Musgrave's Critique of Van Fraassen's Constructive Empiricism

Introduction: Musgrave's Three-Part Critique

Musgrave's essay is a critique of the anti-realism — or constructive empiricism — of Bas van Fraassen. Musgrave divides his essay into three sections: an explanation of van Fraassen's attempt to demolish scientific realism; his insistence that van Fraassen succeeds no better than his predecessors in answering a major objection to anti-realism; and the link between realism and explanation, along with van Fraassen's attempt to sever that link.

Van Fraassen's Challenge to Scientific Realism

According to van Fraassen, realism can be defined in the following way: "Science aims to give us, in its theories, a literally true story of what the world is like, and acceptance of the scientific theory involves the belief that it is true" (1088).

Van Fraassen does not go to the extreme, as some do, of rejecting science absolutely. He accepts that scientific statements have a truth value — that is, they are either true or false. At the same time, however, he rejects the positivist stance that discounts experience not grounded in the observable, and that treats only the observable and empirical as meaningful and valid. His anti-realism operates at the epistemological or methodological level, where he contends that a "theory need not be true to be good." Theories need only be correct as far as observations and experiments go, and this in itself makes a theory adequate — even if it does not conform to positivism (i.e., is not "empirically observable"). A theory can concern unobservables and still be scientifically adequate.

Van Fraassen also distinguishes between realism and strict empiricism. Although both may appear similar in that both rely on observables for testimony, there is a subtle difference. Strict empiricism relies on empiricism at all times; realism, however, insists that scientific statements must make sense and have a truth value whether or not they are observable (and they are not always). Realism accordingly allows non-evidential or "metaphysical" arguments to inform its reasoning.

Sometimes, empirically equivalent but mutually incompatible theories have arisen in real science. For instance, Newton hypothesized that the center of gravity of the solar system is at rest in absolute space. At the same time, he also acknowledged that appearances would be no different if that center were moving through absolute space at any constant velocity. Newton claimed these two theories were empirically equivalent.

Musgrave criticizes van Fraassen's interpretation of Newton's incompatible theories, arguing that van Fraassen does not demonstrate that they have ceased to be empirically equivalent. In several respects, van Fraassen disputes strict empiricism and argues that it does not make sense or does not exist — but Musgrave challenges each of these moves.

Van Fraassen also attacks the realist by arguing that the realist must forge a link between simplicity and truth in order to accept a theory as true. But this link can only be forged by a metaphysical principle, which runs counter to realist philosophy — and this, van Fraassen contends, amounts to the abandonment of strict empiricism.

Anti-Realism's Unresolved Objection: Theory Versus Observation

Musgrave sees the constructive empiricist as occupying a more comfortable position here, since the constructive empiricist will not be troubled by the complexities surrounding simplicity. The constructive empiricist admits that simplicity is needed for good science to occur, but maintains that the ultimate aim of science is empirical adequacy. Returning to van Fraassen's anti-realist arguments, Musgrave nonetheless finds it possible for the realist to counter them through various means. He does not see why the intrusion of simplicity into the scientific construction of experimentation is necessarily harmful. While this may involve a metaphysical leap, the focus should remain on the pragmatic, and the end result can be a form of scientific empiricism that legitimizes the scientific realist's position.

Musgrave argues that anti-realists need to distinguish clearly between theory and observation, and that van Fraassen errs by equivocating between the two. If the best explanation is a theory about the observable, then theory and observation coincide, and we can conclude that there is a theory about an observable variable. If, however, they do not coincide, then empirical adequacy and truth come into conflict, and we cannot conclude that the best explanation in science coheres with realism — because it actually does not.

Realist ways of thinking and talking are difficult to escape entirely. The unconscious acceptance of the truth of a statement and the empirical adequacy of the theory appear to be linked. For instance, as Musgrave points out, van Fraassen speaks of detecting an electron in a cloud chamber. Inherent in this statement is his implicit belief that the object really exists — otherwise, he would not speak of detecting it. In this way, van Fraassen shows a full commitment to the theory of electrons, even though various theories of electrons exist and none has been definitively accepted. Van Fraassen is thus, ironically, as realist as those whose approach he is trying to dismantle.

Finally, Musgrave argues that van Fraassen fails to demolish realism because his argument is sometimes incoherent. Van Fraassen says that what is observable by humans is a "function of facts about us qua organisms in the world" and that "it is for science to tell us what is observable and what not" (1097). However, almost any — if not all — theories necessarily integrate unobservable elements together with observable ones. Therefore, van Fraassen's position would have us accept few if any theories as scientifically valid. The constructive empiricist can only accept a theory if it is empirically adequate, but if part of a theory is not observable by humans, this militates against what the constructive empiricist can accept. Moreover, the consistent constructive empiricist cannot accept that anything is, in principle, unobservable by humans.