Loading field map
Philosophical pragmatics is the branch of the philosophy of language that studies meaning as it arises in actual communicative use. Where formal semantics typically asks what sentences mean in virtue of their grammatical structure and the literal meanings of their words, philosophical pragmatics asks what speakers mean when they utter those sentences on particular occasions, and how hearers are able to figure that out. Its central subject matter is the gap between what is said and what is communicated—a gap that turns out to be far wider, and far more philosophically interesting, than it first appears.
The founding observation of the field is that ordinary communication constantly exceeds the literal content of the sentences we utter. If someone at a dinner table says, "Can you pass the salt?", the sentence literally asks about the hearer's physical ability. Yet the speaker is normally requesting action, not requesting information, and the hearer normally understands this without hesitation. Similarly, if a professor writes "I have office hours on Tuesday" in response to a student's email asking for a meeting, the sentence states a fact about the professor's schedule, but the student understands it as an offer—and, in context, perhaps as a refusal of any other day.
These are not marginal or decorative features of language. They are pervasive, systematic, and essential to how language works. A hearer who could only decode literal sentence meaning would be unable to participate in ordinary conversation. The central task of philosophical pragmatics is to explain how this extra layer of communication is possible: what rules, principles, or cognitive capacities allow speakers to convey more than they say, and allow hearers to recover what was meant.
The field is philosophical rather than merely empirical because these questions connect directly to deeper issues: What is the nature of meaning itself? Is meaning fundamentally a property of sentences or of speaker intentions? How much of what we understand is encoded in language, and how much is supplied by reasoning about the world and about other minds? These questions place pragmatics at the intersection of philosophy of language, philosophy of mind, and the empirical study of language.
The first major systematic approach to these questions grew out of ordinary language philosophy in the mid-twentieth century. J. L. Austin, in his lectures later published as How to Do Things with Words, challenged the assumption that language is primarily in the business of describing reality. He drew attention to utterances that do not describe anything at all but instead perform actions: promising, apologizing, naming a ship, pronouncing a couple married. Austin called these performatives and argued that they reveal something general about language: every utterance is itself an action, performed in accordance with social conventions.
Austin distinguished three levels of any utterance. The locutionary act is the act of saying something with a certain sense and reference. The illocutionary act is what we do in saying it—promising, warning, asserting, requesting. The perlocutionary act is what we bring about by saying it—convincing, frightening, persuading. The heart of the theory is the illocutionary act, which Austin analyzed as governed by conditions of appropriateness: for a promise to be felicitous, the speaker must be able to do the thing promised, the hearer must want it, and so on. This moved the analysis of meaning away from truth conditions and toward the conditions under which utterances count as the performance of particular social acts.
John Searle systematized and extended this framework. He proposed that speaking a language is a rule-governed form of behavior, and that illocutionary acts are constituted by constitutive rules—rules that create the very activities they govern, as the rules of chess create the game. Searle distinguished five basic categories of illocutionary acts—assertives, directives, commissives, expressives, and declarations—and argued that each type is defined by its direction of fit between words and world: assertions aim to make words fit the world, requests aim to make the world fit words, and so on. He also introduced the distinction between direct and indirect speech acts. In an indirect speech act, the speaker performs one illocutionary act by way of performing another: "Can you pass the salt?" is literally a question but indirectly a request. This distinction became central to pragmatics because it made explicit the gap between sentence meaning and utterance meaning that the field seeks to explain.
The speech act tradition's enduring contribution is its insistence that meaning cannot be understood apart from action. Language is not a neutral vehicle for conveying propositions; it is a way of doing things in the world. Its limitation is that it offers little account of the cognitive processes by which hearers recognize what speech act is being performed, especially in indirect cases. The theory describes the structure of speech acts but does not explain how communication succeeds.
