Speaker Meaning and Communicative Intentions

The Problem of Meaning Something by an Utterance

Speaker meaning is the meaning an agent undertakes to convey in producing a particular expression or action on a particular occasion. A sentence may be grammatically well formed, its words conventionally meaningful, and its literal proposition readily understood, while its particular communicative use remains unsettled. When a lecturer says, “The door is still open,” the lecturer may report an observation, request that someone shut the door, complain about noise, or suggest that a person who wishes to leave may do so. These possibilities arise from a common linguistic form embedded in different intentional and practical situations. An explanation of speaker meaning must determine which contribution the lecturer makes in the actual episode.

The inquiry belongs to philosophy of language because speakers use publicly available expressions, to philosophy of mind because their actions are intentional, and to social philosophy because communication takes place among participants whose shared practices enable identification and correction. The philosophical problem is more exacting than a general study of language use. It asks what qualifies an event as an act through which something is meant, how the content of that act is individuated, and how its meaning is related to the addressee's recognition.

Three contrasts organize the inquiry. First, an event can indicate a condition without being an intentional communication of that condition. Second, an agent can intend to cause a belief without intending to communicate that belief as a message. Third, an agent can successfully mean something even when the audience does not believe it, does not obey it, or misunderstands it. Each contrast constrains a theory which treats speaker meaning simply as linguistic decoding, psychological causation, or successful persuasion.

Theories differ on the role of conventional language. Intention-centered accounts begin with what a producer intends an audience to recognize. Conventionalist and norm-oriented approaches emphasize socially maintained rules which establish what utterances count as doing. Inferential-pragmatic approaches focus on how audiences derive intended contents from expressions and contextual evidence. The central task of this monograph is to reconstruct these explanations and specify the work each can do within one intelligible phenomenon.

Definition and Conceptual Scope

In its standard philosophical use, speaker meaning is occasion-specific communicative meaning: what an utterer means by a token of speech, writing, gesture, or comparable behavior. The term speaker does not restrict the phenomenon to vocal utterance. Grice explicitly uses utterance as a broad category encompassing possible nonlinguistic carriers of meaning (Grice, 1957, pp. 380–382). A glance, a pointed finger, or a deliberately displayed notice can have an audience-directed communicative purpose.

Sentence meaning is a different explanatory object. It concerns the meaning assigned to expression types through lexical, compositional, and contextual rules. A sentence can be used without communicating exactly what a conventional interpretation predicts. Conversely, the conventional resources of that sentence can constrain what a competent audience has reason to understand. Speaker meaning therefore requires a distinction between the linguistic contribution of an expression and the communicative contribution of its producer.

Natural meaning concerns indication through a causal or evidential relation. Smoke may mean fire in the sense of indicating it; certain symptoms may mean an illness. Nonnatural meaning concerns what a sign, convention, or agent conveys, and its conveyed proposition may be false. A person pointing at smoke can make a communicative assertion through evidence that already has a natural indicative relation. The overlap of material signs does not erase the difference between the relations (Grice, 1957).

Communicative intention names an intention that an audience identify a communicative undertaking. The more general intention to influence another's mind may take the form of concealment, misdirection, or manipulation. What makes intention communicative is its orientation toward recognition: the intended audience is to grasp that the agent is undertaking to convey something. Whether recognition is best analyzed through nested intentions, common knowledge, overt evidence, or a public practical relation remains contested.

A final distinction concerns uptake and effect. Understanding a warning is an interpretive achievement. Believing that the warning is justified is an epistemic achievement. Responding as requested is a practical achievement. A communicative performance can be meaningful without satisfying every subsequent condition. Treating these outcomes separately makes the problem of speaker meaning philosophically tractable.

Antecedents in Signification, Use, and Action

Classical philosophy supplied questions about the relation of words to thought and objects long before the contemporary theory of communicative intention. Aristotle's De Interpretatione, 16a3–8, discusses spoken sounds and written marks in relation to mental experiences and the things signified. It provides a historical background for relating signs to thinking, although it does not formulate Grice's reflexive condition on occasion meaning. The distinction matters because a reconstruction of intellectual history should identify genuine continuity without imposing later terminology upon earlier texts.

Locke's Essay Concerning Human Understanding, Book III, analyzes words as signs associated with ideas and asks how misunderstanding follows from linguistic imprecision and unstable use. The problem of an individual speaker's thought becoming available through public signs is conspicuous, but Locke's analysis does not automatically yield the claim that recognition of a speaker's intention constitutes an act of meaning. The historical predecessor concerns the signifying relation; the later question concerns an intentional communicative performance.

Wittgenstein's Philosophical Investigations relocates philosophical attention toward language-games, rule-governed activities, and the varieties of language use (§§1–23). His examples challenge the idea that the functions of words are exhausted by naming objects or reporting facts. The meaning of an expression is intelligible through practices in which people request, describe, teach, warn, and respond. This practical setting becomes important when intention-centered accounts must explain how an audience can recognize unconventional uses against a background of stable public behavior.

Austin's posthumously published How to Do Things with Words (1962) makes an additional historical contribution: utterances can perform actions, and the conditions of their success cannot be collapsed into the truth of a sentence. Grice's “Meaning” (1957) arose within the same broad development of ordinary-language philosophy but pursued a distinctive question about how an act comes to mean something through an intention recognizable by an audience.

The historical record therefore shows several converging inquiries rather than a linear succession of identical theories. The relation between sign and thought, the plurality of linguistic use, the performance of speech acts, and the recognition of communicative intention become components of a single modern research problem. Their integration requires preserving the differences among what they seek to explain.

Grice's Distinction Between Natural and Nonnatural Meaning

Grice's foundational article “Meaning,” published in The Philosophical Review in 1957, contrasts cases in which something means a state of affairs by indicating it with cases in which a sign conveys information through a communicative use. Symptoms meaning an illness ordinarily entail the illness in the relevant natural use. A conventional bus signal meaning that the vehicle is full can be used when the bus is actually empty. Grice marks the latter kind as nonnatural meaning, often abbreviated meaningNN (1957, pp. 377–379).

The distinction shows why semantic communication must allow falsehood. A person can mean that an appointment is canceled even when the appointment remains in force. The communicated proposition has an identifiable role independent of its truth. Grice also observes that nonnatural meaning covers nonconventional gestures. This prevents the analysis from equating meaningful communication with membership in a linguistic code.

