ABSTRACT
It is natural to ask, meanwhile, in the light of the results obtained by Rosch and Berlin, whether verbs display any of the same properties as the type of noun we have looked at. Do they, that is, map wholly or partly onto a taxonomic structure? If so, do the relations between categories and their members show the prototype effect, and could similar reasoning be produced to show why this should be so? If this turns out to be the case it would be an interesting result, for there is no a priori reason to expect it to be so: after all, although the categories named by the nouns we have looked at are by and large observable concrete entities, the categories - of events or actions rather than objects - named by verbs are far more complex and abstract, depending on factors which are often institutional or cultural, or to do with hypotheses about personal motives and intentions. One man's taking the life of another, for example, can be classified as 'murder' if done intentionally, as 'execute' if carried out as an officially sanctioned act of revenge in the appropriate circumstances, or as 'manslaughter' if done accidentally. There is usually nothing intrinsic to or observable in the nature of the act itself to determine which of these it will be.