The most influential approach in philosophical pragmatics is due to H. P. Grice, whose William James lectures at Harvard in 1967 reshaped the field. Grice's starting point was a distinction between two kinds of meaning. Sentence meaning is what the words and grammar of a sentence conventionally encode. Speaker meaning is what the speaker intends to communicate by uttering the sentence on a particular occasion. Grice analyzed speaker meaning in terms of intentions: a speaker means something by an utterance if she intends the utterance to produce a certain effect in the hearer, and intends the hearer to recognize that intention, and intends the hearer's recognition of the intention to be part of the reason the effect occurs. This reflexive intention structure—sometimes called "communicative intention"—became the standard way of understanding what it is to mean something.
Grice then asked how hearers can recover speaker meaning when it goes beyond sentence meaning. His answer was that conversation is a rational, cooperative activity, and participants can rely on this rationality. He proposed a general Cooperative Principle: make your conversational contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange. From this principle he derived a set of maxims: be informative (quantity), be truthful (quality), be relevant (relation), and be clear, brief, and orderly (manner).
The crucial move was Grice's account of conversational implicature. When a speaker appears to violate a maxim, the hearer does not simply conclude that the speaker is being uncooperative. Instead, the hearer assumes the speaker is still cooperative and searches for an interpretation that makes the utterance consistent with the maxims. The extra meaning thus derived is the implicature. If someone asks whether a colleague is a good philosopher and receives the answer "He has beautiful handwriting," the answer appears irrelevant. But assuming the speaker is being cooperative, the hearer reasons that the speaker must intend to communicate something—perhaps that the colleague is not a good philosopher—that would make the response relevant. The implicature is not part of what the sentence says; it is something the speaker means over and above what she says, and it can be canceled without contradiction. One can say "He has beautiful handwriting, and he is also an excellent philosopher" without contradicting oneself, which shows that the negative assessment was merely implicated, not asserted.
Grice's framework had several profound consequences. First, it provided a principled explanation of how communication can succeed despite the gap between sentence meaning and speaker meaning: hearers use rational inference, guided by assumptions about cooperativeness, to bridge the gap. Second, it made pragmatics a study of reasoning rather than of conventions. Third, it raised the question of how much of what we ordinarily call "what is said" is actually determined by pragmatic inference rather than by linguistic meaning. Grice himself held that what is said is relatively close to sentence meaning, with only reference assignment and disambiguation requiring contextual input. But later theorists argued that the boundary between semantics and pragmatics is far more porous than Grice supposed.
The decades after Grice saw two main lines of development. One line, associated with Stephen Levinson and Laurence Horn, sought to systematize and refine Grice's maxims into a smaller set of principles. Horn proposed a two-principle system: the Q-principle (say as much as you can, given the need for informativeness) and the R-principle (say no more than you must, given the need for economy). Levinson developed a three-principle system involving quantity, informativeness, and manner. These neo-Gricean approaches aimed to explain systematic patterns in lexical semantics and grammatical structure—for example, why "some" often implicates "not all," or why "try to do X" often implicates "did not succeed in doing X." They treated these as default inferences that arise from general principles of rational communication, though they could be canceled in particular contexts.
The other line, associated with Dan Sperber and Deirdre Wilson, developed into Relevance Theory. Sperber and Wilson rejected Grice's maxims as too numerous and too vague, proposing instead a single cognitive principle: human cognition is geared to the maximization of relevance. An input is relevant to an individual to the extent that it yields cognitive effects (new information, strengthened assumptions, revised beliefs) for the effort required to process it. Communication works because hearers automatically seek the interpretation that yields the most cognitive effects for the least processing effort, and speakers design their utterances to exploit this tendency.
Relevance Theory made two radical claims. First, it rejected the Gricean assumption that communication is a matter of following cooperative norms. Instead, it grounded pragmatics in general facts about human cognition: we are built to attend to relevant information, and speakers exploit this fact. Second, it argued that pragmatic inference is involved not just in deriving implicatures but in determining the very proposition expressed. When a speaker says "I've had breakfast," the hearer must decide whether the proposition is about today's breakfast, this morning's breakfast, or breakfast in general. Relevance Theory calls this process explicature—the pragmatic enrichment of the logical form of the sentence into a full proposition. This blurred the Gricean distinction between what is said and what is implicated, since even the basic content of an utterance is now seen as the product of pragmatic inference.
