Speech-act Theory

From Chapter 2 of my dissertation: "Understand and Act: Classical Rhetoric, Speech Acts, and the Teaching of Critical Democratic Participation." This is a truncated overview of speech-act theory.


       In his famous work, "How to do Things with Words," J. L. Austin outlined his theory of speech acts and the concept of performative language, in which to say something is to do something. To make the statement “I promise that p” (in which p is the propositional content of the utterance) is to perform the act of promising as opposed to making a statement that may be judged true or false. Performatives cannot be true or false, only felicitous or infelicitous. Austin creates a clear distinction between performatives and constantives, statements that attempt to describe reality and can be judged true or false, but he eventually comes to the conclusion that most utterances, at their base, are performative in nature. That is, the speaker is nearly always doing something by saying something.
       For Austin, what the speaker is doing is creating social realities within certain social contexts. For example, using an explicit performative, to say “I now pronounce you man and wife” in the context of a wedding, in which one is marrying two people, is to create a social reality, i.e. in this case a married couple.
       Austin described three characteristics, or acts, of statements that begin with the building blocks of words and end with the effects those words have on an audience. Locutionary acts: “roughly equivalent to uttering a certain sentence with a certain ‘meaning´ in the traditional sense.” Illocutionary acts: “such as informing, ordering, warning, undertaking, &c., i.e. utterances which have a certain (conventional) force.” Perlocutionary acts: “what we bring about or achieve by saying something, such as convincing, persuading, deterring, and even, say, surprising or misleading” (109). Austin focused on illocutionary acts, maintaining that here we might find the “force” of a statement and demonstrate its performative nature. For example, to say “Don´t run with scissors” has the force of a warning when spoken in a certain context. This utterance may be stated in an explicitly performative way, e.g., “I warn you, don´t run with scissors.” This statement is neither true nor false. Instead, it creates a warning. By hearing the statement, and understanding it as a warning, the auditor is warned, which is not to say that the auditor must or will act in any particular way regarding the warning.

 . . .
       Austin maintained that once “we realize that what we have to study is not the sentence but the issuing of an utterance in a speech situation, there can hardly be any longer a possibility of not seeing that stating is performing an act” (139). This conclusion stated his belief that studying words or sentences (locutionary acts) outside of a social context tells us little about communication (illocutionary acts) or its effect on an audience (perlocutionary acts). Speech-act theory and its critiques are often articulated using imagined examples, in which the context of the utterance is erased to make a linguistic point and later added to demonstrate the true complexity of even the simplest statement. To show how statements (performatives) work, linguistic scholars have reduced the illocutionary act to the symbolic expression F(p), in which p is the propositional content and F is the illocutionary force.

    John Searle claims the illocutionary act is “the minimal complete unit of human linguistic communication. Whenever we talk or write to each other, we are performing illocutionary acts” (Mind 136). Illocutionary acts are performed with intentionality. As Bach and Harnish explain, people “don´t speak merely to exercise their vocal cords.” Some reason always exists, and this reason is called the communicative presumption: the mutual belief that whenever one person says something to another, the speaker intends to perform an illocutionary act (7). Further, an “illocutionary act is communicatively successful if the speaker´s illocutionary intention is recognized by the hearer . . . This is what communication is about; anything more is more than just communication” (15). But more does exist because “illocutionary acts are all intentional and are generally performed with the primary intention of achieving some perlocutionary effect” (17). For example, a speaker may say “Shut the window” intending for the auditor to understand this communication as an order and further intending that the auditor should shut the window. According to Searle, a speech act may have any number of effects on the auditor other than those intended by the speaker. For example, the speaker might say “Shut the window,” and the auditor might respond by saying “Shut it yourself.” From this, Searle claims that the “fact that illocutionary acts are essentially intentional, whereas perlocutionary acts may or may not be intentional, is a consequence of the fact that the illocutionary act is the unit of meaning in communication” (Mind 137). This position eliminates intended effects from communication, which, from a rhetorical perspective, is odd. I would argue, in agreement with Bach and Harnish, that it is exactly the perlocutionary effect that drives speaker intention. What is the proper unit of data in communication? What is the unit of meaning? For some speech-act theorists it is F(p), in which the auditor understands the illocutionary force of a statement linked to its propositional content.

    While the number of uses for language is “enormous,” Searle believes that there is a limited number of things we can do with language. In the structure of the illocutionary act F(p), the potential propositional content is limitless. But Searle asks: “How many types of F are there?” This is the same as asking: How many things can be done with language? And this immediately sets us to thinking of the number of verbs that identify illocutionary acts or describe the effect of the illocutionary act on the auditor. So how many Fs there are would be limited to the number of verbs and open to the vagaries of language use and change. To “overcome” this problem, Searle posits the notion of “illocutionary point,” which is the “point or purpose in virtue of its being an act of that type” (Mind 147). In other words, the illocutionary point is the intention behind the illocutionary act, which is stated in a verb that describes the work the sentence is doing. Austin created such a schema, but I am in agreement with Petrey that Searle´s “most serviceable refinement” of Austin´s work is his five-part schema of illocutionary force (59).

    Searle posits five illocutionary points: 1) Assertives: statements that may be judged true or false because they purport to describe a state of affairs in the world; 2) Directives: statements that attempt to make the auditor´s actions fit the propositional content; 3) Commissives: statements which commit the speaker to a course of action as described by the propositional content; 4) Expressives: statements that express the “sincerity condition of the speech act”; and 5) Declaratives: statements that attempt to change the world by “representing it as having been changed” (Mind 148-50). That is, when we speak (or write) we are doing one or more of the following: asserting, directing, commiserating, expressing, or declaring. As Sadock explains, the “illocutionary force of an utterance is always interpreted as having been intended” (10). This statement recognizes the pragmatic contention of Bach and Harnish that people do not speak simply to “exercise their vocal cords.” They would add that speakers have primary intentions that are perlocutionary. The very act of speaking (or writing) rhetorically presupposes an intention, and intentions of a certain kind may be found in the illocutionary force of a statement as it affects the propositional content. If the auditor understands our intended illocutionary point (illocutionary force) in its relation to the propositional content, we can be said to have communicated.

. . .

Back to Rhetorica Critical Meter

The Rhetorica Network