The formation of the kě 可 and kě yǐ 可以 constructions

This paper aims to explain the formation of the kě and kě yǐ constructions in archaic Chinese. We analyze a number of examples from the pre-Qin era to refute previous hypotheses that the kě construction is formed by adding kě to a notional passive, fronting the object in an active sentence including kě, or solely by reanalysis. Subsequently, a verb-moving-backward hypothesis is proposed: kě is used in the underlying structure ‘V-O’ + kě to comment on an already known proposition ‘V-O’, then V is moved to the end to avoid the top-heavy problem. Similarly, this hypothesis also accounts for the kě yǐ construction: kě in the underlying structure ‘yǐ-X-V-Y’ + kě is to comment on the serial verb structure ‘yǐ-X-V-Y’, which is interchangeable with ‘X-yǐ-V-Y’ forming ‘X-yǐ-V-Y’ + kě where ‘yǐ-V-Y’ is moved after kě to avoid the top-heavy problem. Moreover, the “verb moving backward” hypothesis provides new insights into the formation process of similar constructions (e.g., nán 难 ‘be difficult to V,’ yì 易 ‘be easy to V,’ zú 足 ‘be sufficient to V’ constructions) in ancient Chinese, as well as the study of tough constructions.