Lewis Henry Morgan's Comparisons: Reassessing Terminology, Anarchy and Worldview in Indigenous Societies of America, Australia and Highland Middle India 9781789203189

Citation preview

Lewis Henry Morgan’s Comparisons

Lewis Henry Morgan’s Comparisons Reassessing Terminology, Anarchy and Worldview in Indigenous Societies of America, Australia and Highland Middle India Georg Pfeffer

berghahn NEW YORK • OXFORD www.berghahnbooks.com

First published in 2019 by Berghahn Books www.berghahnbooks.com © 2019 Georg Pfeffer All rights reserved. Except for the quotation of short passages for the purposes of criticism and review, no part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system now known or to be invented, without written permission of the publisher. Library of Congress Cataloging-in-Publication Data A C.I.P. cataloging record is available from the Library of Congress Library of Congress Cataloging in Publication Control Number: 2019015097 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library

ISBN 978-1-78920-317-2 hardback ISBN 978-1-78920-318-9 ebook

Contents List of Figures and Tables vii Acknowledgements ix Introduction The Initiative

1

Chapter 1 The Life and Work of Lewis Henry Morgan (1818–81)

9

Chapter 2 Tools and Types

27

Chapter 3 Seneca Revisited

39

Chapter 4 Omaha Skewing Reconsidered

62

Chapter 5 Highland Middle Indian Terminologies

83

Chapter 6 Schneider, Relatedness, ‘Malayan’ and a General Comparison

103

Chapter 7 Social Evolution and the Australian Anomaly

118

Chapter 8 Order in Anarchy: HMI ‘Gentile Organization’ Compared

135

Chapter 9 Bridewealth and Gender in Highland Middle India

164

vi • Contents

Chapter 10 The Dark Side of the Moon

186

Conclusion For the Record

201

Glossary 209 References 210 Index 220

Figures and Tables Figures Affines of two sisters become consanguines 31 Two-line Piaroa relationship terminology (Ego female) 33 Two-line Piaroa relationship terminology (Ego male) 33 Simplified two-line terminological system of the Piaroa and many others 34 2.5 Simplified four-line terminological system of alternating reciprocal affinal relations 36 3.1 Affinal status of Seneca clan categories (example: Bear clan) 45 3.2 The Seneca Iroquois classificatory system 54 3.3 Seneca term [29] Ah-ge⁀ah’ne⁀o equating HZ, WB, BWws, ZHms, FBSWws, FZDHms, FBDHms, FZSWws, MBDHms, MBSWws, MZSWws and MZDHms 56 3.4 Children of affines become consanguines by opposing new affines jointly with consanguines. 58 3.5 Former affines become consanguines by jointly opposing new affines in the subsequent generation 58 4.1 Parental generation: categories to be classified as pairs of either affines or consanguines 69 4.2 Seneca affinal opposition in G+1 69 4.3 Omaha affinal opposition in G+1 69 4.4 Unified Omaha terminology (male Ego) 72 4.5 Unified Omaha terminology (male Ego) emphasizing skewing: consanguineal lines (in black), affinal lines (in grey), undefined terms of G2 in light grey 74 4.6 The second affinal line 76 4.7 Two encompassed symmetric Omaha relationships 77 4.8 The second consanguineal line 77 4.9 Five-line Omaha terminological structure (male Ego) 78 4.10 Unified Omaha terminology (female Ego) 80 4.11 Five-line Omaha terminological structure (female Ego) 81 2.1 2.2 2.3 2.4

viii • Figures and Tables

5.1 Seneca term Ah-ge͒ -ah‘-ne-o͒ indicating encompassed affinal symmetry 85 5.2 Kharia terminology collected in 1982 in the Kolebira village, Gumla district, Jharkhand province 89 5.3 Doubling of encompassed affinal symmetry in G0 of HMI terminologies 94 5.4 Encompassed tetradic structure of affinal relations 95 5.5 Hypothetical symmetric affinal relationships in the Juang and Ho terminologies 97 5.6 HMI terminological structure 97 9.1 Kuttia Kond sacred youth performance when meeting affines 170 9.2 The ‘atom of kinship’ in caste India 181 9.3 The ‘atom of kinship’ as experienced in HMI 181 11.1 Indicators of Seneca and Omaha symmetry 204 11.2 Indicators of HMI symmetry 205

Tables List of Seneca Iroquois terms 53 List of Omaha terms for a male Ego 66 List of Omaha terms for a female Ego 67 List of Kharia terms 88 List of Malay consanguineal terms 110 List of common marker-affixes for Malay terms 112 Features of classificatory terminologies compared 114 Three societal forms: features compared 128 Indigenous seniority concepts of the Chota Nagpur complex 146 8.2 Overview of HMI ‘gentile organization’ 159

3.1 4.1 4.2 5.1 6.1 6.2 6.3 7.1 8.1

Acknowledgements  The ethnographic basis of this book is a result of fieldwork encouraged and supported by Mrs Dugli Nayak and Dr Radhakanth Nayak, I.A.S, Secretary to the Government of India (retired) and founder patron of the National Institute of Social Work and Social Sciences (NISWASS) in Bhubaneswar/Odisha. Over about a quarter of a century, almost every year they enabled me to recover my spirits and my physical energy at their home after spending some months in the highlands of their province. Since 1972, they have allowed me to benefit from their personal integrity, their neverending enthusiasm, their intellectual inspiration and their steadfastness when faced with adverse pressure. My sincere gratitude also goes to Dr Eva Reichel, whose wonderful command of English helped me to improve the language of this book, just as her anthropological queries made me reconsider a good number of its important issues.

Introduction The Initiative

Recently I brought out a book on about forty different societies of which the respective constitutions were determined by kinship as conceived by real people (Pfeffer 2016). The work was based upon outstanding anthropological monographs of the English-speaking world and showed that these social systems displayed a rather limited number of formal variations. It was thus meant to indicate how anthropos, irrespective of the given particular history, had tested and refined a relatively small number of styles to conceive and organize society, i.e. institutional patterns quite different from those of the Weberian legal-bureaucratic mode that had been disseminated across the globe, nowadays overshadowing all others. Thereafter, on noticing that Lewis Henry Morgan’s first major comparative and theoretical work had been completed about 150 years ago, I realized that ever since this outstanding achievement, the additional theoretical insights into the subject had not really been overwhelming compared to Morgan’s giant step in the 1860s. The time seemed to have come to reassess specific aspects of this achievement. For the readers, such a venture may require certain clarifications that can be seen as a major outcome of anthropological efforts since this beginning in the nineteenth century.

Clarifications Kinship is not ‘kinship’. The former, for better or worse, is the basic tie connecting a person’s family members and perhaps more distant relatives, as long as they are identified as kin and understood to share some of the person’s ancestors. By contrast, ‘kinship’ stands for the structure of many societies that used to preoccupy sociocultural anthropology, since they were known as anarchic or else as peasant societies. Kinship terms are labels to address or to refer to kin. The latter are specific persons and as such are the smallest units of our species. They

2 • Lewis Henry Morgan’s Comparisons

are Ego’s kin, with the position of Ego standing for the speaker articulating the terms. A single kin is known to have resulted from physical activities, i.e. procreation and birth, performed by the common parents or more distant forebears that these kin share with Ego. According to this definition, kinship results from acts of nature, though in very rare cases, natural actors may be substituted by official institutions executing formal measures of individual adoption. ‘Kinship’ terms, on the other hand, are – irrespective of genealogical links – labels of sociocultural categories providing public order for numerous past and present societies across the globe. Ego does not simply apply these terms in order to designate a limited number of individuals naturally related to him or her; rather, their application allocates a public status to each person involved in this process. Accordingly, addresses and references provide essential social orientation to members of the given anarchic or peasant community. Such generalized applications of ‘kinship’ terms always follow a specific pattern of the social categories that appears to result from arbitrary principles of classification. Because of such apparent arbitrariness, the pattern is an object of sociocultural research. No known human actors are understood to have introduced these structures of ‘kinship’ terms. The categories and their variant pattern result from a given classification. Morgan recognized and explained this difference between kinship and ‘kinship’, though he did not use my phrases and arguments in his writings. In his own words, he also pointed out that kinship was not a relevant subject of scholarly preoccupation, whereas ‘kinship’ would answer basic questions on the sociocultural existence of anthropos. After his ethnographic book on the Iroquois of New York State (1901 [1851]), Morgan compared the relationship terminologies of many Native North American peoples with each other and with European as well as Middle Eastern nomenclatures. As has been frequently described, he then continued the same kind of comparative inquiries by sending out questionnaires to residents of other continents, before finally bringing out the results of this research on Systems of Consanguinity and Affinity of the Human Family (1871) in his scholarly magnum opus. Some years later, Morgan tried to describe and distinguish several ‘periods’ of human evolution. The focus of these discussions were two ‘plans of government’, the earlier one ‘founded on persons and gentilism’ (societas) and the later one ‘founded on territory and property’ (civitas) as elaborated initially in Part I of his second and more popular work on Ancient Society (1877). In his Part II, he offered for comparison an elaborate account of the stateless ‘gentile organization’ governing the Iroquois. Morgan’s contemporaries and numerous later anthropolo-

Introduction • 3

gists as well as Marxist thinkers intensely and controversially debated his findings in several major contributions. In all quarters, he is considered to be a founding figure of the discipline. A return to Morgan’s comparisons is the primary objective of my book. It will also offer, in the course of basically new theoretical assessments and arguments, elaborate references to the social structure of the many millions of contemporary Indigenous people inhabiting the vast Highlands of Middle India between the River Ganges and, further south, the Godavara, the Aravalli Hills in the west, and those approaching the Bay of Bengal in the east. These ethnographic passages are mostly based upon my personal fieldwork. Almost every year between 1980 and 2002, I had been able to spend several months in this culture area. To understand the general context of Morgan’s comparisons, Chapter 1 will recall his political activism against the ongoing ‘removal’ of the Native North American peoples, his ethnographic research and the theoretical core of the refined twists in his discovery of the distinction between ‘descriptive’ and ‘classificatory’ terminologies. The account will greatly depend upon Thomas R. Trautmann’s seminal work (1987), but will also argue with some of Morgan’s most important critics, notably with the towering figure of Alfred Kroeber (1909), who had probably executed the major blow against the reputation of Systems of Consanguinity and Affinity of the Human Family.

Discussions In order to cope with the different theoretical points of the critics as well as my own divergences from Morgan, the technical prerequisites of my later analyses will have to be understood. Accordingly, Chapter 2 will present an overview and a critical explanation of the analytical tools applied in the subsequent dissections of terminologies. The technical limits of these instruments will also be indicated. At the end of this chapter, I will also add an explanation as to why the well-known ‘terminology types’, such as ‘Iroquois’ or ‘Hawaiian’, which are usually taught in undergraduate seminars, should be removed altogether from the anthropological syllabus. Chapters 3–5 will present formal analyses of relationship terminologies that have either been collected and studied by Morgan’s fieldwork in North America (Seneca and Omaha) or by my own ethnographic endeavours in India, though the Omaha data are taken from the corrected list of Robert H. Barnes (1984). After subjecting them to minute

4 • Lewis Henry Morgan’s Comparisons

inspection, I will contend that these systems must be understood as containing the basic constitutional principles of their respective societies. In other words, these relationship terminologies carry a very different meaning compared to the descriptions of concrete individual kin ties by European nomenclatures. It is indisputable that Morgan had understood and elaborated this critical difference, but the results of my analyses, though praising his discovery, will radically depart from his findings because they take into account the classificatory principle of affinity that he, like most of his successors, failed to conceive at all. Chapter 3 ‘revisits’ the Seneca terminology that had been more familiar to Morgan than any other and had probably instigated his comparisons to become his primary object of demonstration in Systems of Consanguinity and Affinity. My examinations begin by juxtaposing the idea of a ‘kinship terminology’, as defined by the eminent analyst Floyd G. Lounsbury (1964b), with that of a ‘relationship terminology’ in the understanding of the methodological masterpiece written by Alan Barnard and Anthony Good (1984). After a critical discussion of Lounsbury’s findings and Morgan’s misunderstandings, the final result will propose that the Seneca system as a whole must be viewed as an affinal order of the symmetric kind, albeit one proscribing the repetition of intermarriage. Affinity will be the key concept of this book. It is thus advisable for readers to recall the following distinction: There are two kinds of marriage. The first results from the whims of two persons acting as private individuals; the second is a systematically organized affair which forms part of a series of contractual obligations between two social groups. (Leach 1961: 57)

When mentioning ‘affinity’, this book will refer to the second kind of marriage system. Thus, the term is best understood as ‘marriageablity’, a concept that might have been well known in past periods of European history, but is hardly intelligible to modern Western individuals. Affinal relationships are collective ones that allow marriage ties, whereas, for whatsoever reason, consanguineal relationships categorically rule these out. Rather surprisingly, Chapter 4, which contains my second formal analysis examining the Omaha terminology, also concludes that this system is an affinal order of the symmetric kind, again proscribing continued ties to imply a multiplicity of different marriage alliances. The study depends upon the archival work and the publications of Barnes (e.g. Barnes 1984), though only I can be held responsible for the results. It also supports Barnes’ rejection of any general ‘Omaha type’ of worldwide distribution and is only concerned with the empirical Prairie

Introduction • 5

people known as the Omaha. Their particular system of skewing had – mistakenly I suggest – persuaded many other authors to conceive the terminology as resulting from the impact of given patrilineal descent groups. ‘Skewing’ in the Omaha case means that a good number, but not all, relationship terms contain equations that ignore generational boundaries. For example, I refer to my wife’s sister using the same term I apply to my wife’s brother’s daughter. The third formal analytical exercise of this kind in Chapter 5 deals with the terminological data I have collected among the Kharia in Highland Middle India (HMI). It also arrives at the same conclusion. Again, symmetric affinal exchanges must not be repeated in subsequent generations. Accordingly, by three very obviously different terminological constructions, Seneca, Omaha, and Kharia systems respectively imply the same imperative, i.e. the general spreading of intensive social relationships through the constant variations of ties involving symmetric marriage. As in the earlier analyses, I interrelate all Kharia relationship terms within a single matrix to obtain this concentrated ‘message’. To my knowledge, no other HMI terminology has so far been unified in this manner.1 However, the central point in this chapter proposes that all HMI systems contain the same semantic pattern irrespective of the particular tongue belonging to any one of the three inclusive language families of the culture area, i.e. Dravidian, Indo-European and Munda. I come to this conclusion on the basis of the published ethnographic material and after having personally visited some thirty different Indigenous communities in the provinces of Odisha and Jharkhand. Some minor deviations in several languages will also be mentioned. After these rather dry analyses, Chapter 6 is again concerned with the history of anthropology to debate David M. Schneider’s general critique of anthropological kinship studies. Schneider’s rejection is articulated in several contributions culminating in his statement that ‘“kinship” … is a non-subject, since it does not exist in any culture known to man’ (1972: 59). Following Schneider’s course, Janet Carsten (2000) has furthermore developed a reformed approach named ‘New Kinship’ that is equally critical of any kind of formal analysis. Both of these authors had conducted their ethnographic fieldwork among people applying what Morgan had defined as a ‘Malayan’ relationship terminology. In view of these circumstances, Chapter 6 will, in adducing data supplied by David Banks (1974), introduce the specific implications of such a system and relate these to Schneider’s and Carsten’s aversion to Morgan’s legacy. Chapter 7 will also discuss the history of anthropology when taking up Morgan’s proposals on social evolution, or the ‘progress of ­mankind,

6 • Lewis Henry Morgan’s Comparisons

from the bottom of the scale’, as he initially wrote in the Table of Contents of Ancient Society. In this venture, he placed the Indigenous Australians ‘near the bottom of the scale’ (1877: 317). In view of this particular legacy, I will try to point out the major differences between Morgan’s scheme and refined later concepts developed by, among others, Morton H. Fried and Marshall D. Sahlins that became popular during the early period of the Cold War. The endeavour was to compare the particular features of basically different institutional, technological and ideological conditions of individualistic mobile hunter-gatherers with those of settled anarchic societies, or others of a peasantry embedded in a centralized state. The substantively different degree of complexity marking these societal forms implies that the notions and content of the economy, the social order, and the worldview must vary considerably so that the respective basic ideas on kinship, marriage and gender, or those on power or religion must be understood within the given evolutionary context. These three domains of the social, the economic and the power relations will always be closely interrelated. Since the Australian aborigines are usually classified as hunter-gatherers due to the analytic priority Western scholars tend to give to the economy, Chapter 7 tries to show that their supposed anomaly, i.e. the highly complex worldview and social structure compared to their simple technology, on the contrary fits into the general pattern, once the catastrophe of the European invasion and occupation is taken into account that had led to the extreme decline of the Indigenous population and also to the aborigines’ loss of the most fertile regions of the southern continent. A reformed evolutionism is the frame of reference for Chapter 8, which will compare the ‘gentile organization’, or the anarchic public order observed in the Indigenous societies of Highland Middle India. Since I have mostly worked in Odisha and Jharkhand, I will confine myself to these eastern provinces when describing eight different empirical ‘complexes’ of usually interspersed local societies and their members’ respective notions of ‘tribal’ federations, ‘tribes’, moieties and phratries, as well as clans, subclans and local descent groups (LDGs). However, my order of presenting these notions is misleading and, in fact, the opposite one is a better reflection of reality, but Western minds find it very difficult to comprehend such ‘bottom-up’ priorities. In fact, step by step the more inclusive of these enumerated entities lose in empirical relevance. I follow Sahlins (1968a: 15) in insisting that the household is the organizational navel in these societies without rulers and ruled, followed in importance by the larger family that may or may not coincide with a subgroup operating as a halfway house to the fairly influential LDG. As I describe them, all of these empirical

Introduction • 7

phenomena display a number of variations with regard to territorial or descent-related associations or to marriage rules. Some local communities express a clear preference for cross-cousin marriage, i.e. repetitive alliances, while an equally large part of the population is horrified by this idea. In both cases, it should be understood that networking in marriage matters is a major and comprehensive social preoccupation. Given these priorities, it is clear that Chapter 9 on bridewealth and gender touches upon highly relevant interactional features of these societies; it will also, in the light of Morgan’s suggestions, include observations on private property and on the sacrificial rituals that are also discussed in Chapter 8. The latter go together with these major prestations. My data on bridewealth will depend heavily upon the comprehensive and long-term research of Roland Hardenberg (2018) among the Dongria Kond. In a narrative style, I will also recall my eyewitness experiences of confrontations and marriage negotiations among the Kuttia Kond. These will be included because I am not at all sure to what extent earlier research in the culture area, or in India as a whole, has taken notice of the basic institution represented by bridewealth as a ‘circulating fund’. The latter concept has been adopted from Jack Goody’s Africanist studies (1973), which have greatly enriched the general debate. It points to the rather different gender relations in HMI when compared to the well-known ones in mainstream India characterized by the ‘conjugal fund’ of the dowry system. These alternative social values will also be discussed in reference to institutionalized Indigenous youth dormitories allowing and encouraging social interaction of a specific kind that is confined to the domain of unmarried juniors. Considering that Morgan’s life and work had been determined by his political struggle for the cause of Native Americans, Chapter 10 recalls many of my personal observations concerning the far-reaching and multiple social pressures experienced by the HMI population. I will describe lowland land-grabbers and also public or private industrial enterprises that have devastated the hills, and will add observations on the problematic forms of modern healthcare and educational institutions, or the guerrilla movement (‘Naxalites’) of mainstream Indians taking advantage of the absence of state institutions in the area to spread their ‘revolutionary’ message through militant action. Some basic insights will also juxtapose the nature of HMI religion with that of the lowlands and will indicate the rather robust manner in which governmental institutions propagate the established rituals of castes and temples. By contrast, highlanders at times continue to gather in crowds for the sake of Maussian ‘total social facts’, such as the collective

8 • Lewis Henry Morgan’s Comparisons

secondary funeral I could repeatedly observe and then analyse (Pfeffer 2001, 2016: 301f.) in extremely productive discussions with Peter Berger (2010, 2015). Before concluding with results ‘for the record’, this book will thus present three chapters (Chapters 1, 6 and 7) on the history of anthropology, four others (Chapters 2–5) on formal analysis, and the final three (Chapters 8–10) on major empirical features observed in contemporary Indigenous societies of HMI. To put it in a nutshell, I will follow and criticize Morgan’s comparisons when trying to analyse the structures of two Native North American and all HMI relationship terminologies in order to demonstrate that their pattern emphasizes symmetric affinal exchanges between seniors along with those between juniors and that, at the same time, it implies intense social contacts to connect multiple rather than only two ‘sides’. This pattern may be viewed as the abstraction of basic social values that exclude ‘top-down’ and stand for ‘bottom-up’ relationships. It is observed in the many populous Indigenous societies of Highland Middle India. These contemporaries are free from ‘the probability that certain specific commands will be obeyed by a given group of persons’.2 Large or small, they exist without rulers and ruled. In the tradition of Morgan, political commitment for endangered minorities is taken up with the same intensity as has been applied in the other discussions on formal and empirical features of their societies, i.e. the treasures representing important markers of creativity in the evolution of anthropos.

Notes 1. Another kind of such a systematic unification of all relationship terms in a language has been presented by the brilliant Jesuit missionary John Deeney (1975: 128–29) for the terminology of the Ho. 2. ‘[D]ie Chance … für spezifische … Befehle in einer angebbaren Gruppe von Menschen Gehorsam zu finden’ (Weber 1956: 157).

1 The Life and Work of Lewis Henry Morgan (1818–81)

Impressions When I was a student in the 1960s, our beginner courses mentioned Louis Henry Morgan as the founder of anthropology, but regarding the content of his writings, our German teachers had little more than mockery. His books, we were told, had spread absurd theories on periods of social evolution and on kinship systems, two topics that have, without further qualification, always been regarded as eccentric preoccupations to anthropologists of the German1 tradition and worth little more than a condescending smile. Today, such a general attitude has remained unchanged. Fifty years ago, we also learnt about Morgan’s sin of having been posthumously praised by Marx and Engels, but by now this particular lapse may no longer cause the resentment it did at that time. As expected, we followed our elders’ authority. Later, when gradually moving beyond the German ethnological haze that, over many decades, had been known as ‘cultural history’,2 it came as a surprise to notice how, without much common ground in terms of theory, prominent thinkers such as Claude Lévi-Strauss (1949), George Peter Murdock (1949) and Meyer Fortes (1969) had located their own scholarly endeavours in Morgan’s tradition. For the centenary of Systems of Consanguinity and Affinity of the Human Family (Morgan 1871) mutually opposed but equally outstanding specialists composed a joint Festschrift (Reining 1972) for Morgan. The contribution of one of these luminaries, David M. Schneider, had a sustained impact on the future by announcing that: ‘“Kinship” … is a non-subject, since it does not exist in any culture known to man’ (Schneider 1972: 59). Morgan had been the ‘inventor’ of an anthropological subdiscipline that later spread under the label of ‘kinship’, as Trautmann (1987) elaborated in his equally meticulous and exciting account. About a century after Morgan, Schneider had spent more than thirty years of

10 • Lewis Henry Morgan’s Comparisons

his impressive academic career examining his nonsubject. To this day, a number of international kinship experts, all inspired by Morgan’s open mind, meet at conferences, of which the published results (see e.g. Trautmann and Whiteley 2012) are taken, at least in some circles, as a guide to continue on his path. It goes without saying that the following chapter is equally intended to recall Morgan’s genius.

Morgan The fascinating aspect of Morgan is his ability to combine a personal commitment marked by empathy, industry and clarity with a mind for purely formal analyses of semantic patterns within a general sociocultural domain. Very personally and on a long-term basis, he was engaged in fighting for his fellow humans and their equal rights. However, this did not in the least prevent him from intellectual feats such as the worldwide comparison of certain sociocultural features with those of other such general patterns. Never did he ignore contemporary problems of the people he described. At the same time, neither his ethnography nor his social theory could in any way be labelled as ‘problem-oriented’. The political crises of his time were not altogether excluded from his texts, and yet these treatises are, when intended for scholarly readers, articulated in an unemotional or academic language. As demanded by the legal sciences or his primary qualification, Morgan took note of day-to-day behaviour, but gave weight to it only when it could and should be subsumed under some general pattern of social order. For him, observed action was of general interest only as an aspect of a formal framework. When issues of the discipline were being articulated, personal matters remained in the domain of the personal. The overall aim of the research effort was to gradually become familiar with an unknown sociocultural field rather than find facile explanations by reference to individual action. Until his basic shift in 1865, Morgan’s approach was that of an ‘understanding’ (verstehende) rather than an ‘explaining’ kind of anthropology. Thereafter, when proposing certain ‘causes’ of given institutions or patterns, his theoretical refinement suffered, though he was praised by the scholars of his time. In what follows, I will try to recover the twists and turns in his efforts and eventually defend his lasting insights. Whenever he was able to recognize the constructed character of phenomena under study within more general patterns, his discoveries have mostly lasted the test of time. It goes without saying that they were far from obvious. To this day, they remain incomprehensible for many anthropologists.

The Life and Work of Lewis Henry Morgan • 11

Morgan was, of course, fully aware of the sharp differences between Amerindian social formalities and those of the bourgeois establishment in his hometown Rochester, but as an ethnographer, his approach was lacking any educational bent and did not interfere in the regular activities of the people he was trying to understand. At the same time, he was literarily fighting ‘Satan’ in the minds of those Euro-American fellow citizens, whose performances he was bound to witness directly on his extended excursions into the domains of politics, business and economic development. The epoch after the American Revolution had it that the leading lights in these sectors were equally violent and corrupt. Through long-term measures that were always executed pragmatically, Morgan was actively defending the right of Native Americans to be culturally different. When speaking of his legacy, I think of his civil commitment to defend the direct victims of imperial expansion and at the same time of his careful field research in an age of armchair scholarship, covering many material and immaterial details of Amerindian cultures. Last but not least, his theories on the sociocultural potential of anthropos have been substantial contributions to our knowledge on societal order. Morgan spent a restless life full of frictions that affect his political convictions, his inclination towards ethnographic precision and detail, and his ambitious generalizations on human organization and social evolution. Because of these rather different and at times contradictory commitments, his scholarship has survived the rigours of time. Over decades of untiring efforts, he was able to combine personal empathy with exact observations and purely formal analysis.

