In 'Reconstructing Scientific Realism to Rebut the Pessimistic Meta-induction' (Philosophy of Science, 2007), Gerald Doppelt defends an enriched inference-to-the-best-explanation realism: the realist's explanandum (what is to be explained) should include theories' explanatory success and wide scientific acceptance, not just predictive success, and the explanans (what does the explaining) should include, besides truth, the theory's epistemic virtues (unification, simplicity, consilience, consistency, completeness, intuitive plausibility) and the historically evolving standards that govern when those virtues count as evidentially significant; realist truth-commitment is restricted to current theories that are uniquely 'most successful and well established', while the success of superseded theories is explained without attributing truth to them. The rebuttal of Laudan's pessimistic meta-induction that this reconstruction delivers remains incomplete: a new pessimistic meta-induction — from the historical supersession of past standards to the likely supersession of current standards — has yet to be separately stated and empirically evaluated; the virtue/standard apparatus lacks operational criteria; the empirical warrant of the standards is undefended; and the reconstruction has not been comparatively assessed against rival realist and antirealist replies. Completing — or defeating — this reconstruction remains an open problem: the subsequent literature — Doppelt's own refinements (2011, 2014), critiques by Alai and Díez, the proliferation of 'new' pessimistic inductions over methods and unconceived alternatives (Stanford, Wray, Rowbottom, Park), and empirical-historical programs (Mizrahi, Fahrbach, Schindler, Vickers) — has left each component actively contested and unresolved.

Background

Scientific realism is the view that our successful scientific theories are approximately true — that the unobservable entities they posit (electrons, fields, genes) exist roughly as described, and that it is this truth that explains why the theories work. The standard argument for realism is the 'no-miracles' argument, an inference to the best explanation (IBE, the mode of inference on which one infers the hypothesis that would best explain the available evidence): the approximate truth of successful theories best explains their empirical success, so without realism that success would be a miracle. The standard counterargument is the pessimistic meta-induction (PMI), pressed classically by Larry Laudan's 'A Confutation of Convergent Realism' (1981): the history of science contains many theories — phlogiston, caloric, the luminiferous ether — that were empirically successful by the standards of their day yet are now judged outright false; inductively, our present successful theories are probably false too, and explaining past success by truth cannot be right.

Realists have answered in several ways. Selective (or 'preservative') realism, associated with Stathis Psillos, holds that only those components of past theories that were actually responsible for their predictive success were true and were retained by successors. Structural realism holds that only the mathematical structure of successful theories is preserved across theory change and is true. Kyle Stanford's 'problem of unconceived alternatives' (2006) is a distinct historical induction: past scientists could not even conceive of the theories that later replaced theirs, so we too probably cannot conceive of the true successors of our theories.

Gerald Doppelt's 'Reconstructing Scientific Realism to Rebut the Pessimistic Meta-induction' (Philosophy of Science 74(1), 2007) proposes a different reconstruction. He argues (i) that the realist's explanandum — what is to be explained — must include theories' explanatory success — their possession of epistemic virtues such as unification, simplicity, consilience (the coherence of a theory's explanations across previously distinct domains), internal consistency, completeness, and intuitive plausibility — and their achievement of wide acceptance among scientists, not merely predictive success; (ii) that the realist's explanans — what does the explaining — must include, besides the theory's approximate truth, the theory's epistemic virtues and the historically evolving, discipline-specific standards that govern when those virtues count as evidentially significant; and (iii) that realist truth-commitments should be confined to current theories that are uniquely 'most successful and well established' — those that meet and raise the highest standards of empirical success in a mature field — while the success of superseded theories is explained without attributing truth to them, by the virtues those theories realized relative to the standards of their era together with the resources of our current best theories. Because our best current theories are distinguished in kind from past successful theories by the unprecedented standards they satisfy, Laudan's induction over the past cannot be projected onto them.

