Scientific Realism and Antirealism
The realism debate asks whether our best scientific theories should be believed as true descriptions of an observer-independent world, including its unobservable entities.
Definition
Scientific realism is the view that mature, predictively successful scientific theories are approximately true and that the unobservable entities they posit really exist; antirealism denies one or more of these claims, restricting warranted belief to observable phenomena or to theory structure.
Scope
This area covers scientific realism and its main antirealist rivals: constructive empiricism, instrumentalism, and structural realism. It examines the principal arguments — the no-miracles argument for realism and the pessimistic meta-induction and underdetermination arguments against it — together with debates over the epistemic significance of unobservables.
Sub-topics
Core questions
- Should we believe that successful theories are true, or only empirically adequate?
- Do the unobservable entities posited by science really exist?
- Does the success of science require a realist explanation?
- Does the history of discarded-but-successful theories undermine realism?
Key concepts
- approximate truth
- empirical adequacy
- observable/unobservable distinction
- reference
- no-miracles argument
- pessimistic meta-induction
- structural realism
Key theories
- Scientific realism
- Mature scientific theories are approximately true and their central theoretical terms genuinely refer, so we are warranted in believing in unobservables.
- Constructive empiricism
- van Fraassen holds that the aim of science is empirical adequacy, and acceptance of a theory commits one only to belief in what it says about observables.
- Structural realism
- Worrall proposes that what is retained across theory change is mathematical structure, so realism should be confined to the structural content of theories.
History
After the decline of logical positivism, Putnam and others revived realism in the 1970s, grounding it in the no-miracles argument. van Fraassen's 1980 constructive empiricism offered a sophisticated empiricist alternative, while Laudan's 1981 pessimistic meta-induction challenged the realist's success-to-truth inference, prompting structural and selective forms of realism.
Debates
- Truth versus empirical adequacy
- Realists hold that we should believe successful theories true, while van Fraassen argues we need only believe them empirically adequate, suspending judgement about unobservables.
- The threat from theory change
- Laudan argues that many past theories were successful yet false, undermining the inference from success to truth; realists respond by restricting the claim to the parts of theories responsible for success.
Key figures
- Bas van Fraassen
- Hilary Putnam
- Larry Laudan
- Stathis Psillos
- John Worrall
Related topics
Seminal works
- vanfraassen1980
- putnam1975
- laudan1981
- worrall1989
Frequently asked questions
- What is the core disagreement between realists and constructive empiricists?
- Both accept that science aims at and achieves empirical success, but realists infer that successful theories are (approximately) true descriptions of unobservable reality, whereas constructive empiricists hold that we are entitled to believe only that such theories are empirically adequate.