Is Reductionism Necessary for the Unity of Science?
Abstract: This paper looks at Jerry Fodor's argument from his paper “Special Sciences” where he argues that reductionism is not necessary for the unity of science or token physicalism. I argue that Fodor is mistaken and that his attempt at providing a justification for the unity of science that doesn't make use of any type of reductionistic claim is circular and that the unity of science does require some form of reductionism.
Introduction
In his paper “Special Sciences” Jerry Fodor argues that if our main concern is to be “good token physicalists” then we do not need reductionism to be true, which is of course a very fortuitous fact because reductionism happens not to be true. And he also argues that reductionism is not required for the unity of science. In place of classical reductionism that requires bridge laws connecting the reduced and reducing science Fodor offers an account of how we can be good token physicalists and still endorse the unity of science while avoiding the problems of reductionism. In place of reductionism he suggests identity conditions that would guarantee that an event being described in the special sciences is a physical event, thereby justifying token physicalism and the unity of science. He argues that if a description in the language of physics can be given of all the instances being described by a the law of the special sciences then we have enough evidence to support token physicalism and the unity of science.
I will be arguing that Fodor is mistaken and that reductionism is required for token physicalism, because without reductionism we have no reason, other than our intuitions on the subject, to assume that just because an event can be described in the language of physics it is a physical event. I will argue that Fodor's claim that identity conditions are enough to guarantee the unity of science does not adequately answer the question of whether token physicalism and the unity of science are true because merely offering a physical description in place of bridge laws assumes the truth of the unity of science rather than proving it.
Fodor's Argument Against Classical Reductionism
Fodor's uses the following set of formulas to describe the form of reductionism that he will be arguing against:
(1) S1x → S2y
(2a) S1x ↔ P1x
(2b) S2y ↔ P2y
(3) P1x → P2y
(1) refers to a law in the special sciences, and (3) refers to a law in physics. (2a) and (2b) refer to bridge laws that connect predicates of the reduced and the reducing science so that the bridge laws specify the coextensions between the reduced and reducing science. It seems like this account should be true if the special sciences are in fact special cases of physics, but when one actually looks at the laws of the special sciences such a reduction seems impossible.
Fodor uses the example of Gresham's law to illustrate the difficulty with this form of reductionism. The problem in attempting to provide a successful reduction of Gresham's law is that it is a generalization that covers many “wildly disjunctive” events, the physical descriptions of which have nothing in common. As Fodor says, some events of monetary exchange involve, dollar bills, some involve writing your name on a check, and some involve strings of wampum (Fodor 433). But, according to Fodor, the problem for reductionism is even worse than the preceding has indicated because:
. . .reductionism claims not only that all kinds are coextensive with physical kinds, but that the coextensions are nomologically necessary: bridge laws are laws. So, If Gresham's law is true, it follows that there is a (bridge) law of nature such that 'x is a monetary exchange ↔ x is P' is true for every single value of of x, and such that p is a term for a physical kind. (Fodor 433)
This of course seems impossible when one considers, as Fodor says, that P would not only have to cover all systems of monetary exchange but also all possible systems of monetary exchange. Fodor rhetorically asks, “What physical predicate is a candidate for P in ' x is a nomologically possible monetary exchange iff Px'?” (Fodor 433). Therefore, he concludes that economics is not reducible to physics “in the special sense of reduction involved in claims for the unity of science” (Fodor 434). And he thinks economics is not going to be unique amongst the special sciences in its failure to successfully reduce to physics.
Fodor next moves from economics to psychology. He says, “There is, I suspect, nothing peculiar about economics in this respect; the reasons why economics are unlikely to reduce to physics are paralleled by those which suggest that psychology is unlikely to reduce to neurology” (Fodor 434). What would it mean for psychology to be reducible to neurology? It would mean that, “. . .for every psychological kind predicate there is a coextensive neurological kind predicate, and the generalization which states this coextension is a law” (Fodor 434). He gives two separate arguments showing that psychology is not reducible to neurology. The first argument is that there may not be a lawful relationship between brain states and psychological states, or at least any relationship that can be precisely expressed as a law that would reduce psychology to neurology. Fodor gives three reasons why he thinks this is so. Fodor's three reasons are:
1)There are no firm data for any but the grossest correspondence between types of psychological states and types of neurological states. . .
2). . .it is entirely possible that the nervous system of higher organisms characteristically achieves a given psychological end by a wide variety of neurological means.
