Bridging corpus diagnostics and EAP pedagogy: A corpus-driven study on lexical hedges in Chinese scholars' spoken...

As the internationalization of higher education intensifies - particularly under China's "Double First-Class" initiative - Chinese scholars are increasingly expected not only to publish in English but also to perform effectively in high-stakes spoken genres such as conference talks, keynote lectures, and doctoral defences. Yet pragmatic features such as hedging remain a persistent barrier, often leading to pragmatic failure in real-time interactions. This study addresses this challenge through an integrated approach that couples corpus-based diagnostics with pedagogical design. Specifically, it investigates the use of lexical verb hedges in Chinese scholars' spoken academic English by comparing a self-compiled Chinese Scholars' Academic Spoken English Corpus (CASEC; 696,009 words; 279 events) with the Michigan Corpus of Academic Spoken English (MICASE; 1,848,364 words; 152 events). Following Granger's reappraised Contrastive Interlanguage Analysis 2.0, MICASE is treated not as a native-speaker benchmark but as one Anglo-American reference variety against which typical L2 patterns can be described. Quantitative analysis revealed that Chinese scholars significantly underused individualized stance markers (e.g., I think) and significantly overused collective, consensus-building markers (e.g., We see). Qualitative analysis interprets these patterns through the lens of cross-cultural pragmatics and genre constraints, suggesting a rhetorical strategy oriented toward collective harmony and the diffusion of authorial responsibility. Crucially, the study moves beyond diagnosis to propose a "diagnosis-to-treatment" pipeline, culminating in a step-by-step Data-Driven Learning (DDL) module complete with full classroom worksheets. The pedagogical design aims to raise scholars' pragmatic awareness and enhance their international academic communication competence.