Undoing things with words

  • 203 Accesses

  • 1 Citations


Over the last five decades, philosophers of language have looked into the mechanisms for doing things with words. The same attention has not been devoted to how to undo those things, once they have been done. This paper identifies and examines three strategies to make one’s speech acts undone—namely, Annulment, Retraction, and Amendment. In annulling an act, a speaker brings to light its fatal flaws. Annulment amounts to recognizing an act as null, whereas retraction and amendment amount to making it null. Speakers employ retraction to cancel the deontic updates engendered by a given act. They instead use amendment to adjust its degree of strength. I will argue that annulling, retracting, and amending are second-order speech acts, whose felicity conditions vary with the type of illocution they operate on. Undoing is therefore conceived of as a form of doing. Furthermore, I claim that, in calling off our acts, we undo the conventional or illocutionary effects of our words while leaving intact their past causal or perlocutionary outcomes.

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

Access options

Buy single article

Instant unlimited access to the full article PDF.

US$ 39.95

Price includes VAT for USA

Subscribe to journal

Immediate online access to all issues from 2019. Subscription will auto renew annually.

US$ 199

This is the net price. Taxes to be calculated in checkout.


  1. 1.

    A notable exception is the strand of literature on retracting assertives. Bach and Harnish (1979, p. 43) offer an explicit account of claim retraction. For a critical view of their account, see MacFarlane (2011) and Sect. 3.2 below. Retractions also play a crucial role in MacFarlane (2014)’s argument for assessment-relativism. For responses, see Ferrari and Zeman (2014), Marques (2015) and Ferrari (2016). See, moreover, Walton and Krabbe’s work on retracting commitment to statements in the context of critical discussion (Walton and Krabbe 1995; Krabbe 2001).

  2. 2.

    Strictly speaking, in requesting you to do \(\phi \), I try to impute a reason for you to grant the request. Although a request can be appropriately responded to by the hearer granting or refusing it, only the former of such responses satisfies the act’s constitutive goal. In the parlance of conversation analysts, granting is the preferred second. Cf. Levinson (1983, p. 307).

  3. 3.

    I draw here on Bach and Harnish’s distinction between conventional and communicative acts. Conventional acts roughly cover Austin’s ritual performatives. Cf. Bach and Harnish (1979, esp. ch. 6).

  4. 4.

    By uttering (1), the police commissioner may also produce unintended perlocutionary sequels. For example, she may cause John to get anxious because of the secrecy surrounding the case, or disappointed for he thinks it would make more sense to issue a resonant press release.

  5. 5.

    A speaker may succeed in blocking some future consequences of the act she calls off, but can in no way make the perlocutionary outcomes that have already taken place disappear.

  6. 6.

    Cf. Austin (1962, p. 102): «The consequential effects of perlocutions are really consequences, which do not include such conventional effects as, for example, the speaker’s being committed by his promise (which comes into the illocutionary act)».

  7. 7.

    For a more detailed discussion of the illocution-perlocution distinction, see, among others, Cohen (1973), Kissine (2008) and Sbisà (2013).

  8. 8.

    Austin mentions this passage in his preparatory notes for How to Do Things with Words, while making some remarks on annulment. See Sbisà (2007) for an insightful discussion of Austin’s unpublished notes.

  9. 9.

    Langton reads Austin’s point as constitutive. The idea underlining such reading is that (necessary) felicity conditions of speech acts can be supplied in the future, relative to the time of utterance. If such future conditions do not obtain, then the act that they would otherwise have constituted is ‘made undone’. Unlike Langton, I offer here an epistemic reading of Austin’s remarks, according to which a speech act gets undone if some of its past felicity conditions are later discovered not to be fulfilled. Cf. Langton (2018) and her 2015 John Locke Lecture VI, whose title—“How to Undo Things with Words”—is a nod to Austin (1962), as is the title of this paper.

  10. 10.

    Austin’s list of exercitive verbs includes both annul and withdraw. (Austinian withdrawals square with what I refer to as retractions.) Cf. Austin (1962, pp. 155–156).

  11. 11.

    Given a force F, Searle and Vanderveken (1985, p. 50) distinguish between success conditions that are common to all forces within the same illocutionary category as F (general conditions of F) and conditions that are characteristic of F—i.e., not shared by the entire category F falls in (special conditions of F).

  12. 12.

    The adverb ‘infelicitously’, as used in condition i., refers to fatal infelicities (i.e., cases in which the act violates Austin’s A or B rules). Condition ii. should be more precisely formulated as in the starred version below:

    ii.* A was not already annulled at any time between t and \(t_{{n}}\) via an annulment act that has retained its validity up to \(t_{{n}}\).

