Jaegwon Kim, in his superb essay The nonreductivists troubles with mental causation (from the volume Supervenience and Mind), argues that nonreductive materialism (NRM) and emergentism (E) have the same cash value. He identifies four key theses that NRM is committed to: (1) all concrete particulars are physical (2) mental properties are not reducible to physical properties (3) all mental properties are physically realized and (4) mental properties are real properties of objects and events. Kim finds that E is committed to the same four theses: E accepts a materialist ontology (1), accepts that emergent properties are not reducible to their ‘basal conditions’ (2), accepts that higher-level or emergent properties need a physical base; this physical base is itself sufficient for the emergence of these properties (3) and finally, E accepts realism about the mental (4). These agreements are sufficient to show that NRM and E are, more or less, the same thing. Thus, any problems had by one are had by the other. Should NRM face an insurmountable difficulty, E will as well, and should E face its own insurmountable difficulty, then NRM also will.
Kim draws on Samuel Alexander, an early emergentist who to flesh out just what is meant by ‘realism of the mental’: ‘Epiphenomenalism supposes something to exist in nature which has nothing to do, no purpose to serve, a species of noblesse which depends on the work of its inferiors, but is kept for show and might as well, and undoubtedly would in time be abolished‘ (p. 348) and concludes that ‘to be real is to have causal powers’. Now, if the mental is emergent, and realism of the mental entails mental causal powers, then causal powers are also emergent. This shouldn’t be terribly surprising: one of the great struggles on philosophy of mind is to avoid epiphenomenalism, and this is surely one way of doing it. In addition, however, to being emergent, causal powers on this account exercise downward influence (or power):’…mentality, having emerged from physical-biological processes, takes on a causal life of its own and begins to affect what goes on in the underlying physical-biological processes (p. 349). This should, again, not be terribly surprising: downward causation is one of the, if not the, main distinctive of E. Without downward causation, E loses most of its force. So, then, given that NRM and E both accept the reality of the mental, both are committed to the causal power of the mental and downward causation. ‘If M is a mental property, therefore, M must have some new causal powers’ (p. 351)
Here difficulties begin to emerge, however. If being M entails having causal powers, then M is causally efficacious in bringing about another property, N. But recall (2) above: this means that M is physically realized by its base P. If we ask what the cause of whatever M causes (say, M*, another mental property) then we instinctively want to say that M caused it, but it appears that P does some causal work too by realizing M. Should we say that there is a joint causal power, then we contradict the idea that M is realized by P, since P is sufficient to bring about M. The same result follows if we want to say that M* has two distinct, not joint, causes. All this leads to Kim’s conclusion: ‘I believe the only coherent story we can tell here is to suppose that that the M instance caused M* to be instantiated by causing P, M’s physical realization base, to be instantiated. This of course is downward causation, from M to P (p. 352). This downward causation is what NRM and E must be committed to in order to preserve any real causal efficacy and indeed to preserve any realism about the mental at all. There is, unfortunately, according to Kim, a near fatal objection to downward causation:
Suppose that the mental property M is causally efficacious with respect to physical property P*, and in particular that a given instance of M causes a given instance if P*. Given the Physical Realization thesis, this instance of M is there because it is realized by a physical property, say P. Since P is a realization base for M, it is sufficient for M, and it follows that P is sufficient, as a matter of law, for P*. Now the question that must be face is this: What reason is there for not taking P as the cause of P*, bypassing M and treating it as an epiphenomenon? (p. 353-354)
There are a number of reasons according to Kim that this charge is difficult for both NRM and E to answer. The objection above is certainly the most simple: perhaps Occam’s razor trims away unneeded causes. Why accept M as doing causal work when we simply don’t need to? There is also the objection from causal-overdetermination, since P* would be caused by both M and P: this would turn both P and M into sufficient causes of a single event, which doesn’t make much sense. Nor can P and M be joint causes, as we saw above. These objections lead Kim to a rather stark conclusion for NRM and E:
All causal relations are implemented at the physical level, and the causal relations we impute to higher-level processes are derivative from and grounded in the fundamental nomic processes at the physical level…if, as the supervenience thesis claims, all facts are determined by physical facts, then all causal relations involving mental events must be determined by physical facts (p. 355)
Kim lays this line of thinking out more formally as ‘the causal inheritance principle’: If M is instantiated on a given occassion by being realized by P, then the causal powers of this instance of M are identical with the causal powers of P. This is, as Kim notes, a devastating principle to NRM. There is perhaps no starker way to put it than if the principle is true, then there is no way for NRM or E to have real causal powers for the mental. Mental realism is denied, and the mental, according to Alexander’s principle invoked earlier, ought to simply be abolished. This is something of a game-over moment for NRM and E: since physicalism is also committed to the causal closure of the physical, unless one is committed to discarding that closure (and, if one is, as Kim asks, why call yourself a physicalist?) then there appears to be no move for NRM and E to make. Kim notes that most emergentists probably won’t have a problem with the loss of causal closure – this is in all likelihood part of the emergent program. So we have something of a dilemma: in order for NRM and E to avoid epiphenomenalism, realism about the mental must be maintained. But in order for realism about the mental to be maintained, it appears that abandoning the causal close principle is the only way to go, and to abandon this principle is to abandon physicalism.