I
Les Structures élémentaires de la parenté is intended to be an introduction to a general theory of kinship systems (Lévi-Strauss 1949: xi), and it starts with the explanation of a universal phenomenon: the prohibition of incest. The first chapters of the book are devoted to the consideration of this problem and are reprinted without modifications in the second edition of the book (1967). In the preface to this second edition Lévi-Strauss states, however, that ‘many new facts and the development of my own thought mean that nowadays I would no longer express myself in the same way’, although he still believes that ‘the prohibition of incest is to be explained entirely in terms of sociological causes’ (1967: xv).
The reason why we shall consider here Lévi-Strauss’s appraisal of the incest taboo as he expounded it in 1949 is that, although he says he would not express himself in the same way nowadays, he does not propose an alternative solution. Moreover, his consideration of the incest prohibitions still remains as the theoretical counterpart of his definition of ‘elementary structures’. Even though it is possible to consider the validity and applicability of the concept of elementary structures quite independently from Lévi-Strauss’s conception of the incest prohibitions, the ultimate reason for the very existence of elementary structures is referred by Lévi-Strauss to the analytically preliminary existence of the incest prohibitions.
The same kind of considerations apply to the analysis of the concepts of ‘nature’ and ‘culture’, as used by Lévi-Strauss in 1949. He states in the preface to the second edition of his book that his proposal is ‘to trace the line of demarcation between the two orders guided by the presence or absence of articulated speech’, although he would again express himself in this respect more in accordance with the new discoveries in this field (1967: xvi). As a matter of fact, when considering these concepts in Les Structures élémentaires, he does not trace the line of demarcation guided by the presence or absence of articulated speech but, rather, by the presence absence or of the incest prohibitions themselves. The concepts of nature and culture are, on the other hand, related to the incest prohibitions by Lévi-Strauss’s own definition, and we shall try to analyse not their relevance as sociological concepts but the logic of their relationship with the incest prohibitions as established by Lévi-Strauss.
II
The explanation of the incest prohibitions consists, according to Lévi-Strauss, in discovering ‘what profound and omnipresent causes account for the regulation of the relationships between the sexes in every society and age’ (1949: 29).
He criticizes the older theoreticians (Havelock Ellis, Wester-marck, Morgan, Frazer, Durkheim) who have dealt with the problem, because they explained it (1) by natural causes, or (2) exclusively as a cultural phenomenon, or, even when taking into account the double character – natural and cultural – of the phenomenon, (3) by proposing an extrinsic connection between them.
For Lévi-Strauss, the prohibition of incest ‘is in origin neither purely cultural nor purely natural’, although it belongs to both domains; he suggests it might be the source of facts where the bridge between Nature and Culture ought to be studied. It is not possible to distinguish by any experimental research, he explains, what is Nature and what is Culture in human behaviour. No ‘analyse réelle’ can provide an explanation of that bridge. Therefore, only a phenomenon such as the incest prohibitions possessing the characteristics of both domains, i.e. being at the same time universal and governed by rules, offers a suitable point of departure.
But Lévi-Strauss does not intend only a characterization of the ‘incest taboo’, he also tries to explain why it occurs, the ‘profound and omnipresent cause’ for its existence. It exists, Lévi-Strauss explains, because Culture always provides a rule whenever the human group is confronted with a scarce or random distributed value (cf. p. 41). In this case, women are the scarce value, and the incest taboo is the device to maximize it.
III
Let us start with the analysis of Lévi-Strauss’s characterization of the incest taboo as a ‘bridge’ between Nature and Culture. On page 35 of Les Structures élémentaires, he says:
even if the incest prohibition has its roots in nature it is only in the way it affects us as a social rule that it can be fully grasped.
But the fact is not only that it is a phenomenon that has its roots in Nature, and that it has to be apprehended as Culture; it is the domain where:
there only, but there finally culture can and must, under pain of not existing, firmly declare ‘Me first,’ and tell Nature, ‘You go no further’ (1949: 38).
The incest taboo is, therefore, the introduction of ‘Culture’ into human life. Before it, ‘culture is still non-existent; with it, nature’s sovereignty over man is ended’ (p. 31).
What are the definitions of the three terms, namely ‘Culture’, ‘Nature’, and ‘incest prohibitions’, involved in this reasoning? In the first three chapters of Les Structures élémentaire s, the connotations of ‘Nature’ and ‘Culture’ are the following:
| Nature is the domain of: | Culture is the domain of: |
| universality (p. 9) | particularity (p. 9) |
| lack of rules (p. 9) | rules (p. 9) |
| spontaneity (p. 12) | non-spontaneity (p. 12) |
| repetitive processes (p. 37) | cumulative processes (p. 37) |
| biological heredity (p. 37) | alliance (p. 38) |
Whence he derives:
(i) the prohibition of incest is universal;
(ii) the prohibition of incest is a rule;
(iii) therefore, it belongs to both domains.