The debate between neo-Gricean and relevance-theoretic approaches remains active. Neo-Griceans tend to preserve a clearer distinction between semantics and pragmatics, treating implicatures as additional meanings derived from what is said. Relevance theorists argue that the distinction is artificial and that all utterance interpretation is a single process of inference constrained by the search for relevance. The disagreement is not merely terminological; it concerns the architecture of the language faculty and the relationship between linguistic knowledge and general reasoning.
A central preoccupation of philosophical pragmatics has been the question of where semantics ends and pragmatics begins. The classical Gricean picture held that semantics delivers a proposition—a truth-evaluable content—and pragmatics adds further layers of meaning on top. But this picture came under pressure from several directions.
One pressure came from the phenomenon of indexicality and context-dependence. Words like "I," "here," "now," and "that" have their reference fixed by context, but this seemed manageable within a semantic framework: the semantic value of an indexical is a function from context to content. The harder cases involved what philosophers came to call unarticulated constituents. If someone says "It's raining," the sentence does not explicitly mention a location, yet the utterance seems to assert that it is raining at some particular place. Where does that location come from? It cannot be part of the sentence's semantic content, since the sentence contains no expression for it. It must be supplied by pragmatic inference. But then the proposition expressed is not fully determined by semantics.
A related phenomenon is semantic underdetermination. Sentences like "You are not going to die" can mean, depending on context, that the hearer will not die from a minor injury, that the hearer will not die in the immediate future, or that the hearer is immortal. The sentence's linguistic meaning underdetermines which proposition is expressed. John Perry and others argued that this underdetermination is pervasive: almost every utterance requires pragmatic enrichment to yield a determinate proposition. This led to what is sometimes called contextualism—the view that the content of an utterance is always a function of context in ways that go beyond the fixing of indexical reference.
Opposed to contextualism is minimalism, defended by philosophers such as Emma Borg and (in a different form) Herman Cappelen and Ernie Lepore. Minimalists argue that there is a minimal proposition expressed by a sentence—roughly, the content determined by its literal meaning and the fixing of genuine indexicals—and that this proposition is the proper object of semantic theory. Pragmatic processes may add further content, but that content is not part of what is said in the strict sense. The debate between contextualists and minimalists concerns the nature of "what is said" and the proper domain of semantics. It has direct consequences for how we understand truth, since if the proposition expressed varies with context in ways not tracked by the sentence, then the same sentence can express different truths in different contexts without any change in meaning.
A further complication came from the work of Charles Travis, who argued that even the most basic predicates are context-sensitive in ways that resist formalization. Whether a surface counts as "flat" depends on the standards of the context; whether a leaf counts as "green" depends on what counts as the leaf's color for present purposes. Travis's examples suggested that the phenomenon of context-sensitivity is not limited to a few indexical expressions but pervades all of language. This radical contextualism threatens the very idea of a systematic semantics that assigns truth conditions to sentences independent of context.
Philosophical pragmatics has always been intertwined with the philosophy of mind, because its central explanatory notion—speaker meaning—is defined in terms of intentions. Grice's analysis of meaning in terms of communicative intentions made pragmatics depend on a theory of mind: to communicate is to get another person to recognize one's intentions, and to understand communication is to recognize those intentions.
This connection has been explored in several directions. One is the question of whether non-human animals or pre-linguistic infants communicate in the Gricean sense. If communication requires reflexive communicative intentions, then much animal signaling—alarm calls, mating displays—does not count as communication in the full sense, since it lacks the intention to have one's intention recognized. Some researchers, following the developmental psychologist Michael Tomasello, argue that human communication is unique precisely because it involves shared intentionality: the ability to coordinate actions and attention with others around common goals. On this view, the capacity for Gricean communication is a distinctively human achievement, rooted in more basic abilities for joint attention and cooperative interaction.
