Cross-linguistic Metaphor Intelligibility between English and German

Conceptual Metaphor Theory (CMT, Lakoff & Johnson 1980; Lakoff, 1983, 1987, 1993, 2008, 2009), the most prominent cognitive approach to metaphor comprehension, argues that the nature of interconnections within the conceptual system is inherently metaphoric-analogical and that systematic patterns in linguistic metaphor reveal these cognitive interconnections. Relevance Theory (RT, Sperber & Wilson, 1986; Wilson & Sperber, 1993; Sperber & Wilson, 1995; Wilson & Sperber, 2002, 2004) and Graded Salience (GS, Giora, 1997, 1998, 1999, 2003; Peleg et al., 2008; Peleg & Giora, 2011) disagree that systematic patterns in linguistic metaphor can be taken as direct evidence of their cognitive representation. A metaphor consists of two concepts, a source and a target concept. The metaphor implies an analogy between the two concepts. To comprehend a metaphor is to infer under which conditions the implied analogy holds. The meaning of the two concepts is pragmatically enriched by these additional assumptions. Metaphor comprehension is an inferential process. The result of this process is the enriched meaning of the metaphor. This meaning can become conventionalised, in which case it often serves as an inferential shortcut: instead of having to consider all conceptually possible interpretations and their plausibility in the context of the analogy, speakers who are familiar with the conventional (i.e. idiomatic) meaning are provided with a default interpretation. According to CMT, the inferential process is a process of interconnecting primary embodied concepts to ever more complex higher-order concepts. On this view, a metaphoric idiomatic meaning is such a complex concept where the conceptual interconnections are conventional. According to RT, the inferential process is a process of inferring a meaning that is in line with the speaker’s communicative intent, the discourse context, and interlocutors’ expectations of the cognitive relevance of potential inferences. On this view, metaphoric idiomatic meanings are highly salient inferences with a high degree of contextual relevance because speakers’ expectations of relevance are conventionalised. According to GS, the inferential process consists of two modules that work in parallel: a module that infers salient meanings based on linguistic knowledge and a module that enriches the meaning by taking non-linguistic knowledge such as conceptual, experiential, perceptual, contextual, and world knowledge into consideration. On this view, metaphoric idiomatic meanings are highly salient inferences because of speakers’ knowledge of non-conceptual linguistic conventions. This thesis investigates the claims made by CMT, RT and GS by experimentally testing the cross-linguistic communicability of metaphoric proverbs with idiomatic meanings. Proverbs are selected such that the similarity of metaphors’ source and target concepts, expectations of contextual relevance, and the degree of familiarity with proverbs’ conventional wording is cross-linguistically maximised. If CMT is correct, then when cross-linguistic conceptual similarity is maximised in this way, monolingual native speakers should find L2 languagespecific metaphors communicable. If RT and GS are correct, then monolingual native speakers should find L2-specific metaphors less communicable than L1-specific and non-languagespecific metaphoric proverbs because they lack knowledge of the necessary non-conceptual linguistic conventions. Cross-linguistic metaphor communicability is measured in three ways in the experiments: (1) through reading/response times, (2) through plausibility judgements, and (3) through a context creation task. Results show that cross-linguistic metaphor communicability of L2-specific metaphors is lowered for monolingual native speakers on all three measures.