View-Neutral Mapping Adjuncts as Default Rules in Metaphor Interpretation

It is generally accepted that much of everyday language shows evidence of metaphor. We assume a common sense knowledge of the real world in the understanding of metaphorical utterances. We do not address the issue of when an utterance is to be considered metaphorical. Instead, we provide an explanation of how a mtephorical utterance can be processed. Consider the following example: (1) “In the far reaches of her mind, Anne knew Kyle was having an affair.” The interpretation of this metaphorical utterance may involve talking of IDEAS AS PHYSICAL OBJECTS (A) and MIND AS A PHYSICAL SPACE (B). These two metaphorical views might provide mappings from “physical manipulation” concepts in the source domain to “conscious mental processing” concepts in the target domain. We could use (A) to conclude that “Anne physically manipulating an object” in the source maps into “Anne consciously operating on the idea” in the target. However, (A) and (B) are of no direct use in dealing with the “far reaches” aspect of (1). Instead of attempting the creation of new mappings to extend an existing metaphorical view (e.g., adding a mapping for “far reaches” to (A)), we employ within-pretence inferencing, which consists largely of reasoning within the terms of the source domain. In our approach, this reasoning takes place in a special, protected computational context that we call the “pretence cocoon”. Avoiding map-extension is a good default because in some cases there may well be no plausible correspondences. We use the term ‘reality’ to refer to the space outside the pretence where propositions are about reality as the understander sees it.