Straw Men and Windmills

There have certainly been ample well-targeted critiques of constructive empiricism, which had to be taken very seriously since its formulation in 1980. But there are also disconcertingly many attacks on various ‘empiricisms’, apparently designed for the purpose, whose humiliating defeats could insinuate much beyond their literal meaning.

For clear examples of critiques of empiricism biased by a tailor-made definition I can refer here to David Papineau and Alan Musgrave.

3.1 David Papineau’s Empiricist

An example that may have reached many classrooms is in Papineau’s introduction to his collection The Philosophy of Science:

According to van Fraassen’s ‘constructive empiricism’.. we ought never to believe in the truth of any theory which goes beyond the observable phenomena. (Papineau 1996, 8).

That “ought” may be true of some conceivable empiricist philosophy of science, but I don't know of any actually professed, and it is certainly not true of con­structive empiricism.

Papineau's collection includes the article “To Save the Phenomena”, in which the wording is very precise, and never gives leeway to such an implication about what we ought to believe:

anti-realism is a position according to which the aims of science can well be served without giving... a literally true story, and acceptance of a theory may properly involve something less... than belief that it is true. (van Fraassen 1976, 623; Papineau 1996, 82)

This careful formulation respects the distinction between what is involved in acceptance of a theory and what someone who accepts a theory may believe apart from that. That is surely not a distinction difficult for a philosopher to appreciate. Whoever accepts a theory will have many opinions and beliefs that go beyond what is involved in that acceptance, and this philosophical position has no implications for what may or may not be included in those additional opinions and beliefs.

In The Scientific Image this distinction is emphatically observed as well:

Science aims to give us theories which are empirically adequate; and acceptance of a theory involves as belief only that it is empirically adequate. This is the statement of the anti-realist position I advocate; I shall call it constructive empiricism. (van Fraassen 1980, 12).

Much is deliberately left open, including that it is not by any lights irrational to believe in the truth of a scientific theory. The italicized assertion implies only that the criterion of success in science, and hence the belief involved in acceptance of a theory as successful, does not demand so much:

I do not advocate agnosticism about the unobservable, but claim that belief is supererogatory as far as science is concerned; you may if you like, but there is no need. (van Fraassen 2007, 343).

Empiricist leanings will incline one to believe less rather than more, that is true. But if anyone were to have the position that “we ought never to believe in the truth of any theory which goes beyond the observable phenomena”, as Papineau puts it, what could be the reason for that? The only reason there could be is that the constraints of rationality require that disbelief! But what constraint could that be, in the case of a theory consistent with all that we know about the phenomena? The idea is entirely at odds with the liberal form of epistemology that I espouse: rationality is but bridled irrationality.^[21]

3.2 Alan Musgrave’s Empiricist

Leaving epistemology aside now, for a moment, there are also questions about what is involved in theory acceptance other than belief. For any empiricist philosophy of science that must be a crucial question for scientific practice, and the empirical inquiry that it involves, is clearly theory-driven.

Alan Musgrave is much more careful than Papineau, and in his paper “Strict Empiricism versus Explanation in Science” he is careful to define his target ‘strict or hardline empiricism':

What is strict or hardline empiricism? It is the view that only empirical evidence should determine theory-choice.

nomena”..,.^[22]

What I have omitted in this paragraph includes “also known these days as ‘constructive empiricism'”. Rather than cavil about this again, let me address Musgrave's implication that on an empiricist view, theory-acceptance is reduced to a reliance on the evidence at hand. As I see it, Musgrave is quite right that any such view would be woefully inadequate.

Science aims to give us theories which are empirically adequate; and acceptance of a theory involves as belief only that it is empirically adequate (van Fraassen 1980, 12). But what does acceptance involve beyond belief? A practicing scientist who accepts a theory (what Musgrave refers to as theory choice) involves himself in a research program, s/he is committed to meeting the phenomena within the the­oretical framework in question, and this program could be much different if an alternative, though empirically equivalent, theory were accepted (ibid.).^[23] So clearly, it is not only empirical evidence that determines theory-choice, and it is hard to imagine any philosopher naive enough to think so.

Musgrave's strict empiricist is not a good philosopher of science because s/he ignores the pragmatic dimension of theory acceptance. Dennis Dieks' contribution to the same conference provides us with examples of current interest.^[24] Anyone who had quantum mechanics in their formative education years was immersed in a plebeian form of the Copenhagen interpretation. That includes those who, for example, turned later to actively developing Bohmian mechanics or the GRW version of quantum mechanics, and have thus immersed themselves in a new conceptual framework in which to engage the phenomena-new phenomena that the experimentalists are meanwhile creating in fabulous new ways.

Much as Musgrave values explanation, he does not, in the context of this critique of empiricism, explore the pragmatic dimensions of our explanatory practices, nor ask just what explanation is good for, or what purposes it may serve. It seems to me that this might well be an area in which scientific realists and constructive empiricists could cooperate, in order to reach a better understanding of science in practice.^[25]

There can clearly be conceptual and pragmatic differences between theories that are empirically equivalent, and these can play an important role in theory choice. When a practicing scientist accepts a theory, the stakes (personal, communal, intellectual) can be great, there is a gamble here on the future of the scientific discipline. So it is not surprising if Musgrave finds that there will be differences, important to science and in practice, between a given theory T and a theory T* which is true if and only if T is empirically adequate. That will typically be so. (Only typically!^[26]).

Musgrave’s strict empiricist would be embarrassed by this. That is probably why there aren’t any strict empiricists.

4