Biography and Sources The life and work of Morgan may be subdivided into four major phases: • After 1837, as a result of the major and lasting economic crisis in his country, the young Morgan did not have many clients. The cases he was entrusted with did not really excite him. Much of his time was devoted to the politics of his state and to defending the rights of Iroquois Amerindians in his neighbourhood. • From the year before his marriage (1851), all of his efforts were directed towards furthering his professional career. Within a decade, he was able to acquire a considerable fortune. • Being rich was reflected neither in the Puritan conduct of his private life nor in his social ideals. From 1857 onwards, his restless mind was once again preoccupied with Amerindian research. After 1861,

12 • Lewis Henry Morgan’s Comparisons

his regular profession became a secondary matter. With breathtaking intensity, he devoted himself to field excursions and international research campaigns, as well as to the ordering and writing up of the results of these. After a decade of uninterrupted work and the investment of the (then considerable) sum of at least $25,000, his monumental Systems of Consanguinity and Affinity of the Human Family was finally published in 1871. • Though Morgan continued as an activist in defence of Amerindian rights, he also published Ancient Society (1877) and lesser known works of a high quality. In 1881, for example, or more than a century before ‘house societies’ (Carsten and Hugh-Jones 1995) become a general topic of sociocultural anthropology, he published a book on Houses and House-Life of the American Aborigines, a work that stood the test of time. My information on Morgan’s life and family and on the details of his career have been obtained from Leslie H. White’s foreword to The Indian Journals (Morgan 1859), Carl Resek’s biography (1960) and Peter M. Whiteley’s article of 2004, but by far the most important source is Trautmann’s analytically ambitious and historically precise account (1987). Referring to every single one of my adoptions from this latter work would make this volume unreadable. Resek’s account gives a good impression of the general and frequently violent westward drive after the revolutionary impetus in the United States had captured the minds of the many pioneers. This campaign greatly influenced Morgan’s life and work. Whiteley’s essay is a response to current criticism within the discipline, an inclination that has ever so often mistakenly accused elders of complicity in imperialist misdeeds. Whiteley points out obvious gaps in the historical knowledge of these critics by recalling details of Morgan’s civil rights campaigns. As the founding father of anthropology, Morgan was especially engaged in promoting the rights of women. He left his entire fortune to the University of Rochester on condition that it was spent for the promotion of female education.3 After the sister of his Seneca Iroquois consultant Ely Parker had been of great help in his ethnographic work, he financed her college education (Trautmann 1987: 44, 61–63).4

Commitment and Ethnography In 1818, Morgan was born into a settler’s family living in an Iroquois neighbourhood on their old territory. He grew up where, in the previ-

The Life and Work of Lewis Henry Morgan • 13

ous century, these Native Americans had remained allies of the British, but were free from foreign domination and where they were subsequently subjected to the expansionism and greed of the U.S. revolutionaries. The joint appellation Iroquois stands for the five nations who had settled on the five finger-like lakes and were known as the Mohawk, Onondaga, Seneca, Oneida and Cayuga, while the Tuscarora at a later date had joined this so-called league. The army of the new United States of America broke down all Iroquois resistance, forcing these people into reservations, though some of the latter were again dissolved even before the end of the eighteenth century. The land was then distributed among army veterans. After several decades, all Indigenous peoples of the woodlands were ordered to march to the treeless so-called ‘Indian Territory’ to the west of the Mississippi, where they were made to survive on farming and cattlekeeping, even though ploughs and dairies were absolutely alien to their former existence. If they survived the march, they were doomed all the same. But the Iroquois resisted. The U.S. policy towards the Indigenous population was, in short, characterized by two procedures. Initially threats and bribes made a majority of the respective chiefs give up their nation’s inherited lands to the state, even though majority decisions made no sense in Amerindian constitutions, which do not give such territorial sovereignty to chiefs. After the passage of only a few years, the state authorities broke the old contracts and substituted them with new ones of equal validity, until finally the other side of such ‘agreements’ was no more. In 1830, the notorious President Andrew Jackson signed the so-called Removal Acts that concerned Amerindian resettlement, including that of the Iroquois. In their own territory, real estate speculators of the Ogden Land Company hoped to gain massive profits from investments in the promising region between New York and the Great Lakes. The Iroquois resisted and so did their Euro-American neighbours, since Morgan had cleverly and inconspicuously, but without any concessions, won over these settlers for the Indigenous cause. He was the elected leader of the farmers and was simultaneously the advisor of the two Sachem, or leaders, of the Seneca. These dignitaries utilized the translation services of Ely S. Parker, a young man who his nation had sent to college in order to strengthen the resistance movement. Through unexpected circumstances, he and Morgan became friends. The Seneca was able to convince his family to support the ethnographic work of the urban lawyer, who was thereby able to bring out the first major anthropological monograph entitled League of the Ho-De-No-Sau-Nee, or Iroquois in 1851 (1901). On its ‘dedication page’, Morgan explicitly referred to

14 • Lewis Henry Morgan’s Comparisons

Parker, introducing the book as ‘the fruit of our joint researches’.5 To this day, it is quoted as an important piece of empirical research, even if Morgan, over the course of time, discovered some incorrect passages and offered corrections and improvements in later publications. In the subsequent decades, the young Parker was able to fill several high positions. He was sponsored by his college mate Ulysses S. Grant, the commander of the victorious Unionist Army and the eighteenth President of the United States. Much earlier, in 1846, Parker was on Morgan’s side when the latter addressed a note of protest directly to the President and the Senate, but by a very narrow margin, Congress refused to repeal the Removal Acts. However, the Senate did (better late than never) impose a Federal Commissioner to inquire the opinions of the affected Iroquois. Following this, the Ogden Land Company sent in hired thugs, who drove away about 200 Iroquois and killed several others. Because of these interventions, the scandal received country-wide press coverage, leading to a revision of the Removal Acts, so that at least some Iroquois could remain and survive. For years, Morgan spent considerable time and money on canvassing for the defence of the Iroquois. He delivered countless speeches and wrote petitions and articles for popular magazines, the predecessors of our blogs. Through such public activism, he was able to win over the settlers who switched to the side of the earlier despised ‘redskins’. I will only quote two of Morgan’s appeals, clothed as questions, to demonstrate his commitment: ‘ Is this a heathen land that such a sacrifice should be permitted? Have justice and humanity fled? (Emphasis in original)6

Rather than applying the term ‘heathen’ to the ‘redskins’, as was done by others, Morgan was pointing to the real estate sharks and their politicians. After his detailed description of the League, Morgan added a chapter (1901 [1851] II: 108–26) with recommendations on practical policies in relation to the Amerindian population. According to Whiteley (2004: 506): ‘He argued strongly against the national climate for removal.’ The careful account of the Iroquois culture with numerous ethnographic details was completed by this outspoken and constructive political engagement. In general, Morgan’s ‘whole scientific endeavour, in his own conception, had a policy bearing’ (Trautmann 1987: 31). Throughout Morgan’s career, he was engaged as an untiring and yet formally restrained politician and businessman, though primarily as a conscientious scholar. He was definitively assertive and yet his conduct was always disciplined. His puritan sobriety7 and honesty was

The Life and Work of Lewis Henry Morgan • 15

all too obvious and went so far as to make him, a regular participant of Presbyterian Sunday services, omit the second part of the Christian confession of faith, the belief in Christ, because of serious doubts and face the ritual consequences.8

Social Theory It has long been established that Morgan’s major works contain some basically mistaken approaches and untenable conclusions. Endowed with an open mind (as mentioned above), he would have been the first to accept justified criticism. Both his personal modesty and his scholarly consistency are demonstrated in the drama surrounding his most important publication. The manuscript of his voluminous Systems was practically completed by January 1865, but the book only came out six years later and Trautmann’s study offers the precise reasons for what must have been a mind-shattering delay. Trautmann’s evidence includes many references to Morgan’s correspondence, his library acquisitions and his oral presentations, as well as the opinions of other major actors, but I will confine my sketch of event here to the core argument. On the one hand, the eminent physicist Joseph Henry, editor of the renowned ‘Contributions to Knowledge’ series, seems to have delayed the process of acceptance by installing a first and, after the positive reaction somewhat surprisingly, second publication commission that was made to take over some responsibility for the most expensive work in the long history of the publications representing the famous Smithsonian Institution. On the other hand, a change in the chronology of homo sapiens spectacularly altered the preconditions of all historically oriented disciplines, especially those of ethnology. Until Darwin’s work On the Origin of Species (1859), both humanities and sciences followed the Bible when determining the age of anthropos. For philology, the crucial event was, of course, the Babylonian confusion. Different computations assigned this occurrence to a date that was only 6,000–7,000 years before the present. However, almost at the same time as Darwin’s publication, stone-age tools were found in the caves of Brixham.9 The result of this discovery was that practically overnight, the span of human history was multiplied. The renowned and basically cautious geologist Charles Lyell supported such evidence in 1863 and was able to rapidly sell his book in the scholarly community of the United States. As a consequence, the new ‘long’ chronology removed the earlier valid framework of historical research to create the discipline of prehistory.

16 • Lewis Henry Morgan’s Comparisons

In the first draft, Morgan’s Systems had accepted the paradigm of historical linguistics that quasi-genealogically pursued various branches and twigs of languages back to an allegedly unified origin. Researchers at Oxford had not only compared the vocabulary but also the semantic patterns of grammars to reconstruct language families and their nexuses due to the common or otherwise deviating structures. In the same vein, Morgan had discovered another instrument of research that had escaped the linguists. By comparing the semantic patterns of relationship terminologies, he was able to conceive their significance and generalize in order to detect a limited number of such types on a worldwide basis. In the same way as the diffusionist approach of historical linguistics, Morgan interpreted such comparison as formal evidence of his two major hypotheses, namely that of the historical unity of all Indigenous Americans and that of their Asian origin. For Morgan, Amerindian languages of different families, in spite of their diversity, ordered their terminologies in more or less the same manner and were equally deviant when compared to the languages of Europe and the Middle East. After corresponding with and finally meeting the well-informed missionary H.W. Scudder in 1858/59, Morgan further believed that the same Amerindian pattern was behind the structure of the Dravidian languages of South India. For him, this was sufficient proof that the first inhabitants of America had originally entered the continent on a route starting in Asia. Morgan believed that the relationship terminologies of the IndoEuropean and Semitic languages formed ‘natural systems’. He named the strikingly different pattern in Native North America, and also that of the Dravidian languages, ‘artificial systems’, since they appeared to him as arbitrary constructions. Because of this arbitrariness, the semantic structure of relationship terminologies as a new instrument of research, seemed to be highly suitable for worldwide comparisons. As a conspicuous cultural intervention, such artificiality was an obvious marker.10 Within the old or ‘short’ human chronology, its diffusion was to indicate the Asian origin of the Indigenous Americans. But then the project was foiled by the time revolution. The ‘long’ chronology obviously pointed out the absurdity of such diffusionist theories. Within the now indefinitely longer period of human history, the idea of such a particular ‘route’ of a terminology could hardly be supported by valid evidence. Qualified as an anthropologist, historian, Indologist and linguist, Trautmann11 has minutely elaborated the causal connection between the newly discovered dimension of time and the collapse of the migration thesis as presented in the first completed manuscript of Systems in 1865. Morgan’s life and

The Life and Work of Lewis Henry Morgan • 17

work is quite suitable for such detailed research, since his procedures were very systematic, if not at times pedantic, just as he had the habit of noting down his experiences and decisions in political, business and scholarly matters, and of keeping an archive that he left to the University of Rochester. From the side of the publishers, a meticulous account of the drama of Systems is preserved in the archive of the renowned Smithsonian Institution. The change in the chronology of anthropos meant that the content of Morgan’s manuscript, in the form in which it was handed over at the end of January 1865, had become untenable. The benevolent referees of the work and its somewhat less benevolent editor did not, in light of the new evolutionist trend, have any intention of becoming objects of ridicule in the academic world. Today, with the dominant discourse changing ever so often, this position should be easily understood. Newly discovered facts had, at least initially, led to new forms of interpretation known as evolutionism. Morgan’s theoretical compass had been removed, leaving the option for him to dispose of his countless and detailed tables on more than 200 pages, along with his analytical findings resulting from many years of intensive work. The alternative was to revamp his theory in the spirit of the new trend and, at the same time, retain the ethnographic body of some 500 king-sized pages whenever its findings were not obviously contradictory. For Morgan, such a modification of his most valuable research results is unlikely to have been a lighthearted project. By adding a chapter of forty-three pages at the end of the weighty tome, he tried to implement his change of paradigm. The new version added nothing less than ‘general results’ (Morgan 1871: 467–510) on family forms and appropriate terminologies in different periods of human evolution. Though not entirely resulting from free choice, this innovation transformed Morgan into the most famous evolutionist of the discipline. For about a year, Morgan had tinkered with the manuscript. The artificial systems were now, unfortunately, explained as their very opposite or as resulting from human nature, and were relegated to an early stage of anthropos. Perhaps as a compromise, the new name had the neutral label of ‘classificatory system’. And again, what had earlier been the natural systems became the very opposite. They were explained as an achievement of the kind of monogamy that marked humanity’s final period of civilization. Along with this revision, they were rechristened as ‘descriptive systems’. In its new version, the book offers several commonsensical explanations together with the remaining formal comparisons. The different

18 • Lewis Henry Morgan’s Comparisons

terminological phenomena are no longer seen as an outcome of contingency, but rather as plausible results of alleged forms of behaviour, for example, as results of the so-called Punaluan marriage.12 The content of such causal arguments seems to delight the educated world of Morgan’s time, though not later scholars. To this day, similarly plausible explanations of cultural phenomena remain popular, even if the content of the arguments is bound to change. Patterns of social ordering continue to be conceived as the outcome of behavioural impact and not vice versa, so long as such behaviour does not (yet) appear to be as far-fetched as Morgan’s causal arguments on marriage practices became in the due course of time. By adopting the new trend, Morgan was celebrated for about two decades and ridiculed thereafter in most parts of the world. Without his paradigmatic shift, the reactions might have been the opposite, though the very idea is quite improbable, since the first version of Systems is unlikely to have found a publisher with a comparable reputation.

Criticism In the long run, only Marxists have given unqualified and lasting support to Morgan’s theories. From the 1920s onwards in Europe, it was left to outsiders to choose an evolutionist path, while in the United States, the followers of Boas had little difficulty in removing the almost comical excesses of nineteenth-century epigones from standard anthropology courses. A major impact in the process of Morgan’s personal ‘desecration’ was Alfred L. Kroeber’s widely read and extremely influential article on the ‘classificatory systems of relationship’ (1909), though this neither mentions Systems nor its author directly, but polemicizes against anonymous opinions. Because Kroeber does not mention his opponents, he cannot be accused of supplying wrong or misleading evidence, while at the same time, most anthropologists will immediately recognize the position and the author under attack. Initially, Kroeber’s paper created the wrong impression that the unmentioned inventors of the so-called ‘classificatory system of relationships’ defined the latter by the same criteria as his own simplistic introduction did, or by alleging that classificatory terminologies were those through which ‘primitive peoples’, but not Indo-European languages, merged a plurality of kinship relationships within a single term. For example, father (in these other languages) might be merged with father’s brother. Kroeber’s subsequent verdict is far from surprising: ‘Nothing can be more fallacious than this common view’ (1909:

The Life and Work of Lewis Henry Morgan • 19

77). ‘English’, Kroeber continued to suggest, ‘must be not less but more classificatory than the languages of all primitive peoples’ (1909: 78), because altogether, English contains less kinship terms than many other languages and, accordingly, more mergers of the above-mentioned sort. In English, the term ‘uncle’, for example, may refer to father’s brother or to mother’s brother, to husband of father’s sister and many more relationships. Accordingly, Kroeber argued that ‘not only primitive people classify … or fail to distinguish relationships’ (1909: 77), but in fact all societies classify their kinship relationships. In many ways, Kroeber’s article is far-sighted, but his polemics against Morgan’s magnum opus, which he does not cite, are absolutely unfounded. With regard to the English terminology, Systems is very precise: The partial classification of kindred which it contains is in harmony with the principles of the descriptive form, and arises from it legitimately to the extent to which it is carried … it is founded upon conceptions entirely dissimilar from those which govern in the classificatory form. (1871: 12)

As Kroeber was able to recognize forty years later, Morgan had, of course, understood that European languages also classify relationships. At the same time, he elaborated on how English initially classifies individualistic primary relationships as opposed to general or inclusive secondary relationships. The latter certainly do not, like the former, designate specific individuals. Amerindian or Dravidian languages only designate great classes and not primary or individual relationships. This is the critical difference compared to the English terminology. Did Kroeber perhaps have the impression that nobody would bother to read the original definition in Morgan’s monumental oeuvre, since at that time, once again, the trend had changed? It was no longer academically profitable to waste time on the by then obsolete evolutionism, because eclecticism had become the new orthodoxy. To this day, the criteria of the classificatory system of relationships are only very rarely discussed in an appropriate manner. What exactly did Morgan mean by this construct? Misunderstandings, like those spread by Kroeber, seem to linger on. They become problematic when proposed by modern scholars who are Morgan’s friends. Apparently under the influence of Lounsbury’s formal analysis (Lounsbury 1964b), Trautmann follows the interpretations by stating that the Iroquois’ relationship terminology shares the same-sex-sibling merging rule (2012: 34). This regularity would, for example, merge the positions of father and father’s brother and those of other same-sex-siblings to order the system. Several of Trautmann’s works (e.g. 1987: 161–62) emphasize

20 • Lewis Henry Morgan’s Comparisons

this rule as the criterion of the classificatory system. For anyone following Lounsbury’s argument, the different forms of merging would be the key to distinguishing the classificatory from the descriptive. Does this reflect Morgan’s ideas? I depart from the views of Lounsbury, Trautmann and most others13 by setting different priorities. The above-mentioned merging rule appears to me to be a consequence rather than a determining factor; it is a secondary phenomenon. Morgan’s rather extended thoughts on an entirely alien concept of what the terms in question order and how they do so is of primary importance. The mode of classifying relationships is very different, but so is the object of classification. While the technical side of Lounsbury’s formal analysis (Lounsbury 1964b) is beyond reproach and most readers are bound to be impressed by his parsimonious exercise, he was concerned with what he thought was the terminology of Iroquois kinship and nothing else. Explicitly, his formal results are meant to provide an insight into the question of ‘… who gets called what in such a system of kinship terminology and nothing else’ (Lounsbury 1964a: 352, emphasis in original). But Morgan, before turning to evolutionism, was eager to understand the respective meaning, or ‘plan’, conveyed by the specific arrangement of terms: When spread out in detail and examined, every scheme of consanguinity and affinity will be found to rest upon definite ideas, and to be framed, so far as it contains any plan, with reference to particular ends. (Morgan 1871: 11)

In Systems, Morgan explicitly states that ‘the blood relationships, to which specific terms have been assigned under the system of the Aryan family, are few in number’ (1871: 11). His eight primary terms in this system are Mother and Father, Sister and Brother, Daughter and Son, and Wife and Husband. ‘All other terms are secondary’ (1871: 12). Morgan elaborated on how this observation may initially appear to be trivial because, in both his first and also his second manuscript, this system of the Indo-European language family is said to be ‘originating in necessity’ and thus it ‘is a domestic institution, which serves to organize a family by the bond of consanguinity’ (1871: 11). In other words, only the descriptive systems14 and not the classificatory ones represent a domestic institution. The latter kind of terminology is ‘radically distinct’ (1871: 11–12): [It reduces] the consanguinei to great classes by a series of apparently arbitrary generalizations, [and] applies the same terms to all the members of the same class. It thus confounds relationships, which, under the descriptive systems, are distinct, and enlarges signification both of the

The Life and Work of Lewis Henry Morgan • 21

primary and the secondary terms beyond their seemingly appropriate sense. (Morgan 1871: 12)

In my view, this passage points to the decisive criterion. Only the descriptive system subdivides the terms into those of primary and those of secondary status; the classificatory system does not. The latter contains a uniform signification of all terms. Neither Lounsbury nor Trautmann pays attention to this point, but I consider it to be Morgan’s central discovery. It proposes that the European ideas of kinship are very different from the Amerindian conceptions of ‘kinship’. Thus, I suggest that the different merging rules are insufficient, and may even be misleading,15 as an explanation of this difference. The fact that the classificatory system ‘applies the same terms to all members of the same class’ has a specific consequence. It implies that this other kind of classification is not concerned with the issue of whether certain individuals have procreated certain other individuals, whereas this issue is the central criterion of European notions of kinship. Such an Amerindian avoidance of the tracing processes of individual procreation and birth is the reason why the study of ‘kinship’ has become an equally legitimate and relevant exercise of sociocultural research. By distinguishing the criteria of kinship from those I call ‘kinship’, Morgan provided the discipline with the instruments to understand the structures of nonWestern societies. The other major observation in Systems shows that the semantic pattern of a descriptive terminology permanently isolates a single lineal line from its collateral lines, since the latter disperse sideways to be eventually lost. In the classificatory system, however, collateral lines run next to the lineal line to eventually merge with it and thus prevent ‘a dispersion of blood’ (Morgan 1871: 13). Speaking casually for the sake of a more general understanding, this observation might be translated to mean that the collateral lines of European kinship ‘are lost’ in the course of a few generations. Thereafter, only direct descendants of a person are recognized as his or her kin (if at all). In the classificatory system, on the other hand, no lines of ‘kinship’ are ever ‘lost’. The relevance of this observation goes far beyond the issue of a terminological merging of certain relationships such as same-sex siblings. Accordingly, I fail to understand why these extremely important thoughts of Morgan have not received adequate attention. For many generations, teachers seem to have been instructing anthropology students on the classificatory system only in the Kroeberian style, which leaves much to be desired. Morgan’s criteria go far beyond the question of merging rules.

22 • Lewis Henry Morgan’s Comparisons

A single term of a descriptive vocabulary cannot be compared with a single term of a classificatory system. Morgan emphasizes that the eight English primary terms stand for concrete relationships (1871: 13). Ego, the speaker, refers to a concrete father, a concrete mother or a concrete husband. Beyond the circle of these eight, Ego must add more to the single term in order to identify an aunt, a cousin or a sister-in-law as a concrete individual. In an altogether different style, the classificatory system is not concerned with a ‘domestic institution’ and does not serve ‘to organize a family by the bond of consanguinity’ (1871: 11). Therefore, it must always add more to the single term in the event that a concrete individual is to be identified. According to Morgan’s distinction, as mentioned above, Indo-European and Semitic languages designate individuals, whereas Amerindian and ‘Dravidian’16 systems holistically order social categories. The difference between the two kinds of terminological systems amounts to a view of different kinds of social order, though this is only rarely mentioned in anthropology textbooks. Probably, the later tendency to ignore statements that for Morgan must have been of utmost importance arose after the structural-functionalist revolution had gained momentum in the 1930s. Its leader, Alfred R. Radcliffe-Brown, disqualified Morgan’s work for its ‘conjectural history’ (1941: 1). If this point was correct, as far as the speculations on evolution were concerned, it did not touch the structural insights elaborated in Systems. Yet another – and today obsolete – trend was to nullify Morgan’s eminent innovation. Another very important point of criticism has been raised by Trautmann (1987: 237–40). It rejects Morgan’s two main hypotheses of Systems, i.e. that of the basic unity of Amerindian terminologies and that of a rather similar system assigned to speakers of Dravidian languages as mentioned above. The objection is based upon empirical studies of many Indigenous North American languages that do not – like Seneca Iroquois – equate father and father’s brother, but merge father and stepfather. Also, a Tamil man in South India refers to the daughter of his male cross-cousin as his niece and not, like a male Seneca Iroquois, as his daughter. Accordingly, the type of merging is distinct in different Amerindian languages and between all of the latter compared to Tamil or Telugu. Trautmann’s rejection is, in its technical sense, absolutely correct. However, in another sense, Morgan was on the right path, even though he himself was unaware of such a different context. According to Systems, it should be concluded that the descriptive vocabulary covers primary and also, after certain additions, secondary relationships between Ego and individual kinspeople, while the classificatory system covers relationships of a rather different content.

The Life and Work of Lewis Henry Morgan • 23

It certainly does not, like the descriptive system, convey who, as an individual, is related to another individual in the sense of the two being descendants of a common ancestor.17 From the given pattern of a classificatory system, I instead conclude that it points out who is marriageable for Ego and who is not. Marriage, I may add, is a multivalent concept as well as the primary field of power games in societies without rulers. This essential difference in the general function of the terminology has received insufficient attention in most discussions of ‘kinship’ experts. As early as 1953, Dumont showed that ‘Dravidian’ terminologies invariably prescribe alliance. Containing social categories, the terminologies leave no alternative to repetition of intermarriage from one generation to the next. This positive implication constitutes the basic value of the ‘Dravidian’ system. A person is always married to another person who has been categorized as the daughter or son of two persons belonging to the categories of ‘mother’s brother’ and ‘father’s sister’, since these two categories, and only these, simultaneously include parents-in-law. In contrast to the prescription of these ‘Dravidian’ categories, the terminologies of the Amerindian languages studied by Morgan exclude a repetition of intermarriage. Formally, a person is never married to another person if this other person has been categorized as the daughter or son of two persons belonging to the categories of ‘mother’s brother’ and ‘father’s sister’, since these two categories do not simultaneously include parents-in-law. In fact, the categories that do include parents-inlaw are distinct from those that include mother’s brother or father’s sister. This difference is essential not only for the terminologies of the Iroquois but also for those of many other Native North Americans, and was pointed out by Lounsbury (1964b) long ago. In a similar way, the terminologies of the Crow and the Omaha nations, and also others, formally exclude a repetition of intermarriage. For the Hopi, Whiteley (2016) has demonstrated this point together with the fact that this formal exclusion within the analytical frame of reference to categories does not necessarily coincide with the social rules imposed upon empirical marriages. The commonality of classificatory systems in America and India lies in the fact that they, either negatively or positively, indicate who is marriageable. So far, this common ground has not been considered. Indirectly, I propose, it does confirm Morgan’s general outlook of a uniform directionality in Native North American and South Indian systems. To be more explicit: irrespective of whether father’s brother is terminologically merged with father or with stepfather and irrespective of the incompatible fact that a South Indian niece will be a daughter for an Iroquois, all of these classificatory systems characterize ­marriageability,

24 • Lewis Henry Morgan’s Comparisons

though they either prescribe or proscribe affinal relationships. Neither a particular section nor a single term of such a system can serve as the criterion of the classificatory effect, but only the pattern as a whole. Formally, the structure determines the marriage potential. This is the significant aspect of the classificatory systems discussed so far and must be recognized as such, compared to terminologies of Indo-European or Semitic languages. It has been shown above that the content of the latter is concerned with relations between individuals that are based on birth and procreation. Prescriptive terminologies, or those of a structure containing a positive message, have been known since the time of Lévi-Strauss (1949) and have been discussed widely in the discipline, while its ignored negative counterpart, or proscriptive terminologies, is equally frequent. The former type of a classificatory system has been discovered and carefully examined in Amazonia,18 South India, Australia and many regions of Southeast Asia, whereas its negative alternative is mainly found in North America, Melanesia and among the Indigenous people of the Middle Indian highlands (see Chapter 5). Resulting from his lengthy participant observation in an Indigenous village, Roland Hardenberg (2018: 140–251) has, for example, closely examined the common terminology of different Kond units. For some of them, the affinal exclusions of the formal terminological pattern equally apply in the domain of empirical marriage rules and practices, while among other Kond units, norms and behaviour contradict the negative message of the categories. Having toured much of Middle India over several decades, I may say that across the entire culture area, perhaps half of the Indigenous people prefer immediate repetitions of empirical intermarriages, while the other half explicitly prohibit such repetitions, even though all of their different languages, be they of the Dravidian, the Munda or the Indo-European language family, basically share the same semantic pattern or the formal structure of their respective relationship terminology.

