Bu
ui
ibli
CUSTOMARY LEGALISTS
Steps to a reflexive legal anthropology
Peter Pels
M.A. Thesis in Cultural Anthrapology
University
ofAmsterdam
January 1986
Spitsbergenstraat 61, 1013 CL
Amsterdam.
Antropologisch-Sociologisch Centrum,
Sarphatistraat 106a, 1018 GV
Amsterdam.
PREF ACE
As the writing of this thesis has been interspersed with attempts to teach, thinking about how to teach, and with attempts to produce something resembling a historica! quorterly, i t hos taken o long time. The centrol ideos of this manuscript are, in written form, more than two yeors old. This is o sufficient omount of time to get bored with the subject of the thesis, and there have been moments that I have tried to forget I was dealing with i t by stubbornly concentroting upon soma other activity. Those octivities would be interrupted by l i t t l e notes like the following:
Hoi. (28th May 1985)
Your friendly conscience speaking:
-1984- "Niklas Luhmann and his View of the Social Function of Low," Human Studies ? ( 1): 23-38
by John W. Murphy Regards,
Superego. I intend to reply to that note now.
Dear Friendly Conscience,
Sure, "a sneaky way to get me to think about my 'scriptie'", as you said in an earlier note. The thing may be finished now, but I doubt whether you will leave me alone now: you probably will go on nagging me about publications and the like. Never mind. There's one particularly gratifying result of this whole affair: the compliment you gave me when commenting on the first droft, soying that you had a bal! with i t .
Thonks - a lot.
Peter.
But opart from Bob 'Superego' Scholte, I have to thank Carol Greenhouse and Franz von Bendo-Beckmann for their extensive comments of the severol drafts of this thesis. If this paper has any legal anthropological merit, i t is the result of their advice and criticism. I want to thonk Agnes Schreiner and the students of the werk group on lego! anthropology at the Department of Law, too, for interesting discussions that have led me to sharpen my views on several issues. Of course, none of these contri-butors is responsible for the faults and fallacies of this thesis.
CONTENTS
Pref ace Contents
1. Introduction
2. Custom and the Fallacy of the Rule 2.1. Bourdieu's theory of practica;
the fallacy of the rule.
2.2. The fallacy of the rule in legal anthropology
2.3. The dialectic of embodied and
I I I
5
5 9 objectified history 13 3. Modes of communication 20 3.1. Abstraction and control 21 3.2. The distinction between rules andnorms 25
3.3. Epistemology of writing 2? 3.4. Legal anthropology and colonialism 34
4. Conclusions 39
4.1. Legalism as a product of anthropology 39 4.2. The common culture of law and
anthropology and the problem of legal
pluralism 41
When discussing the shift in legol onthropology from 'lego-l i s t i c ' or 'ru'lego-le-centred' approaches to a focus on processes and interests, legal anthropologists commonly attribute the fermer bias to an excessive influence of legal theory against which the latter approaches have been safeguarded [Comaroff and Roberts 1981; 6, ?. Nader and Yngvesson 19?3; 884, 886. Pyle 1981; 20. Roberts 1983; 3). Elsewhere, however, i t has been stated that the relationship between legal and anthropo-logical studies is tenuous, as is shown, for instance, by the naar-absence of interdisciplinary projects [Snyder 1981a; 141, Twining 19?3; 561). This poses the question of whether the influence of the legal establishment on the anthropology of law is indeed as important as anti-legalists suggest; or, to put i t differently, whether 'legalism' in the anthropology of law is a product of the influence of legal theory or a result of legal anthropological practica itself.
In this paper I intand to explore the latter route: I will show that much of the legalistic bias of legal anthropo-logy is closely related to a significant gap in anthropologi-cal discourse: the absence of a reflexive theory of anthropo-logical practica (this is not unique to legal anthropology [ see Schol te 19?2, 19?8]). A consideration of the epistemolo-gical importance of such a theory of practica challenges the current ideas about social practica and the fallacious use of the concepts of rule, norm, and custom that goes with i t .
8acausa anthrcpclcgists have commonly adhered to a view of
their own activity as the following of the correct rules of method (fieldwork) and the subsequent recording of the found dáta in writing (ethnography), they have hardly paid atten-tion to the fact that the whole process of doing research and reporting on i t is, epistemologically, a process of communi-cative interchange, a dialectica of knowledge-production between anthropologist, his or her subjects and/or peers [see Fabian 19?1). This process cannot be reduced to the simple application of methodological rules. Nevertheless, a lot of anthropologists have projected this mistaken idea of field
research into the situotion they studied: their idea of socio! order has usually been based on what Bourdieu calls "the fallacy of the rule", the idea that order is the result of the obedience to shared rules of orderly behavior. A reflexive theory of anthropological practica indicates these epistemological fallacies and systematizes the attempts to criticize them. Moreover, i t indicates the epistemological and ideologicol importance of the failures to distinguish between modes of o~jectification (speaking, writing, buil-ding, et cetera) and logica! levels of communication and metacommunicotion. When these distinctions are acknowledged, they con be used to show that there is a close connection between our idea of "rule" and the wirtten mode of objectifi-cotion. My finol conclusion is that anthropologists are especially vulnerable to the influence of legal theory or ideology becouse anthropologicol practica farces a comparable way of thinking on them.
The epistemological1 approach has to be distinguished
from a positivist 'methodological' one. Where postivist anthropology is mainly concerned with the correct use of scientific methode in order to faithfully represent social reality in an ethnography, an epistemological approach concentrates upon the process of production or constitution of this repre~entation [Fabian 1971;20). The positivist is interested in the analysis of the truth of the representa-t ion, as judged by representa-the use of representa-the righrepresenta-t rules of method. The epistemologist is interested in the constraints and possibi-lities of the anthropologist in the dialectica! process of communication with his/her subjects and peers. As the intersubjectivity of communication is the precondition of the objectivity of scientific knowledge, and the existence of methodological rules only part of the dialectica! process and
The concept of "epistemology" does not refer here to a freischwebende philosophy of knowledge, but to a science of knowledge-production. It is therefore not a metaphysical play of ideas, but a dialectic between theory and empirica! facta, that has the interaction of human beings with their environment by means of their mental capacities as its object (this conception is derived from Bateson [see 1979; 250) and Piaget [ 1972) .
not i t s summary or causa, the epistemological approach encom-posses the positivist methodological view. (As Wilden says, dialectics is of a higher logical type than analysis [ 1980; 123)) .