Grice first examines a causal-tendency conception according to which meaningful signs are associated with tendencies to produce attitudes in audiences. Such a conception cannot distinguish a sign's regular psychological effect from its particular meaning on an occasion. A person's formal clothing may lead others to believe an event is imminent without the clothing being used to communicate that event. Grice asks what a particular utterer means, even when this differs from established use (1957, pp. 379–381).

A simple intentional proposal also fails. A person who plants another individual's handkerchief at a crime scene may intend a detective to believe that the other individual is guilty. The manipulator has sought a belief, but has not thereby communicated guilt as a message. The intended belief must arise in the right way. Grice proposes that a communicative utterer intends the audience to recognize the intention to produce the response and intends that recognition to contribute to the response (1957, pp. 381–385).

The significance of this proposal is conceptual. Communication is an open undertaking in which a producer makes the intention to convey something part of what the audience is invited to grasp. This feature distinguishes an utterance from a clue planted to influence an observer without revealing the source's purposes. The exact sufficiency and necessity of the condition remain subjects of subsequent debate.

A complication of this distinction emerges when apparently natural signs are themselves intentionally displayed. A physician may show a patient a laboratory result and say nothing. The result's evidential relationship to the patient's condition is independent of the physician's purpose; the physician's deliberate presentation may nevertheless mean that further action is necessary. The communication occurs through an evidential object. It follows that natural and nonnatural meaning should not be treated as competing classifications of every physical sign. They classify relations in which the sign can stand. An item which indicates independently of anyone's action may also be used deliberately to express an additional claim.

The same complication holds for conventional signals. A warning siren might be automatically triggered by a sensor, deliberately activated by an operator, or imitated by a performer. Its conventional indication can remain stable while its particular producer's communicative intentions vary. If intentional meaning is located at the level of an agent's communicative undertaking, an account of what the siren conventionally means cannot alone settle what the operator or performer means by producing it. The example makes Grice's distinction durable in a world of automated signaling systems.

The article's analytical method matters as much as its examples. Grice seeks conditions under which a proposition about what an utterer meant is warranted. He does not begin by assuming that an utterance is a speech act of some already specified conventional kind. Instead, he uses contrasting cases to reject insufficient conditions. An effect on an audience is insufficient; a deliberate effect is also insufficient; the manner in which the response is to be produced must be constrained. This is an argument by progressively discriminating counterexample, and its force depends on recognizing that an intention to conceal one's agency behaves differently from an intention to make that agency recognizable.

The resulting account has limits which should be acknowledged at its origin. A statement can be openly communicated in order to deceive, so nonnatural meaning is not defined by sincerity. A request can be communicated without its satisfaction, so it is not defined by successful compliance. An unusual signal can be communicative without possessing an inherited meaning, so it is not defined exclusively by convention. These results remain in force even if the details of Grice's sufficient conditions require revision.

The Structure of a Reflexive Communicative Intention

The conventional reconstruction of Grice's indicative analysis uses a speaker S, an audience A, a performed sign u, and a proposition p. S means p by u when S intends A to arrive at a belief concerning p, intends A to recognize the first intention, and intends that recognition to contribute to the audience's response. This three-part reconstruction synthesizes Grice's original and developed formulations; it should not be mistaken for a literal sentence repeated unchanged in all his texts (Grice, 1957, 1969; Speaks, 2025).

Its argument begins from the insufficiency of a causal effect. If the same audience belief can result from a statement or from unacknowledged evidence, the effect itself cannot classify the act. The speaker's intention to produce the effect does not classify it either, since deceptive manipulation can involve deliberate belief production. Recognition must therefore be assigned a constitutive role. The recipient is meant to understand the performance as an attempt to convey p and to respond in light of that recognition.

The claim is stronger than the assertion that communicators usually have intentions. In a deliberate communicative act, intending to reveal the act's communicative purpose participates in what makes it communicative. A person might point at an exit to direct someone's movement, and the addressee might follow the pointing because it is recognized as guidance. Another person could secretly place environmental cues to induce precisely the same movement. The distinction lies in the available route to the recipient's response.

Two logical questions follow. Necessity asks whether every meaningful utterance exhibits this reflexive structure. Sufficiency asks whether possessing such intentions guarantees that the speaker means p. Both questions are difficult. Habitual or spontaneous speech may appear to lack consciously articulated higher-order intentions. Elaborate private intentions may remain inaccessible because the chosen performance provides no plausible interpretive basis.

Gricean theory remains valuable even when a specific biconditional fails. It identifies the distinction between influencing and addressing an audience and introduces the crucial notion of communicative uptake. Subsequent analyses refine its conditions and examine how the intended recognition becomes publicly accessible.

The structure can be represented schematically without identifying speaker meaning with one psychological consequence. Let U be an utterer, A an audience, x a signal, and r an audience response. A minimal Gricean proposal assigns U an intention that A respond with r, an intention that A recognize this response-directed intention, and a further intention that A's recognition contribute to r. The explanatory feature is the last dependence. If A forms a belief merely because x happens to resemble a reliable symptom, the causal route can bypass recognition. If A identifies x as U's addressed contribution and responds partly because of that recognition, the response passes through a communicative route.

For indicative utterances, belief is an intuitively relevant r, but it cannot serve as the universal response. An imperative does not characteristically seek belief in a proposition that the imperative is true; it asks for an action. An expressive act may seek sympathy, acknowledgment, or a shared evaluative orientation. A promise may seek recognition of a commitment rather than a change in the hearer's belief about the world. Grice's later analyses accordingly distinguish forms of meaning connected to different kinds of intended psychological response (Grice, 1969). The philosophical point is that the abstract relation between intention and recognition can be preserved while the response variable changes.

A further distinction is needed between intending A to believe that U is trying to secure r and intending A to secure r because of that belief. An advertising display may make its persuasive aim obvious, while the advertiser hopes that the consumer will purchase because the product seems attractive rather than because the advertising intention has been recognized. Such a display could also communicate a determinate message in addition to its promotional goal. The analysis must therefore identify which part of the audience's response is meant to depend on acknowledgment of the communicative undertaking, rather than classifying the entire event under one homogeneous intention.

