7 W. H. R. Rivers The aim of these lectures is to demonstrate the close connection which exists between methods of denoting relationship or kinship and forms of social organization, including those based on different varieties of the institution of marriage. In other words, my aim will be to show that the terminology of relationship has been rigorously determined by social conditions and that, if this position has been established and accepted, systems of relationship furnish us with a most valuable instrument in studying the history of social institutions. In the controversy of the present and of recent times, it is the special mode of denoting relationship known as the classificatory system which has formed the chief subject of discussion. It is in connection with this system that there have arisen the various vexed questions which have so excited the interest - I might almost say the passions - of sociologists during the last quarter of a century. I am afraid it would be dangerous to assume your familiarity with this system, and I must therefore begin with a brief description of its main characters. The essential feature of the classificatory system, that to which it owes its name, is the application of its terms, not to single individual persons, but to classes of relatives which may often be very large. Objections have been made to the use of the term 'classificatory' on the ground that our own terms of relationship also apply to classes of persons; the term 'brother', for instance, to all the male children of the same father and n " the term 'uncle' to all the brothers of the and the mother as well as to the husbanc aunt, while the term 'cousin' may denote larger class. It is, of course, true that m. our own terms of relationship apply to ( of persons, but in the systems to whii' word 'classificatory' is usually appliei classificatory principle applies far more \ and in some cases even, more logical! consistently. In the most complete form classificatory system there is not one term of relationship the use of which t that reference is being made to one perse to one person only, whereas in our own s there are six such terms, viz., husband, father, mother, father-in-law and motl law. In those systems in which the classifr principle is carried to its extreme degree term is applied to a class of persons. Th 'father', for instance, is applied to all whom the father would call brother, í all the husbands of those whom the n calls sister, both brother and sister bein; in a far wider sense than among ourseb some forms of the classificatory syste. term 'father' is also used for all those the mother would call brother, and for husbands of those whom the father wou sister, and in other systems the applicat the term may be still more extensive. Sirr the term used for the wife may be applied to those whom the wife would call sister and KINSHIP AND SOCIAL ORGANIZATION 137 s of all those whom the speaker calls ,'\\. mul ier an^ sister aSam being used in a " '. . - «rise than in our own language. ! !-. j,lsyficatory system has many other which mark it off more or less sharply - own mode of denoting relationship, j ■ . not think it would be profitable to , ■ a full description at this stage of our . As I have said, the object of these s to show how the various features of ficatory system have arisen out of, and efore be explained historically by, ■ :ts. If you are not already acquainted , ■ se features, you will learn to know ; more easily if at the same time you i # they have come into existence. i ■ begin with a brief history of the sub-long as it was supposed that all the . of the world denoted relationship in way, namely, that which is customary urselves, there was no problem. There reason why the subject should have d any interest, and so far as I have ; to find, it is only since the discovery Lssificatory system of relationship that the problem now before us was ever raised. I imagine that, if students ever thought about the matter at all, it must have seemed obvious that rhť way in which they and the other known peoples of the world used terms of relationship W.1» conditioned and determined by the social relations which the terms denoted. The state of affairs became very different as si)o:i as it was known that many peoples of the world use terms of relationship in a manner, arid according to rules, so widely different from out own that they seem to belong to an altogether different order, a difference well illustrated by the confusion which is apt to arise v. hen we use English words in the translation of classificatory terms or classificatory terms as tin- equivalents of our own. The difficulty or impossibility of conforming to complete truth .'ml reality, when we attempt this task, is the Ihm witness to the fundamental difference be-tuiľirn the two modes of denoting relationship. I do not know of any discovery in the whole r.inge of science which can be more certainly put to the credit of one man than that of the cLvrificatory system of relationship by Lewis Morgan. By this I mean, not merely that he was i be irst to