the structural role of every sector in a total organic system. Whereas a ruling aristocracy is invariably a numerical minority, a dominant caste may be, and usually is, a majority element in the total population.
It follows that the kind of dominance asserted by individual members of an aristocracy upon individual members of the lower classes is entirely different in quality from intercaste hierarchy, even though both types of relationship are concerned with economic service and even though, in both cases, one of the parties involved is necessarily of 'higher social status' than the other.
I have commented at length upon the special qualities of intercaste relationship because the various contributors, in their treatment of this topic, seem to me particularly illuminating. They have led me to the conclusion that there is something fundamentally wrong about Kroeber's well-known definition: 'A Caste may be defined as an endogamous and hereditary subdivision of an ethnic unit occupying a position of superior or inferior rank or social esteem in comparison with other subdivisions' ( Kroeber 1931). It is wrong because it puts the emphasis in the wrong place -- upon endogamy and rank, and because it slurs the really crucial fact that caste is a system of interrelationship and that every caste in a caste system has its special privileges.
But the principal concern of these authors is not with definition. Each essay has its own individual merits which are both sociologically and ethnographically important. Here I must leave the reader to judge for himself.