Alan Baker's "Parsimony and inference to the best mathematical explanation" (Synthese, 2016) argues that in optimization-style mathematical explanations in science, strengthening the mathematical apparatus can reduce the concrete ontological, ideological, and structural commitments an explanation must carry, posing a challenge to nominalism. The paper itself leaves open how widespread this strength–parsimony trade-off is across the sciences; whether its two flagship cases (periodical cicadas, the honeycomb theorem) are genuinely improved by the stronger mathematics, given unresolved empirical questions about predator periodicity and bee-comb construction; and whether nominalistic paraphrases can in practice reproduce the stronger mathematics — both its statements and its justifying proofs — on which such explanations rely. It argues that parsimony with respect to concreta should be preferred in such theory choice, while how mathematical commitments are to be counted and weighed against concrete commitments remains unresolved.

Background

In the philosophy of mathematics, platonism holds that mathematical objects (numbers, sets) exist abstractly, while nominalism denies this and insists that serious theories should avoid commitment to abstract objects. Indispensability arguments for platonism, descending from Quine and Putnam, hold that if mathematics is an indispensable part of our best scientific theories, then we are committed to the objects it quantifies over. The enhanced (or explanatory) indispensability argument (EIA) sharpens this by appeal to inference to the best explanation (IBE), the mode of reasoning whereby we accept the posits of the theory that best explains the evidence: it argues that some scientific explanations are genuinely mathematical — mathematics does explanatory work that cannot be paraphrased away — so realism about science commits us to mathematical objects. A mathematical explanation in science (MES) is an explanation of an empirical phenomenon in which a mathematical fact carries an indispensable part of the explanatory burden; the much-discussed example from Baker (2005) explains why periodical cicadas of the genus Magicicada have 13- and 17-year life cycles by their primeness minimizing temporal overlap with predators that have shorter, periodic life cycles. An optimization explanation is a subtype in which a phenomenon is presented as the solution of an optimization problem: a target feature to be optimized, a set of constraints, and a class of candidate options. Following standard usage, a theory's ontological commitments are the objects its quantifiers range over; its ideological commitments are the primitive concepts it must deploy; its structural commitments are the structural assumptions (e.g. restrictions on admissible shapes or configurations) it builds in. Baker (2016) adds a twist to the EIA debate: in optimization MESs, a mathematically stronger theorem can apply under weaker physical and structural assumptions, thereby reducing the concrete commitments — unobserved entities, ideologically loaded predicates, or unjustified structural restrictions — that the explanation requires. In the cicada case, the standard number-theoretic lemma (a prime is coprime to, and so maximizes the least common multiple with, every smaller integer) makes primeness uniquely optimal only if predators exist at every shorter period, an implausibly rich inventory of unobserved predator species; Baker's strengthened route combines a biconditional optimality criterion with prime-existence results for multiplicative intervals — Bertrand's postulate guarantees a prime between nn and 2n2n, and El Bachraoui (2006) and Loo (2011) extend this to the intervals [2n,3n][2n,3n] and [3n,4n][3n,4n] — so that optimality follows from a much smaller predator commitment (an upper bound of roughly 3 on predator periods). In the honeycomb case, the classical result attributed to Pappus — the regular hexagon minimizes perimeter among the regular polygons of equal area — requires an auxiliary structural assumption that bees build only regular polygons, whereas the full honeycomb theorem proved by Hales (2001) — any partition of the plane into regions of equal area has perimeter at least that of the regular hexagonal honeycomb tiling — dispenses with that assumption at the cost of markedly heavier mathematics: the published proof is carried out with geometric-measure-theoretic machinery (Hausdorff measure, rectifiable boundaries) and delicate explicit estimates, rather than an elementary comparison of regular polygons. Nominalists have tried to accommodate mathematics-laden science by nominalistic paraphrase: reconstructing the relevant claims, and in stronger programs their justifying proofs, without quantification over abstract objects; Field's program of nominalizing a concrete scientific theory without mathematics and piecemeal case-by-case paraphrases are the main precedents. Baker's paper presents the two cases, argues for a default preference for fewer concrete posits, and raises the issues recorded here, which neither the paper nor the subsequent literature settles. Related work before and after (Baron 2014 on Lévy-walk optimization explanations — Lévy walks being random search patterns with power-law-distributed step lengths, used as models of animal foraging; Saatsi 2016 on the explanatory role of mathematics) frames how widespread genuinely mathematical explanations in science are, which bears directly on the first open issue.