The rebuttal remains incomplete, however, and the open issues pressed in the subsequent debate define the present problem: the anticipated stronger form of the pessimistic induction — over superseded evaluative standards rather than superseded theories — has not been stated or evaluated; no operational criterion fixes which theories count as 'most successful and well established'; the claim that the operative standards are themselves empirically warranted has yet to be defended; and the reconstruction has not been compared with rival replies to the historical challenge. Subsequent work sharpened the stakes without closing them: Doppelt (2011, 2014) developed the account into 'Best Current Theory'/'Best Theory' realism, claiming it beats structural and preservative realism; Alai (2017, 2018) argued the position cannot explain the success of past theories or the failures of current ones; Díez (2018) raised a 'fatal trilemma' for it; and the surrounding literature multiplied and debated 'new' pessimistic inductions — over methods, unconceived alternatives, instruments, and values — while programs by Mizrahi, Fahrbach, Schindler, and Vickers began, without completing, the empirical and operational assessment the question demands.

Problem Statement

Complete — or definitively defeat — Doppelt's reconstructed inference-to-the-best-explanation realism as a rebuttal of the pessimistic meta-induction, by carrying out the evaluation the 2007 paper left undone. The problem has four linked components, all bearing on the overarching question of whether the enriched realism withstands the historical challenge once its open gaps are made good.

(1) The new pessimistic meta-induction over standards. State the anticipated stronger form of the argument precisely: past theories met the highest evaluative standards of their day, and those standards were later superseded by higher ones; inductively, our current standards will likewise be superseded, and our best current theories — which satisfy only current standards — will then be no better placed than their successful-but-false predecessors. Then evaluate this inductive hypothesis empirically against the historical record of theory change and standards change, confirming or refuting it independently of Laudan's original induction.

(2) Operationalization of the virtue/standard apparatus. Supply a precise, non-circular criterion — operationalizable measures of the epistemic virtues (unification, simplicity, consilience — the coherence of a theory's explanations across previously distinct domains — consistency, completeness, intuitive plausibility) and of the historically evolving standards governing when they count as evidentially significant — that determines which theories qualify as 'most successful and well established', so that the realist's truth-commitments have a determinate, contestation-resistant extension.

(3) Empirical warrant of the standards. Determine whether taking particular evaluative standards to be empirically warranted — as well-confirmed claims about what leads to truth — rather than merely methodological or regulative commitments, can be defended with evidence and argument; the evidential work for contested cases remains outstanding.

(4) Comparative assessment. Evaluate the completed reconstruction's strength against the rival replies to the historical challenge — selective/deployment realism, structural realism, Stanford's unconceived-alternatives position, and antirealist positions such as constructive empiricism (van Fraassen's view that science aims at empirical adequacy — saving the observable phenomena — rather than truth) — rather than only against Laudan's original argument.

The historical evidence and philosophical argument may show either that the enriched realism survives the new induction — supplying the operational criteria and warranted standards it currently lacks — or that it does not, with the component that cannot be made good identified. The four components are jointly required, and each admits either an affirmative or a negative resolution — in particular, (2) may be resolved by supplying the criterion or by establishing that no such criterion can be made good, and (4) by showing either that the completed reconstruction outperforms its rivals or that it does not.

