NSM analyses of the semantics of physical qualities: 'sweet', 'hot', 'hard', 'heavy', 'rough', 'sharp' in cross-linguistic perspective

All languages have words, such as English 'hot' and 'cold', 'hard' and 'soft', 'rough' and 'smooth', and 'heavy' and 'light', which attribute qualities to things. This paper maps out how such descriptors can be analysed in the natural semantic metalanguage (NSM) framework, in terms of like and other semantic primes configured into a particular semantic schema: essentially, touching something with a part of the body, feeling something in that part, knowing something about that thing because of it, and thinking about that thing in a certain way because of it. Far from representing objective properties of things "as such", it emerges that physical quality concepts refer to embodied human experiences and embodied human sensations. Comparisons with French, Polish and Korean show that the semantics of such words may differ significantly from language to language.