Problem Statement

Determine the scope, standing, and normative weight of the trade-off, posed by Baker (2016), in which strengthening the mathematical apparatus of an optimization-style mathematical explanation in science reduces the concrete ontological, ideological, and structural commitments the explanation must carry. The common objective holding the problem together is deciding whether this strength–parsimony trade-off obtains widely enough, and carries enough probative and normative force, to sustain Baker's parsimony-based challenge to nominalism. Four components of this question are at issue, and remain contested in the later literature: (1) Extent: determine whether optimization explanations exhibiting the trade-off form a broader class (or classes) of mathematical explanations in science widespread beyond the paper's two illustrative cases — the line of response flagged in the source's endnote (citing Baron 2014) — deciding whether strengthening mathematics substantially reduces concrete commitments generally, occasionally, or rarely. (2) Case standing: for the honeycomb explanation, determine whether independent biological or developmental factors actually justify the structural constraint to regular polygons that the weaker optimality result needs, or whether the empirical facts undercut or displace the parsimony argument for the stronger theorem (Hales's result over all plane partitions); for the cicada explanation, determine whether invoking the strengthened number theory (results guaranteeing primes in intervals such as [3n,4n][3n,4n] — Loo 2011, following El Bachraoui 2006 for [2n,3n][2n,3n] — which reduce the required predator-period commitments) genuinely improves the explanation, or instead replaces one contentious commitment with another — which includes settling the empirical premise about which predator periods must be assumed and the explanatory comparison between weaker and stronger formulations. (3) Nominalist replicability: determine whether nominalistic paraphrases can, in practice, be extended to the stronger mathematical apparatus involved — the geometric-measure-theoretic proof of the honeycomb theorem and the strengthened number-theoretic results — in a way that reproduces both the statements and the justifying proofs that scientists rely upon, and at what cost, or else demonstrate principled limits of such reconstruction. (4) Normative criteria: determine whether parsimony with respect to concreta should be preferred to parsimony with respect to abstracta in such theory choice, and specify and defend criteria for how mathematical commitments are to be counted and weighed against concrete ones, including whether the default preference for parsimony of concrete posits is warranted. Relative language ('substantially reduces', 'widespread', 'improves the explanation') is used as in the source and the surrounding literature. The four components jointly constitute this one research question.

