Supertasks and arithmetical truth


This paper discusses the relevance of supertask computation for the determinacy of arithmetic. Recent work in the philosophy of physics has made plausible the possibility of supertask computers, capable of running through infinitely many individual computations in a finite time. A natural thought is that, if true, this implies that arithmetical truth is determinate (at least for e.g. sentences saying that every number has a certain decidable property). In this paper we argue, via a careful analysis of putative arguments from supertask computations to determinacy, that this natural thought is mistaken: supertasks are of no help in explaining arithmetical determinacy.

This is a preview of subscription content, log in to check access.


  1. 1.

    See Field (1994) and Putnam (1980).

  2. 2.

    See the essays in Salmon (1970) for a collection of classic articles on supertasks.

  3. 3.

    Quoted from Benacerraf and Putnam (1983), p. 20. Weyl (1949) also hints at use of supertasks in arithmetic, though negatively.

  4. 4.

    See Davies (2001) and Earman and Norton (1993, 1996).

  5. 5.

    See Hogarth (2004); see also Button (2009) for discussion of how such generalizations relate to the Church–Turing thesis.

  6. 6.

    As a referee has pointed out, our current practice arguably does include dispositions to (a) “take certain combinations of physical data as evidence for the existence of a supercomputer” and (b) “form new mathematical beliefs on the basis of interactions with that apparent supercomputer”. We don’t deny that in such a scenario, mathematicians would likely take the results of the supertask computations as evidence for arithmetical claims. But although they might play such an epistemic role, our point here is that deference to a supertask computer is not part of the basic metasemantics of actual arithmetic. That is to say: the content of our actual arithmetical claims is surely not fixed by the outcome of any empirical process, let alone (merely possible) supertask computations.

  7. 7.

    See Weyl (1949), p. 42, Bishop (1967), and Dummett (1977), for example.

  8. 8.

    Earman and Norton (1996), p. 251.

  9. 9.

    This is because \({\mathsf {PA}}\) (and even weaker arithmetical theories such as Robinson Arithmetic) are \(\Sigma _{1}\)-complete.

  10. 10.

    Obviously, for this particular case we could work with a restricted version of the rule requiring that \(\varphi \) contain no unbounded quantifiers.

  11. 11.

    For a proof of this that doesn’t require indexing proofs with infinite ordinals, see Tennant (1978). It is worth noting that the result also holds with weak arithmetical theories, such as Robinson arithmetic, in place of \({\mathsf {PA}}\).

  12. 12.

    See Kreisel (1967) for relevant discussion.

  13. 13.

    Typically by way of a second-order categoricity theorem; see Shapiro (1991) for details. An analogous result is available in the open-ended, first-order setting; for this approach, see Lavine (n.d.), McGee (1997), and Parsons (2008), chapter 8.

  14. 14.

    We use this notion rather than categoricity, which is stronger. True arithmetic has nonstandard models, for example.

  15. 15.

    Our discussion concerns only attempts to exploit the computational power of supertask computers to establish determinacy. But, as a referee for this journal has noted, an alternative strategy would be to argue that the existence of Malament–Hogarth machines of the kind envisioned here directly explains the truth of universally quantified claims over infinite physical structures. In essence, this approach appeals to the structure that must be physically realized in order for supertask computation to be possible, as well as the presumed determinacy of the physical vocabulary used to refer to it. Considered as a strategy for securing determinacy this is closely related to the cosmological approach of Field (1994). Accordingly, supertask computations aren’t playing an essential role in this type of approach.


  1. Benacerraf, P., & Putnam, H. (1983). Introduction. In P. Benacerraf & H. Putnam (Eds.), Philosophy of mathematics: Selected readings (2nd ed.). Cambridge: Cambridge University Press.

    Google Scholar 

  2. Berry, S. (2014). Malament–Hogarth machines and Tait’s axiomatic conception of mathematics. Erkenntnis, 79(4), 893–907.

    Article  Google Scholar 

  3. Bishop, E. (1967). Foundations of constructive analysis. New York: McGraw-Hill.

    Google Scholar 

  4. Button, T. (2009). SAD computers and two versions of the Church–Turing thesis. British Journal for the Philosophy of Science, 60(4), 765–792.

    Article  Google Scholar 

  5. Davies, E. B. (2001). Building infinite machines. British Journal for the Philosophy of Science, 52(4), 671–682.

    Article  Google Scholar 

  6. Dummett, M. (1977). Elements of intuitionism. Oxford: Clarendon Press.

    Google Scholar 

  7. Earman, J., & Norton, J. D. (1993). Forever is a day: Supertasks in Pitowsky and Malament–Hogarth spacetimes. Philosophy of Science, 60(1), 22–42.

    Article  Google Scholar 

  8. Earman, J., & Norton, J. D. (1996). Infinite pains: The trouble with supertasks. In A. Morton & S. P. Stich (Eds.), Benacerraf and his critics (pp. 231–259). Oxford: Blackwell.

    Google Scholar 

  9. Field, H. (1994). Are our mathematical and logical concepts highly indeterminate? Midwest Studies in Philosophy, 19(1), 391–429. P. French, T. Uehling, & H. Wettstein (Eds.).

    Article  Google Scholar 

  10. Hogarth, M. (2004). Deciding arithmetic using SAD computers. British Journal for the Philosophy of Science, 55(4), 681–691.

    Article  Google Scholar 

  11. Kreisel, G. (1967). Informal rigour and completeness proofs. In I. Lakatos (Ed.), Problems in the philosophy of mathematics (pp. 138–157). Amsterdam: North-Holland.

    Google Scholar 

  12. Lavine, S. (n.d.). Skolem was wrong.

  13. McGee, V. (1997). How We Learn Mathematical Language. The Philosophical Review, 106(1), 35.

    Article  Google Scholar 

  14. Parsons, C. (2008). Mathematical Thought and its Objects. Cambridge: Cambridge University Press.

    Google Scholar 

  15. Putnam, H. (1980). Models and reality. Journal of Symbolic Logic, 45(3), 464–482.

    Article  Google Scholar 

  16. Salmon, W. C. (Ed.). (1970). Zeno’s paradoxes. Indianapolis and New York: The Bobbs–Merrill Company Inc.

    Google Scholar 

  17. Shapiro, S. (1991). Foundations without foundationalism: A case for second-order logic. Oxford: Oxford University Press.

    Google Scholar 

  18. Tennant, N. (1978). Natural logic. Edinburgh: Edinburgh University Press.

    Google Scholar 

  19. Warren, J., & Waxman, D. (forthcoming). A metasemantic challenge for mathematical determinacy. Synthese.

  20. Warren, J., & Waxman, D. (2016). A metasemantic challenge for mathematical determinacy. Synthese.

    Article  Google Scholar 

  21. Weyl, H. (1949). Philosophy of mathematics and natural science. Princeton: Princeton University Press.

    Google Scholar 

Download references


Thanks to Sharon Berry, Hartry Field, Tomi Francis, Casper Storm Hansen, Beau Madison Mount, James Studd, Jack Woods, and two (possibly identical) referees for this journal.

Author information



Corresponding author

Correspondence to Jared Warren.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Warren, J., Waxman, D. Supertasks and arithmetical truth. Philos Stud 177, 1275–1282 (2020).

Download citation


  • Arithmetical truth
  • Determinacy
  • Supertasks