[page 21]













                                                              Gordon R. Woodman





The concept of Legal Pluralism is increasingly used in legal anthropology, the sociology of law and legal theory. Critical reflection on it may therefore assist research and understanding in these fields. In the process of trying to elucidate a concept, we draw distinctions, formulate general propositions, and consider criticisms of these distinctions and propositions. All of this increases understanding, and frequently suggests hypotheses for further research.


This paper examines certain literature which has discussed the concept of legal pluralism. The object is not to survey comprehensively a particular body of literature, such as that which has used the concept in theoretical or empirical work. Even less is the object to survey all the work which relates to what may be classed as situations of legal pluralism. The object is merely to enable the concept to be used with as much precision as possible.

The discussion starts from the use in the literature of the expression ‘legal pluralism’, although without excluding the possibility that some writers use the same concept but refer to it by a different term. The first section considers [page 22] discussion of the concept prior to the 1980s. The next discusses the developments ensuing from Griffiths’ paper “What is legal pluralism?” (1986). Subsequent contributions are then examined, and the paper finishes with a brief discussion of how we may today be able to elucidate further the field of research designated by the expression.



Discussion of the Concept before the 1980s


In the arenas of debate with which we are concerned the dominant theories about law in this period were promulgated and accepted within the legal professions. In some aspects these theories were antagonistic to all notions of legal pluralism, although they were more opposed to some than to others. These antagonistic aspects are adequately subsumed under the title of the ideology of legal centralism, introduced by Griffiths (1986), who summed it up as the view that:


law is and should be the law of the state, uniform for all persons, exclusive of all other law, and administered by a single set of state institutions (Griffiths 1986: 3).


This view was expressed or implied in the writing of most western legal theorists of the nineteenth and twentieth centuries until roughly the last 20 years. It seems clear that still today the tenets of legal centralism are accepted by the great majority of legal theorists. The prevalence of legal centralism has had as its corollary a slowness in the development of a concept of legal pluralism. This may explain the relative lack of elaboration in the versions of the concept which have been advanced. It may also account for the manner of advocacy with which they have been advanced. The proponents of these ideas had reason to believe that they were liable to be rejected without serious consideration. The concept had, as Griffiths has described it, a “combative infancy” (1986: 1 - the period referred to is the late 1970s or early 1980s). A purpose of his article was avowedly “simple debunking” of opposed views (1986: 5). Combativeness is self-generating, and the debates have tended to continue in this style. It is fun, but one may wonder whether the process always takes us to understanding as rapidly as possible.


We may separate the direct discussion of the concept of legal pluralism, largely by lawyers, from the analyses of empirical investigations, largely by anthropologists, which are relevant to understanding of the concept.


[page 23]  

Legal studies of the concept of legal pluralism


Three writers made important contributions to the emergence of the concept. Their versions of the concept are broad and not on the whole precisely delineated. That indefiniteness may have been a merit in early formulations of the concept, since it left open possibilities for later scholars to refine it in ways appropriate to their own concerns. That these three writers were legal scholars rather than anthropologists or sociologists may have a bearing on the nature of their views. Their arguments have been discussed by Griffiths (1986) in detail. Generally my readings of these writers coincide with those of Griffiths, although we differ as to the relative significance of certain aspects of their ideas.


I try to set out the essential elements in each writer’s notion of legal pluralism. This enables them to be compared with each other and with the writers considered later.



(a) Gilissen


John Gilissen edited a volume of essays published under the title Le pluralisme juridique in 1971. His introduction to the volume does not present a definition of the expression, but indicates some propositions contained within the implied definition of the field of the collection.


1.          The category ‘law’ includes non-state law. Gilissen suggests that the use of the concept of legal pluralism entails a recognition of non-state law. He contrasts legal pluralism with monism, which he associates with the view that the state is the sole source of law. He emphasises the legal nature of customary law, as a form of non-state law.


2.          There is a form of state law pluralism. His examples suggest that, while state law may be one element in an instance of legal pluralism, there are also instances of pluralism within the law of a state, for example, when different rules, standards of proof or judges operate with respect to commercial issues from those with respect to other issues. I call this ‘state law pluralism’. It may be inferred that in Gilissen’s view it is the less significant form of legal pluralism.


3.          The constituent elements of legal pluralism are ‘laws’. Each element of an instance of legal pluralism in his view is a ‘law” (droit). By this he seems to mean primarily a body of rules, although the term may include institutions such as courts, as his observations on the instance of commercial issues show. It is clear that a ‘law’ need not be so [page 24] extensive as to regulate all social life comprehensively, although his examples are of substantial bodies of norms, or institutions which apply substantial bodies of norms.


These three propositions do not fully define a concept of legal pluralism. The only general conclusion to be drawn from them is that various types of laws exist. Everyone who uses the expression believes that legal pluralism entails also some sort of contact between different types of laws. Gilissen’s discussion refers to one category of contact - that when two or more laws exist within one state law - and holds that that is a type of legal pluralism, but he denies that all instances of legal pluralism are of this type.



(b) Vanderlinden


Jacques Vanderlinden contributed to the volume edited by Gilissen an Essai de synthèse. He there defines legal pluralism as:


L’existence, au sein d’une société déterminée, de mécanismes juridiques différents s’appliquant à des situations identiques (Vanderlinden 1971: 19).


This in some respects echoes Gilissen’s implied definition, but in others is more specific.


1.          Recognition of non-state law. Like Gilissen, Vanderlinden recognises non-state law. In many of his examples of legal pluralism the two constituent elements are drawn from non-state law and state law respectively.


2.          State law pluralism. Like Gilissen, Vanderlinden considers that there are instances of legal pluralism within state laws, as some of his other examples show. He suggests that instances of legal pluralism may be classified as pluralisme contrôlé and pluralisme indépendant: these seem to correspond to pluralism within state law and pluralism in which state law is one element. He notes further that instances of pluralisme contrôlé are almost necessarily complémentaires, by which he means that the elements can always be reconciled into a self-consistent scheme, if only by an overarching norm which determines which is to prevail in particular situations. In contrast, cases of pluralisme indépendant may, but need not be antagonistes, by which he means that the elements are mutually contradictory. (Vanderlinden 1971: 47-50)

[page 25]  

3.          Constituent elements of legal pluralism. Vanderlinden designates the elements of each instance of legal pluralism mécanismes juridiques. This might appear to be a narrower and more precise conception than Gilissen’s ‘laws’. However, his initial explanation for using the term is merely that he wishes to avoid reference to legal ‘systems’, which would imply structured and relatively comprehensive bodies of law. As Griffiths shows, Vanderlinden’s examples indicate that by mécanismes juridiques he means single rules or clusters of rules or institutions (Griffiths 1986: 12).


Vanderlinden adds two further propositions which do not correspond to aspects of Gilissen’s discussion. By these additions he completes a comprehensive definition.


4.          Each instance of legal pluralism is located “au sein d’une société déterminée”. This specification of the locus of legal pluralism seems helpful. Perhaps inevitably, it poses further issues. It requires a delimitation of the notion of a ‘society’. This is a difficult task. The requirement implies a view of the human universe as divided into distinct ‘societies’, a view which is open to question. These issues are mentioned again below.


5.          It is of the essence of legal pluralism that in each instance the different constituent elements (legal mechanisms) apply to identical situations. The existence of different mechanisms applying to different situations in one society is an instance of plurality of law, rather than of legal pluralism (Vanderlinden 1971: 20). This might appear to require, if legal pluralism is to exist, a contradiction between the constituent elements, either in their respective normative provisions, or at least in the purported scope of application of each. However, Vanderlinden includes among his examples cases such as that where a defined category of officials in a state are in certain circumstances governed by a different body of state law from that which governs the remainder of the citizens. In such cases it is common for each body of norms to define its own scope of application in terms which are complementary to the other, thus excluding contradictions. It would seem that by “identical situations” Vanderlinden means those in which all the relevant categories of facts are identical except for one, the presence or absence of which determines which legal mechanism applies. Thus in the example of state officials who are subject to a different legal regime, “identical situations” exist if the legally relevant facts are the same except for the status of the subjects, who in some instances are state officials and in other instances not. In this argument the ‘relevant’ [page 26] categories of facts are, it would appear, those recognised as relevant by either of the legal mechanisms.[1]


Thus Vanderlinden’s discussion recognises various types of law, and states that legal pluralism exists when, au sein d’une société déterminée, two types apply to identical situations.



(c) Hooker


Barry Hooker’s Legal Pluralism (1975) is an important study in comparative law. The present discussion concerns only Hooker’s contribution to the development of the concept of legal pluralism. Such a contribution is not a major aim of Hooker’s book. Its object is to give an account and analysis of a particular phenomenon in state laws, namely “the systems of legal pluralism... which have resulted from the transfer of whole legal systems across cultural boundaries” as a consequence of certain types of colonialism (Hooker 1975: 1). Having thus identified clearly his field of study, he did not need to formulate and defend in depth a generally applicable meaning for the expression ‘legal pluralism’.


The clearest indication of the general meaning which he would give to the term appears in the statement:


     The term ‘legal pluralism’ refers to the situation in which two or more laws interact (1975: 6).


The book contains various other express and implicit indications. Taking account of these, we may compare Hooker’s view with those of Gilissen and Vanderlinden.


1.          Recognition of non-state law. Hooker also recognises non-state law. At the start he criticises, as did Gilissen, the view that only state law is “properly law” (1975: 1-2). His position emerges more fully in his Chapter I, entitled “Legal pluralism and the ethnography of law”. However, his book does not attempt to survey legal pluralism in general.