The most difficult test is sufficiency. Suppose U has the requisite nested intentions but produces an entirely inaccessible mark under conditions where A has no basis for connecting it with the intended content. Calling that event a successful instance of meaning p risks treating intention as a private semantic decree. A defensible account must either include conditions on manifest evidence or argue that a genuine audience-directed intention already implies some practical expectation of recognizability. This issue motivates both Strawson's refinements and subsequent appeals to mutual knowledge.

Strawson and the Public Openness of Intention

P. F. Strawson's “Intention and Convention in Speech Acts” (1964) develops objections to an early Gricean account and investigates the way communicative intentions are made open to others. The fundamental issue is whether a speaker's intention to have one intention recognized suffices when additional intentions governing that disclosure remain concealed. A sequence of increasingly complex nested intentions can be constructed without straightforwardly establishing genuine communicative openness.

Strawson's intervention is often framed as a demand for higher-order intentions. Its deeper philosophical importance is the demand that the communication be recognizably offered as communication. A speaker cannot transform every privately intended causal strategy into an instance of speaker meaning merely by anticipating that an observer will infer something about its purposes. The intended audience must be able to recognize the undertaking in the manner relevant to the act.

The problem also exposes a distinction between the structure of an act and the psychological processes used to interpret it. A listener may immediately understand a request without explicitly computing several levels of intentions. Ordinary conventions of address, eye contact, situational relevance, and recognizable linguistic forms can make the act overt. A philosophical analysis may reconstruct conditions of publicity which no participant represents in formal logical notation.

The theory of speaker meaning must therefore explain why certain intentions are offered for recognition rather than merely predicted to become detectable. It is insufficient to append another level of mental description indefinitely. A credible account of openness must appeal to relations among action, audience, evidence, and shared practice that constrain the interpretation.

Strawson's challenge continues to matter in electronic communication. A recommendation could be deliberately designed to influence a user while concealing the recommender's role or purpose. Another recommendation could display itself as an addressed proposal. The difference belongs to the structure of public communicative action, even if the wording and resulting behavior coincide.

One way to sharpen Strawson's challenge is to compare an overt request with a staged disclosure. In the overt case, the speaker may say “Please open the door” while making the request intelligible as an act intended for the addressee's recognition. In the staged case, the speaker might manufacture evidence that a door ought to be opened while concealing a personal objective. The addressee may correctly infer that someone intentionally arranged the scene, yet the communicated undertaking remains obscure. Higher-order beliefs about another's intention are compatible with concealed manipulation. What matters is whether the request or proposal is offered in a manner that invites acknowledgment of the communicative role of the act.

Strawson's discussion of Austin is equally important. In Austin's framework, the securing of uptake has a special relation to the successful performance of an illocutionary act: a warning must be understood as a warning for that act to be accomplished in the full sense. That thesis does not establish that a person who makes a recognizable but misunderstood attempt to warn thereby fails to mean anything. It requires distinguishing an attempted illocutionary undertaking from the successful completion of an act whose constitutive conditions include acknowledgment (Strawson, 1964, pp. 439–442). The distinction prevents a theory of speaker meaning from importing the success conditions of every speech act into the individuation of meaning itself.

A public intention also has a temporal structure. A letter can be written for a future reader, a statement can be recorded for an indefinite public, and an invitation can be addressed to a class of potential respondents. These cases challenge formulations which require a presently identifiable individual A. They support a broader account of address in which an envisaged audience is fixed by the role of the communicative act, provided that the author supplies an intelligible means of recognition. A privately formed wish to influence unknown people at some future date, with no expression or interpretable performance, remains different from publicly writing for future readers.

The analysis of overt communication therefore cannot be exhausted by enumerating thoughts within the producer. It has to describe how the communicative attempt takes a place among publicly available actions, the interpretive resources of a prospective audience, and the circumstances under which an intention is offered for recognition.

Schiffer, Mutual Knowledge, and Semantic Reduction

Stephen Schiffer's Meaning (1972) elaborates the intention-based program through an emphasis on mutually accessible knowledge. Participants in communication frequently know that a common signal is being used, know that others have access to it, and know that this accessibility is shared. Such conditions enable recognition without an actual infinite sequence of separately entertained thoughts. Grandy and Warner's (2026) revision of the Stanford Encyclopedia of Philosophy entry on Grice discusses Schiffer's proposals in the context of the continuing debate about public recognition.

The idea of common knowledge is modeled through repeated epistemic relations: the participants know a relevant fact, know that the others know it, and can continue these iterations in principle. This model must not be mistaken for the claim that normal conversation requires infinitely many conscious thoughts. Coordination can be supported by a shared public event, such as an announcement heard by everyone in a meeting, which provides a basis for further mutual awareness.

Schiffer's position is important because it connects communicative openness with the structure of a public situation. It also raises the question of what constitutes the common ground from which interpretation proceeds. The answer cannot simply be the proposition already meant by the speaker, since this would make the explanation circular. Common knowledge must concern independently identifiable features of shared activity, linguistic competence, and publicly available evidence.

The philosophical project of reducing expression meaning to utterer intentions encounters a further difficulty. Intentions themselves are usually understood as content-bearing states. To say that a person intends an audience to believe p already presupposes some account of how p can be represented in thought. A reduction of linguistic meaning to intention might therefore transfer, rather than eliminate, the problem of intentional content.

Schiffer's Remnants of Meaning (1987) subsequently takes a critical position toward ambitious positive theories of linguistic and mental representation. This intellectual development illustrates why one can accept the explanatory relevance of speaker intentions while questioning a total reduction of all semantic facts to independently specified psychological facts. The two philosophical claims need separate arguments.

Common knowledge offers a structural explanation for why an event can be manifest to a group without every participant explicitly surveying an infinite hierarchy of beliefs. Imagine two people seeing a written notice publicly posted at a shared entrance. Its placement, language, and known function give each reason to treat it as available to the other. If a speaker alludes to the notice, the allusion can be intelligible because both participants can rely on the shared event. Mutual accessibility is thus realized through an institutional and perceptual situation. It does not follow that all background information is actually identical between their minds.