References

  1. Doppelt, G. (2007). Reconstructing Scientific Realism to Rebut the Pessimistic Meta-induction. Philosophy of Science 74(1): 96–118. DOI 10.1086/520685. https://doi.org/10.1086/520685
  2. Laudan, L. (1981). A Confutation of Convergent Realism. Philosophy of Science 48(1): 19–49. DOI 10.1086/288975. https://doi.org/10.1086/288975
  3. Psillos, S. (1996). Scientific Realism and the 'Pessimistic Induction'. Philosophy of Science 63(S3): S306–S314. DOI 10.1086/289965. https://doi.org/10.1086/289965
  4. Stanford, P. K. (2006). Exceeding Our Grasp: Science, History, and the Problem of Unconceived Alternatives. Oxford University Press. DOI 10.1093/0195174089.001.0001. https://doi.org/10.1093/0195174089.001.0001
  5. Doppelt, G. (2011). From Standard Scientific Realism and Structural Realism to Best Current Theory Realism. Journal for General Philosophy of Science 42(2): 295–316. DOI 10.1007/s10838-011-9167-8. https://doi.org/10.1007/s10838-011-9167-8
  6. Doppelt, G. (2014). Best theory scientific realism. European Journal for Philosophy of Science 4(2): 271–291. DOI 10.1007/s13194-014-0090-9. https://doi.org/10.1007/s13194-014-0090-9
  7. Alai, M. (2017). Resisting the historical objections to realism: Is Doppelt's a viable solution? Synthese 194(9): 3267–3290. DOI 10.1007/s11229-016-1087-z. https://doi.org/10.1007/s11229-016-1087-z
  8. Díez, J. (2018). A (Fatal) Trilemma for best theory realism. European Journal for Philosophy of Science 8(2): 271–291. DOI 10.1007/s13194-017-0185-1. https://doi.org/10.1007/s13194-017-0185-1
  9. Alai, M. (2018). How Deployment Realism withstands Doppelt's Criticisms. Spontaneous Generations 9(1): 122–135. DOI 10.4245/sponge.v9i1.27046. https://doi.org/10.4245/sponge.v9i1.27046
  10. Alai, M. (2021). The Historical Challenge to Realism and Essential Deployment. In Contemporary Scientific Realism, pp. 183–215. Oxford University Press. DOI 10.1093/oso/9780190946814.003.0009. https://doi.org/10.1093/oso/9780190946814.003.0009
  11. Ruhmkorff, S. (2013). Global and Local Pessimistic Meta-inductions. International Studies in the Philosophy of Science 27(4): 409–428. DOI 10.1080/02698595.2013.868185. https://doi.org/10.1080/02698595.2013.868185
  12. Wray, K. B. (2015). Pessimistic Inductions: Four Varieties. International Studies in the Philosophy of Science 29(1): 61–73. DOI 10.1080/02698595.2015.1071551. https://doi.org/10.1080/02698595.2015.1071551
  13. Mizrahi, M. (2015). Historical Inductions: New Cherries, Same Old Cherry-picking. International Studies in the Philosophy of Science 29(2): 129–148. DOI 10.1080/02698595.2015.1119413. https://doi.org/10.1080/02698595.2015.1119413
  14. Mizrahi, M. (2016). Historical Inductions, Unconceived Alternatives, and Unconceived Objections. Journal for General Philosophy of Science 47(1): 59–68. DOI 10.1007/s10838-015-9295-7. https://doi.org/10.1007/s10838-015-9295-7
  15. Fahrbach, L. (2017). Scientific revolutions and the explosion of scientific evidence. Synthese 194(12): 5039–5072. DOI 10.1007/s11229-016-1193-y. https://doi.org/10.1007/s11229-016-1193-y
  16. Rowbottom, D. P. (2019). Extending the argument from unconceived alternatives: observations, models, predictions, explanations, methods, instruments, experiments, and values. Synthese 196(10): 3947–3959. DOI 10.1007/s11229-016-1132-y. https://doi.org/10.1007/s11229-016-1132-y
  17. Park, S. (2018). Justifying the Special Theory of Relativity with Unconceived Methods. Axiomathes 28(1): 53–62. DOI 10.1007/s10516-017-9336-4. https://doi.org/10.1007/s10516-017-9336-4
  18. Park, S. (2017). Why Should We Be Pessimistic About Antirealists and Pessimists? Foundations of Science 22(3): 613–625. DOI 10.1007/s10699-016-9490-y. https://doi.org/10.1007/s10699-016-9490-y
  19. Park, S. (2022). Critiques of the Pessimistic Induction. In S. Park (Ed.), Embracing Scientific Realism (Synthese Library, vol. 445, pp. 23–43). Springer. DOI 10.1007/978-3-030-87813-9_2. https://doi.org/10.1007/978-3-030-87813-9_2
  20. Schindler, S. (2018). Theoretical Virtues in Science: Uncovering Reality Through Theory. Cambridge University Press. DOI 10.1017/9781108381352. https://doi.org/10.1017/9781108381352
  21. Vickers, P. (2022). Identifying Future-Proof Science. Oxford University Press. DOI 10.1093/oso/9780192862730.001.0001. https://doi.org/10.1093/oso/9780192862730.001.0001
  22. Lyons, T. D. (2015). Scientific Realism. In The Oxford Handbook of Philosophy of Science (Oxford Handbooks Online). DOI 10.1093/oxfordhb/9780199368815.013.30. https://doi.org/10.1093/oxfordhb/9780199368815.013.30
  23. Cronin, P. (2025). An A Priori Refutation of the Classical Pessimistic Induction. Erkenntnis 90(7): 3049–3069. DOI 10.1007/s10670-024-00832-5. https://doi.org/10.1007/s10670-024-00832-5