respect (thus, in that world nothing manifests mentality, or mentality is radically redistributed in other ways). The existence of such a world does not violate the constraints of global supervenience; since they are not physically indiscernible from the actual world, they could, under global supervenience, differ radically from this world in psychological characteristics (loc. cit., p. 277).
He then points out that it is a perfectly legitimate question, if global supervenience holds, to ask why it holds. Why do "physical truths determine all the truths", if that is what they do? And surely it is a legitimate question. If you endorse the global supervenience thesis – also if you endorse the Strict Implication thesis – the onus is on you to provide an explanation of why this kind of determination holds. But I think that although Kim is right to press the question, the answer he suggests is wrong. He starts from the reasonable assumption that "if psychophysical determination or dependence means anything, it ought to mean that the psychological nature of each [psychological] unit is wholly determined by its physical nature" (loc. cit.). But he goes on to draw what seems to be an unjustified and, indeed, false conclusion. He says, "The trouble is that once we begin talking about correlations and dependencies between specific psychological and physical properties, we are in effect talking about psychophysical laws, and these laws raise the specter of unwanted physical reductionism. Where there are psychophysical laws, there is always the threat, or promise, of psychophysical reduction" (op. cit., pp. 278f.). By exploiting the Strict Implication thesis we can see that nothing compels us to follow this route.
Recall what he says about the possibilities he alleges are left open by the global supervenience thesis – such as a possible world where there is just a tiny physical difference from the actual world (a displaced hydrogen atom in deep space), but huge mental differences. It is true that neither global supervenience nor Strict Implication includes any explicit statement to the effect that such worlds are impossible. But nor do they include statements to the effect that there are no worlds containing square circles, or where pi is a rational number; and clearly that is no objection to them. In general, necessary truths and necessary relations do not have to be explicitly stated (otherwise, since there are infinitely many of them, we could never state any thesis at all). By making use of this fact, exponents of global supervenience and Strict Implication can escape Kim's objection.
Presumably everyone will agree that such things as details of the distribution of hydrogen atoms scattered in deep space are irrelevant to people's mental states: that is why Kim's objection looks plausible. Of course many features of the world do have a bearing on our mental states; but many do not: they are nothing to do with them. So both global supervenience and Strict Implication theorists will concede that, although it is simpler to state their position in terms of determination of mental states by the totality of physical ones, what actually fixes our mental states is some subset of the latter. Only some physical facts are relevant. Now, since these theorists maintain that it is absolutely impossible that the relevant physical facts should have been as they actually are while the mental facts were different, they owe an explanation of why that is so. Merely asserting either thesis is not enough (as Kim insists). But (and this is where I think he goes wrong) the explanation of these necessary relations from the physical to the mental can be expected to make clear that what he imagines are possibilities are not genuine possibilities at all. He appears to have overlooked a whole alternative approach to one which goes via the notion of psychophysical laws. The alternative goes via philosophical explanations.
To illustrate this point, consider behaviourism. I take this to be the view that an individual's mental states depend entirely on how they behave or are disposed to behave, regardless of the nature of the internal processes underlying behaviour and dispositions. If it turns out that the purely physical facts about organisms are enough to account for behaviour and dispositions, then behaviourism in effect provides a philosophical explanation of how it is that the purely physical facts fix the mental facts. I am not saying behaviourism is right about this (in fact I think it is wrong), but that it is an example of a philosophical explanation of the sort that could answer Kim's question. Why do the physical truths determine the mental truths? Because, firstly, the physical truths about any organism determine the facts about its behaviour and behavioural dispositions; and secondly, the facts about an organism's behaviour and dispositions constitute the facts about its mental states. (They constitute the latter facts in the sense that there is no more to its having its mental states than its behaving as it does and having those behavioural dispositions.) The first part of the answer is empirical; but the second is philosophical. In this way a behaviourist defence of the Strict Implication thesis, or for that matter of the global supervenience thesis, would require no appeal to any putative psychophysical laws. According to the Strict Implication thesis, it is absolutely impossible that P should be true and Q false. Behaviourism would explain this by saying that the physical facts represented by P ensure that human beings and other sentient creatures have the right behaviour and dispositions; and then the relevant philosophical considerations, dealing with the relations between behavioural dispositions and mental states, ensure that creatures with those dispositions necessarily have those mental states. (Notice that the behaviourist can add that if dualism had been true, the physical facts alone would not have been enough to guarantee that people had the right dispositions. So even a behaviourist could concede that a thesis corresponding to the Strict Implication thesis might have been false.)
