The Logic of the Laws

Preface

The research upon which this thesis is based was financially supported by the Dutch and Canadian Governments. I received four consecutive doctoral fellowships from the Canada Council in the academic years from 1971 to 1975. I also received two of the Netherlands Government Scholarships which are awarded to Canadian nationals by the Dutch Ministry of Education and Science. The International Affairs Section of the Ministry has also provided financial assistance to help defray the printing costs of the thesis. Without this support the research, writing, and publication of the following could have been neither undertaken nor completed.

Academic tradition at Leiden University does not allow me to thank the individual members of the teaching staff who have assisted me in the preparation of this thesis. I wish, therefore, to collectively thank the members of the Faculty of Letters and the Faculty of Social Sciences who have directly and indirectly contributed to the preparation of the thesis. The staff of the Koninklyk Instituut voor Taal-, Landen Volkenkunde have graciously assisted me in the innumerable ways that are only possible in such a research institution. In particular, F. G. P. Jaquet introduced me to and helped me to decipher the often confusing patterns of Dutch archival and manuscript collections. My wife Gretchen, to whom this thesis is dedicated, in addition to typing and retyping drafts of the thesis, carefully edited the entire text. My friend and colleague F. E. Tjon Sie Fat provided much valuable assistance in puzzling through many of the mathematical subtleties of the legal codes of this study. Additionally, our lengthy discussions which encompassed some of the most improbable topics were simultaneously refreshing and enlightening. To all of these individuals, many of whom must remain anonymous, I wish to express my thanks for without them this work would have never come to fruition.

A NOTE ON ORTHOGRAPHY

In an anthropological study such as this where the majority of the material is drawn from sources dating from the nineteenth century, the problem of orthography is particularly acute. Thus I have attempted to adhere to the following rules. When there was confusion as to which rule to apply I must admit that I have tended to follow the procedure advocated by T. E. Lawrence (1940, pp. 18-20) for the transcription of Arabic.

1) When referring to indigenous concepts in a non-specific or general sense, I have used the orthographic convention for Malay and Indonesian that has been in use since August 1972. The basic rules of this system are set forth in a pamphlet published by the Indonesian government and entitled Pedoman Edjaan Bahasa Indonesia Jang Disempurnakan. The main conventions that involve changes from previous usage are in the following table.

Malay (previously) Indonesian (previously) Combined System j dj

y j y ny nj ny sh sj sy ch tj c kh ch kh

The approximate English equivalents of the combined system are as follows: j as the j in jump; y as the y in yes; sy as the sh in show; c as the ch in chop; kh as the ch in the Scottish pronunciation of loch. Though the ny sound does not normally occur in English it can be approximated by the combination of the last phoneme of “green” and the initial phoneme of “yacht? when the words “green yacht” are pronounced rapidly.

VIII THE LOGIC OF THE LAWS

2) When transcribing Arabic script into a Romanized form I have also used the 1972 convention. This means that my transcription of the Arabic forms is not a completely accurate philological evaluation of the original. For example, though in Arabic script the “s’s” in the words “semendo” and “fasal” are consistently represented by different forms I have only used a single “s” a philological study I have taken the view that the exact transcription of the Arabic script into Romanized form either would have been unnecessarily confusing for the anthropologist or would have made this book prohibitively expensive.

3) When directly citing published Dutch and English language sources I have retained the spelling of indigenous terms as in the original. A useful guide for converting these to the modern spelling can be found in paragraph 1. The Malay system roughly corresponds to English practice and the Indonesian system roughly corresponds to Dutch practice.

4) When directly citing manuscripts written in Romanized Malay I have retained the spelling of the original. All of the Romanized Malay manuscripts used in this study were prepared using Dutch orthography as a reference. Thus the consonant values tend to follow the pattern presented in paragraph 1 for the previous Indonesian system. The patterns for vowels and diphthongs is unfortunately perversely inconsistent and especially confusing for the native English speaker. The following observations on Dutch orthographic practice should provide some assistance.

Dutch oe, e.g., boek (book) like the vowel in the English word “tool” but shorter (longer before r), e.g. loeka = luka (Ind. wound).