[page 27]  

2.          State law pluralism. Hooker like Gilissen and Vanderlinden considers that there are instances of legal pluralism within state laws. His book is in the main an examination of certain aspects of the doctrines of state laws, which he regards as instances of legal pluralism. He claims, perhaps in explanation or justification of his choice, that the social reality today is that frequently state law prevails when its injunctions or interpretations of events differ from those of non-state laws. A critic might question his choice of focus, and his view of the effect of state law in social reality, but these issues do not seem to be relevant to his concept of legal pluralism.


3.          Constituent elements of legal pluralism. The elements of each instance of legal pluralism in Hooker’s conception are ‘laws’, or ‘legal systems’. A law is a “system of obligation”, a phrase which appears to refer to systems of norms. Different laws in this sense are distinguished by their sources. Hooker does not find it necessary to elaborate on this conceptual scheme.


4.          The locus of legal pluralism. Like Gilissen and unlike Vanderlinden, Hooker does not specify that legal pluralism is located within a ‘society’ or any other type of field. In this respect his view is potentially less restrictive than that of Vanderlinden. However, by renouncing this restriction Hooker gives up one possibility of forming a complete definition.


If the location of instances of legal pluralism is not to be delimited, something more than mere coexistence of ‘laws’ (as defined) is needed. This additional factor is provided in the last aspect of Hooker’s concept.


5.          The nature of instances of legal pluralism: the different constituent elements (laws) interact. Whereas Vanderlinden specifies that the constituent elements of any instance of legal pluralism (in his view, legal mechanisms) must each apply to identical situations, Hooker requires that the elements (in his view, “laws”) must “interact”. This feature of Hooker’s discussion enables him, unlike Gilissen, to offer a complete definition. He does not explicitly elaborate on the general notion of interaction of laws. That again was unnecessary for his purpose. The objects of his study are the doctrines of particular state legal systems which consist of multiple ‘laws’. In the particular context of this study the general nature of interaction between laws is obvious.


[page 28]  

Anthropological and sociological conceptions regarding non-state law


The sociologist of law Georges Gurvitch writes of legal pluralism, although he is concerned more with establishing its existence than exploring its detailed meaning and implications (Gurvitch 1935, 1947). A brief summary of his model of sources of law, set out principally in his essay “Théorie pluraliste des sources du droit positif” (Gurvitch 1935: 138-152) may be useful.[2] He argued that the rules customarily referred to as sources of law, “formal” sources such as statute and judicial practice, should be seen as secondary sources. Behind them lay “primary or material” sources, which he classified as faits normatifs, and which alone could confer on the formal sources both authority and effectiveness. Primary or material sources could validate any number of formal sources. To limit the number of formal sources to certain types (such as state enactment - loi étatique) alone had no scientific validity and could only be pure dogmatism. Similarly there was a priori no necessary hierarchy among the formal sources. This plurality of formal sources did not of itself entail insoluble conflicts, anarchy of sources, since they might all be derived from one material source, which would provide the means of solving conflicts. Thus


on pourrait, en principe, admettre le pluralisme des sources formelles, même en restant moniste (par exemple étatiste) quant aux sources primaires, aux ‘faits normatifs’. La véritable opposition entre le monisme et le pluralisme juridique ne commence donc qu’à propos de la couche profonde des sources primaires et matérielles.... (Gurvitch 1935: 145)



But true legal pluralism existed:


[I]l est impossible d’argumenter en faveur de la liaison indispensable entre l’Etat et le droit positif, sans affirmer que l’Etat est le seul ‘fait normatif’. Or, une telle affirmation est contredite, non seulement par toute notre expérience actuelle, qui enregistre l’existence d’innombrables centres générateurs du droit, de foyers autonomes du droit, (syndicats, coopératives, trusts, usines, églises, services publics décentralisés, unions internationales, administratives, [page 29] Organisation international de travail, S.d.N., etc.),[3] elle est infirmée non seulement par l’enseignement de toute l’histoire du droit qui nous dévoile combien le fait normatif de l’Etat a pu subir la concurrence d’autres ‘faits normatifs’, dont il se contentait d’appliquer le droit, mais elle se détruit encore par des considérations théoriques; celles-ci en effet démontrent avec clarté que chaque communion active et chaque rapport avec autrui qui réalisent, de par leur existence même, des valeurs juridiques positives, peuvent s’affirmer comme des autorités qualifiées et efficientes, qui fondent la force obligatoire du droit positif.... (Gurvitch 1935: 146-147)


The discussion then proceeds to propose means of solving conflicts between norms derived from different material sources by establishing an order of priority between those sources, using arguments which are not relevant to the present discussion.


The model proposed by Gurvitch is relevant to some of the five issues raised by the legal literature just discussed. (1) He emphatically recognises non-state law. (2) He notes state law pluralism to the extent that he recognises the possibility of multiple formal sources of a state’s law, although he does not seem to regard this as an important instance of pluralism in law. (3) The constituent elements of the more important instances of legal pluralism are for him the bodies of norms owing their authority and effectiveness to different material sources. (4) He does not locate legal pluralism within a particular field. When he discusses the relationship between international law and the laws of individual states, he shows that he views the entire world as his field of analysis (Gurvitch 1935: 149-151). (5) He does not analyse the nature of instances of legal pluralism. He indicates that norms derived from different material sources may coexist. He argues that conflicts between material sources may be resolved, and “la destruction de l’unité de la notion du droit” avoided (Gurvitch 1935: 149), by constructing a hierarchy of material sources.


None of the anthropological literature of this period expressly discusses the concept of legal pluralism. The writing which comes closest to using the concept in this period is Nader and Yngvesson (1973). It may not be helpful to devote much time to searching for implied references to the concept in anthropological writings (but cf. Rouland 1994: 46-56). However, there is no lack of statements of conceptions of law. At this point we need only note briefly the relevance of [page 30] anthropological work to the five issues raised by the legal literature just discussed.


1.          Recognition of non-state law. This is the central pillar of legal anthropology. The literature contains a great deal of discussion about what features of social life, in the societies studied by anthropologists, should be identified as law. Some were inclined to abandon the use of the concept of law (e.g. Abel 1973-74: 221-226; Roberts 1979: 9, Chap. 2; Comaroff and Roberts 1981: 12, 243). Those who adhered to it were divided for some years between the ‘rules’ and ‘process’ schools. This dispute is mirrored in legal theory in the controversy over whether state law is best seen as bodies of norms (a view generally taken by legal positivists such as Hart (1961) as well as by the natural law tradition exemplified by Finnis (1980)), or as centred on judicial activity (the view of inter alia ‘American realism’ examined e.g. in Twining (1973)). Most of the questions discussed in this paper could be put in terms either of rules or of dispute processes. By the 1980s the issue was no longer impeding progress, since adherents of both camps were prepared to continue their investigations with little concern over whether it could properly pass under the title ‘law’. (It has been said that Comaroff and Roberts (1981) “resolved” the debate: Starr and Collier 1987: 367.)


2.          State law pluralism. There was no serious discussion among anthropologists of the notion of state law pluralism. While legal anthropologists took account of the existence of state law, none of their studies focused exclusively on that area of activity.[4]


3.          Constituent elements of legal pluralism. In the absence of explicit analyses of the concept of legal pluralism, there are no statements of the views writers take as to its constituent elements. However, there are indications of the writers’ various views about the nature of the individual legal order, or the grouping of legal phenomena which might be distinguished from another such grouping, with which it could [page 31] sometimes interact. The texts have been sufficiently examined in other literature (e.g. Griffiths 1986, discussed below). Four writers need mention because their views are relevant to issues discussed later. Ehrlich (1936), a lawyer, but writing from a social scientific standpoint, saw rules of law as emerging from the internal practices of associations. These norms he called “living law”. Legal pluralism, had Ehrlich explicitly considered the concept, might have been regarded as involving contact between the living laws of different associations. Smith (1974) claims that the ordering of society can be understood in terms of corporations. Each corporation has its own legal order. Pospisil (1971, 1978) sees a society as composed of groups and subgroups at different levels, each group at an upper level comprising the total membership of a number of groups at the level below. Every functioning group or sub-group, at whatever level, is distinguishable by its possession of its own legal order. Moore (1978) sees laws as emerging from the semi-autonomous social field. Such a field, she says, is “defined and its boundaries identified... by a processual characteristic, the fact that it can generate rules and coerce or induce compliance to them” (1978: 57).


4.          The locus of legal pluralism. As under the previous head, it is necessary to infer. Ehrlich and Smith both have a general notion of ‘a society’ within which a number of legal orders exist. Pospisil has a more specific vision. Groups and sub-groups with legal orders are, according to him, grouped within and compose the entirety of a society. It does not follow that a society must have a general legal order, and indeed the Kapauku society which he studies does not (1978). Moore sees nation-states as themselves semi-autonomous social fields, with other such fields existing within them (1978: 56). Her conception gives little attention to the wider pattern of the arrangement of fields except for the all-important fact that they overlap, her concern being to argue that the semi-autonomous social field is a suitable subject of study in anthropology.



5.          The nature of instances of legal pluralism. Again by inference, it may be stated that Ehrlich and Smith see these as situations in which some individuals are members of two associations or corporations. Pospisil sees the groups and subgroups of a society as organised hierarchically in a series of levels. At each level membership of groups is exclusive, while each group at one level is included within a group at the next higher level. Thus the entire society is legally pluralist, since individuals are subject to legal systems at each level. The particular instances which he mentions are those where an individual is subject to [page 32] two systems which contradict each other. Moore conveys the clearest concept of legal pluralism, while not using this term. For her, where two semi-autonomous social fields overlap, individuals within the area of overlap are subject to two different bodies of law, which may or may not be mutually consistent.



