Science and Values
Science and Values: The Aims of Science and Their Role in Scientific Debate was published by the University of California Press in 1984. The book extended the framework of Progress and Its Problems by addressing a question the earlier work had left open: how do the aims of inquiry themselves evolve? Positivists had assumed aims were fixed; Kuhnians had suggested that aims changed with paradigms in ways that made inter-paradigmatic evaluation impossible. Laudan proposed a third option: aims, methods, and theories form a reticulated web in which each level constrains the others and each can be modified through coordinated revision of the whole. The book established the philosophical framework within which Laudan's subsequent work operated, and provided the resources this volume extends into the AI discourse.
In the AI Story
The book responded to critics who had argued that Progress and Its Problems assumed but did not defend the criteria by which problem-solving effectiveness itself was judged. If what counts as a problem depends on the tradition's aims, and the aims themselves are part of what the tradition can modify, then the problem-solving model seemed to face a regress. Science and Values addressed the regress by showing that the aims are not arbitrary — they are constrained by the theories and methods they coexist with, and they can be revised rationally through the same comparative procedure Laudan had developed for theories.
The book's central contribution is the reticulated model, which treats theories, methods, and aims as mutually constraining. A method is progressive when it serves the theory well given the aims; a theory is progressive when it is well-supported given the method and the aims; an aim is progressive when it is achievable given the methods and theories available. Changes at any level can drive revision at the others, and the cumulative effect is the rational development of the tradition.
Applied to AI, the book's framework clarifies why the discourse is so difficult to navigate. The transition is forcing modification at all three levels. The theories of creative work are changing (what is creativity, given that AI can produce outputs once considered creative?). The methods are changing (how is work done, given that AI can do much of it?). The aims are changing (what is work for, given that its traditional purposes — earning a living, demonstrating competence, building identity — are being destabilized?).
The reticulated model does not predict how the revision will resolve. It provides the structure within which rational revision is possible. The AI discourse will make progress to the extent that its participants are willing to hold some elements fixed long enough to evaluate others, and then to revise what was previously stable as the evaluation reveals new tensions. This is harder than it sounds, because the cultural pressure to choose a side works against the patience the reticulated model requires.
Origin
The book emerged from Laudan's responses to critics of Progress and Its Problems, including sustained engagement with realist and relativist alternatives. It was refined through seminars at the University of Pittsburgh and Virginia Polytechnic Institute, and represents his most mature statement of the framework he developed across five decades.
Key Ideas
Three coordinated levels. Theories, methods, and aims form the structure of rational inquiry.
No fixed foundation. No level is privileged as the unchanging basis for the others.
Aims evolve rationally. Aims are modified when they become incompatible with the methods and theories available.
Progress is cumulative. The web evolves through many small revisions, not through revolutionary replacement.
Appears in the Orange Pill Cycle
Further reading
- Larry Laudan, Science and Values (University of California Press, 1984).
- Larry Laudan, Beyond Positivism and Relativism (Westview, 1996).
- Ernan McMullin, "The Shaping of Scientific Rationality" (1988).