Context-Limited Simulation and Metaphor

Abstract

The traditional theories of metaphor discussed in Chapter 2, like the theories of language and cognition on which they draw, reflect the influence of propositional logic and computer programming (Barsalou, 1999a; Brandt, 2005; Gibbs, 2006). These approaches assume that thought is accomplished by algorithms, similar to programs that run on digital computers. Perceptions are translated into signals that are transmitted to the brain, where they are translated again into an abstract, code-like, non-representational ‘language of the brain’, and processed by rule-governed algorithms, based on formal logic. The output of these algorithms is then translated into signals that are transmitted to the motor-control system for further translation into actions. In these models, cognitive processing is separate from both perception and action, and emotion is antithetical to reasoning (Barsalou, 1999a; Clark, 1997; Brandt, 2005). Categories are likewise well-defined and organized in a neat hierarchy, with precise rules (truth-conditions) for determining category membership.