Answers are grounded in this paper and cite their sources.
This study investigates the bèi - gěi passive alternation in Mainland and Taiwan Chinese, drawing on methods inspired by work in comparative sociolinguistics. We specifically investigate a dataset covering n = 1,800 hand-coded tokens drawn from the Chinese Web Corpus 2017. On the technical level, we evaluate (dis)similarities in how writers choose between passives by considering three lines of evidence (constraint significance, strength, and hierarchy) via the random-forest based Boruta method in conjunction with mixed-effects logistic regression, plus Variation-Based Distance and Similarity Modeling on the side. Analysis uncovers a shared underlying probabilistic grammar concerning constraint significance, alongside fluctuations in the relative importance and in effect strength of constraints, particularly among syntactic, semantic, and stylistic factors. On the methodological plane, this paper highlights how an original mix of state-of-the-art corpus-based methods with a computational twist can reveal theoretically important variational patterns that had not previously been identified in two, in principle, well-researched varieties.
No takes yet. Share an insight, caveat, or question.
Guan et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: