{"id":1963,"date":"2025-06-17T17:55:44","date_gmt":"2025-06-17T15:55:44","guid":{"rendered":"https:\/\/linguistica.info\/b\/lei\/?page_id=1963"},"modified":"2025-09-26T08:35:26","modified_gmt":"2025-09-26T06:35:26","slug":"9-3-illocutionary-force","status":"publish","type":"page","link":"https:\/\/linguistica.info\/b\/leiwp\/toc\/9-pragmatics\/9-3-illocutionary-force\/","title":{"rendered":"9.3. Illocutionary force"},"content":{"rendered":"

It is an old idea in linguistics that by uttering a linguistic expression, we perform actions that have consequences in the real world. The British philosopher J.L. Austin was the first to discuss this extensively. He pointed out cases like the following, which he called performatives<\/strong>:<\/p>\n

\n
(1)<\/div>\n
For obstructing traffic on the M4, I sentence<\/strong> you to five weeks’ imprisonment.<\/em><\/div>\n<\/div>\n
\n
(2)<\/div>\n
I order<\/strong> you to report to Pentonville Prison tomorrow morning<\/em>.<\/div>\n<\/div>\n
\n
(3)<\/div>\n
I’m asking<\/strong> you one last time whether you acted alone.<\/em><\/div>\n<\/div>\n
\n
(4)<\/div>\n
I apologize<\/strong> for blocking the traffic.<\/em><\/div>\n<\/div>\n
\n
(5)<\/div>\n
I promise<\/strong> I will not block roads again.<\/em><\/div>\n<\/div>\n

By saying\u00a0I sentence you to five weeks’ imprisonment<\/em>, a judge would of course be performing a linguistic act \u2014 she would assert the truth of the proposition in (6):<\/p>\n

\n
(6)<\/div>\n
SENTENCE-TO(x, y, z), where x is the speaker, y is the hearer and z is a five-month prison term<\/div>\n<\/div>\n

However, she would also be performing a non-linguistic act: she would actually sentence H to a prison term. As a consequence of this, S would make a change to the world: H would now have to spend 5 weeks in jail.<\/p>\n

Austin proposed terms for the different acts speakers perform when they utter a linguistic expression:<\/p>\n

    \n
  • locutionary act<\/strong> for the linguistic act of uttering an expression with a particular form and meaning (for example, the sentence in (1) with the meaning in (6)),<\/li>\n
  • illocutionary act<\/strong> for the non-linguistic act performed by doing so (for example, sentencing H to five weeks in jail); note that the term illocutionary force<\/strong> is often used instead to refer to S’s\u00a0intention<\/em> to perform a particular illocutionary act;<\/li>\n
  • and perlocutionary act or effect<\/strong> for the effect or consequence of the illocutionary act (e.g. the fact that H now has to go to jail).\u00a0<\/li>\n<\/ul>\n

    All three acts taken together make up a speech act<\/strong> (but note that this term is often used to refer just to the illocutionary act\/force).<\/p>\n

    Let us look at another example, the one in (2):<\/p>\n

      \n
    1. The judge says I order you to report to Pentonville prison<\/em>. This is the locutionary act, its meaning can be determined based on the verb\u00a0order<\/em>, which means \u2018tell someone that they have to do something\u2019, where the judge is the one performing this action and the Hearer is the one who should perform the action; the action is described by the verb\u00a0report<\/em>, which in this case means \u2018go to a prison in order to be taken in as a prisoner\u2019, the Hearer is the one performing this action and Pentonville Prison<\/em> is the prison.<\/li>\n
    2. By performing this locutionary act, the judge is not simply describing a situation in the world, she is actually doing the thing she is describing: issuing an order that H must report to Pentonville Prison. This is the illocutionary act.<\/li>\n
    3. As a result of this illocutionary act, the judge has created an obligation for H to report to Pentonville Prison. This is the perlocutionary act.<\/li>\n<\/ol>\n
      \n