Conclusion After Morgan had completed his major contribution to anthropological knowledge in 1865, he was forced to add lengthy and far-reaching corrections to the manuscript. As Trautmann (1987) has shown, he changed the text of several passages and added a last chapter to adjust the work to the newly discovered ‘long’ chronology of anthropos and to match the results with the evolutionist trend of his time. And yet, I think that

The Life and Work of Lewis Henry Morgan • 25

Morgan did retain his epochal innovation as his major contribution in the final version of Systems. During the course of his ‘general observations upon systems of relationship’ (Morgan 1871: 10–16), he identified only the descriptive terminology, not the classificatory terminology, as a ‘domestic institution, which serves to organize a family by the bond of consanguinity’. However, in the later added corrections (1871: 470), he articulated a contradictory statement. Within the new evolutionist frame of reference, the descriptive vocabulary of the Europeans could no longer be rated as a natural order that was overshadowed by the artificiality and complexity of the multivalent classificatory system. Though Trautmann (1987: 237–40) has correctly pointed out the differences between certain equations in several Amerindian terminologies as well as their general distinctions compared to the ‘Dravidian’ pattern, I propose that all classificatory systems in America and India, through the pattern of their categories, convey either a positive or a negative message concerning the general marriage potential of the respective speaker. In this respect, Morgan’s thesis about their common character remains valid, even though the founder of the discipline never probed into the specific dimension of affinal classification. Morgan saw that ‘kinship’ was much more than certain naturally determined individual relationships. He recognized that terminological systems could represent a specific ‘plan’ and that the classificatory system was ‘artificial’ and thus a relevant object of anthropological research. Meanwhile, it has been established that over the last 10,000 years, ‘kinship’ must have determined, for most humans and for most of the time, the dominant sociocultural constitution of anthropos. It should be emphasized that this constitution was only determined by ‘kinship’ (Sahlins 1968a: 15). After 150 years, this chapter intends to acknowledge Morgan’s discovery and his essential contribution to an understanding of pre-state patterns of social organization.

Notes Part of this chapter has been delivered in German as the Frobenius Lecture (7 July 2017) in Frankfurt am Main and published in Paideuma 63: 7–28 (2017).  1. The few surviving departments in the German Democratic Republic followed the Soviet line.  2. As far as I know, absolutely nothing has remained of the general approach and its theoretical variations that have constituted the tradition of Germanlanguage ethnology for about half a century.

26 • Lewis Henry Morgan’s Comparisons

 3. He also saw to it that his widow and his mentally ill son could continue as before.  4. It may be recalled that, even fifty years later, many German universities or their most important subjects were closed to women.  5. Quoted after Whiteley (2004: 504), who must have seen the original edition with this ‘dedication page’, whereas I had only access to a reprint of the 1901 edition of ‘League of the Ho-De-No-Sau-Nee, or Iroquois’.  6. See Whiteley (2004: 505).  7. Morgan was also an active campaigner for the temperance movement.  8. According to Resek (1960: 51), Morgan was a ‘deist’.  9. Situated in the English county of Devon. 10. For Morgan’s later alterations, see Trautmann (1987: 170–71) and for Lubbock’s critique regarding these points and Morgan’s reaction, see Trautmann (1987: 256). 11. See especially Trautmann 1987: 205–30. 12. This hearsay construct was suggested to Morgan by the clergyman Joshua H. McIlvaine, who was also one of the referees on the publication committee. Deviating from his proposition in terms of details, Morgan adopted it for his final version of 1867 (Trautmann 1987: 160–69). As a ‘slight extension’ of this (never observed) custom during the age of savagery, McIlvaine mentioned the ‘promiscuous intercourse of cross cousins’, but Morgan did not take up this suggestion. As a possible reason for this rejection, Trautmann (1987: 241–45) mentions Morgan’s own cross-cousin marriage. As Sabean (1998: 81–82) has shown elsewhere, during the nineteenth century, bourgeois circles of the Western world preferred cousin marriages. Thus, it would have been difficult for Morgan to connect the state of savagery with this custom. 13. The exception is Louis Dumont (1983: 179, fn. 6), who has greatly influenced me. Only after I had noticed the discrepancies between Systems and the usual interpretations of Morgan did I leaf through Dumont’s book to recover the footnote that must have been rather important for its author. 14. Both the manuscripts of 1865 and 1967 mention this domestic institution to organize the family ‘under the system of the Aryan family’ (Morgan 1871: 11). However, the last chapter (‘General Results’), which had finally been added to the earlier version, also endows the classificatory terminology with the main qualities of a domestic institution (1871: 470). By this later addition, the main distinction between the two kinds of systems, as elaborated in the second chapter of Part I, contradicts the added evolutionist chapter V of Part III at the end of the book. 15. In Chapter 5, I will try to demonstrate how some classificatory systems lack the same-sex merging rule. 16. I have added inverted commas because some non-Dravidian languages offer such systems and some Dravidian languages differ from them. 17. The German Civil Code (§ 1589 BGB) defines ‘kinship’ in this manner. 18. See, for example, Rivière (1969); and Overing (1975).

2 Tools and Types

‘Tribe’ and ‘Tribal’ Long ago, the discipline abandoned derogatory jargon (e.g. ‘savagery’ and ‘barbarism’) as had been applied by Morgan and the other established scholars of his age. More recent disqualifications of certain technical expressions such as ‘tribe’ and ‘tribal society’ seem to convey other problems, though their content lacks any derogatory connotations. However, they are easily and frequently misunderstood as referring to substantive units or groups of humans jointly involved in activities such as organizing or migrating, fighting or observing the implications of a common calendar. This is wrong. In fact, references to peoples such as the Santal or Kond, like those to a caste for mainstream Indians but of different value, convey status categories in relation to others of a comparable kind belonging to a ‘tribal’ kind of society, in contrast to the caste system. In their own languages, Santal and Kond are ‘humans’ (horo) or ‘Indigenous’ (desia) and not the feared or despised lowlanders i.e., ‘others’. None of these ‘tribal’ names ever represents a corporate unit, even though the Indian state has ‘bureaucratized’ such units in administrative documents on substantive ‘Scheduled Tribes’, ‘Scheduled Castes’ and ‘Other Backward Classes’, individual membership of which must be certified by state officials. To avoid misunderstandings, this book will thus try to omit the word ‘tribe’ or will give it in inverted commas. I will rather characterize such a specific sociocultural concept as a given ‘anarchic society’, since the people concerned are actively engaged in a rich communal life without any form of government, administration or empirical demarcation. Below the radar of state institutions, they may or may not control a unified territory and will very often live interspersed with others of a similar category, but with somewhat different housing patterns, dress and tattoo styles, dances, ritual priorities or marriage rules. Joint

28 • Lewis Henry Morgan’s Comparisons

actions may be carried out more frequently together with these neighbours than with those of the same ‘tribe’ who inhabit more distant settlements. The concept of citizenship with rights and duties assigned to Indian nationals is alien to these millions of Indigenous highlanders. The latter avoid, as best as they can, the impact of the state and its officials, or any other external authority, while retaining their own rather specific norms and conventions of a social and ideological context without rulers and the ruled. All sanctions against deviant behaviour are executed at a local level. The women and men produce their food by growing crops and raising animals, while all relevant social contacts follow ‘kinship’ principles, organized locally between ‘humans’. ‘Nonhumans’ are known to appear as army, police or businessmen, bus drivers, development officers, or anthropologists. One has to cope with them in one way or another, but normative action based on Indigenous values concerns the own people, primarily those of the same household, the own local descent group (LDG) within the own village, and the formalized external exchange relationships based on ‘kinship’ with those of other settlements in the wider neighbourhood who are defined as agnates or affines. Many of these Indigenous HMI peoples, whether small or large in population, will be very well embedded in their religious convictions without a body of dogmatic demands like those of the Abrahamic creeds, without temples and without the iconic deities worshipped in the lowlands. Their shifting cultivation of millet or their dry or wet rice farming ignores the market forces and the formal setup of these anarchic societies follows only ‘kinship’ regulations, even if members are forced to face numerous and powerful ‘nonkin’ in their neighbourhood.

Letters Morgan was among the first ethnographers to notice that: ‘American Indians always speak to each other, when related, by the term of relationship, and never by the personal name of the individual addressed’ (1871: 132). The importance of this observation has escaped many later commentators, even ‘kinship’ specialists, though it seems to emphasize the (in view of European appellations) rather different character of ‘kinship’ terms that is equally valid for references. An Indigenous person would never say ‘Jane has gone to the market’, but rather ‘elder sister has gone to the market’, even if in innumerable cases Jane would not share the same parents with the speaker. Proper names are not normally applied.

Tools and Types • 29

What Morgan observed among the remaining representatives of Indigenous America holds true for the infinitely larger anarchic units of contemporary Highland Middle India. In the village context, individual names are applied to outsiders only, while – irrespective of his or her ‘tribe’ – each person familiar with the speaker is referred to and addressed either by a nickname or by the ‘kinship’ term, or not at all, in the event that an avoidance rule must be observed. Obviously, the terms specifying a ‘kinship’ relation cannot be translated as conveying the meaning of a kinship tie. The difference must be attended to, otherwise problem-prone translations are bound to influence almost all academic discussions on terminologies. Anthropological books on either kinship or ‘kinship’ regularly use eight one-lettered abbreviations when mentioning relatives or terms for relatives. To facilitate the reading for nonspecialists, ethnographic monographs will introduce these short forms with an initial glossary standing for mother (M), father (F), sister (Z), brother (B), daughter (D), son (S), wife (W) and husband (H) so that they must be assumed to indicate genealogical positions (‘kin-types’). Some scholars even upgrade these eight positions to the rank of ‘primary kin-types’ when comparing them to – potentially endless – ‘kin-types’ represented by amalgamations of these eight letters, such as, FZHB and SWMZD, or father’s sister’s husband’s brother and son’s wife’s mother’s sister’s daughter. All of this seems to convey meticulous if cumbersome recording, but the impression may be deceptive. Morgan had shown that ‘Aryan, Semitic and Uralian families’ have as their ‘primary terms of relationship … those for husband and wife, father and mother, brother and sister, and son and daughter’, with each containing a meaning that is ‘restricted to the primary sense in which they are here employed’ (1871: 12). On the other hand, ‘Turanian, American Indian, and Malayan families’ are ‘rejecting descriptive phrases in every instance, and reducing the consanguinei to great classes by a series of apparently arbitrary generalizations’. Their respective classificatory system: confounds relationships, which, under the descriptive systems, are distinct, applies the same terms to all the members of the same class, and enlarges signification both of the primary and the secondary terms beyond their seemingly appropriate sense. (1871: 11)

Thus, readers cannot avoid noticing how the eight basic abbreviations of contemporary anthropological convention are identical, with the eight genealogical positions employed only by the people of Europe and the Middle East as the primary ones. Highlanders in Middle India, like Morgan’s Iroquois, do not share the latter evaluation; they classify

30 • Lewis Henry Morgan’s Comparisons

differently. This implies that the basic technical tools of the subdiscipline carry the bias of specifically European notions into studies on the meaning of rather important terms of non-European languages. The problem is a delicate one and will require more extended comparisons of the way in which kinship and ‘kinship’ terms are being applied. For a European, a kinship term identifies a concrete person. Ego, the speaker, designates Alter, the person spoken of. By the kinship term, the relationship between Ego and Alter amounts to one between two individual relatives – and nothing else. On the other hand, in a classificatory system of Indigenous North America or South Asia, a term is applied to ‘all members of the same class’ (Morgan 1871: 11). Ego, the speaker, does refer to or address Alter, a concrete individual, but the application of the term is not – as in European understanding – confined to only one or else a very ­limited number of individuals known to be genealogically related to Ego. The ‘kinship’ term applied by Ego designates a category, irrespective of the number of physical persons classified by it. Charting out a list of the many daughters to whom an Indigenous Ego would regularly refer or had referred to in the past would be essentially impossible, but we do know that Morgan, on several occasions, mentioned his two daughters when using this term and not any other persons. A classificatory system offers ‘apparently arbitrary generalizations’ and ‘enlarges signification both of the primary and the secondary terms beyond their seemingly appropriate sense’ (Morgan 1871: 11). In Highland Middle India, Ego may refer to innumerable persons by ‘kinship’ terms if they are ‘of the land’ (desia) or ‘of the human kind’ (horo), and not the usually hostile or incomprehensible outsiders (diku) who have moved in from the lowlands. I argue that such a speaker does not remember his or her specific – existing or nonexistent – genealogical relations to those we would call ‘distant’ relatives or those who are not genealogically related. Ego instead designates as relatives the persons of established contact in one way or another. As the most relevant criterion of selecting a specific term, the person’s status as a consanguine or as an affine is conceived, while the externally recognizable attributes of relative generation, age and sex go without saying. In short, Alter will be evaluated as belonging to the marriageable or the nonmarriageable category of established contacts and thereafter by additional criteria. Other classificatory systems may set different priorities, but in all, Ego will indicate such a status within a publicly recognized status system that includes genealogical relationships, but does not reflect them. To ignore the difference between such systemic ‘kinship’ and genealogical kinship means impairing analysis, since its commonly used

Tools and Types • 31

Figure 2.1 Affines of two sisters become consanguines. Figure created by the author.

tools – the eight above-mentioned letters – are bound to create misunderstandings. For example, Z for sister will appear as a ‘close’ relative, while FZD will stand for father’s sister’s daughter as a more ‘distant’ relative in European genealogical reckoning. But in many classificatory systems, Z simply stands for a nonmarriageable female of Ego’s generation, whereas FZD only means that she is marriageable. In either case, the opposition is valid irrespective of the person’s genealogical distance from Ego. Both Z and FZD may be extremely ‘remote’ relatives in the European sense or even not genealogical relatives at all. The ‘apparently arbitrary generalizations’ (Morgan 1871: 11) may trace genealogical reckoning, but are founded upon given or not given marriageablity, i.e. affinity in the metalanguage of anthropology. If, for example, two unrelated men marry two genealogically related sisters, these two men become brothers. Their respective sisters will then be nonmarriageable for their other brothers, even if some of these women had been eligible for some of these men before the weddings (see Figure 2.1).

Symbols and Matrixes The distinction between kinship terms designating ‘concrete relationships’ (Morgan 1871: 13) and ‘kinship’ terms designating ‘apparently arbitrary generalizations’ (1871:11) is also crucial with regard to the common use of technical symbols in such studies. Nick Allen has pointed to this difference when opposing ‘person symbols’ on the one hand and matrixes permitting a ‘classificatory reading’ on the other (Allen 1998: 315). The former comprise a circle for a woman and triangle for a man. In any context, they should only be employed to represent individual humans. As such, they can be introduced to exemplify the application of a particular kinship or ‘kinship’ term, or several of them, by symbolizing a particular speaker (Ego) in relation to another person (Alter).

32 • Lewis Henry Morgan’s Comparisons

As an example taken from the English nomenclature, one can show Ego’s symbol, conventionally blackened, as a triangle. This would represent a male individual who designates a female individual as sister, because a line above the symbol connects it with a circle at the top. No other meaning is conveyed. If the line would connect the symbols at the bottom, the circle would mean wife. This restriction implies that a classificatory system of relationships cannot be represented by such person symbols, whereas a descriptive nomenclature assembling terms for ‘concrete relationships’, or genealogical links, can be illustrated as a simple and potentially endless addition of triangles and circles connected by lines above the symbols as siblings or by others below them as spouses. On the other hand, in the case of classificatory systems, the matrix representation is suitable for those based upon the primary opposition between consanguineal and affinal categories, such as are found in Indigenous North America, South Asia and many other parts of the world. Several other systems of ‘apparently arbitrary generalizations’ (Morgan 1871: 11) seem to exist as well, though I cannot introduce most of them or suggest an equally correct and applicable method of their display. For any formal representation of a classificatory ‘kinship’ terminology, all terms must be included. Among those systems based upon affinal opposition, the simplest is that containing only one line of consanguineal terms and one of affinal terms. For the matrix given in Figure 2.2 below, I have selected and arranged the Piaroa1 pattern. Joanna Overing Kaplan (1975: 74, 131) had introduced its practical application by person symbols and the systematic application by a box-type chart of the matrix kind. All terms of such a system must be examined because it infers logical generalizations rather than ‘concrete relations’. Thus, its classification by the sex distinction has it that the BD (brother’s daughter) of a man is equated with the ZD (sister’s daughter) of a woman and not with the BD of a female speaker. The generalization implies that same-sex sibling’s children are equated (and also other-sex sibling’s children), but not the concrete genealogical link to a female child of a sibling irrespective of the speaker’s sex. Though the relativism of the generalization (i.e. same-sex versus other-sex) is the same for a man and a woman, the technical style of our biased way to translate systemic generalizations into genealogical links may force us to chart out two different matrixes, one for a female and one for a male speaker, in order to do justice to the generalization. This is also necessary for the Piaroa system (see Figure 2.3).

Tools and Types • 33

Figure 2.2 Two-line Piaroa relationship terminology (Ego female). Figure created by the author.

Figure 2.3 Two-line Piaroa relationship terminology (Ego male). Figure created by the author.

34 • Lewis Henry Morgan’s Comparisons

Figure 2.4 Simplified two-line terminological system of the Piaroa and many others. Figure created by the author.

A first look at the two matrixes above reveals that the generation of grandparents (or G+2) and of grandchildren (or G-2) supplies only one term for a female and another one for a male category, respectively implying that no affinal relationship is articulated for these generations. Only the principles of sex and of generation are applied for classification in this external frame of the system, not those of affinal and of age distinction. The terms may be considered as being ‘undetermined’ with respect to affinity. In Ego’s generation (or G0), consanguineal categories are internally discriminated by relative age (elder/younger), but affinal ones are not. Nor is this age opposition found in the adjoining generations of G+1 (parental) and G-1 (filial). In the three medial generations, comprising the core of the system, the opposition between consanguineal categories and affinal categories dominates. Each one of the former is opposed by one of the latter. The system does not allow for variation; it formally prescribes the affinal relationship. Any person to whom Ego refers to as cha’o is married to a person Ego refers to as chimin’hu. In the same way, a person Ego calls chubu’a must be married to one who is called chiaspo by Ego. The system as a whole is classified by the simple opposition of these categories, of which one terminological line comprises consanguines and the other line comprises affines. However, there is more to this. Since, in the three medial generations, any one category in the consanguineal line is linked to only one female and only one male category of the affinal line, the relational structure formally indicates symmetric exchange. The marriageable category for one line is the nonmarriageable category for the other. Finally, the fact that the same symmetric exchange relationship is formally indicated in the parental, the speaker’s and the filial generation means that it is invariantly repeated from one generation to the next. The symmetric affinal exchange is diachronically prescribed, which is conveyed in the simplified matrix above (see Figure 2.4). Formally a

Tools and Types • 35

woman referred to as a consanguine always ‘takes’ for marriage a man known as an affine and ‘gives’ a consanguineal man for marriage to an affinal woman. This is repeated from one generation to the next as the only way to classify ‘kinship’. The most difficult problem for a European mind is perhaps the observation that such formal affinal prescription need not coincide with empirical marriage rules or behaviour. As in the Piaroa case, the terminology may prescribe diachronic affinity or the invariant repetition of the given affinal relations, while social norms and practices contradict such an idea of a continued symmetric exchange. In other ethnographic cases, the classificatory system will exclude any repetition of the same affinal relations in Ego’s lower generations, while repeated intermarriage is in conformity with legal regulations. On the other hand, the imperatives set by the structure of categories and those of legal regulations do coincide in many other Indigenous societies. Obviously, the domain of language, and thus of affinal prescription, is not as prone to innovations as is the domain of either social norms or practices. Since changes in the three fields are unlikely to occur synchronically, Rodney Needham (1967) has suggested2 an evolutionary sequence beginning with the constraints of affinal exchanges in all three spheres of categories, norms and practices, to be followed by a gradual emancipation in the practical and normative domain, which would eventually have its repercussions upon the order of the language. Similarly, Allen’s many contributions (e.g. Allen 1998) have proposed an original tetradic structure in both the formal and the behavioural sphere that has been partially dissolved over the course of human history. However, since the very concepts of terminological prescription and of inherited affinity have been highly controversial in the discipline, such evolutionary considerations have – ­unfortunately – failed to encourage later scholars to continue this research. A terminological structure, like one of grammar, is a ‘given’ classification. Several such systems are found on more than one continent, so they are unlikely to have been deliberately constructed by known persons or others assumed to have existed in history. However, European ideology has it that not just cathedrals but also social constitutions and concepts must have been ‘invented’ by thinkers or technicians who may be identified. In order to relativize such assumptions, it may be recalled that a particular North American people has maintained the idea of its ‘eternal’ social order being represented by the camp circle of its tents, even after the practical arrangements of camping did not agree with this order. The same Omaha people continue to apply a classificatory system of ‘kinship’ terms that has stunned European

36 • Lewis Henry Morgan’s Comparisons

Figure 2.5 Simplified four-line terminological system of alternating reciprocal affinal relations. Figure created by the author.

ethnographers and is yet to be explained in a convincing manner. Similar variations in the manner of classifying phenomena may be found by open minds. For the time being, I will only show that the very simple structure of a two-line terminology, as exemplified above by the Piaroa terms, may in other regions such as Australia appear as a doubled version or a system of four terminological lines. This would be only one of several more complex variations. However, I will not burden the reader with ethnographic details of the case and will only mention that such a four-line system implies symmetric affinal relations not being immediately repeated, but rather in alternating generations. Thus, two distinct lines of consanguineal categories in turn relate to two distinct affinal lines (see Figure 2.5). The introduction of such a ‘doubling’ is only meant to warn readers of further variations in the following chapters. Classificatory systems offer several different generalizations and perhaps different combinations of the generalizing principles that are yet to be discovered.

Terminology Types Not infrequently, anthropology students are made to learn so-called terminology ‘types’, of which that called Iroquois is probably known best. Other labels such as Eskimo, Hawaiian, Sudanese, Omaha and Crow are also assigned according to ethnic or regional entities and thus seem to contain some ethnographically relevant meaning. Leslie Spier (1925: 77) was probably the first to propose such types, including the one named Iroquois. He restricted his comparisons to Native North Americans and began his text with the following uncharitable and subsequently unexplained remark: ‘Lewis H. Morgan published sufficient to cover the region east of the Rockies in his Systems … some fifty years ago. But his unfortunate manner of presentation rather prejudiced reworking the data’ (Spier 1925: 71).

Tools and Types • 37

No such negative references can be found in Robert H. Lowie’s articles on relationship terminologies. Like Spier, he selected six terms (three for males and three for females)3 out of any given nomenclature to distinguish them as the ‘Generation’ type, merging F, FB and MB, from a second one named ‘Bifurcate Merging’ and containing a single term for F and FB, with another one for MB and yet more from a third type named ‘Bifurcate Collateral’, with distinct terms for F, FB, and MB. Finally, his ‘Lineal’ type merged FB and MB, but distinguishing F (Lowie 1929: 84). In Native North America, he associated the ‘varieties’ of Iroquois, Omaha and Crow with his ‘Bifurcate Merging’ type, Eskimo with the ‘Bifurcate Collateral’ type, and Salish with the ‘Lineal’ type (1929: 85). Independently and almost at the same time, Paul Kirchhoff wrote on similar types, but did not attach ethnic labels to them. His ‘Type D’ (1932: 49) may be identified as that called Iroquois by Lowie. Whereas Lowie had confined his selection of terms to G+1, the parental generation, Kirchhoff simultaneously studied and elaborated upon the distinctions of some terms chosen from G0, or Ego’s generation, comprising those for siblings and cousins. Undoubtedly, the climax of these typological exercises involving ethnic labels was reached by Murdock (1949: 223–24), who saw the Iroquois type merging all cross-cousins (FZD, FZS, MBS and MBD) and distinguishing them from parallel cousins (FBD, FBS, MZS and MZD), who were classified as siblings. Murdock’s far-reaching influence within the anthropological community arose from his comprehensive statistics assembled in the Human Relations Area Files, which, in his time, were thought by many to function as the key to all empirical research on kinship. In particular, German anthropologists were (probably for the first and only time) impressed by studies in this subdiscipline due to Murdock’s approach dealing with a massive amount of data from innumerable societies. Statistical calculations would relate his ‘types’ to other sociocultural phenomena in the political, economic and religious domains in order to arrive at general conclusions. However, I cannot see that Murdock’s generalizations involving terminology ‘types’ have been accepted as analytically valuable by anyone over a longer period of time. Whenever names of ethnic units were mentioned, the labelling may have misled readers to believe that some local, ideological or theoretical significance could be associated with the selected few terminological equations mentioned above. Such an impression is wrong. I justify my critical remarks with the observation that, irrespective of past criticism by renowned specialists (e.g. Needham 1971: 13f.), these ‘types’ continue to be taught in anthropology classes.4 I reject them for the following reasons:

38 • Lewis Henry Morgan’s Comparisons

• All of the scholars propagating certain ‘types’ of terminologies have arbitrarily selected a very limited number of terms out of many more of the respective nomenclature, but have failed to explain – or even consider – why their specific selections (or others) should deserve such an extraordinary attention. • Out of nomenclatures containing many more terms than those chosen for the particular ‘types’, these scholars have only selected a few terms that in English and other European languages are associated with the word kinship. In everyday language, this means consanguinity. None of the selected terms is associated with affinity. Without further considerations, they have thus simply disqualified the classificatory potential of affinal terms and thereby propagated the cultural bias characterizing European languages. • Morgan’s classificatory system is ‘reducing the consanguinei to great classes by a series of apparently arbitrary generalizations’, just as it ‘applies the same terms to all the members of the same class and enlarges signification both of the primary and the secondary terms beyond their seemingly appropriate sense’ (1871: 12). Accordingly, terms of such holistic structures are simply not comparable to a vocabulary in which each ‘relationship is thus made independent and distinct from every other’ (1871: 12). As Morgan has shown, the latter kind of nomenclature is a simple assembly of independent parts. Between them, these parts are not interrelated. Since these ‘terminology types’ lack a relation to reality, they should be dismissed altogether. I must suggest such a measure because readers not familiar with the subject might confuse such worldwide comparisons of a very limited number of samples selected from both descriptive (e.g. English) and classificatory nomenclatures (e.g. Iroquois) with the formal analysis of an entire classificatory terminology, e.g. all terms applied by the Iroquois.

