Bardache was the word applied by early French explorers to some "native" individuals they encountered, mostly male, who adopted (or were assigned) feminine dress, mannerisms, and work assignments. They also took male sexual partners. I'm not absolutely certain about this yet, but it appears that "berdaches" did not copulate with other "berdaches": the trade/queer divide applied, as in later American and other cultures.
It's not clear why the French thought that berdache was the right word for these men. Its history has been traced to the Arabic (or Persian?) bardaj, "slave." Supposedly it eventually came to refer to a "catamite," from a Latinized form of Ganymede, the beautiful boy who was raptured by Zeus to serve as his, erm, cupbearer: in practice it meant the "passive" partner in a male homosexual relation, and that is why the (evidently classically-educated) French trappers applied it to the effeminate Indians they encountered, because that was what they were. But here's something odd, and an indicator for future research on my part: it just occurred to me to look up berdache and bardache in an online French-language dictionary and encyclopedia. The bucket came up dry, with no references, historical or contemporary, for either word. Van Cleve, however, claims that berdaj was "a Persian word for 'intimate male friend'" (17), which I have never heard before and can't find attested elsewhere.
The first thing to remember is that an etymology doesn't tell you the meaning of a word, it only tells you its history. Words do funny things over time; they may reverse meanings or take on other seemingly unrelated ones. Eventually "Western" anthropologists adopted and adapted berdache to mean specifically Native North American males who took on feminine dress and work assignments. (By "specifically" I mean that if a similar role occurred in other societies -- South or Southeast Asia, say -- anthropologists would not call them berdaches but something else, hijra or katoey or bakla perhaps. So, while the term's origin is less than felicitous, it no longer means or connotes "catamite," especially when the anthropologist is gay or lesbian, and it was the best word available to the profession before two-spirit was invented. I should think that the fiercest defenders of indigenous intellectual property wouldn't want non-indigenous anthropologists to use two-spirit anyway.
The role of the berdache in Land of 10,000 Loves relates to a famous painting by George Catlin, "Dance to the Berdache":
Like many Euro-Americans who encountered sexual practices they disapproved of, Catlin was repulsed by the ceremony and its focus. According to Van Cleve, a late 18th-century fur trader named Alexander Henry said that "this creature, called Ozaw-wen-dib, was now near fifty years old and had lived with many husbands" (16). Participants in the dance depicted in the painting were Ozaw-wen-dib's present and former "husbands" who "voluntarily got up to dance." There are over a dozen of them in the painting; it would seem that by modern Homo-American standards, the indigenous term A-go-kwa should be translated "slut," a Homo-American term which means "someone who's getting more than you are."
But Van Cleve overlooks something important. By his criteria, and according to the standard Anglo-American scholarly consensus, two-spirit people were not queer in their own societies. Far from violating indigenous gender or sexual norms, they occupied a more or less comfortable niche; there was a place for them -- an honored place, according to the current consensus. If they represented "an alternative to the male-female gender dichotomy," it was an alternative that their societies had provided for them as part of their gender systems. They did not call the assumptions of their social orders into question, they accepted them. (Some of us feel the same way about well-known analogues of the two-spirit in our own culture; I wish I could remember who it was who called drag queens the loyal opposition to our gender norms.)
Needless to say, I'm skeptical about the scholarly consensus on the two-spirit; to me it comes suspiciously close to romanticizing the Noble Savage who rejected simplistic binaries and intolerance of those who were different. But I still have a lot to learn before I can confirm or reject my suspicions. As I say, it's a subject for ongoing and future research.
(The title of the post is culturally appropriated from Sherman Alexie's poem "How to Write the Great American Indian Novel." Thanks, Mr. Alexie.)