Advance by Combat


This section examines primarily the argument of Griffiths (1986). There is a parallel discussion by Galanter (1981). The contrast between those two discussions reflects the distinction between the views of law as norm and process. Griffiths’ argument generally, subject to exceptions, treats legal systems as consisting largely of norms. Galanter’s discussion is generally, and subject to exceptions, an exploration of pluralism in dispute processing.


The aim of Griffiths (1986, first drafted and delivered seven years earlier: 39) is stated to be “to establish a descriptive conception of legal pluralism” (1986: 1). This means a conception which will enable the student to compare the degrees and types of legal pluralism in different societies. His definition of legal pluralism is:


that state of affairs, for any social field, in which behavior pursuant to more than one legal order occurs (1986: 2).[5]


Since he elaborates his conception through critical analysis of other writers, including those discussed in the previous section, it is relevant to notice whether they have different objectives from his.


Hooker’s objective is to write an account of that aspect of state legal orders which has arisen from the transnational reception of laws. He needs to identify his field of interest, but for this purpose it is unnecessary to define the concept of legal pluralism with any precision. In contrast, Gilissen and Vanderlinden seek to introduce the concept of legal pluralism in general as a field of study. They need for this purpose to define the subject in terms which set its outermost boundaries. While they are less explicit and precise than Griffiths about their [page 33] orientation, it seems clear that they, like him, set out to describe and analyse social facts, not to formulate legal doctrine. However, Griffiths’ objective, while in the same general category as theirs, is more narrowly focused. They seek to initiate the investigation of a certain aspect of social order. Griffiths seeks a tightly delineated concept which will firmly direct future investigation and analysis. In opening up a field, their approach is extensive, tending to include any phenomenon which may appear to be worthy of consideration. In formulating a programme, Griffiths’ approach is exclusionary.


How does Griffiths differ from them on the particular issues which have been enumerated?


1.          Recognition of non-state law. Griffiths recognises non-state law more emphatically than the previous writers, because a central part of his programme is to combat the ideology of legal centralism and its denial of the character of law to normative orders other than that of the state.[6] In parallel, Galanter’s article is devoted primarily to the exploration of non-state modes of dispute-processing, which he calls “indigenous law’.


Griffiths’ emphasis leads him, unlike the earlier writers, to consider explicitly the concept of ‘law’. The considerable discussion of this by jurists has in the main been limited to the examination of state law. Conceptions of law formulated on this basis are unlikely to be of help with respect to the present issue without modification, since they rely, implicitly or expressly, on distinctions between state and non-state normative orders to draw a boundary around what they conceive to be law. The removal of that boundary by the recognition of non-state law requires the formulation of some other distinction between legal and non-legal social phenomena. Griffiths observes that Ehrlich, Pospisil and Smith have seen law as the self-regulation of corporate groups. For reasons noted below he prefers Moore’s notion that law consists of the self-regulation of semi-autonomous social fields (Griffiths 1986: 37-38).


A definition of law as the self-regulation of either a corporate group or a semi-autonomous social field may need further development. Such definitions without qualification include features of social life such as positive morality, etiquette, fashion and custom. If law can be distinguished from these, the concept will be understood more precisely, and more in accord with some normal linguistic usage. Griffiths addresses this problem, largely in another writing (1984). [page 34] Galanter suggests that regulation below a certain level of differentiation of norms and sanctions may be excluded, the determination of the level for particular purposes being a matter of convenience (1981: 18-19, n26). The issue will be discussed further below, when it will be seen that the difficulty of establishing a satisfactory distinction has been advanced as an argument against the use of the concept of legal pluralism.



2.          State law pluralism. Griffiths excludes from his conception of legal pluralism situations analysed solely in terms of a state legal order. This follows from his attack on legal centralism and its insistence that the only law is state law. He argues that, in formulating a “descriptive conception” of legal pluralism, it is necessary to identify “legal pluralism in the strong sense”. This consists in each instance of a relationship (possibly but not necessarily of conflict) between two or more separate and distinct legal orders, one of which may be that of the state. “Legal pluralism in the weak sense” exists entirely within a state legal order. An example is the case where a state law regulates two categories of sales contracts differently, according to whether they are made by merchants (under commercial law) or ordinary persons (under general contract law). I have referred to the former as “deep legal pluralism” and the latter as “state law pluralism” respectively (Woodman 1988a.) It is worth noting that ‘legal pluralism in the weak sense’ should be capable of existing within a non-state legal order, although this has hitherto received little attention. If we are to avoid conceding a privileged status to state law, general discussion should perhaps always be about pluralism within legal orders, not just state law pluralism. As Griffiths shows, Gilissen and Vanderlinden discuss without distinction numbers of both classes of instances, while Hooker’s book consists largely of accounts of instances of state law pluralism.[7]

[page 35]  

Griffiths claims that a clear distinction can be drawn. Deep legal pluralism is a socially observable fact. State law pluralism is mere legal doctrine. Thus he notes that Hooker lists his sources of information as various types of documentary evidence of the formal rules of state law. He is, charges Griffiths, concerned not with “an empirical phenomenon”, but with “conceptual analysis, in doctrinal legal terms, of a feature of state-law in particular circumstances.” A “descriptive conception of legal pluralism” requires “an empirical intent”. One who adopts, in contrast, a “juridical intent” conforms to the ideology of state law, accepting its claims to exclusive jurisdiction over and regulation of social life as conclusive statements of reality (Griffiths 1986: 9-10).


This is a particular instance, in respect of the concept of legal pluralism, of the wider issue as to the relationship between social fact and legal doctrine. While not now questioning the distinction between the two forms of knowledge, it may be suggested that the line perhaps need not be drawn where Griffiths suggests. State law need not be conceptualised as doctrine alone. State power is frequently effective in certain significant, if limited, fields of the social world. It is frequently exercised in accordance with the norms of state law, if not always. Thus in some circumstances a statement of the formal law of the state can yield accurate description of certain aspects of social reality.


Such a conception of state law as a social fact, and so comparable with non-state law, has operational uses. For example, it is the basis of the literature on the distinction between folk law and lawyers’ customary law (e.g. Woodman 1985) and on the ‘invention’ of customary law in Africa (e.g. Chanock 1989; Starr and Collier 1987: 368).


The argument carries an implication which is important in the combat with legal centralism. The thesis of legal centralism, by denying that non-state law is law, asserts a clear distinction between state law and other normative orders. Griffiths’ argument gives some support to that distinction. But if the distinction [page 36] is seen to be blurred, the possibility of maintaining that only state law should be classed as ‘law’ is reduced. This is taken up again below.


Even if state law is classed, with non-state law, as an empirically verifiable phenomenon, a difficult issue remains. Griffiths’ rejection of state law pluralism is based not only on the ground that the rules of state law are ideology, but also on the ground that a conception of legal pluralism as existing within one legal order is incoherent. This is discussed under the following head.


3.          Constituent elements of legal pluralism. For Griffiths these are ‘legal orders’. He also approves of Vanderlinden’s notion of mécanismes juridiques. He prefers these terms to Hooker’s ‘systems’ because this has “overtones of completeness, orderliness, institutionalization, and static equilibrium” (1986: 12). But it would seem that a mécanisme could be a fraction of an ‘order’. The remainder of his discussion shows that Griffiths intends to refer not to fractions of orders but to entire legal orders, with the proviso that an order need not regulate or claim to regulate social life comprehensively. Thus he expresses some approval of Smith’s notion of the constituent element as the (entire) self-regulation of a corporation (1986: 18-23, discussing Smith 1974), and of Ehrlich’s notion of the (entire) self-generated regulation of an association (1986: 23-29, discussing Ehrlich 1936). He finally prefers Moore’s notion of the (entire) self-regulation of a semi-autonomous social field (1986: 29-37, discussing Moore 1978: Chap. 2). Legal pluralism thus occurs when there is some coincidence between legal orders, or the self-regulatory orders of semi-autonomous social fields. His statements that legal pluralism occurs whenever there is “law of various provenance”, or “more than one ‘source’ of law” are not developed, and seem to be references to the self-regulations of different social fields (1986: 38). The concepts of a legal order and a semi-autonomous social field are considered later.


According to this view the law of a state is a single legal order. It follows that a state law can be only one element in an instance of legal pluralism. This appears to be the principal reason for Griffiths’ insistence that the concept of legal pluralism entails the recognition of non-state law. It is also the main reason for his denial of the possibility of legal pluralism within a state law. This would seem open to question.


In considering critically Griffiths’ view that legal pluralism cannot exist within a state law, it is worth noting that the notion he is attacking has an intellectual appeal. As already seen, Gilissen, Vanderlinden and Hooker all apply the term legal pluralism to include state law pluralism. It would seem prima facie [page 37] appropriate to use it for a situation in which, for example, state law contains one set of rules on marriage and divorce derived from Islamic law, perhaps several other sets derived from different ethnic customary laws, and a set derived from English law; or in which sales transactions between merchants are governed by different rules (derived from the custom of the merchant community) from those between other persons. We might note also that the quality of some studies of pluralism within local legal orders suggests that this is an identifiable field for worthwhile study. One example is the impressive contribution to the understanding of African urban life by Johan Pauwels’ work in Kinshasa (1968).[8]


Griffiths’ argument against this use of the concept is that it depends upon categorisations which are doctrinal, not empirical. He criticises Vanderlinden’s references to different legal mechanisms for ‘the same situation’ on the ground that “the concepts of ‘difference’ and ‘sameness’ are not empirical....”, but rather stipulated by rules of law. “There is nothing in the nature of the world or of social life which requires anyone to agree with Vanderlinden that [for example] the acts of an ordinary person and of a businessman... are really the same” (1986: 13). Griffiths concludes that situations which are said to be instances of legal pluralism within a single legal order are no more than instances of lack of uniformity of law, or legal diversity (1986: 14, 11). These merely demonstrate “the obvious fact that every legal system, no matter how unitary its law, provides different rules for [what it categorises as] different situations” (1986: 11, criticising Gilissen). Griffiths thus doubts the possibility of sustaining a distinction between pluralism within a legal order and diversity within a legal order.


