In view of the special interest in anthropology on the problem
of relationship terms, I would like to add an appendix to the
present book as a supplement to the chapter on kinship
extension.

It is essential to make clear that a structural analysis of
relationship terms, at best, covers only a part of the whole
problem of kinship system and that a mere presentation of a
chart of terms is of little use by itself because it fails to show
their sociological implications. Such a treatment, which most
of the previous studies have followed, from the old work of
Morgan and Hart up to the recent publication of H. Y. Feng, ¹
is resulted from the unsound conception of language which
views words as representations of reality. Therefore it is
believed that an analysis of the relationship terms will be
enough to understand the organization of kinship.

Relationship terms, like all other linguistic data, should be
studied in their contexts. They are used for progmatic
purposes in establishing claims, in expressing affectional
attitudes and, in short, as a part of the behaviour of the speaker
towards his relatives. An adequate analysis must be carried
out by direct observation of how the terms are actually used. ²
But in the present note it is not possible to treat the problem
in detail; I only intend to suggest an outline for further
investigation.

Chinese relationship terms can be classified into four cate-
gories based on the general contexts of speech: (1) the context

-287-