Recent approaches to bridging: truth, coherence, relevance
Deirdre Wilson, Tomoko Matsui · UCL Discovery (University College London) · 2000
© Deirdre Wilson and Dan Sperber 2012. Introduction Some classic examples of bridging reference are given in (1): In interpreting the italicised NPs the front door, the steak, the hearer has to supply a referent which is not explicitly mentioned in the preceding discourse, but whose existence can be inferred from what has gone before. One way of doing this would be to add the following assumptions to the context in which the italicised NPs are interpreted: Such assumptions may be seen as ‘bridging the gap’ between what has been explicitly stated in the first part of the utterance and what is needed for reference assignment in the second. The need for such assumptions is what mainly distinguishes bridging reference from other types of reference. We will call them bridging assumptions, and the italicised NPs bridging NPs. Early analyses of bridging (e.g. Clark and Haviland 1977; Sanford and Garrod 1981) were conducted in a largely Gricean pragmatic framework, with its machinery of Cooperative Principle and maxims, maxim-violation, inferential intention-recognition, and so on. Bridging assumptions such as those in (2) were treated as Gricean implicatures: beliefs that had to be attributed to the speaker in order to preserve the assumption that she was obeying the Cooperative Principle and maxims in saying what she said. In developing their account of bridging, Clark and Haviland added further maxims to the framework, stating felicity conditions on the use of specific linguistic constructions, whose violation (real or apparent) would trigger the bridging process. For example, the bridging implicatures in (2) would be triggered by violation of a Maxim of Antecedence governing the use of definite NPs.