ABDUCTIVE INFERENCES IN PRAGMATIC PROCESSES

Abstract

In pragmatic theories, the notion of inference plays a central role, together with the communicative act in which it is activated. Although some scholars, such as Levinson, Sperber and Wilson, propose detailed and accurate analyses of this notion, we will maintain that these analyses can be better systematized if seen through Peirce’s notion of abduction. We will try to maintain that the variety of inferential processes in play in a linguistic act is mostly of an abductive nature. Moreover, we will maintain that the typological tripartition of abductions discussed by Eco (1981) allows to account for a significant part of the mechanisms involved in the comprehension of an utterance, ranging from quasi-immediate and spontaneous levels of understanding to processes that draw on creative resources. In our proposal the vast majority of our linguistic activities implies the automatic retrieval of a habit of action (automatic abduction). In the other cases we need more onerous processes. We might need to identify, among a range of possibilities, the appropriate rule (habit of action/linguistic routine) to be applied to the contextual situation (abduction by selection) or, depending on the context and on our background knowledge, we might be forced to create ex novo a new linguistic routine (creative abduction). In our view, this typology of abductive inferences (Eco, 1981; see also Bonfantini and Proni, 1980) provides us with all the necessary tools to account for the different inferential demands entailed by different levels of the process of language comprehension. On the basis of this typology we can, develop a fine-grained model of linguistic inferences and, thus, simplify the terms of some problematic nodes debated within contextualist approaches.

The benefits of this fine-grained model is that it provides a unitary framework in which all the levels of the process of language comprehension are shown to be inferential and have the same logic structure (the structure of abduction). At the same time, it is also able to account for all the different inferential efforts entailed by different linguistic processes. Thus, this model gives us both a unitary account of the structure of linguistic inferences and a fine-grained description of the differences existing between them.