Liu, Wei-feng

Language Center, Ming Chi University of Technology, Taiwan

Abstract: Grammatical metaphor (GM), a characteristic of academic discourse, enables writers to compress complex ideas into compact expressions and construct objective arguments. While L2 writing research has emphasized one GM form, nominalizations, little attention has been paid to how multiple GM types function across rhetorical structures. This study addresses the gap by analyzing GM use in thirty L2 argumentative essays passing advanced level of Taiwan’s General English Proficiency Test. Drawing on Halliday and Matthiessen’s (1999) typology and Hyland’s (1990) move-based model, the study examines how overall GM frequency and GM types vary across paragraphs and rhetorical moves, focusing on their rhetorical functions. Analyses revealed that GM density was highest in Retrospect and Close, whereas Claims relied more on congruent forms. Nominalizations were prevalent across sections, particularly in Retrospect; relators were key in argument development. Adjectives were used to characterize in Support and to persuade in stance-making. Emergent noun modifiers were used strategically in Background and Support; emergent adjectives yielded compact reformulations in Retrospect. These patterns show how GM types are selectively deployed to fulfill rhetorical purposes. By mapping GM use to rhetorical functions across moves, the study offers pedagogical insights for developing discourse-level abstraction and conceptually rich representation in L2 writing.

Key words: grammatical metaphor, rhetorical functions, L2 argumentative essays

Note: This is a pre-proof version and is subject to change during the editing process.