Notes 1. During the 1960 and 1970s, Overing Kaplan worked with the Piaroa of Venezuela before they were dispersed. 2. He has exemplified his proposal by empirical cases refering to the terminologies and social norms of the Garo and the Mapuche. 3. The terms for the three female relatives have the same pattern as those for the three male relatives I introduce. 4. Due to the general importance of Google, I should mention that most entries under ‘Iroquois’ contain grave mistakes.

3 Seneca Revisited

The Basic Difference During the 150 years after Morgan’s ‘invention of kinship’ (Trautmann 1987), many scholars have devoted their undivided attention to terminologies of the kind analysed in Systems of Consanguinity and Affinity of the Human Family. Some of them discussed ‘kinship terminologies’, while others instead focused on ‘relationship terminologies’. The difference between these two approaches is rarely clarified. For the first camp, Lounsbury’s influential article1 on Seneca terminology probably offers the most explicit form of defining the object of his research: The set of KIN-TYPE designations – such as father, father’s brother, mother’s brother, father’ sister’s son, etc. – specifying the genealogical positions of one’s known kin in relation to himself, can be regarded as constituting a semantic field. (Lounsbury 1964b: 1073, emphasis in original)

Clearly the analysis is confined to designations of ‘the genealogical positions of one’s known kin in relation to himself’.2 Perhaps for technical reasons, Lounsbury excludes all other notions on the meaning of such terminologies, or else he presupposes a worldwide consensus about the nature of the phrase ‘one’s own kin’. Such an agreement would universalize Euro-American usage. In the same culturally restricted mindset, Lounsbury as well as innumerable other publications explicitly confine their analyses of kinship terms to those understood to be consanguineal kin-types.3 For these authors, only such genealogical positions seem to matter, or at least be of superior relevance, compared to those of affines, and again such a grading agrees with the common notion of the English language assigning ‘in-laws’ a minor or secondary status, or one not provided by nature, compared to ‘real’ relatives. The German Civil Code4 and

40 • Lewis Henry Morgan’s Comparisons

­ robably other Western law books explicitly distinguish in-laws by p such a logical downgrading. Following these notions conveyed by the English language, Lounsbury and many others formally delimit the object of terminological analysis. The authors of the second camp, on the other hand, offer reasons for analysing ‘relationship terminologies’. They prefer such a label: because it defines the field of relevance more broadly and does not assume that persons denoted by such terms are ‘kin’ as we understand them, nor indeed that they are ‘related’ in any genealogical sense whatever. (Barnard and Good 1984: 37)

These authors seem to suggest that European notions of kin or kinship are not held universally. It follows that analysis must take note of the semantic differences in the same way as culturally rather different notions of ‘religion’, ‘work’ or ‘death’ are recognized and examined by anthropologists. To ignore such distinctions of semantic content and conceive all other terminologies in the manner of European languages would amount to a wrong premise and would be responsible for subsequent analytical mistakes. It would be an unfounded generalization of the analysts’ cultural values. In the same vein, members belonging to the second camp do not grade the linguistic units of a terminology by culture-based criteria to exclude some terms (i.e. affinal terms) of the set from their analysis while retaining others. In fact, initially they may not decide on the consanguineal or affinal character of a term until they have studied the particular relationships constituting the terminological whole, since they conceive of such a whole as one of several available classificatory alternatives. In a further step, pioneers of research in the Guyanas (e.g. Rivière 1984: 56) have differentiated the meaning of affinity in view of the original Latin concept ad finis or ‘on the boundary’ (1984: 70). As an additional consequence of his empirical research in Amazonia, Eduardo Viveiros de Castro does not restrict the application of affinal terms to humans. Taking up an initiative by Overing Kaplan (1975: 2), he proposes three different culture zones of the globe comprising respectively: those societies in which consanguinity encompasses affinity, those in which the two principles stand in an equistatutory relation, and those in which affinity encompasses consanguinity. (Viveiros de Castro 2001: 22–23)

Seneca Revisited • 41

Preliminaries: Morgan’s Initiative For Morgan, the founder of the discipline, the arrangement of European terms, though logically imperfect, was designed to realize concrete relationships. He assigned what he called ‘purely descriptive forms’ to Erse, because Celtic languages lack terms for uncle and aunt, nephew and niece, or cousin (1871: 13). Erse only describes genealogical relationships, whereas beyond the eight primary terms, English or French maintain a partial classification that, however, follows quite dissimilar principles compared to those of a classificatory system. As to such a classificatory system, textbooks usually refer to Morgan’s work on Seneca when he states ‘my father’s brother is my father’ (1871: 13). However, things are not that simple. The more relevant – and more complicated – formal context of this equation shows that in Seneca: the several collateral lines, near and remote, are finally brought into, and merged in the lineal line, thus theoretically, if not practically, preventing a dispersion of the blood. The relationships of collaterals by this means are both appreciated and preserved. (1871: 14)

Hopefully, this quotation will indicate Morgan’s focus on a different object compared to Lounsbury’s semantic field. The latter is confined to the terms for ‘one’s own kin’, or how kin (according to the EuroAmerican understanding) are being addressed and referred to. Instead, Morgan’s analytical interest is directed towards the classificatory mode of a terminology. The Seneca mode, for example, does not point to a limited number of individuals as kin, while the English one does. At the same time, Morgan and Lounsbury share the bias that, all over the world, a secondary or lower status should be assigned to affinal relationships when compared to those based on ‘blood’ or, in modern phrasing, on common genes as the basis of kinship. In his analytical text, though not in his tables, Morgan’s book ignores all of the terms he ranks as ‘Marriage Relatives’ (1871: 169). He also lists on the tables what he understands to be consanguineal terms, such as FZDH, though, given a non-European cultural context, I might rank them as affinal ones. As to his diagrams (Plates I–XIII), Morgan altogether omits the terms he has earlier assigned to the ‘Marriage Relatives’. Clearly, they remain of secondary importance for him. As Trautmann (1987: 52) has shown in a very important observation, Morgan’s many tables enumerate relationship terms by adopting the model of the famous British legal scholar Blackstone (1893). This highly influential style initially reports terms for members of the

42 • Lewis Henry Morgan’s Comparisons

lineal line (viz. Ego’s direct forebears and descendants), followed by those referring by degree to the four nearest collateral lines. After he has completed all of these lengthy presentations, Morgan (e.g. 1871: 378–82) – in his own meticulous way – adds a few terms for ‘Marriage Relatives’ without further discussion. Even if he postulates that terminologies may consist of ‘apparently arbitrary generalizations’, he remains unconcerned about the possibility that such generalizations might include affinal terms. Thus, Morgan retains the specifically European style to separate ‘marriage relatives’ from ‘blood relatives’. For example, he defines FZD and MBS as consanguines of the second collateral line (e.g. 1871: 168), whereas these genealogical positions figure as affines in many nonEuropean terminologies. The consequences are far-reaching, as may be demonstrated by Morgan’s famous – but mistaken – observation that the Tamil system in South India is ‘substantially identical with that of Seneca Iroquois’ (1871: 385). His detailed comparison of these two terminologies (1871: 511–14) was a high point in the first version of Systems and remained unchanged in the second version. By referring to one term only, I will try to point out Morgan’s cultural bias that was later retained by Lounsbury and many others, even if, for different reasons, they rejected the equation of the Tamil and the Seneca structure. When Morgan translates a Seneca word as ‘daughter-in-law’, he introduces the respective Tamil one as ‘daughter-in-law and niece’,5 but in the lengthy Table III, where all Dravidian terms are being presented according to Morgan’s (or rather Blackstone’s) regular procedure, this twofold Tamil meaning is noted only three times (1871: 527, 531 and 563), while the very same term is only translated as ‘niece’ for fifteen other positions in Table III. Clearly, for Morgan, the meaning of ‘niece’ has priority, being considered as a consanguineal relative. In his tables, consanguines are always those understood to be kin in European languages, and affines are only equated with the insignificant number of English in-laws. In Tamil, however, a single term merges ‘niece’ and ‘daughter-inlaw’, whereas in Seneca, one term means ‘niece’ and a different term means ‘daughter-in-law’. Formally as well as empirically, a ‘niece’ is unmarriageable in Seneca conceptions, while a Tamilian ‘daughter-inlaw and niece’ is formally and empirically the only or the best relative to marry Ego’s son. Just because Morgan mostly ignores these (and corresponding) mergers by giving undue priority to the meaning of ‘niece’ for the Tamil term, he is able to equate the Seneca and the Tamil structure. Apparently he set this priority because it coincides with the value contained in European kinship conventions.

Seneca Revisited • 43

Lounsbury (1964b) was the first to point out the difference between the Seneca and the Tamil structure, and subsequently Trautmann (1981, 1987), as an Indologist, elaborated the basis of this difference. However, Lounsbury too followed a comparable procedure by leaving out affinal terms from his analysis altogether. Until today, the concept of inherited affinity, or an affinal status not arising from a single marital union between two individuals, is unthinkable for most anthropologists.

Preliminaries: The Descent Bias For Morgan, European kinship is a ‘domestic institution’ with in-laws as relatively insignificant and ephemeral appendices of consanguines. By contrast, Viveiros de Castro points to affinity as the encompassing dimension of Amazonian notions of ‘kinship’. Rhetorically, the latter scholar goes so far as to introduce ‘affinity without affines’, which is another way of emphasizing ‘potential affinity’ (2001: 24), as relations ‘on the boundary’. Against ‘our deep-seated extensionist prejudices’, Viveiros de Castro offers the following argument: rather than being a metaphorical extension, a semantic and pragmatic attenuation of matrimonial affinity, ‘figurative’ affinity is the source of both ‘literal’ affinity and the consanguinity the latter breeds. (2001: 25, emphasis in original)

In the same spirit, I suggest, the Seneca relationship system should be studied. Originally, it must have been designed within an ideological context very different from contemporary conditions, but those sociocultural systems unimpaired by the lethal consequences of European expansionism can hardly be reconstructed. However, irrespective of the comprehensive and lasting European onslaught upon the physical existence and the social order of Native Americans, I contend that the dimensions of consanguinity and affinity in the Seneca ego-centric terminology are outlined by a nonterminological framework, the sociocentric categories, as equally formal and empirical phenomena. To this day, these clans classify all Seneca, all other Iroquois and many more Amerindians by the general concept of relative affinity. Such an affinal scheme, i.e. the Seneca clan system, became obvious to me only after extended ethnographic fieldwork in Highland Middle India.. To Euro-American thinking, the rigid social norm of clan exogamy has always been difficult to understand and is generally ignored or belittled in anthropological textbooks. Morgan, meaning ‘clan’ when he wrote ‘tribe’, offers the first but not the last example of such a failure

44 • Lewis Henry Morgan’s Comparisons

to grasp the structural significance of clanship: ‘Each tribe [i.e. clan] is a great family of consanguinei … the lineal descendants of the same person’ (1871: 139). This empirically unfounded, but, to this day, rather common idea pointing to descent from a common clan ancestor6 is hardened in Ancient Society (1877), where he chooses the word ‘gens’ for ‘clan’ and adds what even to contemporary Western minds is regularly meant to ‘explain’ the existence of clans and clan exogamy: ‘As intermarriage in the gens [i.e. clan] was prohibited, it withdrew its members from the evils of consanguine marriages and thus tended to increase the vigor of the stock’ (Morgan 1877: 65). In later decades, the most prominent African studies of British social anthropology had the effect that the clan was wrongly conceived as a corporate unilineal descent group, although in the seminal work on ‘African Political Systems’, E.E. Evans-Pritchard was clearly limiting his characterization to Nuer clanship, contrasting it to ‘many African clans’ of an altogether different kind. If this scholar confined his definition to the Nuer case, many subsequent anthropological discussions simply ignored the restriction conveyed by the following passage: The Nuer clan is not an undifferentiated group of persons who recognize their common kinship, as are many African clans, but is highly segmented. The segments are genealogical structures, and we therefore refer to them as lineages and to the clan as an exogamous system of lineages which trace their descent to a common ancestor. (Evans-Pritchard 1940: 285)

Seneca clans, on the contrary, like innumerable more in African and other societies all over the world, lack pedigrees, segmentary lineages, corporateness or any other significant social construct linked to a ‘vertical’ or genealogical reckoning of descent, so they should not be confused with the basically different Nuer units. In the same vein, Claude Lévi-Strauss wrote on the subject of Native American clan names: It becomes impossible to define membership by genealogical means … On the other hand, differentiating marks in any society, though varying one from the other in content, must be formally of the same type. (1964: 12)

The ‘differentiating marks’ of Seneca clans are ‘formally of the same type’ carrying labels like bear and beaver, and hawk and heron. As the decisive point of distinction, a wolf of the animal world can only procreate with another wolf, whereas in the sociocultural cosmos of anthropos, a wolf can never marry another one. Even if a Seneca does, at times, marry a non-Seneca,7 such a spouse must never belong to the same

Seneca Revisited • 45

Figure 3.1 Affinal status of Seneca clan categories (example: Bear clan). Figure created by the author.

clan. A Seneca wolf, for example, cannot marry a Cayuga wolf or one belonging to any other wolf clan of different Native American nations. However, in the past, such a negative phrasing has been misleading and should thus be substituted by a positive observation that, astonishingly, I have failed to find8 in any publication on sociocultural anthropology. In other words, a Seneca of, for example, the bear clan must marry someone belonging to any of the seven other clans of the societal whole (see Figure 3.1). Being different sociocentric categories, these seven others represent the inherited affines of the bear category, irrespective of any actual marriage between two individual humans. All bears understand themselves to be consanguines, because they cannot intermarry. In other words, clanship defines categorical affinity by reference to all Seneca, though such a general assignment of affinity is alien to the languages and unimagined by European minds. The website of the Seneca nation9 enumerates the first four clan categories of Figure 3.1 as ‘animal clans’ being juxtaposed by a unit of the second four ‘bird clans’,10 so that one may speculate about a historical system of exogamous moieties11 in the sense of two inclusive categories defined by mutually exchanging partners for marriages,12 which in the unknown past may or may not have involved further selection refinements. Significantly, the deer people are counted among the ‘bird clans’; thus, any naturalistic explanation of these labels is uncalled for. Together, these eight clan categories represent a finite system of ordered marriageability, providing the Seneca with an all-inclusive frame of reference or a formal document of affinal relations that articulates their structural principle of ‘kinship’ within an omnipresent system.

Preliminaries: Lounsbury’s Footnote Seneca clans are sociocentric affinal categories. I contend that they do not determine the structural details of the egocentric terminological

46 • Lewis Henry Morgan’s Comparisons

c­ lassification, but rather provide and emphasize affinity as the general principle of all social structuring or as the dominant ideological feature. The essential difference between the Ego-centric classificatory design and the society-based technique of ordering is the relativity of the former and the absolute or constant relational setup of the latter. Irrespective of any empirical considerations, this distinction will always formally prevent an exact fit between the two systems. The detailed pattern of the terminology can never be identical with the general pattern of affinity contained in clanship. Lounsbury (1964b) must be credited with the central insight that, contrary to Morgan’s theory, Seneca is altogether different from the socalled ‘Dravidian’ terminological structure. He is also the first scholar to demonstrate convincingly how – according to Morgan’s data – a Seneca speaker will find terminological ‘fathers’ as well as ‘maternal uncles’ in her or his own clan and also in other clans among the ‘more distant’ relatives (1964b: 1073n). Lounsbury’s highly influential critique of the ‘classic but erroneous anthropological view’ contains no reference to the criticized authors. He has articulated his reservations in an elaborate footnote,13 where he does mention ‘the anthropological theoretical writings on the subject’ that spread a contention ‘about which so much has been written’ in a ‘classic theory’. Morgan is said to be ‘partly responsible for this error’. Under attack is the thesis that Seneca, just as any other terminology of the Iroquois type, ‘classifies kin by membership in unilineal descent groups’. The critic shows meticulously how, after his elaborate inspection of all the extensions mentioned by Morgan, Ego’s ‘more distant’ terminological ‘fathers’ and ‘maternal uncles’ are equally represented in both Seneca moieties and in any of the Seneca clans. In a reference to Morgan’s League of the Ho-De-No-Sau-Nee, or Iroquois (1901: probably 77), the false view is said to ‘be derived from a metaphoric use of “sibling” terms and from ignoring all but the closest kin-types included in his own tables of data he published in Systems’ (1964b: 1079n). Morgan14 had reported: ‘In the eyes of the Iroquois, every member of his own tribe [i.e. clan], in whatever nation, was as much his brother or his sister as if children of the same mother’ (1901: 77). For Lounsbury, such practice amounts to a metaphoric use of the words ‘brother’ and ‘sister’, since it is obviously not used in the same way as one of his narrowly defined ‘kin-type designations’. For a system (like Seneca) assigning clan membership matrilaterally, the critic would probably have indicated the same metaphoric status in cases when a Seneca speaker referred to all males of the paternal clan as ‘fathers’, i.e. clan members of all generations, as is common practice

Seneca Revisited • 47

among Native North Americans and other anarchic societies of other continents. In short, the prominent linguistic anthropologist locates a ‘metaphoric’ application as a rather different and somewhat secondary concept compared to the ‘kin-type designations’ that Lounsbury has defined as the sole object of any terminological analysis. At the same time, Lounsbury criticises the (anonymous) observation that a term like ‘father’,15 when designating distant kin-types, should apply only to those who are ‘men of one’s father’s clan in his generation’. He expresses surprise that: the essential data pertinent to the subject about which so much has been written should have been in print and available to all for nearly a century without anyone’s having taken account of the classification of any but the closest collateral kin types. (Lounsbury 1964b: 1079n)

As outlined above, Lounsbury demonstrates that in reality, such ‘more distant’ fathers are found in all clans. However, Barnes (1984: 174; 2012: 72) seems to forward a rather different view based on empirical observations of the way in which Amerindians refer to people. He arrives at these conclusions after extensive archival work on the Native American Omaha. This has allowed him to question the very idea of Amerindians remembering their respective ancestors or ‘the marriages of many collateral relatives’ for a longer period of time. His critique implies that their relationship systems are simply not meant to serve as a correct method of identifying persons who, according to European understanding, are ‘distant’ relatives. This insight should mean that the Ego-centric terminology serves for the references to the known genealogical relatives who are regularly contacted by a speaker, whereas all other acquaintances would be addressed or referred to by taking into account their sociocentric category or clan. By such a ‘shortcut’, many persons familiar to the speaker, including (but not limited to) genealogical relatives assumed to be ‘distant’ by European reckoning and belonging to any generation would become ‘fathers’ or ‘maternal uncles’ in day-to-day speech. However, in Lounsbury’s critique, the Seneca system does not classify ‘by membership in unilineal descent groups’ and I do not dispute but fully support this statement, though from an entirely different point of view. For the time being, it should be remembered that in some applications, Lounsbury rates Seneca terms (‘brother’, ‘father’) as designations of kin-types. In other applications, he provides them with the rather different status of being metaphoric conventions. Such a rating is the very opposite of that used by Viveiros de Castro, as noted above.

48 • Lewis Henry Morgan’s Comparisons

To be more precise, on the one hand, Lounsbury proposes that ‘distant’ relatives are classified by the Ego-centric terminology in a particular manner that had been ignored by previous analysts, but on the other hand, he grades all known Seneca applications of terms into (a) designations of kin-types and (b) metaphorical usages of a different and secondary value and of no concern for analysis. In empirical reality, however, the latter terms are known to be applied to numerous persons, including ‘distant’ relatives. As such, it must be concluded that Lounsbury discriminates the empirical application of terms by rating some situations (when genealogical relatives are referred to) as involving suitable terms for the analysis of kinship terms and other situations (when clan membership determines the reference) as involving metaphors and thus unsuitable expressions. Formally, however, Lounsbury sees both ‘close’ and ‘distant’ kin-types being designated by the Seneca kinship terminology. However, Barnes seems to suggest that such equations of ‘distant’ relatives are empirically out of the question because Native North Americans do not work with extended pedigrees and do not remember the marriages of many collateral relatives. In following the latter view, I contend that Seneca clanship conveys the general principle of affinal classification that also determines the classification of the Ego-centric terminology. Nevertheless, the latter is not a lower level in a logical hierarchy of which clan and moiety systems form the same kind of structure at higher or more inclusive levels. Rather, I contend that the same value-idea of alliance, or symmetric affinal exchange, simultaneously governs the different structures of moieties, clans and Ego-centric relationship terms. Accordingly, a Seneca speaker employs clan membership as a ‘shortcut’ when referring to a more distant male acquaintance as a ‘father’, or else as a ‘mother’s brother’, in cases when this distant acquaintance belongs to the paternal, or otherwise to the speaker’s clan. Any idea of a Seneca engaging in lengthy algebraic exercises to identify a distant relative according to Blackstone’s model would be plausible for a situation in which an ethnographer (Morgan?) is asking endless systematically ordered questions on relationship terms by following this model. In other or ‘uninfluenced’ situations, the idea that a Seneca speaker, lacking a genealogical memory worth mentioning, would somehow activate Blackstone’s genealogical calculations to find the proper term of addressing or referring to a very distant relative is not plausible. The mistake, in my view, has been made by the anonymous proponents of the ‘classic view’ and also in the same way by their critic

Seneca Revisited • 49

Lounsbury, since both define Seneca clans and moieties as ‘unilineal descent groups’ (Lounsbury 1964b: 1073n). However, clans are not groups and have never have been so. They are exogamous (i.e. affinal) social categories, whether or not any one of them is applied to the Seneca people only or simultaneously to ‘whatever nation’ (Morgan 1851: 77) of Native North America. The mistaken view of most Euro-American observers may have resulted from the rather small number of individuals associated with a single ‘nation’ in Native North America and also with the culturespecific (or ‘bureaucratic’) idea that a clan should be a subunit of a ‘tribe’ and not an affinal category irrespective of its particular association with a certain ‘tribe’. The demographic conditions have resulted from the European destruction and occupation of the land found to the west of the Atlantic, while the logical mistake of conceiving of a clan as ‘members of a unilineal descent group’ has arisen from our tendency to ignore ‘affinity as a value’ (Dumont 1983). The wrong Euro-American inclination to deal with ‘tribes’, moieties and clans as substantive entities or corporate pluralities of concrete human individuals forming inclusive and included units has caused this misunderstanding. My observations in Highland Middle India, on the other hand, show that several such ‘tribes’ may, at times, merge within some juxtaposition or the other, that ritual ‘brotherhood’ of persons or local descent groups applies equally within and between ‘tribes’, and that ‘intertribal’ marriages too must observe clan exogamy. I have been working in a district where several anarchic societies16 employed an identical clan classification. Three of them shared the same exogamous moieties of the cobra and the leopard. In spite of the very different – contemporary – demographic situation, I contend that the Seneca terminology had been developed under conditions comparable to those of Highland Middle India. In the same way, the assigned categories, such as moieties and clans, were also to be employed as ‘shortcuts’ to choose the proper ‘kinship’ term for a person. Such a system must have been continued even after the genocide executed by Europeans resulted in the existence of demographic units comprising only a few hundred people. As a consequence of the external onslaught, each clan of such a unit would then appear to anthropological observers as if it was a ‘group’ consisting of a very limited number of individuals rather than as representing a general concept of the value of affinity. The above conjectures have resulted from my ethnographic work in contemporary Indigenous Middle India, where I have found a speaker calling a person ‘maternal uncle’ if the latter belonged to the clan of

50 • Lewis Henry Morgan’s Comparisons

Ego’s mother, irrespective of the genealogical ‘distance’ or whether a genealogical relationship existed at all. Thus, certain inhabitants of a village are collectively called ‘sisters’ children’ because of their clanship and not because they can be traced genealogically to identifiable female persons. Similarly, a man of a different clan becomes Ego’s brother and his father thus becomes Ego’s father, if this man has married the genealogical sister of Ego’s wife (see Figure 2.1 above) and then the same holds true for the respective other relationships. When comparing the very limited number of European kinship metaphors with those Lounsbury assumes to be Seneca ones, I also see a basic logical difference. Europeans conceive some kind of a behavioural likeness in a highly unsystematic manner. One finds a Catholic ‘father’ leading a ritual, or a ‘sister’ as a nurse in a hospital, and also a ‘brother at arms’, but no system relating these three to each other. Because she is working for the patients as a nurse, the woman becomes a ‘sister’, but because he is a ‘maternal uncle’, an Indigenous person behaves differently compared to a ‘father’. Both of these roles only make sense in relation to each other and the entire system of ‘kinship’.

