Questions: Scientific Progress and Convergence to Truth

Question 1

The caloric fluid theory successfully predicted heat flow from hot to cold bodies and was the best available theory of heat for decades — then it was completely abandoned. Which position in the scientific realism debate treats this history as strong evidence against convergent realism?

• A. The no-miracles argument, which explains such predictive success by positing approximate truth
• B. Structural realism, which claims mathematical structure is preserved even when ontology changes
• C. The pessimistic meta-induction, which argues that past predictive success is no guarantee of truth-tracking
• D. Instrumentalism, which was actually the dominant view among caloric theorists

Answer: (C)

The pessimistic meta-induction (associated with Laudan) takes exactly this form: caloric fluid, phlogiston, luminiferous ether, and Newtonian absolute space were all the best available theories, made successful predictions, and were abandoned. If the track record of science is one of successful-then-abandoned theories, why believe our current best theories are any different? The no-miracles argument (option A) is the convergent realist response *to* this challenge, not the challenge itself.

Question 2

Newton's laws of motion are not exactly true — they break down at relativistic speeds. A structural realist uses this fact to argue which of the following?

• A. Newton's laws should be abandoned, confirming the pessimistic meta-induction
• B. Newton's laws were never approximately true and only had instrumental value
• C. The mathematical structure of Newton's laws is approximately preserved as a limiting case within general relativity, showing that structure survives even when ontology changes
• D. Scientific progress is impossible because every theory is eventually replaced

Answer: (C)

Structural realism holds that what science converges on is not the identity of fundamental entities (Newtonian absolute space, for example, was abandoned) but the mathematical relations among them. Newton's laws survive as a limiting case of general relativity (when v << c), and Fresnel's equations survive the wave-to-photon transition. The structural realist argues this preservation of structure is what the pessimistic meta-induction misses: the ontology changes, but the equations don't disappear — they are embedded in the successor theory.

Question 3

The no-miracles argument claims that the predictive success of our best theories would be miraculous if those theories were not approximately true.

• T. True
• F. False

Answer: True

This is the standard formulation of the no-miracles argument (associated with Putnam). The argument is: novel predictive successes — theories predicting phenomena they were not designed to account for, like Neptune's position or antiseptics' effectiveness — are most plausibly explained by those theories tracking something real. If the theories were radically false, such successes would be inexplicable coincidences, i.e., miracles. Convergent realism is thus presented as the best explanation of science's empirical success.

Question 4

The pessimistic meta-induction and the no-miracles argument are empirical disputes that can be resolved by gathering more data about scientific history.

• T. True
• F. False

Answer: False

Both arguments involve philosophical interpretation of the same historical data, not a gap in that data. Realists and anti-realists agree on the history: theories are eventually replaced, earlier theories had predictive success. They disagree about what this pattern implies about truth-tracking. The pessimistic meta-inductionist sees past abandonment as evidence against current theories being true; the realist sees novel prediction as evidence for approximate truth. This is a philosophical disagreement about inference and explanation, not a factual question that more historical data would settle.

Question 5

What is the structural realist's response to the pessimistic meta-induction, and what does it require us to say is and is not preserved across scientific revolutions?

Model answer: The structural realist concedes that the ontology of science — the specific entities posited (caloric fluid, ether, Newtonian absolute space) — is routinely abandoned across revolutions, as the pessimistic meta-induction claims. But they argue that the mathematical structure of successful theories is approximately preserved: Fresnel's equations for light survive the transition from wave to photon theories; Newton's laws survive as a limiting case of relativity. What science converges on, the structural realist claims, is not the nature of entities but the abstract relational structure of nature.

The key move is distinguishing ontological commitment (what kinds of things exist) from structural commitment (what the equations say about relations among measurable quantities). The structural realist argues that the pessimistic meta-induction correctly targets the former but misses convergence in the latter. Critics of structural realism challenge whether 'structure without objects' is coherent, and whether the structural preservation claim is actually borne out by the historical cases cited.