Loading field map
Speech act theory is the branch of the philosophy of language that studies how utterances do things in the world, not just say things about it. When someone says "I promise to help," "I name this ship," or "I apologize," they are not describing a pre-existing state of affairs; they are performing an action. The theory asks what such actions are, how they work, what conditions make them succeed or fail, and how they relate to meaning, intention, and social institutions.
For much of the early twentieth century, the dominant approach in the philosophy of language treated meaning primarily as a matter of truth and reference. A sentence was meaningful if it could be evaluated as true or false, and its meaning was understood in terms of the conditions under which it would be true. This worked well for statements like "Snow is white," but it left a large class of ordinary utterances unexplained. Questions, commands, promises, warnings, apologies, and declarations all seem meaningful, yet they are not in the business of being true or false. Speech act theory arose to account for this gap: it treats these utterances as actions performed with words, and it asks what makes such actions possible.
The central insight is that when a speaker utters a sentence, they typically perform several acts at once. The most basic is the locutionary act: the physical production of sounds or marks and the construction of a grammatical sentence with a determinate sense and reference. The illocutionary act is what the speaker does in uttering the sentence: promising, warning, requesting, asserting, apologizing. The perlocutionary act is what the speaker achieves by uttering the sentence: convincing, frightening, persuading, or annoying the hearer. The heart of the theory is the illocutionary act, because it is the level at which language connects to social action.
The modern field begins with the work of J. L. Austin, a British philosopher who delivered the William James Lectures at Harvard in 1955, published posthumously in 1962 as How to Do Things with Words. Austin started with a distinction between constative utterances, which describe states of affairs and can be true or false, and performative utterances, which perform an action. His early examples included "I do" in a marriage ceremony, "I bet you five pounds," and "I bequeath this watch to my brother." These sentences do not describe anything; they accomplish something.
Austin soon realized that the distinction was unstable. Many constative utterances also perform actions: to say "The cat is on the mat" is to make an assertion, which is itself an action. Rather than abandoning the insight, he generalized it. All utterances, he argued, are performative in the sense that they all perform illocutionary acts. The constative/performative distinction collapsed into a general theory of illocutionary force.
Austin also introduced the idea of felicity conditions. Unlike truth conditions, which determine whether a statement is true or false, felicity conditions determine whether an illocutionary act is successful or "happy." For a promise to be felicitous, the speaker must be able to perform the promised action, the hearer must want it, and the speaker must intend to be bound by it. For a marriage ceremony to be felicitous, the speaker must be authorized to perform the ceremony, and the participants must be eligible. When these conditions fail, the act does not fail to be true; it fails to be performed at all, or it is performed in a defective way. Austin called such failures infelicities.
John Searle, an American philosopher who studied under Austin at Oxford, transformed the theory into a more systematic and general framework. His 1969 book Speech Acts and his 1975 essay "A Taxonomy of Illocutionary Acts" are the most influential formulations. Searle's central move was to connect speech acts to the philosophy of mind and to the theory of intentionality. He argued that an illocutionary act is a matter of expressing a mental state. An assertion expresses a belief; a promise expresses an intention; a request expresses a desire. The sincerity condition of an act is that the speaker actually has the mental state expressed.
Searle also proposed a taxonomy of illocutionary acts, which has become the standard classification. He distinguished five categories:
Searle's taxonomy is based on the "illocutionary point" of the act, which is the purpose that the act is designed to achieve. He also introduced the distinction between direct and indirect speech acts. A direct speech act is one in which the literal meaning of the sentence matches the illocutionary force. An indirect speech act is one in which the speaker performs one act by way of performing another. For example, "Can you pass the salt?" is literally a question about ability, but it is typically used as a request. The hearer must infer the intended force from the context, the conversational rules, and the shared background of the participants.
Searle's framework was an attempt to give a complete theory of how language works, from the smallest units of meaning to the largest structures of conversation. He argued that speech acts are the basic units of linguistic communication, and that the rules for performing them are constitutive rules, meaning that they create the very activity they govern. The rules of promising, for example, do not regulate an existing activity; they make the activity of promising possible.
A parallel development, closely related to speech act theory, came from the philosopher H. P. Grice. Grice's work on meaning and conversation is not strictly a speech act theory, but it is deeply intertwined with it. Grice distinguished between natural meaning (smoke means fire) and non-natural meaning (a speaker means something by an utterance). He argued that non-natural meaning is a matter of the speaker's intention to produce a belief in the hearer by means of the hearer's recognition of that intention.
Grice also proposed the cooperative principle and the maxims of conversation: the maxims of quantity (be as informative as required), quality (be truthful), relation (be relevant), and manner (be clear). When a speaker flouts a maxim, the hearer must infer an implicature, a meaning that is not literally said but is intended. This account of how hearers recover intended meaning is crucial for understanding indirect speech acts. When someone says "It's cold in here" and the hearer closes the window, the hearer has inferred an indirect request from a statement. Grice's framework explains how such inferences work.