However, os o reflexive concentrates on the octivity
epistemology of anthropology of the reseorcher, i t will alwoys remain merely a criticol assessment of the product of research; i t can never be on empirica! argument. This implies that the arguments advanced here will always tand to the negative and critica!; I hope that my critica! orguments do not completely overshadow my respect for the intellectual craftmanship of the people criticized. Moreover, the empiri-cal content of a definition of law can, as a consequence, not be subject of discussion here. Some use of the concepts of 'low' ond 'legol' is necessory, though. As my main concern, following the discussion on reflexivity and the fallacy of rule, is the difference between 'legol' and 'customary' communicotion, I will use these words moinly to refer to a context in which written rules are used by on odjudicotor with power of sonction, without wonting to stress thot this covers all instonces of legality thot con be imogined. In foct, I ogree with Geertz thot we con rest content with exploring bath differences and similorities between our low and their custom, "the one lighting whot the ether dorkens", without recourse to on encompossing definition [Geertz 1983; 233). To try to estoblish a "baseline for cross-culturol purposes" moy even be o disodvontoge to comparotive projects when the similorities assembled in an encompassing definition make us ignore relevant differences. But whatever choice of definitional strategy I prefer is not relevant to the discussion in this paper.
I will start with the exploration of the reflexive understanding of anthropological practica developed by Pierre Bourdieu. By means of this argument we can identify and circumvent the "follocy of the rule", the idea that "rules govern practica". I will show that the fallocy of the rule is not just o chorocteristic of the 'legolist' or 'rule-centred' approaches in legol anthropology, but thot more or less explicit anti-legalist approaches commit i t as well.
Further-more, I wil! argue that the latter tend towards another type of reification I wil! call the "fallacy of interests". A reconsideration of Moore's fameus "Law and Social Change'' [ 1973) wil! lead to the conclusion that closes this section: that the fallacy of the rule consists of two related mista-kes: the failure to distinguish modes of objectification and the failure to see levels of logica! typing.
These points are further elaborated in the next section, where I will try to distinguish the epistemological differen-ces between customary, normative and regulative communica-tion; these are differences in logica! levels in the first case and differences of mode of objectification in the second. In this section I will argue that our idea of the rule is congruent with the epistemology of writing. This section wil! be concluded by a considering the ideological function of these epistemological fallacies in a colonial
situation.
Finally, I wil! return to the question with which I started this introduction, to see in
lism' of legal anthropologists is the
what sense the 'lega-result of their own practica and discourse. I wil! finish with some suggestions for the further elaboration of the reflexive argument in the direction of a theory of capitalist culture, with special attention to the currently popular topic of legal pluralism.
2. CUSTOM AND THE FALLACY OF THE AULE.
Pierre Bourdieu's Outline of ~ theory of Qractice [ 1972] has not o~ yet ottrocted much ottention in legal anthropology [but see Pyle 1981). Same have tried to adopt his theory of practica [Bentley 1984), and others ottempted to develop a comporoble theory [Comaroff and Floberts 1981), but none of them succeeded in understanding this theory completely. This is the result of the foct thot they ignored the ref lexive aspect of his theory: i t does not just describe the activity of ethers, but also our own octivities os onthropologists. Bourdieu's theory of practica cannot be adequately applied without these reflexive insights1 • I wil 1 start with
outli-ning Bourdieu's "Outline of a theory of practica" and the identification of the "fallocy of the rule" thot fellows from i t . This will be followed by a consideration of the meaning of these insights for legal anthropological theorizing. The implications of these insights for the distinction of different modes of communication and objectification in legal and social practica will be made clear in a discussion of Moore's fameus essay "Law and Social Change", which concludes this section.
2.1. Bourdieu'& theory of practica; the fallocy of the rula.
All readers of anthropologicol confession literoture are familiar with the image of the "innocent anthropologist"~: the blundering reseorcher disrupting the smoothly running rhythm of native life, offending everyone with his ignorance and treated like a child becouse of his loek of mastery of loco! customs. Most of the onthropologist's energy is consumed by attempts to overcome this state of affairs; and i t is easy to forget that the raad to mostery of the onthro-pologist is very different from the one a native child has to
See for the discussion of Floberts' "follacies of the nota 4, and subsection 3. 1. 2 The expression comes from the
confessions [ 1983). Bentley's and rule" respectively Comaroff and page 5 and t i t l e of Barley's fieldwork
go: he or she lacks the native competences to deal with social life, and therefore has to adept a representation of action that can substitute for i t : a repertoire
2.f.
rules or rOle, a prescription of action in the form of a "semi-forma-lized" set of discourses, a "stage-part" that compensates for the lack of practical mastery of native affairs.It is significant that "culture" is sometimes described as a ~= i t is the analogy which occurs to an outsider who has to find his way around in a foreign landscape and who compensates for his lack of practical mastery, the prerogative of the native, by the use of a model of all possible routes [Bourdieu 1972; 2. Emphasis in original].
