A physicist uses quantum field theory to make extraordinarily accurate predictions about collider experiments. According to van Fraassen's constructive empiricism, the appropriate epistemic attitude toward QFT is:
AFull belief: predictive success means we should believe QFT's claims about quarks and virtual particles are true
BRejection: theories about unobservables have no real content and are mere calculation tools
CAcceptance: believe what QFT says about observable outcomes while remaining agnostic about its unobservable ontology
DSuspension: withhold both belief and acceptance until direct observation of all posited entities is possible
Constructive empiricism recommends *acceptance* — using QFT to guide action and belief about observables — without requiring full *belief* in its unobservable ontology (quarks, virtual particles). Option A describes scientific realism. Option B is instrumentalism, which van Fraassen explicitly distinguishes from his view: constructive empiricism holds that claims about unobservables are genuinely true or false, not mere fictions. Option D conflates agnosticism about unobservables with rejecting the theory entirely.
Question 2 Multiple Choice
The no-miracles argument claims: if electrons didn't exist, theories positing them wouldn't make such accurate predictions. Van Fraassen's response is:
AConcede the argument and endorse scientific realism about well-confirmed entities like electrons
BArgue that electrons are observable under favorable conditions, so the argument doesn't apply
CArgue that empirical adequacy explains predictive success without positing unobservables, and that inference-to-best-explanation itself needs justification before licensing beliefs about unobservables
DArgue that since all theories eventually fail, the argument has no force for any theory
Van Fraassen questions the legitimacy of inference-to-best-explanation itself. The no-miracles argument assumes we should believe the best explanation of a phenomenon — but whether inference-to-best-explanation licenses beliefs about unobservables is exactly what's at issue. The argument thus begs the question. Van Fraassen adds the pessimistic meta-induction: previously successful theories have been abandoned, so predictive success doesn't reliably track truth about unobservables.
Question 3 True / False
On van Fraassen's view, a theory that is empirically adequate but makes false claims about unobservable entities can still be rationally accepted.
TTrue
FFalse
Answer: True
This follows directly from constructive empiricism. Rational acceptance requires only empirical adequacy — that the theory's observable predictions are true. If a theory correctly describes all observable phenomena but its posited unobservables (e.g., the ether) turn out not to exist, the theory was still acceptable in van Fraassen's sense. This is not a defect but the point: it allows science to be useful and rational even given the history of radical theory change.
Question 4 True / False
Constructive empiricism treats scientific theories as mere instruments for calculation, denying that their claims about unobservables have any truth value.
TTrue
FFalse
Answer: False
This describes instrumentalism, which van Fraassen explicitly distinguishes from constructive empiricism. For van Fraassen, a theory's claims about unobservables are genuinely true or false — they have full semantic content. He argues only that *rational belief* should track observability. The distinction between acceptance (using the theory) and belief (taking all its claims as true) is the heart of the position and marks it off from instrumentalism.
Question 5 Short Answer
What is the difference between 'accepting' a scientific theory and 'believing' it, according to van Fraassen? Why does this distinction matter?
Think about your answer, then reveal below.
Model answer: Accepting a theory means using it to guide action and form beliefs about observable phenomena — treating it as a reliable guide to what we can observe. Believing a theory means taking all its claims, including claims about unobservable entities, to be true. Van Fraassen argues these are rationally separable: we can accept theories as tools for navigating the observable world without committing to the existence of their unobservable posits. The distinction matters because it allows us to be rational about scientific success without over-extending our ontological commitments to things we can never directly observe.
Van Fraassen's analogy: a navigator who relies on a map for accurate coastlines while remaining agnostic about depth contours no one has measured accepts the map without fully believing it. This preserves the empiricist spirit — belief should be grounded in what is accessible to experience — even while scientific practice routinely extends beyond direct observation.