point out clearly the existence of this mode of denoting relationship, but that it was he who collected the vast mass of material by which the essential characters of the system were demonstrated, and it was he who was the first to recognize the great theoretical importance of his new discovery. It is the denial of this importance by his contemporaries and successors which furnishes the best proof of the credit which is due to him for the discovery. The very extent of the material he collected [1871] has probably done much to obstruct the recognition of the importance of his work. It is a somewhat discouraging thought that, if Morgan had been less industrious and had amassed a smaller collection of material which could have been embodied in a more available form, the value of his work would probably have been far more widely recognized than it is today. The volume of his materia! is, however, only a subsidiary factor in the process which has led to the neglect or rejection of the importance of Morgan's discovery. The chief cause of the neglect is one for which Morgan must himself largely bear the blame. He was not content to demonstrate, as he might to some extent have done from his own material, the close connection between the terminology of the classificatory system of relationship and forms of social organization. There can be little doubt that he recognized this connection, but he was not content to demonstrate the dependence of the terminology of relationship upon social forms the existence of which was already known, or which were capable of demonstration with the material at his disposal. He passed over all these early stages of the argument, and proceeded directly to refer the origin of the terminology to forms of social organization which were not known to exist anywhere on the earth and of which there was no direct evidence in the past. When, further, the social condition which Morgan was led to formulate was one of general promiscuity developing into group-marriage, conditions bitterly repugnant to the sentiments of most civilized persons, it is not surprising that he aroused a mass of heated opposition which led, not merely to widespread rejection of his views, but also to the neglect of lessons to be learnt from his new discovery which must have received general recognition long before this, if they had not been obscured by other issues. 138 W. H. R. RIVERS The first to take up the cudgels in opposition to Morgan was our own pioneer in the study of the early forms of human society, John Ferguson McLennan [1876: 331]. He criticized the views of Morgan severely and often justly, and then pointing out, as was then believed to be the case, that no duties or rights were connected with the relationships of the classificatory system, he concluded that the terms formed merely a code of courtesies and ceremonial addresses for social intercourse. Those who have followed him have usually been content to repeat the conclusion that the classificatory system is nothing more than a body of mutual salutations and terms of address. They have failed to see that it still remains necessary to explain how the terms of the classificatory system came to be used in mutual salutation. They have failed to recognize that they were either rejecting the principle of determinism in sociology, or were only putting back to a conveniently remote distance the consideration of the problem how and why the classificatory terms came to be used in the way now customary among so many peoples of the earth. This aspect of the problem, which has been neglected or put on one side by the followers of McLennan, was not so treated by McLennan himself. As we should expect from the general character of his work, McLennan clearly recognized that the classificatory system must have been determined by social conditions, and he tried to show how it might have arisen as the result of the change from the Nair to the Tibetan form of polyandry [1876:373]. He even went so far as to formulate varieties of this process by means of which there might have been produced the chief varieties of the classificatory system, the existence of which had been demonstrated by Morgan. It is quite clear that McLennan had no doubts about the necessity of tracing back the social institution of the classificatory system of relationship to social causes, a necessity which has been ignored or even explicitly denied by those who have followed him in rejecting the views of Morgan. It is one of the many unfortunate consequences of McLennan's belief in the importance of polyandry in the history of human society that it has helped to prevent his followers from seeing the social importance of the classificatory system. They have failed to see that the classificatory system may be the r neither of promiscuity nor of polyanorv yet has been determined, both in its ^ei ■. character and in its details, by forms of s organization. Since the time of Morgan and Mel oi few have attempted to deal with the que- . in any comprehensive manner. The pro . . has inevitably been involved in the control *, which has raged between the advocates o- ■. original promiscuity or the primitive mo:., amy of mankind, but most of the former been ready to accept Morgan's views hli. while the latter have been content to u explain away the importance of conJiľ .. derived from the classificatory system \\ ii ■ attempting any real study of the evidence ■' the side of Morgan there has been one e> tion in the person of Professor j. Köhler j If ' " who has recognized the lines on wiucr problem must be studied, while on the ( side there has been, so far as I am aw arc. ■ ■ one writer who has recognized that the evidence from the nature of the classificalorv system of relationship cannot be ignored or belittled, but must be faced and some explanation alternative to that of Morgan provided. This attempt was made four years ago by Professor Kroeber [1909], of the University of California. The line he takes is absolutely to reject the view common to both Morgan and McLennan that the nature of the classificirory system has been determined by soci.il conditions. He explicitly rejects the view dial the mode of using terms of relationship depends on social causes, and puts forward as ihe »liter-native that they are conditioned b\ causes purely linguistic and psychological. It is not quite easy to understand what is meant by the linguistic causation of renn-» of relationship. In the summary at the end of his paper Kroeber concludes that 'they ircrm-» of relationship) are determined primári] > by language'. Terms of relationship, however. are elements of language, so that Kroebefs proposition is that elements of language are determined primarily by language. In so far as this proposition has any meaning, it must he í hat. in the process of seeking the origin of I ingiiisiic phenomena, it is our business to ignore any hut linguistic facts. It would follow that the student of the subject should seek the* antecedents of KINSHIP AND SOCIAL ORGANIZATION 139 ■uiisric phenomena in other linguistic phe- tneiia and put on one side as not germane . his task all reference to the objects and rela- (. which the words denote and connote. Professor Kroeber's alternative proposition hat terms of relationship reflect psychology, - sociology, or, in other words, that the way v h ich terms of relationship are used depends ■i chain of causation in which psychological )ce»es are the direct antecedents of this use. . . ill trv to make his meaning clear by means of instance which he himself gives. He says r at the present time there is a tendency om: ourselves to speak of the brother-in-■.■ as a brother; in other words, we tend to *s the brother-in-law and the brother toiler in the nomenclature of our own system relationship. He supposes that we do this hum.' there is a psychological similarity be-. :en the two relationships which leads us to ;s tliem together in our customary nomen-;l:i-c. I shall return both to this and other of his examples later. We have now seen that the opponents of Moi'iian have takenup two mainpositions which [t is possible to attack: one, that the classifies -torv svstem is nothing more than a body of terms of address; the other, that it and other moiies of denoting relationship are determined by psychological and not by sociological causes. I propose to consider these two positions in turn. Morgan himself was evidently deeply impressed by the function of the classificatory system of relationship as a body of salutations. His own experience was derived from the Nonli American Indians, and he notes the exclusive use of terms of relationship in address, a iisa^e so habitual that an omission to recognize a relative in this manner would amount almost to au a i front. Morgan also points out, as one moriu1 ior the custom, the presence of a reluctance u» utter personal names. McLennan had to rely entirely on the evidence collected by Morgan, and there can be no doubt that he was greatly influenced by the stress Morgan himsell laid on the function of the classificatory terms a1- mutual salutations. That in rude societies certain relatives have social functions definitely assigned to them by custom was known m Morgan's time, and I think it might even men have been discovered that the relation- ships which carried these functions were of the classificatory kind. It is, however, only by more recent work, beginning with that of Howitt, of Spencer and Gillen, and of Roth in Australia, and of the Cambridge Expedition to Torres Straits, that the great importance of the functions of relatives through the classificatory system has been forced upon the attention of sociologists. The social and ceremonial proceedings of the Australian aborigines abound in features in which special functions are performed by such relatives as the elder brother or the brother of the mother, while in Torres Straits I was able to record large groups of duties, privileges and restrictions associated with different classificatory relationships. Further work has shown that widely, though not universally, the nomenclature of the classificatory system carries with it a number of clearly