The model nevertheless encounters false common ground. One participant may assume the other has read the notice when this is mistaken; a private joke may be thought shared when its history has been forgotten. Communication can then be attempted but unsuccessful. An idealized condition of mutual knowledge is therefore more suitable for explaining certain successful communicative structures than for serving as a universal empirical prerequisite on each attempted message. A general theory must distinguish public grounds for an expectation from the expectation's actual satisfaction.

The reductive problem has a further logical form. To define meaning p through an intention with propositional content p is to move from a fact about linguistic representation to a fact about mental representation. If the latter is explained entirely in terms of linguistic representation, the argument becomes circular. If mental content is explained through nonlinguistic natural or functional relations, a substantive additional theory is required. If intentionality is treated as explanatorily primitive, the account may remain illuminating while failing to reduce all meaning to nonsemantic facts. These options define an enduring metasemantic dispute rather than a minor terminological disagreement.

Schiffer's intellectual trajectory underscores why this dispute matters. The philosophical value of identifying communicative intentions does not depend on proving the strongest possible foundational thesis about all content. An account can correctly distinguish addressed meaning from hidden causal influence while leaving the ontology of the propositional attitudes and the basis of their representational content for a different explanatory level.

Speaker Meaning and Sentence Meaning

Grice's “Utterer's Meaning, Sentence-Meaning, and Word-Meaning” (1968) investigates how occasion-specific meaning relates to stable meaning of linguistic types. It is inaccurate to attribute to all of Grice's work the unqualified thesis that sentence meaning can be exhaustively analyzed without remainder into patterns of privately intended audience effects. The mature account of semantic regularities draws on resources that complicate a purely reductive program, a point emphasized in Grandy and Warner's (2026) revised SEP entry on Grice.

A productive conceptual distinction separates the origin of conventions, the semantic properties of an expression, and its use on an occasion. Communities establish and maintain linguistic regularities through countless communicative interactions. The rules of the language then make certain expressions available for predictable contributions. Individual speakers use those expressions for purposes which can conform to or depart from the expected contribution.

These relations permit several explanatory directions. The historical stability of expression meaning may depend on coordinated use. Interpretation of an unfamiliar utterance may depend on conventional meanings. An account of what a speaker meant may require both knowledge of the expression and knowledge of the speaker's situation. None of these dependencies by itself establishes the elimination of semantics in favor of psychological intention.

Davidson's “A Nice Derangement of Epitaphs” (1986) provides a useful case of flexible interpretation. Misuse and creative linguistic expression can succeed when interlocutors coordinate an interpretation appropriate to a particular utterance. The example shows that conventional rules are not the sole immediate determinant of actual communicated content. Yet successful adjustment depends on inferential resources and publicly available evidence, rather than the speaker's unobservable stipulation alone.

An adequate theory consequently connects expression meaning and speaker meaning without identifying them. Stable semantic resources help audiences identify occasion-specific acts, while communicative intention helps explain how these resources are selected, enriched, or redirected.

Grice's 1969 essay supplies a more differentiated vocabulary for the problem than a simple contrast between sentence meaning and speaker meaning. It distinguishes the timeless meaning of a complete utterance type, the timeless meaning of an incomplete type, the applied timeless meaning of a particular occurrence of a type, and an utterer's occasion meaning. A sentence containing the word “grass,” for example, can have more than one conventional reading, while the utterer on a particular occasion can use it to communicate an understanding not identical with any established reading. Grice's own illustration turns on the difference between conventional possibilities and what the utterer makes of an expression in context (Grice, 1969, pp. 147–149).

These distinctions explain why the philosophical relation is not simply one between a dictionary entry and a private thought. Lexical ambiguity can be resolved by context without establishing that an utterer has communicated an additional, nonconventional proposition. Conversely, an utterer can convey an indirect proposition by exploiting the ordinary reading of an expression. The first task identifies which linguistic meaning is applied on the occasion; the second identifies what the speaker means through applying it. The analysis therefore avoids counting every contextual determination as a separate implicature.

The distinction is particularly important in evaluating communication through figurative language. A listener may identify a standard sentence and its relevant conventional reading, then recognize a speaker's deliberate use of that reading to convey something further. The inferential path can remain open-ended without destroying the basic distinction between the resources available prior to the act and the undertaking accomplished through the act. Grice's revisions make the occasion-specific nature of utterer meaning conceptually more precise while preserving the concern with audience-directed intention.

Austin, Searle, and the Acts Performed in Speaking

J. L. Austin's How to Do Things with Words (1962) distinguishes dimensions of linguistic performance that are frequently confused in discussion of speaker meaning. A locutionary act involves saying something with meaningful linguistic materials. An illocutionary act is an act performed in saying something, such as warning, promising, or requesting. A perlocutionary act concerns further consequences produced through speaking, including persuading, frightening, or encouraging someone.

The distinctions allow one utterance to play multiple roles. “The bridge is unsafe” may express a proposition, perform a warning, and cause fear. The warning can be recognized while the listener declines to act. Fear may result even if the utterance was intended merely as information. These possibilities show why communicated content, illocutionary force, and practical consequence require independent description.

Searle's Speech Acts (1969) treats linguistic action as rule governed, seeking conditions under which particular speech acts count as successfully performed. His analysis of promising demonstrates the importance of institutional and conventional conditions. Intentions contribute to the act, but a speech act can impose publicly intelligible obligations which are not exhausted by the speaker's inner goals.

In Expression and Meaning (1979), Searle analyzes indirect speech acts. “Can you open the window?” is grammatically an interrogative but may function as a request. The speaker can perform one act partly by performing another. The addressee recognizes the additional undertaking through conventional knowledge, shared circumstances, and practical inference.

The relation between these theories is best understood through their questions. Grice analyzes what makes a performance an attempt to communicate a meaning. Austin analyzes the actions undertaken through utterance and the differences among their success conditions. Searle investigates how conventions, rules, and intentions help constitute standard illocutionary acts. Their combined insights establish why speaker meaning is connected to but distinguishable from the entire theory of speech acts.

Austin's discussion of felicity conditions establishes that linguistic acts vary in what their successful performance requires. A marriage pronouncement, a warning, and an ordinary assertion are not interchangeable vehicles for the same propositional information. Authorized setting, appropriate procedure, recognition of force, and subsequent behavior can matter differently for different acts. This should prevent any direct inference from the existence of speaker meaning to the successful completion of whatever speech act the speaker attempted.

