16
RlCE UNIVERSITY STUDIES
up these investigations with intensive studies of typical segmental units
selected by national sampling.
I am not suggesting that anthropologists seriously attempt to record the
entire behavioral profile of Japanese life. But I am implying that more care
could be given to the selection of part-phenomena for study so that the
results are of greater significance or representativeness. Key patterns of
social relations could be studied in several social contexts, not just one.
Studies like Nakane’s and Befu’s (see references) of the relations between
blood kinship and economic phenomena could be extended to social seg-
ments that are not rural. Although a number of studies of hierarchical
authority patterns have been produced, these are often done by using
different conceptual schemes (e.g., compare Bennett and Ishino 1963 with
DeVos and Mizushima 1967), and comparison is difficult.
Sociologists or anthropologists writing in the sociological frame have
done much better at representing the nation. Sociology has a traditional
concern for the largest, not the smallest, social units. The sociological
approach is, in fact, that approach in the behavior sciences which seeks to
know the social whole—despite anthropology’s well-known claim to be
the science of cultural wholes. Anthropology’s definition of the “whole” is
based on the concept of an integrated tribal culture, and this form of human
behavior is ambiguously present in large complex social systems.
Therefore such studies as Dore’s of a Tokyo ward (1959) and Vogel’s on
the middle class, Koyama on family (1962), or Norbeck on associations
(1967), aim at more comprehensive statements. They often succeed because
the sociological approach focuses on phenomena with universal significance:
stratification, achievement, urbanization, migration, mobility, and demo-
cratic interaction. All of these processes affect the national entity and often
operate in the various social segments or microsocial realms with sufficient
sameness and regularity to make findings appropriate to one or two seg-
ments roughly representative of the whole.
Sociopsychological research on Japanese society stands roughly in
the middle between the anthropological and sociological modes. DeVos,
Wagatsuma, Doi, and Mizushima (see references) address themselves to
topics of large magnitude — achievement, hierarchy, anxiety. But they often
study these topics in an “anthropological” context, that is, as microsocial
segments. These segments are chosen partly for ease of management, and
partly because of the same curiosity about unknown or obscure social
phenomena that guides the work of many anthropologists.
The most impressive single category of Western anthropological research
on Japan, kinship and family studies, not only reflects the deep interest in
this subject during the 1950’s but also is based solidly on Japanese rural
ethnological investigations (e.g., Fukutake 1967; Nagai 1953, reporting