This article aims to classify the various types of accents that are distributed throughout Tokyo's northeast region and clarify their characteristics using speaker typology. Moreover, it attempts to grasp the process of change among the accents in the region based on the relationships between types. The data for this study come from 44 elderly persons in Tokyo's northeast region reading approximately 28 2-mora words of Kindaichi types I~V from a word list. For the speaker typology, cluster analysis was used.
 The results of the analysis classify the accents in Tokyo's northeast region into 4 groups, “Saitama Distinctive Accent Speakers,” “Quasi-Saitama Distinctive Accent Speakers,” “Quasi-Standardized Language/Tokyo Central Region Accent Speakers” and “Standardized Language/Tokyo Central Region Accent Speakers.” Furthermore, in regards to the division of the groups, I found important the conditions under which the following three characteristics manifested: 1) Standardized Language/Tokyo Central Region Accent types, 2) Formal variation, 3) Saitama Distinctive Accent types.
 Based on the above, I argue that the process of change for accents in Tokyo's northeast region consists of movement towards the Standardized Language/Tokyo Central Region Accent. In this process, accents move from “Saitama Distinctive Accent Speakers” -> “ Quasi-Saitama Distinct Accent Speakers” -> “Quasi- Standardized Language/Tokyo Central Region Accent Speakers” -> “Standardized Language/Tokyo Central Region Accent Speakers.” Also, I argue that standardization, or movement towards Tokyo Central Region Accent, is finalized by “a change in short phrases of types IV and V from the LHL accent pattern (-2 accent pattern) characteristic of Tokyo Northeastern Accent to the HLL accent pattern (1 accent pattern) of the Standardized Language/Tokyo Central Region Accent” and “the disappearance of formal variation.”

View full abstract

In previous studies about Japanese sentence length, two different types of the model for sentence generation have been proposed by Sasaki: one is a multiplicative stochastic model resulting in log-normal sentence length distribution and the other is an additive stochastic model resulting in the negative binomial sentence length distribution.
  In the present study, motivated by Sasaki's suggestion, we examined the structure of dependency trees and checked whether those models could explain the obtained structure of dependency trees. To do that, we used Kyoto University Text Corpus (33,082 sentences) which includes the information of dependency relations among segments.
  As a result, we found that the structure of the dependency trees did not accord with the expectation of multiplicative nor additive stochastic process.

View full abstract