However, the idea of culture as a map is wrong: the map is not the territory [Bateson, quoting Korzybski, in 1979; 32-33]. A native does not "follow" a map when dealing with his social environment. A child does not learn the inventive competences necessary for participating in the life of the community by memorizing rules and applying them in practica; instead, "the essential part of the modus operandi w~ich defines practical mastery is transmitted in practica, in its practical state, without attaining the level of discourse" [ Bourdieu 1972; 87 - my emphasis]. The child acquires habits of interaction produced in interaction with and by imitation of ethers, and in this way "embodies" the product of the "objective structures"3 i t was confronted with (as Bateson
hos stated: terological Growing up,
"the unit of interaction and the unit of charac-learning [ . . . ] are the same" [ 1979; 146)) .
a child embodies in this way the "generative schemes" to deal with his social context. Custom is, then, not just
females,
the product of "shared" hunters or gatherers,
schemes: children, males, headmen or sorcerer, all learn different practical schemes, and i t is the interaction of these schemes that produces the regularities of behavior
3 We must note here that "objective" is not meant in the traditional sense of philosophies of knowledge, as refer-ring to unchanging reality. Following Marx, Bourdieu uses the term to refer to the environment the individual is confronted with. "Objective" is, then, not a quality of knowledge, but an element in interaction: the reality the subject confronts and deals with. I will use "objective" throughout the paper in this sense.
chorocteristic of o certain community. Also, these generative schemes (the "habitus", as Bourdieu calls i t [ 1972; ?Bff.]) have nothing in common with "rules" that can be followed: Bourdieu gives the example of the Kabyl of Algeria, where a man who receives o challenge hos no prescribed course of oction; he has to ossess the elements of the situation and make inventive choices between possible courses of action on the basis of an intuitive "sense of honour" that decides, unconsciously, between the "endlessly redundant material" he was confronted with during his learning years. Thus, the "dialectic of objectification and embodiment" that teaches the child how to behave in his environment gradually gives way to a dialectic between the striving or acting subject and the objective structures he or she is confronted with. This is, in general, the thrust of Bourdieu's theory of practica.
The reflexive importonce of this theory of practica becomes clear when we direct our attention to the dialectic of knowledge-production in which the anthropologist is engaged. We can say that he or she has "embodied" the generative schemes of his or her peculiar academie environ-ment to such on extent thot the onthropologist's intention is to tronsform the observed regulorities of social life of the people studied into discourse on these regulorities. The "objective structure" of the situotion is thot he or she connot inscribe these regularities in discourse on the basis of the inventive competences his or her "objects" passes. The~efore, s/he hos to produce a substitute, a "map" of social life, a total overview of the life of the community studied (not so much because the anthropologist wants to attoin practical mastery - though that is during fieldwork a comforting spin-off - but because i t is the necessary form of the product that hos to be presented in the community of peers) .
As an outsider, the anthropologist hos to ensage in a dialectic with the natives, confronting him/herself with their actions or asking his informants about their reasons for them. but as a totalizing outsider, a mapmaker, the anthropologist hos to try to overcome the effects of time and of the consequent changes in his or her situation
brought about by this dialectic of knowledge-production. Even more, anthropologists induce a similar "semi-theoreti-cal" stance in their informants, and together they produce something very different from the actual practica of the community studied: a repertoire of rules, a map, generalisa-tiens supposed to cover all possible courses of action.
But just as the native is usually unconscious of the generative schemes which shape courses of action, so the anthropologists have been unconscious of their own; and this lack of reflexivity has the result that they confuse their product; discourse on the social life of a certain community, with their object: social life itself (this is the reflexive critique that Bentley [ 1984] failed to see in Bourdieu's work4 ) • Because anthropologists tend to forget their own
project, the dialectic of knowledge-production in which they have been engaged, they see their "map", "model", "repertoire of rules", "rOles" or "codes" as the real reason behind all the phenomena they are supposed to cover. It is this reifica-tion which Bourdieu calls the fallacy·of the rule:
To consider regularity, that is, what recurs with a certain statistically measurable freguency, as the product of a consciously laid-down and consciously respected ruling (which implies explaining its genesis and efficacy), or as the product of an unconscious regulating by a mysterieus cerebral and/or social mechanism, is to slip from the model of reality to the reality of the model [Bourdieu 19?2; 29].
Unconscious of their own dialectic of knowledge-produc~ion,
the anthropologists fail to see the dialectics of social life of the people they study, because they think practica is determined by the rule or model. The fallacy of the rule implies a theory of practica that reduces it to an epipheno-menon of the rule or model, a result or spin-off of the only "real" reality: the rule or model itself. The anthropologist erroneously confuses metacommunication with communication, or discourse on the community studied with the life of the community itself. The practica studied "does not attain
4 Bentley's misunderstanding of Bourdieu's reflexivity is clear from the fact that he imputes "shared assumptions" on the basis of his own constructions of the regularities of social life, not on the basis of observed, stated norms
the level of discourse":
what is essential i..QJt!_ without saying because i t comes without saying: the tradition is silent, not least about itself as a tradition: customary law is content to enumerate specific applications of principles which remain implicit and unformulated, because unquestioned [ibidem; 16?].
thus the precepts of custom [ . . . ] have nothing in common with the transcendent rules of a juridical code: everyone is able, not so much to cite and recite them from memory, as to reproduce them ( fairly accurately) [ibidem; 17).
The last two quotes emphasize the relevance of Bourdieu's werk for legal anthropologists: we can say that the rules of law "come with saying" while the regularities of custom do not. In fact, the distinctive character of custom lies not in its "precepts" (these have to be formulated by the anthropo-logist), but in the fact that i t is done, not stated. Custom must be "said" by the outsider who observes practica and then "totalizes" i t , abstracts the regularity from i t s context. It is a fallacy of the rule to say that custom is "a body of norms governing practica". Laws, on the contrary, are written and explicitly invoked. From this can be concluded that laws are fundamentally different from customs.
2.2. The fallacy of the rule in legal anthropology.
A lot of legal anthropologist thought otherwise, however. They argued that law is custom, but only a special kind of custom, distinguishad the power of sanction, the fact that i t is "restated", or soma such special mark. They disagreed permanently over what "law" really consisted of, but were remarkably unanimous in regarding customs as rules or norms followed in practica. Comaroff and Roberts have contrasted the most obvious of these ideas, what they call the "rule-centred" approach, with another, "processual" approach:
the rule-centred paradigm is grounded in a conception of social life as rule-governed and of normal behavior as the product of compliance with esta-blished normative precepts [ . . . ] The opposed stand-point [ . . . ] envisages man as a self-seeking being, whose willing cooperation with his fellows is an expression of enlightened self-interest [ . . . ] The
analytica! corollary of this is to seek the dynamics of order in the social process itself [ 1981; 5).
