Odors, words and objects

The paper focuses on concepts and words referring to odors and to objects that have an odor. We argue that odors are an interesting object of study since they are evanescent, and since odor words do not refer to concrete and manipulable objects, but to scents evoked by objects. A second reason why odors are interesting is that some languages, as the Western ones, lack a specific odor lexicon, comparable in richness and variety to the color lexicon, and that performance on odors naming is typically worse than performance in color naming. In this work we discuss three main issues. First, we illustrate literature showing that, even if odor words do not have concrete referents, many languages encode them quite easily: the case of odors suggests that word meaning cannot be exhausted by the relationship with a referent, and highlights the importance of the social sharing of meaning. Second, we have discussed the peculiar status of odor concepts and words. Given their ambiguous status, their simple existence poses problems both to theories according to which concrete and abstract concepts do not differ, and to theories according to which they represent a dichotomy. Finally, we present an experiment in which we show that names of objects evoke their smell, and that these smells evoke approach and avoidance movements, in line with theories according to which words are grounded in both sensorial and motor systems.