Abstract
This article aims to clarify the reality of bunke （stemming and stemmed family branch） in an early modern Japanese village, using the database of various documents including the Shumon Aratame-cho（Religious Faith Registers） and family trees from Kami-shiojiri, Ueda, Nagano.
Compared to the results of other studies, Kami-shiojiri Shumon Aratame-cho indicates that the bunkes by the second and third sons account for less than one tenth of the total of 94 cases and ‘cousins’ account for more than two fifths. The current article has found that, regarding the ‘cousins’ on the Shumon Aratame-cho who became bunke, only one third of the total are real cousins. The others are 5 and more degrees apart, i.e. the fathers or even grandfathers were cousins. Therefore, referring to the family trees, the bunkes had been set up in the previous generation and there are often generation gaps between the Shumon Aratame-cho and family trees for the timing of the bunkes. This suggests that they had been substantially stemmed for they were independent enough. The family trees as informal documents recorded such a reality. However, the bunkes were not established enough to be independent as an administrative unit and the Shumon Aratame-cho, as more formal documents, do not show them. Nevertheless, over some generations, they became economically independent enough to be considered administratively independent of the village society and stemmed on the Shumon Aratame-cho, too. Since bunkes were informal, they came in different circumstances and the contemporaries made some notes on them. To understand bunkes requires cross-referencing with the Shumon Aratame-cho, family trees and other such documents, for the reality can be discerned between these various documents.