In Lévi-Strauss’s own words:
This rule [the prohibition of incest] is at once social, in that it is a rule, and pre-social, in its universality and the type of relationships upon which it imposes a norm (1949: 13).
Let us examine the consistency of these definitions. The particular concept that Lévi-Strauss is going to explain is defined by reference to two sets, the defining characteristics of which are pairs of opposites. The set of elements labelled as ‘incest prohibitions’, i.e. the rules concerning sexual relations in a group, share two defining characteristics of the set Nature, namely, universality and the sort of relations to which they refer, and all the characteristics of the set Culture. In this sense, one would say that, in fact, incest prohibitions do not belong to either of the referential sets. ‘Incest prohibitions’ do not belong to Nature because they possess only two of the defining characteristics of this set; and they do not belong to Culture either, because while possessing all the defining characteristics of this set, they also possess two characteristics of the opposite set. Lévi-Strauss is aware of this problem when he says, for instance, that the incest taboo ‘is neither purely cultural, nor purely natural’ (1949: 30).
IV
It seems at this point that there are some problems here concerning the demarcation of the sets or the application of the defining criteria, and that these are responsible for the conceptualization of something as belonging and not belonging to each of the opposite sets as defined. The problem becomes more evident if one thinks of any other possible rule. If a rule concerning incest prohibitions belongs to Nature because of ‘the type of relationships upon which it imposes a norm’, we can consider other examples in which the connection with Nature in this sense is also obvious, thus, rules concerning ‘alimentation’ or ‘cooking’. They would be considered by Lévi-Strauss as being part of Nature because of the kind of needs they satisfy and because they would doubtless be ‘universal’, but they would be also ‘particular’ and therefore part of Culture as well.
In general, any rule whatsoever can be thought of as related to ‘biology’ or ‘basic needs’ in some way. In this sense, all rules are bound to fall within the province of Nature as well as being part of Culture because of the very fact that they are rules. Any rule can be included in a larger class defined by the sort of ‘needs’ it satisfies, and as such it would be considered ‘universal’ and as belonging to Nature, because it would then constitute ‘a general condition of Culture’ (Lévi-Strauss 1949: 30).
V
If the analysis of human behaviour is made by reference to Nature and Culture as opposite concepts, it seems that these concepts ought to permit a clear categorization of any fact under analysis. In other words, one of the inherent characteristics of any sort of typology or opposition used in any kind of analysis is that the opposite concepts or types have to be mutually exclusive. If they are not, this does not reveal an anomalous characteristic in the phenomenon under study but merely a defect in the typology proposed.
In the case of Lévi-Strauss’s use of the concepts of Nature and Culture, the defect comes mainly from the application of the first pair of opposites listed for them, namely, universality and particularity. The fact that there are different cultural answers to a unique universal ‘need’ does not make these different answers ‘universal’: what remains universal is the ‘need’. If the answers are different for each society, they are particular. The fact that in any known society there exists a certain regulation of sexual relations is no different, in a sense, from the fact that any society possesses a certain kind of political organization or a special type of economic system. But the fact that political institutions or economic systems can be defined for any known society does not lead us to consider these institutions as a part of Nature. The large variety of institutions that can be labelled as ‘political’ are obviously not universal. The human ‘needs’ to which these institutions respond are probably universal, but not the institutions themselves. And even this last phrase is controversial, because different institutions that can be categorized under the same label, i.e. ‘political’, ‘economic’, etc., can respond to different ‘needs’ and can therefore have very different meanings.
The case of the incest prohibitions is no different. The reasons why they do exist have to be sought in each particular case. In this sense, or even following Lévi-Strauss’s own argument, the references to ‘Nature’ and ‘Culture’ when explaining the incest prohibitions do not throw any light on the problem. Moreover, it seems completely out of the question that they could do so.
VI
Apart from the definition of the incest prohibitions by reference to ‘Nature’ and ‘Culture’, the universal reason to which these rules respond, according to Lévi-Strauss, still remains to be analysed.
If the ‘incest taboo’ constitutes a rule, it ought to be an institutionalized deviation from random behaviour in a particular sphere of action. The sphere of action involved is sexual relations among certain social categories, and the question is to discover why in this sphere behaviour deviates from the random at all.
For Lévi-Strauss, these prohibitions have a functional value, namely to ‘freeze’ women within the family ‘so that their distribution, or the competition for them, is within the group, and under group and not private control’ (Lévi-Strauss 1949: 55).
The same idea is developed by White in a paper published in 1948. For him, the so-called ‘incest taboo’ should properly be considered as a form of social organization where the units involved are ‘sociological rather than biological relationships’ (1948: 418). Incest prohibitions and exogamy are explained in terms of cooperation (which can only take place when the evolution of articulated speech makes communication possible) and alliances. The prohibition of marriage with certain categories of people is explained, following Tylor’s argument, by the imperative of ‘marrying out or being killed out’, as the basis of alliances between groups, the groups themselves having been formed previously because of the advantages of human cooperation.