Preliminaries: The Alliance Bias In 1949, Lévi-Strauss overcame the European descent bias, though he sometimes used a language that is no longer acceptable according to the standards of gender equality. He showed that the ‘prohibition of incest is … a rule obliging the mother, sister or daughter to be given to others’, since this is ‘the supreme rule of the gift’ (1969: 481).17 His magnum opus could demonstrate that it is ‘always a system of exchange that we find at the origin of rules of marriage’ (1969: 478). He introduced his ‘elementary structures of kinship’ or systems of the nomenclature: which prescribe marriage with a certain type of relative or alternatively, those which, while defining all members of the society as relatives, divide them into two categories, viz., possible spouses and prohibited spouses. (1969: xxiii)

By contrast, ‘complex structures … limit themselves to defining the circle of relatives and leave the determination to other mechanisms, economic or psychological’ (Lévi-Strauss 1969: xxiii). In the preface to the second edition, he also took up what he called ‘intermediary types’ (1969: xxxvi) in the shape of those normally called ‘Crow-Omaha systems’, but I agree with the observation18 that his criteria apply equally to the Seneca terminology, since they refer to:

Seneca Revisited • 51

systems which only set up preventions to marriage, but apply them so widely through constraints inherent in their kinship nomenclature that, because of the relatively small population, consisting of no more than a few thousand persons, it might be possible to obtain the converse, viz., a system of unconscious prescriptions which would reproduce exactly … the contours of the mould formed by the system of conscious prohibitions. (1969: xxxvi)

Unfortunately, by confounding the demographic19 situation in the society of the speakers with formal aspects of their language, Lévi-Strauss reduces the value of his statement that otherwise rightly refers to the systemic ‘constraints inherent in their kinship nomenclature’ that such ‘intermediary types’ share with his ‘elementary structures’. Differing from Lévi-Strauss, other scholars20 of the alliance school, when writing on prescription, confine an analysis to the formal structure of terminological categories. The latter must leave no uncertainty about the position of the affinal ‘return gift’ for Ego. For these authors, the terminological structure articulates a positive message when characterized by affinal prescription. Because such prescription only refers to positive imperatives, I suggest that apparently negative commands can be interpreted in the same way. Classificatory systems like those of the North American Seneca, the Omaha and innumerable others containing a negative message in their respective terminological structure are in fact based upon the same value-ideas, namely the spirit of the gift, as those containing a positive statement. Such a proscriptive terminology unconditionally excludes the repetition of affinal reciprocity in generations following that of Ego, since its formal structure categorically commands new alliances in the later generations, thereby communicating the ideological and technical imperative to socialize by exchange with more than just a single affinal opposite. Excluding repetition thus implies a widening of social contacts by reciprocal affinal relations or the unconditional extension of the pool of possible relationships. This implication is emphasized by the terminological whole, whether or not it is equally given by the set of behavioural norms. While prescriptive terminologies, or those containing a positive message, command an intergenerational alliance with the same affinal ‘other’, proscriptive terminologies contain the order that a different exchange relationship is to be initiated in each new generation, thereby making the spirit of the gift, or the value of social reciprocity, even more explicit than a prescriptive structure. At the same time, my suggestion to give equal recognition to proscriptive and to prescriptive terminologies will clarify the fundamental

52 • Lewis Henry Morgan’s Comparisons

distinction between, on the one hand, those classificatory terminologies that are proscribing repetition and, on the other hand, descriptive vocabularies, such as the English one, referring to ‘concrete relations’. Both are worlds apart from each other, as Morgan has discussed at length, since all classificatory systems based on affinity have it that ‘the several collateral lines, near and remote, are finally brought into, and merged in the lineal line’ (Morgan 1871: 14), whereas the descriptive vocabulary cannot ‘preserve’ the collateral relationships and will ‘lose’ them over time. To conflate proscriptive classificatory systems with a descriptive vocabulary under the label of ‘complex systems’ (Lévi-Strauss 1969) is untenable because it confounds systems and nonsystems. Like prescriptive systems, proscriptive systems formally determine reciprocal affinal relationships, as will be confirmed below by the analysis of the Seneca terminology. Such a structure lacks formally composed links of Ego’s spouse connecting him or her to other generations, but the pattern still articulates formal relationships of reciprocal or symmetric affinity. Even if it does not clarify the relational network of Ego’s spouse, it does identify symmetric affinal reciprocity in other sectors of the terminological whole, as will be discussed in the analysis below. To conclude: apart from the so-called ‘Malayan systems’21 (Morgan 1871: 453), I will rate the classificatory terminologies I know, be they of the prescriptive or the proscriptive kind, as systems of affinal reciprocity based upon the Maussian ‘spirit of the gift’. Moreover, since Morgan’s ‘descriptive’ nomenclatures contain no formal or empirical commonalities with classificatory terminologies of the proscriptive kind, the two should not, as Lévi-Strauss had suggested, be united under a common label. In the following, I will examine the Seneca system as an example of such a proscriptive order. It appears to be quite suitable for such a demonstration, since in the past probably no other terminology has been studied more often and more intensely.

Seneca Terminology The following discussion of Seneca will demonstrate how the structure of a classificatory terminology implies what Whiteley has called the ‘conjoint tendency to preserve and disperse alliances’ (2012: 107). As reported by Morgan (1871: 167–69), relationships between all terms will be listed below, though I do omit the many redundancies and also the words for twins, as well as those provided for widowed people in his Seneca tables. Since the system is ‘reducing consanguinei to great classes by a series of apparently arbitrary generalizations’ (1871: 13), I

Seneca Revisited • 53

Table 3.1 List of Seneca Iroquois terms (Morgan 1871: 167–69).  1.  2.  3.  4.  5.  6.  7.  8.  9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.

Hoc’-sote Oc’-sote Hä’-nih No-yeh’ Ha-ah’-wuk

FF, MF, FFF, MFF, FMF, MFF, W’s grandfather MM, FM, MMM, FMM, FFM, MFM, W’s grandmother F, FB, FFBS M, MZ S, BSms, ZSws, FBSSms, FZSSms, FBDSws, FZDSws, MBSSms, MBDSws, MZSSms, MZDSws, MMBSSSms Ka-ah’-wuk D, BDms, ZDws, FBSDms, FBDDws, FZSDms, MBDDws, MBSDms, FZDDws, MZSDms, MZDDws, MMBSSDms Ha-yä’-da all grandsons and great-grandsons Ka-yä’-da all granddaughters and great-granddaughters Hä’-je eB, FBSe, MZSe, FFBSS Ah’-je eZ; FBDe, MZDe Ha’-gă yB, FBSy, MZSy Ka’-gă yZ, FBDy. MZDy Ha-yă’-wan-da ZSms, FZDSms, MBDSms, FBDSms, MZDSms Ka-yă’wan-da ZDms, FZDDms, MBDDms, FBDDms, MZDDms Ha-soh’-neh BSws, FBSSws, FZSSws, MBSSws, MZSSws, MMBSSSws Ka-soh’-neh BDws, FBSDws, FZSDws, MBSDws, MZSDws, MMBSSDws Ah-gah’-huc FZ Hoc-no’-seh MB, MMBS Ah-gä’ni-ah MBW Ah-găre’-seh FZD, FZS, MBS, MBD, MMBSS Da-yake’-ne H, W Hä-ga-sä HF, H’s grandfather On-ya’-sä HM, H’s, grandmother Oc-na’-hose WF, WM, ZDH, BDH, son-in-law Hoc-no’-ese MZH, FZH, stepfather Oc-no’-ese FBW, stepmother Ka’-sä ZSW, BSW, daughter-in-law Ah-ge-ah’-ne-ah BWms, FBSWms, FZSWms, MBSWms, MZSWms Ah-ge⁀ah‘ne⁀o HZ, WB, BWws, ZHms, FBSWws, FZDHms, FBDHms, FZSWws, MBDHms, MBSWws, MZSWws, MZDHms Ha-yă’-o HB, ZHws, FZDHws, MBDHws, MZDHws Ka-yä’-o WZ Ka’-no stepdaughter Ha’-no stepson, Hä’-je (o) ha’-gă(y) stepbrother Ah’je(o ka’-gă(y) stepsister

represent the interdependent parts of such a whole in a unified matrix. Person symbols (circles and triangles) will only be introduced to highlight a particular relationship. They cannot represent a system, since they stand for individuals rather than categorical relations. Behind each term in the matrix below, its particular number from the above list is added in square brackets to facilitate the checking of all given

54 • Lewis Henry Morgan’s Comparisons

Figure 3.2 The Seneca Iroquois classificatory system. Figure created by the author.

equations. For reasons of space, not all equations from the list could be entered into the boxes of the matrix, but the missing ones can be easily traced by the numbers. In the affinal lines, the central generation G0 is not divided by the male/female opposition, but, following Seneca classification, its terms for categories include males or females without distinction.

Analysis of the Seneca System What Morgan22 called the ‘plan’ (1871: 11) and Dumont (1983) introduced as the ‘value’23 of a terminology is, in the Seneca case, the implication not to repeat one and the same affinal relationship. Formally, symmetric affinal exchanges are dispersed. In each new generation, a new affinal line is invariably represented as ‘giving’ a member to the

Seneca Revisited • 55

only consanguinal line as a spouse and as ‘taking’ one in return. Whereas ‘Dravidian’ terminologies of Hindu South India (see Trautmann 1981) prescribe diachronic affinity by implying the immediate repetition of such a reciprocal affinal relationship between two lines (Dumont 1983: 3–17), the language of the Seneca radically excludes such a possibility. Like ‘Dravidian’, the Seneca pattern implies the symmetry of all affinal relations. ‘Givers’ are formalized as reciprocal ‘takers’. However, what is clearly classified as ‘give-and-take’ between two lines is always encompassed by the overall imperative of affinal dispersal as an invariant value. The following will consider these relations in detail. In G+1 both M [4] and F [3], in spite of being spouses of each other, are classified as Ego’s consanguines. MB [18] is represented as the affine of F and thus of Ego, just as FZ [17] (along with MBW [19]) is shown to be the affine of M and thereby of Ego. This affinal relationship implies that F is replacing MB as a male in the same way as M is replacing FZ within the duo on the female side. Formally, such a relationship means that a husband is substituting a brother, while a wife comes in for the outgoing sister of the female-male pair. The fact that MZH [25] is equated with ‘stepfather’ and FBW [26] with ‘stepmother’ within the consanguineal line, as if they were surrogate parents, confirms the system, whereas I fail to detect the logic contained in the equation of MZH and FZH [25]. By facing the consanguineal line of M and F, FZ and MB of the first affinal line are initially represented in the symmetric affinal relationship of G+1. Subsequently, their children MZS/MZD/FZD/FZS, all equated by the single term Ah-găre’-she [20], do not continue this relationship. Significantly, this category of a cross-cousin [20] also figures as the spouse of Ah-ge⁀ah’ne⁀o [29] of the second affinal line. Thus, it is not equated with that of W/H [21]. Nor is it equated with either sibling’s spouse or with spouse’s sibling [28], [30], [31]. Though the term for cross-cousin Ah-găre’-she [20] differs from the term for sibling to mark the different parental line, it shares with siblings the same affinal opposition. In short, it is positioned like a consanguine, or ‘consanguinized’. The children of Ah-găre’-she [20] in the subsequent generation G-1 then no longer differ in any way from the children of siblings. In the second affinal line, a woman applies a single term Ha-yă’-o [30] for both HB and ZH to show that two brothers may be the husbands of two sisters, but a man does discriminate BW as Ah-ge-ah’-ne-ah [28] from WZ as Ka-yä’-o [31] to exclude this possibility. Encompassed or in the centre of the entire system, a single category receives special exposure. Ah-ge⁀ah’ne⁀o [29] stands for the equation of HZ, WB, BWws, ZHms, FBSWws, FZSWws, FBDHms, FZDHms,

56 • Lewis Henry Morgan’s Comparisons

Figure 3.3 Seneca term [29] Ah-ge⁀ah‘ne⁀o equating HZ, WB, BWws, ZHms, FBSWws, FZDHms, FBDHms, FZSWws, MBDHms, MBSWws, MZSWws and MZDHms. Figure created by the author.

MBSWws, MBDHms, MZSWws and MZDHms. Though the introduction of person symbols (circles and triangles) may be misleading in terminological analysis, in this case (see Figure 3.3) it is a useful tool to illustrate the symmetry of relationships as being enforced by a single term [29]. Figure 3.3 is not meant to demonstrate anything else. Irrespective of sex, Ego calls Alter ‘affine’, or Ah-ge⁀ah’ne⁀o [29], and is reciprocally called the same by Alter. The generality of the category makes it include persons of either sex. This affinal category Ah-ge⁀ah’ne⁀o [29] of G0 provides the model for the symmetric relations of the whole. Its nondistinction of sex disallows potentially awkward ties. It equates other-sex sibling’s spouse and spouse’s sibling (same-sex), thus clearly indicating the process of a symmetric give and take. It is quite distinct from the two different terms for spouse’s sibling (other-sex) [30], [31], from that for same-sex sibling’s spouse [28] and also from the term for spouse [21]. None of these five categories [21], [28], [29], [30] or [31] could in any way be mapped in the style of a ‘Dravidian’ equation.24 Of all-important significance is the reciprocal affinal relationship between a cross-cousin [20] and Ah-ge⁀ah’ne⁀o [29], since the latter term not only includes FZDHms, FZSWws, MBDHms and MBSWws, or spouse of cross-cousin, but also MZSWws, MZDHms, FBSWws, and FBDHms, or spouse of sibling-like parallel cousin. The meaning of the term Ah-ge⁀ah’ne⁀o [29] also includes: • BWws, ZHms or other-sex sibling’s spouse; • HZ, WB or spouse’s sibling (same sex).

Seneca Revisited • 57

However, this affinal category Ah-ge⁀ah’ne⁀o [29] is distinct from the one that includes HB and ZHws [30] and is equally distinct from WZ [31], as well as from BWms [28]. The formal implications of these equations and discriminations are as follows: • The central equation [29] includes BWws or the wife ‘entering’ the new relationship to replace the ‘outgoing sister’ Ego (female), as well as including ZHms, the husband ‘entering’ the new relationship to replace the ‘outgoing brother’ Ego (male), and also including HZ and WB, the sibling who is ‘replaced’ by Ego in the relationship. The equation Ah-ge⁀ah’ne⁀o [29] implies BW=HZ, or symmetry. The category is the affinal counterpart of siblings in one reciprocal relationship and also – separately though simultaneously – the affinal counterpart of cross-cousins, whereas Ego is individualized and supplied with an individualized spouse. Furthermore: • In G0 terms for cross-cousins [20] continue the first affinal line. However, by virtue of sharing the same affines with Z and B of the consanguineal line, these cross-cousins become the same as consanguines. Terminologically, they are distinct from consanguines, but relationally they are not distinct since they, together with siblings, jointly oppose the second affinal line. For this reason, their children are no longer distinguished from those of siblings. • BWms [28], ZHws [30], WZ [31] and W/H [21] are not obviously involved in any reciprocal pattern, even though they certainly do not formally exclude relations of symmetry. They are not mapped in any relation of affinal reciprocity so that one might rate them as ­individualistic appendices. However, since Morgan (1871: 169) explicitly marks the absence of specifications for WBW, HBW, WZH and HZH, the reciprocal ‘gift’ to WB, HB, WZ and HZ must be Z and B. For the consanguineal line, the category Ah-ge⁀ah’ne⁀o [29] represents the new or second affinal line in G0, but the term must also be seen as succeeding other terms of G+1 and G+2, since the skewed HF/HFF [22] together with the skewed HM/HFM [23] imply even in G+2 a discrimination of the second affinal line from the consanguineal line. While FF and MF [1] along with MM and FM [2] respectively are equated as consanguines, HFF [22] and HFM [23] demonstrate an affinal distinction of categories even in the upper marginal generation. As an alternative to the ‘upward’ skewing of the husband’s forbears, both WM and WF Oc-na’-hose [24] take the ‘downward’ direction by being the self-reciprocals of DH [24].

58 • Lewis Henry Morgan’s Comparisons

In the filial generation of G-1, Oc-na’-hose [24] equates DH, BDH and also ZDH. In the same way, Ka’-sä [27] equates SW, ZSW and BSW. Both again (as for cross-cousins) reduce the affinal status of other-sex sibling’s children, since the categories for ZSms, etc. [13] and that for BSws, etc. [15] share affinal counterparts with Ego’s own children. The same is the case for ZDms, etc. [14] and for BDws, etc. [16]. These four terms [13], [14], [15] and 16] continue the second affinal line and yet, together with members of the consanguineal line, they share the same affinal counterparts of the third affinal line. By sharing these with consanguines, they become like consanguines or are ‘consanguinized’. All of this sounds complicated, so I will simplify the demonstration of my thesis that affines of one generation are systematically ‘consanguinized’ in the following generation using an illustration. On the left side of Figure 3.4 below, the parents of all those relational positions that are represented by grey letters belong to the second affinal line, while the parents of those printed in black letters belong to the consanguineal line. However, for the respective children this distinction of the parental status will become irrelevant because of their joint reciprocal affinal relation when facing the new affines of G-1 This joint position assembles all on the left-hand side of Figure 3.4 as counterparts of DH/BDH/ZDH [24] and SW/ZSW/BSW [27] of the third affinal line positioned on the right-hand side.

Figure 3.4 Children of affines become consanguines by opposing new affines jointly with consanguines. Figure created by the author.

Figure 3.5 Former affines become consanguines by jointly opposing new affines in the subsequent generation. Figure created by the author.

Seneca Revisited • 59

In short, over the course of the three medial generations, Seneca terminology has transformed affines into consanguines by joining them in opposition against the new affines of the subsequent generation (see Figure 3.4). Compared to the three medial generations, G-2 does not contain any surprises. Long ago, Morgan observed that: the several collateral lines near or remote, are finally brought into, and merged in the lineal line, thus theoretically, if not practically, preventing the dispersion of blood … This mergence is, on like manner, one of the characteristics of the classificatory system. (1871: 13)

The phrasing ‘theoretically, if not practically’ allows us to consider that Morgan may have been aware of the distinction between the formal and the empirical, or the analytical frame of reference designed for terminological categories as distinguished from the frame of reference concerned with social norms and practices. In conclusion, I suggest that the systemic ideas or the values of the Seneca classification (see Figure 3.2) maintain that: • within any one generation, affinal relations are symmetric; • affinal relations cannot be repeated in subsequent generations; • in the generation following the first reciprocal affinal relations, consanguines and the terminologically distinct children of their first affines face identical counterparts of a second affinal line; • in the generation following this second relation of reciprocity, consanguines and the terminologically distinct children of the second affines likewise face identical counterparts of a third affinal line. • thus, along with the change of generation, former affines come to share the same side as consanguines when facing the new affines; • by sharing the same side in the binary opposition, affines eventually become consanguines. Seneca social norms for marriage seem to maintain the same value by enforcing new alliances in every new generation or by generalizing the spirit of the gift by exchanges with many different affinal partners rather than repeating the alliance. By the combination of these proscriptions and prescriptions the structure of the societal whole is maintained.

Notes  1. In this article, Lounsbury was the first to point out the elementary difference between the Seneca terminology and the ‘Dravidian’ nomenclature

60 • Lewis Henry Morgan’s Comparisons

 2.  3.

 4.  5.  6.

 7.  8.  9. 10. 11. 12.

13. 14. 15. 16.

of South India. Unfortunately, the two patterns continue to be equated in many contemporary publications. Earlier versions of his contribution (or parts of it) were originally presented in 1956. In almost all publications of that time, such gendered language can be criticized. After writing ‘We restrict our discussion to consanguineal types’ (1964b: 1075), Lounsbury does not explain his omission. Several contributors to the pioneer volume edited by Eggan, as well as Eggan himself (1970a [1937]: 43, 44), initially present a major diagram of what they conceive as ‘kin terms’ along with a minor diagram of ‘affinal terms’ or else ‘in-law terms’. More recently, Trautmann and Barnes (1998: 31) and Tjon Sie Fat (1998: 71) also explicitly confine their analyses to consanguineal terms. See Bürgerliches Gesetzbuch (BGB) §§ 1589, 1590. The corresponding procedure is followed for the term ‘son-in-law’. The lasting attraction of this wrong notion may have resulted from the patriarchal concept of the Bible, with figures like Adam, Sem, Abraham and Judah who I consider to be of an altogether different social order compared to that of the clan system. Morgan’s example (1901: 77; 1871: 139) can easily be confirmed by those in many other anarchic societies. I may not have read the books containing such a positive turn. See the homepage for the Seneca Nation of Indians. Retrieved 29 January 2019 from https://sni.org. Morgan (1901: 75–78; 1871: 139) refers to them in League of the Ho-De-NoSau-Nee, or Iroquois. Morgan even goes so far as to state that ‘originally’ (1851: 79) two exogamous moieties existed, though he applies the technical term ‘phratries’ for them. Better than any other example, the system of moieties demonstrates the difference between the empirical and the formal, the substantive and the categorical. Empirically, it is absolutely impossible to maintain exogamy/ intermarriage, since demographic conditions never provide an equal size of the two halves within a society based upon the imperative of universal marriage. Some technical deviation must always be found, though it will not touch the omnipresence of the affinal idea. I have been confronted with several cases of this distinction of formal and empirical moieties in Highland Middle India. Trautmann and Barnes qualify Lounsbury’s comment as ‘the most famous footnote in anthropology’ (1998: 27). In none of Morgan’s works could I find a passage indicating that Seneca relationship terms were structured according to the clan or moiety membership of an individual speaker. In a comparable context, the same holds for the term ‘maternal uncle’ or any other. The persons involved spoke different languages belonging to the three different language families.

Seneca Revisited • 61

17. It must be remembered that Lévi-Strauss was writing on systems and not on actors. Technically, the same rule refers to the father, the brother and the son, and could equally well be articulated as such. 18. ‘[T]he proper terrain for analysis of semi-complex marriage structures ought to be all cases of semi-complex marriage prohibitions, whether or not they are accompanied by terminologies containing skewing’ (Trautmann and Whiteley 2012: 21). 19. In Highland Middle India, such ‘intermediate’ terminologies (e.g. in a Dravidian language called Gondi or a Munda language called Santali) are applied by several million and not just ‘no more than a few thousand’ people. 20. See especially Needham 1969: xvii–xix; 1971; xiii–cxvii; 1973. 21. They will be discussed in Chapter 6. 22. ‘[E]very scheme of consanguinity and affinity will be found to rest upon definite ideas, and to be framed, so far as it contains any plan, with reference to particular ends’ (Morgan 1871: 11). 23. Morgan too remarks that ‘the adoption of some method to distinguish one relative from another, and to express the value of the relationship, would be one of the earliest acts of human intelligence’ (1871: 10). 24. The basic difference between ‘Iroquois’ and ‘Dravidian’ terminologies has been discussed at length by Trautmann (1981: 87).

4 Omaha Skewing Reconsidered

Afterwards in 1860, while at the Iowa reservation in Nebraska, I had an opportunity to test it … through White Cloud a native Iowa well versed in English. While discussing these relationships he pointed out a boy near us, and remarked that he was his uncle, and the son of his mother’s brother who was also his uncle. —Lewis Henry Morgan, Systems of Consanguinity and Affinity of the Human Family

The Initial Mistake All terminological studies imply the ‘fundamental work on the distinction between the formal and the empirical’ (Needham 1966: 141). In case of terminological skewing, this basic premise seems to have become even more relevant than before. It is at the root of all debates that, since Morgan’s days, have preoccupied scholars deciphering the classificatory system of the Native American Omaha. Europeans are simply stunned by the nomenclature overriding the classificatory principle of generation. An Omaha FZH is referred to by the same term as a WB and a WBS, while an MFZ is referred to by that of an MBD. Nobody is astonished if some other system does not distinguish relative age to equate, for example, the elder and the younger sister, and few observers are amazed if it happens to ignore the sex-difference between cousins. But the Omaha way of equating, for example, WFZ and BW has largely confused Euro-American anthropologists. The Seneca equation of lineals and collaterals (i.e. that of parents and their siblings) must have been a breathtaking discovery. As an additional variation, Omaha skewing was probably most annoying to European minds. For many, it may have been among the reasons to discard terminological studies altogether. By contrast, the fact that kinship means consanguinity only for speakers of English or other European languages that provide ‘in-law’ terms with a subordinate

Omaha Skewing Reconsidered • 63

or secondary value has preoccupied only a small minority of ‘kinship’ researchers. Recently, following a conference of experts, a group of scholars published a thick description of ‘Crow/Omaha’, the ‘classic problem of kinship analysis’, debating the history of several controversies as well as a number of paths towards future insights (Trautmann and Whiteley 2012), but without a consensual solution for the problem. The ‘type’ in question ‘has come to mean a relationship terminology containing equations that might be termed “patrilineal” (e.g. MB=MBS=MBSS)’ (Barnes 2012: 74). The ‘skewing/lineal equations’ in anthropological jargon is defined in the glossary of this seminal publication as ‘the merging of kin down a unilineal descent line, as in terminologies of the Crow or Omaha type’ (Trautmann and Whiteley 2012: 306). However, the definition seems to create several problems because – apart from other critical points – the equations applied in the language of the Prairie people known as Omaha include social categories that definitely do not refer to members of the same unilineal descent line. For example, Ita’ hon, which, among others, includes FZH=ZH=WB, does not ‘merge kin down a unilineal descent line’, nor does Iti’zhun, which means FZD and also ZD. A definition of the above-mentioned kind can, of course, be adjusted without major problems. More relevant seems to be what I consider the ‘initial mistake’ of many terminological studies. It is responsible for the existence of an ‘Omaha problem’. The usual criterion of ‘Omaha’ terminologies, i.e. the equation MB=MBS=MBSS, refers only to a very partial extract from the entire nomenclature, whereas I contend that any classificatory relationship terminology can only make sense as a structural whole or when all of its parts are examined according to their mutual relationships. Selecting bits and pieces out of this whole and theorizing over such a self-selected sample may be suitable for a descriptive terminology like Spanish or French, because the parts of these amount to designations of several concrete persons, whereas the logic of a classificatory system is unconcerned about individuals. Since Morgan’s initiative, the analysis of such a system intends to discover certain generalizations ordering the social whole. The epochal quality of his work is also the result of his meticulous analysis addressed to all parts of a terminological whole rather than being content with a sample of those that appear to be relevant or astonishing to a Euro-American observer.