Though its history goes further usually connected with the two
back, the opposition is digures of Malinowski and Radcliffe-Brown. The latter is the anthropological oncestor of the rule-centred opprooch, which can olso be troced to his own intellectual oncestor and Malinowski's scapegoat, Durkheim, who
by o "totolity
saw order in primitive society oso constituted of beliefs ond sentiments" shared by the members of the group [ 1933; 27). Practica is seen as the result of this obedience to shared beliefs and sentiments.
Malinowski was very critica! of this doctrine of "automa-tic obedience" to customs, and in his work we see the first indications of the "processual" opproach. This wos morked, however, by inconsistencies and contradictions and repeated shifts of theoretica! position [see Schapero 1957; 139' 147) . While on the one hond the rules of civil low were "kept in force by a specific mechanism of reciprocity and publicity" that was driven by self-interest [Malinowski
1926; 66), on the ether "civil low [ . . . ] comprises all the rules governing sociol life" [quoted in Schapero 1957; 140 -my emphosis]. Attacking Durkheim, Malinowski is very
sensi-tive to the importance of reciprocal practica in the consti-tution of sociol order, but in a less combotive mood he describes these activities as governed by rules [see also Malinowski 1926; 39, 51, 55)
was not yet fully articulated.
The "processual" approach
The fallacy of the rule was a dominant custom in "rule-cent red" approaches in legal anthropology: Hoebel, for instance, though at first hesitant to decide whether "custom" is a "norm" or a "way" [ 1954; 14), shows his ignorance of the difference between practica and discourse in his search for "legal postulates" in all cultures, even where, as in the case of the Eskimo, he hos to acknowledge that these were never "postulated" by the natives themselves [ibidem; 68). Gluckman produced a "corpus juris" and a norm of the "reasonable man" where clearly none existed [see 1955; 93], and the follacy of the rule is clear from his statementa that "failure to maintain norms of a relationship sets up
disturbances in the relationships themselves" [ibidem; 78]. The views of Gluckman's critic, Paul Bohannan, seem to have been written to f i t Boudieu's critique: "A norm is a rule, more or less overt, which expresses 'ought' aspects of relationships between human beings. Custom is o body of such norms [ . . . ] octuolly followed in practica much of the time" [ 1965; 45) Finally, in those recent works where law is seen os "locol knowledge" [Geertz 1963; 167 ff.] or "indigenous social theory" [O'Connor 1981), social practica is not even on epiphenomenon anymore; i t has been deleted altogether (for the reduction of the object of anthropology to discour-se, see Asad [ 1979)) .
One can imagine that by stressing the enlightened self-interest of actors directing the interaction in judicia! affairs, the processual approach skirts the dangerous issue of the fallacy of the rule. However, this is hardly if ever the case. Such a founding father of the processual approach as Malinowski slipped, as we have seen, regularly into the idiom of the fallacy of the rule. This is also the case with ether proponents of the processual approach, as is particu-larly well illustrated by a passage from Gulliver's fameus essay on the Ndendeuli:
The Ndendeuli say and sa, for brevity and with caution, may the anthropologist that men expect help from one another and accede to recruitment of an action-set to assist and support Ego because of their kinship relation with him. A more realistic explanation, however, is that men are willing to help Ego, and to become involved in collectiva oction, because of the help tney
tnemselves have
received and expect in the futura to receive from Ego. The principle of reciprocity is crucial. Kinsmen are in practica these more or less regulorly help a persen; these who do not are in effect not kinsmen [ 1969b; 30-31 - my emphasis].The Lévi-Straussian wording of the passage states that the "law" of reciprocity is somehow more "real" than the practica from which i t was abstracted. It is, again, the argument that regularity is the result of a model, low or (in this case) principle that underlies i t a description deemed more "realistic" thon that of the participants themselves. This fallacy of the rule is echoed by several ether proponent& of the processual approach: Nader and Yngvesson state that the
description of sociol processes deeply embedded in sociol contexts is concerned with
modes of perceiving" and rul es of the game" [ 1973; Moore [ 1973; 744) talks gome"15 •
"the mapping of certoin shared "an understanding of the underlying 886, 887 - my emphases]. Similarly, of the "operotive rules of the
The follocy of the rule is, however, not the centrol issue in the "processuol" approaches. We con soy that the "principle of reciprocity'' that is sa crucial to Gulliver is derived from the enlightened self-intarast of the portici-ponts: i t is odvontagaous for tham to "follow" this prin-ciple. This touches Bourdieu's orguments, toa, os ha soys thot the inte~ests of the octors are the "primary principle" from which sociol interaction must be viewed, and the symbolic process is by implication delegoted to a secondory plone [see Bourdieu 1972; 6, 76]. If we return to Gulliver's exompla, we see thot he thinks his explonotion is more rao-listic than the native one, because kinship terms do not "really" refer to kinship relations, but to werk-alliances insteod. Gulliver's cleorly is an ethnocentric point of view in the sen se that ~ woy of judging the reality of kinship terminology ( i.e. thot i t has to refer to octual bloed- or marrioge-relotions and not hing e 1 se) is taken to judge the Ndendeuli custom (i.e. thot "kinship is 0 short hond woy of expressing mutually advantogeous interaction" [Gulliver 1969b; 31)) By denying the meaning of the Ndendeuli use of kinship terms in favor of our own, the farmer becomes necessarily "fictive" because the latter is more "real" - ~ result of Gulliver's failure to see that "interests" are embedded in the customary and symbolic processes and can, for instance, be expressed in terms of kinship terminology. Interests that are not expressed in i t , may not be perceived and therefore do not exist: they are simply categories of
5 For an extended discussion of Moore's fallacies of the rule, see the next subsection ( 2. 3). Though they ara not directly lagol anthropologists, from Stanley Moora's account of Marxian thaories of low in primitiva society i t oppears that bath Marx and Engels did not distinguish rulas and regularitias [ 1960; 654-5).
the researcher, having only epistemological but no ontologi-cal status.