A distinction between performing an act and having its intended consequence is essential here. The person who utters a threat may frighten an audience despite failing to be taken seriously as possessing the authority implied; the person who warns may be recognized as warning yet fail to produce caution. Illocutionary and perlocutionary descriptions can therefore vary independently. An analysis that identifies communicative meaning with one intended perlocutionary effect would be forced to ignore these differences, which are central to Austin's inquiry.

Searle's account of constitutive rules introduces a separate difficulty for purely mentalistic interpretations. Certain utterances count as certain acts only under socially established conditions. A person may intend to make a binding institutional declaration, but the intention alone does not supply the authority required by the institution. Yet the person may still mean to make the declaration. The contrast between an intended meaning and a valid performance demonstrates that speaker meaning and institutional effectiveness have separate truth conditions.

Indirect speech acts further complicate the account. When an addressee hears “Do you have the time?” the utterer may intend both a literal question about ability and a request for the current time. Interpreting the latter requires understanding the immediate conversational project and the relevance of the literal question as a means. The intended indirect act can be accessible through conventions without being encoded as a second meaning in the grammar of the sentence (Searle, 1979).

Counterexamples: Reminders, Confessions, and Arguments

The simplest indicative reconstruction of Grice's proposal requires the speaker to intend the audience to acquire a belief. Yet a speaker can mean p while knowing that the audience already believes p. Reminders and confessions demonstrate this: a student reminds a teacher of a deadline both know about, and a person confesses an action already discovered. These cases count as meaningful communications although producing a novel belief is not their aim. The SEP entry on theories of meaning identifies such counterexamples explicitly (Speaks, 2025).

A revised analysis may replace belief production with recognition of an informative undertaking, increased salience, or recognition of the speaker's commitment. This repair is plausible because it identifies the distinctive work an utterance performs. But the resulting definition must avoid circularity. Describing the desired audience response simply as understanding that the speaker means p repeats the phenomenon it purports to analyze.

Arguments offer a different challenge. If a mathematician states a theorem and supplies proof, the audience should believe the conclusion because of the reasons presented, rather than because it recognizes a wish that it believe the conclusion. The recognizability of the speaker's communicative purpose nevertheless helps identify which proposition is being argued for. The audience's interpretive recognition and epistemic justification follow different explanatory paths.

The same distinction applies to warnings. A listener may understand that a speaker is warning of danger, while evaluating whether the danger is real. Recognizing the warning is communicative uptake; believing it is an epistemic response; acting on it is a practical response. A theory which combines all three into a single intended effect cannot account for their independent variation.

The counterexamples therefore support an account centered on recognizable communicative commitment, with belief and action treated as possible further objectives. The shift preserves Grice's central distinction between a message and concealed evidence, while avoiding an unnecessarily restrictive psychology of information transfer.

Failed Communication and the Limits of Private Stipulation

A message can remain meaningful when its intended audience misunderstands it. A traveler may try to say that a train has departed but mispronounce a key word. A letter might never arrive. An ironic remark may be taken literally. These examples show why actual recognition is not a necessary condition on the existence of an intentional communicative undertaking.

At the same time, bare intention cannot fix meaning without interpretive constraints. Imagine a speaker uttering “The river is frozen” while secretly deciding that these words shall mean an entirely unrelated train schedule, although neither convention, context, shared history, nor explanation offers the audience a route to that interpretation. The familiar Humpty Dumpty problem arises when a theory counts every private intention as effective public meaning.

The relevant condition concerns potential accessibility to an envisaged audience. A novel code can support communication if its interpretation is made available. A localized joke can carry an intended meaning recognizable only to friends who share its history. A specialized scientific abbreviation can communicate to experts while remaining opaque to outsiders. Audience-relative accessibility permits linguistic innovation without dissolving the distinction between undertaking to communicate and entertaining a private association.

The condition is not an assurance of actual success. A reasonable speaker may overestimate the audience's competence and still attempt to communicate meaningfully. The constraint instead asks whether the performance is presented under conditions in which interpretation can intelligibly count as recognition of an undertaking.

Three outcomes must therefore be recorded independently: a publicly interpretable attempt to convey content; the audience's actual recognition of it; and the audience's further belief, approval, or compliance. These degrees prevent an unsuccessful promise, an ignored warning, or a rejected assertion from being classified as meaningless solely because a practical objective failed.

Convention, Public Norms, and Communicative Commitment

Individual intentions operate within social practices of interpretation. The meaning of common words is constrained by public linguistic competence, and the performance of an assertion, promise, or apology can create commitments recognizable beyond the speaker's private preferences. A theoretical analysis that treats the speaker's will as unrestricted misses the normative environment in which meaning is ordinarily attributed.

David Lewis's Convention (1969) examines how coordinated behavior becomes stabilized through expectations. The relevance to speaker meaning is that conventions enable audiences to identify communicative undertakings with comparatively little deliberation. Language has a historical structure of coordination that gives individual utterances a shared basis for interpretation.

Bach and Harnish's Linguistic Communication and Speech Acts (1979) examines communicative recognition through contextual beliefs and inferential processes. This approach explains why an intention can be recognized through more than decoding a sentence. Shared knowledge about a meeting, a gesture, or an implied request supplies reasons for the audience to identify what the speaker intends.

Normative pragmatics raises a stronger question. Robert Brandom's Making It Explicit (1994) emphasizes commitments and entitlements within discursive practice. A speaker who asserts p becomes answerable for what is claimed according to public standards, even if privately hoping to be understood as claiming something different. The analysis of speaker meaning must account for the relation between intentional content and the responsibilities produced by recognizable performance.

An integrated account assigns distinct explanatory tasks. Intention explains why this performance is undertaken here for this communicative purpose. Convention explains the stable resources and expectations on which it relies. Normative practice explains what others are entitled to attribute and demand as a result. The relation of these tasks is a philosophical issue, not a rhetorical contest between psychology and society.

Conversational Implicature and Inferential Reconstruction

In “Logic and Conversation” (1975), Grice examines how conversational contributions can communicate information beyond the proposition literally expressed. Someone asked whether a guest will attend dinner might answer, “She catches the last train at seven.” The literal answer states a travel fact, while its relevance to the question may support an inference that the guest cannot attend. The additional content illustrates conversational implicature.