As already seen, Vanderlinden seeks explicitly to distinguish between instances of legal pluralism, as he has defined it, and instances of plurality of law. The latter are situations in which conflict between legal norms is avoided by an overriding rule stipulating that the respective norms are to apply to different situations. Vanderlinden does not seem to have been suggesting that the distinction between identical situations and different situations was a matter of empirical fact. While the point is not spelt out, he seems to imply that situations are to be categorised as ‘identical’ for the present purpose if one or both of the legal mechanisms in issue so categorise them.

[page 38]  

The development of theory requires that where possible the perceptions of both sides to a combat be used. On this particular argument a tentative suggestion for compromise may be made. We may find it instructive at this stage to follow the more general usage, and continue to classify as instances of legal pluralism the cases which I have called state law pluralism. Moreover, recalling the earlier suggestion that rules contained in an operative state law are social facts as well as doctrine, the classification by legal norms of situations as ‘the same’ may be regarded as an empirically observable phenomenon. Moreover, we may need to go even further than Vanderlinden’s position. Although situations legally classified as the same may be distinguished from those legally classified as different, it may be inconvenient for the purpose of research to exclude the latter category from the sphere of legal pluralism. Vanderlinden’s argument excludes, for example, the widespread state of affairs in medieval Europe in which both church and state laws affected large numbers of individuals, although they rarely both applied to situations which they classified as ‘the same’ (see also Griffiths 1986: 13-14). Thus, if state law pluralism be seen as a type of legal pluralism, as I suggest, it seems preferable to include all instances of diversity in state law in the category. Thus we might classify every state legal order (and every other legal order) as internally pluralist, although the degrees and types of pluralism may vary. These variations are likely to be of interest in the study of comparative law.


However, we should note also Griffiths’ argument that instances of state law pluralism may also be seen as instances of legal diversity. We must note that, according to the arguments considered so far, pluralism within a legal order is distinguishable from deep legal pluralism. The suggested compromise view is that, while recognising state law pluralism as a form of legal pluralism, we may need to distinguish between this form and deep legal pluralism. It follows that instances from the two categories may for some purposes not be comparable. This is a provisional proposal at this point in the argument. The issue is discussed further below, when this interim position is modified.


4.          The locus of legal pluralism according to Griffiths is a social field. He approves of Vanderlinden’s statement that it occurs au sein d’une société déterminée. However, it seems that Griffiths’ formulation is not intended to incorporate the degree of definiteness of Vanderlinden’s formulation. His approval of Vanderlinden’s statement is given for its acknowledgement that legal pluralism occurs in social terrain. When he comes to contrast different views of the locus of legal pluralism, Griffiths argues that legal pluralism is to be seen as an attribute of a “specified social group”, and not “of ‘law’ or of a ‘legal system’; nor does it have a necessary connection to territorial or other non-social entities” (1986: 12; also 38, referring to a “social field”). He does not [page 39] endorse Vanderlinden’s specification of the field as a société déterminée. Subsequently he prefers Moore’s view of interacting semi-autonomous social fields to the models of Pospisil and Smith which adopt a ‘whole-society’ preconception and seek to locate legal orders and associations within a given society.[9]


When Griffiths refers to a ‘social field’, he sometimes uses the term in Moore’s sense of a semi-autonomous social field identifiable by the fact of its self-regulation. This indicates a field within which instances of legal pluralism are likely to occur, but it does not indicate the extent of any particular instance. In other places he uses the term to refer to the area of overlap between two or more legal orders. This latter is a field of which the limits are known only when an investigation has been completed. The conclusion to be drawn from the view that legal pluralism occurs in a social field seems to be nothing more than that it is a social fact. Legal pluralism is not a characteristic of a specific type of society or of a given social field.


In this respect Griffiths’ conception of legal pluralism goes beyond the conceptions of pluralism in the social sciences generally. While the nature of pluralism and the most useful ways to study it have been the subjects of dispute, all debate has been on the premise that pluralism is a characteristic of a society. The prevalent conceptualisation supposes that there are ‘plural’ societies and (at least in theory) ‘monistic’ societies, and that to study an instance of pluralism is to study a particular feature of a single, given society (e.g. van den Berghe 1973; for indications of the movement of anthropological theory towards the recognition of both types of legal pluralism, see Nader and Yngvesson 1973: 887-892, 903-906).


5.          The nature of instances of legal pluralism. Griffiths does not accept Vanderlinden’s requirement that the elements of an instance of legal pluralism apply to an ‘identical situation’. We have seen his argument for rejecting that.

[page 40]  

Griffiths further questions Pospisil’s account of the pattern of relationships between the different legal orders in instances of legal pluralism (1971, 1978: see Griffiths 1986: 15-18). He objects that this is an idealised picture which cannot accommodate “the laws of groups within a society which are not part of an overall hierarchical arrangement and cannot be assigned to a particular ‘legal level’ - groups such as clubs, guilds, churches, factories and gangs” (Griffiths 1986: 17).


The effect of the last two points is that Griffiths’ view of legal pluralism is that it is simply a feature of social life at large. There is no particular field of social life within which instances of legal pluralism exclusively occur. It has no locus in quo, but rather is present in any instance where two or more legal orders ‘overlap’ (1986: 38). This is a broadening of the concept, which opens the possibility of using it to refer to a wider range of phenomena, and which may help to reveal the common features of many different instances. In this respect this view advances the theory. On the other hand, it poses a risk. The identification of instances of legal pluralism now depends crucially on the identification of separate legal orders or of semi-autonomous social fields. If there is any uncertainty in these concepts, Griffiths’ vision of legal pluralism loses its clarity.



Recent Critique and Refinement


It might have been thought that the stage of development just described left ample room for further debate over the elucidation of the concept of legal pluralism, and so for further progress. It might have been thought also that further debate could have been provoked by the controversial value of the concept: anyone who has discussed this issue informally in recent years must be aware that the views just considered are not universally accepted in legal and academic circles. Yet in the literature there has been relatively little criticism of the foundations or the details of the propositions associated with the concept (Benda-Beckmann 1988: 897; an important exception to this is Tamanaha 1993, whose arguments are considered below).[10] It may be that lawyers have preferred to ignore the subject since it challenges their accepted ideologies, and does not appear to assist them in performing their usual work. Legal anthropologists are perhaps generally agreed in approval of the concept, and have seen no need to elaborate it. The lack of further theoretical progress is evidenced in the survey of the topic by Merry for the Law & Society Review. [page 41] This takes Griffiths’ formulation of the concept (first propounded in 1979) as generally accepted (1988: 870), and states that Moore’s conception of legal orders as semi-autonomous social fields (first published in 1973) is still “the most enduring, generalizable, and widely-used” (1988: 878). This is despite the considerable research on instances of legal pluralism completed in the 1980s, noted by Merry.


An exception to the lack of recent theoretical debate is the continuing work of Chiba (Chiba 1989, Kitamura n.d.). However, his work aims less to develop or clarify a definition of legal pluralism than to develop or clarify the features of certain instances of legal pluralism. Because of this it is not referred to frequently in the following pages. The value of Chiba (1989) as an analysis of manifestations of legal pluralism has been noted elsewhere (Woodman 1992, n.d.).


This section considers five issues which have been discussed since the early 1980s, and which appear to have relatively great significance for the development of the theory of legal pluralism.



1. The legal nature of non-state law: (a) the question of the distinctiveness of state law


It has been seen that a constant refrain in the theory of legal pluralism has been the recognition of non-state law. Admittedly some of the leading writers have equivocated over the use of the term ‘law’ to refer to non-state normative orders (Moore 1978: 62, noted Griffiths 1986: 37; Merry 1988: 870, 874, noted Tamanaha 1993: 193). Nevertheless the version of legal pluralism which I have called deep legal pluralism, and which receives the greatest attention today, claims that there are types of law other than state law. Writers on legal pluralism continue to assert this with numerous examples (e.g. Vanderlinden 1993; a strong statement of the argument with reference to Japanese culture appears in Chiba 1989). Tamanaha (1993) argues, to the contrary, that the concept of law should be reserved for state law. His case is based upon two arguments. One will be considered here, and the other in the next section.


Tamanaha claims that there is an empirical distinction between state law and other forms of normative social ordering. The essence of state law is its identification and enforcement by institutions of the state. Non-state normative orderings, while they may also include institutional apparatus for the identification and enforcement of norms, consist essentially of “concrete patterns of social ordering” (a phrase originating in Galanter 1981: 17). Social norms exist as such by virtue of being part of the social life of the group rather than [page 42] through institutional recognition. Non-state norms may pass across the divide and become part of state law by virtue of their recognition by designated state officials. State law norms and non-state norms are thus ontologically distinct. Once this distinction is established, argues Tamanaha, there are strong arguments for reserving the term ‘law’ for state law norms.


This argument is important for the present purpose. The tendency to suppose a gulf between state and non-state normative orders is strong even among many of those who approve of the concept of legal pluralism.[11] This tendency is pursued notwithstanding that it is also common to use the word ‘law’ for non-state normative orders, with a variety of qualifying terms, such as customary law, folk law, religious law and international law (Merry 1988: 875-877).[12] Moreover it may seem that there is a difference in the forms of argument commonly used to establish the content of state law and non-state law respectively. Much discussion of state law emanates from professional lawyers reasoning in terms of legal doctrine, and is concerned with issues of validity. [page 43] Most discussion of non-state law found in the literature comes from social scientists and is about social conduct.


