Speech Act Theory
What it is
Speech Act Theory is an account of language as action. It holds that an utterance is not merely a description that can be true or false but a deed performed in the act of speaking. The philosopher J.L. Austin called these performative utterances, and his student John Searle built them into a systematic theory of how words accomplish social acts such as requesting, promising, and declaring.
The core idea
Every utterance carries three layers. The locutionary act is the literal saying of words with sense and reference. The illocutionary act is what the speaker does in saying them, such as warning or promising, and this is the heart of the theory. The perlocutionary act is the effect produced on the listener, such as persuading or frightening. Meaning therefore lives in intended social force, not in literal content alone.
How it is used
Scholars use the theory to analyze what an utterance accomplishes rather than only what it states. By coding messages for illocutionary force (assertive, directive, commissive, expressive, or declarative in Searle's taxonomy) and checking felicity conditions (the contextual requirements that make an act succeed), researchers explain misfires, indirect requests, and the gap between literal wording and pragmatic intent across conversation, institutions, and technology.
In practice
A person at a dinner table says, "It is cold in here." The locutionary act is simply uttering that grammatical sentence about temperature. The illocutionary act, given the setting, is a request that someone close the window. The perlocutionary act occurs if a listener actually rises and shuts it. The literal statement and the accomplished deed are plainly not the same thing.
Key studies & evidence
J.L. Austin laid the foundation in his 1955 William James Lectures at Harvard, published posthumously in 1962 as How to Do Things with Words. He began by separating constative utterances, which describe and can be true or false, from performatives, which do something, then collapsed the distinction by arguing all speech has illocutionary force. John Searle, Austin's student, formalized the program in Speech Acts (1969), grounding it in constitutive rules and felicity conditions. In a 1976 paper Searle proposed the influential five-part taxonomy of illocutionary acts: assertives (representatives), directives, commissives, expressives, and declarations. H.P. Grice's parallel work on conversational implicature complemented the theory by explaining how listeners infer indirect illocutionary force from what is left unsaid.
Critiques & limitations
Critics note that classifying illocutionary force is harder than the taxonomy suggests, since context and tone can make one utterance several acts at once, and indirect speech acts blur the categories. Jacques Derrida challenged Austin's reliance on serious, sincere, first-person utterances, arguing that quotation, fiction, and citation are not deviant but central to how language works. The theory was built largely on isolated sentences and idealized Western, individualist speakers, so it underplays culture, power, and conversational sequence. Conversation analysts counter that meaning is negotiated turn by turn between participants, not fixed by a single speaker's intention, which the theory tends to privilege.
Applications
Speech Act Theory is a staple of pragmatics, discourse analysis, and communication pedagogy, where it teaches students to read intent and force rather than literal wording. It informs politeness research, cross-cultural communication, and the study of how directives and commitments function in organizations and law. In computer-mediated and mediated-presence settings it helps explain how requests, promises, and self-disclosures are performed through text, emoji, and platform features where vocal and bodily cues are thin. Social-media analytics increasingly classify posts by illocutionary intent (a complaint, a threat, a pledge) rather than topic alone, and the theory is now used to audit how conversational AI systems perform and recognize acts like promising, apologizing, and asserting, which is directly relevant to AURA Lab work on mediated and machine communication.