This project concerns the distinction between semantics and pragmatics, i.e. the question of which aspects of meaning are conventional and encoded in language, and which are pragmatically inferred in context. In this connection, 'scalar implicatures' (SIs) have always been a matter of debate, because on the one hand they seem to be conventionally associated with certain expressions, but on the other they are context-dependent. For example, "some" is said to generally implicate "not all", e.g. in "Some of my friends are Dutch", but there is no such implicature in "If you eat some of her cake, she will be mad". Moreover, the "not all" implicature can be denied "without contradiction" (Levinson 2000: 68), as in "Some, in fact all, of my friends are Dutch". When the implicature is absent (due to the context or due to explicit denial), it is said to be 'canceled'. Cancelability is often evoked as one of the main arguments for regarding SIs as pragmatic (e.g. Geurts 2010: 82; Levinson 2000: 57; Recanati 2010: 152). A second important argument is supplied by the results of psycholinguistic experiments, which often favor pragmatic theories of SIs (e.g. Breheny et al. 2006). In the present project, a 'conventionalist' account of SIs will be developed by questioning the standard arguments for viewing SIs as pragmatically inferred and by reanalyzing scalar implicatures and their cancelation in the light of Anscombre and Ducrot's (1983) theory of argumentation. The development of this new account will imply the formulation of novel hypotheses, which will be tested by means of a detailed corpus study and a series of questionnaires. The project will not only consider frequently studied terms such as "or", "some" and numerals (e.g. "two"), but also other scalar expressions such as "warm" and "possible". Also, it will be verified whether the results of existing psycholinguistic experiments may have been skewed by a neglect of the argumentative aspect of sentences. Thus, the present project brings together two traditions of linguistic analysis that have not very frequently been confronted with one another: the Gricean tradition (initiated by Grice 1975), and the argumentative one (initiated by Anscombre & Ducrot 1983). The conventionalist theory that will be developed not only challenges the standard Gricean arguments against conventionalism, but also invites conventionalists and theories inspired by Anscombre and Ducrot (1983) to reflect on the context-dependency of lexical and argumentative meaning. It will thus constitute an innovative contribution to the ongoing debates on the interface between pragmatics and semantics and on the nature of lexical meaning.
Anscombre, J.-C., and Ducrot, O. 1983. L'Argumentation dans la Langue. Mardaga, Collection "Philosophie et Langage".
Breheny, R., Katsos, N., Williams, J. 2006. Are Generalised Scalar Implicatures Generated by Default? An On-Line Investigation into the Role of Context in Generating Pragmatic Inferences. Cognition 100, 434–463.
Geurts, B. 2010. Quantity Implicatures. Cambridge University Press.
Grice, H. P. 1975. Logic and Conversation. In P. Cole & J. L. Morgan (eds.) Syntax and Semantics (vol. 3): Speech Acts. New York: Academic Press, 41–58.
Levinson, S. 2000. Presumptive Meanings: The Theory of Generalized Conversational Implicature. Massachusetts: The MIT Press.
Recanati, F. 2010. Truth-Conditional Pragmatics. Oxford University Press.