8ourdieu's formulation runs the some risk of reifying the observer's perception of the other's interests as the "primary" causa of sociol interaction. If we take his "theory of practica" seriously, i t has to be clear that the partici-pant in o certain social process has learned the competences of participating, including the perception of his interests, in the "dialectic of objectification and embodiment". That implies thot interests - in the way the participants perceive them are always embedded in ether "embodied" schemes, symbolic, emotional, or physical, and act with and against their realization. Therefore i t is an arbitrary and ethnocen-tric judgment to say that interests are primary: as perceived interests, they are not, and os interests imputed by the observer they imply again an underlying regulating mechanism that is somehow more
themselves. "Objective interests" politicus, and his dialectica! the implementation of strotegies.
real thon the regularities reduce man to a homo interaction with ethers to "Human motivation is now attributed to the crude monocausal principle of maximization, and social control is viewed as an epiphenomenon of strategie interaction" [ Comaroff and Roberts 1981; 16]. It implies a theory of practica that is as meagre os the one characteris-tic of the follocy of the rule.
2.3. The dialectic of embodied and objectified history.
The follacies of the rule and of interests are bath reductio-nist and anti-dialectica!; influence is exerted only in one direction, from rules or interests on social practica. I have shown that we have to distinguish between rules and practica, that the latter is not simply the result of the farmer and can not be reduced to i t . The relation between
the two is dialectical, social practica (and
laws are relatively autonomous from vice versa). Sally Falk Moere hos pointed this out in her fameus essay "Low and Social Change", in which she develops the idea of the "semi-autonomy" of social fields from the attempts to regulate them. In
discus-sing this essay, I wil 1 try to show thot the relativa or "semi-outonomy" of the low can best be conceived of on the basis of the difference between customary regularities and legal rules; o difference in mode of objectification. In Bourdieu's terminology, we can say that customary behavior is objectified in human bodies, in their inolienoble potterns of hobituol reoctions; embodied history. On the ether hond, human societies produce olienable objectificotions thot are relatively autonomous from whot is done with them, houses, books, tools, etcetera; objectified history. The dialectic between embodied and objectified history [see Bourdieu 1981; 305) is essential to understanding the dialectics of customs and laws, and thus, of the relativa autonomy of the low.
The explicit intention of Moore's essay is to dispel the "legalistic" fallocy of the rule: the idea thot by means of legislotion, legol innovotion, one con change traditional customs or loco! practica what is called "legal engi-neering". By giving examples of the influence of legal rules on the customs of the clothing industry in the United States, and the woy the Chaggo of Mount Kilimanjaro raspend to official legislation, she shows thot this non-diolecticol view of low is wrong and thot legol innovotion does not engineer social chongee. Against this legalistic view she poses the interaction of low and custom in society: "A court or legislature can make custom law. A semi-autonomous field can make low its custom". And bath con refuse the influence of the ether. Moere thus hos to distinguish between laws and customs: the first are rules consciously mode by formol ogencies, and the second "rules" that evolve spontaneously in certain sociol settings [ 19?3; ?44). The influence of legal rules is thus dependent on mediation by regular practica.
The ideo of the mediation of (newly mode) rules by the local customs to which they are maant to opply is not new: already in 1906, Sumner wrote that "legislation cannot make
6 Following Lévi-Strauss [ 1962; 26 ff.] the bricolage of social change.
mores" [quoted in Aubert 1966; 117) and Aubert has tested this assumption by an inquiry into the socio! influence of the Norwegian Housemaid Law of 1948. He concluded that
the law was, at least for some years, ineffective in the sense that actual conditions of werk remained at vorionce with the norms loid down, and olso in the sense thot even conformity to the lego! norms was rarely due to the influence from the low [Aubert 1966; 121].
However, this was s t i l l research into the relativa "effecti-veness" of legal engineering. What is new in Moore's approach is the ideo that the community evolves its own ordering customs vis-~-vis the laws: not just non-legal ordering, but sometimes even manifestly illegal. This community is what Moore calls the "semi-autonomous social field", that is defined and delimited for research purposes by its capacity to "generate ( i t s own) rules and coerce or induce compliance to them" [Moore 1973; 722] . Thus, the clothing industry in the United States is supposed to be governed by union laws specifying the hours and days of werk of its employees. However, the objective situation calls for a more flexible arrangement: the vagaries of fashion necessitate arrangements in which the employees of the jobber or contractor are able to work for a much langer period when necessary, which is compensated by free hours when times are slaak. The jobber and contractor therefore cultivate a friendly relationship with the union representative, strengthened by gifts and ether tekens of friendship, so he will look the ether way when union laws are braken in a time of hard work. The same customary, non-legal or illegal relationships are cultivated between jobber and contractor, when, for instance, the second gives interest-free loans by not insisting on the immediate fulfilling of contractual obligations [ibidem; 723-729].
However, Moere formulates these customs of the clothing industry in terms of the fallacy of the rule. The relativa autonomy of legal rules from the social setting in which they are meont to apply is not just the result of regulari-ties of social practica, but, Moere says, of the "generation of rules" by the social field and "coercing or inducing compliance to them". The semi-autonomous social field thus becomes a substitute legislative and law-enforcing agency,
instead of a network of people trying to order their mutual affairs. From Moore's description is nevertheless clear that clothing gifts and the relationships between the members of the
industry are maintained, not by rules, but by
ether friendly gestures, by means of a "sense of friendship" as indeterminate as the "sense of honour" of Bourdieu's Kobyl. Also, foilure to maintain friendships is not "punish-ed" by the detour of a judgment about a braken rule, but by direct action: ostracism from the network of relations, resulting in economie ruin.
The difficulties of conceiving of relativa autonomy of and from the law in terms of the fallocies of the rules are especially clear when applied to the legal establishment itself. A consequence of Moore's views would be thot we con conceive of the community of court-officials in terms of o "semi-autonomous social field": as Blumberg hos shown, the octions in court do not simply reflect the rules of low and due process that the law-student has leorned at the univer-sity. On the contrary, the novice will find that he or she has s t i l l to learn the customs of negotiotion between lawyers ond judge in criminal, and among the lawyers them-selves in civil cases. This custom is created and sustoined in the face of features of the objective situotion, such as the common culture of ottorney, defense lawyer and judge and their common interest in a speedy handling of cases. the student of law has to learn in practica the custom of using due process and the rule of law as legitimation and cover-up for "bureaucratie practica" [Blumberg 196?; 330), and the customs of questioning a witness in the appropriate ways (in order to, for instance, convey information to the jury directly, with only apparent interference of the witness [ see Woodbury 1984; 215]) . Even more,
Courts ore the site of administrotive processing, record-keeping, ceremonial changes of status, settlement negotiations, mediation, orbitration, and "warfare" (the threatening, overpowering, and disobling of opponents), as well as of adjudication. Indeed, in most courts moves into the formol odjudi-catory mode are for purposes ether thon securing on adjudicated outcome [Galanter 1981; 3).
situation where he s t i l l has a lot of - unwritten - "rules" to learn.
