Definition
Scientific realism is the position that the physical world exists independently of our minds and is accessible to rational inquiry. Our best theories describe real entities — including unobservable ones — and successive theories get progressively closer to describing reality as it is.
Why It Matters
Scientific realism is the ‘stake in the ground’ for objective truth; it asserts that science is not just a useful story but a progressively more accurate map of a real, mind-independent world.
Core Concepts
- The core claim: There is a mind-independent physical reality, and science is our best method for discovering its structure. Claims about quarks, gravitational waves, and parallel universes are claims about real things.
- The methodological criterion: If an entity is indispensable to our best explanation of observed phenomena, it is real. This is how Deutsch argues for the reality of the multiverse — it is required by the best explanation of quantum interference.
- Telescopes and reality: A common objection is that more instrumentation distances us from direct reality. Deutsch inverts this: telescopes reveal more of reality by adding more layers of corrective theory. They don’t bypass reality — they give us better access to it by correcting the distortions that limit naked-eye observation.
- Unobservables are real: Quarks have never been directly observed, yet they are real because they figure essentially in our best explanations. The same argument applies to other galaxies, the interior of black holes, and parallel universes.
- Against instrumentalism: The instrumentalist says theories are just tools. The realist says tools work because they track real structures. The explanation of why prediction succeeds requires realism.
- Against positivism: Positivism restricts meaningful claims to those directly verifiable by observation. Realism rejects this restriction — the whole point of science is to go beyond the immediately observable.
- Fallibilist realism: Our current best theories may be wrong — realism is compatible with fallibilism. We don’t claim our theories are certainly true, only that they aim at and partially describe a mind-independent reality.