Key Principle
The empirical basis of science rests not on infallible observations but on conventionally accepted decisions to accept or reject basic statements. A basic statement is a singular existential statement ("There is a so-and-so in region k") that is inter-subjectively testable and involves macroscopic physical bodies.
This resolves Fries's trilemma: justifying empirical statements faces three traps — dogmatism (accepting without justification), infinite regress (every justification needs justification), or psychologism (grounding in subjective experience). Popper's solution: basic statements are indeed "dogmas," but innocuous ones, since any basic statement can itself be retested. The regress is halted by decision, not by reaching bedrock.
Why This Matters
If the empirical basis were infallible, science would have a secure foundation but would also be dogmatic — no observation could ever be questioned. If the basis were merely subjective experience, science would collapse into psychologism — your experience vs. mine, with no resolution. Popper's conventional basis preserves the revisability that science requires while maintaining enough stability for testing to proceed.
This extends Popper's fallibilism to its deepest level: not only are theories tentative guesses, but even the observations that test them are provisional agreements. Science has no foundations — only an endlessly revisable structure of conjectures and tests.
Good Examples
The table-thumping analogy: "Experiences can motivate a decision, and hence an acceptance or a rejection of a statement, but a basic statement cannot be justified by them — no more than by thumping the table" (Chapter 5). Perceptual experiences are psychologically relevant but logically irrelevant to justification.
The verdict analogy: Accepting a basic statement is like a jury's verdict. The jury reaches a decision by applying rules of procedure (not by direct access to truth). The verdict can be challenged and overturned. Similarly, scientists agree to accept a basic statement provisionally, following methodological rules, and can revise it if further tests warrant (Chapter 5).
Theory-ladenness of observation: "Theory dominates the experimental work from its initial planning up to the finishing touches in the laboratory" (Chapter 3). There is no pure observation — all testing is theory-directed. This is not a weakness but a feature: it makes testing severe rather than aimless.
Counterpoints
Protocol sentences (Neurath/Carnap): The Vienna Circle proposed "protocol sentences" — records of immediate observations — as science's foundation. Popper rejects these as psychologism in formal dress: they smuggle subjective experience into logic by disguising it as linguistic convention (Chapter 5).
Neurath's ship metaphor: Neurath argued we can revise any statement, even basic ones, like repairing a ship at sea. Popper agrees with the revisability but warns that Neurath's version — where any observation can be freely deleted — "throws empiricism overboard." Some discipline is needed: basic statements must be inter-subjectively testable and accepted only after attempts at refutation (Chapter 5).
Foundationalism: Any claim that observations are certain or self-justifying reintroduces the psychologism Popper aims to eliminate. Even the most confident observation report is a hypothesis about what happened, subject to error and revision.
Key Quotes
"Experiences can motivate a decision, and hence an acceptance or a rejection of a statement, but a basic statement cannot be justified by them — no more than by thumping the table." — Karl Popper, Chapter 5
"The empirical basis of objective science has thus nothing 'absolute' about it. Science does not rest upon solid bedrock." — Karl Popper, Chapter 5
Rules of Thumb
- Treat every observation report as a testable hypothesis, not as a self-evident truth
- When observations conflict with a well-tested theory, test the observations before abandoning the theory
- Inter-subjective reproducibility, not subjective conviction, is the mark of a legitimate basic statement
- The decision to accept a basic statement should itself be subject to review
Related References
- Falsifiability and the Logic of Deductive Testing — The deductive testing method that basic statements serve
- Conventionalist Stratagems and Methodological Rules — Where conventional decisions belong (basic statements) vs. where they don't (shielding theories)
- Corroboration — How a Theory Stands Up to Tests — How accepted basic statements feed into the appraisal of theories