※ 引述《mankiller (English teacher is me)》之铭言： : morphophonemic rules提到英文单字单复数的变化方式 : 有[-s],[-z],和[-iz] (第三个那个schwa打不出来...= =") : 有两条rules，R1>R2 之前老师有讲过好几次 : 但我有一点点忘记原因是为什麽...课本也有一点看不太懂... : 而且要举例子说明好像要举结尾是sibilants的字，才有办法prove R1<R2 : 有没有哪位大大知道原因的吗? 为何R1要先於R2? R1: an epenthesis rule that inserts a schwa between sibilants 0 → schwa / C # __ C [+strident] [+strident] R2: a voice assimilation rule that makes the suffix agree in [voiced] with the last segment of the preceding word /-z/ → [αvoiced] / [αvoiced] # __ R1 must take the priority to apply, and thereby it can bleed the application of R2, which would otherwise make the suffix devoiced and derive the unattested allomorph *[-is], if the sibilant preceding it is also voiceless, as illustrated below: /baks-z/ /baks-z/ R1 baks-iz R2 baks-s R2 -- R1 baks-is output [baksiz] output *[baksis] : 另外还有一个问题...是这个月初去考元智的语概问题， : 题目大致上是这样: : Please explain why in English the indefinite article 'a' is : used before words that begin a consonant sound,and why'an'is : used before words which begin with a vowel sound. : 感谢大家!! :) Suppose the base form of the indefinite article is /an/, the choice of allomorph falls out naturally as a result of the unmarkedness of CV syllable. Specifically, the coda nasal of /an/ is dispreferable in terms of the notion that CV syllable is the most unmarked syllable type. It follows that this nasal coda should be truncated except when followed by words that begins with a vowel sound, in which case it has been resyllabified as the onset of the following syllable. Note however that CVC syllable is in fact widespread in English, so an alternative is to assume the indefinite article /an/ as a function word may prefer to be monomoraic, namely, a light syllable. It is probably the reason that the nasal coda, which is argued to contribute to syllable weight, has to be deleted unless it is resyllabified as the onset of the following syllable. As for why to choose [an] rather than [a] as the base form, there are two reasons: 1. Assuming [a] as the base requires an [n]-insertion rule when preceding words that begins with a vowel sound. However, in so doing it is hard to explain why the sound to be inserted is [n], not others. 2. Assuming [an] as the base requires an [n]-deletion rule when preceding words that begins with a consonant sound. Such a rule seems to be natural relative to the other way around as the arguements displayed above. --

※ 发信站: 批踢踢实业坊(ptt.cc) ◆ From: 140.119.203.39 ※ 编辑: Lhanas 来自: 140.119.235.214 (03/24 16:20)