A constructive empiricist argues that all we need to demand of a successful theory is empirical adequacy — truth about observables. How does the no-miracles argument respond?
AIt concedes that empirical adequacy is sufficient and drops the claim of approximate truth
BIt argues that if theories were only empirically adequate fictions with no connection to unobservable reality, their novel predictive successes — far outside their original domain — would be an unexplained miracle
CIt shows that empirical adequacy and approximate truth are logically equivalent
DIt appeals to Occam's razor: truth is simpler than empirical adequacy
The no-miracles argument targets exactly this gap between empirical adequacy and truth. Quantum electrodynamics predicts the electron's magnetic moment to eleven decimal places in domains far removed from the data it was built to fit. If the theory were just a useful fiction with no real connection to unobservable structure, this unprecedented novel success would be an extraordinary coincidence — a miracle. The best explanation is that the theory is approximately true, including its claims about unobservables. The anti-realist cannot explain away the pattern of novel success without invoking luck on a cosmic scale.
Question 2 Multiple Choice
The pessimistic meta-induction challenges the no-miracles argument primarily by pointing out that:
AScientific theories are too mathematically complex to be literally true
BInference to the best explanation is not a valid logical rule
CHistorically successful theories — caloric, phlogiston, Newtonian mechanics — later turned out to be false, so by induction our current successful theories are probably false too
DThe no-miracles argument only applies to physics, not biology or chemistry
The pessimistic meta-induction uses the historical record as evidence against realism. Caloric theory predicted heat flow, phlogiston organized combustion, Newtonian mechanics predicted orbits with extraordinary precision — all were predictively successful and all were later shown to be approximately false in their core claims. If success in the past didn't track truth, the NMA's inference from success to truth is undermined for our current theories too. Structural realists respond by arguing that mathematical structure is preserved across theory change, even when the ontology changes.
Question 3 True / False
The no-miracles argument is a deductive proof that our best scientific theories are true.
TTrue
FFalse
Answer: False
False — the NMA is an abductive argument (inference to the best explanation). It does not deductively establish truth; it argues that truth is the *best explanation* of predictive success. Abductive arguments can be rational and compelling without being deductively valid. The conclusion ('theories are approximately true') could be false even if the premises are true — alternative explanations (like van Fraassen's empirical adequacy account) are logically possible. The NMA's strength lies in its explanatory power, not logical entailment.
Question 4 True / False
Structural realism preserves the core intuition of the no-miracles argument while conceding ground to the pessimistic meta-induction.
TTrue
FFalse
Answer: True
True. Structural realism holds that what successful theories likely get right is not the nature of the entities (electrons, caloric, ether) but the mathematical structure of the relationships between them. This explains why Fresnel's optical equations survived into Maxwell's electromagnetism even as the ether was abandoned: the structure was preserved. This concedes to the pessimist that ontological claims are often overturned, while preserving the no-miracles intuition that something in the theory — specifically its structural content — is approximately true and explains success.
Question 5 Short Answer
Explain the logical structure of the no-miracles argument and identify its main vulnerability.
Think about your answer, then reveal below.
Model answer: The NMA is an inference to the best explanation: (1) our best theories make precise novel predictions that succeed far outside the conditions in which they were formulated; (2) the best explanation of this success is that the theories are approximately true, including their claims about unobservables; therefore (3) theories are approximately true. Its main vulnerability is twofold: first, it is abductive (not deductive), so the conclusion can be denied without contradiction if one accepts an alternative explanation like empirical adequacy. Second, the pessimistic meta-induction shows that historically successful theories regularly turned out false, weakening the inference from success to truth.
The NMA also faces a circularity worry: it employs inference to the best explanation to justify scientific methods, but IBE is itself a scientific method. Whether IBE reliably tracks truth for unobservables is exactly what's in dispute between realists and anti-realists. Realists must either accept this circularity or argue for IBE's reliability on independent grounds.