Utdrag fra Making Place: Houses, Lands, and Relationships among Ngadha, Central Flores - Smedal 1994):

If adequate sociocultural anthropology is based not on observation alone but also on participation, if by participation we mean interactive engagement in other peoples' lives, and if by such engagement we imply that other peoples' practical and discursively expressed knowledge is not merely relevant but pivotal to our endeavour, then the term 'informant' seems both inept and unfortunate; conjuring [up] an image of a provider of raw data for advanced processing much as Third World countries deliver raw materials to industrially advanced states. Some anthropologists refer incessantly to what their 'friends' have said or done - while I may well lack these researchers' capacity for making friends I have also learnt much from Ngadha men and women I would not dream of placing in this category. 'Partner in dialogue' is fine but invokes one-to-one conversations rather than the 'multilogues' I suspect characterise much anthropological field research; 'conversationsal partner' is perhaps better but remains, with 'partner in dialogue', within an implicit framework of spoken words (before one knows it frozen in texts). I alternate in this work between 'Ngadha' or 'Ngadha people', 'Ngadha men/women', 'specialist', 'well informed person' and, sometimes, 'consultant'. The choice of the last after Basso (1979:99 ff.) is tentative, for objections can doubtless be raised with respect to this term, too - generating as it perhaps does images in some peoples' minds of Business School graduates or attaché case-equipped sociologists.