Dutch 1, e.g. mijn (mine). This diphthong is problematical because there is considerable variation among Dutch speakers. Thus to the English speaker the diphthong may sound like the vowel in “mine” or “main”. However, with regard to texts from South Sumatra the form is found in association with the modern Indonesian “ai”, which may be approximated by the vowel in the English word “mine”, e.g., soengij = sungai (Ind. river) or sampaij = sempai (Ind. reach, extend to).

Dutch te, e.g., ziek (sick) like the vowel in the English word “week” but shorter (longer before r), e.g., lakie lakie = laki laki (Ind. man).

Dutch ou, e.g., koud (cold), similar to the vowel in the English word “how” and the Indonesian diphthong au. However, because this may be represented in a variety of ways in Dutch, e.g., blauw (blue) ; hout (wood); saus (sauce); and bouw (building, structure) there is

. Since this is an anthropological and not

A NOTE ON ORTHOGRAPHY IX considerable variation in South Sumatran texts, e.g., atouw = atau (Ind. or); karbauw = karbau (Ind. buffalo); and Soengij Lemou = Sungai Lemau (Ind. place name).

5) When referring to a marriage form or legal concept specifically (i.e., with reference to a single law text) I have regarded the reference as a direct citation and thus either the rule in paragraph 2 or 4 applies, depending on the nature of the text.

Preface

Appendix IT .

Key to Manuscript References

Bibliography .

Maps

Indexes

Chapter 1 Introduction

In 1888 Prof. G. A. Wilken published an article in the Bijdragen tot de Taal-, Land- en Volkenkunde van Nederlandsch-Indié entitled “De Verbreiding van het Matriarchaat op Sumatra” (The Spread of Matriarchy on Sumatra). The Royal Institute, which was responsible for the publication of the article, sent a number of copies of the article to the Minister of Colonies together with a request that they be forwarded to civil servants in Sumatra so that they might conduct research in the field in order to fill any gaps in the data (Wilken, 1912, II, p. 220; 1891, p. 150). As a result the Ministry received a number of memoranda that were forwarded to the Royal Institute and ultimately to Wilken. Nine of these memoranda were from the Residency of Palembang and provided excellent ethnographic data covering most of the region." With these memoranda available, Wilken wrote an article entitled “Over het Huwelijks- en Erfrecht bij de Volken van Zuid- Sumatra” (Concerning the Marriage and Inheritance Law of the Peoples of South Sumatra) which was published in Volume 40 of the Bijdragen in 1891.7 This article was the first major attempt to produce an anthropological synthesis of the apparently widely disparate marriage forms to be found in South Sumatra.

Among the early Dutch ethnological scholars Wilken was exceptional in that he “firmly placed the study of Indonesian societies in the context of the dominant ethnological theory of his time, evolutionism” (De Josselin de Jong, 1968, p. 3). The marriage forms of South Sumatra, or as Wilken points out (Wilken, 1912, II, p. 219; 1891, p. 149), that part of Southern Sumatra formerly encompassed by Palembang, Bengkulu, and the Lampung Districts but excluding Jambi, presented an analytical problem that was bound to fascinate any nineteenth century evolutionist. In Wilken’s own words, “Nowhere does one come across so many marriage forms so close together that are in the most divergent stages of development and with so many nuances and transitional forms as in South Sumatra” (Wilken, 1912, II, p. 223; 1891, p. 153).

For any of the evolutionists the most significant marriage forms in South Sumatra would have been jujur and ambil anak. These two types of marriage, when considered in their most extreme forms, are virtually ideal types for the evolutionists’ notion of patriarchy and matriarchy. This is so much the case that it is perhaps useful to use William Marsden’s description of these forms because it is rather difficult to argue that Marsden, who was writing in the eighteenth century, was influenced by the dominant theoretical themes of the nineteenth century. “The jujur is a certain sum of money, given by one man to another, as consideration for the person of his daughter, whose situation, in this case, differs not much from that of a slave to the man she marries, and to his family” (Marsden’s History, 1811, p 257). In direct contrast to this there is ambil anak. “In the mode of marriage by ambel anak, the father of a virgin makes choice of some young man for her husband, generally from an inferior family, which renounces all further right to, or interest in, him... He lives in the family, in a state between that of a son and a debtor. He partakes as a son of what the house affords, but has no property in himself... He is liable to be divorced at their pleasure, and though he has children must leave all, and return naked as he came” (Marsden’s History, 1811, p. 262f.). What made this problem interesting to the evolutionist, and continues to make the problem interesting for the modern anthropologist, is the fact that these forms occurred not only in societies which were in close geographical proximity but also in the same society, simultaneously and in direct contrast to each other.

