The Timing of Language Change
The Timing of Language Change
复制标题
语言改变的时机
DOI：
10.1002/9781118257227.ch23
复制
发表时间：
2012
期刊：
影响因子：
--
通讯作者：
Mieko Ogura
中科院分区：
文献类型：
--
作者：
Mieko Ogura
19世纪70年代中期，在德国工作的新语法学家提出了关于音变如何发生的最明确的假设。新语法学派的假说主要有两个方面：词汇规律性和语音渐变性。我们不会重复最近提出的各种论点来表明新语法假说的困难--见小仓（1987），以及本手册的第22章。鉴于新语法学派所提倡的机制的不令人满意的状态，人们提出了实现声音变化的其他过程。在过去的四十年里，对各种语言的实证研究，使用大量的数据，表明必须有一个过程，这是以词汇渐进的方式实现的，在整个词汇中扩散。这是承认语音不规则性的必然结果;在他的开创性文章中，王（1969）称这一过程为“词汇扩散”。词汇扩散的时间分布可以用S曲线斜率来表示。当变化第一次进入语言时，它影响的单词数量可能很小。变化逐渐扩散，起初进展缓慢。然后，随着它的传播，它会加速，在中途加快速度。在最活跃的时期，这种变化通过大量的词项迅速移动。然后它又逐渐变慢，并在结束时逐渐变细（Chen 1972）。扩散过程类似于传染病的流行，流行病的标准模型产生一个S曲线。
The most explicit hypothesis on how sound change comes about was proposed by the Neogrammarians working in Germany in the mid-1870s. The Neogrammarian hypothesis has essentially two parts: lexical regularity and phonetic gradualness. We will not repeat the various arguments that have been offered recently to show the difficulties of the Neogrammarian hypothesis–see Ogura (1987), and also Chapter 22 in this Handbook. In view of the unsatisfactory state of the mechanism that the Neogrammarians advocated, other processes for implementing sound change have been proposed. Empirical investigations over the past four decades on a variety of languages, using large amounts of data, have shown that there must be a process which is implemented in a manner that is lexically gradual, diffusing across the lexicon. This is an inevitable consequence of admitting phonetic abruptness; in his seminal article Wang (1969) called this process ‘lexical diffusion.’The chronological profile of lexical diffusion may be represented by the S-curve slope. When the change first enters the language, the number of words it affects may be small. The change gradually diffuses, going slowly at first. Then, as it spreads, it accelerates, picking up speed in mid-stream. In the most active period, the change moves quickly through a large number of lexical items. It then gradually slows down again, and tapers off at the end (Chen 1972). The diffusion process is comparable to epidemics of infectious diseases, and the standard model of an epidemic produces an S-curve.