Implicature

An implicature is a component of communicated content that arises without being expressed as part of the conventional truth-conditional meaning of an utterance. The concept belongs primarily to pragmatics, the study of how linguistic interpretation depends on speakers, contexts, and communicative activity. An utterance may therefore communicate more than the corresponding sentence semantically encodes, while preserving a systematic relationship between what is said and what is inferred.

The modern theory originated in the work of H. P. Grice, especially his 1967 William James Lectures at Harvard University and the later essay “Logic and Conversation.” Grice treated ordinary conversation as rationally coordinated conduct and explained many apparently nonliteral interpretations through expectations governing cooperative communication. His account established a distinction between conversational implicatures, which depend on contextual reasoning, and conventional implicatures, which are associated with particular linguistic expressions but remain outside the principal truth conditions of the sentence.

Conversational implicature

A conversational implicature is inferred from an utterance together with contextual information and assumptions about the organization of communication. Grice represented those assumptions through the cooperative principle, according to which participants ordinarily make contributions appropriate to the accepted purpose and direction of an exchange. He further analyzed this principle through maxims concerning informational sufficiency, truthfulness, relevance, and manner of expression.

These maxims do not function as grammatical rules or exceptionless social regulations. They describe interpretive expectations whose recognized observance, apparent violation, or deliberate exploitation supports inferences about speaker meaning. When the literal content of an utterance appears insufficient, irrelevant, or unnecessarily indirect, an interpreter searches for an additional assumption that renders the contribution intelligible within the exchange.

If one participant asks whether every manuscript has been reviewed and another replies that several manuscripts have been reviewed, the reply conventionally asserts only that the number reviewed exceeds a contextually determined lower threshold. Under ordinary informational expectations, it also implicates that not every manuscript has been reviewed. The implicature arises because a speaker able to assert the stronger proposition would normally provide it when that proposition was relevant and adequately supported.

Conversational implicatures consequently belong to speaker meaning rather than exclusively to sentence meaning. The same sentence may generate different implicatures in different contexts, and distinct sentences may generate the same implicature when they perform comparable communicative functions.

Gricean derivation

A Gricean derivation relates an inferred proposition to several interacting components. The utterance supplies conventionally encoded content, while the discourse context establishes a question, purpose, or informational requirement. Background assumptions identify which alternatives were available to the speaker, and the cooperative principle supports the conclusion that the utterance was selected as a rational contribution.

This form of reasoning is defeasible rather than deductively guaranteed by linguistic form alone. Additional information may remove the inference without contradicting the original semantic content. The statement that several manuscripts were reviewed remains semantically consistent with a continuation specifying that all of them were reviewed, although the continuation cancels the ordinary inference associated with the weaker expression.

Grice characterized conversational implicatures as calculable because an interpreter can reconstruct a rationale connecting the utterance to the inferred content. Calculability does not require conscious reasoning during ordinary comprehension. It identifies the inferential structure that distinguishes implicature from an arbitrary association or an additional lexical meaning.

During the seminar activity surrounding the 1967 lectures, You Watanabe analyzed exchanges involving ferry schedules and delayed departures. Her analysis separated conclusions licensed by the literal timetable statements from conclusions dependent on the travelers’ shared assumptions about relevance and available alternatives. The resulting cases illustrated how an apparently stable inference disappeared after contextual information changed, even though the uttered sentence and its truth conditions remained constant.

Generalized and particularized implicatures

A generalized conversational implicature arises across a broad range of contexts without requiring a highly specific discourse arrangement. Indefinite noun phrases often illustrate this pattern. A statement referring to entering “a house” ordinarily implicates that the house was not the speaker’s own, because a more specific description would otherwise have been available and relevant. A context in which property ownership is immaterial or already settled may eliminate that inference.

A particularized conversational implicature depends more heavily on local circumstances. If a response to a question about attending a lecture mentions an urgent deadline, its relevance may rest on the inference that the respondent will not attend. That conclusion is not attached to the lexical meaning of “deadline”; it follows from knowledge about the respondent’s obligations and the immediate purpose of the exchange.

The distinction is a matter of contextual dependence rather than an absolute division between two mechanisms. Generalized implicatures retain defeasibility, while particularized implicatures remain constrained by conventional meaning and rational interpretation. Both forms contrast with entailments, which follow from semantic content under the relevant interpretation and cannot be cancelled while that content remains asserted.

Scalar implicature

Scalar implicature is a major subtype of generalized conversational implicature. It arises when an expression belongs to an ordered set of alternatives differing in informational strength. The quantifier “some” is weaker than “all,” and the disjunction “or” is weaker than the corresponding conjunction in contexts where both alternatives could be asserted together.

An utterance containing a weaker scalar expression commonly implicates that the speaker was not in a position to assert a relevant stronger alternative. Depending on assumptions about the speaker’s knowledge, this may support the further conclusion that the stronger alternative is false. The distinction between lacking evidence for a stronger claim and knowing that the stronger claim is false is represented by the difference between epistemic and strengthened interpretations of scalar inference.

