Complexity of grammatical metaphor: an entropy-based approach

Abstract Grammatical metaphor in M. A. K. Halliday’s sense has long been extensively investigated by researchers in terms of theoretical and empirical studies. Regarding the empirical studies, they have predominantly employed observed or normalized frequencies of grammatical metaphor to uncover its distribution in different text types. Few studies, however, were conducted to quantitatively examine the complexity of grammatical metaphor in that no indicator presently is proposed to measure the degree of complexity in grammatical metaphor. This paper targeted at investigating the feasibility of implementing entropy to measure this complexity. The findings demonstrate that the entropy value could be afforded as an indicator of complexity of grammatical metaphor to efficiently uncover the way that disciplines of different registers build knowledge by means of employing such linguistic features as grammatical metaphor. This research is significant in that it sheds light on the quantitative comparison of complexity of grammatical metaphor in text analysis.