摘要
在原型范畴理论的视域下,以626条现代汉语"X+化"派生词为研究对象,对其形式结构及成因进行认知分析,发现,基于"量"的语义特征,"性质形容词"最容易进入"X+化"结构,"名词"在表示抽象内涵义时可进入该结构,"动词"在表示延续性动作时可少量进入该结构。典型成员"形容词+化"在词频上并不占优势,实际上"名词+化"所占比例最大,这与典型成员通常表征为词频高的特点是相矛盾的。由于词频统计的客观局限性以及词类本身使用频度的影响,语义内省才是确定原型之关键。
This paper conducts a cognitive study of the formal structure and its causes of "X+hua (化) " derivatives in modern Chinese by examining 626 items from the perspective of the prototype category theory. The study finds that the character adjective can most easily enter the structure of "X+hua(化) ", the noun can enter the structure when it expresses an abstract meaning, and a few verbs can enter the structure when it refers to the durative action. The typical member "Adj+hua(化) " has no advantage in the word frequency, Instead, the "N+hua (化) " accounts for the most, which contradicts with the view that the typical member of a category always bears a high frequency. Considering the limitations of the statistics and the frequency of use of the word class itself, the intro- spection based on semantic features is the key to identify the prototype.
出处
《湖南师范大学社会科学学报》
CSSCI
北大核心
2018年第5期120-127,共8页
Journal of Social Science of Hunan Normal University
基金
湖南省研究生科研创新项目"原型范畴理论视域下的现代汉语‘X+化’派生词研究"(CX2015B157)
关键词
现代汉语“X+化”派生词
原型范畴
连续统
量
"X+hua (化) " derivatives in modern Chinese
prototype category
eonfinuum
quantity