Jaegwon Kim criticizes John Searle’s biological naturalism for the problem of causal overdetermination. Kim suggests the identity-theory to clean up the mess, but Searle responds with an incoherent recapitulation of his biological naturalism using two senses of causation as a linguistic trick to avoid the warranted charge of overdetermination. Furthermore, Searle’s response reveals that he is a property dualist and refuses to believe it himself.
Searle’s biological naturalism from his piece Mind: A Brief Introduction, in his own terms, which supposedly make all the difference, claims that “conscious” states (he will not call them mental states) are real phenomena in the real world that cannot be ontologically reduced to their biological processes because they bear a subjective, first-person ontology for which such reductions will neglect to account. These conscious states, however, are caused by lower-level neurobiological processes, which is to say, Searle claims, that they are causally, but not ontologically, reducible to these processes. Conscious states are real features of the brain “system”, only realized at a higher level and are real features of the real world, so they function causally, flat-out rejecting epiphenomenalism. As Searle regularly demonstrates, “I want to raise my arm, and the damn thing goes up!”
Furthermore, the following system is allegedly entirely compatible with Searle’s biological naturalism, however whether it metaphysically resembles the world is another issue. Being a real feature of the world, a “conscious” state, s1, a pain, let us suppose, causes another conscious state, s2, the desire to take aspirin. Since s1 and s2 are both higher-level phenomena of the brain “system”, they must each be caused by a lower-level neurobiological process, let us say p1 and p2, respectively. However, p1 also causes p2 at the lower-level description of the brain system.
Kim’s warranted criticism of biological naturalism lies in question as to what causes s2. According to the above situation, s2 is caused by s1, that is, the desire to take aspirin is caused by the pain. However, then s2 is casually overdetermined, because every conscious state is caused by a lower-level neurobiological process, in the case of s2, by neurobiological process p2, but we just noted that it is also caused by the conscious state of pain, s1. So s2 has two causes.
It seems at this point Searle has two options to clean up his apparently incoherent view: to allow for epiphenomenalism, or, as Kim suggests, the identity theory.
If Searle allow that the conscious higher-level states are causally inert, then as for the conscious states, Searle’s system will not be causally overdetermined. If he concedes that s1 does not actually cause s2, then s2 is not overdetermined. p1 will then cause s1 and p2, and p2 will cause s2. However, Searle’s biological naturalism stresses that each lower level causes the manifestations of the higher level, so even though epiphenomenalism would eliminate overdetermination in the higher-level of conscious states, the neurobiological processes would be overdetermined.
“[W]e could keep going on down to the level of quarks and muons” as Searle says in his reply to Kim, and if we did we would find that p2 would not only be caused by p1 but also by some even lower-level quark/muon process, q2, let us say (“Reply to Kim” 2). All this said, Searle’s own illustration of biological naturalism makes explicitly clear that he believes conscious states are quite causally potent—“… and the damn thing goes up!”
Kim suggests that Searle’s theory’s structure can make sense if he appeal to the identity theory. If Searle wants to claim that “pain causes the desire to take aspirin” and “neurobiological process p1 causes neurobiological process p2” are descriptions at different levels of the same phenomena, then to avoid overdetermination, the pain must be the very same thing as p1 and the desire to take aspirin be p2, not caused by them. s1 and s2 just are p1 and p2, respectively, that is, they are referring to the exact same phenomena. In that case s2 will not be causally overdetermined, for it will only have one cause, s1.
However, in his response letter to Kim, Searle rejects Kim’s suggestion and makes use of a two notions of causation to supposedly avoid causal overdetermination. There is what Searle refers to as, left-to-right causation over time, of a later event by an earlier event but there is also bottom-up causation that is not over time—it is simultaneous, rather. s1 causes s2 and p1 causes p2 in the left-to-right, over time sort of causation, whereas p1 causes s1 and p2 causes s2 in the bottom-up sense of causation.
