从语言特征分布图到方言分区图以沂南方-言为例.pdf

Modern Linguistics 现代语言学, 2018, 6(1), 61-76 Published Online February 2018 in Hans. /journal/ml /10.12677/ml.2018.61008 From Linguistic Feature Maps to Dialect-Area Map: A Case Study of Yi’nan Dialect Mengbing Xiang Key Laboratory of Computational Linguistics, Department of Chinese Language and Literature Center for Chinese Linguistics PKU, Peking University, Beijing th th th Received: Feb. 7 , 2018; accepted: Feb. 20 , 2018; published: Feb. 27 , 2018 Abstract In order to effectively divide a dialect area into subareas, we must select the appropriate number of Linguistic items that can better reflect the geographical differences of a region. The basic work is to draw a series of isoglosses based on the Linguistic feature maps and make a further compre- hensive analysis. Taking Yi’nan dialect as an example, this paper shows the process in detail and proofs it with the clustering trees generated by MEGA software. It is pointed out that the weight of linguistic items is a key issue for dialectal subdivision. Keywords Isogloss, Dialectal Subdivision, Weight of Items, Cluster Analysis 从语言特征分布图到方言分区图——以沂南方 言为例 项梦冰 北京大学中文系暨中国语言学研究中心，计算语言学重点实验室，北京 收稿日期：2018年2月7 日；录用日期：2018年2月20 日；发布日期：2018年2月27 日 摘 要 为了有效地进行方言分区，必须选择适当数量、能较好反映地理差异的语言项目，在语言特征分布图的 文章引用: 项梦冰. 从语言特征分布图到方言分区图——以沂南方言为例[J]. 现代语言学, 2018, 6(1): 61-76. DOI: 10.12677/ml.2018.61008 项梦冰 基础上绘制同言线并进行综合分析。论文以沂南方言的分区为例，详细展示了这一过程，并用MEGA软 件所生成的聚类树进行检验，指出语言项目如何权重是方言分区所要面对的一个关键问题。 关键词 同言线，方言分区，项目的权重，聚类分析 Copyright © 2018 by author and Hans Publishers Inc. This work is licensed under the Creative Commons Attribution International License (CC BY). /licenses/by/4.0/ Open Access 1. 引言 本文以《沂南县志》( 山东