2005 Volume 56 Issue 1 Pages 147-164
Ever since Claude Fischer predicted the rise of friendship in cities in his “subcultural theory, ” many surveys on urbanism and friendship were conducted in Japan. Yet the results were mixed. This article addresses the empirical and theoretical issues on urbanism and friendship by analyzing the data of Nagoya Metropolitan Survey conducted in 2000. The results of the analysis show that, on the contrary to “the rise of friendship” hypothesis, the number of friends decreased with increased urbanism, mainly for those who had grown up in the region, because of the decline of local peer groups in the urban areas. Nevertheless, the more urban the area, the more middle-distance friends they had. The number of long-distance friends, on the other hand, was not affected by urbanism but by the respondent's history of residential mobility. The fact that many empirical associations are conditioned by residential history supports the “structuration” model of social networks, instead of the “choice-constraint” one, since the former emphasizes that the geographic distribution of relational resources is varied by one's history of residential mobility and that the reproduction of friendship is affected by urbanism as far as he or she has a lot of relational resources in the metropolitan area. Further, it is speculated that, on the macro level, while social networks within the metropolis may have been scarce in the earlier stage of urbanization due to the great share of immigrants, middle-distance friendship networks may flourish after one generation due to the increasing number of those who grew up in the region. Thus, this study brings temporal and spatial perspective into the theory of urbanism.