This puts the theory of "semi-autonomous fields" in the paradoxical position that i t has to account for a social field thot "generates its own (customary) rules" to
circum-vent the rules that i t is supposed to opply and generote in the first place! This confusion is, of course, the result of the failure to distinguish between modes of objectification; while the rules of law are objectified in law books and in their invocation by judge and counsel, the customs of court life are objectified in the actions of the members of that particular legal community. We can say that the learning process (the dialectic of objectification and embodiment) of the law student at the university is very different from from the one during court practica; while in the first the most important mode of objectification is the code of law laid down and commented upon in writing, in the second i t is the octions of the court personnel thot confront the learning subject7 •
That implies that a distinction between modes of objecti-fication is crucial for a consistent view of the dialectics of laws and customs and the relativa autonomy of law. When regularities are objectified in the actions of, for instonce, the members of the clothing industry, thay are ephamere but directly influence the maintanonce of social order (or, better, they ~ social
objectified in writing order) . are, on Aegularities the contrary, expressed and more permanent but also leas influential in themselves, because, as writing, they can be ignored, laid aside, reflected upon, etcetera. They are about social order.
Laws are autonomous from existing social practica because they are objectified in writing and not in human action, not embodied. The important epistemological
differen-? Thus, we must distinguish between "embodiment" as objecti-fication of social regularities and "embodiment" as a learning process. They are related, but in the first we deal with "embodied" social structures that confront the acting subject, and in the second, with the learning process of the acting subject itself.
ce is the mode of objectification: embodied customs are merely intuitively known or not at all conscious, objectified laws are consciously and explicitly known. To say that law is custom or "shades imperceptably into it", to postulate a merely quantitative difference in "degree of explicitness" (os, for instance, Galanter seems to do [ 1981; 18)) is to ignore this very significant difference in quality, ond thereby to confuse regularities of social behavior with descriptions of these regularities, what is done with whot is stated obout whot is or should be done.
At this point one can justifiably object that this is a far toa simple view on the matter: I have tended to reduce social practica in generol to the written legol rule on the one hond, ond the embodied and unreflected social regulori-ties on the ether. I have os yet not disucssed bahovior thot falls short of the formulation of specific legal rules: descriptive and normative communication. Even more, to point to a difference in modes of objectification does not exhaust the differences between regulor behavior and statements obout regulor behovior: we ore olso dealing with a difference in what Boteson (following Aussell ond Whitehead) hos colled "logica! level" or "logicol type".
As seen from the perspective of the medium, customs differ from lows in the sense thot the first ore embodied and the second are written. But os saen from the perspective of the massage, o
custom to which i t
low is of a higher logicol refers or is maant to
~
thon the
apply; i t is metacommunicotion. The difference between communication and metacommunication is o difference between "oction in context and oction or behovior which defines context and makes context intelligible" [Boteson 1979; 129) And oction thot defines context con itself be defined by o metomessoge of o higher logicol type: "In all such cases, the step from one logica! type to the next higher is a step from information about on event to information obout a class of events or from considering classes" [ibidem; the class to 13?. See also considering Bateson 19?2; the closs of 279-308 andWilden 1980; ch. 4 and 5).
These bosics of epistemology specify the differences with which we are dealing here: the steps from customary behavior to communicotion about i t , and to communication about this communication; and the steps from acted to stated to written regulorities. The specificotion of these differences in the special case of legol anthropology is the subject of the next section.
3.
MODESOF COMMUNICATION.
In the preceding section I have shown that when we apply reflexive insights we cannot continua to spaak of rules or customs "governing" practica. I tried to argue that distinc-tions between modes of objectificotion and levels of logicol typing are useful and necessory to distinguish legol rules from customary behavior, objectified from embodied history. Though this is an improvement on the legalistic of idealistic forms of (not only legal) anthropology that tand to reduce social life on the "normative order" or "system of shored meanings", i.e. discourse on social life [ see Asad 19?9], i t is s t i l l too simple. In legal anthropology, we are not dealing with one unit of analysis called the "normative order", nor are we solely occupied with legol rules on the one hond and customory behavior on the ether. Usually, "rules", "norms" and "customs" have not been sharply distin-guished os they seemed to point to the same unit of onalysis, but I think i t can be profitably argued that distinctions between modes of objectification and logica! levels epistemologicol distinctions - show that these distinctions are important to make. Epistemologically, customs, norms and rules belang to different modes of communication that, though they are similor in soma respects, differ fundamen-tally in mode of objectification or logical level. Whereas in the previous section I was mainly interested in distin-guishin& customary behavior from legalistic communication, I intand to concentrate solely on communication about behavior in this section. Therefore I will skip the first fundamental differences: in modes of objectification, from the body to language, and in logical level from behovior (which is, I agree with Bateson, itself communicotion [quoted in Wilden 1980; 1]) to tolking or writing obout behavior (thus, a form of metacommunicotion). Instead, I will concentrote on the difference in logica! type between customary communication about behavior and normative communication about behavior, and on the difference in mode of objectification between the spoken norm and the written rule.
logica! type and the critique of the idea of "implicit norms" thot fellows from a failure to see this difference. The second will deal with some differences between norms and rules that follow from common-sense and linguistic considera-tions, and in the third subsection I will try to relote these differences to the difference between the spoken ond written modes of objectification of longuoge, ond coma to the conclusion that the idea of the rule is specific to a literate civilisation. As this may all seem to be a singular-ly academie preoccupation, I will try to point to the historica! relevance of these epistemological ideos in the fourth section, which will deal with the ideological role of the fialures to distinguish modes of objectification and logica! types in the "colonial encounter" [ see Asad 19?3).
