Van Fraassen Quote on Empiricism: Empirical Adequacy

Quote

"Science aims to give us theories which are empirically adequate; and acceptance of a theory involves as belief only that it is empirically adequate."

— Bas van Fraassen, The Scientific Image (1980, Chapter 2)

Context

This statement appears in Chapter 2 of The Scientific Image (1980), the book that introduced constructive empiricism and reinvigorated the debate between scientific realism and antirealism. Van Fraassen was responding to a consensus, dominant in mid-twentieth-century philosophy of science, that the success of science — especially its predictive success — could only be explained by the truth of scientific theories, including their claims about unobservable entities like electrons, genes, and quarks.

Scientific realism holds that mature scientific theories should be accepted as approximately true, including their claims about unobservable entities. The "no miracles" argument, advanced by Hilary Putnam, held that the success of science would be a miracle unless its theories were at least approximately true. Against this, van Fraassen argued that science does not aim at truth about unobservables but at empirical adequacy — agreement with what can be observed. A theory is empirically adequate if everything it says about observable phenomena is true, even if its claims about unobservable entities are false or have no determinate truth value.

The Scientific Image was awarded the Lakatos Award and became one of the most influential works in the philosophy of science. Van Fraassen's strategy was not to deny that science is successful but to offer an account of that success that does not require belief in unobservables. Constructive empiricism holds that science is a rational, truth-seeking enterprise, but that its aim is truth about the observable, not truth about everything the theory posits.

Meaning

Van Fraassen's dictum has two parts that together define constructive empiricism:

"Science aims to give us theories which are empirically adequate": The aim of science is not truth, wholesale, but empirical adequacy — the property of a theory that correctly describes all observable phenomena. A theory may posit electrons, curved spacetime, or selfish genes; what matters, epistemically, is whether what it says about things we can observe is correct. The unobservable posits are instruments for generating empirical predictions, not objects of belief.

"Acceptance of a theory involves as belief only that it is empirically adequate": When a scientist accepts a theory, they commit to the belief that it is empirically adequate — that it will survive all possible observational tests. They do not commit to the truth of the theory's claims about unobservables. Acceptance also involves a practical commitment — to use the theory in inquiry, to trust its predictions, to build on it — but the epistemic commitment is limited to empirical adequacy.

The distinction between truth and empirical adequacy is crucial. A theory could be empirically adequate (all its observable predictions true) yet false in its unobservable claims. Consider two theories that agree on all observable consequences but differ in what they say about unobservable structure. Both are equally empirically adequate; the realist must hold that one is true and the other false, but the constructive empiricist holds that we have no grounds — and no need — to choose between them on epistemic grounds. The observable evidence underdetermines the unobservable truth, and van Fraassen takes this underdetermination to show that belief in unobservables goes beyond what science, properly understood, requires.

This makes constructive empiricism a form of epistemic voluntarism: belief in unobservables is permissible but not rationally required. Van Fraassen does not claim that unobservable entities do not exist; he claims that science does not oblige us to believe they do. The scientist who accepts a theory is epistemically entitled to believe only its empirical adequacy, and everything else — including the existence of electrons — is a matter of personal epistemic risk, not scientific obligation.

Philosophical Significance

Constructive empiricism reshaped the realism-antirealism debate:

The challenge to scientific realism: Van Fraassen offered the most sophisticated and influential challenge to scientific realism, forcing realists to clarify and defend their position. The realist must explain why the success of science warrants belief in unobservables, not just in empirical adequacy. Van Fraassen's argument from underdetermination — that multiple, incompatible theories can be equally empirically adequate — put pressure on the realist's claim that science converges on the truth about unobservable structure.

The epistemic significance of observability: Constructive empiricism turns on the distinction between observable and unobservable entities. Van Fraassen argues that this distinction, while vague at the margins, is scientifically meaningful: some entities (planets, cells) are observable, others (electrons, quarks) are not. The debate over whether this distinction can bear the epistemic weight van Fraassen places on it has generated a substantial literature on observation, instruments, and the limits of the unaided senses.

Explanation and inference to the best explanation: Realists often defend their position by appeal to inference to the best explanation — the success of a theory is best explained by its truth. Van Fraassen countered with a "pragmatic" account of explanation, arguing that explanation is not a matter of inference to truth but of answering why-questions relative to a context. This reframed the debate over the epistemic force of explanatory considerations and influenced a large literature on the nature of explanation.

The voluntarist conception of rationality: Van Fraassen's epistemic voluntarism — the view that rationality permits a range of epistemic stances, not all of which are compelled by the evidence — has influenced Bayesian epistemology and the philosophy of science more broadly. The idea that acceptance of a theory involves pragmatic commitments beyond belief, and that rational agents may legitimately differ in the epistemic risks they take, has shaped discussions of rationality, belief revision, and the ethics of scientific practice.

Sources

1. Van Fraassen, Bas C. The Scientific Image (Oxford University Press, 1980), Chapter 2. Context in: https://plato.stanford.edu/entries/fraassen/
2. "Bas van Fraassen," Stanford Encyclopedia of Philosophy. Available at: https://plato.stanford.edu/entries/fraassen/
3. "Scientific Realism," Stanford Encyclopedia of Philosophy. Available at: https://plato.stanford.edu/entries/scientific-realism/
4. "Scientific Realism," Internet Encyclopedia of Philosophy. Available at: https://iep.utm.edu/scientific-realism/
5. Musgrave, Alan. "Realism versus Constructive Empiricism," in Essays on Realism and Rationalism (Rodopi, 1999).