Mauss, Polanyi and Sahlins took issue with a view of humans which assumed that they were individualistic, maximising and fundamentally selfish creatu… - Thomas Hylland Eriksen

The anthropological production of knowledge has at least two elements: fieldwork and analysis. Some might want to add a third one, namely description; you first collect a body of empirical material through various field methods, you then describe whatever it is that you’ve discovered and, finally, you analyse the findings. Many, including the author, are sceptical of the distinction between description and analysis because the (anthropological) analysis inevitably begins in the (ethnographic) description itself and, indeed, already with observation. No all-encompassing, neutral description exists of anything, and nothing has a meaning independently of that ascribed to it. Already the delineation of the field of enquiry – socially, thematically, spatially, with respect to the concepts used – necessarily entails that reality ‘out there’ is presented in a selective and theoretically biased way. It is impossible to describe everything, or to give equal emphasis to everything one has observed.

The relationships between mother, father and children, family trees and genealogies, preferential treatment of relatives and alliances through marriage furnish us with some of the few really good and useful comparative concepts we have in anthropology. They exist everywhere in one form or another, and they differ in interesting ways. If the ultimate goal is to discover the unity of humanity through its manifold appearances, the profession cannot afford to let go of the still rich gold mine of kinship.

The world, as it is perceived by human beings, is to a certain extent shaped by language. However, there is no agreement as to just what the relationship between language and non-linguistic reality is.