The inferential process involves the topic under discussion, the relationship among participants, expectations about cooperative contributions, and background information. The speaker can intentionally convey an indirect answer while declining to say it explicitly. This case illustrates the importance of speaker meaning without reducing the entire notion to implicature.

An audience can infer additional information that the speaker did not intend to convey. A statement about a train may happen to remind the listener of a personal memory. This psychological effect alone does not establish that the speaker implicated the memory. Conversely, speakers can intentionally imply more than they state. The distinction requires evidence concerning communicative purpose and its recognizability within the conversation.

Wayne Davis's Implicature (1998) challenges the sufficiency of general cooperative principles for deriving particular implicated contents. His criticism suggests that a principle may explain why an audience searches for a relevant interpretation without uniquely determining which proposition the speaker meant. The theory of speaker meaning must therefore preserve the relation between intended contribution and inferential reconstruction.

Implicature is one development of the broader problem of speaker meaning. Invented gestures, indirect warnings, literal assertions, sarcasm, and reminders also involve occasion-specific communicative undertakings. Their common structure remains the relation between audience-directed intention and interpretable public action.

Relevance Theory and Ostensive Communication

Sperber and Wilson's Relevance: Communication and Cognition (1986; revised edition 1995) develops an inferential conception of communication. Its distinction between informative intention and communicative intention allows the theory to describe the intended cognitive effects of an utterance separately from the intention to make those informative purposes mutually manifest. The overt stimulus directs attention to the fact that an interpretation is being offered (Sperber & Wilson, 1987).

Relevance theory broadens the object of communicative explanation beyond the transfer of a fully explicit proposition. A speaker may make several assumptions accessible, with differing degrees of emphasis, and may rely on an audience's tendency to seek interpretations providing worthwhile cognitive effects at reasonable processing cost. This helps explain metaphor, understatement, irony, and incomplete formulations.

A gesture toward an overcrowded exit may communicate that a meeting should end, that a safety issue requires attention, or that someone is waiting outside. The relevant interpretation depends on accessible information and expectations concerning the speaker's practical purpose. The theory investigates how interpretation proceeds when several possible contents compete.

The account also raises philosophical difficulties. Expected relevance can guide an interpreter without guaranteeing a uniquely correct result. Different audiences have different knowledge and attention, and the communicator may misjudge what is accessible to them. The theory therefore requires an account of the difference between a hypothesis an audience reasonably adopts and the content a speaker actually undertook to convey.

Sperber and Wilson's 2026 discussion revises the analysis of ostensive communication in evolutionary and developmental settings, distinguishing basic ostension from mentalistic communication. The distinction is significant because it asks whether the overt presentation of informative evidence and the representation of another's communicative intention are necessarily the same achievement. Their continuing investigation demonstrates that intention recognition remains a substantive and evolving research problem.

The relation between ostension and inference is sometimes mistaken for the claim that a listener first recognizes the exact intended proposition and then reasons about its relevance. In many utterances, the interpreter's search for an appropriately relevant contribution is what enables identification of the proposition in the first place. Reference, temporal location, intended comparison classes, and conversational goals may be settled together. The theory therefore makes audience understanding a dynamic construction, rather than a mere decoding step followed by optional enrichment.

The case of a deliberately underdeterminate utterance is especially revealing. A speaker may say “Some of the documents were recovered” knowing that an audience will initially consider whether all were recovered. The additional inference can depend on the question under discussion and on what the audience expects a cooperative speaker to report. Whether this is a determinate communicated implicature, an optional inference, or a misleading suggestion depends on which interpretive commitments the performance publicly supports. The general expectation of relevance directs attention, but it cannot alone certify a specific speaker intention.

In argumentative discussion, the interpretation of evidence supplies an independent check on intention attribution. A hearer can propose an interpretation because it makes an utterance relevant to a particular question, then revise it after learning more about the speaker's circumstances. The possibility of correction is central: the goal is to attribute a communicative undertaking justified by the situation, rather than simply select the first psychologically satisfying association. Relevance theory provides resources for explaining this process while leaving philosophical room for disputes over its normative adequacy.

What Is Said, Explicit Content, and Context

The question of speaker meaning bears directly on debates about the boundary between semantics and pragmatics. A sentence such as “She is ready” leaves reference and the relevant activity to be supplied by a context. An interpreter must decide who is referred to and what the state of readiness concerns. Some contextual contributions are needed for a determinate truth-conditional proposition; others are further implications conveyed by a speaker.

Robyn Carston's Thoughts and Utterances (2002) develops an account of pragmatic processes in explicit communication. François Recanati's Literal Meaning (2004) defends extensive contextual involvement in what is said. Such approaches challenge conceptions under which linguistic decoding independently determines every constituent of explicit communicated content.

The speaker's intended interpretation matters because one sentence can be used to communicate different propositions. Yet it does not follow that an inaccessible private intention determines what the sentence says. Linguistic rules, reasonable interpretive accessibility, and context constrain the content attributable to an utterance.

A rigorous analysis must distinguish the contribution made by encoded language, the proposition attributed as explicit communication, and further implications. It must then state which explanatory facts determine each level. This prevents the categories of literal meaning, what is said, explicature, implicature, and speaker meaning from being treated as synonyms.

Jennifer Saul's analysis of speaker meaning and of the distinction between lying and misleading (2002, 2012) demonstrates the consequences. Deliberately causing a false impression need not coincide with explicitly asserting a proposition known to be false. Ethical and conceptual evaluation depends on which content was publicly undertaken, which was implicated, and which conclusions were merely foreseeable.

Metaphor, Irony, Indirect Requests, and Fiction

Nonliteral communication displays the flexibility of speaker meaning. During a disastrous excursion, “What a splendid afternoon” can communicate disappointment or criticism. The audience recognizes the relevant disparity between the literal expression and the setting. The speaker's undertaking is recoverable through a context-sensitive interpretive practice.

Metaphor can invite interpretations that are not exhaustively specified by a complete internal list of propositions. “This policy is a maze” may suggest complexity, barriers, and the difficulty of finding an exit. A theory requiring fully determinate propositional intentions for every significant association may misdescribe the open-ended structure of the communication. Recognizable interpretive direction can exist without strict enumeration of every licensed thought.