While the forms of jujur and ambil anak marriage, both in extreme and attenuated forms, dominate the ethnographic picture of South Sumatra, there is a multiplicity of forms that may be roughly labelled, if not in essence then by implication, as cognatic.* Such cognatic forms are usually matrilocal; however, the status of the children may vary substantially. Sometimes only one child must return to the father’s family, sometimes the children are divided, and sometimes both parents have equal rights to all of the children. And in some cases, as if the whole system was far too complicated for any definite rules, the decision is left to the child himself. Still further complicating the picture is the fact that in certain divorce situations jujur and ambil anak forms have acquired cognatic features (c.f. Dulken, 1862, pp. 303-304).

Wilken’s solution to this problem, not surprisingly, follows strict evolutionary lines. As a starting point he argues that, “In contrast to the inhabitants of Middle Sumatra, i.e., the various tribes of the Minangkebau Malays who practice matriarchy, the people of South Sumatra have patriarchy” (Wilken, 1912, II, p. 223; 1891, p. 153). Thus Wilken believed that the people of South Sumatra had passed through the matriarchal phase and had progressed far enough to be considered in the patriarchal phase. The key transitional step was when matriliny and patriliny existed side by side, i.e., a lineage could be continued by either a man or a woman. However, at least according to Wilken, the system gradually progressed until the rule became that the lineage was continued by sons unless there were no sons, in which case it was continued by a daughter (c.f. Wilken, 1912, II, p. 247; 1891, p. 176f.). At this point, by implication, the stage of patriarchy had been reached. Wilken’s argument is supported by the fact that in many cases it appears that jujur marriage was considered to be the norm, while ambil anak marriage was an optional rule most frequently applied when a family lacked sons. In this regard Wilken singles out Van der Meulen’s memorandum which applied to Ogan Ulu and Kumering Ulu. It is worth quoting from the original document in detail not only because it is relevant to Wilken’s argument but also because it helps to answer the still important question of what was the relation between patrilineal and matrilineal forms when they appeared as alternatives in the same society. However, any judgments on this latter point must be tempered by the fact that the memorandum was written in response to a query related to Wilken’s original article on “The ‘Spread’ of Matriarchy in Sumatra”.

When the brideprice (djoedjoer) is paid, the woman follows her husband to his family. Completely in agreement with the frequently expressed idea that the man has purchased (beli) his wife, the children stay with the family of the father.

If the father does not have any sons then one of the daughters (if possible the eldest) marries by “terambil anak lepas”. In this case the man is obliged to live with his wife [i.e., matrilocally]. The woman is the head of the family (gezin = strictly speaking, nuclear family) “tegah djoerei” because she is completely equated with a son. The children stay with the family of the mother. If there are marriageable daughters and only one young son then one of the daughters marries by terambil or “ka'ambil” anak (without, however, establishing a descent line (?) (stam)) until the son can marry and there is no fear that the descent line will die out in the male line. The woman is then free to follow her husband, who returns to his own family. If the son dies before he has any descendants, then the daughter’s marriage is changed to the “ka’ambil anak lepas” form and with a feast it is made known that she has established a “djoerei”.

By way of exception, it sometimes happens that an elder daughter establishes a “djoerei” while the father still has a young son, but according to the current superstition this may result in the sickness or death of the son. The reason for this exception is that the father is rich and powerful and fears that his “djoerei” will die out or that because his daughter has followed her husband she may be forced to do heavy work in the fields. (Algemeen Rijksarchief, Ministry of Colonies Archives (non-secret), 27 Sept. 1894, No. 41, Memorandum Van der Meulen, 12 Nov. 1889).