      For each of the following verbs, give an example of a locutionary act containing it, and state the illocutionary force and the perlocutionary effect: pronounce<\/em> (as in pronounce someone husband and wife<\/em>, pronounce someone dead<\/em>), quit<\/em>, fire<\/em>, appoint<\/em>, predict<\/em>, warn<\/em>, tell<\/em>.<\/p>\n<\/div>\n

      Note that sentences in (1) to (5), and the ones you created yourself just now, are special in that they indicate the illocutionary force explicitly using the verbs sentence<\/em>, ask<\/em>, order<\/em>, apologize <\/em>and promise<\/em> (such verbs are sometimes called performative verbs<\/strong>). However, utterances do not need to contain such verbs in order to have an illocutionary force. Consider the following paraphrases of (2) to (5):<\/p>\n

      \n
      (2′)<\/div>\n
      Did you act alone?<\/em><\/div>\n<\/div>\n
      \n
      (3′)<\/div>\n
      Report to Pentonville Prison tomorrow morning!<\/em><\/div>\n<\/div>\n
      \n
      (4′)<\/div>\n
      I feel bad for blocking the traffic, I hope you can forgive me.<\/em><\/div>\n<\/div>\n
      \n
      (5′)<\/div>\n
      I will not block roads again<\/em><\/div>\n<\/div>\n

      They have the same illocutionary forces as their performative counterparts. In the first two cases, this force is indicated by the sentence type: (2′) is an interrogative and (3′) is an imperative \u2014 interrogatives function to ask questions and imperatives function to give orders.<\/p>\n

      Because they can indicate illocutionary force, sentence types and performative verbs are sometimes referred to as illocutionary force indicating devices<\/strong> (IFIDs). But as (4′) and (5′) show, not all utterances have such IFIDs. In Section 9.4, we will return to the question how hearers figure out that a sentence like (4′) can be an apology and sentence (5′) can be a promise.<\/p>\n

      So, how many possible illocutionary forces are there? It depends on how specific we want to be: at the most specific level, there are at least as many illocutionary forces as there are performative verbs in a given language \u2014 since cultures differ with respect to what actions they consider important, languages will differ with respect to the verbs they have for describing actions. However, it would be nice if we could categorize all these illocutionary forces into a few general categories, and many linguists and philosophers have attempted to do so.<\/p>\n

      There is no general agreement, but the following categories, which were first suggested by the American philosopher John R. Searle, are widely used:<\/p>\n

        \n
      • assertive<\/strong> \u2014 illocutionary acts by which S asserts the truth of the expressed proposition (statements);<\/li>\n
      • directives<\/strong> \u2014 illocutionary acts by which S intends to make the hearer do something (orders, requests, advice, but also questions, which are requests for information);<\/li>\n
      • commissives<\/strong> \u2014 illocutionary acts by which the speaker commits to a future action (promises, warnings, oaths, bets)<\/li>\n
      • expressives<\/strong> \u2014 illocutionary acts by which the speaker expresses a particular attitudes or emotions (apologies, congratulations, condolences, excuses, greetings, thanks)<\/li>\n
      • declarations<\/strong> \u2014 illocutionary acts by which the speaker performs a conventionalized cultural action (baptisms, sentencing, quitting a job, firing someone, marriage pronouncements)<\/li>\n<\/ul>\n

        A given utterance can fall into more than one of these categories \u2014 they often make a statement in addition to one of the other functions, a commissive may also be expressive (I am sorry, it won\u2019t happen again!<\/em>), etc.<\/p>\n

        Categorize examples (1) to (5) and the examples you came up with yourself according to these five categories.<\/div>\n

        The fact that not all utterances are statements poses a problem for the idea that we can treat the meaning of sentences in terms of their truth conditions: This approach works only for assertives! What would it even mean to say that a question, an order or a promise are \u201ctrue\u201d or \u201cfalse\u201d?<\/p>\n

        Austin proposed a more general way of dealing with the meaning of utterances: instead of thinking of them merely in terms of truth conditions, we should think of them in terms of a more general set of conditions that must be met in order for an utterance to be successful (or, as Austin called it, felicitous<\/em>).\u00a0<\/p>\n

        These conditions are referred to as felicity conditions<\/strong>, and are usually regarded as falling into three types (following suggestions by Searle):<\/p>\n