Identités relationnelles
Les localités multiples et l'identité d'un groupe social

Débats et controverses 5 décembre 2016

Edmund Leach, Political Systems of Highland Burma.
A Study of Kachin Social Structure
Réédition avec une nouvelle introduction,
London, LSE & G. Bell and sons, 1964

Eclairé par:

François Robinne and Mandy Sadan, Eds.,
Social Dynamics in the Highlands of Southeast Asia.
Reconsidering Political Systems of
Highland Burma by E. R. Leach
Leiden, Brill, 2007

Leach veut dissoudre l'idée reçue selon laquelle une culture ou une tribu forme une unité discrète:

(281) A part of my objective in writing this book has been to demonstrate that in contexts such as we find in North Burma the ordinary ethnographic conventions as to what constitutes a culture or a tribe are hopelessly inappropriate. I am not suggesting that this is a wholly new idea but I consider it to be a matter of general theoretical importance that calls for emphasis. A great deal of modern anthropological field work [sic] has been carried out in areas where there is a palpable lack of coincidence /282/ between the boundaries of cultural and structural phenomena. My thesis is simply that the conventional use of the concept of unit cultures serves to obscure the significance of such facts.

Leach écrit ce livre contre l'idée reçue selon laquelle les frontières du groupe social et les frontières de la culture coincident. Les ethnologues de la vieille école, refusant de voir que dans bon nombre de territoires habités par des tribus les phénomènes structurels et les phénomènes culturels ne coincident pas (there is a palpable lack of coincidence between the boundaries of cultural and structural phenomena), font une fixation sur les tribus qu'ils conçoivent chacune comme une entité culturelle (cultural entity) bien délimitée. Leach fait voler en éclat ces fictions ethnographiques (ethnographic fictions).

(290) Finally let me make the plea that the general type of analysis which I have attempted in this book may have value in other contexts besides that of the Kachin Hills Area. The cultural situation in the Kachin Hills, as I have described it, is both confused and confusing but it is not exceptionaL On the contrary I would claim that it is largely an academic fiction to suppose that in a 'normal' ethnographic situation one ordinarily finds distinct 'tribes' distributed about the map in orderly fashion with clear-cut boundaries between them. I agree of /291/ course that ethnographic monographs frequently suggest that this is the case, but are the facts proved? My own view is that the ethnographer has often only managed to discern the existence of 'a tribe' because he took it as axiomatic that this kind of cultural entity must exist. Many such tribes are, in a sense, ethnographic fictions.

Les débats et controverses qu'a suscités le chef d'œuvre de Leach sur les Kachin datent de plus d'un demi-siècle mais il trouve aujourd'hui une nouvelle actualité, comme le montre la relecture à laquelle se sont livrés les meilleurs spécialistes des hautes terres d'Asie du sud-est sous la conduite de François Robinne et Mandy Sadan. On ne parle plus de «groupes ethniques» comme dans les années 50, mais d'identités relationnelles et de localités multiples. L'ethnographie multisite — éventuellement multilingue — transforme radicalement la notion de «frontières» en anthropologie et recourt à des jeux d'échelles entre le local et le global.

Dan Sperber, Edmund Leach et les anthropologues, Cahiers Internationaux de Sociologie, nouvelle série, vol.43 (juillet-décembre 1967), pp.123–142.
Adam Kuper, An Interview with Edmund Leach, Current Anthropology, Vol.27, No.4 (Aug.-Oct., 1986), pp.375–382.
Charles Keyes, Presidential Address: "The Peoples of Asia" – Science and Politics in the Classification of Ethnic Groups in Thailand, China, and Vietnam, The Journal of Asian Studies, Vol.61, No.4 (Nov., 2002), pp.1163–1203.

Leach combattait dans tous les domaines — la parenté, les populations et les frontières, les tribus, la culture — une approche taxonomique des faits sociaux et culturels, fondée selon lui sur un présupposé fallacieux selon lequel, en assignant des noms à des relations de parenté, des groupes ethniques et leurs frontières, des traits culturels, etc., on réduit des réalités fluctuantes à des entités discrètes qu'on peut ensuite ranger dans les cases d'un tableau à double entrée.

Tabulated categories, tabulated nonsense
Controverse frontale entre américains et britanniques années 60

Edmund Leach, [Comment] in Raoul Naroll, On ethnic unit classification,
Current Anthropology
, Vol.5, No.4 (Oct., 1964):283-312, p.299. (Extrait.)

Murdock's World Ethnographic Sample [American Anthropologist 59 (1957):664-87 and all work that derives from it rest on the fundamental assumption that the units of discourse, call them "tribes," "cultures," "cultunits," or whatever, are "species objects" which can be described taxonomically by a list of characteristics. Just as a species of beetle may be described. This is a proposition which I simply do not accept. If I write a monograph about the Kachin (as I have done) and Murdock chooses (as he has done) to have that book treated as if it were a taxonomic description of a particular cultural unit (Murdock 1957:680), he is acting within his rights, but from my point of view he is producing tabulated nonsense.

Naroll tells us that the object of a comparative statistical study, a cross cultural survey, is "to compare culture patterns in order to see if there are general tendencies governing their construction." This objective presupposes that "culture patterns" are discrete entities composed of further discrete entities which are capable of being tabulated by code letters or numbers. I do not accept either of these suppositions, and it seems to me that tabulations which are made on this basis are bound to be grossly misleading no matter how the units are defined. Let us take the Kachin as a case in point.

Murdock (1957:680) tells us that (a) fishing is "absent, insignificant, or sporadic;" (b) Kachins live in compact villages or towns; (c) there is "normal occupancy of a single dwelling by an entire lineal (small extended) family;" (d) residence is "normally" patrilocal, with no "patterned alternatives;" (e) matrilateral cross-cousin marriage is "preferred;" (f) the system includes hereditary aristocrats and hereditary slaves. Are these statements true or false? I suppose it all depends upon what you mean by "normality." I know of Kachin villages where fishing is the major economic activity; in 1940, most Kachin communities were small hamlets of three or four houses, but a small minority were compact villages of over 100 houses; most houses were occupied by "an entire lineal family," but some houses contained 50 or more people, and in the days when (f) was true, the majority of the population lived in such "long houses;" while most couples reside patrilocally, the political structure is entirely dependent on the fact that a significant proportion of households reside uxorilocally; statistically speaking, marriage with a matrilateral cross-cousin is rare, though marriage with a kinswoman of the category to which a matrilateral cross-cousin belongs is obligatory. It is not that Murdock's tabulators misread their sources; it is simply that the ethnographic facts of the case will not fit tidily into tabulated categories. I believe that this is true of all human societies.

I have written in the first person, but I would suppose that most British social anthropologists would share my views on this particular issue.