defined social practices. One who applies a given term of relationship to another person has to behave towards that person in certain definite ways. He has to perform certain duties towards him, and enjoys certain privileges, and is subject to certain restrictions in his conduct in relation to him. These duties, privileges and restrictions vary greatly in number among different peoples, but wherever they exist, I know of no exception to their importance and to the regard in which they are held by all members of the community. You doubtless know of many examples of such functions associated with relationship, and I need give only one example. In the Banks Islands the term used between two brothers-in-law is wulus, walus, or walui, and a man who applies one of these terms to another may not utter his name, nor may the two behave familiarly towards one another in any way. In one island, Merlav, these relatives have all their possessions in common, and it is the duty of one to help the other in any difficulty, to warn him in danger, and, if need be, to die with him. If one dies, the other has to help to support his widow and has to abstain from certain foods. Further, there are a number of curious regulations in which the sanctity of the head plays a great part. A man must take nothing from above the head of his brother-in-law, nor may he even eat a bird which has flown over his head. A person has only to say of an object 'That is the head of your brother-in-law', and 140 W. H. R. RIVERS the person addressed will have to desist from the use of the object. If the object is edible, it may not be eaten; if it is one which is being manufactured, such as a mat, the person addressed will have to cease from his work if the object be thus called the head of his brother-in-law. He will only be allowed to finish it on making compensation, not to the person who has prevented the work by reference to the head, but to the brother-in-law whose head had been mentioned. Ludicrous as some of these customs may seem to us, they are very far from being so to those who practise them. They show clearly the very important part taken in the lives of those who use the classifica-tory system by the social functions associated with relationship. As I have said, these functions are not universally associated with the classificatory system, but they are very general in many parts of the world and only need more careful investigation to be found even more general and more important than appears at present. Let us now look at our own system of relationship from this point of view. Two striking features present themselves. First, the great paucity of definite social functions associated with relationship, and secondly, the almost complete limitation of such functions to those relationships which apply only to individual persons and not to classes of persons. Of such relationships as cousin, uncle, aunt, father-in-law, or mother-in-law there may be said to be no definite social functions. A schoolboy believes it is the duty of his uncle to tip him, but this is about as near as one can get to any social obligation on the part of this relative. The same will be found to hold good to a large extent if we turn to those social regulations which have been embodied in our laws. It is only in the case of the transmission of hereditary rank and of the property of a person dying intestate that more distant relatives are brought into any legal relationship with one another, and then only if there is an absence of nearer relatives. It is only when forced to do so by exceptional circumstances that the law recognizes any of the persons to whom the more classificatory of our terms of relationship apply. If we pay regard to the social functions associated with relationship, it is our own system, rather than the classificatory, which is open to the reproach that its relationships Ca into them no rights and duties. In the course of the recent work of the pe Sladěn Trust Expedition in Melanesia ' Polynesia I have been able to collect a bntk facts which bring out, even more clc' than has hitherto been recognized, the depc ence of classificatory terms on social j-j» [Rivers 1919]. The classificatory systems Oceania vary greatly in character. In sC places relationships are definitely distingue in nomenclature which are classed wirb t)| relationships elsewhere. Thus, whik- n] Melanesian and some Polynesian systems h a definite term for the mother's brother and the class of relatives whom the mother c brother, in other systems this relative is clas with, and is denoted by, the same term as father. The point to which I now call your ,nt tion is that there is a very close correlation tween the presence of a special term for i relative and the presence of special f unci i-attached to the relationship. In Polynesia, both the Hawaiians and inhabitants of Niue class the mother's brot with the father, and in neither place w.i able to discover that there were any spei duties, privileges or restrictions ascribed the mother's brother. In the Polynesian islai of Tonga and Tikopia, on the