The dissection and structural mapping of Cantonese sentence final particles

This paper explores the possibilities of dissecting the particles that we find in sentence final position in Cantonese into sub-syllabic semantic units. After characterizing the 40-odd most common particles, we study groups of particles, which share one feature (e.g., initial g-) to see what semantic feature they have in common, as well as groups of minimal pairs (gaa3-gaa2 and ge3-ge2) to find out whether there are any correlative systemic semantic differences. We end up proposing 12 minimal semantic units, 4 initials (l, z, l/n, m), 2 rhymes (aa and o), 3 tones (1, 4, 5), 1 coda (k) and 2 such elements incorporating a tone (g3 and aa4). Aside from these, we assume that there is one default rhyme (e) and one default tone (3). In the final part of the paper, all the minimal semantic units are given a place in the structure of the sentence, notably in the CP domain.