By Daniel A. Kaufman
On tap this week is one of the most influential essays in the philosophy of science, since the Second World War: Jerry Fodor’s “Special Sciences,” which appeared in the journal Synthese, in 1974.
The paper did two very important things. First, it struck a crippling blow against a certain kind of positivist view of the relationship between the special sciences and the physical sciences. Second, it undermined a common positivist fantasy about the sciences as a whole, one that it is commonly held by people who otherwise would not identify themselves as positivists (the physicist Lawrence Krauss immediately comes to mind).
What counts as a special science is not entirely clear. When I studied this paper, back in college and graduate school, several decades ago, the special sciences were presented as consisting of the social sciences and psychology. I have sometimes heard the term used in a way that also includes biology. The Wikipedia entry on the subject deems any science other than physics a special science, including chemistry. Given, however, that the reason for talking about the special sciences at all has to do with the question of their reducibility to the physical sciences, it seems strange to describe chemistry, a physical science — and the one for which there is the least question about reducibility – as being a special science. Of course, there are distinctive problems that arise with respect to the reduction of biology to chemistry and physics, but insofar as Fodor’s focus is on the relationship of the social sciences and psychology to the physical sciences, I will use the expression ‘special sciences’ to refer to the social sciences and psychology.
Reductionism, as Fodor understands it (and his understanding is drawn, largely, from the canonical account of it given by Ernest Nagel, in his 1951 book, The Structure of Science), describes the following relationship between sciences:
Science A can be said to reduce to Science B, if the laws of Science A can be shown to be equivalent, in some relevant sense, to the laws of science B. The laws of Science A can be shown to be equivalent to the laws of Science B, if both the antecedents and consequents of the laws of Science A can be shown to be materially equivalent (or identical with) the antecedents and consequents of laws of Science B.
The statements describing these material equivalencies between the predicates of the reduced science and the reducing science are commonly referred to as “Bridge Laws.”
It should be made clear at the outset that the ambition of the reductionist is more than simply to assert a kind of “token physicalism,” by which we mean the view that every individual thing or event is a physical thing or event. To be a token physicalist about money, for example, is simply to hold the view that every instance of money is some sort of physical object or event, whether it be paper, gold, precious stones, or electronic codes. The reductionist’s claim, however, is much stronger than this: it is that the property of being money is itself a physical property and that therefore, any laws regarding money – such as those one might find in economics – are ultimately physical laws. This “type physicalism” is what the reductionist is ultimately after, since his main aim is to demonstrate the explanatory and ontological generality of physics. To accept that there are explanations — and therefore, laws, and therefore, types or kinds — that cannot be reduced to physical explanations, laws, types, and kinds, on the other hand, would be to reject the idea that physics is general in this way, something that the positivists were loath to admit, as they thought it meant opening the door to any number of woolly varieties of metaphysics.
Fodor’s argument is relatively simple. For any special science predicate, there is going to be an indefinite number of materially equivalent physical predicates. The trouble, then, is twofold: Not only is that disjunction of physical predicates not, in itself, a predicate of any physical science, but the resulting law, in which both the antecedent and consequent consist of disjunctions of physical predicates, is not itself a law of any physical science.
The example that Fodor uses is a law of economics — Gresham’s law — which says, essentially, that bad money drives out good money. That is, if a nation’s currency includes commodities of varying degrees of value, but which are worth the same amount in exchange – say, silver coins versus paper notes – the less valuable money – i.e. the paper – will chase the more valuable money – the silver – out of circulation and into hoards. Now ‘money’ is a term that refers to a type or kind in economics, and certainly, any particular instance of money is going to be some physical object or other (or a physical event, in the case, say, of an electronic transfer). But notice that there are indefinitely many possible physical instantiations of money: money can be instantiated in paper; gold; silver; diamonds; electrical impulses; etc. If I want to reduce Gresham’s law to some law of a physical science, then, I need to provide a bridge law, in which ‘money’ will appear on the left side of a bi-conditional and the corresponding physical type or kind appears on the right. The result is something like this.
M ↔ P1 or P2 or P3 or P4 or P5… or Pn.
When done with both the antecedent and consequent of Gresham’s Law, the fully reduced law will look something like this.
P1 or P2 or P3 or P4 or P5… or Pn → P1a or P2a or P3a or P4a or P5a… or Pna
But as already mentioned, neither the antecedent or consequent of this law is a type or kind of any physical science and the law itself, is not a law of any physical science. That is, the indefinite disjunction, consisting of paper, gold, silver, diamonds, electrical impulses, etc., is not a type or kind of any physical science, and laws containing disjunctions like these are not laws of any physical science.
The lessons of this are manifold. One is that the positivistic idea of a unified science gives way to a picture of highly decentralized, mutually autonomous sciences, each with their own distinctive vocabularies, taxonomies, ontologies, and laws. Another is that while every object and event may admit of a physical description and thus, fall under some physical law or other – as paper and gold, for example, certainly do – when it comes to the sorts of things the special sciences talk about and explain, these physical descriptions, laws, and explanations are largely beside the point. They do not speak to what interests us about these things. As Fodor puts it with regard to monetary exchanges, surely what is interesting about them is not their commonalities or behavior under some physical description or law, but rather, how they function within economies.
It is fascinating to me that this rather simple – and I would say, obvious – point has received so much pushback and that so many eminent scientists are still shilling not just for reductionism, but for the larger “unity of the sciences” thesis. Fascinating, but not surprising. Hope, after all, springs eternal, and for those whose entire careers have been invested in “theories of everything,” the unity of the sciences thesis is a crucial component of that hope.