other ha where there are special terms for the moiln brother, this relative has also special functio The only place in Melanesia where I failed find a special term for the mother's brother \ in the western Solomon Islands, and that v also the only part of Melanesia where í ŕai to find any trace of special social funcik ascribed to this relative. I do not knim such functions in Santa Cruz, but my inlorr tion about the system of that island is derb from others, and further research will aim certainly show that they are present. In my own experience, then, among i different peoples, I have been able to esiabl a definite correlation between the presence a term of relationship and special funcrii associated with the relationship. Informal kindly given to me by Father Egidi, huwu' seems to show that the correlation among Melanesians is not complete. In Mek en, mother's brother has the duty of putting the first perineal garment of his nephtu, 1 KINSHIP AND SOCIAL ORGANIZATION 141 no special term and is classed with the Among the Kuni, on the other hand, p is a definite term for the mother's brother nguishing him from the father, but yet he not so far as Father Egidi knows, any ial functions. rn jn Melanesia and Polynesia a similar elation comes out in connection with í relationships, the most prominent excep-being the absence of a special term for the er's sister in the Banks Islands, although this -ive has very definite and important func-s In these islands the father's sister is -ed with the mother as vev or veve, but i here, where the generalization seems to Jc down, it does not do so completely, for father's sister is distinguished from the her as veve vus rawe, the mother who kills , as opposed to the simple veve used for the her and her sisters. here is thus definite evidence, not only for association of classificatory terms of rela-ship with special social functions, but from part of the world we now have evidence ;h shows that the presence or absence of ial terms is largely dependent on whether e are or are not such functions. We may it as established that the terms of the classificatory system are not, as McLennan supposed, merely terms of address and modes of mutual salutation. McLennan came to this conclusion because he believed that the classificatory terms were associated with no such functions as those of which we now have abundant evidence. He asks, 'What duties or rights are affected by the relationships comprised in the classificatory system?' and answers himself -iccording to the knowledge at his disposal, 'Absolutely none' [1876: 366], This passage makes it clear that, if McLennan had known what we know today, he would never have taken, up the line of attack upon Morgan's position in which he has had, and still has, so many followers. 1 c.u i now turn to the second line of attack, that h Inch boldly discards the origin of the terminology of relationship in social conditions, and seeks for its explanation in psychology. The line of argument I propose to follow is first to show that many details of classificatory systems have been directly determined by social factors. If that task can be accomplished, we shall have firm ground from which to take off in the attempt to refer the general characters of the classificatory and other systems of relationship to forms of social organization. Any complete theory of a social institution has not only to account for its general characters, but also for its details, and I propose to begin with the details. I must first return to the history of the subject, and stay for a moment to ask why the line of argument I propose to follow was not adopted by Morgan and has been so largely disregarded by others. Whenever a new phenomenon is discovered in any part of the world, there is a natural tendency to seek for its parallels elsewhere. Morgan lived at a time when the unity of human culture was a topic which greatly excited ethnologists, and it is evident that one of his chief interests in the new discovery arose from the possibility it seemed to open of showing the uniformity of human culture. He hoped to demonstrate the uniformity of the classificatory system throughout the world, and he was content to observe certain broad varieties of the system and refer them to supposed stages in the history of human society. He paid but little attention to such varieties of the classificatory system as are illustrated in his own record of North American systems, and seems to have overlooked entirely certain features of the Indian and Oceanic systems he recorded, which might have enabled him to demonstrate the close relation between the terminology of relationship and social institutions. Morgan's neglect to attend to these differences must be ascribed in some measure to the ignorance of rude forms of social organization which existed when he wrote, but the failure of others to recognize the dependence of the details of classificatory systems upon social institutions is rather to be ascribed to the absence of interest in the subject induced by their adherence to McLennan's primary error. Those