Multi-Dimensional Regularity Analysis: How the Entrenchment-and-Conventionalization Model can be applied to corpus data

Abstract

Abstract This paper introduces and illustrates a methodology for analyzing corpus data with the goal of identifying regularities in the use of a given element or pattern. In line with the research tradition in construction grammar, one type of these regularities pertains to form-meaning and/or form-function pairings. In addition to this, however, further types of regularities are explored, among them the reciprocal type of regularity that links communicative goals and linguistic forms as well as contextual and social regularities. I pursue three main goals: 1) to lay out the conceptual groundwork of the methodology; 2) to present a manual for applying it; and 3) to illustrate its application with a case study on the sequence in here. The motivation for developing this methodology, labelled Multi-Dimensional Regularity Analysis (MDRA), lies in the dynamic, flexible and multi-dimensional notion of linguistic conventions and conventional constructions proposed as part of the Entrenchment-and-Conventionalization Model (Schmid 2020).