A hybrid methodology of linguistic metaphor identification in elicited data and its conceptual implications
This chapter develops a hybrid methodology of linguistic metaphor identification in elicited data, which is a tentative solution to two related problems inherent in much of the research into elicited A is B metaphors, namely definitions of metaphoricity which are based solely on conceptual criteria and approaches to the stability of concepts (possibly) underlying linguistic metaphors which neglect the impact of sentence-level context on categorisation. In order to reveal the idea behind the hybrid model, the chapter is divided into three sections. First, criteria for simple and complex procedures are proposed, which then serve as parameters along which representative methodologies for linguistic metaphor recognition are compared. These juxtapositions make it evident that approaches to metaphor classification form a family-resemblance category, where the multiplicity of perspectives is a norm. To meet this standard, a hybrid approach to elicited data related to educational research is developed. The methodology enables the classification of linguistic examples into five categories, whose stability is taken as an indicator of the informants’ convictions (Low, Chap. 1 this volume, Sec. 4.1). Finally, extrapolations from the linguistic to the conceptual highlight the unique role of attenuated lexico-grammatical categories and the arguable position of conceptual metaphors in motivating people’s beliefs about education.