A sensorimotor interpretation of Logical Form , and its application in a model of Māori sentences

The guiding hypothesis behind the model is that the logical form (LF) of a sentence reporting some directly observable event in the world can be interpreted as a description of the sensory and/or motor processes involved in experiencing this event. This proposal is described in more detail elsewhere (see e.g. Knott 2012; Takac et al. 2012). We hypothesise that language is intimately connected with the sensorimotor mechanisms through which we experience the world. This idea has received a lot of attention in cognitive science, within ‘embodied’ models of cognition (see e.g. Feldman & Narayanan 2004; Barsalou 2008). But the idea has some interesting implications about language universals that cognitive scientists do not typically pursue. If language is connected to sensorimotor mechanisms, then we expect structural similarities between all languages, because speakers of all languages have the same sensorimotor mechanisms. If language is strongly connected to sensorimotor mechanisms, as many embodied linguists believe, then we should expect a substantial set of structural similarities between languages. Such similarities are clearly not visible ‘on the surface’, so the only way to maintain a strongly embodied model of language is to adopt some linguistic theory that posits cross-linguistic universals at some ‘underlying’ level of structural representation. This argument provides an interesting way of thinking about Chomskyan models of syntax. Chomskyan models take linguistic universals seriously: identifying underlying structures that obtain in many languages is at the heart of the Chomsykan research programme. From this perspective, a Chomskyan account of syntax might provide an ideal vehicle for the expression of ‘strongly’ embodied models of language.