who believe that the classificatory system is merely an unimportant code of mutual salutations are not likely to attend to relatively minute differences in the customs they despise. The credit of having been the first fully to recognize the social importance of these differences belongs to j. Kohler. In his book Zur Urgeschichte der Ehe, which I have 142 W. H. R. RIVERS already mentioned, he studied minutely the details of many different systems, and showed that they could be explained by certain forms of marriage practised by those who use the terms. I propose now to deal with classificatory terminology from this point of view. My procedure will be first to show that the details which distinguish different forms of the classificatory system from one another have been directly determined by the social institutions of those who use the systems, and only when this has been established, shall I attempt to bring the more general characters of the classificatory and other systems into relation with social institutions. I am able to carry out this task more fully than has hitherto been possible because I have collected in Melanesia a number of systems of relationship which differ far more widely from one another than those recorded in Morgan's book or others which have been collected since. Some of the features which characterize these Melanesian systems will be wholly new to ethnologists, not having yet been recorded elsewhere, but I propose to begin with a long familiar mode of terminology which accompanies that widely distributed custom known as the cross-cousin marriage. In the more frequent form of this marriage a man marries the daughter either of his mother's brother or of his father's sister; more rarely his choice is limited to one of these relatives. Such a marriage will have certain definite consequences. Let us take a case in which a man marries the daughter of his mother's brother, as is represented in Figure 7.1. One consequence of the marriage between C and d will be that A, who before the marriage of C was only his mother's brother, now becomes also his wife's father, while b, who before the marriage was the mother's brother's wife of C, now becomes his wife's mother. I i B=a A=b C d E f Figure 7.1 Key-, Capital letters are used to represent men and the smaller letters women. Reciprocally, C, who before his marriage \ been the sister's son of A and the husbar sister's son of b, now becomes their son-law. Further, E and f, the other children o and b, who before the marriage had been o the cousins of C, now become his wi brother and sister. Similarly, a, who before the marriage o was her father's sister, now becomes also husband's mother, and B, her father's sist husband, comes to stand in the relation of h band's father; if C should have any broil-and sisters, these cousins now become brothers- and sisters-in-law. The combinations of relationship wh follow from the marriage of a man with daughter of his mother's brother thus differ a man and a woman, but if, as is usual, a n may marry the daughter either of his moth brother or of his father's sister, these collations of relationship will hold good for b men and women. Another and more remote consequence the cross-cousin marriage, if this become;, established institution, is that the relations!1 of mother's brother and father's sisters li band will come to be combined in one r the same person, and that there will be a si lar combination of the relationships of faih sister and mother's brother's wife. If the en cousin marriage be the habitual custom, l>; b in Diagram 1 will be brother and sister: consequence A will be at once the modi brother and the father's sister's husband ot while b will be both his father's sister and mother's brother's wife. Since, however, mother's brother is also the father-in-law,; the father's sister the mother-in-law, three ■ ferent relationships will be combined in e. case. Through the cross-cousin marriage relationships of mother's brother, f.irh sister's husband and father-in-law will be a bined in one and the same person, :ind relationships of father's sister, mother's liner's wife and mother-in-law will be simik combined. In many places where we know the crc cousin marriage to be an established msti tion, we find just those common designate which I have just described. Thus, in the Ml dialect of Fiji the word vungo is applied to mother's brother, the husband "of the fatli' KINSHIP AND SOCIAL ORGANIZATION ________ 143 ■uid the father-in-iaw. The word nganei is for the father's sister, the mother's broth-v-ilV and the mother-in-law. The term , is used by a man for the son of the sr'a brother or of the father's sister as js for the wife's brother and the sister's .jid. Ndavola is used not only for the child . mother's brother or father's sister when in" in sex from the speaker, but this word »ušed by a man for his wife's sister and his ťr"s wife, and by a woman for her hus-s brother and her sister's husband. Every f these details of the Mbau system is the and inevitable consequence of the cross-i marriage, if it becomes an established abitual practice. s |:ijian system does not stand alone in iL-sia. In the southern islands of the New de.