Nevertheless Tamanaha’s distinction seems open to question. For the purposes of social science a concept of state law refers to social facts. State law is generally seen from this viewpoint as a combination of two social facts. First is the fact of the effectiveness, to a certain extent and subject to the limits of the reach of state institutions, of the body of state law norms. Second is the social fact of the daily use of certain types of doctrinal argument in state law fora. The concept of non-state law also is generally a concept of social fact. And again it is seen as a combination of social control of a certain degree of effectiveness (‘concrete patterns of social ordering’, although never effective in absolutely all instances), and the use of certain forms of argument whereby appropriate, or ‘sound’ answers are found to particular issues. The distinction between state law as doctrine and non-state law as social ordering is no more than a distinction of relative emphasis in the sources of information most readily available, not an ontological divide.[13]


If there were such a difference as Tamanaha claims, then, whatever our decision as to the use of the word ‘law’, it would be difficult to maintain that every instance of legal pluralism arose from the coincidence of two or more objects falling into the same category, such as ‘legal orders’ or ‘laws’. A useful contribution in this respect is that of van den Bergh (1986), which questions “simplistic binary concepts like law-custom, written-unwritten law, state-folk law”, and shows the difficulty of specifying what exactly is meant by state law and non-state law. Tamanaha’s argument shows that more work is needed to clarify the relevant concepts of state law and non-state law. Some further investigation of the issue is attempted below.



2. The legal nature of non-state law: (b) the question of the distinction between non-state law and social order generally


Just as the boundary between state law and non-state law has long been [page 44] problematic, so has been the boundary between non-state law and similar, non-legal social phenomena. Social scientists note that in many societies there are wide varieties of social norms and processes. Social norms range from prohibitions regularly enforced by severe sanctions imposed by formally authorised (although non-state) officials, to norms of etiquette and good manners, vaguely formulated and enforced by relatively mild expressions of disapproval. It has seemed to many analysts that it should be possible to formulate a criterion to distinguish the legal from the non-legal among these non-state phenomena.


This has proved difficult. Attempts to adapt definitions devised for state law have generally proved unsuccessful, but neither have entirely new modes of distinction been helpful. (De Jong (1994) attributes the problem largely to Malinowski and his definition of rules of law simply as “all the rules conceived and acted upon as binding obligations” (Malinowski 1926: 55, also 58). This problem in Malinowski’s definition has been noticed by many commentators: see e.g. Moore 1978: 220). Consequently some have doubted whether outside the modern state a discrete field of law could be said to exist. Others continue to express worry about the difficulty (e.g. de Sousa Santos 1985: 279; Merry 1988: 871, 878-79; Greenhouse and Strijbosch 1993: 5). A not uncommon reaction has been that already noted: abandonment of attempts at definition, and the selection for research of subject-matter which could be delimited without reference to ‘law’, such as disputing processes (Abel 1973-74: 221-226; Roberts 1979: 9, Chap. 2; Comaroff and Roberts 1981: 12, 243).


Tamanaha argues that this difficulty, the “Malinowski problem” (1993: 206, 207), provides another ground for holding that the concept of legal pluralism is neither definable nor useful. The concept of legal pluralism, he argues, requires that legal orders or systems should be clearly identified. For this it is necessary that law be precisely defined. Legal pluralism rejects a definition of law which is limited to state law. But once non-state law is admitted, it is impossible to define law precisely, and any attempt “slippery slides to the conclusion that all forms of social control are law”. It is, then, impossible to define ‘law’ in a way which allows the concept of legal pluralism to be sustained (Tamanaha 1993: 193). This argument is related to that of Tamanaha concerning the distinction between state law and non-state law discussed in the previous section. (De Jong’s conclusion was similar to Tamanaha’s, although his target was the concept of customary law. He argued that it was necessary to abandon the concept of customary law and revert to the formal criteria of state law for the identification of law: de Jong 1994.)


It does not seem possible to meet this criticism by defining law as a distinct form of social control which is clearly distinguishable from the others. Attempts to do [page 45] this have failed in consequence of the variety of known social orders. A more defensible answer is that, if there is no empirically discoverable dividing line running across the field of social control, we must simply accept that all social control is part of the subject-matter of legal pluralism. This conclusion is not convenient, but it may be necessary. To invent a dividing line which did not accord with a factual distinction would be irrational and unscientific. For the same reason Tamanaha’s argument, that it is essential to distinguish state laws from other social norms because there is no other means of differentiating law, is also not a valid social scientific argument. The conclusion must be that law covers a continuum which runs from the clearest form of state law through to the vaguest forms of informal social control.


An acceptance of this truth need not hinder social scientific research. Legal anthropologists will naturally seek to define their own fields of investigation for particular projects within the total field of law. The contribution of legal anthropological work to our understanding of human society shows that this is practicable.[14]



3. Pluralism within a legal system


This issue turns primarily on new arguments recently developed by Vanderlinden (1989, 1993). His main contention is discussed below. It will be seen that he eventually rejects a notion of legal pluralism which classifies situations by reference to particular legal systems. However, his references to ‘systems’ in these papers help to clarify this subject.


It has been seen that Griffiths (1986) in criticism of Vanderlinden’s earlier essay [page 46] (1971) argues that the constituent elements of any instance of legal pluralism are entire legal orders. (Griffiths prefers to speak of ‘orders’ rather than ‘systems’.) Pluralism within one legal order, argues Griffiths, can exist only as doctrine, not as social fact. Earlier in this paper I defended Vanderlinden (1971) against that criticism. Important research has been done within conceptual frameworks according to which legal pluralism could exist within state laws (e.g. Arthurs 1985). Vanderlinden’s definition has been explicitly repeated and employed by others (e.g. Belley 1993). But now Vanderlinden states: “The more I try to think about it, the more I am convinced that the idea of a pluralistic legal system is impossible” (Vanderlinden 1989: 154). In consequence he now excludes from the category of legal pluralism a formidable list of instances which he previously included, and many of which, it might be thought, legal theorists would have expected to be included (Vanderlinden 1993).


Saying that by ‘legal system’ he means a coordinated set of practices tending to regulate the social network (the latter term being preferred to ‘social field’), Vanderlinden argues that “to speak of a pluralistic legal system is either self-contradictory or redundant” (Vanderlinden 1989: 152). A pluralistic legal system would have to exist either in a totally autonomous, isolated social network or in a semi-autonomous social network. In a totally autonomous network (which is in any case unlikely to exist), the members would be “totally free from the influences of other legal systems”, and so not in situations of legal pluralism. A semi-autonomous network necessarily has to live with competing orders. If any of these legal orders competes with its own legal order, then its members are in a situation of legal pluralism. “Hence to speak of a legally pluralistic system as being a system in which many competing legal orders exist is redundant” (Vanderlinden 1989: 152).


With respect, this argument seems to fail to establish the conclusion. Vanderlinden impliedly defines legal pluralism as the situation where an individual is a member of two or more networks, each of which is regulated by its own, single legal system. That very definition excludes the possibility of legal pluralism within one system.


We may here also note the development in the past six years by scholars in the Nordic countries of the concept of Polycentricity of Law (Bentzon 1992). The term seems to refer to a category of instances of legal pluralism. They are described as the use of sources of law in different sectors of the state administration. The principal hypothesis, which is currently the subject of research, is that different authorities frequently use different sources of law, and that even when they use the same sources they observe different orders of priority between them. Bentzon writes:

[page 47]  

The concept of Polycentricity thus supplements Sally Falk Moore’s picture of the semi-autonomous social fields inside the apparatus of the State (Bentzon 1992).


This is a concept of state law pluralism alone. The references to the differing sources of law suggest the possibility of conflict between different state sectors as to the degrees of authority to be given to different sources. It is also possible that the use of different sources results in some sectors deciding issues by applying norms which contradict those applied by other sectors. The possibility of incompatible conflicts within state law is important. It is considered below, where the issue is approached by a different method. It is not yet clear whether research on polycentricity has revealed incompatible contradictions between the observance of sources of law, or between norms claiming to regulate social behaviour. It is possible that conflicts at the level of particular sectors are resolved by the application of overriding norms at higher levels. We may be faced with some instances of plurality of laws, together with some instances of differences between the norms applied in practice and those which are officially declared to be applicable. There is room for debate as to whether any of these matters are to be regarded as legal pluralism.


On this issue we are still left with the incomplete conclusion stated in the previous section: the notion of state law pluralism, or any other pluralism within one legal order, seems to have meaning and to be potentially useful; however, it appears to be distinct from deep legal pluralism; but the issue needs further investigation. I investigate it further, if briefly, in the concluding section.



4. The locus of instances of legal pluralism


We have seen that according to Vanderlinden (1971) legal pluralism occurs au sein d’une société déterminée. Griffiths (1986) sees it as occurring “in a social field”, which we saw means little more than that it is a social fact. Vanderlinden now argues that it is more satisfactory to focus on the individual as the sujet de droit.


Instead of looking at the legal pyramid from the top, from the centres of decision, from the standpoint of power, one is brought to contemplate it at the level of ordinary men in their daily activities (Vanderlinden 1989: 153).


