De-lexicalizing di: The story of how a Chinese lexical item becomes a determiner, and a nominalizer with attitude

A common phenomenon—at least among East Asian languages—is the recruitment of determiners into nominalizers that often also evolve into mood markers (very often sentence-final particles that encode speaker attitude). Examples include Japanese no, Korean –n (geot), classical Chinese zhi and zhe, and contemporary Mandarin Chinese de. The lexical sources of these grammatical/pragmatic markers are often no longer traceable—for example, although geot in Korean is still identifiable as the general noun ‘thing’,-n is a nominal linker whose origin will perhaps forever remain a mystery. Chinese written records, however, provide a telescopic window back into the remote past, and with new computerized databases such as CHANT (i.e., the Chinese Ancient Texts Database), allow more comprehensive and in-depth analysis of language change phenomena. The present paper identifies a once-prolific determiner and nominalizer in classical Chinese, namely di, and using both computerized databases and extant literature, traces the rise of this Chinese morpheme from its lexical origin meaning ‘foundation, base, bottom ’ to its more grammatical use as noun-modifying phrase linker (e.g., genitive, associative and relative clause marker) and eventually as a ‘sentence nominalizer ’ and sentence-final stance particle