Quote
C Chiu
Most tone changes are brought about by the effects of adjacent tones of a compound. In other words, the changes are most probably based on phonetical (e.g. 前門 with two adjacent tone 4) and morphological (e.g. tone 4 for first syllable of kinship terms) grounds.
However, for tone changes based on semantical and syntactical grounds, they are applicable even to single characters. For example, 糖 has tong4 (sugar) vs tong2 (candy); 袋 has doi6 (to pocket) vs doi2 (a pocket). Such changes involve those characters that do not constitute a part of a compound.
Can you think of a semantic or syntactical tone change for a 入聲 character, in which this change is observed for an individual character? My hope is that a mid-rising (35) entering-tone would never be necessary to yield a Chinese character, and that any 入聲 character would have a level/even citation tone (7, 8, or 9).
When using Cantonese pronunciations to summon Chinese characters on a computer, is distinguishing 'sugar' from 'candy' even relevant if the result would be precisely the same character? I can understand why you'd want to offer the user both possibilities to maximise flexibility, though.