You might object that a proper behaviouristic account of mental concepts would yield psychophysical biconditionals, or bridge laws, that would be sufficient for reduction. Two considerations undermine this objection. First, even if there were such things as adequate behaviouristic analyses of mental concepts, they would not yield psychophysical biconditionals, but only psychodispositional ones. It would take further argument to show that the behavioural side of the equivalences could in each case be provided for only by some single physical condition. And to do that it would first be necessary to produce an a priori refutation of dualism. In addition, we noticed one good reason to the contrary when discussing multiple realizability. The second consideration is that it is perfectly possible to maintain a general behaviouristic account of the nature of mental states without also maintaining that each particular mental state is associated with its own proprietary set of behavioural dispositions. Indeed, such global or holistic behaviourism seems to be the least objectionable kind.* So, given certain empirical facts about the workings of actual organisms, philosophical considerations may be able to support the Strict Implication thesis without risking reduction.
That illustrates one way in which we might be able to show that the Strict Implication thesis held without knowing just which members of P or Q were actually true. If those general (behaviouristic) philosophical considerations were valid, and if we knew enough about the physical facts to be sure they provided for the right sorts of dispositions, we could immediately conclude that the thesis held: no such detailed knowledge would be required.
The same example helps to undercut another possible objection. If P really does necessitate Q, then mustn't it be possible, in principle, to derive Q from P? If so, how could that be done unless there were equivalences between mental statements and physical ones? Global or holistic behaviourism is a counterexample. Such behaviourism does not require any such equivalences. Yet if it were true, the physical facts could unproblematically be seen to necessitate the mental ones. Notice particularly how this example shows that strict implication does not require it to be possible to `read off' the mental statements from an examination of the physical ones – no matter how much hard philosophical work we might put in.
A related comment is in order. In contrast to some forms of physicalism, the Strict Implication thesis does not imply that the concepts of physics supply a basis for all other ways of describing reality. Knowing the physical facts represented by P will not in general endow the knower with the concepts required for other vocabularies, such as that of everyday psychology. A creature might be capable of grasping the statements in P without being the right sort of creature to grasp those in Q. (Here – though only here – I am in agreement with Nagel, 1974). To be able to grasp the concepts of everyday psychology, a person needs to have a nature, capacities, tendencies, and interests that are sufficiently close to those of human beings. But knowledge of physically stated truths could not by itself endow a creature with that particular nature or capacities or interests. For example it could not endow it with colour vision. Arguably a full grasp of human colour concepts requires the ability to see colours as human beings do. If that is right, it provides one illustration of why a knowledge of P would not necessarily enable the knower to move on from it to a grasp of the concepts involved in Q. In this way acceptance of the Strict Implication thesis is consistent with maintaining that psychology has a fair degree of autonomy. (Yet it might still be possible for a person who was incapable of a full grasp of certain mental statements to discover, by deep reflection on the physical facts, which mental statements were actually true. The true statements used by intelligent bats to describe their echolocatory experiences would not, I think, be comprehensible by us – – or not unless we had made some very special arrangements. But we might still be able to discover that they were true.)
Now we have seen some of the work that global supervenience and strict implication theorists have to do, we can also see that Kim was wrong to assume that the only way they could possibly explain the dependence of the mental on the physical would be via psychophysical laws. There is an alternative route: via a philosophical explanation of the nature of the mental.1