3)It is also possible that given neurological structures subserve many different psychological functions at different times, depending upon the character of the activities in which the organism is engaged. (Fodor 434)
Fodor's second argument is that there are other nomologically possible systems, automata, “which satisfy the kind predicates of psychology but which satisfy no neurological predicates at all” (Fodor 434). Taken together Fodor's arguments against the reducibility of Gresham's law and of psychology are used as grounds for rejecting reductionism, however he does not see this as a problem for the unity of science or for token physicalism. And we get a hint of the general approach that he will take early on when he says, speaking about Gresham's Law, “I am willing to believe that physics is general in the sense that it implies that any event which consists of a monetary exchange (hence any event which falls under Gresham's law) has a true description in the vocabulary of physics and in virtue of which it falls under the laws of physics” (Fodor 433).
Fodor's Solution to the Problem of Reductionism
Reductionism is, according to Fodor, “more than we need to assume if what we primarily want, from an ontological point of view, is just to be good token physicalists” (Fodor 436), and what Fodor attempts to do in the second part of his paper is sketch a view that could provide the conditions under which we could consider any law of the special sciences to be explained by physics, thereby preserving token physicalism and the unity of science. Fodor rejects reductionism but cannot completely abandon the unity of science because to do so would undermine token physicalism. If he were to abandon unification completely it would be questionable whether token physicalism were true. The unity of the sciences is after all supposed to be the proof of physicalism. The question is “Can Fodor's methodology preserve the unity of science despite its rejection of reductionism?”
Fodor assumed that the problem for reductionism was the heterogeneous and disjunctive nature of predicates in the laws of the special sciences, and he attempts to capture this disjunctive nature by substituting a disjunctive identity statement for the bridge laws of classical reductionism. The new bridge statements take the following form (Fodor 436).
(4)Sx ↔ P1x ∨ P2x ∨. . . Pnx
In explaining this new bridge law Fodor says:
I am thus assuming that it is enough, for purpose of the unity of science, that every law of the special sciences should be reducible to physics by bridge statements which express true empirical generalizations. Bearing in mind that bridge statements are to be construed as species of identity statements, formula (4) will be read as something like 'every event which consists of x's satisfying S is identical with some event which consists of of x's satisfying some or other predicate belonging to the disjunction P1 ∨ P2 ∨ . . . ∨ Pn'. (Fodor 436)
This account of reductionism or unification is supposed to express the basic token physicalist intuition that all things are physical things. So, in place of a bridge laws that would accomplish this reduction of taxonomies Fodor uses bridge “laws” that makes identity claims identifying any event in the reduced science with some physical event.
However I do not believe that this account of the unity of science can be used, as Fodor wants it to, to support token physicalism. It appears that what Fodor wants to say is that any event that can be described in the language of physics is a physical event. But my concern is that without reductionism it is not clear that a physical description is a description of the event that is to be reduced by physics. The key word in Fodor's definition is “identical.” He translates his bridge “law” as saying “every event which consists of x's satisfying S is identical with some event which consists of of x's satisfying some or other predicate belonging to the disjunction P1 ∨ P2 ∨ . . . ∨ Pn”, but my concern is how could one would actually know that the events are identical, and to me it seems the answer is that one would not know this without assuming token physicalism to be true.
Using the example of Gresham's law let us see if Fodor's identity conditions actually accomplish what he wants them to, justification of token physicalism and the unity of science. Remember, the problem with classical reductionism was that it was impossible for there to ever be a physical law that could possibly cover all the different instances of monetary exchange that Gresham's law applies to. Fodor's solution is to replace the traditional notion of bridge laws with a disjunctive identity statement that covers all possible instances of Gresham's law. The underlying intuition here being that, as Fodor says at the conclusion of the paper, “If physics is to be a basic science, then each of these things had better be a physical thing” (Fodor 440).
However, as I suggested earlier, it is not clear to me that this approach really accomplishes what Fodor wants. My concern with this line of thinking is that Fodor's attempt at unification without reduction doesn't really accomplish what he needs it to. Because it is not enough to provide a physical description of an event to prove that the event is a physical event. How do we know the physical event is “identical” with the event being described?
It seems to me at least a valid concern that a description of the physical objects involved in a law like Gresham's law may be missing something very important, namely Gresham's law. Gresham's law does not seem to be talking about physical objects. It is not a description of the way physical objects move under certain conditions, otherwise it would be easily reducible to physical laws. Fodor assumed that the problem for reductionism was the disjunctive nature of the physical descriptions of the laws of the special sciences but that cannot be the fundamental problem. No form of classical reductionism is going to be able to reduce Gresham's law to physics in a way that justifies the unity of science, but neither is Fodor's suggestion. Gresham's law is not talking about physical objects so how is Fodor's suggestion that we identify Gresham's law with the physical descriptions of objects that happen to be described by Gresham's law supposed to prove that the things being described by Gresham's law are physical. If Gresham's law is not talking about physical objects how is a description of physical objects supposed to be a be sufficient evidence for the unity of science? Gresham's law and other laws of the special sciences are problematic for traditional reductionism, but Fodor's solution does not solve the problem. Rather his solution merely obscures the fact that he is not really providing any explanation of the special sciences. Rather than a true explanation of the phenomena, Fodor explains it away.