3.1. Abstraction and Control.
The concept of "norm" has been variously used in legal anthropology: "It is sometimes a euphemism for law, or a synonym for custom, or bath, or neither" [Greenhouse 1982; 59]. Only recently have soma legal anthropologists tried to formulate this concept in a less ambiguous way [ibidem], or tried to find out how norms are used in practica. Comaroff and Roberts have adressed the latter topic [ 1977, 1981; 84 ff. See also Greenhouse 1985) . Several cases from their fieldwork among the Tswana are used to ascertain when norms are explicitly invoked in disputes and when they "remain implicit". The cases show that in the presentation of his case to the Tswana public the complainant seldom uses explicit norms. He is content to state the facts in such a way that their arrangement "will convey to ethers present his paradigm of argument". A defendant or accused who feels that this arrangement is right will not question i t , either. Bath stay within the confines concrete statements about events; their discourse is purely contextual: customary communica-tion.
Only when the defendant or accused wants to present a competing paradigm of argument, he states norms explicitly. In this way he tries to assert control over or change the
terms in which debate is proceeding [ 1981; 103]. Judges invoke norms explicitly when they are compelled or think i t is necessary to decide between competing paradigms of argument [ibidem; 105].
However important these conclusions are, they suffer from o lack of raflexivity that projects norms into o situotion whara they do not axist. When Comaroff and Robarts spaak of "implicit norms", they forget thot these "norms" ara constructions of an observer ottempting to organise an interoction for which he lacks the customary intuition to understand i t in itself. It is the observer who formulates these norms to get a grasp on this situation; Comaroff and Roberts commit the fallacy of the rule (or, in this case, of the norm) when they project these norms into the situation as ontological categories, even as causol ogents:
The fact that normative propositions should give form to the arguments formuloted in the course of a dispute is understandable [ . . . ] Without i t (the "normative repertoire"/PP) poradigms of argument could not be constructed and no comprehensible universa of discourse would be established [ 19??; 105) .
Comaroff and Roberts fail to see a crucial difference: between the different modes of communication in which norms are and are not used. A norm is an abstraction, formulating the "essence" of a certain set of communicated facts in order to define and possibly change i t . That implies that a statement of a norm is communication about a communication, metacommunication. When norms are used, one talks about a situation, when they are not, one communicates the situation itself. It is cleorly a difference in logica! level or logical type,. A "paradigm of argument"2 is a way in which
argument is ordered, not an argument obout its ordering.
Arno [ 1985) develops a related argument on the basis of Bateson's work. Though he is conscious of the difference in logical levels, he fails to apply this consistently, as is shown by his equation of paradigms of argument ("ac-counts of problematic behavior") with norms to control argument ( "control communicotion") [ 1985; 43. See olso Bilmes 19?6; 55 and nota two of this section].
2 I take "porodigm" to be an epistemological concept without direct ontological basis in fact. It refers to a set of basic assumptions imputed to a certain social field by its
If we take care not to muddle this distinction, the conclusions of Comaroff and Floberts are particularly impor-tant: they show that normative metacommunication is used precisely in these situations where the one who formulates norms trias to grasp a total situation: the anthropologist who lacks that practical mastery to understand i t
intuiti-vely, the disputant who tries to change i t , or the judge who has to define i t in the end. In all cases, abstraction - the use of a norm - is a means of control, an attempt to place oneself above the situation at hand and to direct i t from the top downwards. Thus, the efficacy of explicitly formula-ted norms "does not lie - or does not only lie - in giving voice to what otherwise literally goes without saying". They take on "tactical importance" and are powerful in a way that customary communication is not [Greenhouse 1985; 100-101] . This power is first of all based on their difference from customary communication with so-called "implicit norms". It is, again, not just a difference in quantity (in the sense of "more (or less) explicitly formulated") but one of quality: explicit norms are not a rendering of an otherwise unformulated paradigm of argument but an interpretation of an account of problematical events. This interpretation can be used to change the paradigm of argument and thereby influence the transformation of a dispute by and before an audience ( what Mat her and Yngvesson [ 1981)
sion" of the subject of the dispute) .
call the
"expan-This power of explicit norms is not guaranteed: the expansion of argument i t provokes may be ignored, disclaimed or otherwise made harmless by opponents and third porties (aftar all, the stated norm is relatively autonomous from its practical implementation). Even more, i t may in a sense be less powerful than a communication of problematic events that remains bound to concrete happenings and the concrete
observer on the basis of observed regularities. It therefore does not refer to ideas used in practica but to ordered practica itself. These orders are usually unconscious and therefore possess a "matter-of-factness" that resists change. A paradigm is a regularity of communication, not a corpus of metacommunicative rules.
relationships in which i t is expressed, as i t proposes a paradigm of argument of which the listeners are not con-scious. Though we can say that all norms and rules have the intended relationship between abstraction and control in common, this is far from saying that there is a transcendent relation between normative communication ond power. When norms are powerful,
transcendent power;
however, they usually accepted norms imply more
suggest such thon just on interpretation of a set of problematic events, they go beyond the speaker bound to place and time; the "collectiva good", "society in the abstract" or "state" [Greenhouse 1985; 100-101), a "totalized" situation of a higher level of justification than that of the speaker alone (see also subsection 2.1). Indeed, without this connotative surplus o"f explicit norms the neutrality of for instance a mediator could, when questioned, hardly be upheld [ibidem; 102)3 •
Gulliver is right in arguing that "there is greater reliance on, appeal to, and operation of rules, standards and norms where adjudication rather than negotiation is the mode of dispute settlement" [ 1969a; 18]. This is not just because compromises are best reached without conflicting norms, and negotiations always tand to conciliation and compromise; Starr and Yngvesson [ 19?5] show conclusively that such an idyllic situation does not exist. Neither must the use of "rules, norms and standards" be exclusively related to the existence of an authority that determines which norms apply (as Gulliver [ibidem] seems to do) . Instead, one should relate the use of norms and rules to their abstract nature, implying an attempt to gain control over a situation. Thus, in adjudication, a judge has to determine which definition of a situation (in the case of competing interpretations) is the "right" one. But during negotiation, toa, a participant may state a norm in order to get the audience to accept his view of the situation. We can expect that in a situation where conciliation is the goal, the use of norms is hardly
necessa-3 Gulliver [ 1969b; 43, 45) gives an example of an attempt to "mediate" (that is, accuse someone of witchcraft) that is not accorded authority by the participants in the dispute.