The relationship between Grice and speech act theory is one of mutual support. Speech act theory explains what kinds of acts are possible; Grice's theory explains how hearers figure out which act is being performed. The two are often combined in a broader theory of pragmatics, the study of how context contributes to meaning.
A later development, associated with Searle and Daniel Vanderveken, attempted to formalize speech act theory. In Foundations of Illocutionary Logic (1985), Searle and Vanderveken tried to construct a formal system in which illocutionary acts could be represented as logical entities with conditions of satisfaction. This project aimed to give a rigorous, systematic account of the relations between different speech acts, such as the fact that a promise entails a commitment, or that a request and a command share a common illocutionary point but differ in force.
This formal approach has been influential in computational linguistics and in the design of artificial agents that need to reason about communication. However, it has also been criticized for being too rigid. The formal system tends to assume that speech acts are discrete, well-defined entities, whereas in practice they are often vague, overlapping, and context-dependent. The formal approach has not replaced the more flexible, descriptive accounts of Austin and Searle.
A different line of development came from outside the analytic tradition. The French philosopher Jacques Derrida, in his 1971 essay "Signature Event Context," engaged directly with Austin's theory. Derrida argued that Austin's account of speech acts was too dependent on the idea of a "context" that could be fully specified and controlled. Derrida pointed out that utterances are iterable: they can be repeated in new contexts, and this repetition is essential to their functioning. A promise, for example, must be recognizable as a promise even when it is quoted, performed on stage, or written in a book. This iterability means that the meaning of an utterance is never fully fixed by the speaker's intention or the original context.
Derrida's critique has been taken up in literary theory and cultural studies, where it has been used to question the stability of meaning and the authority of the speaker. It has also been influential in the philosophy of law and in the study of political speech. The feminist philosopher Judith Butler, for example, has used speech act theory to analyze hate speech and the way that language can be used to constitute social identities. Butler argues that the power of hate speech is not just that it expresses a hostile attitude, but that it performs an act of subordination. This is a direct application of the idea of the illocutionary act, but it also draws on Derrida's insight that the act is not fully controlled by the speaker.
This critical turn is not a replacement for the analytic tradition; it is a different way of using the theory. It is more concerned with the social and political effects of speech acts than with the conditions of their success. It has also been criticized by analytic philosophers for being imprecise and for conflating the illocutionary act with the perlocutionary effect.
In the late twentieth century, speech act theory was taken up by linguists and psychologists, who used it as a framework for empirical research. In linguistics, the theory became a foundation of pragmatics, the study of language use. The concept of the illocutionary act is used to analyze the function of sentences in discourse, and the taxonomy of speech acts is used to classify the functions of utterances in conversation. The theory has also been used in the study of language acquisition, where it is used to describe how children learn to perform and understand speech acts.
In psychology, speech act theory has been used to study the development of communicative competence and the nature of social cognition. The theory has also been used in the study of speech act comprehension, which investigates how hearers recognize the intended illocutionary force of an utterance. This research has shown that hearers use a variety of cues, including intonation, facial expression, and context, to infer the force of an utterance.
The empirical turn has also revealed some of the limitations of the theory. The taxonomy of speech acts, for example, is not always easy to apply to real-world data. Many utterances do not fit neatly into one category, and the boundaries between categories are often fuzzy. The theory also tends to assume a single speaker and a single hearer, but many real-world speech acts are performed by groups or in complex institutional settings.
Speech act theory is not a single, unified doctrine but a family of approaches that share a common core: the idea that language is a form of action. The field is organized around a set of enduring questions rather than a single paradigm. These questions include:
The analytic tradition, from Austin to Searle, remains the most influential and the most widely taught. It provides the basic vocabulary and the basic framework. The Gricean tradition provides the theory of inference that explains how speech acts are understood. The formal tradition has been useful in computational applications. The critical tradition has opened up the theory to questions of power and politics.
The field is also characterized by a persistent tension between the universal and the particular. On the one hand, speech act theory aims to describe the general conditions of communication, which are assumed to be universal. On the other hand, the theory has been criticized for being based on a narrow, Western, and often English-speaking model of communication. The study of speech acts in other languages and cultures has shown that the categories and the conditions of success are not universal. For example, the act of promising is not the same in all cultures, and the conditions for a successful apology vary widely.
The most durable contribution of speech act theory is its insistence that language is not a neutral medium for describing the world but a tool for acting in it. This insight has been absorbed into the broader philosophy of language, into linguistics, and into the social sciences. The theory has been modified, criticized, and extended, but its central question remains: what are we doing when we speak?