The Method of Science
The method of science, in Peirce's 1877 formulation, is distinguished from tenacity, authority, and the a priori method by a single feature: it is self-correcting. The scientist accepts a belief because predictions derived from the belief have been checked against observations, and she revises the belief when the predictions fail. Tenacity cannot detect its own errors because it refuses to consider alternatives. Authority cannot detect errors because it suppresses dissent. The a priori method cannot detect errors because it evaluates beliefs against cultural assumptions rather than evidence. Only the method of science has a built-in mechanism for error detection: the confrontation between belief and experience, the willingness to revise when the confrontation goes badly, and the community of inquirers whose collective scrutiny exposes errors individuals would miss.
In the AI Story
The method rests on four features that the Peirce volume identifies as essential: genuine confrontation with experience (beliefs tested against an independent reality capable of contradicting them), capacity for self-correction (revision driven by normative commitment to evidence), the irritation of doubt (the uncomfortable awareness that beliefs may be wrong), and commitment to truth as a normative ideal (valuing correspondence over comfort).
The method requires a community — the community of inquiry — whose collective scrutiny sustains the self-correcting process. No individual can maintain the irritation of doubt indefinitely against the seductive confidence of settled conviction. The community provides institutional friction: the critical questions, the alternative perspectives, the demand for evidence.
The Peirce volume argues that the method of computation — belief-fixation through fluent AI output — can substitute for the method of science in daily practice without anyone noticing the substitution, because the outputs can be superficially indistinguishable. The difference lies not in the surface but in the process. The scientific analysis is the product of a method that has subjected its claims to the test of experience; the AI output is the product of a method that has generated its claims from statistical patterns.
The method's superiority over its alternatives is not immediate or guaranteed. It is superior only in the long run, because only over time does the self-correcting process accumulate into genuine knowledge. In the short run, the method of science is slower, more uncomfortable, and less emotionally satisfying than the alternatives. Its defense requires commitment to the long run — to truth over convenience, testing over fluency, the sustained discomfort of genuine inquiry over the seductive comfort of confident output.
Origin
Peirce's classical statement appears in "The Fixation of Belief" (1877), developed further in the five subsequent essays of the Illustrations of the Logic of Science series.
The method as Peirce describes it is not a procedure but a normative ideal — the account of what science should be, against which actual scientific practice can be evaluated.
Key Ideas
Self-correcting. The only method with a built-in mechanism for detecting and revising its own errors.
Confrontation with experience. Beliefs tested against independent reality that can contradict them.
Requires community. The self-correcting process is collective; no individual can sustain it alone.
Long-run convergence. The method's superiority emerges over time through accumulated self-correction, not immediately in any particular inquiry.
Appears in the Orange Pill Cycle
Further reading
- Charles Sanders Peirce, "The Fixation of Belief" (1877)
- Charles Sanders Peirce, "How to Make Our Ideas Clear" (1878)
- Cheryl Misak, Truth and the End of Inquiry (Oxford, 1991)
- Nicholas Rescher, Peirce's Philosophy of Science (Notre Dame, 1978)
- Isaac Levi, Pragmatism as a Way of Life (Harvard, 2012)