Authorial voice in Q1 linguistics and hard science journals: a corpus-based comparative study of stance, engagement, and genre conventions

This study offers a corpus-based comparative analysis of authorial voice in Q1 linguistics and hard science journal articles, examining how stance, engagement, and genre conventions are shaped by disciplinary epistemologies and rhetorical norms. Drawing on Hyland’s stance and engagement model and Swales’s genre theory, the research analyzes 100 IMRAD-structured articles across two corpora totaling over one million tokens. Quantitative frequency analysis and qualitative discourse examination reveal striking disciplinary asymmetries. Linguistics articles exhibit greater rhetorical density and interpersonal alignment, marked by extensive use of self-mention, hedges, boosters, directives, and reader pronouns—constructing a reflexive, dialogic authorial identity. In contrast, hard science writing is characterized by evidential saturation, procedural detachment, and epistemic restraint, foregrounding authorial effacement and methodological fidelity. Genre analysis shows linguistics favors pronounced gap-identification and metadiscursive commentary, while hard sciences adhere to compressed, data-driven exposition. Q1 journal conventions function as discursive gatekeepers, regulating authorial visibility in alignment with field-specific communicative values. The study reconceptualizes authorial voice as a genre-bound, ideologically embedded construct shaped by disciplinary traditions and institutional expectations. Implications are drawn for genre-based writing pedagogy and for understanding voice as a regulated performance of academic identity and epistemic legitimacy.