If the discussion so far has been unconvincing I hope the following example will make more clear the inherent problem with Fodor's solution. The problem becomes clear when we consider the reduction of consciousness because there we run into the well known problem of how a reduction, or any non-reductive physicalist account of consciousness, can really be an description of experience. Even if we assume a complete reduction of psychology to neuroscience, it is still an open question as to whether or not a successful reduction of consciousness has been accomplished because it seems that even a complete neurological description does not explain the experience of consciousness. If it is possible that a description of c-fibers firing is not an adequate description of pain then it is possible a physical description of the objects involved in a monetary transaction is not an adequate description of Gresham's law. It certainly seems possible that a physical description of a phenomena might not turn out to actually be a description of the phenomena in question and therefore it would not be sufficient evidence for token physicalism or the unity of science.
If that is true then we have good reason to doubt Fodor's thesis that token physicalism is not dependent on the truth of reductionism. As a conclusion to his paper Fodor summarizes his main point as follows:
If science is to be unified, then all such taxonomies must apply to the same things. If physics is to be basic science, then each of these things had better be a physical thing. But it is not further required that the taxonomies which the special sciences employ must themselves reduce to the taxonomy of physics. It is not required, and it is probably not true. (Fodor 440)
Fodor suggests that what is required for the unity of science is all things that the special sciences refer to better be physical things, or that everything the special sciences talk about can also be talked about in the language of physics. According to Fodor the unity of science does not require reductionism to be true, but only that all things be physical things. But how do we know that all things are physical things! How do we know we are talking about the same thing when we are simply providing a physical description? The classical form of reductionism attempts to actually connect the reducing and reduced sciences through bridges laws which included predicates of both sciences. The assumption being that this would guarantee that we were actually talking about the same phenomena. But without this genuine connection between the reducing and reduced sciences it is far from clear that one is actually talking about the same thing when one provides a physical description. Of course Fodor wants to say it doesn't matter that we are not talking about the same thing, as long as what the special sciences are talking about can be described in the language of physics we are justified in assuming they are not referring to anything non-physical. However, we would only know that all things were physical things if reductionism were true. If reductionism were false we wouldn't know if all things really are physical things. Reductionism does seem to be false and therefore, we cannot assume that token physicalism is true. Let us imagine the following conversation:
Fodor: We don't need reductionism for token physicalism.
Response: If reductionism is not true how do we know token physicalism is true?
Fodor: All we need to be token physicalists is that all things that the special sciences refer to have a true physical description.
Response: The only way we would know if they had a true physical description is if we could prove reductionism to be true.
Fodor: We would know that token physicalism is true if we could give a physical description of all events described in the special sciences. Take Gresham's law for example, although there are no bridge laws that state lawful coextensions between monetary exchange kinds and physical kinds, and therefore a classical reduction is impossible, it is enough for token physicalism to know that there is a disjunction that could state a physical description of every possible monetary exchange. We could state a new type of bridge “law” would serve as an identity statement, identifying every instance of Gresham's law with a physical description.
Psychology is similar to economics in that the problem in reducing psychology to neuroscience is that the kinds that psychology posit do not correspond to any natural kinds in physics, and therefore no law could possibly reduce psychology to neurology. Additionally consciousness can never be reduced to neurology because there are nomologically possible systems which satisfy the kind predicates of psychology but none of the kind predicates of neurology. However if we can offer a true physical description of all the instances of consciousness then that is sufficient enough reduction for token physicalism.
Response: The problem with your account is that it is not clear that just by providing a description of an event in the language of physics you have actually described the event. How do we know the events are identical as you claim. In fact it seems like a physical description of physical objects has very little to do with Gresham's law. And when we turn to psychology it become even more problematic because it is not clear that a physical description of neurological process, or any other physical process is a description of consciousness. It seems that something is lost in a merely physical reduction. And at least we must grant it as a logical possibility that there is something non-physical. So, consider: If it is a logical possibility that something could be non-physical then we cannot assume that a physical description is a description of a certain phenomena. If we cannot assume that a physical description of a phenomena is a true description then simply providing a physical description is not good enough to support token physicalism or the unity of science. And therefore the identity conditions you suggest fail to support the conclusion you want them to. Your argument only works if you assume token physicalism to be true. Of course if you assume token physicalism is true than any physical description of a phenomena is guaranteed to be a complete description. But since you cannot assume token physicalism to be true, at least not in an argument for token physicalism, your argument fails.