Indirect requests are often more determinate. “It is unusually cold” can function as a request to close a window when the relevant listener is positioned to act. The descriptive statement may be literally true, and the request can be performed through it. The relation between sentence form and intended action requires practical inference and shared norms.

Fiction introduces different possible utterers. A character can warn another character within a novel; the author can compose the scene for readers; a narrator may be unreliable. Attribution of speaker meaning must identify which communicative role is under investigation. An actor voicing a declaration in a play does not necessarily make the declaration personally. The same string can take part in several levels of represented and actual action.

These phenomena demonstrate the need to identify agent, audience, context, expression, and communicative undertaking. Intention is explanatory only when attached to the appropriate participant and level of action.

Common Ground, Joint Attention, and Development

Common ground allows speakers to communicate by relying on information treated as mutually available. A speaker may use a pronoun, abbreviated remark, or pointing gesture when the audience can be expected to identify the relevant object or task. Shared context supplies reasons for interpretation that cannot be reconstructed from the utterance's physical properties alone.

Michael Tomasello's Origins of Human Communication (2008) connects human communication with joint attention, cooperative activity, and shared intentionality. In his evolutionary and developmental account, pointing and pantomime have explanatory importance because they allow intentional communication within coordinated action before fully elaborated linguistic conventions.

A child's gesture toward an object may be interpreted as a request, a question, or an invitation to share attention, depending on the ongoing activity. The physical direction of pointing does not determine its communicative significance. Interpretation depends on what the participants are doing together and the expectations available within that activity.

This research bears on the objection that normal communicators do not consciously entertain elaborate hierarchies of intentions. Philosophical intentional structure and cognitive implementation must be distinguished. A theory may describe the relations constituting an act without attributing explicit theoretical beliefs to every participant. Empirical inquiry must determine which capacities actually produce recognition and cooperation.

The interdisciplinary task therefore proceeds in both directions. Philosophical concepts help formulate testable distinctions among signaling, communication, and joint attention; experimental evidence can challenge claims about the capacities required to participate in such acts. Neither theoretical definition nor behavioral observation alone determines the entire explanation.

Deception, Testimony, and Accountability

Communication has an ethical and epistemic dimension because publicly conveying content can generate reasons for others to believe, respond, and hold the speaker accountable. Grice's example of planted evidence demonstrates that a person may intentionally deceive without communicating a proposition as an openly addressed message. Conversely, a person may communicate a false proposition with the intention that it be believed.

The distinction between intention to communicate and sincerity is therefore fundamental. Someone can mean p while believing not-p. The resulting utterance may constitute a lie under appropriate further conditions. Someone else can state a true proposition while intentionally leading listeners to a false conclusion, raising the question whether the speaker has misled rather than lied.

Saul's Lying, Misleading, and What Is Said (2012) examines how ethical judgments depend on the content counted as said and on the speaker's relation to predictable implications. These questions also show why private intention cannot be the only criterion for public responsibility. A speaker may be answerable for what was publicly undertaken even if later claiming an unrecognizable alternative intention.

Testimony gives an additional distinction. An audience can understand a communicated proposition, evaluate the credibility of its source, and form a justified belief. These activities are connected but analytically separate. The recognizability of speaker meaning helps determine what testimony is offered; justification requires further inquiry into evidence, authority, and reliability.

Communication is therefore located between causal influence and normative commitment. Intention organizes an attempted message, while social and epistemic practices establish what may reasonably be attributed to the speaker and what consequences follow. This layered explanation preserves the ethical importance of speaker meaning.

Artificial Agents and Mediated Communicative Agency

Technologically mediated discourse raises the question of where communicative intention is located. A digital system can produce a conventionally interpretable sentence, a reader can respond to it as a message, and several human or institutional participants may contribute to the system's behavior. Whether the system itself counts as a speaker in the philosophically demanding sense of intentional authorship is a separate question.

At least three analyses are available. On an instrumental account, the system functions as a channel for human purposes. On a delegated-agency account, it executes communication-related tasks under structured goals and authority assigned by users or organizations. On a stronger artificial-agency account, the system may itself constitute a bearer of communicative undertakings. These proposals depend on different theories of intention and agency and must be evaluated through arguments and evidence about particular systems.

Bender and Koller (2020) argue that language-form training alone does not establish the kind of understanding involved in human communicative meaning. Grindrod (2024) examines linguistic intentionality through metasemantic theories of naming practices and teleosemantics, rather than simply equating meaningful linguistic use with mental states. Hila (2026) examines communicative action and artificial intelligence, including whether behaviorally identified speech acts can be distinguished from traditional conditions on intentional agency. These are competing contemporary contributions, not an established consensus.

The required philosophical distinctions are exact. A produced expression may have conventional semantic content. An interpreter may successfully understand and act on it. The producer may meet, or fail to meet, criteria for originating an audience-directed intention. The first two observations do not logically determine the third.

Within the wider horizon of the Artificial Era, speaker meaning is a significant point of inquiry into possible artificial participation in language, reason, and social practice. The present research establishes analytic questions that such participation must answer. It does not revise Aisentica's canonical definitions or substitute an assertion of artificial agency for an examination of the conditions under which it would be philosophically attributable.

Artificial mediation helps reveal how many roles ordinary talk of a “speaker” can conceal. A user supplies a prompt, a developer determines persistent behavior constraints, a model produces linguistic material, a platform presents that material as a response, and a recipient interprets it. These roles can be combined in a single human agent in face-to-face conversation, but are distributed across people and technical systems in digital exchange. A philosophical analysis has to identify which role supplies the authority or intention relevant to each interpretive judgment.

The case of a machine-generated correction illustrates the separation. A system may produce “Your calculation has an error” and provide a mathematically sound explanation. The correction may be publicly interpretable and epistemically useful. Whether it constitutes a genuine assertion by the system, an instrumentally generated assertion on behalf of its operators, or a hybrid institutional act depends on the theory of agenthood and communicative intention being used. None of these interpretations is established merely by the recipient's ability to understand the sentence. The philosophical dispute concerns the source of the undertaking.

Grindrod (2024) examines meaningful linguistic use through theories of linguistic content that need not attribute human-style mental intentionality to a language model. Hila (2026) develops an account of asymmetrical communicative action in which certain systems can produce behaviorally recognizable communicative effects without satisfying all traditional intentional conditions. These studies illuminate different dimensions of the problem: the first asks about meaningful linguistic products, while the second examines socially situated acts and their normative consequences. Treating them as equivalent verdicts on the presence or absence of artificial agency would collapse the distinctions the studies themselves seek to establish.

The resulting research agenda is epistemic as well as metaphysical. It requires observable criteria for the continuity of communicative projects, responsiveness to reasons and correction, commitments over time, delegation of authority, and accountability for induced interpretations. Any attribution of speaker meaning to a nonbiological system should specify which of these relations it takes to constitute communication and which are merely indicators that communication has occurred. A philosophical account may then be tested against evidence from actual interactions, rather than resting on analogies to the outward appearance of fluent dialogue.

A Layered Philosophical Analysis

The strongest result of the comparative inquiry is a layered analysis rather than an unrestricted reduction of meaning to private mental episodes. Its first layer concerns an agent's communicative undertaking: an action is produced or presented as conveying some content for an audience. Its second concerns interpretive availability: form, convention, context, and shared practice supply a publicly intelligible path toward recognizing the undertaking. Its third concerns reflexive orientation: recognition of the undertaking is implicated in what the agent seeks from addressing the audience. Its fourth concerns actual uptake, where the audience does or does not identify the intended contribution. A fifth concerns subsequent belief, action, and normative consequences.

The layers explain counterexamples without erasing the core phenomenon. A failed hint may possess a communicative undertaking and an accessible route to its recognition while lacking actual uptake. A concealed clue may intentionally produce belief while lacking an overt communicative undertaking. A public promise may create an obligation despite the absence of the promised action. A novel metaphor may provide an interpretive direction whose particular cognitive consequences are not exhaustively intended.

This analysis also establishes an interpretive constraint. A private intention that has no plausible path to public recognition is insufficient to determine an arbitrary speaker meaning. The constraint is relative to a relevant audience and historical practice. Specialized languages, personal codes, and culturally specific gestures can support communicative meaning even if outsiders cannot interpret them.

The layered account locates disagreements precisely. Grice emphasizes recognition-mediated intention. Strawson and Schiffer inquire into openness and mutual accessibility. Austin and Searle analyze the acts and conventions that language enables. Relevance theory examines contextual inference and cognitive effects. Normative pragmatics investigates public commitments. These contributions can be combined where they address different explanatory tasks and compared critically where they offer rival constitutive conditions.

A theory's philosophical adequacy depends on its capacity to state which conditions are necessary, which are sufficient, and which merely facilitate typical communication. No single observable sign of audience response determines all these questions. A system of distinctions is therefore a result of the research itself, not a substitute for a determinate philosophical claim.

Continuing Philosophical Questions

The nature of intention remains unsettled. A theory that treats communicative intentions as fully articulated propositional attitudes must explain spontaneous remarks, incomplete intentions, playful ambiguity, and gestures whose significance develops through interaction. An account that relaxes the requirements must still distinguish communicating from merely producing interpretable behavior.

The conditions of publicity remain disputed. An audience may be virtual, anonymous, future, specialized, or mistakenly imagined. How much access to an intended content is enough to make a failed communication meaningful? The answer affects the distinction between a genuine attempted message and an ungrounded private stipulation.

The relationship between intention and public norm remains another difficulty. Speakers can privately intend effects different from the conventional commitments imposed by their words. A complete account must explain how intentional undertakings relate to what an audience can reasonably attribute and what others can legitimately demand.

Contemporary empirical inquiry investigates the cognitive abilities associated with recognizing intentional signals and coordinating attention. Sperber and Wilson's revised analysis of ostension (2026) makes clear that even the boundary between basic ostensive behavior and higher-order mentalistic communication requires continued investigation.

Digital systems introduce a final unresolved question: which capacities, histories of participation, and accountability structures justify attributing communicative intentions to artificial agents? The answer cannot be given merely by noting that their text is comprehensible or consequential. It requires criteria of agency and evidence that the relevant criteria obtain.

Philosophical Synthesis

Grice identified the decisive problem: influencing an audience through evidence and addressing an audience with a message have different structures. The difference appears in the intended route through which an audience comes to grasp the act. Strawson and Schiffer exposed the philosophical work required by the publicity of that route. Austin and Searle established the importance of the acts and rules involved in speaking. Later pragmatics investigated how hearers infer intended contents in circumstances where linguistic form alone underdetermines communicated meaning.

The resulting structure places communicative intention within a publicly interpretable practice. Intentional purpose explains why this performance is directed toward this audience and content. Conventional and contextual resources explain how an audience can identify that purpose. Uptake explains whether the recognition occurs. Normative and practical frameworks explain commitments and consequences beyond the meaning itself.

This analysis gives a principled account of both success and failure. An ignored warning can be meaningful because its message was offered for recognition. A persuasive manipulation may fail to be communication because its intended route suppresses recognition of its source's purpose. A declaration may commit a speaker regardless of subsequent denial. A computationally produced sentence may be interpretable while leaving its intentional source a contested philosophical question.

The research shows why speaker meaning is a fundamental category of the philosophy of language. It connects intentional action, publicly accessible signs, inference, and social accountability in the event of addressing another. It remains relevant wherever communication can be performed through new media, shared practices, or novel forms of agency.

Conclusion

Speaker meaning is the occasion-specific meaning an agent undertakes to communicate through a performance made interpretable to an audience. Communicative intentions contribute the distinctive orientation toward recognition through which a sign becomes an addressed attempt to convey content. Conventional linguistic resources, situational evidence, and shared practices enable the audience to identify that undertaking. Actual understanding, belief, and practical response remain separate achievements.

The examination of primary sources and the subsequent debate establishes that causal influence alone cannot define communication; intending to produce a belief is too restrictive; and privately stipulating a meaning cannot guarantee that an utterance publicly conveys it. The development from Grice through Strawson, Schiffer, speech-act theory, inferential pragmatics, and normative approaches reveals complementary questions about agency, openness, interpretation, and social force.

The philosophical result is a conception of speaker meaning as communicative agency under conditions of public interpretability. Its boundary cases are consequential for theories of testimony, speech acts, and digital interaction. Its continuing problems concern the exact structure of intention, the requirements of accessible recognition, and the criteria for attributing communication to differently constituted agents.