Accordingly his definition of legal pluralism today is:


la situation, pour un individu, dans laquelle des mécanismes [page 48] juridiques relevant d’ordonnancements différents sont susceptibles de s’appliquer à cette situation (Vanderlinden 1993: 583).[15]


It has been suggested already that little is gained by specifying that situations of legal pluralism occur within a society or a social field. Is it helpful to specify that they are situations pour un individu? The perspective of the individual may certainly be informative, and in some circumstances practically useful (such as when seeking to assist a particular individual). But it would also seem informative and useful to adopt the top-down perspective. If we wish to understand the meaning of legal pluralism for entire categories of individuals, or for a legislator (who surely needs all the help possible), for the purpose of comparing situations of legal pluralism in different social orders, or for the development of the understanding of human society, a project which looks at social groups as a whole has merits. Vanderlinden’s current insistence on defining and studying instances of legal pluralism in terms of the individual subject is most useful as an addition to other perspectives, but should not be adopted as a substitute for them. (This conclusion for the study of law is the same as that of Pospisil (1978: 97-99) for the study of social structure.)



5. Criticisms as to the practical usefulness of the concept


A reading of the debates reveals strong ideological currents. The ‘debunking’ exercise of Griffiths was aimed at demolishing the ideology of legal centralism, and strong criticisms of that ideology continue. But ideological issues are not part of the subject-matter of this paper. The elucidation of concepts needs to be distinguished from the combat of ideologies. No doubt the analysis of concepts requires an awareness of the danger of being misled by ideologically inspired prejudices. No doubt also concepts are employed in the formation of ideologies. But a concept cannot itself be an ideology. A statement, for example, that “legal pluralism... involves an ideological commitment” (Sack 1986: 1), is, strictly speaking, without sense. At best it is a confusing way of saying that an assertion of a need to be strongly aware of the (alleged) ubiquity of situations which some [page 49] call legal pluralism is an ideological statement.[16] This way of talking can impede clarity of thought, because it assumes we already have a clear view of the referent of the expression ‘legal pluralism’, and so conceals the need to analyse the concept and relate it to empirical data. The consequence, in my view, is well illustrated by the failure of the work just cited to clarify the concept (Woodman 1988b).


Conceptual analysis can be of practical assistance in the enlargement of our knowledge. As Merry argues, the definition and use of the concept of legal pluralism may stimulate some types of research by directing attention to the relations between legal systems (Merry 1988). A focus on these, she argues, may reduce an excessive concentration on the law/society dichotomy, and on studies of the effects of one on the other, and it may result in study of the “more complex and interactive relationship between official and unofficial forms of ordering” (Merry 1988: 873). It may encourage the search for external factors to explain changes in legal orders, and offset the monist view of each legal system as evolving alone, through internal necessity (Merry 1988: 879-886; and see Merry 1992, as another illustration, additional to those she lists).[17]


She argues in effect that these merits have corresponding disadvantages. By focusing attention on relations between legal systems, the concept of legal pluralism draws attention away from the analysis of internally generated development within particular social fields, and may lead to a neglect of local [page 50] variations within a system (Merry 1988: 891).


It is of course true that if, when there is a scarcity of intellectual resources, attention is drawn to one meritorious subject of study, another equally valuable may be neglected. However, it is doubtful whether the concept of legal pluralism necessarily draws attention away from the internal development of social fields. On the contrary, to see the social universe as containing numbers of semi-autonomous social fields, each generating a social order, could lead to an increased attention to the internal workings of fields. Merry subtly departs from Griffiths’ view when she refers to “analyses of systems [sic] to the neglect of the variation in particular local places”. Griffiths prefers the term ‘legal order’ to ‘legal system’ to avoid assumptions of systematic bodies of law. This terminological choice draws attention to the possibility of internal variation between localities within legal orders.


In another respect, however, Griffiths’ approach may contribute to the deficiency Merry sees. As we have seen, in rejecting the concept of state law pluralism Griffiths rejects the notion of pluralism within a legal order as suitable for consideration as a form of legal pluralism. If we accept that the concept of pluralism within state law, as within any other legal order, can be part of a coherent sociological analysis of law, as Gilissen, Vanderlinden and Hooker do, we shall have a useful tool in the study of single legal orders.



Conclusion: What is Legal Pluralism Really?[18]


On the assumption that the arguments advanced in the foregoing sections have some validity, this section considers the concept of legal pluralism which emerges, and the further issues which it suggests for debate.


The discussion has noted two major, related questions about the concept. These are whether there is a fundamental difference between state law and non-state law, and whether there is a fundamental difference between non-state law and other elements of social ordering. The issues might equally well be expressed as concerning the meanings of ‘law’ and ‘legal pluralism’, although that does not show that they are merely verbal.


It is not intended to add to the foregoing discussion of the issue of the distinction between non-state law and other social orderings. The onus is surely on those who claim that there is a distinction to elaborate and explain it.

[page 51]  

On the distinction between state and non-state law progress may perhaps be made through the study of the social facts of state law. We have seen that the debate about legal pluralism began as an assault on legal monolithism and continued as a combat with legal centralism, the ideology of state law. A major argument has been that statements within this ideology asserting the exclusiveness of state power can be shown to be untrue. The attack on this ideology has been conducted largely by demonstrations of the existence and characteristics of non-state laws, while the contestants have tacitly accepted the integrity of state laws and the claims made for state law, other than that to exclusivity. Thus much of the research discussed in Merry’s survey has taken state law as a constant; and, as she notes, the ways in which other normative orders shape state law are “particularly unstudied” (Merry 1988: 884). Yet we have available enormous quantities of information about state law in the literature used by lawyers. Records of ideological claims must not be confused with records of informative statements about social facts. But, as argued above, statements of state law can, if used with care, be used to glean factual information.


There is strong evidence that state laws are (a) not internally self-consistent, logical systems, and (b) not clearly bounded and distinct from other social normative orders. It is not possible to refer comprehensively to the literature here. Most of the arguments may be found in the work of Dworkin (1978, 1986), who developed them in relation to the common law systems of state law.


1. The boundaries of state law. It is convenient to take issue (b) first. A large literature in the field of jurisprudence in this century has studied the processes of law-finding, especially the judicial process in the common law. It has frequently been claimed that a wide degree of discretion is exercised in many or all instances of law-application. However, it has also been argued that the law-finder does not exercise unlimited discretion, but rather incorporates social and political norms into the law. Consequently norms which do not formally qualify as norms of state law, having not been enacted by the legislature nor laid down in precedent, are in reality part of state law. On this analysis the boundary between state law and what some have called positive morality disappears. The consequence is a view advanced 50 years ago by Gurvitch, who wrote that is was necessary in the sociology of law to


... break down the positivist separation of law from other forms of social control and... picture this broad concept of law as a continuum rather than a discrete series of ideal types, models or paradigms (Gurvitch 1947: 30).


2. The self-consistency of state law. The analysis of the law-finding process just [page 52] mentioned renders issue (a) even more acute, for it negates the possibility of a cohering, unifying factor within a state law, which could guarantee internal self-consistency. It is a central tenet of orthodox legal thought that the law of every state is derived ultimately from a single source, or from a few which are organised in a recognised hierarchy. Thus in cases of apparent conflict between norms, the proper application of doctrine will always show that there is really no conflict because one of the competing norms is applicable to a particular fact-situation and the other not. This view also may be an ideological claim, not a report of reality. If the law-finder routinely draws standards from the general social and political norms of the community, it becomes impossible to claim that every legal norm is derived from a legislative code, or some ultimate, basic norm of the system.


The same conclusion may emerge for another reason than the use of outside standards. The doctrinally acknowledged sources of law quite possibly do not have the mutual self-consistency claimed. In the common law system, for example, the principal modern sources of authority are statute and precedent. It seems that these are independent sources. It is doubtful whether there is in reality a body of norms which determines in every circumstance which takes priority over the other. It seems that the same conclusion is being demonstrated by the study of polycentricity of law. Furthermore, there are numerous instances of specific norms, whether derived from the same or from different sources, which conflict with each other, again with doubt as to whether the legal system of the state places them in a clear hierarchy of precedence. All this can be seen once the observer abandons the professional, doctrinal habit of mind which requires a belief in the logical coherence of the system. Thus, as Dworkin has put it, the norms of a state’s law do not have a common pedigree, providing tests which


...can be used to distinguish valid legal rules from spurious legal rules... and also from other social rules (generally lumped together as ‘moral rules’) that the community follows but does not enforce through public power (Dworkin 1978: 17).


The usual conceptions of deep legal pluralism assume that state law is a well-defined, consistent whole which can be one, clear part of a plural situation. If state law does not have this character, conceptions of legal pluralism of the current type will need to fasten upon some other well-defined aspects of the legal world. The argument raises a pressing question as to the constituent elements of situations of legal pluralism.


3. Constituent elements of legal pluralism. We have seen that the writers have [page 53] defined legal pluralism in terms of the coincidence of, respectively, droits (Gilissen), mécanismes juridiques (Vanderlinden 1971), ‘systems’ (Hooker), and ‘legal orders’ (Griffiths). We have seen that the most elaborated scheme, that of Griffiths, associates legal orders with semi-autonomous social fields, a concept preferred to associations, corporations or groups and sub-groups. We have seen that Vanderlinden (1993) refers to mécanismes juridiques relevant d’ordonnancements différents. The important concept here seems to be that of the ordonnancement, or legal order.


Griffiths in preferring ‘legal orders’ to ‘legal systems’ has already acknowledged the argument that state law is not systematic. Non-state legal orders are also, according to the anthropological studies, not perfectly logically coherent. These observations suggest that legal orders cannot be identified by reference to their internal coherence. Scholars who because of this have declined to refer to legal doctrine to define the legal order have turned to alternative defining facts. The concept of the semi-autonomous social field was considered helpful in maintaining the notion of ‘pluralism’ as a coincidence of two or more entities.


