Cycles and Stability in Linguistic Signaling

Language change is evidenced by a difference between the linguistic expressions of one generation and the next. Given that the output from one generation serves as the input to the next, both language acquisition and use are crucial to the process of change. While the input to acquisition arises from the interplay of many factors, it is not arbitrary. Rather, it is governed by a pragmatic competence that “underlies the ability to use [grammatical competence] along with the conceptual system to achieve certain ends or purposes” (Chomsky, 1980, 59). In the terms of Chomsky (1986), the process of language acquisition maps the E-Language of one generation to the I-Language of the next, whereas pragmatic competence is what governs the mapping from each generation’s I-Language to its E-Language. In the Gricean tradition (Grice, 1975; Levinson, 1983; Horn, 1984), this pragmatic competence has been framed in terms of speakers’ beliefs, preferences, and intentions. Linguistic expressions are used according to a Cooperative Principle whereby interlocutors are taken to make the appropriate contribution to the conversation at the appropriate time towards “the accepted purpose or direction of the talk exchange” (Grice, 1975, 26). This framework deftly captures the systematic relationship between what is said and what is meant, but is clearly an idealization: speakers and hearers might, but need not have the same preferences or goals in a given exchange. This dissertation aims at understanding the consequences of loosening the assumption of Gricean commonality. It will examine how differences in speakers’ and hearers’ preferences might impact the use and acquisition of linguistic signals over time. At the center of this endeavor will be a diachronic process that implicates both use and acquisition, the development in the expression of negation over time known as Jespersen’s Cycle (1917). The main components of this dissertation address the main components of the cycle in turn. First, we consider how the expression of negation transitions from a purely preverbal negator with an optional emphatic postverbal element towards a system with obligatory preverbal and postverbal elements. We present a formal model that derives this transition from even a slight preference for exaggeration on the part of speakers. If