Lévi-Strauss’s own explanation does not differ very much from White’s, and can be summarized as follows:
1 [Men live in groups];
2 When something is scarce (or stochastically distributed) and yet necessary for the biological continuance of a group, it becomes an ‘economic good’;
3 In order to ‘maximize’ that economic good, a group creates and transmits ‘rules’ about it. That is, it conceptualizes that good as an economic good and establishes some sort of patterned behaviour in order to distribute it;
4 Women, in primitive societies, constitute an economic good because they are stochastically distributed and necessary for reproduction;
5 Some kind of organization (set of rules) is, therefore, necessary for the distribution of women;
6 The incest prohibitions are the first necessary step for such an organization.
In Lévi-Strauss’s words:
This problem of intervention [Culture introducing prohibitions of marriage] is not raised just in this particular case. It is raised, and resolved in the affirmative, every time the group is faced with the insufficiency or the risky distribution of a valuable of fundamental importance (Lévi-Strauss 1949: 39).
VII
But, for Lévi-Strauss, if the incest prohibitions constitute some sort of organization, this is not due to the prohibitions themselves but to the fact that:
the prohibition of incest is a rule of reciprocity. The woman whom one does not take, and whom one may not take, is for that very reason, offered up (1949: 45).
Moreover,
every prohibition is at the same time, and under another aspect, a prescription (1949: 56).1
This prescription is expressed in certain kinds of kinship system, those systems ‘in which the nomenclature permits the immediate determination of the circle of kin and that of affines, that is those systems which prescribe marriage with a certain type of relative’ and which he calls ‘elementary structures’ (Lévi-Strauss 1949: ix).
This is why, some years later, Lévi-Strauss explains the sequence of the themes in Les Structures élémentaires as follows:
The question we asked ourselves was that of the ‘meaning’ of the incest prohibition . . . it was necessary, then, to establish the systematic nature of each kinship terminology and its corresponding set of marriage rules (1960: 28).
VIII
Let us analyse some questions involved in Lévi-Strauss’s argument. The main idea is that the incest prohibitions constitute the basis for the distribution of women because they imply at the same time a principle of reciprocity. This principle of reciprocity is expressed, as we have seen above, in the ‘kinship terminology and its corresponding set of marriage rules’. But not any kinship terminology expresses the principle of reciprocity. Only those that classify, within the social group, all the possible spouses into two categories: prohibited and prescribed. That is, only those kinship systems that can be defined as ‘elementary structures’. These structures contain, according to Lévi-Strauss, a distinction between parallel cousins and cross-cousins which constitutes an ‘extremely simple and efficient method for maintaining an indefinite extensible balance of matrimonial exchanges between consanguineal groups’. This distinction is, on the other hand, ‘nothing else than the positive aspect of a rule of which the prohibition of incest . . . represents only the negative aspect’ (Lévi-Strauss 1955: 108).
Therefore the elementary structures are the reverse of the incest prohibitions; they are the conversion of a negative rule into a set of stipulations of a different order (cf. Lévi-Strauss 1949:56).
But, in order to demonstrate that systems which prescribe a certain social category as spouse are the reverse of incest prohibitions, one should be able to prove the connection between them independently. If prescriptive systems (elementary structures) exist because there is a ‘basic principle of reciprocity’ underlying the incest prohibitions, one cannot give as a proof of the existence of this principle the fact that prescriptive systems do exist.
Furthermore, there is always the question of the universality of the incest prohibitions and the comparatively rare cases of societies with prescriptive systems. If ‘every prohibition is at the same time . . . a prescription’, then, whenever one finds prohibitions of incest one should be able to find a prescriptive rule of marriage. This correlation is obviously not valid, and neither therefore is the argument. But the argument seems to be invalid not only because it contains logical inconsistencies, but because it includes also an erroneous conceptualization of the relationship between ‘incest’ and ‘marriage’.
IX
We remarked above on the convenience of searching for the reasons for the existence of prohibitions of incest in each particular case. The ‘meaning’ of each rule prohibiting incest could vary according to the sort of social categories it involves and the kind of society where it is expressed (cf. Needham 1971b).
But even with these considerations in mind, a rule that prohibits incest cannot be the counterpart of a rule that prescribes marriage because they refer to two different things. The former refers to sexual relations, the latter to marriage. It is true that this latter concept (marriage) does not have a consolidated meaning in the anthropological literature (cf. Rivière 1971), but whatever its uses it can never be reduced to the counterpart of incest prohibitions.
1 This phrase does not appear in the English edition (1969: 45). The editor informs me that the omission was not made at the instruction of Professor Lévi-Strauss but is the result of a regrettable oversight at one stage in the retyping of the draft translation.