modern political realities (see also Galey 1989). Even Raheja, whose brilliant
ethnography exposed the faultlines of Dumont’s theory more clearly than any
other recent work, muddies the waters somewhat when considering the nature
of caste in contemporary India. She writes that “[k]ingship no longer exists, but
it has been, perhaps, replaced by the ritual centrality of the dominant caste”
(Raheja 1988b:517). The word “replaced” is unfortunate because it has always
been the function of members of dominant castes to patronize ritual, replicating
on a diminished scale the role of the king. Put differently, members of dominant
castes continue to be mini-kings, not just materially, but in terms of their politi-
cal/ritual function. They are “at the center of a complex ritual organization that
permeates nearly every aspect of the everyday life of the village” (ibid.). This real-
world fact of kingship/patronage sits uneasily with Dumont’s insistence that the
king’s role has been secularized, and much more comfortably with Hocart’s claim
that the king/patron is the “principal” of the ritual (Hocart 1970 [1936]: 61).
A phrase which I have elided from the two quotes from Raheja in the previ-
ous paragraph further confuses understanding of both kingship, and relations
between patrons of rituals and the priests who carry them out. In the village of
her fieldwork, she says, “as in many of the textual traditions on kingship, the
Bra ̄hman.a is hierarchically superior, yet the dominant landholding caste stands
at the center of a complex ritual organization ...” (ibid.). It is difficult to see what
the foundation for this alleged superiority is, other than an engrained idea that
Bra ̄hman.asmustbe superior because that is what everyone else seems to believe
these days. In practice, superiority seems to attach to members of the landhold-
ing caste in both ritual and nonritual domains. If the Bra ̄hman.a is “hierarchi-
cally superior” in an ideological sense, it is a Bra ̄hman.a who is so idealized that
he cannot possibly belong to the world of caste relations where he would be
tainted by the receipt of prestations for performing ritual services. It cannot be
the Bra ̄hman.a of the village who is necessarily immersed in caste (i.e. interde-
pendent) relations, and who is therefore, like everyone else, caught up in the web
of inauspiciousness which is the product of the dealings of normal social life.
Interestingly, Dumont’s theory predicts just this eventuality: “In theory, power
is ultimately subordinate to priesthood, whereas in fact priesthood submits to
power” (Dumont 1980: 71–2).
One of the problems with any understanding of caste is that the word “caste”
itself has been used to translate two quite different Sanskrit concepts which are
assumed, quite wrongly, to have an automatic connection. We have already
encountered the concept ofja ̄tiand its sense of “kind” or “species.” Another way
of glossing this concept might be to say “the group that one was born into,” the
relativity of this gloss conveying the contextual nature ofja ̄tiascriptions. When
asked to name their ja ̄ti, people may name their patrilineage, the name of the
group of lineages they conventionally marry into, or even the name of what
would now be called an ethnic group. However, even though the concept is
elastic, the idea of origin by birth is constant. There is no mystery about this. If
a British person is asked: “where are you from?,” he or she may give the name
502 declan quigley