Loading field map
When someone says "I promise to be there," are they describing a future action or performing the act of promising itself? Speech act theory begins with the insight that many utterances do not describe the world—they change it. The subfield's history is shaped by a persistent disagreement: is the force of a speech act best explained by the speaker's intention and shared conventions, or by the social power structures and institutional contexts that authorize some speakers to act while silencing others?
J. L. Austin opened the subfield by challenging the dominant truth-conditional paradigm, which treated sentences as true or false descriptions of reality. In lectures at Oxford in the 1950s, he identified a class of utterances he called performatives—sentences that do something rather than report something. Saying "I name this ship the Queen Elizabeth" during a christening ceremony does not describe a naming; it accomplishes the naming. Austin argued that performatives are not true or false but felicitous or infelicitous: they succeed or fail depending on whether certain conditions are met. These felicity conditions include the existence of an accepted conventional procedure, the appropriateness of the participants and circumstances, and the correct execution of the procedure. This framework carved out a domain of language that truth-conditional semantics could not handle, but it left a puzzle: what about sentences that look like descriptions but also perform actions?
Austin soon realized that the performative/constative distinction could not be maintained. A sentence like "I warn you that the bull is about to charge" looks like a description (it can be true or false) but also performs the act of warning. The pressure to account for such mixed cases forced Austin to collapse the distinction and replace it with a more general theory of speech acts. In How to Do Things with Words (1962, posthumous), he introduced a tripartite structure: every utterance has a locutionary act (the act of saying something with a certain meaning), an illocutionary act (the act performed in saying it, such as promising, warning, or ordering), and a perlocutionary act (the effect produced by saying it, such as persuading or alarming). The illocutionary act became the core of the theory. By absorbing performatives into a broader framework, Austin showed that all utterances have illocutionary force, not just a special class. Felicity conditions now applied to all speech acts, not only to explicit performatives. This framework gave the subfield its foundational vocabulary and set the agenda: what kinds of illocutionary acts are there, and what makes them succeed or fail?
John Searle took Austin's loose list of illocutionary verbs and systematized it. In Speech Acts (1969) and later work, he argued that illocutionary acts can be classified by their illocutionary point (the purpose of the act) and their direction of fit between words and world. His taxonomy identified five categories: assertives (commit the speaker to the truth of a proposition, e.g., stating, claiming), directives (attempt to get the hearer to do something, e.g., requesting, ordering), commissives (commit the speaker to a future course of action, e.g., promising, offering), expressives (express a psychological state, e.g., thanking, apologizing), and declarations (change the world by representing it as changed, e.g., declaring war, firing an employee). Declarations were especially important because they require an extra-linguistic institution—a boss, a judge, a referee—to succeed. Searle's taxonomy did not replace Austin's framework; it narrowed and refined it by providing a principled basis for classification. It also served as infrastructure for later work: the declaration category, in particular, pointed toward a theory of how language creates social reality.
Searle himself noticed that many utterances perform one illocutionary act indirectly while appearing to perform another. When someone asks "Can you pass the salt?" the literal form is a question about ability (a directive, in Searle's taxonomy), but the intended illocutionary act is a request. Indirect speech act theory analyzed how hearers infer the speaker's real intention from the literal meaning, the context, and shared background knowledge. This framework drew heavily on Gricean Pragmatics, which had already shown that communication relies on inferential reasoning guided by cooperative principles. The connection was mutually reinforcing: Grice's framework explained the mechanism of indirectness (the hearer infers the intended force by assuming the speaker is cooperative), while indirect speech acts provided Grice's theory with a rich domain of cases. Indirect speech act theory shifted attention toward the hearer's perspective and the inferential processes that bridge literal meaning and intended force. It coexisted with Searle's taxonomy by showing that the same sentence can realize multiple illocutionary points depending on context, and it narrowed the earlier frameworks' focus on speaker intention by emphasizing the hearer's interpretive work.
Since 1990, speech act theory has branched into two active traditions that respond to the Austin-Searle legacy in divergent ways.
Feminist philosophers such as Rae Langton and Jennifer Hornsby argued that the classical frameworks overlooked how social power shapes the conditions for felicitous speech acts. Langton's analysis of pornography as a speech act that subordinates and silences women drew directly on Austin's felicity conditions: if a speaker lacks the authority required by the conventional procedure, the speech act is infelicitous. Feminist speech act theory showed that silencing is not merely the absence of speech but the systematic failure of a speaker's illocutionary acts to achieve their intended force due to the hearer's refusal to recognize the speaker's authority. This framework transformed Austin's concept of felicity by revealing that the conditions for successful speech acts are not neutral conventions but are shaped by gender, race, and other power structures. It also challenged Searle's taxonomy by arguing that some speech acts (like subordination) do not fit neatly into his five categories because their illocutionary point depends on the social position of the speaker and hearer.
Searle himself extended his earlier work into a theory of social ontology. In The Construction of Social Reality (1995) and later writings, he argued that institutional facts—money, marriage, property—are created by declarations that assign status functions to objects and people. The formula "X counts as Y in context C" captures how brute facts (a piece of paper) become institutional facts (a dollar bill) through collective acceptance. Institutional speech act theory grew directly out of Searle's taxonomy's declaration category: declarations are the mechanism by which institutions are created and maintained. This framework narrowed the earlier focus on individual speech acts by emphasizing the constitutive rules that underlie entire domains of social reality. It coexists with feminist speech act theory by offering a complementary account of authority: for Searle, authority is a matter of occupying a position within an institutional structure; for feminist theorists, authority is contested and often denied to marginalized speakers.
Today, feminist and institutional speech act theorists agree that speech acts are not merely individual actions but are embedded in social and institutional contexts. They agree that felicity conditions include not only speaker intention and convention but also the speaker's social position and the hearer's recognition. However, they disagree about the primary locus of illocutionary force. Institutional theory locates force in the constitutive rules and collective acceptance that define institutional roles: a judge can declare a verdict because the institution of law authorizes her. Feminist theory locates force in the power relations that determine whose speech acts are recognized and whose are silenced: a woman's protest may fail as a speech act not because she lacks institutional authority but because the hearer refuses to grant her the standing to perform that act. This disagreement is not merely theoretical; it shapes how each framework analyzes real-world cases of harm, protest, and political speech. The subfield remains divided between those who see institutions as the foundation of illocutionary force and those who see power as the hidden condition that institutions often encode.