On the continuous debate about discreteness

Abstract The issue of discreteness vs. continuity comes into play in all domains of linguistic analysis and at multiple levels. The distinction’s experiential basis is discussed, as well as various means of discretization and continuization. Most phenomena are sufficiently complex that treatments emphasizing discreteness and continuity both have some validity—it is not a matter of choosing between them, but of determining what each contributes and how they relate to one another. These notions are applied to a number of specific problems, including genetic relationships, constituency, constructions, grammaticality, and grammatical categories. Special attention is devoted to the network model of complex categories. Like any metaphor, it can be misleading if pushed too far. An alternative is proposed which arguably represents a more appropriate mixture of discreteness and continuity.