Moore’s concept has helped many of us to grasp the notion of a plurality of bodies of law overlapping with each other in such ways that individuals may find themselves members of the areas of operation of more than one. But close examination suggests that the concept may be too vague to be of assistance beyond that. The semi-autonomous social field is identified by its possession of “rule-making capacities, and the means to induce or coerce compliance” (Moore 1978: 55-56). There are several difficulties. It is common for rule-making capacity to be held by a smaller number than those who are induced or coerced to comply with the rules. It is not unusual for some or all of the persons with the rule-making capacity not to be among those who are induced or coerced to comply. The primary definition of the field would seem to be in terms of those who are induced or coerced to comply, but it is doubtful whether we can simply leave out the reference to those with the rule-making capacity. Furthermore, a group of persons may be induced or coerced to comply with rules emanating from two quite different sources. This might well be regarded as a clear instance of legal pluralism.


There are further reasons for thinking that Moore’s field may have ragged, vague or undiscoverable edges. Of a number of rules emanating from one group with rule-making capacity, some may be complied with by a certain group, others by a smaller group, others by a group which coincides partly with each of the first two groups but includes persons who are in neither. We can combine with those variations the frequent situation in which the rule-making capacity is exercised by a group with a constantly fluctuating membership. It seems unlikely that the concept of the semi-autonomous social field can restore to legal [page 54] pluralism that feature of defined constituent elements which was lost when the bounded, self-consistent ‘legal system’ dissolved into the indiscriminate ocean of social norms.


The concept of polycentricity of laws may also need reconsideration. It may be helpful in so far as it questions the claims of state law to be internally unitary and consistent. But in so far as it ignores non-state law and non-state law-application fora, its potential is limited. Still more seriously, the title ‘polycentricity’ suggests that there has been substituted for one, monolithic structure a number of centres. There is little reason to assume that the centres thus identified will themselves operate as minor monoliths. Each is liable to be self-contradictory in operation, and to constitute a field of varying size as the law-making and law-application functions are exercised by and imposed upon different, fluctuating populations.


It must be admitted that the argument has led us to a new problem rather than to a solution of a previous problem. The notion of legal pluralism seems to reflect a part of the experience of observers of the legal universe. But analysis seems to show that all ideas of a tightly structured model of legal pluralism are indefensible. This, it may be suggested, reduces the significance of the criticism of the notion of state law pluralism. If systems of law do not exist, it can hardly be contended that every instance of legal pluralism must involve more than one legal system. The concepts of legal pluralism and plurality of laws have merged.


The implications may be tolerable. Plurality of laws exists everywhere, because everywhere there are “different rules for different situations” (Griffiths), or “cultural heterogeneity and normative dissensus” (Geertz 1983: 225). Consequently legal pluralism exists everywhere. The interesting questions for our attention may involve comparison of legal fields in terms of the degrees of legal pluralism present. ‘Legal fields’ for specific research purposes may be designated in whatever ways are convenient, although research which is sensitive to reality will not forget the connections of the research subject to the rest of this vast field. A straightforward distinction between unitary and plural legal situations will not be possible because unitary situations do not exist. But this is only to suggest that legal pluralism is a non-taxonomic conception, a continuous variable, just as, according to Griffiths’ well-founded and helpful observation, ‘law’ is.


[page 55]  




ABEL, Richard L.

1973-74  ‘A comparative theory of dispute institutions in society.’ Law & Society Review 8: 217‑342.


1985     Without the Law: Administrative Justice and Legal Pluralism in Nineteenth‑Century England. Toronto and Buffalo: University of Toronto Press.

BELLEY, Jean-Guy

1993     ‘Pluralisme juridique.’ Pp. 446-449 in André-Jean Arnaud (dir.), Dictionnaire encyclopédique de théorie et de sociologie du droit (2nd ed.). Paris: Librairie Générale de Droit et de Jurisprudence.


1988     ‘Comment on Merry.’ Law & Society Review 22: 897-901.

BENTZON, Agnete Weis

1992     ‘A note on the polycentricity of law. A Danish version of legal pluralism.’ Commission on Folk Law and Legal Pluralism Newsletter XXI: 30.

BERGH, G.C.J.J. van den

1986     ‘The concept of folk law in historical context: a brief outline.’ Pp. 67-89 in Keebet von Benda-Beckmann and Fons Strijbosch (eds.), Anthropology of Law in the Netherlands: Essays in Legal Pluralism. Dordrecht: Foris.

BERGHE, Pierre L. van den

1973     ‘Pluralism.’ Pp. 959-977 in John J. Honigmann (ed.), Handbook of Social and Cultural Anthropology. Chicago: Rand McNally.


1972     Sociologie juridique. Paris: A. Collin.


1989     ‘Neither customary nor legal: African customary law in an era of family law reform.’ International Journal of Law and the Family 3: 72-88.

CHIBA, Masaji

1989     Legal Pluralism: Toward a General Theory through Japanese Legal Culture. Tokyo: Tokai University Press.


1981    Rules and Processes: The Cultural Logic of Dispute in an African Context. Chicago and London: University of Chicago Press.

DWORKIN, Ronald M.

1978     Taking Rights Seriously. Cambridge, Massachusetts: Harvard University Press.

[page 56]  

1986     Law’s Empire. Cambridge, Massachusetts: Harvard University Press.


1936     Fundamental Principles of the Sociology of Law (trans. W.L. Moll), Harvard Studies in Jurisprudence. Cambridge, Massachusetts: Harvard University Press.


1980     Natural Law and Natural Rights. Oxford: Clarendon.

GEERTZ, Clifford

1983     Local Knowledge: Further Essays in Interpretive Anthropology. New York: Basic Books.


1971     ‘Introduction à l’étude comparée du pluralisme juridique.’ Pp. 7-17 in John Gilissen (ed.), Le pluralisme juridique. Bruxelles: Editions de l’Université de Bruxelles.


1993     ‘Legal pluralism in industrialized societies.’ Journal of Legal Pluralism 33: 1-9.


1984     ‘The division of labor in social control.’ Pp. 37-70 in Donald Black (ed.), Toward a General Theory of Social Control. New York: Academic Press.

1986     ‘What is Legal Pluralism?’ Journal of Legal Pluralism 24: 1-55.


1935     L’Expérience juridique et la philosophie pluraliste du droit. Paris: Éditions A. Pedone, Librairie de la Cour d'Appel et de l'Ordre des Avocats.

1947     Sociology of Law. London: Routledge & Kegan Paul. (First published New York, 1942).


1961     The Concept of Law. Oxford: Clarendon.

HENRY, Stuart

1983     Private Justice: Towards Integrated Theorising in the Sociology of Law. London: Routledge & Kegan Paul.

HOOKER, M. Barry

1975     Legal Pluralism: An Introduction to Colonial and Neo-colonial Laws. Oxford: Clarendon Press.

HUND, John

1982     ‘Legal and sociological approaches to indigenous law in Southern Africa.’ Social Dynamics 8: 29-40.


1971     ‘Le pluralisme juridique dans l’œvre des philosophes du droit.’ Pp. 57-88 in John Gilissen (ed.), Le pluralisme juridique. Bruxelles: Editions de l’Université de Bruxelles.

[page 57]  

JONG, J.P.B. de Josselin de

1994     ‘Customary law: a confusing fiction.’ Pp. 111-117 in Alison Dundes Renteln and Alan Dundes (eds.), Folk Law: Essays in the Theory and Practice of Lex Non Scripta. New York and London: Garland Publishing Inc.. (Originally published as pp. 3-8 in IIIe Congrès international des sciences anthropologiques et ethnologiques. Koninklijke Vereeniging Indisch Instituut: Mededeling No. LXXX, Afd. Volkenkunde No. 29, 1948. Amsterdam: Indisch Instituut.)

KITAMURA, Takanori

n.d.      ‘Interview avec M. Chiba Masaji, le professeur de la sociologie du droit.’ In Wanda Capeller and Takanori Kitamura (eds.), Introduction aux cultures non-occidentales. Paris: Librairie Générale de Droit et de Jurisprudence. (Forthcoming)

LLEWELLYN, K.N., and E. Adamson HOEBEL

1941    The Cheyenne Way: Conflict and Case Law in Primitive Jurisprudence. Norman: University of Oklahoma Press.


1926      Crime and Custom in Savage Society. London: Routledge & Kegan Paul.

MERRY, Sally E.

1988      ‘Legal Pluralism.’ Law & Society Review 22: 869-896.

1992      ‘Anthropology, law, and transitional processes.’ Annual Review of Anthropology 21: 357.

MOORE, Sally F.

1978       Law as Process: An Anthropological Approach. London: Routledge & Kegan Paul.

NADER, Laura and Barbara YNGVESSON

1973     ‘On studying the ethnography of law and its consequences.’ Pp. 883-921 in John J. Honigmann (ed.), Handbook of Social and Cultural Anthropology Chicago: Rand McNally.


1967      Rechtskeuze en wording van een eenvormiges stadsgewoonte in de inlandse rechtbanken te Leopoldstad (Kinshasa), 1926-1940. Tervuren: Musée royal de l'Afrique centrale.

1968      ‘Les origines du droit urbain de Kinshasa.’ Revue juridique du Congo 44: 320-345.


1971     Anthropology of Law: A Comparative Theory. New York: Harper & Row.

1978     ‘The structure of society and its multiple legal systems.’ Pp. 96-109 in Philip H. Gulliver (ed.), Cross-Examinations: Essays in Memory of Max Gluckman. Leiden: E.J. Brill.

[page 58]  


1979     Order and Dispute: An Introduction to Legal Anthropology. Harmondsworth, England: Penguin Books.

ROULAND, Norbert

1994     Legal Anthropology (trans. Phippe G. Planel). London: Athlone Press.

SACK, Peter

1986     ‘Legal pluralism: introductory comments.’ Pp. 1-16 in Peter Sack and Elizabeth Minchin (eds.), Legal Pluralism. Proceedings of the Canberra Law Workshop VII. Canberra: Law Department, Research School of Social Sciences, Australian National University.