References

  1. Baker, A. (2016). Parsimony and inference to the best mathematical explanation. Synthese 193(2), 333–350. DOI: 10.1007/s11229-015-0723-3. https://doi.org/10.1007/s11229-015-0723-3
  2. Pataut, F. (2016). Comments on "Parsimony and inference to the best mathematical explanation". Synthese 193(2), 351–363. DOI: 10.1007/s11229-015-0706-4. https://doi.org/10.1007/s11229-015-0706-4
  3. Baker, A. (2005). Are there Genuine Mathematical Explanations of Physical Phenomena? Mind 114(454), 223–238. DOI: 10.1093/mind/fzi223. https://doi.org/10.1093/mind/fzi223
  4. Baron, S. (2014). Optimisation and mathematical explanation: doing the Lévy Walk. Synthese 191(3), 459–479. DOI: 10.1007/s11229-013-0284-2. https://doi.org/10.1007/s11229-013-0284-2
  5. Hales, T. C. (2001). The Honeycomb Conjecture. Discrete & Computational Geometry 25(1), 1–22. DOI: 10.1007/s004540010071. https://doi.org/10.1007/s004540010071 (preprint: arXiv:math/9906042, https://arxiv.org/abs/math/9906042)
  6. Saatsi, J. (2016). On the 'Indispensable Explanatory Role' of Mathematics. Mind 125(500), 1045–1070. DOI: 10.1093/mind/fzv175. https://doi.org/10.1093/mind/fzv175
  7. Räz, T. (2017). The silent hexagon: explaining comb structures. Synthese 194(5), 1703–1724 (online 2016). DOI: 10.1007/s11229-016-1014-3. https://doi.org/10.1007/s11229-016-1014-3
  8. Wakil, S., & Justus, J. (2017). Mathematical Explanation and the Biological Optimality Fallacy. Philosophy of Science 84(5), 916–930. DOI: 10.1086/694154. https://doi.org/10.1086/694154
  9. Barrantes, M. (2019). Optimal representations and the Enhanced Indispensability Argument. Synthese 196(1), 247–263 (online 2017). DOI: 10.1007/s11229-017-1470-4. https://doi.org/10.1007/s11229-017-1470-4
  10. Dieveney, P. (2021). Scientific explanation, unifying mathematics, and indispensability arguments (A new Cicada MES). Synthese 198(1), 57–77 (online 2018). DOI: 10.1007/s11229-018-01979-9. https://doi.org/10.1007/s11229-018-01979-9
  11. Molinini, D. (2018). Parsimony, Ontological Commitment and the Import of Mathematics. In Truth, Existence and Explanation (Boston Studies in the Philosophy and History of Science, Springer), 189–207. ISBN 978-3-319-93341-2. DOI: 10.1007/978-3-319-93342-9_11. https://doi.org/10.1007/978-3-319-93342-9_11
  12. Boyce, K. (2021). Why inference to the best explanation doesn't secure empirical grounds for mathematical platonism. Synthese 198(1), 583–595 (online 2018). DOI: 10.1007/s11229-018-02043-2. https://doi.org/10.1007/s11229-018-02043-2
  13. Heylen, J., & Tump, L. A. (2021). The Enhanced Indispensability Argument, the circularity problem, and the interpretability strategy. Synthese 198(4), 3033–3045 (online 2019). DOI: 10.1007/s11229-019-02263-0. https://doi.org/10.1007/s11229-019-02263-0
  14. Leng, M. (2021). Models, structures, and the explanatory role of mathematics in empirical science. Synthese 199(3–4), 10415–10440. DOI: 10.1007/s11229-021-03253-x. https://doi.org/10.1007/s11229-021-03253-x
  15. Kuorikoski, J. (2021). There Are No Mathematical Explanations. Philosophy of Science 88(2), 189–212. DOI: 10.1086/711479. https://doi.org/10.1086/711479
  16. Baker, A. (2021). Circularity, indispensability, and mathematical explanation in science. Studies in History and Philosophy of Science 88, 156–163. DOI: 10.1016/j.shpsa.2021.05.017. https://doi.org/10.1016/j.shpsa.2021.05.017
  17. Paseau, A. C., & Baker, A. (2023). Indispensability (Cambridge Elements in the Philosophy of Mathematics). Cambridge University Press. DOI: 10.1017/9781009099042. https://doi.org/10.1017/9781009099042
  18. Himelright, J. (2025). Reassessing the Explanatory Indispensability Argument: A Bayesian Defense of Nominalism. The British Journal for the Philosophy of Science, article 737197. DOI: 10.1086/737197. https://doi.org/10.1086/737197
  19. Fraser, R. (2026). The problem of extra-mathematical explanation. Synthese 207(5), article 193. DOI: 10.1007/s11229-026-05560-7. https://doi.org/10.1007/s11229-026-05560-7
  20. Nazzi, F. (2016). The hexagonal shape of the honeycomb cells depends on the construction behavior of bees. Scientific Reports 6, article 28341. DOI: 10.1038/srep28341. https://doi.org/10.1038/srep28341
  21. Cardoso, S. S. S., Cartwright, J. H. E., Checa, A. G., Escribano, B., Osuna-Mascaró, A. J., & Sainz-Díaz, C. I. (2020). The bee Tetragonula builds its comb like a crystal. Journal of the Royal Society Interface 17(168), 20200187. DOI: 10.1098/rsif.2020.0187. https://doi.org/10.1098/rsif.2020.0187
  22. Narumi, T., Ogawa, R., & Osaki, K. (2026). Honeycomb construction on designed foundations reveals geometry-dependent developmental pathways. Scientific Reports 16, article 25446. DOI: 10.1038/s41598-026-55752-x. https://doi.org/10.1038/s41598-026-55752-x
  23. Toivonen, J., & Fromhage, L. (2020). Hybridization selects for prime-numbered life cycles in Magicicada: An individual-based simulation model of a structured periodical cicada population. Ecology and Evolution 10(12), 5259–5269. DOI: 10.1002/ece3.6270. https://doi.org/10.1002/ece3.6270
  24. Ito, H., Kakishima, S., Uehara, T., Morita, S., Koyama, T., Sota, T., Cooley, J. R., & Yoshimura, J. (2015). Evolution of periodicity in periodical cicadas. Scientific Reports 5, article 14094. DOI: 10.1038/srep14094. https://doi.org/10.1038/srep14094
  25. El Bachraoui, M. (2006). Primes in the interval [2n,3n]. International Journal of Contemporary Mathematical Sciences 1(13), 617–621. DOI: 10.12988/ijcms.2006.06065. https://doi.org/10.12988/ijcms.2006.06065
  26. Loo, A. (2011). On the primes in the interval [3n,4n]. International Journal of Contemporary Mathematical Sciences 6, no. 38, 1871–1882. No Crossref DOI. https://www.m-hikari.com/ijcms-2011/37-40-2011/looIJCMS37-40-2011.pdf
  27. Tump, L. A. (2025). On Mathematical Explanations of Empirical Phenomena: Indispensability, Number Theory, and Mathematical Counterfactual Dependence (Synthese Library 517). Springer. DOI: 10.1007/978-3-031-98860-8. https://doi.org/10.1007/978-3-031-98860-8
  28. Pincock, C. (2012). Mathematics and Scientific Representation. Oxford University Press. DOI: 10.1093/acprof:oso/9780199757107.001.0001. https://doi.org/10.1093/acprof:oso/9780199757107.001.0001
  29. Field, H. (2016). Science without Numbers: A Defense of Nominalism (2nd ed.). Oxford University Press. DOI: 10.1093/acprof:oso/9780198777915.001.0001. https://doi.org/10.1093/acprof:oso/9780198777915.001.0001
  30. Colyvan, M. (2023). Indispensability Arguments in the Philosophy of Mathematics. In E. N. Zalta & U. Nodelman (Eds.), The Stanford Encyclopedia of Philosophy (substantive revision of 6 March 2023). https://plato.stanford.edu/entries/mathphil-indis/