64 • Lewis Henry Morgan’s Comparisons

The Additional Fault As Barnes (2012: 74) has pointed out, some of the very prominent scholars (Kohler and Durkheim), but neither Morgan nor Kroeber, argued that Omaha skewing had resulted from social practices emphasizing unilineal descent, or clanship, but in fact, such a causal relationship is unwarranted and many terminological aspects point to the contrary. To be more explicit, I contend that an Omaha clan, assigned unilaterally to a child, is a sociocentric category of affinal exchange or of the quintessential Indigenous value characterizing the specific form of a holistic social order. By its affinal relations, the clan system contains the ‘spirit of the gift’, which, at the same time, is the ideological foundation of terminological classification. This does not mean that the nomenclature in any way reflects the empirical specifications of the exchanges between clans or any kind of unilineal institution, even if such an impression may be created in a European mind. This impression is wrong. A common ideological basis in no way implies a common technical execution of ‘the spirit of the gift’ within different domains. Thus, the ‘Omaha problem’ should only be approached after such a causal relationship between clanship and nomenclature has been excluded from subsequent discussions.

Type ‘Omaha’ and Omaha Terms About half a century ago and with special reference to the ‘Omaha’ label, among others, Needham criticized the ‘premature urge to construct empirical typologies’ of ‘mistakenly grouped’ phenomena. His verdict explicitly stated that ‘there exists no useful generalization about this fictitious class of Omaha terminologies’ (1971: 14). The same opinion was expressed by Barnes (1975, 1984 and 2012), who has carried out intensive archival research on the ethnographic accounts of the Omaha. Long before Malinowski, the social structure of this Prairie people had been reported very conscientiously after long-term participant observation by ethnographers whose unpublished materials (fieldnotes, etc.) are available in the archives. In the following, I will adopt Needham’s and Barnes’ careful arguments against the very existence and worldwide distribution of a type called ‘Omaha’ to discuss only the relationship terminology of empirical Native Americans known as Omaha in anthropological literature. The outstanding ethnographic accounts of these Indigenous people,

Omaha Skewing Reconsidered • 65

especially that of Dorsey (1884) and of Fletcher and La Fleche (1911), will not be taken into account here, since I will confine my discussion to formal issues. As to terminological details, Kronenfeld (1989) questioned some of Morgan’s data and identified twelve errors that Trautmann and Barnes have examined in detail to find that ‘the ethnological discussion of Omaha type terminologies in North America has not been contaminated by Morgan’s errors’ (Trautmann and Barnes 1998: 54). Since I lack any expertise in Amerindian anthropology, this chapter will, after some general observations, only raise the issue of Omaha skewing in the light of affinal relationships. For this purpose, I will make use of the Omaha vocabulary on the lists below provided by Barnes, who is the specialist in this field. As far as I know, he is the only author to examine the complete relationship terminology (2012: 78) and is one of the few who takes the terminological whole into consideration rather than selecting only a few portions of it. He also separates the terms recorded for a male Ego from those for a female by supplying two different lists (Barnes 1984: 132–37). Such a sex-wise specification is an indispensable step towards an understanding of the ‘classic problem’. I find it quite amazing that modern gender studies have never taken up the issue of these separate formal perspectives. The technical requirement to translate a single Native American classificatory system into two different representations for European understanding concerns sociocultural categories beyond those of the Western tradition in the humanities or the social sciences and thus seems to belong to the given anthropological domains rather than to those of any other discipline.

Who Is a Consanguine, Who Is an Affine? Anthropology courses on social structure usually teach how ‘patrilineal’ clans differ from ‘matrilineal’ clans. As long as these clans are descent constructs like the Nuer units, such usage of the metalanguage is correct. However, most Native Americans, and certainly the Omaha, seem to lack clans based upon lineages or any other genealogical reckoning of the Nuer kind. From among the ten Omaha clans, the Thatada, for example, are subdivided into four subclans, while others include only two. In the inclusive moieties of the ‘Sky People’ facing the ‘Earth People’, and the five clans within each one of these, and also in a subclan, membership is not determined by someone’s pedigree. Moieties, clans and subclans lack any tendency towards a systematic segmentation. The moment the descent bias of the European tradition

66 • Lewis Henry Morgan’s Comparisons

Table 4.1  List of Omaha terms for a male Ego (Barnes 1984: 132–49).  1. Iti’gon

 2. Ikon’

 3. Itha’di  4. Ihon’  5. Iti’mi  6. Ine’gi  7. Izhin’the  8. Içun’ga  9. Itonge 10. Izhin’ge 11. Izhun’ge 12. Iton’shka 13. Iti’zhun 14. Itu’shpa

15. Ita’ hon 16. Ihon’ga 17. Iti’ni 18. Iton’de 19. Igaxthon 20. WZH = Ego = WFZH

FF, FFB, FFF, FFFF, FFFBS, FMF, FMB, FMBS, FMBSS, FMZH, MF, MFB, MFF, MMF, MMMF, MMB, MMBS, MMBSS, MMBSSS, WF, WFF, WFB,WMF, WMB, WMS, WMBSS, WMZH, SWF FM, FMZ, FMM, FMBW, FMBD, FMBSD, FFM, FFZ, FFFM, MM, MMZ, MFM, MMM, MMBD, MMBSD, MMBSSD, MMMM, MMMZ, MMMZD, WM, WMM, WMZ, WMBW, WMBSD, SWM F, FB, FMZS, FMBDS, FFBS, FFFBSS, MZH M, MZ, MBD, MBSD, MBSSD, MFBD, MFBSD, MFBSSD, MFZ, MMBSDD, MMBDD, MMBDSD, MMZD, MMZSD, MMMZDD, FBW FZ, FMZD, FMBDD, FFBD, MBW MB, MBS, MBSS, MBSSS, MFBS, MFBSS, MFBSSS, MMBSDS, MMBDS, MMBDSS, MMZS, MMZS, MMZSS eB, FBSe, FMZSSe, FFBSSe, FFFBSSe, MZSe, MFZSe, MMBDDSe, MMZDSe yB, FBSy, FMZSSy, FFBSSy, MZSy, MFZSy, MMBDDSy, MMZDSy Z, FBD, FMZSD, FFBSD, MZD, MFZD, MMBDDD, MMZDD, MMMZDDD S, BS, FBSS, MZSS, MFZSS, MFZSS, FFFBSSSS, WBDS, WFBDS, WZS, WMBDDS, WMZDS D, BD, FBSD, MZSD, MFZSD, WBDD, WFZD, WFBDD, WZD, WMBDDD, WMZDD ZS, FBDS, FFBDS, MZDS, MFZDS, FZS, FMZDS, FFZS ZD, FBDD, MZDD, MFZDD, MMMZDDDD, FZD, FMZDD, FFBDD, FFZD, ZHBD SS, SD, SSS, SSD, SDS, SDD, DS, DD, DDS, DDD, DSD, DSS, BSS, BSD, BDS, BDD, ZSS, ZSD, ZDS, ZDD, MFZSSS, MFZSDD, MFZDSD, MFZDDS, MMMZDDDD, FZSS, FZSD, FZDS, FZDD, FFZSS, FFTSD, FFZDS, FFZDD, FFFBSSSSS, SWB, SWBS, DHZ, DHFZ, WZSS, WZSD, WZDS, WZDD, WFZSS, WFZSD, WFZDS, WFZDD, ZHZ, ZHFBD, ZHFBDD, ZHFBSD WB, WBS, WBSS, WFBS, WFBSS, WMBDS, WMBDSS, WMZS, WMZSS, ZH, FZH, FMZDH, FFZH WZ, WMBDSD, WMBDD, WMZSD, WMZD, WBD, WBSD, WFZ, WFBD, WFBSD, BW SW, SWZ, SWBD, SSW, WBW, DHM, DSW, ZSW, FZSW, FFZSW DH, DDH, SDH, FZDH, FFZDH, ZDH, DHB, DHF, DHFF W

Omaha Skewing Reconsidered • 67

Table 4.2  List of Omaha terms for a female Ego (Barnes 1984: 135–37).  1. Iti’gon

 2. Ikon’

 3. Itha’di  4. Ihon’  5. Iti’mi  6. Ine’gi  7. Iti’nu  8. Içun’ga  9. Izhun’the 10. Itonge 11. Izhin’ge 12. Izhun’ge 13. Itu’shka 14. Itu’zhunge 15. Itu’shpa

16. Ishi’e 17. Ishi’ko 18. Iti’ni 19. Iton’de 20. E’gthunge ^ 21. HBW = ego = HFBSW

FF, FFB, FFF, FFFF, FFFBS, FMF, FMB, FMBS, FMBSS, FMZH, MF, MFB, MFF, MMF, MMMF, MMB, MMBS, MMBSS, MMBSSS, WF, WFF, WFB,WMF, WMB, WMS, WMBSS, WMZH, SWF FM, FMZ, FMM, FMBW, FMBD, FMBSD, FFM, FFZ, FFFM, MM, MMZ, MFM, MMM, MMBD, MMBSD, MMBSSD, MMMM, MMMZ, MMMZD, WM, WMM, WMZ, WMBW, WMBSD, SWM F, FB, FMZS, FMBDS, FFBS, FFFBSS, MZH M, MZ, MBD, MBSD, MBSSD, MFBD, MFBSD, MFBSSD, MFZ, MMBSDD, MMBDD, MMBDSD, MMZD, MMZSD, MMMZDD, FBW FZ, FMZD, FMBDD, FFBD, MBW MB, MBS, MBSS, MBSSS, MFBS, MFBSS, MFBSSS, MMBSDS, MMBDS, MMBDSS, MMZS, MMZS, MMZSS eB, FBSe, FMZSSe, FFBSSe, FFFBSSe, MZSe, MFBDe, MFZSe, MMBDDSe, MMZDSe yB, FBSy, FMZSSy, FFBSSy, MZSy, MFZSy, MMBDDSy, MMZDSy  eZ, FBDe, FMZSDe, FFSDe, MZDe, MFBDDe, MFZDe, MMBDDDe, MMZDDe, MMMZDDDe yZ, FBDy, FMZSDy, FFBSDy, MZDy, MFBDDy, MMZDDy, MMMZDDDy S, ZS, BDS, FBDS, FZS, FMZDS, FFBDS, FFZS, MZDS, MFBDDS, MFZDS, HBS, WMZDS, HFBSS, HMZSS, HMBDSS D, ZD, BDD, FBDD, FZD, FMZDD, FFBDD, FFZD, MZDD, MFBDDD, MFZDD, HBD, HFBSD, HMZSD, HMBDSD BS, FBSS, FFFBSSSS, MZSS, MFBDSS, MFZSS, HZS, HFBDS, HFZS, HMBDDS, HMZDS BD, FBD, MZSD, MFBDSD, MFZSD, HZD, HFBDD, HFZD, HMBDDD, HMZDD SS, SD, SSS, SSD, SDS, SDD, DS, DD, DDS, DDD, DSD, DSS, BSS, BSD, BDS, BDD, ZSS, ZSD, ZDS, ZDD, MFZSSS, MFZSDD, MFZDSD, MFZDDS, MMMZDDDD, FBDS, FBDD, FZSS, FZSD, FZDS, FZDD, FFZSS, FFZSD, FFZDS, FFZDD, FFFBSSSSS, SWB, SWBS, DHZ, DHFZ, HZSS, HZSD, HZDS, HZDD, HFZSS, HFZSD, HFZDS, HFZDD, HFBSSD, HFBSDS, HFBDSS, HFBDDD, HMZSSS, HMZSDD, HMZDSD, HMZDDS HB, HFBS, HMZS, HMBDS, BDH, ZH, FZH, FMZDH, FFZH HZ, HFBD, HMZD, BW, HMBDD SW, SWZ, SWBD, SSW, BSW, DHM, HFZSW, HZSW, FZSW DH, HFZDH, DHB, DHF, DHFF, DDH, HZDH, FFZDH H

68 • Lewis Henry Morgan’s Comparisons

is put aside, they must be viewed as being patrilaterally assigned to children, while members of Seneca moieties and clans are ordered by a matrilateral recruitment rule, since a child is classified on the maternal rather than the paternal side. In the first of these two systems, a child belongs to the father’s side and in the second to the mother’s side, though such unilateral ordering is far from universal in anarchic societies of either North America or the rest of the world. Many of them lack a finite number of formal sociocentric categories. As an alternative, a newborn child may, according to a decision taken by the people involved, be associated with either the local group of the father or that of the mother, or perhaps with a local group that includes both parents. In the latter case, the anthropological jargon has named the bilaterally organized group ‘kindred’. Among the Plains peoples, the ‘Cheyenne kindred weaves its individual families into a finely knit cooperative body’ (Hoebel 1960: 30) and several of these join together in a ‘band’ that is not exogamous (Eggan 1970a: 37). Though the realm of such a sociocentric ordering is not mirrored by the classification of relationship terms, a terminological structure based upon the idea of affinal exchange is logically bound to offer similar indicators of the ‘give and take’, even if such an Ego-centric pattern of relationships must1 deviate from the sociocentric assignments. Apart from the so-called ‘Malayan’ system,2 other structures of classificatory terminologies seem to indicate the difference by making affines face consanguines on the opposite side of the relationship. For the people concerned and for ordinary outsiders, the affinal/ consanguineal opposition may appear to be given by nature, but a simple comparison of terminological systems leaves no doubt that classificatory intervention is variable and will always have to cope with the same fundamental impediment: Ego is aware of siblingship and of parenthood. Brother and sister are understood to be of the same kind and should thus be classified together on the same side as Ego, just as father and mother are related to Ego in the same way, so this fact should be formalized by the terminology. However, spouses can never be classified as brother and sister and siblings are never classified as joint parents of their offspring. This Sisyphean task to classify together both pairs of spouses and pairs of siblings in as minimally awkward a style as possible seems to have led to a number of different solutions. Terminologies based upon the idea of affinal exchange may classify the term for M as an affine or as a consanguine in the same way as her B, but then the latter cannot be Ego’s F. Similarly, if F is classified on the same side as M, this will exclude his classification on the same

Omaha Skewing Reconsidered • 69

F

FZ

MB

M

Figure 4.1 Parental generation: categories to be classified as pairs of either affines or consanguines. Figure created by the author. Consanguines

F

M

Affines

MB

FZ

Figure 4.2 Seneca affinal opposition in G+1. Figure created by the author.

Consanguines

F

FZ

Affines

MB

M

Figure 4.3 Omaha affinal opposition in G+1. Figure created by the author.

side as FZ. In short, an affinal system must indicate whether both M and F are consanguines or else M and MB. Placing F along with FZ on the consanguineal side is a third possibility, which would mean that MB and M become affines. Because the relationship terms of anarchic societies like the Omaha do not amount to ‘a classification of kindred … according to their degrees of nearness’, but rather stand for social categories, systematically designed by ‘a series of apparently arbitrary generalizations’ (Morgan 1871: 13), the affinal opposition in the parental generation must involve the following four categories (Figure 4.1) with a female and a male in each pair: The Seneca relationships are clear. In spite of the fact that their clans are assigned matrilaterally, their egocentric classification places F and M as consanguines and the opposing MB and FZ as affines in a reciprocal affinal relationship (see Figure 4.2). However, the Omaha are classified in a different manner, though the designation of FZ is not without its ambiguities. The term includes the equation FZ=MBW, whereas the term for MB does not include the equation MB=FZH. The discrepancy may indicate a certain tendency, but the overall pattern leaves no doubt that F and FZ are jointly classified as Ego’s consanguines and affinally opposed by MB and M (see Figure 4.3). The evidence for this proposition is supplied by the general configuration of terms and will be elaborated upon below.

70 • Lewis Henry Morgan’s Comparisons

Terminological Lines When discussing Amerindian and Indian terminologies in this book, I refer to opposed terminological lines indicating the intergenerational features of affinal relationships in the decisive three medial generations. Thus, in the system that is frequently called ‘Dravidian’ and universal in South India, also being represented by many Amazonian terminologies, two lines are displayed to cover the reciprocal affinal relationships of all three levels, as exemplified by the Piaroa system.3 In Seneca, however, a single consanguineal line is formally committed to change reciprocal relationships from one generation to the next;4 thus, it depends upon three different affinal lines. The Seneca structure is one formally prescribing affinal symmetry on the same genealogical level and formally proscribing a continuation of it in distinct generations. As elaborated above in chapter three, Seneca does contain some skewing (HF=HFF), but this only concerns the marginal generation of G+2 rather than involving all generations, as in Omaha. Such terminological lines of consanguines and affines should not be confused with descent lines, since the former only refer to affinal oppositions in an intergenerational perspective. Long ago, Needham elaborated upon this difference after referring to Dumont’s ‘fundamental work on the distinction between the formal and the empirical’5 (1966: 141) and reached the following conclusions: • the type of relationship terminology is not necessarily correlated with any type of social grouping; • lineal descent terminologies, with the obvious exceptions of those comparatively rare types which make lineal equations (Crow, Omaha, Purum, Ungarinyin, etc.) are not necessarily correlated with particular rules of descent. Speaking on a certain terminology, he stated that: we cannot infer from its lineal character that there will be a definite rule of descent, let alone whether such a rule will be patrilineal or matrilineal, or that it will denote any kind of descent group. (Needham 1966: 142)

By adopting Needham’s position, it should be clear that skewing cannot be understood as ‘the merging of kin down a unilineal descent line’ (Trautmann and Whiteley 2012: 306). Taking the above examples into

Omaha Skewing Reconsidered • 71

account, M and F, in the Seneca case, belong to the same terminological line, but not to the same descent line, while in the Omaha case, M and F neither belong to the same terminological line nor to the same descent line. M shares both types of lines with MB, but the children of M and MB neither share the same terminological nor the same descent line. In short, terminological lines are intergenerational constructs referring to a classificatory system by affinal opposition. They should be studied irrespective of any physical or social grouping, and also irrespective of sociocentric categories, since they happen to be purely formal regularities.

Technicalities Skewing is central to Omaha terminology. Compared to Seneca, this classificatory intervention implies an altogether different order of terminological lines that, at first sight, seems to confuse Europeans considerably. To facilitate understanding, my reinterpretation will adopt the following technical procedures: • I propose that skewing is the Omaha classificatory method to represent terminological lines as formal indicators of given or excluded intergenerational affinal reciprocity. • In the same style as introduced for the Seneca terminology,6 I will add one or two translations of each Omaha term in the matrix below, as well as the number (in square brackets) taken from the above lists provided by Barnes for male or female speakers. This will simplify the checking of equations. • Initially, the terminological structure for a male Ego will be offered in a matrix. It will be followed by partial presentations of it to clarify relational details. Finally, the overview of the terms assigned to a female Ego7 will be given by another matrix. The latter differs from the matrix of a male Ego in details, but follows the same principles, so that the analysis can be less elaborate. • To clarify the difference between the empirical and the formal frame of reference, I have avoided the phrase ‘alliance’ and rather use that of ‘reciprocal affinal relation’, marking each one of these relations by a numbered double-arrow () in the matrixes. • In each matrix that includes Omaha terms, all consanguineal lines will be printed in black, all affinal lines in dark grey letters, and undefined terms (in the two external generations) in light grey, as is initially demonstrated in Figure 4.4.

Figure 4.4 Unified Omaha terminology (male Ego). Figure created by the author.

Omaha Skewing Reconsidered • 73

Generation G+1 Compared to the Seneca pattern, Omaha indicates the major difference in G+1, since M [4] and F [3] are not classified as the consanguineal pair of that generation and consequently FZ [5] and MB [6] do not form the affinal pair. Instead, MB [6] and M [4] are classified as affines opposed by F [3] and FZ [5], the consanguines. Omaha G+1 is all the same, like G+1 of Seneca, characterized by the first reciprocal affinal relationship, the one between the first consanguineal line and the first affinal line, with FZ [5] being classified as the wife of MB [6] and M [4] as the wife of F [3]. The above contention is supported by the indisputable observation that only the first consanguineal line is constructed without skewing, whereas all other terminological lines are emphasized by this measure. Since textbook accounts of Omaha tend to present such an intergenerational equation for the maternal line only, I repeat that all lines involved in an affinal reciprocal relationship with Ego’s line, and also with the second consanguineal line, are formally identified by skewing. In other words, the first consanguineal line, containing distinct terms for Ego’s parents, siblings and children in the three medial generations, provides the system with its ‘vertical’ axis of genealogical levels. Because of this provision, the other lines, including the second consanguineal line, can be classified by intergenerational equations. These skewing measures indicate diachronic affinal constancy, or the categorical impact of regularly ensuring the non-repetition of reciprocal affinal relations. Accordingly, this impact could also be communicated by the somewhat different representation in the matrix in Figure 4.5 below. The term for FZH [15] presents special difficulties. For FZH [15] and for MB [6], FZ [5] is classified as wife, a constellation amounting to a formal contradiction. The doubling is also a unique technical move. It is a consequence of M [4] being classified as an affine. This status of M [4] is attested by major intergenerational equations: the single terminological pair joining terms [6] and [4] not only refers to MB and M, but also to MBS/MBD and to MBSS/MBSD in the three medial generations, thereby formally proscribing a repetition of the first reciprocal affinal relation () in the parental generation in order to cause other reciprocal affinal relations of the first consanguinal line in Go and G-1 respectively. The affinal status of M [4] implies that FZ [5] must be the consanguine on the other side of the opposition, even though FZ [5] cannot be the wife of F [3] or the mother of his children. However, these children

Figure 4.5 Unified Omaha terminology (male Ego) emphasizing skewing: consanguineal lines (in black), affinal lines (in grey), undefined terms of G2 in light grey. Figure created by the author.

Omaha Skewing Reconsidered • 75

of F [3] must be consanguines, as must be her own children, even though these two pairs of children cannot share a single terminological line. As a result, Omaha classification contains a second consanguineal line with the children of FZ [5] in spite of the fact that her husband (FZH, MB) must be an affine. In Seneca, the children of M and F are consanguines, whereas in Omaha, the children of FZ and – separately – those of F are consanguines. They must be separated, because FZ and F cannot be spouses of each other. The solution of the latter dilemma is a second consanguineal line. The children of MB [6] must be equated with him and M [4], but not with their mother FZ [5]. Nor can the children of FZ [5] be equated with FZH [15], because he must be an affine. Also, the children of FZ [5] cannot be in the terminological line of their mother because, due to the first affinal opposition, she shares that of F [3] and his children. For systematic reasons, then, FZ [5] ‘requires’ two terminological husbands. This is a formal measure. The first husband MB [6] must be an affine having affinal children, while the second husband is ‘required’ for her consanguineal children, since they cannot be those of her brother. FZH [15] is the formal ersatz8 for F [3]. The quintessential cause of elementary structural innovations of society, the fact that two spouses cannot procreate those who in subsequent generations will again be two spouses, disallows a ‘pure’ classification of Omaha relationships. All other affinal classificatory systems also have their logically ‘impure’ spots. The rather awkward Omaha solution classifies FZH [15] as ‘some kind of an affine’ or formally unattached. He must be positioned somewhere. In a system employing skewing to represent terminological lines, MB [6] in the first affinal line of G+1 cannot be equated with FZH. As a consequence, and in spite of the fact that genealogically (or ‘by descent’), FZH [15] is not the father of WB/ZH [15], he is classified – in a camouflaging manoeuvre that avoids open embarrassment – within the equation representing the second affinal line. It includes the spouses of Ego’s siblings. Ego’s own spouse is not involved. His wife is an individualized category. Whereas equations of a spouse are central to prescriptive terminologies, they may be individual appendices in proscriptive terminologies (like Seneca and Omaha) in which siblings and siblings’ spouses take the centre stage. A less visible but equally ‘awkward’ solution is found in the Piaroa, the Seneca and the HMI9 systems and many others. These other constructions contain in both the consanguineal line and the affinal line of G+1 couples of spouses (F and M; MB and FZ), but pairs of siblings (B and Z; FZS and FZD) continue their terminological line in the subsequent

76 • Lewis Henry Morgan’s Comparisons

generations. By contrast, all Omaha levels contain pairs of siblings (e.g. F and FZ; MB and M as well as B and Z and FZS and FZD) or categories of a sibling-like relationship.

Generation G0: The Second Affinal Line To indicate its particular kind of prescription and proscription beyond doubt, the matrix in Figure 4.6 will show only a part of the Omaha structure. The second reciprocal relationship () of the system, connecting the first consanguineal line with a second affinal line, is articulated in G0 by the opposition between Ego’s siblings (but not Ego) and Ego’s siblings’ spouses. Thereby, G+1 and GO are jointly characterized by formal documents of affinal symmetry, which, however, leave out Ego. As in the Seneca case, they are ‘encompassed’ by the constellation of the whole (see Figure 4.7). The person symbols of Figure 4.7 may help to illustrate the following information conveyed by the pattern: even though the terminological whole of the Omaha prescribes the symmetric character of relationships connecting Z with BW [16] as well as B with ZH [15] just as, via Ego, this reciprocal affinal relation of G0 is linked to the parental generation of G+1, it only provides an unrelated category as Ego’s wife or husband. Neither W/H nor BW [16] and ZH [15] is directly linked by any other route to the upper generation, and nor can either be mapped as consanguines. For this reason, their very existence formally proscribes any repetition of Ego’s parental relationships. Formally, Ego is left alone with her/his husband/wife, whereas in the all-important

1.CONSANG. LINE

1.AFFINAL LINE male

female

Ine'gi MB[6]

Ihon' M[4]

0 Ine'gi

n'

G+1 G

Iho MBS[6] MBD[4]

male

1.

Itha'di Iti'mi F[3] FZ/MBW[5] Izhin'the eBms[7] Iton'ge Ego n' Icu ga Z[9]

2.AFFINAL LINE male

female

? 2.

female n'

Ita'ho FZH[15]

Ihon'ga WFZ[16]

Ita'hon' Ihon'ga WB/ZH[15]WZ/BW[16]

yBms[8]

Ihon' -1 Ine'gi MBSS[6] MBSD[4]

G

Ita'hon' WBS[15]

Figure 4.6 The second affinal line. Figure created by the author.

