Page 460 - Encyclopedia of Philosophy of Language

Pragmatics and Speech Act Theory
juridical definition, whereby the basic function of pre- suppositions is to 'establish a frame for further dis- course' (1972:94; trans. CaflS). Presuppositions are based on a mutual, tacit agreement which has not been given before, but which is constantly renewed or revoked during interaction. Presuppositions are grounded on complicity.
After having been the focus of lively discussions on the part of linguists and philosophers of language during the 1970s, presuppositions seemed to have gone out of fashion by the 1990s, and to have main- tained their vitality only in the thin air of certain circles, such as generative grammar, far removed from pragmatic concerns. This decline isjustified only inso- far as more subtle distinctions between different phenomena are suggested, and a descriptively adequate typology of implicata is built. It is less jus- tified if the substitution is only terminological, that is, if the concept of presupposition is replaced by a gen- eral concept such as implicature or inference without any increase in explanatory power. In the latter case, it is not clear what advantages might result from the replacement of the cover term 'presupposition.' The question is whether the term 'presupposition' refers to a range of heterogeneous phenomena or to a par- ticular type of implication to which other types can be
added.
But even before that, it should be asked whether,
between the more credited semantic presupposition and the more recent notion of implicature, there is room for pragmatic presupposition. Before answering this question, the problem of the distinction of the latter from the two mentioned types of adjacent implicata, semantic presupposition and implicature, must be addressed briefly.
1. Relation with Semantic Presupposition
The difference between semantic presuppositions and conventional implicatures is that the latter, unlike the former, are irrelevant to truth conditions.
Unlike semantic presuppositions and some con- ventional implicatures, pragmatic presuppositions are not directly linked to the lexicon, to the syntax, or to prosodic facts (cf. the contrastive accent in 'MARY has come'), but to the utterance act.
Types of implicata recorded by dictionaries and which are part of the semantic representation of a lexeme or are conveyed by given syntactic or prosodic structures, should be distinguished from implicata that are independent of both dictionary and grammar. Pragmatic presuppositions are triggered by the utter- ance and speech act; they are neither the pre- suppositions of a lexeme, of a proposition, nor of a sentence. They are presuppositions that a speaker activates through the utterance, the speech act, and the conversational or textual move which that speech act performs. Thus, pragmatic presuppositions do not concern imperative sentences but orders, not declara- tive sentences but assertions, and so on. In other words, one ought to stress the distinction between syntactic moods and pragmatic functions, where there is no one-to-one correspondence: the same linguistic structure, the same utterance can perform different functions, convey different speech acts and different sets of presuppositions. The acknowledgment of a further theoretical level is the sine qua non for the analysis of pragmatic presuppositions. Pragmatic pre- suppositions are related to knowledge which is not
grammatical but encyclopedic, i.e., concerns our being in the world. Rather, they do not consist in knowledge, in something which is already known, but in some- thing that is given as such by the speaker, in something that is assumed as such and is therefore considered irrefutable (Van der Auwera 1979).
Once the semantic nature of lexical and gram- matical (syntactic and prosodic) presuppositions has been recognized, the connection between semantic and pragmatic presuppositions should be stressed. First, the connection is of a general nature and concerns the obvious (but not to be neglected) fact that, when not dealing with an abstraction but with a real utterance, the phenomenon of semantic presuppositions becomes one of the available means by which the speaker can change the communicative situation. If the link between, for instance, a certain lexeme and the presupposition which is triggered by it is semantic, the latter's analysis within an utterance and a context must face the pragmatic problem of their use and effects.
Second, at a closer distance, the connection regards the specific pragmatic functions performed by phenomena which can be labeled as semantic pre- suppositions. For instance, it is important to recognize the effectiveness of semantic implicata (lexical pre- suppositions, conventional implicatures, pre-
The concept of semantic presupposition
clear. Its parentage is accredited: Frege, Russell, Strawson. Its lineage seems to be traceable back to Xenophanes, quoted in Aristotle's Rhetoric, via Port- Royal and John Stuart Mill. There is a substantial agreement about its definition: the presupposition of a sentence is what remains valid even if the sentence is negated; its truth is a necessary condition for a declarative sentence to have a truth value or to be used in order to make a statement. A respectable test has been devised to identify it—the negation test. There is a list of the linguistic facts (Levinson 1983:181-85) which trigger the phenomenon of pre- supposition: 31 have been listed, from factive verbs (e.g., know, regret) to change-of-state verbs (e.g., stop, arrive), to cleft sentences, etc. Semantic pre- suppositions and conventional implicaturesidentified by Grice (1975), who exemplifies them with therefore, have much in common; both depend on a surface linguistic element which releases the presupposition.
438
is relatively