Abstract

1. It must be emphasized at the outset that this paper is concerned with phonemics and not with transcriptions. This is not an attempt to add one more segment to that remarkable product of Natural Selection, the Wade-GilesGardner-Dubs System.' The distinction between transcriptions and orthographies has been clearly discussed by Denzel Carr in connection with the Romaji controversy2 and by others in both practical and theoretical papers devoted to linguistic techniques. The point need not be elaborated here. This investigation was begun primarily with the purpose of discovering the phonemic relationships of the so-called 'palatal series' of initial consonants in the Peiping dialect. It was soon found, however, that a resolution of this problem required a general study of the phonemic system of the dialect.3 The problem itself is simply stated. There exists a set of palatalized initials [ts, tW', s] which always occur before the vowels [i, y] or the corresponding semivowels. There also exist three distinctive sets of initials, respectively alveolars [ts, ts', s], cacuminals [ts, t?', s], and medio-velars [k, k', x], none of which may occur in this position. Although two or more phonetically similar sounds in complementary distribution may usually be assigned to the same phoneme, this situation serves here merely to state the problem, for there are three cases of complementary distribution, each involving the same set of sounds. On the other hand, it certainly cannot be concluded immediately that the members of the palatal series correspond to three independent phonemes, since there is never contrast between this series and any of the other three. This dilemma has long proved to be annoying to students of the language. As long ago as 1814 Marshman was aware of some difficulty when he wrote, 'The character A khee, ... which the Imperial Dictionary and all others, unite in beginning with the initial k, the inhabitants of Pekin sound as though beginning with ch. This however is no wonder: even London has its provincialisms which a good speaker would

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call