functions which are widely regarded as inherently dangerous – mean that there
are four mistakes in the oft-repeated formula that “Bra ̄hman.as are the highest
caste because they are priests.”^4
1 Not all priests are Bra ̄hman.as;
2 Many Bra ̄hman.as are not priests. They are Bra ̄hman.as because they are
related to other Bra ̄hman.as (who may or may not be priests themselves; if
they are assumed to be related to priests at some point, this may be in the
distant, or forgotten, past);
3 At least some priests are degraded by their ritual activities and, according to
some authorities, allof them are;
4 If the performance of ritual functions is degrading or compromising, this
implies that only the renouncer, who is by definition outside caste-organized
society, can claim to be “pure.” The concept of “most pure” cannot then be
equated with “highest” since “highest” indicates “inside.”
The most important conclusion one can draw from these observations is that
the idea that “Bra ̄hman.as are the highest caste” makes no sense (irrespective of
the fact that it is very widely held by Hindus themselves and by commentators
on Hinduism). There is no doubt that ritual activity is directed at the ridding of
inauspicious qualities, but this does not indicate a capacity to rank groups as
higher and lower. On the contrary, careful study of the ethnographic record
shows that it is always impossible to rank castes unambiguously or without con-
testation: ambiguity and dispute are built in to the structure of caste relations.
This is seen both in the relations between patrons of rituals and priests of various
kinds, and among patron castes themselves.^5
Once these points are admitted, much of Dumont’s theory of caste quickly
starts to unravel. For Dumont, the opposition of the pure and the impure
depends on priests monopolizing ritual functions which were once the preserve
of the king: “power in India became secular at a very early date” (Dumont 1980:
76). But in spite of the fact that his entire theory of caste hangs on this claim,
Dumont provides no evidence to support it – bar his own assertions that
Bra ̄hman.as are now unambiguously the “highest” caste and Untouchables the
“lowest.” With much textual and ethnographic evidence suggesting otherwise,
it is fortunate that there is an alternative way of approaching the problem which
does not depend on concepts of highest and lowest.
The pollution concepts which are the hallmark of caste are conventionally
said to be an expression of “Hinduism,” even though we know perfectly well that
this is not a monolithic set of beliefs. An alternative representation of caste
values portrays them as an expression of institutions associated with kingship.
This approach, which is often associated with the seminal work of Hocart (1950
[1938]), has certain advantages over any other explanation of caste. First, it
offers an explanation of why caste organization is found in some parts of the
Indian subcontinent but not others. Typically, caste originates in regions which
500 declan quigley