Joint Seminar l Introducing MSDIP: A Method for coding source domains in metaphor analysis

Conceptual metaphor analysis is an important part of many metaphor studies (e.g., Ahrens & Jiang, 2020; Dodge, Jisup, & Sitckles, 2015; Steen, 1999, 2011), because conceptual metaphors are considered to provide a window onto human thinking (Lakoff & Johnson, 1980). Over the years, however, the analytical process of attributing source-domain to linguistic metaphors has received relatively limited attention (Deignan, 2016). In this talk, we introduce a systematic, reliable method for source-domain coding in metaphor analysis. This method is based on the assumption that metaphorical words can have multiple potential source domains that need to be identified.
The method is illustrated through various examples from the Corpus of Historical American English and the Corpus of Contemporary American English. We also report the results of a series of reliability tests to show that the method constitutes a reliable tool for source-domain identification. Finally, we illustrate situations in which the linguistic context or aspects related to genre or time of publication may restrict the range of potential source domains. In such cases, linguistic and/or contextual information thus serve to further specify the source domain.