Thus Wilken’s argument that the dominant form is patriarchal in nature is based on the fact that at the level of norms the patrilineal form of marriage takes precedence over the matrilineal form. It is interesting to note that most modern social anthropologists, while avoiding the term patriarchy, would interpret this passage from Van der Meulen's memorandum as being primary evidence for the existence of a patrilineal descent system. Wilken’s argument is not based solely on the material provided by this memorandum but reflects a general feature associated with South Sumatran marriage systems that is to be perceived in the early literature, i.e., there is a strong general preference for the jujur or patrilineal and patrilocal form of marriage. The term preference is used advisedly. Although with regard to important positions and titles, the preference for patrilineal descent seems to acquire the strength of a prescription, even here the exceptions are frequent and important. And further, the status implications of jujur marriage derive not only from the positive values surrounding the principle of patrilineal descent, but also from the ability to pay what in many cases was an exorbitant bride price. Indeed, one has the impression that the major obstacle to the extensive use of jujur was the financial implications of the marriage payment. Thus, in the Lampung Districts where the size of jujur payments reached the most extreme proportions, Forbes (1885, p. 152) suggests that only the eldest son married by jujur while all of the others married by one or the other of the matrilocal forms.

In contrast to his view that patriarchy has replaced matriarchy as the dominant form in South Sumatra, Wilken argues that the cognatic system has begun to develop but has not yet replaced the patriarchal system. Wilken uses a very precise definition of cognatic, i.e., a system in which both parents have equal rights to all of the children (Wilken, 1912, II, p. 265; 1891, p. 193). While one might quibble about the precise definition of transition points, Wilken’s argument accords well with one aspect of the early literature. Where one finds internal reform pressures there is often evidence that either a marriage form already cognatic in its implications acquires a new and special emphasis or a cognatic form is introduced, or the jujur and ambil anak forms are reshaped in such a way that in certain circumstances cognatic rules apparently apply. For example, in John Marsden’s laws for Lais a semendo form cognatic in its implications is presented explicitly as a reform that is to overcome certain harmful effects of both jujur and ambil anak marriage (Marsden’s History, 1811, p. 225). Consistent with his view that the evolutionary sequence is matriarchy, patriarchy and cognatic Wilken mentions this example (Wilken, 1912, II, p. 267; 1891, p. 196) and emphasises the equality aspects of the marriage form by quoting Marsden, i.e., “This marriage is a regular treaty between the parties, on the footing of equality” (Marsden’s History, 1811, p. 263).

There are, however, a number of marriage forms in South Sumatra which are neither exclusively matrilineal nor exclusively patrilineal but do not imply legal equality between the marriage partners. The most common of these is a form frequently referred to as semendo balik jurai. This is a form of matrilocal marriage according to which one child belongs to the family of his father while the rest belong to the mother’s family. Another situation that frequently arises either as a distinct marriage form or as a circumstantial aspect of another form is the division of the children, i.e., the child belongs to his mother’s family or to his father’s family but not to both. Largely on the grounds that such forms are usually matrilocal and the man has an inferior legal position with respect to his wife, Wilken views forms such as these as being representative of a transition stage between patriarchy and matriarchy. i

Contrary to Wilken, largely as the result of the structural analysis of marriage laws, I have grouped together all forms that are neither strictly matrilineal nor strictly patrilineal. However, modern British Social Anthropology, using different criteria, would partition these nonlineal forms into two groups that are virtually identical to Wilken’s division. On the one hand, the definition of cognatic as equal rights to all children is still largely accepted. On the other, systems in which some of the children belong exclusively to one group and some of the children belong exclusively to another group, are often described as ambilineal.

Wilken’s solution to the problem presented by the multiplicity of South Sumatran marriage forms was consistent with the dominant evolutionary theories of his day. However, to regard this analysis only as an historical curiosity from a bygone anthropological era, is to do a fundamental injustice to his work. In particular, the manner in which he differentiated between types of forms is in many respects still acceptable. For example, the implicit distinction that he makes between cognatic and ambilineal types of marriage is still valid. And further, his assertion that the marriage system of South Sumatra is essentially patrilineal anticipates a style of argument that is still dominant in British Social Anthropology. However, sophisticated as these arguments are, they are nonetheless dominated by a theoretical bias that is no longer accepted in modern social anthropology. Thus when Wilken attempted his synthesis of the marriage systems of South Sumatra and found the solution in evolutionism, one might say that he was asking the right question but providing the wrong answer.