    It is at least plausible to imagine a scenario in which A gets annulled at \(t_{{\textit{1}}}>t\) via an annulment act that is declared void at \(t_{{\textit{2}}}>t_{{\textit{1}}}\). In this scenario, A might be annulled again at \(t_{{n}}>t_{{\textit{2}}}\) via a second annulment act that re-enables the first. Condition ii.* is meant to account for such cases (as well as for more standard ones).

  13. 13. (accessed October 24, 2017).

  14. 14.

    In the time frame between the performance of an act and its retraction, the deontic statuses of the conversational participants may have changed in diverse ways for reasons that have nothing to do with the act at stake. It goes without saying that those updates will remain in force even once the act is retracted. Every retracting maneuver erases only the normative facts that exist in virtue of the act to be retracted.

  15. 15.

    Under the party’s constitution, a party leader must “communicate his decision to resign in writing to the party chairman, who must then summon an emergency meeting of the NEC within 28 days” (art. 7.7). Notice that the party’s rules do not foresee the possibility for the NEC to refuse a written resignation. Even in the case everyone on the NEC is opposed to it, an official resignation should count as such.

  16. 16.

    This is an oversimplification that serves the purpose of drawing a sharp line between annulment and retraction. Every speech act (originally taken as felicitous) can, in principle, turn out to be null and void should some fatal flaws be discovered (or a known flaw be reassessed and deemed to be fatal). It ensues that annullable acts are practically indistinguishable from valid ones until their deficiencies are brought to light—and that infelicitous acts, contrary to what I stated above, may be retracted too as long as they are not recognized as such. In my view, however, retractions of this sort would be annullable acts themselves, for it may later come up that the original act was void and the retraction, therefore, mistargeted. For this reason and for the sake of clarity, in what follows I will put this complication aside.

  17. 17.

    I thank an anonymous reviewer for pressing me to clarify this point.

  18. 18.

    See n. 1 above.

  19. 19.

    This is not the same as obtaining an annulment. According to the Catholic Church Canon Law, to annul is to state that no marriage ever existed; an annulment is a declaration that the couple, who apparently got married, was not truly married after all. To dissolve, by contrast, is to declare that the two parties were really married, but their marriage was not indissoluble. (The indissolubility of a valid marriage attains through its consummation.) See, esp., cc. 1141 and 1142.

  20. 20.

    The sincerity conditions of a given act are defined by specifying the psychological state(s) the speaker expresses in performing that act. For example, in asserting one expresses a belief, in making a promise one expresses an intention, and so on. Cf. Searle (1969, p. 65).

  21. 21.

    Though assertions imply the expression of a belief (as shown by Moore’s Paradox—one cannot assert that P while disavowing the belief that P), retractions of assertions do not express any lack of belief (after all, it makes perfect sense to say “I retract that, but I keep believing it”).

  22. 22.

    I mean this to account for both synchronic and diachronic retractions. A synchronic retraction is one performed in the same conversational turn as the original speech act; a diachronic retraction is one performed in some subsequent turn. Notice that the adjective ‘synchronic’ is used here in a loose sense, for synchronic retractions (e.g., “I’ll pick you up at the station at noon...Oh, no, wait. Scratch that. I can’t!”) are still after the original act.

  23. 23.

    The expression ‘felicitously performed acts’, as used in the general felicity conditions of retraction and amendment, refers to both acts that comply with all their felicity rules (fully felicitous acts) and acts that violate some \(\varGamma \) rules (partially felicitous acts). For a seeming exception to condition i. for retraction, see n. 16 above.

  24. 24.

    Of course, you can disobey the order—but that would cause the act to perlocutionarily fail (while remaining illocutionarily intact).

  25. 25.

    I am concerned with direct speech acts. If an assertive is used to indirectly perform a request, a promise, or a speech act of another kind, then its amendment conditions will be completely different.

  26. 26.

    Pleas are also typically performed in a humbler manner than requests. Cf. Searle and Vanderveken (1985, p. 204).

  27. 27.

    Again, this is meant to cover both synchronic and diachronic amendments. Utterances (2)–(5) are examples of synchronic amendment; (7) is instead a diachronic amendment. See n. 22 above.

  28. 28.

    As in the case of retraction, this condition allows for exceptions. Indeed, infelicitous acts seem to be amendable too as long as they are not recognized as such. One can, for example, amend a strong claim into a weaker one before anybody figures out that one did not have sufficient evidence to back it up. See n. 16 above.