ry because the disputants want to agree over the paradigm of argument. But in negotiations where one of the porties attempts to gain control over a resource or audience [see Starr and Yngvesson 19?5; 558), norms may be stated when this control is disputed and participants aim at a zero-sum outcome. And norms are not just potentially powerful meons of deciding for or against on interpretation of a case; they function in ether ways as well. The norm of Lozi judges thot they try to reconcile the disputants - a norm shared by most community members and unquestioningly adopted by the anthro-pologist (Gluckman 1955; 21) - while the majority of cases they decide on has a zero-sum
19?5; 556-?] may be a case in
outcome [Starr and Yngvesson point: to say one preserves social relationships, guards the "collectiva goed" or soma such "totality" referred to above, is a powerful way to justify oneself while at the soma time putting pressure on ethers.
3.2. The distinction between rules and norms.
In the previous subsection I have argued qualitative difference between customary
that there is a communication in the account of problematic events, and the norms or rules thot interpret or clossify this account: a difference in logical type pointing to differences in social usage. Though both norms and rules have this logical level and the subse-quent intention of control by abstraction in common, they are s t i l l , I think, different.
The Penguin dictionary defines a rule as an "established principle of guidance and control" or an "accepted convention to be followed" [ 1969; 615]. A norm is described as a "recog-nized standard" [ibidem; 486] This may not seem to amount to much, but we can deduce two points from i t : first, that both rules and norms are "established", "accepted" or "recognized", that is, explicitly acknowledged conventions. In the second place we can note the fact that in the descrip-tion of "norm", the reference to "guidance and control" or the "following" of the precept is absent. Bath "guidance" and "following" imply that the rule is E.!:.,!.- or proscriptive.
It exists before the
act
it
is
supposed
to
regulate. As
Greenhouse
states,
a
rule
can
be
breken,
a norm cannot
[ 1982; 61].
Rules are
objectified in
such a
way that they
exist
before
the
act
and
can
be
used to measure it. In
contrast, norms are used as reasons
[ibidem], to
be applied
in different
and changing
situations. Therefore, norms must
be contextual, able
to
shift
referents
from
situation to
situation. Norms
are "adverbia!"; "o rule is a noun", it con
stand on its own [ibidem).
These rother condensed statements
must
be
clarified. A
particularly
apt
illustration
is
the use of
norm~in what
Greenhouse calls "duelling with scripture" among the Baptists
in Georgia
[ibidem;
64) .She describes
asituation in which
a teenager accuses another
of
hypocrisy
in
urging
one of
their friends
to attend
Sunday School,
by quoting from the
Bible: "Judge not lest you
yourself
be
judged".
The ether
rep lied: "But
if
Isee my
brother wonder from the way, and
do not call him back, am I truly a Christian?".
Compare this to two examples of what are clearly "rules":
a
passage
from
Roman
Catholic
Canon Law [Canon
1075-
my
translation from the Dutch original] :
No lawful marriage can be contracted between:
1. These, who have
committed adultery
and have promised
eachother
to
contract
a
marriage
or
have
tried
to
contract a marriage,
also
by
means
of
an exclusively
civil action,
while the
same lawful marriage continues.
2. These, who ...
Or a similor passage from
the
Dutch
crimina!
code
of law
[Art. 249] :He,
who
fornicates
with
his
infant, stepchild or
fosterchild, his pupil, a persen under
age entrusted
to his
care or
upbringing, an
underaged servant or
inferior, will be punished
with
imprisonment
of at
most six years or a fine of the fourth category.
The
most
telling
difference
between
these
statements of
norms and rul es is, to me,
their
use
of
pronouns:
in the
case of
the norms from Holy Scripture one refers to "you" or
"I", but in
the
legal
rules,
reference
is
made
only to
"they" ("these,
who ... ") or
"he". In ether words, the norms
use linguistic shifters while rules
are
not
adressed
to a
concrete
persen
but
name
an
absent
third.
Shifters are
"lexica! items that change their meaning, taking their colour
from the context, [ Seidel 1975; 207] .
or from the situation of the speaker" "They refer only to the reality of only be defined, that is, indicated, in discourse and can
terms of location in [Fernandez 1965; 16]. persen' is opposed to
the particular instance of discourse" In contrast, "the grammarian's 'third
the first and second person as a nonparticipant in the dialogue.
'persen'; i t is really the verbal
The "'third persen' is not o form whose function i t is to express the non-persen" [Fabion quoting Benveniste, 1983; 65]. The "third persen" as o "non-persen" is absent, outside the dialogue between disputants or lowyers. It contrasts sharply with the concrete indications of "I" and "you" [ibidem) The linguistic shifters only have meaning in a concrete context, indicating the persen to which they refer. The so-called ''third persen" is, on the contrary, depersona-lized and decontextuadepersona-lized; i t can refer to anyone. We could say thot the third person is an object, a "persen" reified. In contrast, the first and second persen pronouns indicate the dialectical situation of dialogue [see Fernandez 1965; 18, 22 on "reversibility" and "turn-taking"].
Because a norm is contextual, i t connot have predictive volue outside context. When its referents connot be indica-ted, i t is just an unused proverb among many ethers. In contrast, the rule has meaning in itself; i t is objectified meaning, removed from context and therefore predicting all situations to coma. The norm is contextuol, situated in a diologue where on "I" holds up ("objectifies") a norm t o a "you". In this sense i t is a "reoson": i t is on attempt to justify its speaker and by justifying him or her i t gains (partial) control of the situotion. While the norm can be compared to the weopon used in o duel (cf. Greenhouse's "duelling with scripture"), thQ rule more resembles a referee of a fencing ossociation.
3.3. Epistamology of writing.
In the judicial processes that form the object of the legol anthropologists' attention we can
of objectification: writing, speech
distinguish several modes (varying from relotively