As Wilken correctly observed, the marriage systems of South Sumatra provide an unusually complex but exceptionally interesting problem for analysis. While modern anthropology rejects Wilken's approach, the problem is to find a useful and hopefully valid alternative to evolutionism. In contemporary anthropology there are two basic trends that may be employed to describe the relationships between the various marriage forms occuring in South Sumatra. One is the British social anthropological approach; the other is the Paris-Leiden structural approach. The social anthropological approach focuses on social interaction, especially with regard to institutionalized forms of behaviour. The structural approach focuses on the conceptual aspects of the social system, especially with regard to formally organized systems of categories.

In social anthropological terms the marriage forms of South Sumatra represent a series of institutionalized alternatives that regulate marriage. Accordingly, the study of the alternatives is primarily concerned with an examination of the factors that influence or govern the choice of marriage type. Thus the passage from Van der Meulen's memorandum quoted above can be interpreted as meaning that jujur marriage is used unless: 1) there are no sons to continue the family line, in which case a matrilineal and matrilocal form is used; 2) there is some doubt as to whether a son will survive, in which case a daughter marries by a matrilocal form which can be converted to a strict matrilineal form or to a patrilocal form as circumstances require; or 3) a wealthy man does not wish to marry his daughter out of his family.

Van der Meulen clearly indicates that the lack of sons or the concern for their survival is the dominant factor governing the choice of alternatives. In more general terms, the choice of alternatives is ultimately related to the demographic characteristics of the population.

In particular, how many individuals are there in the sibling group and what is the probability that at least one will be a male who will survive to produce offspring? However, this is not a simple demographic problem but is itself in turn dependent on certain aspects of the marriage system.

The Substance of the Report on the Condition of Society published in The Proceedings of the Agricultural Society established in Sumatra presents a detailed discussion of the situation in the early nineteenth century (+ 1820). “It appears that on an average of marriageable persons of both sexes, no less than one fourth, and often more, are in a state of celibacy, of which the majority are females. The excesses of unmarried females is chiefly owing to the emigration of young men, who wanting the means of marrying in their own country, pass into the neighbouring ones, where the difficulty of obtaining wives is less (p. 3f.) ... The late age at which marriages frequently take place may have a further effect in diminishing the number of children (p. 5)... The proofs of the obstacle which this custom, [jujur marriage payments] combined with poverty, presented to marriage, are obvious in every village throughout the country, where the number of Gadises or virgins, many of whom are bending under the weight of years, is not a little remarkable” (p. 15) (Proceedings II).° The essential point made by this material is that the analysis of the factors influencing the choice of alternatives requires the study and evaluation of circumstances and variables external to the marriage forms themselves. In South Sumatra the most significant of the external factors are of a demographic and economic nature.

Another approach within the general realm of social anthropology bases the analysis not so much on the factors governing the selection of alternatives but on the cumulative effects of the making of choices among the forms. Thus Jaspan’s (1964) argument that in the twentieth century the Rejang people have passed from a patrilineal system to a matrilineal one is in a large way based on the fact that the pattern of choice has changed from the regular selection of a patrilineal form to a regular selection of a matrilineal one.

Thus the social anthropological approach treats the various marriage forms of South Sumatra as institutionalized alternatives. By viewing these forms as alternatives one is obliged either to analyse the factors governing the choice between alternatives, or to analyse the cumulative effects of making choices, and preferably both. Furthermore, the analysis of these factors must ultimately become involved with circumstances and conditions existing in the society which are external to the system of formally organized marriage categories. On the other hand, the structural approach treats the marriage forms not as alternatives but as categories. This approach has one important additional aspect and that is the insistence that the relationships between categories be analysed in so far as possible in terms of the system itself and with as little reference as possible to external circumstances.

The structural approach in anthropology can be succinctly summarized by a postulate, a specification, and an application rule. The postulate states that the relations between elements are more important than the elements themselves. The specification is that elements are named categories. And, the application rule states that given a closed system of formally organized categories, the structural approach is especially appropriate to the analysis of that system.