6 The general argument from "supervenient causation"
Kim's last move in his attempt to force physicalists to choose between eliminativism and reductionism is a general argument dealing in relations between physical and mental causes. Given mental realism, given that some mental events cause physical events `in virtue of' their mental properties, and given that every physical event has a physical cause ("the causal closure of the physical domain"), there is a question about the relation between mental properties and physical ones. An obvious way out would be to say that each mental property was identical with some physical property. But for Kim that would be reductionism: "These property identities would serve as bridge laws par excellence" (op. cit., p. 282). Is there any other way out? He maintains that "the only way other than the identity solution is to give a general account of causal relations involving macro-events as `supervenient causal relations', causal relations that are supervenient on micro-causal processes" (loc. cit.).
He explains that a macro-event is a cause or effect of another event "in virtue of the fact that" the former is supervenient on some micro-event which is a cause or effect of the latter. And he asserts that according to this view "mental properties are seen as deriving their causal potential from the physical properties on which they supervene" (op. cit., p. 283). He goes on to say that the only sort of supervenience that would provide for the needs of this sort of account would be "strong", involving "specific mental properties supervening on specific physical base properties". And he claims that that is possible only if there are laws correlating psychological properties and physical properties. He concludes that "If nonreductive physicalists accept the causal closure of the physical domain, ... they have no visible way of accounting for the possibility of psychophysical causation. This means they must either give up their antireductionism or else reject the possibility of psychophysical causal relations" (op. cit., p. 284).
I would query his reasoning at several points; but here let me focus on just one. I think he is mistaken in assuming that physicalists are compelled to accept the doctrine of supervenient causation. This is another place where the Strict Implication thesis does useful work. For if the Strict Implication thesis is true, and a conjunction of all actual physical truths strictly implies a conjunction of all the mental truths about us, then we have a reasonably clear sense in which mental language is just a way of talking about physical events. And if there are any true statements of mental causation, those statements too are among the statements strictly implied by the physical truths. Now, although it takes philosophical work to show that this is indeed the case, the thought is that once that work has been done, the existence of mental things and properties will be seen to be provided for by the existence of certain physical things and properties. So far as I can see, the strict implication theorist is not compelled to concede that mental things and properties are even token-identical with physical ones. For one thing, it is consistent with the actual psychophysical relation being one of strict implication that there are possible worlds where the mental involves non-physical things or processes. But nor is the strict implication theorist compelled to go along with Kim's doctrine of supervenient causation. A simple example may help to clarify these points.
The phenomenon of relief rain is normally explained as follows. Prevailing winds drive masses of relatively humid air against the slopes of mountains, forcing the air upwards, so that in the lower temperatures at high altitudes water vapour condenses and rain falls on the leeward side of the mountain. In brief, relief rain is caused by humid air being blown up mountain sides. That explanation, given elementary knowledge about humidity, temperature, condensation, and so on, is entirely adequate. Now, it may be possible to construct an explanation of the phenomenon in micro-terms, dealing in conglomerations of H2O molecules and treating mountains as complex structures of molecules or even atoms. But what would be the point of the exercise?2 It would not result in an explanation that was more perspicuous. On the contrary, we should hardly be able to find our way about in it: the high-level explanation is vastly more perspicuous. So would the micro-explanation be more accurate? Not at all. On the contrary, it would introduce information that was actually irrelevant. For to specify the micro-composition of the mountain and the water vapour would be to mask the fact that what matters in such cases is not the fine details but the overall shape and size of the mountains, and general points about relative humidity and the like. Those are the things actually doing the `causal work'. At the same time, as a (minimal) physicalist I am not going to say that relief rain involves something more than whatever micro-events are involved. Nor do I want to say that when the shape and size of the mountains are doing their causal work, other causal factors are involved than are involved in the micro-events doing their causal work.
So how can we account for the fact (as I take it to be) that the large-scale events involving the mountain can enter into genuine causal relations – or rather, how can we account for it if we reject both the claim that those large-scale events are identical with those micro-ones, and the claim that `supervenient causation' is involved? We can exploit the still available notion of multiple realizability, and say that the micro-events realize or constitute the large-scale ones; but that they are not identical with them because exactly similar large-scale events could have been realized or constituted by different micro-events. Clearly this does not imply that mountains might have involved something non-physical. The point is that a whole spectrum of different micro-structures – conceivably, even different systems of physical laws – could have provided for mountains that caused relief rain in similar ways. Further, if the micro-events actually constitute the macro-events, there is no sense in saying, in Kim's terms, that the latter "derive their causal potential from the physical properties on which they supervene". To say that is to imply that the micro-events have some kind of causal priority; which we have seen not to be the case for such properties as the shape and size of mountains. Armed with these concepts of strict implication and constitution, therefore, the physicalist is well placed to resist Kim's claim that nonreductive physicalists must accept supervenient causation, which then leads back to reductionism. There is a good alternative for physicalists: physicalism based on the Strict Implication thesis.3



Footnotes

1. For more detailed exposition and discussion see Kirk 1994, pp. 71-86.
2. See e.g. Block, 1981, for reasons.
3. For an attempt to supply a satisfactory account (as contrasted with a behaviouristic one, which is not satisfactory) see Kirk, 1994.



REFERENCES

Block, Ned: 1981, `Psychologism and Behaviourism'. Philosophical Review 90, 5-43.

Kim, Jaegwon: 1993, `The Myth of Nonreductive Materialism', in his Supervenience and Mind, Cambridge University Press, Cambridge, pp. 265-84.

Kirk, Robert: 1994, Raw Feeling: a philosophical account of the essence of consciousness, Clarendon Press, Oxford.

Nagel, Thomas: 1974, `What Is It Like to Be a Bat?', Philosophical Review 83, 435-50. Reprinted in his Mortal Questions, Cambridge University Press, Cambridge, 1979.

Putnam, Hilary: 1979, `Philosophy and Our Mental Life', in his Mind, Language and Reality: Philosophical Papers vol. ii, Cambridge University Press, Cambridge, pp. 291-303.

Department of Philosophy
University of Nottingham
Nottingham – NG7 2RD U.K.
Robert.Kirk@nottingham.ac.uk