1992    ‘Introduction.’ Pp. xiii-xxxi in Peter Sack and Jonathan Aleck (eds.), Law and Anthropology, International Library of Essays in Law and Social Theory. Aldershot: Dartmouth Publishing Co.


1974     Corporations and Society. London: Duckworth.

SOUSA SANTOS, Boaventura de

1985     ‘On modes of production of law and social power.’ International Journal of the Sociology of Law 13: 299-336.

STARR, June and Jane F. COLLIER

1987      ‘Historical studies of legal change.’ Current Anthropology 28: 367-372.


1993      ‘The folly of the “social scientific concept” of legal pluralism.’ Journal of Law & Society 20: 192-217.

TWINING, William L.

1973      Karl Llewellyn and the Realist Movement. London: Weidenfeld & Nicolson.


1971     ‘Le pluralism juridique: essai de synthèse.’ Pp. 19-56 in John Gilissen (ed.), Le pluralisme juridique. Bruxelles: l’Université Bruxelles.

1989      ‘Return to legal pluralism: twenty years later.’ Journal of Legal Pluralism 28: 149-157.

1993     ‘Vers une nouvelle conception du pluralism juridique.’ Revue de la recherche juridique - droit prospectif XVIII: 573-583.


1993     ‘Trading places: recognizing and recreating legal pluralism in colonial Uganda.’ Journal of Legal Pluralism 33: 147-159.

WOODMAN, Gordon R.

1985     ‘Customary law, state courts, and the notion of institutionalization of norms in Ghana and Nigeria.’ Pp. 143‑163 in Antony Allott and Gordon R. Woodman (eds.), People's Law and State Law: the Bellagio Papers. Dordrecht: Foris.

1988a   ‘Unification or continuing pluralism in family law in Anglophone Africa: past experience, present realities, and future possibilities.’ [page 59] Lesotho Law Journal 4 (2): 33‑79.

1988b     ‘Book review.’ Journal of Legal Pluralism 27: 173-178.

1992       ‘Book review.’ Journal of Legal Pluralism 32: 141-148.

1993a     ‘Folk Law’ (trans. S. Boyron). Pp. 262-265 in André-Jean Arnaud (dir.), Dictionnaire encyclopédique de théorie et de sociologie du droit (2nd ed.). Paris: Librairie Générale de Droit et de Jurisprudence.

1993b     ‘Non-state, unbounded, unsystematic, non-western law.’ Pp. 103-115 in Masaji Chiba (ed.), Sociology of Law in Non-Western Countries. Oñati Proceedings: 15.

n.d.        ‘Droit comparé général.’ In Wanda Capeller and Takanori Kitamura (eds.), Les cultures juridiques non-occidentals. Collection Droit et Société. Paris: Librairie Générale de Droit et de Jurisprudence. (Forthcoming)


1993       Virtuous Citizens, Disruptive Subjects: Order and Complaint in a New England Court. New York: Routledge.


[1]Vanderlinden’s essay includes a wideranging and stimulating exploration of the origins and varieties of legal pluralism according to his definition. This seems to me to be of lasting value, but a full discussion of it would not be relevant here.

[2]A fuller, and helpful account of relevant aspects of this book is given in Ingber 1971. There is a useful critical discussion of the argument of Gurvitch 1947 in Henry 1973: 51-57.

[3]See also Gurvitch 1947: 5-6, propounding the “well-known facts of the origins and persistence of jural relations entirely independent of the state”.

[4]It would not have been impossible for legal anthropology to conduct such studies: cf. e.g. Yngvesson (1993). In the period now under consideration Llewellyn and Hoebel make reference to situations of legal pluralism (Llewellyn and Hoebel 1941: 27-28, 50-55). They assert the existence of various types of ‘law-stuff’ at various levels in a society in terms which could include state law pluralism. Given Llewellyn’s background as a lawyer, it seems likely that he, if not Hoebel, had in mind at least some instances of state law pluralism.

[5]He also states it to be: “the presence in a social field of more than one legal order...” (1986: 1). There seems to be no significant difference between these. I make use of that quoted in the text partly because it appears in the main text of his paper, whereas the other appears in the abstract, and because the reference to behaviour emphasises that the aim is to contribute to social science.

[6]Cf. another effective attack on the monopolistic legal claims of the state in Hund (1982: 31).

[7]My reading of Griffiths leads to the conclusion that he is clear and consistent on this issue. This appears to be doubted by Rouland, who writes that:


[f]or Griffiths, medieval law is not pluralist because the geographical diversity of custom and the internal law of corporations are areas of law to which the state was prepared to accommodate itself. We find that this is taking things too far...  (1994: 56-57).


I have been unable to find the passage in Griffiths’ article to which this comment is directed. Griffiths’ view would seem to imply that “custom and the internal law of corporations” are distinct legal orders. If the state “accommodated itself” to them, this meant presumably that there were few or no contradictions between those legal orders and the legal order of the state. But neither Griffiths nor any other writer has suggested that legal pluralism exists only when there are contradictions between the constituent elements, nor that the state or any other legal order cannot accommodate itself to other legal orders with which it coexists. Moreover, such non-state legal orders do not become elements in state law pluralism merely because the state formally incorporates them into its own body of norms. They are transformed into state law, and ‘legal pluralism in the strong sense’ is avoided, only if the state succeeds in eradicating the autonomy of the social fields in which they have been effective.

[8]Pauwels 1968 is a summary in French of his publication of 1967, which has not been available to me. While Pauwels does not appear to use the expression pluralisme juridique, his analysis of the types of legal scenarios in the local courts of Kinshasa employs a carefully defined category of pluralisme (1968: 336-340).

[9]It could be argued that Smith ‘starts with’ the segmentary lineage (Smith 1974: Chap. 1). But he certainly finishes with a picture of a society of which pluralism is a feature, writing:


Pluralism is a condition in which members of a common society are internally distinguished by fundamental differences in their institutional practice. (Smith 1974: 205; emphasis added)

[10]Another exception is Carbonnier 1972: 150. That argument is not developed in detail, and is answered by Rouland 1994: 58-59.


[11]Thus Merry writes:


I think it is essential to see state law as fundamentally different [from all other forms of social ordering] in that it exercises the coercive power of the state and monopolizes the symbolic power associated with state authority (1988: 879).


However, it is not clear whether she intends to draw a distinction as fundamental as Tamanaha’s. Her overt reason, that state law involves the power and authority of the state, is not a reason, but a tautology. Her implied reason is perhaps that state law is more effective than other types of ordering. As a general proposition that is highly questionable (cf. her own discussion, 1988: 879-880), and in any case refers only to differences in degree between common characteristics of different normative orders. Cf Benda-Beckmann 1988: 900.

[12]Another sign of the special status given to state law, despite the wish to exclude the claims of legal centralism, is the fact that studies of legal pluralism almost invariably focus on instances of the coincidence of a state law with one or more non-state laws. It is frequently assumed that every instance of legal pluralism takes this form (Merry: 873, 881-886). If the concept of legal pluralism were taken more seriously, there might be more studies of the relations between two or more non-state laws. Another instance of legal pluralism theory which accepts the distinction and which may in consequence confer an exaggerated status on state law may be found in Chiba 1989 (discussed Woodman 1992; see also Woodman n.d.).

[13]Tamanaha (1993: 209) notes my suggestion that the criterion of existence of state law is different from that of folk (non-state) law, citing a preliminary publication of what was later Woodman 1993a. In so far as this suggested a strict distinction between the essential natures of state law and folk law, rather than a difference in the degrees in which certain common characteristics are present, I now believe it was mistaken. Tamanaha’s discussion has helped me to see this.

[14]Tamanaha (1993: 194), referring to the suggestion that we can proceed with research without formulating a precise, universal definition of law (which now appears in Woodman, 1993a: para. 3), argues that this “rather blithe attitude” is inadequate. Unless we have a definition of law at the outset, he contends, the question of borderline cases cannot even arise. This seems questionable. The recognition that the borders extend widely need not result in a fatal lack of definition of the field of research. It may still be possible to formulate a view of the subject which will be adequate for the identification of objects of inquiry, and for most other purposes. The discussion of non-state law in the literature at present proceeds meaningfully and constructively on the basis of express or implied definitions of law which do not delineate its boundaries with great precision. Among anthropologists these have commonly been in terms either of social norms or of disputing processes.

[15]This appears to be in substance the same as that in his 1989 paper:


[T]he condition of the person who, in his daily life, is confronted in his behaviour with various, possibly conflicting, [legal] regulatory orders... emanating from the various social networks of which he is, voluntarily or not, a member. (Vanderlinden 1989: 153-154).

[16]I suggest this formulation of the “ideological commitment” which Sack has in view because it makes use of words and ideas which seem to be implied or stated in the passage cited. But cf. Sack 1992: xxi, where he refers to “a switch from legal positivism and its bias in favour of an increasing centralization, unification and uniformization of ‘law’” which is involved in “legal pluralism as [an] ideology”. This suggests that the ideological commitment involved in legal pluralism means support for developments whereby law becomes more localised, multiple and diverse.

[17]I do not list the other merits which Merry claims for the concept, partly because this paper is concerned with criticism rather than praise, but partly also because I am not convinced that so many of the interesting recent developments in legal thought can be attributed to the concept of legal pluralism. It is perhaps possible that realisation of the weakness of the law/society dichotomy was hastened by use of the concept. It seems unlikely that, for example, the move away from an exclusive focus on disputes to social ordering in non-dispute situations, or the interpretive perspective of Geertz (1983), would not have emerged even if the concept of legal pluralism had never been proposed.

[18]Some of the arguments presented here were considered in Woodman 1993b.