Because there is substantial evidence that the marriage forms of South Sumatra form a closed system of formally organized categories, I have adopted a structural approach in an effort to elucidate the nature of the marriage systems. While the chapters that follow are a demonstration of how structural analysis can be used to describe the relations between the marriage forms of South Sumatra, it is perhaps useful at this point to indicate some of the major contrasts with the social anthropological approach. The example from Van der Meulen's memorandum (and this is not an isolated example) indicates that ambil anak marriage is an alternative only to be used when jujur cannot be used, thus suggesting that jujur is the dominant form and ambil anak is only a subsidiary alternative. However, at the conceptual level there is abundant evidence indicating that the two forms are of equal value and opposed. Hazairin expresses this eloquently and succinctly. “Oorspronkelijk kende het Redjangsche volk slechts als huwelijksvormen de semendo-ambil-anak huwelijken en de djoedjoer huwelijken. Beide huwelijksvormen staan in nauw verband met elkaar, de eene is de contravorm van de andere. De eene kan niet zonder den andere bestaan op straffe van vernietiging van het volksideaal” (Originally, the Rejang people only had semendo ambil anak marriage and jujur marriage as marriage forms. The one is the antithesis of the other. The one cannot exist without the other except at the risk of destroying the ideal order of the people (volksideaal) (Hazairin, 1936, p. 37). This is not an isolated idiosyncratic opinion on the part of the author but, as we shall see in the following analyses, a prevalent idea in South Sumatra.

Another difference between the social anthropological approach and the structural approach relates to the analytical treatment of the non-unilineal forms of marriage, i.e., those other than jujur and ambil anak. Among these non-unilineal forms two basic types can be readily distinguished. First, there are those forms in which both parents have egual rights to all children. Secondly, there are those forms in which both parents have rights to the children but each child is assigned to either the group of the mother or the group of the father but not to both. The first is usually labelled cognatic and the second ambilineal. However, as the passage from Hazairin suggests, and detailed analysis confirms, such a partitioning of the non-unilineal marriage forms does a fundamental injustice to the conceptual order. These forms are conceptually defined by their relation to the basic jujur/ambil anak opposition and as such their key feature is their lack of a strict, exclusive unilineal principle. To insist upon a further dichotomy is to impose an unwarranted and external logical notion on the data.

Thus the structural approach is more concerned with how the people of South Sumatra think about their marriage system than with how they use that system. The basic problem, therefore, is to find data that elucidate the conceptual aspects of the marriage system. Prior to the publication of Wilken's synthesis, the majority of the available literature concerning South Sumatra had been written by Dutch civil servants or before 1825 by English officials. From these sources one can construct a reasonably accurate statement as to the nature of the individual marriage forms and how they were used. However, with respect to the conceptual order the only substantive material to emerge relates to the conceptual systems of the Europeans and not the Sumatrans. Indeed, the biases and personal views of the various writers are so pronounced that they can be controlled and the sources used with considerable confidence. Paradoxically, this is in marked contrast to many of the later sources where anthropological thinking had begun to make an impact. Here the theoretical biases of writers (both civil servants and anthropologists) influenced their interpretation of phenomena to such a degree that reinterpretation is usually difficult and hazardous, if not impossible.

In an effort to better understand the ideas implicit in each marriage form, I began by looking for early sources that gave descriptions of each marriage form in an indigenous language (usually Middle Malay). Initially, the goal was an extremely modest one: how are the essential aspects of each marriage form expressed in the local language? In particular, I was interested in the choice of vocabulary and syntactic structures that were employed to differentiate one form from another.

The most important published source that provides such material is L. W. C. van den Berg's Rechtsbronnen van Zuid Sumatra (Legal Sources of South Sumatra) (1894). Here Van den Berg published a series of Malay language legal texts in Arabic script together with their translations. Though the quality and usefulness of the texts varies considerably, it is very clear that the Malay texts contain much more anthropologically significant information than do the translations. Additionally, this material indicates that at least some legal codes can be used to provide insights into the nature of the marriage systems of South Sumatra. However, the texts are not uniformly useful. The most serious problem affecting the use of Malay language legal codes for anthropological purposes is the fact that in some cases colonial authorities used legal codes not so much as a means for the documentation of local legal usages but as one way in which to implement certain “reforms”. |