The semantic content of a declarative statement is its minimal proposition, exhaustively determined by the standing meaning of words and how they are syntactically combined, with occasional contextual enrichment obviously required to disambiguate or assign reference to certain syntactic elements. All other content-effects caused by context-sensitive mechanisms therefore belong to pragmatics. I will start by illustrating how content-causes map onto the minimalist distinction. Then, I will address two objections to minimalism: First, that all literal meaning is context-sensitive, and second, that minimally truth evaluable propositions are epistemically otiose in communication.
The content of an utterance is the total information a person would need to have a complete understanding of it, which typically involves understanding a proposition (or multiple) which the speaker intends to convey. Content-causes, the knowledge or mechanisms involved in linguistic processing, can be divided in to literal or “semantic” causes, and context-sensitive “pragmatic” causes. Consider the sentence “Jill is tall.” Understanding the literal meaning of an utterance of this sentence requires having knowledge of the meanings of the words and how they combine syntactically to form a sentence; in simpler terms, knowledge of English. This sometimes requires narrow context to complete the meanings of certain syntactical elements. For example, to disambiguate a word with multiple meanings (like “lock”), or to assign a reference to indexicals (like “she”). Such syntactically driven content-causes are sufficient for understanding the literal proposition expressed by the sentence (in this case, that Jill is tall) and for assigning a truth value. The literal proposition derived by these processes exhausts the semantic content of the utterance.
However, this is not the only proposition that a hearer may retrieve from an utterance of the sentence. Imagine Jill is 5’5. If “Jill is tall” is uttered at a kindergarten and Jill is a five-year-old, the sentence comes out true. But, if “Jill is tall” is uttered at a WNBA game and Jill is a player on the court, the sentence comes out false. These examples show that the truth assignment of the proposition expressed by “Jill is tall” apparently shifts depending on context, and more than solely linguistic mechanisms are required to fully understand its content; interpreting the utterance depends on things like the setting of the utterance, the average heights of five-year-olds and WNBA players, and so on. If so, “Jill is tall” must express different propositions at different times (“Jill is tall for a five-year-old”; “Jill is tall for a WNBA player”). Such processing, which draws on context beyond the literal meaning of the words and how they are syntactically combined, is pragmatic processing; any proposition determined by these processes therefore belongs to pragmatic content.
The semantic-pragmatic boundary thus drawn implies that the literal meaning of a sentence is not necessarily what a speaker intends to communicate. Irony is one obvious example of this: If Jill is behaving rudely and someone rolls their eyes and utters “she’s nice,” it seems obvious that they don’t literally mean to convey that Jill is nice– the speaker intends to express quite the opposite. The implicated or non-literal meaning (that Jill is not nice) depends on context regarding speaker intentions and therefore is not part of semantic content. But, the literal proposition of the sentence is nevertheless that Jill is nice, only this proposition is not understood to have been meant by the speaker. In Gricean terms, semantic mechanisms in a dialectical context determine the proposition expressed by “what is said,” and pragmatic mechanisms, including knowledge about speaker beliefs and intentions, determine the proposition of “what is meant.”
John Searle objects to the claim that every sentence has a context-free literal meaning that determines a truth-evaluable literal proposition. The literal proposition expressed by what is said is always context-sensitive says he. In a simple case, he argues that the literal meaning of “the cat is on the mat” is context-sensitive because it is only truth evaluable relative to the context of an indefinite set of background assumptions including, for example, that the cat and mat are in earth’s gravitational field, and that the cat is not suspended by hundreds of thin metal wires right above the mat. Minimalists can avoid this objection by including Borg’s tennet that semantic content is totally distinct from speech-act content and therefore that it does not need to be what a hearer takes to have been literally said. Since knowledge about gravitational fields and metal wires belong to the belief-sets of speaker-hearers, they do not affect semantic content. The semantic proposition that “the cat is on the mat” is literally true if and only if the cat is on the mat; any conditions beyond this affect only whether a speaker-hearer believes that the cat is on the mat.
If the context-free minimal proposition is not identical to what is said, then the further objection can be raised that it is epistemically otiose in linguistic communication. Returning to our example, the proposition expressed by uttering “Jill is tall” at the kindergarten is understood by a hearer to be that “Jill is tall for a five-year-old.” Robyn Carston argues that, since this is the proposition taken by the hearer to have been literally stated, then, in Gricean terms, this is the literal meaning of what is said. This renders the minimal truth proposition “Jill is tall” otiose as it is not represented at all by the hearer in conversation. However, Emma Borg explains that the minimal truth proposition can nonetheless play a fallback role in linguistic communication: If for some reason the hearer cannot access the context needed to determine what is said, their knowledge of the language lets them defer to the minimal proposition; “If I know you said ‘Jill is tall’ but I don’t know what comparison you had in mind, I can nevertheless report you as having said ‘Jill is tall’” (p442). Therefore, although distinct from what is said, the semantic proposition can be exhaustively determined by the formal, systematic rules of the language and nonetheless play a role in linguistic communication.
Read:
Ezcurdia, Maite, and Robert J. Stainton, eds. The semantics-pragmatics boundary in philosophy. Broadview Press, 2013.
Nadia Miller 