定中式形名组合词长搭配的量化研究

A Quantitative Study of Word-length Patterns in Chinese[A N]Compounds

本文从量化研究的角度来讨论定中式形名组合[1+1]、[1+2]、[2+1]、[2+2]结构的词长搭配问题。本文使用兰卡斯特汉语语料库(Lancaster Corpus of Mandarin Chinese),通过分析,得出以下结果:一、在定中式形名组合[1+1]、[1+2]、[2+2]、[2+1]四种组配模式中,[1+1]出现的频率最高,其次是[1+2]或[2+2],[2+1]最低。二、[2+1]相对于其他三种组配模式而言,出现的数量极少,而且主要局限于少数特殊结构;它跟名名组合中的[1+2]和动名组合中的[2+1]一样,都不是一种能产结构。三、语料的语体差异会导致统计数据的部分不同,但是不影响总体结论。

We offer a quantitative study of word-length patterns in 1＋1, 1 1＋2, 2＋ 1 and 2＋2 [A N] compounds in Chinese, using the Lancaster Corpus of Mandarin Chinese. Three results are found： （i） The occurrence frequency of 1＋1 is the highest, followed by 1＋2 and 2＋2, and the occurrence frequency of 2＋1 is the lowest. （ii） The occurrence of 2＋1 is extremely low and is limited to a few specific structures; it is similar to those of 1＋2 [N N] and 2＋1 [V N], which are widely thought to be unproductive. （iii） Different text styles have some effect on the occurrence frequency of various length patterns, but they have no effect on （i） and （ii）.