Source
Worrall (1989), LIT-315 (Dialectica 43(1–2), 99–124), read in full in NOTE-297.
What was actually shown
The setting (pp. 101–111). Two old arguments pull apart. The no-miracles argument says predictive success would be a miracle unless the theory were essentially correct; Worrall weakens it to “a prima facie plausibility argument” (p. 102). The argument from scientific revolutions says successful theories have been overturned at the theoretical level. Convergent realism (Boyd, Putnam) denies the history and fails on Fresnel’s ether, which was “entirely overthrown” (p. 116). Instrumentalism and Popperian conjectural realism give up no-miracles.
Maturity (pp. 113–115). A science is mature once it has a theory that predicted new types of phenomena not written into it. This removes some of Laudan’s counterexamples (Hartley’s and LeSage’s ethers). Fresnel’s ether remains the sharpest case.
The case (p. 119). Fresnel’s four equations for reflected and refracted amplitude ratios, tan(i−r)/tan(i+r) and its companions, are “directly and fully entailed by Maxwell’s theory” once the amplitudes are read as those of the electric vector. Fresnel was wrong about what vibrates and right that the phenomena “depend on something or other that undergoes periodic change at right angles to the light” (p. 120). The reader checked the four ratios against the standard Fresnel coefficients. They match under the modern reading of “polarised in the plane of incidence”, and Maxwell’s theory yields exactly these forms only for media of equal magnetic permeability, a condition the paper does not state.
The general pattern (pp. 120–121). Usually old equations reappear as limiting cases of new ones (Newton from general relativity) while the mechanism does not: there is “no clear sense” in which action at a distance approximates space-time curvature. So the correspondence principle is “not evidence for full-blown realism – but instead only for structural realism” (p. 121).
What could have come out otherwise. Fresnel’s equations could have been lost in Maxwell’s theory, or his ether kept. The general claim could fail case by case, but no survey was run.
What this does not say
- That the world is structure. The view is epistemic. It is “a mistake to think that we can ever ‘understand’ the nature of the basic furniture of the universe” (p. 122), and Poincaré’s “real objects which Nature will hide for ever” is quoted with approval. The ontic turn is Ladyman’s (LIT-342, LIT-217).
- Invariance under a symmetry group. “Structure” here is equations preserved across theory change, identically or in a limit. Group-theoretic structure does not appear (NOTE-297).
- Anything about Ramsey sentences or the Newman objection. Those belong to Worrall & Zahar 2001 and later, as the SEP entry (LIT-045) reports.
- That retention is the general pattern, shown. One detailed case, called unrepresentative by its author, and one example of limiting-case recovery. “The only available account” (p. 121) claims more than the elimination argument shows.
- What structural approximation is. “Much clarificatory work needs to be done … concerning the notion of one theory’s structure approximating that of another” (p. 121). Worrall’s own objection to approximate truth, that it is intransitive, is not shown to spare structural approximation.
- That empirical content always accumulates. That premise is asserted and defended elsewhere.
Connections
- French (THEORY-594) and De Haro and Butterfield (THEORY-690) concern ontic structural realism and its physics. This is the epistemic position those later debates start from.
- THEORY-034 places the ontic and informational structural realisms against Chakravartty’s dilemma. Worrall’s view is not of the kind that dilemma targets: it makes no claim about relata.
- THEORY-017 holds that in a Hilbert-space model only unitarily invariant structure is intrinsic and a basis is extra data. NOTE-297 calls the link an analogy: a basis plays the role Worrall gives the interpretation of the terms, but his structure is equations up to limits, not invariants of a group.