Although there seems to be some consistency in Wolf Roder's argument
that a tribe constitutes a nation, this view misses the point that a
nation may also be seen as a more inclusive entity than the tribe. So
your the from a conceptualization of tribe to the definition of a
nation does not provide us with the analytical tools to incorporate
all the elements that constitute a nation. Roder's view can be used
as an analytical starting point though, in the sense that it evokes
the dynamics that determine the making of a nation.
Are there such identities as Anglo-Saxons or Latin people ? One even
speak of Indo-European languages, Semitic people. On the basis of
what ? On the basis of the fact that as a group of people they have
something--the language and the culture--which distinguishes them
from other groups of people. The term Bantu, despite what Mr. Newman
thinks, allows for the identification of a group of people whose
languages--and certainly cultures--are "closely related".
The fabrication of the concept is one thing and the actual existence
of a group of people and the cultural elements that characterize them
is another thing. A reality forces us to identify this group of
people by using a term. It could have been another term than Bantu.
The point is that identification and differentiation are proper to
the making of human societies. What may be problematic, however, is
the origin of the concept we use to identify and differentiate people
and their languages, the intentionality behind the concept and its
interpretation.