In this paper a new technique for representing dialectal differences will be introduced. Two techniques of simplification are applied to represent geographical distribution patterns of standard Japanese. The first one is a representation of two-dimensional geographical distribution patterns by one dimension. The second one is plotting the geographical locations making use of railway distances from cultural centers. By simplifying the two-dimensional geographical distribution into one by railway distance, another dimension can show the average percentage of usage of standard Japanese forms for each prefecture. In order to see the patterns of distribution of the Japanese standard forms, “Kasai data”, or numerical data of the “Linguistic Atlas of Japan”, was utilized.
By comparing railway distances from Tokyo and from Kyoto in the form of scattergrams, the degree of linguistic influence of each of the capitals of Japan was analyzed. The overall pattern showed that railway distances surely correspond to standardization of dialects. The standardization of language in Japan proceeded in two historical stages. The first stage was standardization from Kyoto. The second stage appeared later from Tokyo. These two stages were-ascertained by classifying 82 words into 4 clusters. West cluster words showed Kyoto-centered diffusion and east cluster words showed Tokyo-centered diffusion.
An analysis of younger Japanese people through a countrywide postal survey showed that standardization has proceeded with greater speed in recent years. Most of the words of the Kasai data approached the end of diffusion. This third stage of standardization seems to have been influenced by the mass media.