-., in Tanna, Eromanga, Aneityum and .i. í he cross-cousin marriage is practised \c'ir systems of relationship have features r to those of Fiji. Thus, in Aneityum the matak applies to the mother's brother, ther's sister's husband and the father-in-vhile the word engak used for the cross-i is not only used for the wife's sister and other's wife, but also for the wife herself, dn, in the island of Guadalcanal in the Lons the system of relationship is just s would result from the cross-cousin mar-One term, «id, is used for the mother's ;r and the wife's father, and probably also e father's sister's husband and the hus-> father, though my stay in the island was ■ng enough to enable me to collect suffi-genealogical material to demonstrate points completely. Similarly, tarunga in-in its connotation the father's sister, the :r's brother's wife and the wife's mother, robably also the husband's mother, while >rd iva is used for both cross-cousins and :rs- and sisters-in-law. Corresponding to ...- ^rminology there seemed to be no doubt that it was the custom for a man to marry the daughter of his mother's brother or his father's sister though I was not able to demonstrate this tain of marriage genealogically. These three regions, Fiji, the southern New Hcnrides and Guadalcanal, are the only parts ot Melanesia included in my survey where I «uind the practice of the cross-cousin mar-"ajie, and in all three regions the systems of relationship are just such as would follow from this form of marriage. Let us now turn to inquire how far it is possible to explain these features of Melanesian systems of relationship by psychological similarity. If it were not for the cross-cousin marriage, what can there be to give the mother's brother a greater psychological similarity to the father-in-law than the father's brother, or the father's sister a greater similarity to the mother-in-law than the mother's sister? Why should it be two special kinds of cousin who are classed with two special kinds of brother-and sister-in-law or with the husband or wife? Once granted the presence of the cross-cousin marriage, and there are psychological similarities certainly, though even here the matter is not quite straightforward from the point of view of the believer in their importance, for we have to do not merely with the similarity of two relatives, but with their identity, with the combination of two or more relationships in one and the same person. Even if we put this on one side, however, it remains to ask how it is possible to say that terms of relationship do not reflect sociology, if such psychological similarities are themselves the result of the cross-cousin marriage? What point is there in bringing in hypothetical psychological similarities which are only at the best intermediate links in the chain of causation connecting the terminology of relationship with antecedent social conditions? If you concede the causal relation between the characteristic features of a Fijian or Aneityum or Guadalcanal system and the cross-cousin marriage, there can be no question that it is the cross-cousin marriage which is the antecedent and the features of the system of relationship the consequences. I do not suppose that, even in this subject, there will be found anyone to claim that the Fijians took to marrying their cross-cousins because such a marriage was suggested to them by the nature of their system of relationship. We have to do in this case, not merely with one or two features which might be the consequence of the cross-cousin marriage, but with a large and complicated meshwork of resemblances and differences in the nomenclature of relationship, each and every element of which follows directly from such a marriage, while no one of the 144 W. H. R. RIVERS systems I have considered possesses a single feature which is not compatible with social conditions arising out of this marriage. Apart from quantitative verification, I doubt whether it would be possibie in the whole range oi science to find a case where we can be more confident that one phenomenon has been conditioned by another. I feel almost guilty of wasting your time by going into it so fully, and should hardly have ventured to do so if this case of social causation had not been explicitly denied by one with so high a reputation as Professor Kroeber. I hope, however, that the argument will be useful as an example of the method I shall apply to other cases in which the evidence is less conclusive. The features of terminology which follow from the cross-cousin marriage were known to Morgan, being present in three of the systems he recorded from Southern India and in the Fijian system collected for him by Mr Fison. The earliest reference [Grant 1870: 276] to the cross-cousin marriage which I have been able to discover is among the Gond of Central India. This marriage was recorded in 1870, which, though earlier than the appearance of Morgan's book, was after it had been accepted for publication, so