  1. Alston, W. P. (2000). Illocutionary acts and sentence meaning. Ithaca, NY: Cornell University Press.

  2. Austin, J. L. (1962). How to do things with words. London: Oxford University Press.

  3. Bach, K., & Harnish, R. M. (1979). Linguistic communication and speech acts. Cambridge, MA: MIT Press.

  4. Cohen, T. (1973). Illocutions and perlocutions. Foundations of Language, 9(4), 492–503.

  5. Ferrari, F. (2016). Assessment-sensitivity. Analysis, 76(4), 516–527.

  6. Ferrari, F., & Zeman, D. (2014). Radical relativism, retraction and ‘being at fault’. In S. Caputo, M. Dell’Utri, & F. Bacchini (Eds.), New frontiers in truth (pp. 80–102). Newcastle: Cambridge Scholar.

  7. Kissine, M. (2008). Locutionary, illocutionary, perlocutionary. Language and Linguistics Compass, 2(6), 1189–1202.

  8. Krabbe, E. C. (2001). The problem of retraction in critical discussion. Synthese, 127, 141–159.

  9. Lance, M., & Kukla, R. (2013). Leave the gun; Take the cannoli! The pragmatic topography of second-person calls. Ethics, 123, 456–478.

  10. Langton, R. (2018). Blocking as counter-speech. In D. Harris, D. Fogal, & M. Moss (Eds.), New work on speech acts. New York: Oxford University Press.

  11. Levinson, S. (1983). Pragmatics. Cambridge: Cambridge University Press.

  12. MacFarlane, J. (2011). What is assertion? In J. Brown & H. Cappelen (Eds.), Assertion. New philosophical essays (pp. 79–96). New York: Oxford University Press.

  13. MacFarlane, J. (2014). Assessment sensitivity: Relative truth and its applications. Oxford: Clarendon Press.

  14. Marques, T. (2015). Retractions. Synthese.

  15. McGowan, M. K. (2004). Conversational exercitives: Something else we do with our words. Linguistics and Philosophy, 27, 93–111.

  16. Sbisà, M. (1984). On illocutionary types. Journal of Pragmatics, 8, 93–112.

  17. Sbisà, M. (2007). How to read Austin. Pragmatics, 17, 461–473.

  18. Sbisà, M. (2013). Locution, illocution, perlocution. In M. Sbisà & K. Turner (Eds.), Pragmatics of speech actions (pp. 25–75). Berlin: Mouton de Gruyter.

  19. Sbisà, M. (2014). The Austinian conception of illocution and its implications for value judgments and social ontology. Ethics and Politics, 16(2), 619–631.

  20. Searle, J. R. (1969). Speech acts. An essay in the philosophy of language. Cambridge: Cambridge University Press.

  21. Searle, J. R. (1975a). Indirect speech acts. In P. Cole & J. L. Morgan (Eds.), Syntax and semantics: Speech acts (pp. 59–82). New York: Academic Press.

  22. Searle, J. R. (1975b). A taxonomy of illocutionary acts. In K. Gunderson (Ed.), Minnesota studies in the philosophy of science (Vol. VII, pp. 344–369). Minneapolis: University of Minnesota Press.

  23. Searle, J. R., & Vanderveken, D. (1985). Foundations of illocutionary logic. New York: Cambridge University Press.

  24. Strawson, P. F. (1964). Intention and convention in speech acts. Philosophical Review, 73(4), 439–460.

  25. Walton, D. N., & Krabbe, E. C. (1995). Commitment in dialogue: Basic concepts of interpersonal reasoning. Albany, NY: State University of New York Press.

Download references


Conversations with Claudia Bianchi, Bianca Cepollaro, Fabio Del Prete, Rae Langton, Neri Marsili, Jennifer Saul, and Marina Sbisà were extremely helpful, as were comments from two anonymous reviewers for this journal. I am grateful to the audiences at the 4th PLM Conference in Bochum 2017, the 24th SFL Conference in Milan 2018 and 12th IUWMAM “Language & Power” in Barcelona 2018, where this paper was partially presented. Special thanks to the Sheffield Language RG 2017 participants: the paper has greatly benefitted from the lively discussion.

Author information

Correspondence to Laura Caponetto.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Caponetto, L. Undoing things with words. Synthese (2018) doi:10.1007/s11229-018-1805-9

Download citation


  • Annulment
  • Retraction
  • Amendment
  • Undo
  • Speech acts