Ihon'ga WBD[16]

Omaha Skewing Reconsidered • 77

Figure 4.7 Two encompassed symmetric Omaha relationships. Figure created by the author.

reciprocal relationship, her or his sister and brother equally ‘give to and take from’ another sibling pair. One sibling out of each pair is classified as crossing over to transform them into two couples of spouses. However, FZH [15] of the parental generation, though equated with ZH/WB [15], is in no conceivable way involved in this symmetric relationship of G0.

Generation G-1: Different Opposites of Affines Become Consanguines In G-1, the Omaha terminology (for a male Ego) indicates the crucial twist. The terms for categories S [10] and D [11] of the first ­consanguineal

Figure 4.8 The second consanguineal line. Figure created by the author.

78 • Lewis Henry Morgan’s Comparisons

line are supplied with their affinal counterparts DH [18] and SW [17] of the third affinal line to become the third reciprocal affinal relationship () of the system in the third of the three medial generations. However, DH [18] and SW [17] are in the same way symmetrically allied with ZS [12] and ZD [13], who, because of this fourth reciprocal affinal relationship (), become the second consanguineal line consisting of equations in the G0 and G-1 categories. As in the Seneca case, the joint affinal opposition of S [10] and ZSms [12] together with D [11] and ZDms [13] when facing DH [18] and SW [17] creates consanguinity. ZS/ FZS [12] and ZD/FZD [13] occupy a consanguineal status by opposing affines of other consanguines.

Generations G+2/G-2 and Relational Content Beyond the three medial generations, the equations of the marginal generations only partially unify consanguineal and affinal categories by the undetermined categories FF/MF/MMB [1], FM/MM/FFB [2] and SS/SD/DD/DS [14], while skewing of the affinal lines is significantly continued by term MFZ [4] and term WBSS [15]. In this respect, the pattern resembles the Seneca terminology in which the skewing of HP/ HPP marks a distinction in G+2. However, such complications have no effect upon the following ‘message’ of the system:

Figure 4.9 Five-line Omaha terminological structure (male Ego). Figure created by the author.

Omaha Skewing Reconsidered • 79

• The first consanguineal line lacks skewing to provide the vertical axis or the generational depth for the system as a whole. • All of the other four terminological lines in the three medial generations contain only two terms each: one male and one female. Skewing is equivalent to opposing consanguineal and affinal terminological lines, thereby clarifying the system of affinal relations as a whole rather than complicating it. Complications arise only when EuroAmericans define terms as designations of kin-types, i.e. as referring to a very limited number of human individuals. • The general categorical impact of the system is the same as that of Seneca and other Amerindian systems. It excludes a repetition of reciprocal affinal relationships. This means that consanguines must formally be involved in a mutual ‘give and take’ with several lines rather than just one line of affines. It implies a multiplicity of social exchanges or the proliferation of ‘the spirit of the gift’.

Differences for a Female Ego From the perspective of a female Ego, the crucial twist of generating the second consanguineal line is realized by the two separate oppositions of the second affinal line. BS [13] and BD [14] become the second consanguineal line by sharing the same opposition against the spouses of siblings [7], [8], [9] and [10] of the first consanguineal line. Again, different opposites of the same affines become consanguines. Significantly, in the second affinal line term [16] equates FZH, ZH/HB and BDH for the second () reciprocal affinal relations, even though BS [13] and BD [14] represent only G-1 and, as shown above, FZH is an awkward fit. The terms for FZS/FZD [12/13] of a man’s second consanguineal line are shifted as FZS/FZD [11/12] (being equated with S and D) to the first consanguineal line in the terminology of a female Ego to end up in the lower generation of G-1, since their classification must follow the law of reciprocal terms. In both of the gendered perspectives, they are thus being classified as belonging to consanguineal lines, though they are distinguished from siblings.

Conclusion The differences in the terminological scheme for a female Ego do not affect the general structure of the Omaha terminology. This

Figure 4.10 Unified Omaha terminology (female Ego). Figure created by the author.

Omaha Skewing Reconsidered • 81

Figure 4.11 Five-line Omaha terminological structure (female Ego). Figure created by the author.

­ omenclature has been a ‘classic problem’ in anthropology because n scholars have been tied to the European concept of kinship, which is based upon relations between parents and children, or those ‘vertical’ relations determined by physical procreation and birth. Scholars have not seen the generalizations of marriage relations or the ‘­horizontal’ ­relations that only have a secondary and ephemeral status in European kinship. In innumerable anarchic societies in America, Asia and Oceania, marriage is a universal social norm of overwhelming importance and ‘kinship’ is ordered by alliance. Accordingly, relationship terminologies are classified by affinal or ‘horizontal’ relations, while ‘vertical’ relations may have a secondary or subordinate status. In past analyses, major breakthroughs and lasting contributions to knowledge have been achieved when positive marriage rules and the classificatory method of affinal prescription were discovered, differentiated and defended against scores of sceptics. So far, however, the idea of negative marriage rules and their formalization as affinal proscription has not been elaborated. In several variations, the latter concept may equally well constitute the logic of terminologies. All classification based upon affinal relationships must cope with a fundamental contradiction: the inability of sibling units to continue their particular affinal opposition against other sibling units in subsequent generations. The ‘vertical’ dimension is the basic problem and is thus comparable to the ‘horizontal’ problem of European terminologies,

82 • Lewis Henry Morgan’s Comparisons

leading to a downgrading of affinity. Omaha skewing is a classificatory mode to continue affinal opposition of one sibling unit against another one in all generations, but at the same time, a new unit is added with each generation and an old one disappears. Euro-American scholars have since long noticed how different relationship terminologies tend to neglect or ignore any one classificatory principle (e.g. age and sex), while promoting others. Even when Morgan and others discovered that many contemporaries merge the category of a mother’s sister with that of a mother, this way of ignoring collaterality became understandable after some time. But the negation of the generational distinction as a systematic way to clarify affinal oppositions continues to amaze the discipline, unless it is simply waved aside, as if did not exist and as if non-European values were irrelevant. However, Morgan was open to foreign ways of ordering society. He was talking to White Cloud a native Iowa well versed in English. While discussing these relationships he pointed out a boy near us, and remarked that he was his uncle, and the son of his mother’s brother who was also his uncle. (1871: 179n1)

Obviously both White Cloud and Morgan understood that the second uncle was the genitor of the first uncle. For White Cloud, it must have gone without saying that he could never marry the sister of either of his two uncles because then the bride would be – in both cases – the mother of the bridegroom. However, this issue of marriageability or nonmarriageability was not considered by Morgan, or at least was not discussed in his publications. Such an idea of ‘affinity as a value’ (Dumont 1983) was, at that time, too big a step.

Notes 1. Formal preconditions differ basically. 2. ‘Malayan systems’ (Morgan 1871: 448f.) may distinguish sex and generation, but not lineals and collaterals or consanguines and affines (see Chapter six below). 3. See Chapter two above. 4. See Chapter three above. 5. The reference was to Dumont (1961 and 1964). 6. See Chapter three above. 7. Since the Omaha assign clans and moieties patrilaterally, I have chosen this order. 8. German for ‘surrogate’. 9. For details, see the next Chapter five below.

5 Highland Middle Indian Terminologies

The Issue of Form Beyond Europe, classifying relative age is a fairly frequent principle of relationship terminologies, but has received very little attention from analysts. The following will try to show its relevance in a widespread terminological structure that earlier might have been subsumed under ‘Type Iroquois’, or Type B,1 if its specifications of relative age and their logical consequences had not ruled out this pigeonhole. The pattern may also be identified as a ‘semi-complex’ structure of affinal alliance, though again this point may become problematic due to certain definitions in the discipline. Lévi-Strauss had conceived semi-complex structures of kinship as being of the ‘intermediate’ or ‘compromise’ kind, belonging neither to the so-called elementary nor to the complex structures, but forming a ‘bridge’ (1969: xl) between the two. These ideas (1969: xxxvii‒xxxix) were confined to terminologies and marriage rules of the so-called Crow-Omaha ‘type’, which he equated with the semi-complex structure. Clarifying the context, Trautmann and Whiteley (2012: 20‒21) have pointed out more recently that we must keep in mind the two different ways of defining Crow-Omaha: on the one hand, as terminologies containing skewing and, on the other hand, as semi-complex marriage structures. The two specialists note that systems of the latter kind are numerous and that the proper terrain for their analyses ought to be such terminologies, whether or not they contain skewing. Following this lead, I will discuss certain affinal proscriptions in Highland Middle Indian (HMI) terminologies, which include a kind of skewing2 that, like Seneca skewing, is only of marginal importance. The terminological analysis will be separated from, and later briefly compared with, the marriage rules in different societies of this culture area. In the past, Bouez had added two of these Indigenous nomenclatures

84 • Lewis Henry Morgan’s Comparisons

to the semi-complex systems of the volume edited by Héritier-Augé (1990). Like the ones discussed below, they are not of the Crow-Omaha ‘type’, but their terminological structure does have certain commonalities with Type B or Iroquois. The intricate issue is that Trautmann (2012: 2) could not find traces of Type B in South India, and his urgent search for other evidence on the subcontinent (‘is there Iroquois elsewhere in the vicinity?’) did not yield results. The question is important for him, because of his concern with the proximity and transformations of the ‘Dravidian’, Iroquois and Crow-Omaha structures, but again established definitions may become impediments. Following Lounsbury’s classic analyses of Crow-Omaha and Seneca terminologies (1964a, 1964b), Trautmann had identified the ‘same-sex-sibling merging rule’ as one of the criteria of his Iroquois pattern and added as another its particular crossness, which is rather different from ‘Dravidian’ (or Type A) crossness (Trautmann 2012: 35). As a partial response to Trautmann’s search, I will show how the terminological structure of HMIs on the one hand lacks the ‘same-sex-sibling merging rule’, or the general principle underlying the merger of FB3 and F or of MZ and M. On the other hand, it contains certain similarities to Iroquois crossness. The classificatory principle of relative age is comprehensively applied, thereby ruling out the merger of same-sex siblings. The Iroquois terminology, especially its Seneca version, has probably been discussed more than any other in anthropology, and analysts have described its canonical type as well as its generational varieties in great detail4. In comparison: Dravidian is based on the rule of cross cousin marriage, whereas the Iroquois do not have and are not known to have had such a rule. (Trautmann 1987: 241)

The equations ssGC=MBssCC=FZssCC as well as osGC=​MBosCC=​ FZosCC are diagnostic of Iroquois (Trautmann 1981: 87‒8), as is the specification EF≠MB or EM≠FZ. The above-mentioned scholars map the canonical Iroquois structure in figures displaying ‘cross or parallel consanguineal kin only’ (Tjon Sie Fat 1998: 60). Surprisingly, they do not consider Morgan’s reference (1871: 169) to the affinal term Ah-ge͒-ah’-ne-o͒ of Seneca5 that equates HZ=♀BW=WB=♂ZH to convey a pattern of symmetric affinal exchange within G0. Much of the following discussion will relate to this equation representing affinal symmetry (see Figure 5.1). As to age classification, Seneca only distinguishes eB from yB and eZ from yZ, and is thus rather different when compared to HMI. ‘Dravidian’ terminologies of mainstream Hindu South Indians have their own specific style to express intragenerational seniority.

Highland Middle Indian Terminologies • 85

Figure 5.1 Seneca term Ah-ge͒ -ah‘-ne-o͒ indicating affinal symmetry. Figure created by the author.

Trautmann has pointed to the prefixes of a ‘big’ and a ‘little’ father as obligatory distinctions from the ‘father himself’. The relevant structural aspect is the common reciprocal for all three (1981: 37). In contrast, the Indigenous people of Middle India clearly distinguish between these three reciprocals. Also, MyZ and MeZ, or FeB and FyB, as well as the other inbuilt age distinctions are represented by basic terms and not simply by prefixes. In other words, the very reason for Morgan (1871) to initiate his grand comparative project ‒ the difference between Iroquois crossness and English lineality (Trautmann 2012: 31) ‒ is not apparent at first sight in Highland Middle India, where collateral terms (i.e. those for MeZ, MyZ, FyB, FeB, as well as essGC and yssGC respectively) are distinguished from the lineal ones (i.e. those for M and F or for D and S). In the following I hope to show that first impressions may be quite deceptive. The terminological structure to be discussed has little in common with the North Indian assemblage of terms (Dumont 1966; Trautmann 1981: 91‒104) and differs from the ‘Dravidian’ structure in South India (Dumont 1953; Trautmann 1981). Like the latter, it is relativistic, but as the crucial point of difference, it excludes any repetition of a reciprocal affinal relationship between the same terminological lines, in spite of the fact that affinal symmetry is an invariant feature in each of the three medial generations. Implied by the language, the formal effect of this exclusion is, as in Seneca and in Omaha, the dispersal of affinal relations. Affines of the parental generation differ from affines of Ego and those of the filial generation. The latter two also lack any other formally provided connection. Accordingly, HMI terminologies are not characterized by affinal prescription. Their structure invariantly determines

86 • Lewis Henry Morgan’s Comparisons

who is to be avoided, implying at the same time that the – respectively separate – affinal relations of Ego, his or her parents, siblings and children can only be of the symmetric kind. The social rules of marriage differ from the prohibitions of North India6 and also – at least in many cases – from the norms demanding cross-cousin or oblique marriage in the south. Symmetry and seniority, being Dumontian value-ideas (Dumont 1980), are identified in Highland Middle India as the ideological basis of the terminological order and of anarchic social organization on the ground. However, they appear within a variety of empirical arrangements. Even though both terminology and society are structured by the same basic values, the given pattern of the terms is not a reflection of the social organization found in Indigenous villages. The languages offer a unified and condensed structure, while the observed rules of social organization are more variable. I will introduce the latter only after analysing the constellation of categories. As an example, the Kharia7 nomenclature8 will be analysed in detail before its crucial aspects are compared to those of other Indigenous languages of the culture area as ‘the issue of propinquity’.9 Thereafter, I will contrast this structure of the terminologies with the marriage rules of empirical groups and individual speakers. Initially, however, a very short introduction10 on the ethnographic basis and its scant anthropological coverage will be provided.

Ethnography and Ethnographers The highlands of ‘Middle India’11 are not the central part of the subcontinent because they almost extend to the Bay of Bengal12 in the east, while in the west, they continue beyond the Aravalli Hills. At least 100 million people13 belong to the Indigenous population of the subcontinent. A single ethnic unit may be defined as a Scheduled Tribe by the Indian government and may number several million individuals or only a few thousand. Before the colonial regime gradually introduced some state institutions, these Indigenous people were acephalous and organized by multivalent institutions defined by ‘kinship’. Sahlins’ (1968a) criteria of the ‘tribal’ ideal type seem to fit. They imply that production, polity and piety are not separated, but are different functions of the same institutions. Similarly, ‘from one vantage … the tribe presents itself as a pyramid of social groups … a “segmentary system” not simply because it is built of compounded segments, but also because it is only so built’ (Sahlins 1968a: 15).

Highland Middle Indian Terminologies • 87

Viewed from another angle, it ‘is divided into concentric circles of kith and kin: the household in central position, a circle of … kinsmen surrounding it, a wider circle of village relations, and then furthermore the tribal and inter-tribal spheres’ (ibid.). Each sphere, otherwise a level of organization, in this second perspective is a sector of social relations and the latter become ‘increasingly broad and dilute as one moves outward from the familial navel’ (1968a: 15/16). To this day, Sahlins’ ‘familial mode of production’ (1968a: 75) is empirical reality in many subregions of the highlands. In colonial times, administrators and missionaries described Indigenous India at length. Some dedicated outsiders14 devoted much of their lives to the study of these societies. Whereas anthropology had been taught in metropolitan India much earlier, after independence it was also established in the more remote provinces, which led to numerous publications on Indigenous societies in the respective neighbourhood.15 The collection and linguistic comparison of kinship terms by Bhattacharya (1970) as well as the findings on social organization in the older and more recent literature have been evaluated by Parkin (1992) for speakers of different Munda languages.16 Earlier, Bouez had himself conducted field research in the culture area, leading to several works (1975, 1985, and 1990) on Ho and Santal affinity, but without a unified matrix of the terminological whole. Like the publications resulting from my own ethnographic work,17 Bouez’s volumes were marred by some assumptions that can no longer be supported.

List and Lines The Kharia terminology on the list below is applied by about half a million people living in the southern border region of Jharkhand province and the adjoining parts of Odisha. In this book, the given pattern of the Kharia terms is meant to represent all nomenclatures of the culture area and will be ordered in the subsequent matrix for a ‘classificatory reading’ (Allen 1998: 315). Within any one of the boxes, a number (in square brackets) will lead to the corresponding term of the list, while the added specification is to convey an initial impression of its relational meaning. The latter can be fully attained by an examination of the other specifications on the list. As a whole, the matrix will present two consanguineal lines being interrelated with four affinal lines. So far, I have not come across such a terminological structure elsewhere and I am not aware of earlier attempts that, like the present one, offer a matrix to represent the

88 • Lewis Henry Morgan’s Comparisons

Table 5.1  List of Kharia terms.  1.  2.  3.  4.  5.  6.  7.  8.  9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39.

aja: PF, PFB yaya: PM, PMZ mama: MB, FZH mami: FZ, MBW kaka: FyB, MyZH kaki: MyZ, FyBW mij: MeZ, FeBW, ♀yZD, ♂yBD, WyZD, HyBD, ♀MBDyD, ♀FZDyD, ♂MBSyD, ♂FZSeD dada: FeB, MeZH, ♂yBS, ♀yZS, WyZS, HyBS, ♂MBSyS, ♂FZSyS, ♀MBDyS, ♀FZDyS apa: F, MH ma: M, FW bhai: MBS, FZS boi: MBD, FZD baing: yB, FyBS, MyZS bahining: yZ, FyBD, MyZD dai: eZ, MeZD, FeBD dada: eB, FeBS, MeZS, WeZH kulumra: eBW kimin: yBW, SW bhati: eZH araming: yZH, DH bhatiji: ♂eBD, ♀eZD, WeZD, HeBD, ♀MBDyD, ♀FZDyD, ♂MBSyD, ♂FZSyD bhatija, ♂eBS, ♀eZS, WeZS, HeBS, ♂MBSyS, ♂FZSyS, ♀MBDyS, ♀FZDyS betain: S beting: D bhagni: ♂ZD, ♀BD, HZD, WBD, ♂MBDD, ♂FZDD, ♀MBSD, ♀FZSD bhagna: ♂ZS, ♀BS, HZS, WBS, ♂MBDS, ♂FZDS, ♀MBSS, ♀FZSS bokudi: DD, SD bokuda: SS, DS kanrai: W ajiker: WeZ, HeZ boksel: WyZ, HyZ baotang: WeB, HeB kulumday: WeBW, HeBW boker: WyB, HyB bokdera: WyBW, HyBW somdhi: SWF, DHF, SWM, DHM sasuro: WF, HF kinkaraing: HM, WM kendor: H

­structure of all Indigenous Middle Indian terminologies. Similarly, I have not found a single and unified matrix of all Kharia terms. Nor, with the exception of Deeney’s most important contribution,18 have I seen a structural representation of all relationship terms of a single Middle Indian language. The ‘classificatory reading’ of my matrix

Figure 5.2  Kharia terminology collected in 1982 in Kolebira village, Gumla district and Jharkhand province. Figure created by the author.

90 • Lewis Henry Morgan’s Comparisons

­ ifferentiates six terminological lines. For readers unfamiliar with such d a matrix, added double arrows () indicate the invariant reciprocal affinal relations between the categories depicted in the boxes. The emerging pattern may be interpreted as if it were established for marriage exchanges between people.

Analysis of the Kharia Terminology The following analysis will ‘focus on the equations and discriminations’ made by the lexical forms so that the ‘type of terminology can be characterized by the types of equation and discrimination it makes (Allen 1998: 317; emphasis in original). Thus, term [36] CEP of Kharia ignores the sex19 difference, and terms [18] yBW, SW and [20] yZH, DH the generation difference, as do the self-reciprocal terms [7] MeZ, FeBW, ♀yZD, ♂yBD, etc. and [8] FeB, MeZH, ♀yZS, ♂yBS, etc. Irrespective of these partial aspects, which are frequent but not universal in the Indigenous Middle Indian languages, the system will now be analysed in a generation-by-generation inspection of the matrix. Frame To begin with, each of the two external generations of the matrix (G+2, G-2) contains only two unifying categories: one male and one female. Thus, they are not distinguished as consanguines and affines, but represent the undetermined frame of the terminological core in the three medial generations. A small minority of nomenclatures (for example, the Gondi terminology) contain more terms either on only one or on both of these margins specifying maternal and paternal grandparents, or else DC in contrast to SC, but most of the nomenclatures reduce both G+2 and G-2, as shown above. The three central generations contain six terminological lines: two consanguineal and four affinal. The fact that the affinal lines are only interconnected by the consanguineal lines, but not by any other link, defines the terminological whole as a semi-complex or nonprescriptive structure. Odd Generations G+1 and G-1 In G+1, a symmetric affinal relationship invariantly connects the first affinal line with both consanguineal lines. The terms for M and MB, together with those for FZ and F, stand for this exchange, though PessG

Highland Middle Indian Terminologies • 91

and PyssG are equally involved on the consanguineal side. By their joint opposition within the same affinal relationship, the latter two obtain the same relational status as the categories of parents. Their respective affinal counterparts PosG determine the inclusiveness on the consanguineal side and the symmetry of the relationship. Formally, Ego’s M and F are consanguines, while FZ and MB are affines (see Dumont 1983: 75‒79). I have emphasized the symmetric affinal relationship within G+1 of the matrix by inserting a double arrow named ‘Reciprocal affinal relations’ to indicate this connection. The same kind of symbol stands for the relations within G-1, where D and S, and also essGC and yssGC, are related to SW and DH in a symmetric pattern. Again, the latter two, as the affinal counterparts of the former, determine the inclusiveness and symmetry of the relationship. The most meaningful aspect in these two formal documents of affinal reciprocity is the discrimination of EP [37, 38], or their exclusion from the ‘1. Reciprocal affinal relations’ in the same way as the two N [25, 26] are not involved in the ‘4. Reciprocal affinal relations’. Like CEP [36] and PosGC [11, 12], they have no part in the above-mentioned affinal symmetry of their respective generation. As ‘third parties’, they either indicate the unavoidable beginning of a new affinal relationship in the subsequent generation (i.e. EP and CEP) or the end of the affinal relationship in the previous generation (i.e. PosGC and N). Thus, EP, CEP, PosGC and N are diagnostic of the semi-complex or nonprescriptive structure of the whole, since they imply that each affinal relationship of the terminological whole is subsequently discontinued. Terminations of Affinal Lines What happens to an affinal line in the generation below that of the invariant ‘Reciprocal affinal relations’? In the first affinal line, cross-cousins are, as in Iroquois (Trautmann 1981: 86‒87; Trautmann and Whiteley 2012: 2), conceived as if they had the status of siblings. Formally, their affinal character is lost. This becomes apparent from the fact that their children (PosGCC) are terminologically equated with the children of siblings (GC), either as niblings or as children. However, cross-cousins themselves are not (as in the Cheyenne variation) equated with siblings. Their continued discrimination is implied, since their relative age is not distinguished, whereas siblings only appear as either elder or younger ones. Briefly, the children resulting from the ‘1. Reciprocal affinal relations’ in G+1 are not affines. Those of F and M are, like all consanguines, distinguished by relative age, whereas those of MB and FZ are not. Thus, these two pairs of children continue to be distinguished. The

92 • Lewis Henry Morgan’s Comparisons

latter belongs to an affinal line, though male and female PosGC are neither affines of siblings as they are in ‘Dravidian’ systems) nor equated with siblings. For the time being, their status is betwixt and between,20 whereas their children (PosGCC) are finally ‘consanguinized’. In G0, the third affinal line offers equations according to sex and relative age (WeZ=HeZ, HeB=WeB, HyB=WyB and WyZ=HyZ), but those of the respective children in G-1 follow another principle, since all WZCh and all HBCh become ssGCh of either the elder or the younger specification. Their opposites, WBCh and HZCh, become N without age distinction. In G0, same-sex and same relative age cause the equations, while the sex-difference of the sibling link (HB and WB versus WZ and HZ) is ignored. However, below them in G-1, the sex-sameness or sex-otherness of the sibling link (HBCh and WZCh versus WBCh and HZCh) determines the equations of the respective ssGCh and N. Then the status of affinal sibling-pairs comprising this link (WZ and HB versus WB and HZ) is determined in the same way as the status of consanguineal sibling-pairs (Ego plus ssG versus Ego plus osG), so that the affinal status of third affinal line is finally terminated in G-1. Central Generation G0 Two named arrows represent the invariant second and third Reciprocal affinal relations of G0. No. 2 opposes siblings’ spouses and siblings. Symmetrically, it connects the second affinal line with the first consanguineal line. At the same time, term CEP of G0 is not involved in any of the ‘Reciprocal affinal relations’. Its discrimination implies the distinction between CE and PosGCC, or the fact that each new generation must contain new affinal relationships of the consanguineal lines. Such affinal dispersal coincides with the crucial discrimination of G0, which will now be discussed. ZH is not – as in Type B (Iroquois) – equated with WB, nor is BW equated with HZ. Normally, such specifications would exclude symmetric exchange. Given these obviously problematic oppositions, the relations of all categories in the three medial generations will now have to be reviewed. Review Terminologically, M is not equated with MZ. Three separate basic terms stand for M, MeZ and MyZ respectively. In the very same way, the term for F is distinguished from those for FeB or FyB, even though no samesex collaterals exist for FZ and MB. This shows that the dimension of collaterality is not the relevant aspect of the classification; rather, its sig-

Highland Middle Indian Terminologies • 93

nificance lies in the distinction of relative age in each of the three medial generations of the first consanguineal line. It is the special feature of this line. The distinction of lineals and collaterals is nothing other than an unavoidable side-effect of discriminating age in what is basically a classificatory system in the relativistic sense of Morgan’s Systems of Consanguinity and Affinity of the Human Family (1871: 12). The ‘2. Reciprocal affinal relations’ in Go contain the difference of relative age in the first consanguineal line and in the second affinal line. Clearly, eBW is the affinal opposite of eZ, as is yZH of yB. This distinction obviously indicates that an elder brother cannot be the husband of the younger brother’s wife, nor can a younger sister be the wife of the elder sister’s husband. The unavoidable outcome of this discrimination between seniors and juniors is that Ego’s affinal relations must be segregated from the affinal relations of Ego’s siblings, since Ego will always be either a senior or a junior sibling. Accordingly, neither Ego’s spouse nor the siblings of Ego’s spouse can be terminologically equated with the spouses of Ego’s siblings. Thus, the classificatory principle of relative age is not simply a distinction of siblings; in fact, it emphasizes the discrimination of the respective siblings’ relations. Had the affinal terms implied that Ego’s sibling was the spouse of Ego’s spouse’s sibling, they would have constituted sibling equivalence, a value that is formally excluded by the distinction of age. In systems specifying HZ≠BW and WB≠ZH, such nonequations may point to an affinal prescription of the asymmetric kind. However, in Highland Middle India, a separate second consanguineal line rules out such a possibility. In G0, this second line of consanguines contains the spouses of the siblings of Ego’s spouse. Their consanguineal status is, in the Kharia language as in others, emphasized by the equation HeZH=eB, which has caused me to graphically represent eZ and HeZH as siblings by linking their two boxes in the matrix. At the same time, the basic terms for EeBW and EyBW are not equated with siblings. In other highland languages too, such additional consanguineal terms differ from those of the first consanguineal line. They specifically refer to the spouses of the siblings of Ego’s spouse, as surrogate-consanguines, so to speak, stabilizing the paradox that Ego is part of the ‘3. Reciprocal affinal relations’ without simultaneously involving his or her elder or younger siblings of the first consanguineal line whose spouses in the second affinal line have nothing in common with Ego’s spouse in the third affinal line. To be more precise, formally, the symmetric ‘return gift’ for W or H is EGE. The status of the latter is undoubtedly siblinglike, though at the same time, it is explicitly distinct from siblings of the first consanguineal line.