that I think we can be confident that Morgan was unacquainted with the form of marriage which would have explained the peculiar features of the Indian and Fijian systems. It is evident, however, that Morgan was so absorbed in his demonstration of the similarity of these systems to those of America that he paid but little, if any, attention to their peculiarities. He thus lost a great opportunity; if he had attended to these peculiarities and had seen their meaning, he might have predicted a form of marriage which would soon afterwards have been independently discovered. Such an example of successful prediction would have forced the social significance of the tern ■ ology of relationship upon the attention of J " dents in such a way that we should have V spared much of the controversy which ha long obstructed progress in this branch of c ology. It must at the very least have acted ■. stimulus to the collection of systems of r ■■ tionship. It would hardly have been pass that now, more than forty years after ihe .. pearance of Morgan's book, we are stil complete ignorance of the terminology of r tionship of many peoples about whom vohir have been written. It would seem impossi. ■ for instance, that our knowledge of lnts. systems of relationship could have been \\ ■. it is today. India would have been the com --in which the success of Morgan's predici would first have shown itself, and such event must have prevented the almost t< neglect which the subject of relationship . .. suffered at the hands of students of lnc ■ sociology. REFERENCES Grant, C., 1870. Gazetteer of Central Provin 2nd ed., Nagpur. Kohler, J., 1897. Zur Urgeschichte der Ehr. l . schrift für vergleichende Rechtswissenich 12, pp. 187-353. Kroeber, A. L., 1909. Classificatory System' Relationship. Journal of the Royal Anthrn logical Institute, 39, pp. 77-84. McLennan, J. F., 1876. Studies in Ancient /: tory, London. Morgan, L. H., 1871. Systems of Consangni. and Affinity of the Human Family. Smith-.or ■ Contributions to Knowledge, 17, Washingrt Rivers, W. H. R., 1914. The History of Mclm ion Society, 2 vols. Cambridge. 8 Structural Analysis in Linguistics and in Anthropology Claude Lévi-Strauss listics occupies a special place among the 1 sciences, to whose ranks it unquestion-belongs. It is not merely a social science he others, but, rather, the one in which by ie greatest progress has been made. It is ibly the only one which can truly claim to science and which has achieved both the ■ ulation of an empirical method and an rstanding of the nature of the data submit-o its analysis. This privileged position ;s with it several obligations. The linguist jften find scientists from related but differ-Jisciplines drawing inspiration from his .pie and trying to follow his lead. Noblesse p.. A linguistic journal like Word cannot ne itself to the illustration of strictly linie theories and points of view. It must also ">me psychologists, sociologists, and an-lologists eager to learn from modern lin-ics the road which leads to the empirical .'ledge of social phenomena. As Marcel ■is wrote already forty years ago: "Soci-' would certainly have progressed much er if it had everywhere followed the lead : linguists___"l The close methodological »gy which exists between the two discip-imposes a special obligation of collaboration upon them. F\er since the work of Schrader2 it has been mi necessary to demonstrate the assistance which linguistics can render to the anthropologist in the study of kinship. It was a linguist and a philologist (Schrader and Rose)3 who showed the improbability of the hypothesis of matrilin-eal survivals in the family in antiquity, to which so many anthropologists still clung at that time. The linguist provides the anthropologist with etymologies which permit him to establish between certain kinship terms relationships that were not immediately apparent. The anthropologist, on the other hand, can bring to the attention of the linguist customs, prescriptions, and prohibitions that help him to understand the persistence of certain features of language or the instability of terms or groups of terms. At a meeting of the Linguistic Circle of New York, Julien Bonfante once illustrated this point of view by reviewing the etymology of the word for uncle in several Romance languages. The Greek Osioc, corresponds in Italian, Spanish, and Portuguese to zio and tio; and he added that in certain regions of Italy the uncle is called barba. The "beard," the "divine" uncle -what a wealth of suggestions for the anthropologist! The investigations of the late A. M. Hocart into the religious character of the avuncular relationship and the "theft of the sacrifice" by the maternal kinsmen immediately come to mind.4 Whatever interpretation is given to the data collected by Hocart (and his own interpretation is not entirely satisfactory), there is no doubt that the linguist contributes to the solution of the problem by reveaiing the tenacious survival in contemporary vocabulary of relationships which have long since disappeared. At the same time, the anthropologist