Revisiting two finiteness phenomena for Mandarin Chinese complementation structures: An empirically oriented approach via systematic hypothesis testing

The literature has debated whether Mandarin Chinese exhibits a finiteness distinction despite the absence of overt tense and agreement marking. C.-T. J. Huang (2022), along with other Generative studies, has re-affirmed this distinction and repeatedly rejected Hu et al. (2001), which presents opposing views. If the finiteness distinction exists, it should be a detectable empirical fact, transcending theoretical perspectives. To empirically test this, we first formulate finiteness as a relative concept using a methodological criterion, which is inspired by Lowe (2019) and potentially cross-linguistically applicable. We devise two testable hypotheses to investigate the distributions of modals (hui/yao) and aspectual markers (le/ guo), which have been seen as having implications for finiteness distinction. After conducting tests on diverse complementation verbs, we have identified empirical deficiencies in past studies. We have shown that it is possible to synthesise insights from apparently opposite camps to form integrated perspectives which can account for a wider range of new empirical data. Under a broader concept of finiteness, the distributions of hui/yao and le/ guo reveal Chinese finiteness as semantic dependency where non-finite complements are more semantically dependent on the matrix clause than finite complements. Overall, we employ a descriptive, evidence-based approach to address the Chinese finiteness conundrum.