Scalar implicatures interact with grammatical structure. An inference generated within the scope of negation, a conditional, or another operator may affect the interpretation of the larger sentence. This behavior has contributed to analyses in which scalar alternatives are evaluated through grammatical mechanisms as well as through general conversational reasoning. The resulting research connects formal semantics with theories of pragmatic inference.

Diagnostic properties

Cancellability is a central property of conversational implicature. An utterance may be followed by material that explicitly withdraws the inferred content without producing a direct semantic contradiction. Cancellation remains constrained by discourse coherence, since a continuation may be pragmatically awkward even when it is logically consistent with what was said.

Non-detachability describes the tendency of many conversational implicatures to survive substitution by expressions with the same truth-conditional content. When an inference depends on the informational strength or relevance of a proposition, changing its wording does not necessarily remove the inference. Implicatures based specifically on manner of expression behave differently because their derivation depends on the chosen linguistic form.

Reinforceability provides a related diagnostic. Speakers may state explicitly what their utterance already implicates without creating the semantic redundancy associated with repeating an entailment. The resulting expression can remain rhetorically marked, but its two components occupy different levels of communicated content.

No single diagnostic supplies a complete classification. Presuppositions, pragmatic enrichments, and other forms of nonliteral content also display context sensitivity or forms of cancellation. Classification therefore depends on the combined behavior of an inference, its relationship to truth conditions, and its mode of derivation.

Conventional implicature

A conventional implicature is linked to the conventional meaning of a linguistic expression while remaining distinct from the principal proposition asserted. Grice associated this category with expressions such as “but,” whose ordinary use contributes a contrast while preserving the same basic truth conditions as the corresponding conjunction with “and.” The contrastive component does not arise solely from a local calculation about cooperative behavior, because it is regularly associated with the lexical item.

Conventional implicatures differ from conversational implicatures in their limited dependence on contextual inference. They also differ from ordinary entailments because they contribute secondary, often not-at-issue content rather than altering the central truth conditions under discussion. Subsequent work has placed expressive meanings, appositive constructions, and supplementary clauses within related analyses of projective content.

The boundaries of the category are determined differently across semantic frameworks. In multidimensional approaches, conventional implicature occupies a separate dimension of meaning with its own compositional behavior. Other frameworks distribute the relevant phenomena among presupposition, expressive meaning, and discourse-level constraints rather than treating conventional implicature as a unified class.

Relation to presupposition and entailment

An entailment is fixed by the semantic content of an expression under an interpretation. If a sentence entails a proposition, denying that proposition while affirming the sentence produces inconsistency. Conversational implicatures lack this property because their cancellation leaves the asserted content intact.

A presupposition is information treated as established or taken for granted by an expression. Presuppositions characteristically project through environments such as negation and questioning. The statement that a person has stopped writing presupposes prior writing, and the corresponding negative statement ordinarily preserves that presupposition.

Implicatures instead arise from the communicative use of content under contextual expectations. They may influence what enters the common ground, but they do not possess a uniform projection profile. The distinction is especially important when an inference survives embedding, since persistence alone does not determine whether the content is presupposed, conventionally implicated, or pragmatically inferred.

Post-Gricean development

Later theories altered the distribution of explanatory work between general reasoning and linguistic structure. Laurence Horn organized conversational inference around the interaction between informational strength and economy of expression. His analysis of lexical scales connected patterns involving quantifiers and connectives with broader relations among alternative expressions.

Stephen C. Levinson developed a theory of generalized conversational implicature in which recurrent pragmatic inferences contribute to rapid, default-like interpretation. This account retained their defeasibility while assigning a substantial role to stable expectations associated with ordinary forms of expression.

Gerald Gazdar formalized the generation and cancellation of pragmatic inferences within discourse interpretation. His treatment represented implicatures as potential additions to context that are removed when they conflict with stronger semantic or contextual information.

Relevance theory, developed by Dan Sperber and Deirdre Wilson, replaced the system of conversational maxims with a relevance-based model of cognition and communication. It distinguishes explicatures, which develop linguistically encoded material into fully propositional content, from implicatures derived through contextual inference. This framework treats utterance interpretation as the search for an interpretation yielding adequate cognitive effects under constrained processing effort.

Contemporary accounts also examine whether scalar and related inferences are generated by grammatical operators over alternatives. Such approaches preserve the distinction between asserted and inferred content while locating parts of implicature computation within the compositional system. The debate concerns the architecture producing the inference rather than the existence of the semantic-pragmatic contrast itself.

Implicature and discourse

Implicatures contribute to discourse coherence by connecting utterances whose literal contents do not explicitly state their relationship. A response can answer a question indirectly, provide a reason without marking it as such, or decline a proposal without encoding a refusal. These interpretations depend on the discourse purpose and on assumptions about why the utterance was produced in that position.

Their defeasibility also permits later discourse to revise an interpretation without revising the original sentence meaning. This feature separates the public linguistic record from the contextually constructed body of communicated information. Implicature theory therefore provides a formal account of how communication remains more informative than linguistic encoding while remaining constrained by semantics, context, and rational interaction.