This article discusses which of the family formation patterns described by
John Hajnal obtained in traditional Japan. The issue was confronted in 1985
by Arthur Wolf and Susan Hanley, who concluded that 'China is to Japan as
Eastern is to Western Europe.' By examining a body of available empirical e
vidence concerning proportion married, age at first marriage, celibacy, fam
ily life-cycle patterns, servanthood, family labour and coresidence pattern
s, it would appear that their claim is misleading. japan's stem family syst
em and its corresponding marriage pattern should be seen as representing a
type structurally distinct from either the eastern- or the western-European
model.