Anthropos 88.1993: 323-336 
Middle Indian Kinship Systems 
A Critique of Georg Pfeifer’s Interpretation 
Robert Parkin 
Abstract. - The question of how the kinship systems of the 
tribes of middle India should be interpreted is examined with 
explicit reference to the work of Georg Pfeffer. It is argued that 
his hypotheses of the existence of four- and eight-line kinship 
terminologies in the region do not correspond to the evidence 
but reflect instead an excessive reliance on the dictates of struc 
turalist theory. These terminologies are best seen as modified 
two-line prescriptive ones, broadly intermediate between those 
of south and north India. [India, tribes, kinship, structuralism] 
Robert Parkin, Ph. D., Research Associate of the Institute of 
Social Anthropology, University of Oxford; until recently, he 
taught and carried out research in the Institut für Ethnologie, 
Freie Universität Berlin. - His major work is “The Munda of 
Central India: An Account of their Social Organization” (1992); 
see also References Cited. 
I 
How should the kinship systems of tribal middle 
India be interpreted? Georg Pfeffer and myself 
have long been concerned, independently, with this 
question but have come up with very different 
answers. In this article, I wish to draw atten 
tion to these differences and to address them in 
the form of a critique of the rival interpretation. 1 
They cannot be dismissed as theoretical, since 
both Pfeffer and myself share the same, basically 
structuralist approach to kinship drawn ultimately 
from Lévi-Strauss through such figures as Louis 
Dumont, Rodney Needham, and N. J. Allen. They 
are rather substantive, i.e., they concern the nature 
of the systems themselves, something which in this 
case ultimately reflects methodological issues such 
as the proper use of evidence, the coherence of 
different sets of data, and the consistency of logical 
reasoning. 
First, however, a word as to our respective writ 
ings; For myself, the major work is my book (Par 
kin 1992«), supplemented by a group of articles 
(Parkin 1988a, 1990, forthcoming), all of which 
should be seen as partially superseding earlier 
work of a more provisional and tentative nature 
(Parkin 1985, 1986, 1988b). Pfeifer’s corpus con 
sists first of his book “Status and Affinity” (1982a) 
which has been the subject of two scathing reviews 
(Fruzzetti 1984; Bouez 1985a), both of which see 
it basically as a boldly conceived design that goes 
seriously wrong in its execution. There are also a 
couple of related articles (Pfeffer 1982b, 1983), an 
article postulating the existence of dual organiza 
tion in India (1982c), a paper on Santal totemism 
(1984a), and a comparative study of kinship in the 
south Asian area (1985a). Occasional references 
to Pfeffer’s other writings will be made here, but 
I leave aside for detailed discussion his work on 
gods in Orissa (1987a) and the secondary funerals 
of the Gadaba (1984b), as well as that on German 
ic kinship terminologies (Pfeffer 1985b, 1987b, 
1987c; butcf. Parkin 1993b). 
It would be surprising, of course, if there was no 
convergence between Pfeffer’s work and my own, 
given the convergence in region, ethnic groups, 
topic, and theoretical approach. For the most part, 
however, this is limited to generalities - the sig 
nificance of middle Indian kinship systems as in 
termediate between the better defined regions of 
north and south India (a point also made by Bouez 
1985b: 153); the existence of otherwise prescrip 
tive terminologies with separate affinal terms, both 
in middle India and, sporadically, elsewhere in 
south Asia; and the practice of dispersing alli 
ances among several alliance groups. Conversely, 
we do not agree on the nature of the tribal sys 
tems themselves, and it follows from that that we 
do not agree either on how they may be relat 
ed, whether logically or historically, to those of 
north and south India respectively, nor on what 
the transformational paths between these various 
systems might be. I should add that Pfeffer has one 
apparent advantage over myself - he has visited 
and travelled in the area extensively, whereas I 
1 In part, this paper takes further points that emerged from a 
previous exchange but which could not then be developed 
properly (see Parkin 19926, Pfeffer 1993a, Parkin 1993a).