Adding transparency to the identification of cross-domain mappings in real language data

In recent years much progress has been made in developing systematic protocols for finding linguistic metaphors in authentic language data. The description of conceptual structures, however, has not been placed on equally firm footing. One existing proposal, known as the five-step method, introduces systematicity to the process of determining conceptual structures of metaphors in discourse. However, it does not take sufficient steps to minimize intuition and to maximize transparency. This paper seeks to reduce these weaknesses by introducing the systematic use of dictionaries and a lexical database. The result is a more transparent and constrained method. © John Benjamins Publishing Company.