The contested impact of cognitive linguistic research on the psycholinguistics of metaphor understanding
Problems with Cognitive Linguistic studies
Methodological suggestions
There are several ways that cognitive linguistic work on metaphor can be presented in a way to make this research more amenable to empirical validation (adapted from Gibbs in press).
1. Don’t rely solely on your own intuitions. Systematically ask others, including nonlinguists, for their judgments, and report the degree of similarity of response.
2. Don’t just use made-up examples Examine a large sample of linguistic materials related to the topic or phenomenon of interest. Although individual case studies and the analysis of individual exceptions to broader trends are always interesting and, indeed, often important, determining the extent to which some linguistic phenomenon exists, and its frequency, is critical to establishing the legitimacy of one’s intuitions about some aspect of language.
3. Be as concrete as possible about the exact criteria used for classifying some language item as an instance of some larger linguistic or conceptual category. For instance, describe exactly what criteria have been employed to determine that a particular expression is motivated by one conceptual metaphor, as opposed to another.
4. Acknowledge the context in which linguistic statements appear. Again, cognitive linguistics has a tendency to focus on isolated examples that are either made-up or stripped from their contexts. But just as judgments of grammaticality and semantic anomaly can vary depending on the context in which word strings appear, so too may judgments about matters such as metaphoricity and conventionality alter with the specifics of context.
5. Do not assume that linguistic analyses are isomorphic with individual mental representations. Articulate falsifiable hypotheses about the role of unsconscious conceptual knowledge in language use. Thus, make sure that each speculative possibility mentioned (e.g., a particular blending account of how counterfactuals are understood) can, in principle, shown to be false. By trying to disprove a theory, and failing to do so, this lead to a tentative acceptance of the idea as having a reasonable degree of empirical support.
6. Always specify alternative hypotheses. Most cognitive linguistic proposals do not consider alternatives, which is a special problem when making claims for postulated conceptual entities like image schemas, conceptual metaphors, and the like, as well as hypothetical cognitive processes like conceptual integration.
7. Do not assume that complex meanings require complex mental processes to produce or understand those meanings. People may be able to correctly infer what a linguistic expressions means without having to engage in complex psychological processing because of the familiarity and/or frequency of the individual words and expressions. People may learn and correctly understand the complex meanings of many words and expressions by various means that have little to do with the possible cognitive motivation for such speech within a linguistic community (Langacker 1987). Acknowledge these other forces, such as historical convention, cultural norms, and social context, as relevant reasons for why people speak and understand as they do, and suggest ways of how these factors may even interact with conceptual schemes in explaining realistic linguistic behavior.
Paying close attention to these recommendations would enhance the empirical credibility of cognitive linguistic research on metaphor, at least in the eyes of many cognitive scientists, and likely lead to more experimental studies arising from cognitive linguistic analyses.