Representation Through Collaboration Anthropology has a poor reputation in some circles. It denotes a discipline that, while ostensibly a descriptive and appreciative of difference in the intolerant New World Order, is frequently criticized for perpetuating a monologic that promotes another form of neocolonialism. The problem is with the word whereby an outsider speaks with authority for an Other's culture. And although some anthropologists of a postmodern bent have educated us to the epistemological problems of (re)presenting Others (Clifford 1988; Marcus and Fischer 1986), an equally damaging critique is being lodged by Indians themselves It must be acknowledged that all social artifacts (Deloria 1969, Vizenor 1989). One issue is over who gets to, if anyone, possess a dialogic component. (re)present the lives of Native Americans. The act of representation, we are told by the above critics, replicates entrenched modes of domination. The question is how we are to proceed in the current crisis of representation, if we believe that presenting other ways of living is important. The social science's is attacked because of its emphasis on objective data collection, validity claims, and generalizability. Its written works represent by reducing individuality and indeterminacy to abstract cultural principles. This article does not escape the representation problem because we will attempt to say something about O'odham culture. How we have resolved the voice problem, at least for the two of us, is to work collaboratively, and to present our work in a dialogic format. In this way we offer an approach that situates discussion, the act of knowledge production, the opening of the text, at the forefront of our research and writing process. In this way what we offer is incomplete and partial, even idiosyncratic and anecdotal because conversations are incomplete and partial. We agree with Gudeman and Rivera (1989: 267) who have observed: Good conversations have no ending, and often no beginning. One aim, then, is to chime into discussions that have no essential beginning nor ending. Another aim is to offer a modest alternative to the monologue that permeates social scientific production. It must be acknowledged that all social artifacts possess a dialogic component. The creation of social artifacts material and otherwise are at least partially realized through a process of verbal communication. This communication does not occur in a vacuum, but is the product of real people with differentially expressed identities merging in a context where they are transformed by the encounter itself. Much anthropological research, this work included, is the product of (at least) two individual's unique contributions. But it must also be admitted that each of our verbal contributions are the cumulative products of previous dialogues unique to our own personal histories. Kevin Dwyer (1977, 1979, 1982) recognizes this dyadic (and presumably triadic, quadratic, and historical) element of the anthropological process. In speaking of the research project, Dwyer suggests that