Searle claims that m2 is not overdetermined because “we are not talking about two causes, but one system of causation described at different levels” (“Reply to Kim” 3; emphasis his). Though he may say that there is just one system of causation and implies that this somehow provides the awaited solution, and though he never states it in the same sentence, he does say that m2 has two causes, one bottom up and one left-to-right. So he must be using two distinct senses of causation. Searle even concedes that “it’s okay with [him] if [Kim] say that the pain is explained by neuronal processes” and not bottom-up caused by it (2). He clearly admits that he uses two different senses of cause, one explanation and the other our standard “Humean”, as Searle accuses it, causation.
As a result, it is clear that Kim is not mistaken in finding something confusing about Searle’s using the same word for two different notions. In fact, there seems little different about Searle’s bottom-up notion of causation and the explanation used in reductive analyses, like that of the identity theory. In such reductions, the higher-level features are explained by their lower- level features, so Searle has reduced pain to the neurobiological process. Moreover, if m1 and m2 are only explained by p1 and p2, then Searle should not so reluctant to concede that m1 and p1 are referring to the same phenomena, much in the same way a certain configuration of atoms is a hammer’s head. Kim is right: the identity theory is not a bad idea.
However, Searle tries to avoid the identity theory because he believes that conscious states are ontologically subjective, and therefore are not reducible to the physical, neurobiological processes. He does this by introducing a nonstandard notion of causation, which resembles a common notion of reductive explanation. Just because he uses a different word does not allow him to escape the fundamental problems in his view.
The typical, and according to Searle, misled, notion of identity, if m1 be identified with p1, will leave out the subjective qualitative character, the first-person ontology of m1. Such an identification will claim that there is nothing about the pain, m1, that is “over and above” the neurobiological process, p1, but he believes that there is something “over and above” p1 about m1. The pain is ontologically irreducible, because any ontological reduction leaves out the “touchy-feely” qualities of the higher-level conscious state.
Kim is right in saying that Searle can save his biological naturalism if he appeals to the identity theory, but he must also abandon his conviction that conscious states are ontologically irreducible. If he does not, then his view as proposed is property dualism. ‘Conscious states are caused, in the nonstandard, explanatory sense, by lower-level neurobiological process’ which is a physicalist view, married with, ‘conscious states have something special about them, a unique property that causes them to resist reduction’, entails property dualism. He clearly holds these views, but would certainly object to my use of this language.
At this point he would likely say that the reduction of conscious states to neurobiology is not an ontological reduction like their ontological irreducibility. Though may have averted attention away from the identity theory, he has now argued for property dualism. Even though the bottom-up causation can be thought of as explanation, it does not explain anything. He acknowledges that conscious states are different than other higher-level descriptions of other systems: “in the case of consciousness, unlike say, solidity, the causal reduction does not lead to an ontological reduction”. So this bottom-up causation or explanation does not actually explain anything.
What is the reason for this unique feature of conscious states? Even if we allow that they are caused by lower-level neurobiological processes, there is still the mind-body question: how are they caused by lower-level states? How on earth does subjective experience arise from neurobiological processes? He does not know, and his theory sheds light on none of the fundamental issues of the mind-body problem.
In Searle’s response to Kim, I think it is quite telling that in his last section of points addressing “Dualism” he says that “because the first person ontology is not ontologically reduced to a third person ontology we are tempted to think that there must be some kind of dualist ontology leftover… But of course that is a massive mistake” and does not refute the charge of dualism any further (“Reply to Kim” 5-6).
Nothing is keeping us from distinguishing Searle’s biological naturalism from property dualism. If m1 and m2 are reducible in every way except ontologically to p1 and p2, then obviously there is another property, that is, the ontology, of the higher-level conscious states that resists reduction. Though he may not wish to call these special states mental, and prefers to call them “conscious”, and though we can allegedly reduce them causally, there are two kinds of properties in his view: sticking to Searle’s arbitrary terminology replacement, conscious properties and physical properties that are obviously as distinct as Descartes’s mental and physical substances.
