9058009550

castration and penectomy operations, never have relations with men, never take on new feminine names, and never speak as men. Megha, who has a high profile in her district of Banaras, is very aware of how her own self-presentation affects societal opinion, especially in light of the recent increase of anti-hijra violence in northern India; she is morę interested than the other hijras in pro-jecting a self that conforms to societal expectations - a self that is determined by both ascetic and anatomical considerations. Megha’s insistence that the hijras were not only given feminine names at birth but have also never spoken in the masculine serves to support the claim that the hijras’ femininity is innate, affirming a larger cultural belief that the hijra lifestyle is not socially con-structed, but rather something that begins at (or even before) birth.

Most of the hijras we interviewed, with the exception of Rupa who became a hijra as an adult, primarily employ feminine-marked verbs when speaking in the first person or when addressing other hijras in the second person. When using the third person to refer to other hijras, however, the hijras are much less consistent, their choice of marking dependent on the relative social status of the referent in ąuestion. When the hijras speak in the third person and express distance from the referent, specifically when the referent is perceived to be either a superior or a subordinate, they tend to make greater use of the masculine; in contrast, when the hijras express solidarity or familiarity with a referent of equal status, they tend to make greater use of the feminine. Hijras rely not only upon their own internal systems of law and order, but also upon elabo-rate familial structures which delegate various feminine roles to different members of the group, among them dadi [‘paternal grandmother’], rtant [‘maternal grandmother’], ma [‘mother’], mausl [‘mother’s sister’], caci [‘uncle’s wife’], dtdi [‘older sister’] and bahin [‘younger sister’]. Fundamental to this system is the guru-disciple relationship; the initiate pledges life-long devotion to an older, morę experienced hijra, who in turn gives her a share of the community’s earnings. The affected kinship situation created by the hijras is unique, in that the guru acts symbolically as both sas [‘mother-in-law’] and suhag [‘state of being in a husband’s pro-tection’]. Having abandoned all worldly ties upon entry into the community, the hijras appear to transfer every auspicious łife-rela-tionship to their guru, regardless of the fact that such a transferral, in the eyes of society at least, results in a superficially incestuous system.¹⁷

The hierarchical naturę of the community becomes transparent