ABSTRACT
Relations such as ‘similarity’ or ‘difference’ are abstract by nature. They do not refer to objects, but rather to the mental activities of observing, comparing, and reasoning. In some sense, everything is similar to or is like everything else. Yet words like ‘similar’, resemble’, and ‘like’ and their cognates are useful in normal discourse and are applied automatically. Nevertheless, similarity elicits puzzling logical and philosophical questions. These longstanding issues are explored here by a semantic analysis of the use of similarity words and their meaning relations. The analysis shows that ‘similarity’ is a cluster of notions. The main challenge is to find out what unifies similarity subtypes; in other words, the common mechanism that underlies them and justifies subsuming them under one general, although vague, concept. This chapter shows how similarity words contribute to the flexibility of the conceptual system and its ability to expand. The puzzling nature of the concept of similarity is shown to be rooted in these specific cognitive functions.