Another direction concerns the relationship between pragmatics and the theory of mind. Understanding an utterance often requires reasoning about what the speaker believes, desires, and intends. This has led some theorists to ask whether pragmatic abilities are a special case of mindreading—the general capacity to attribute mental states to others—or whether they constitute a distinct cognitive system. The question is empirical as well as philosophical, and it connects to research on autism, where difficulties with pragmatic communication are common and are often attributed to impairments in theory of mind.
A third direction concerns the normativity of meaning. Grice's framework treats communication as a rational activity governed by norms of cooperation. But what kind of norms are these? Are they moral norms, rational norms, or merely regularities of behavior? This question connects to broader debates about the normativity of meaning and the nature of rule-following. Some philosophers, following Wittgenstein, have argued that meaning is essentially normative: to mean something by a word is to be bound by standards of correct use. Pragmatics inherits this question in a specific form: when a speaker implicates something, is she bound by norms governing what she has communicated? If a speaker says "He has beautiful handwriting" and thereby implicates that the colleague is a poor philosopher, can she later deny having said anything negative? The answer depends on how we understand the normative force of implicature.
The present state of philosophical pragmatics is characterized by several overlapping research programs rather than a single dominant paradigm. The Gricean tradition remains foundational, but it has been transformed by the debates described above. The semantics-pragmatics boundary dispute continues to generate sophisticated work on both sides, with contextualists and minimalists refining their positions in response to each other. Relevance Theory has developed into a comprehensive cognitive theory with applications beyond philosophy, including in literary studies and psychology. Experimental pragmatics has emerged as a field that tests pragmatic theories against behavioral data, investigating how people actually interpret utterances in controlled settings.
One notable development is the increasing attention to the social and political dimensions of pragmatic phenomena. Work on slurs, dog whistles, and other forms of covert communication has shown that speakers can exploit the gap between what is said and what is communicated to convey content while maintaining deniability. A dog whistle—a term or phrase that carries a specific meaning for a particular audience while appearing innocuous to others—works precisely because pragmatic inference is audience-relative. This raises questions about responsibility for what one communicates, the ethics of indirect speech, and the limits of the Gricean framework in accounting for communication that is deliberately opaque.
Another development is the growing interest in the relationship between pragmatics and the philosophy of language more broadly. The classical view that semantics is primary and pragmatics is a supplementary theory of "extra" meaning has been increasingly challenged. Many philosophers now hold that pragmatic considerations are central to understanding not just communication but also the nature of linguistic meaning itself. If the content of an utterance is always shaped by pragmatic inference, then the idea that words have meanings independent of use becomes harder to maintain. This has led some philosophers back toward a broadly use-theoretic conception of meaning, in which the meaning of an expression is constituted by its role in communicative practices.
The field also continues to grapple with its own foundations. The Gricean analysis of speaker meaning in terms of reflexive intentions has been criticized on various grounds: it may be too intellectualist, attributing to ordinary speakers a level of intentional complexity they do not consciously possess; it may be too individualist, ignoring the social and conventional dimensions of communication; and it may be too narrow, failing to account for forms of communication that do not fit the model of one speaker intending to produce a belief in one hearer. These criticisms have not produced a consensus alternative, but they have kept the field alive to the possibility that its founding assumptions may need revision.
Philosophical pragmatics remains, in its current form, a field defined by a set of questions rather than a single method or doctrine. Its practitioners share a conviction that meaning cannot be understood apart from use, and that the study of language must therefore include the study of speakers, hearers, and the reasoning that connects them. The field's enduring contribution to philosophy has been to show that the gap between what we say and what we mean is not a defect of language but one of its most important features—a space in which speakers exercise creativity, hearers exercise judgment, and the full complexity of human communication comes into view.