94 • Lewis Henry Morgan’s Comparisons

Figure 5.3 Doubling of encompassed affinal symmetry in G0 of HMI terminologies. Figure created by the author.

Accordingly, the second symmetric relationship in G0 (the ‘3. Reciprocal affinal relations’ of the whole) is as invariant as the first. Such a doubling of the affinal exchange relationship is the unique feature of HMI. Ego’s spouse is not a sibling of a spouse of Ego’s siblings, and yet symmetry is retained in both affinal relationships (see Figure 5.3). Thus, Ego’s affinal relations are not only discriminated from those of F and M and those of S and D, but also from those of B and Z. Summary In the Kharia terminological structure and all others of HMI terminologies, P and PssG jointly oppose PosG, while EP are excluded from this relationship. Likewise N, or osGC, have no part in the opposition of C and ssGC jointly facing DH and SW. PosGC are discriminated from both G and E, while the specification of CEP implies that SW and DH are not children of PosGC. Though neither BW and WZ nor ZH and HB are equated, surrogate consanguines determine the doubling (see Figure 5.3) of the Seneca relationship (see Figure 5.1) of encompassed affinal symmetry. In the event that the four encompassing pairs of categories (EP, CEP, PosGC and N) that structure the system are ignored for a moment, i.e. those indicating the dispersal of affinal relations within the unrelated cosmos of marriageable positions, the remaining encompassed four, separated by sex in Figure 5.4, reveal a tetradic structure of symmetric affinal relations rather than positions. They are encompassed by the other four pairs standing for new or external relations on each of the three medial generations.

Highland Middle Indian Terminologies • 95

Figure 5.4 Encompassed tetradic structure of affinal relations. Figure created by the author.

The Issue of Propinquity Compared to the ‘Dravidian’ system, or diachronic affinal symmetry between two lines only, classifying relative age in Highland Middle India reveals synchronic affinal symmetry as a feature encompassed by an overall structure comprising six lines. At the same time, this pattern formally excludes diachronic affinity or consecutive affinal relations between the same two lines. Such affinal dispersal is also implied by the data on the Ho (Deeney 1975: 106‒7). Hardenberg (2018: 171‒73) has found these features among the Dongria Kond and I have recorded them among several other Kond units21 and also among the Bhumij, the Sora, the Bondo, the Parenga and the Didayi.22 However, among the Bhil, the two terms for PosG are equated with those for EP (Naik 1956: 58‒60), and I have also found the same merger among the Gadaba. But the latter lack the ‘Dravidian’-style equation of reciprocals, since osGC differ from CE. The Gadaba also discriminate the terms for PosGC from those for E or EG. Thus, their terminology lacks the ‘Dravidian’ affinal prescription.

96 • Lewis Henry Morgan’s Comparisons

In the vast area of the central provinces, the terminology of the widespread Dravidian Gondi language has been meticulously collected by: • Christoph and Elizabeth von Fürer-Haimendorf (1979: 351‒53) for the Gonds of Telangana province; • M.P. Buradkar (1940: 150‒62) for those in Maharashtra and Madhya Pradesh; • Stephen Fuchs (1960) for Gonds and Baiga of the northern highlands of Chattisgarh. Added to this must also be Chris Gregory’s informed study (2009) focusing on the manual labourers, craftspeople and petty commercial agents and their Halbi language, which (as others in other regions) is an Indo-European lingua franca in the entire south of the Chattisgarh province. His article contains the extremely important information that is equally applicable for similar languages (Desia and Sadri) in western Odisha and Jharkhand: if we compare the Halbi kinship system with that of Gondi we find that the only difference is the lexicon; from the semantic point of view they are identical. (Gregory 2009: 68)

The terminologies of these four contributions clearly indicate the allimportant structural contrast between the six-line HMI and the two-line ‘Dravidian’ system. It is articulated by the following nonequations in the former, which appear as equations in the latter: GE≠EG, MB/FZ≠EF/ EM, MBC/FZC≠E/GE, MBC/FZC≠CEP, ossC≠CE. The same structure is found in the Dravidian language of the people usually known as Oraon.23 Both in Gondi and in Halbi, four terms are found in G+2, whereas this grandparental generation is usually – but certainly not always – reduced to only two terms in many Indigenous languages of Odisha and Jharkhand. However, this difference is structurally insignificant. Among the Juang MBC are – irrespective of their biological age – equated with elder siblings, while FZC are always equated with younger siblings (McDougal 1964: 447‒48). As shown in the matrix (Figure 5.2) above, the Kharia terminology and most others contain senior and junior siblings, but no age distinction among the specifications for cross-cousins. It follows that like Kharia, most terminologies of the culture area distinguish siblings from sibling-like cross-cousins by classifying relative age in one way or another.

Highland Middle Indian Terminologies • 97

Figure 5.5 Hypothetical symmetric affinal relationships in the Juang and Ho terminologies. Figure created by the author.

All HMI terminologies contain separate terms for EGE carrying a consanguineal status. Thus, Eva Reichel, after long-term field research, told me that the Ho equate eB and HeZH as bau. Additionally, however, saragi stands for WZH without any equation and without any distinction of relative age. Among the Bondo, I found the term bailo denoting WyZH and HyZH, having nothing in common with mé, which denotes yB. In the same manner, iong stands for WyBW and HyBW, whereas kui is the term for yZ. Simultaneously, however, eB and WeZH or HeZH are equated with mang. Clearly, the non-equation WB≠ZH or HZ≠BW is the crux of the HMI systems. Seemingly, the Juang offer an exception with sango or WeB=yZH. Their bao stands for eZH, whereas inibao (=junior bao) is WyB (McDougal (1964: 447‒48). Reichel has discussed these points with me when reporting the Ho term teya (WyB/eZH). I think that the difference between bao and inibao offers the clue here by indicating the added qualification of seniority as the factor excluding affinal symmetry (see Figure 5.5). Had sango implied symmetry, the distance of seniority between WeB and his wife would be doubled. Similarly, in the case of teya, WyB would not only be a junior of his wife; her formal age-distance would be multiplied by two. All of this would formally contradict the general implication of the given terminologies in the culture area that matches seniors with seniors only, and juniors with juniors. To summarize: with minor deviations, the structural features of the Kharia nomenclature are confirmed as those of all terminologies in the culture area, irrespective of the language family assigned to their speakers. This HMI pattern shares the crossness of Type B (or Iroquois), but lacks the same-sex-sibling merging rule, since it discriminates lineal and collateral terms in a cohesive manner. Reduced further, this overall HMI pattern may be represented as in Figure 5.6 below.

98 • Lewis Henry Morgan’s Comparisons

Figure 5.6 HMI terminological structure. Figure created by the author.

Practice: Seniority All classification of social groups and collective categories in Highland Middle India employs the principle of seniority as the crucial distinction.24 For example, the Kharia are said to be the ‘seniors’ of the Munda, who in turn rank the neighbouring Oraon as their ‘juniors’ or as latecomers in the region. The seniority principle is applied everywhere to distinguish adjacent or interspersed communities and subdivide major and minor units within each of them. Even within a village, local descent groups, ritual positions and, last but not least, ritual food and alcoholic drink consumed during rituals are separated as the ‘senior’ and the ‘junior’ kind. The distinction by seniority, which has nothing to do with power relations, is omnipresent in the Indigenous worldview.

Practice: Affinity Clan membership is patrilaterally assigned, but clans differ from those of Indian upper castes (see Parry 1979: 132) or from those of the Nuer (Evans-Pritchard 1934), in that they lack a concept of lineal ramification involving elaborate pedigrees. Though its representatives may perform ritual functions within a village, a clan is primarily a sociocentric category of affinal exchange and never a corporate group involved in any kind of action.25 My rather extended comparative enquiries on this topic show that marriage relations are organized by an agnatic local descent group of a certain clan. Either such an LDG or a subdivision of it is the decisive unit of social action in Indigenous politics. Generally, the pressure to

Highland Middle Indian Terminologies • 99

repeat a marriage is a common feature, but the tendency to delay such a repetition and disperse alliances is equally strong in other communities. Either one of these rules is applied, though people are conscious of both options. Similarly, Whiteley, writing on other marriage systems, has identified: a characteristic tension in semi-complex systems to both disperse alliance among a more encompassing social pool (than in an elementary system) and to reproduce existing alliances within a partially closed social space. (2016: 6)

General Conclusions The perplexity arising from the particular classificatory twists in HMI terminologies provokes some general considerations, since this structure is found in a contemporary sociocultural milieu that seems to contain many features resembling Morgan’s Native North America, though the demographic situation is quite different and to this day vast subregions in the culture area are only superficially controlled by the state. The common terminological structure of HMI is basically different from the arrangement of a descriptive nomenclature and is not of the prescriptive kind. At the same time, a definite ‘family resemblance’ with Seneca and Omaha can hardly be denied. The two Native North American terminologies as well as HMI, each of them by a different patterning, are classificatory systems that formally exclude a repetition of affinal relations in subsequent generations. Accordingly, all three could be subsumed under the label of ‘semi-complex’ systems, but this label does not appear to be adequate, since it would indicate (and had been meant to indicate) an intermediate, or compromise, or imperfect condition due to the Lévi-Straussian dictum that I am unable to accept. Their status must be reconsidered. Decades ago, Needham demanded the analysis of relationship systems as the first step in a research venture because ‘terminologies, especially prescriptive ones, are classifications by which people order their social lives’ (1973: 174). He defined a prescriptive terminology as one that is ‘constituted by the regularity of a constant relation that articulates lines and categories’ (Needham 1971: 32). The relation: serves as a module in the constitution of the system. It is this invariant factor’ in cases when ‘the system of social classification can be viewed as being established on a prescription to marry a woman of a certain category. (Needham 1973: 175, emphasis in original).

100 • Lewis Henry Morgan’s Comparisons

HMI cannot be viewed as being established on a prescription to marry a woman of a certain category. On the contrary, the structure implies a proscription to marry such a woman, i.e. one identified by the given categories. The terminologies of the culture area can be viewed as being established to exclude given categories of women from marriage by a constant relation that articulates lines and categories. This relation serves as a module by providing the following: • A pattern of symmetric affinal exchange encompassed in each of the three medial generations. This pattern is doubled in G0. • Additional categories on each of these levels (EP on G+1, CEP and PosGC on G0, and osGC on G-1), which invariantly imply new external affinal ties with the change of generation, or the distinction of Ego’s affinal relations from those in previous or subsequent generations. Though native English speakers should select the proper expression for such a purely negative impact, I would suggest the label ‘proscriptive terminology’ for a system of social classification that can be viewed as being established on a proscription to marry a woman of diachronically interrelated categories. Apart from ‘Malayan systems’, I believe that all of Morgan’s classificatory systems are equivalent to those ‘constituted by the regularity of a constant relation that articulates lines and categories’ (Needham 1971: 32) and should be subdivided into prescriptive and proscriptive terminologies. By adopting these two subclasses, we would avoid the unfounded and misleading evolutionist tinge in the sequence suggested by Lévi-Strauss, which ranges from elementary via semi-complex to complex structures. We would also join what belongs together. In this new context, the relationship of affinal symmetry – being repeated between the same two terminological lines from one generation to the next – is diagnostic of prescriptive ‘Dravidian’ (or Type A) terminologies as are found in South India, whereas the relationship of affinal symmetry not being repeated between the same two terminological lines from one generation to the next characterizes the proscriptive terminologies of Seneca, Omaha and also of HMI.

Notes  1. Trautmann and Barnes (1998: 30) give good reasons for avoiding culturally specific labels.

Highland Middle Indian Terminologies • 101

 2. The HMI skewing involves junior affines: yZH=DH and yBW=SW.  3. Along with the usual kinship notations, this chapter will introduce for the rest of the book P=parent, C=child, G=sibling, E=spouse, XC=cross-cousin, //C=parallel cousin, N=(cross-nephew, cross-niece)=nibling, ♂=male speakers, ♀=female speakers, ss = same sex, os = other sex.  4. See Trautmann and Barnes (1998: 31); Tjon Sie Fat (1998:70‒72).  5. See Chapter three above.  6. My initial comparative article on the subject (Pfeffer 1983: 115) referred to Tiemann (1970: 171f.). Earlier, I had personally observed these rules (Pfeffer 1970: 83) during my fieldwork. I have also referred to Blunt (1931: 60‒62) as probably being the first to mention them.  7. Some 500 000 Kharia live in the southeast of Jharkhand and the northwest of Odisha province as plough-cultivators, being sometimes confused with the hunter-gatherer Kheria on the move in the same border region.  8. The list supplied by Roy and Roy (1937: 148‒54) contains more terms taken from Hindi. My terms were collected in a Kharia village, before Dr and Mrs Nabor Soreng, as native speakers, ultimately confirmed it.  9. Juxtaposing issues of form and propinquity is adopted from Godelier, Trautmann and Tjon Sie Fat (1998: 6). 10. For a more elaborate account, see Pfeffer (2014). 11. The designation is adopted from Elwin (1951). 12. Only the small flat coastal strip separates the highlands from the ocean. 13. If the dimensions are known, figures can only be estimated, since the Census of India classifies some Indigenous people as ‘Scheduled Tribes’, others as ‘Scheduled Castes’ and yet others as ‘Other Backward Classes’. The latter two categories mainly (but not only) include Hindu lowlanders displaying basically different cultural features. 14. Prominent examples include the Bengali lawyer S.C. Roy, the Oxford theologian V. Elwin, a self-styled ‘philanthropologist’, and the Austrian aristocrat C. von Fürer-Haimendorf, who in the 1950s became a professor of cultural anthropology at SOAS in London. 15. Indian anthropologists conduct surveys, but do not engage in longterm participant observation in Indigenous villages. One reason for this avoidance may be the fact that villagers eat and store beef in their houses. 16. The majority of the Indigenous people speak Dravidian languages; the others speak those of the Munda and the Indo-European families. Thus, the junior Gadaba speak a Dravidian language, their seniors a Munda language, and all tribes in the region communicate through an Indo-European lingua franca. A single terminology often combines the lexica of all three linguistic families. 17. Though I have retained some ideas of an earlier contribution (Pfeffer 2004), I hereby altogether withdraw it. 18. Deeney, a missionary cum philologist, had spent more than fifty years with the rural Ho. On his own, he had devised a highly coherent, correct and informative diagram (1975: 128‒29) of all Ho relationship terms, which I later passed on to kinship specialists of the region. For their own reasons,

102 • Lewis Henry Morgan’s Comparisons

they have never mentioned his work. Reichel (2018) offers an elaborate and proper recognition of Deeney’s most valuable research. 19. This word ‘when used in the context of a kinship terminology … is restricted to a category of classification [as] a more apt descriptor than “gender”’. For the entire argument, see Barraud (2015: 229). 20. For the Dongria Kond buffalo sacrifice (a total social fact), Hardenberg (2018: 412‒608) describes the ritual status of ‘protectors’ (mudrenga), who are nonmarriageable members of affinal clans, as being neither affinal sacrificers nor consanguineal sacrifiers. 21. However, in G0 of the Kuttia Kond terminology, I found what Tjon Sie Fat (1998: 76) calls ‘variant 2’ of Iroquois or ‘merging of all first and second cousins’ with siblings. 22. Some of these names are applied by the state and most anthropologists, but differ from the Indigenous designations. I have chosen them all the same to facilitate cross-checking with other accounts. 23. The outstanding authority on Kurux, the language of the very large Scheduled Tribe also known as Oraon, is Martin Pfeiffer (2018). He has kindly supplied me with the data on the Kurux-terminology he had found in Grignard (1924) and also in Kobayashi and Tirkey (2017). They reveal – not just by the distinctive nonequations of HMI – the same semantic pattern as Gondi and Halbi. 24. For details, see Chapter 8 below. 25. For details, see Chapter 8 below.

6 Schneider, Relatedness, ‘Malayan’ and a General Comparison

Schneider Over many decades Morgan’s Systems of Consanguinity and Affinity of the Human Family had, even among his most outspoken critics, initiated extended comparative research on terminologies of either kinship or relationship, before David M. Schneider’s book (1984) contributed to the sharp decline of such studies and generally of all anthropological research on ‘kinship’.1 For most of his distinguished career, he had asked ‘what is kinship all about?’ (1972) and at times supplied answers with some positive content before his final and outright negative conclusion. ‘Kinship’, for him, was not structure, but ongoing work of culture. People were to be studied as actors when constructing, perhaps doubting, abandoning and applying ‘kinship’ in words and behaviour, or when following certain general concepts of a culture that might, for example, include particular wife‒husband or mother‒daughter relationships, to be discovered in each of the given ethnographic cases. Process rather than pattern was to be understood by the research effort. Schneider’s approach has been examined at length in Chapter 4 of Adam Kuper’s book Culture (1999), and his life and work have been recalled in the equally comprehensive and critical work edited by Richard Feinberg and Martin Ottenheimer (2002). The following will rely upon these scholarly contributions and try to point to so far unmentioned ethnographic involvements that may have influenced Schneider’s work. By 1970, Schneider tried to remove ‘kinship’ studies from the anthropological agenda and most members of the discipline, especially those who had never wanted to learn ‘what kinship is all about’, were happy to follow this course. Finally, they were able to get rid of the unloved analytical exercises and the esoteric technical debates that, in the history of the discipline, had annoyed many excellently trained

104 • Lewis Henry Morgan’s Comparisons

a­ nthropologists. In a very limited number of patterns, Morgan had uncovered the existence of globally distributed formal modes of classifying ‘kinship’, whereas Schneider was out to annihilate the formal character of any observation related to it and to remain within an empirical context. The diversity of observed behaviour was to be examined, as it was indicating variable concepts and practices of ‘kinship’ within specific cultures. A short sketch of Schneider’s progressively more pronounced stand may illustrate this summary. Schneider’s earlier work had a bent towards the then popular sociological role theory. Ethnographically, he engaged himself on the tiny island of Truk in Micronesia. As a result of this research, he announced that, in contrast to descent, kinship ‘defines a number of statuses and their interrelationships according to a variety of rules or principles and distinguishes kinsmen from non-kin’ (Schneider 1961: 2).2 In his discussion of the Trukese terminology, Schneider mentioned ‘Hawaiian or generation type’ (1961: 222‒25) without further comment in a statement that was entirely adopted from the work of Ward Goodenough (1951: 94‒96). In other words: Schneider referred to the formal aspects of the terminology, but declined to analyse it or contribute a remark of his own. Later (Schneider 1965a), he severely criticized the entire direction of ‘componential analysis’ developed by his friend Goodenough and remained an outspoken critic of ‘ethnoscience’, ‘cognitive anthropology’ and similar versions of formal analysis popular at that time (see Fogelson 2002). Schneider had achieved lasting publicity due to an earlier publication. In response to Lévi-Strauss (1949) and the very different kind of formalism in the French school founded by Durkheim and Mauss, he published a small book together with sociologist George C. Homans (1955), in which they decided that Lévi-Strauss had presented the final causes of the worldwide distribution of matrilateral cross-cousin marriage, whereas they were proposing the efficient causes of the (compared to others) far more frequent occurrence of this kind of union. They simply did not take into account the Durkheimian sociological context of Lévi-Strauss’ contribution, but rather based their explanatory argument on a psychological ‘extension of sentiments’ thesis that went as follows: because most societies were said to be patrilineal, the role of the father was associated (as Homans and Schneider thought) with one demanding respect and distance from his son, whereas the mother was considered to be emotionally close to a young man and, by extension, so would be her brother who, accordingly, would rather give his daughter in marriage to the sister’s son than to anyone else.3

Schneider, Relatedness, ‘Malayan’ and a General Comparison • 105

At that time, this short and clear text in response to the magnum opus of Lévi-Strauss, composed in a rather different style and based upon a kind of thinking that was diametrically opposed to commonsensical ‘explanations’, may have been welcomed on many undergraduate courses. However, its psychologism and individualism were, quite apart from certain basic technical defects,4 untenable in the longer run. Needham (1962) criticized the booklet in detail, whereupon Schneider published a direct attack against all ‘kinship’ studies conducted in the context of British social anthropology. The critique consisted of jibes against both the school of Africanists favouring ‘descent theory’ and against others specializing on Asia, Oceania and South America when applying their ‘alliance theory’. Both approaches were out to understand formal social structure, even if this word did not have the same meaning for the two schools. In this article, Schneider also accused Needham of ‘gross manipulations’ of ethnographic data (1965b: 79), though without providing any evidence, either then or later. The climax of Schneider’s scholarly reputation was reached after a long-term empirical research project on American Kinship (1968) had been completed. He published the results as ‘a cultural account’, or by working out the cultural symbols of ‘kinship’ and their specific applications in the United States. Though his project had employed refined methods to find out what was understood as ‘kinship’ by the people of the United States, a close colleague in Chicago suggests that Schneider’s account did not refer: to the thousands of pages of field data collected by his assistants and carefully distilled and summarized by his wife, Addy. She accused him of dishonesty and burnt her compilations. He justifies his procedure by arguing that it would be equally fraudulent to bolster conclusions already formulated with selected supporting materials. Indeed, there is honesty in his dishonesty, but one is left wondering, why he went to the trouble of mounting a massive research project if he intended to ignore the results of that research. (Fogelson 2002: 36)

A formal analysis of a terminology was not part of this exercise, whereas the meaning of several terms and concepts in use, or the desire to obtain a certain bride (rather than another), or the collectives related in one way or the other were examined in detail. ‘Kinship’ meant for Schneider what the majority of the members belonging to a given culture (such as that of the United States) had in mind when talking about the subject. Non-American concepts might differ greatly from those of his homeland and, accordingly, the author was not tied to the idea of Euro-American kinship.

106 • Lewis Henry Morgan’s Comparisons

By that time, any universal definition of the domain was out of the question for Schneider. Four years later, his dictum ‘“Kinship” … is a non-subject’ (1972: 59) no longer surprised anyone. The statement was part of an article in a small book in memory of Systems and thus did not really attract a great deal of attention, but in 1984, when Schneider’s more elaborate and much publicized ‘critique’ came out, anthropologists who had been indifferent or adverse to the arid debates of ‘kinship’ specialists were relieved. In many anthropology departments of German-speaking countries, where ethnological traditions had always avoided ‘kinship’ and, of course, discussions of either ‘descent theory’ or ‘alliance theory’, Schneider became the apostle of a new anti-kinship creed banning the subject from the discipline once and for all. Schneider had, over the course of a few decades, realized that the Radcliffe-Brownian categories (‘patrilineal societies’) of his early days could not be maintained for long. For his own reasons, he was unwilling to adopt the idea that physical relationships, such as those through procreation, birth, and relative age, were transformed all over the world and figured in one way or another as formal terminological schemes of social classification. He was also averse to the idea that these data could be analysed comparatively, even after those who had applied the terms had died or had adopted the dominant European concepts. Instead, Schneider’s empirical research was meant to lead to a ‘cultural account’ on each ethnographic case in its own right, while denying any relevance of what is called kinship terminology by some or relationship terminology by others. Earlier, Morgan’s ideas in Systems had been repudiated in the discipline, because Kroeber (1909) had somewhat arrogantly published a harsh if indirect critique5 that caused most U.S. anthropologists to reunite classificatory and descriptive terminologies as a universal way of designating genealogical kin. While Morgan had shown how classificatory terminologies could formally regulate ‘gentile societies’ (1877: 60) without rulers and without a special apparatus to legitimize such an order, Kroeber had confined his analyses to a universal practice of applying terms to genealogical kin according to certain general principles. Schneider, however, finally rejected any such universalism and biologism along with any other criteria of kinship. On the other hand, the critic was willing to work on specific classificatory tools as incomparable creations of given cultures, such as generally conceived notions classifying those of common blood and bones, milk and semen, or seed and soil,6 or other substances, i.e. one binary combination or the other, which went together with a code to refer to

Schneider, Relatedness, ‘Malayan’ and a General Comparison • 107

such discriminations. But Schneider was not prepared to admit that relationship terminologies could work as transcultural and, at the same time, more refined tools of this kind, whether or not they were applied side by side with the more general classificatory measures within a particular culture area. Briefly, Schneider rejected terminological differentiation as anthropos’ achievement to classify the social. He was interested in particular people, since his impetus was meant to change the focus of ‘kinship’ studies ‘from structure to practice and from practice to discourse’ (Carsten 2000: 2). If his early ideas on certain kinship sentiments within a certain social framework of a worldwide distribution could not survive, Schneider still retained his focus on individuals in a quantitatively measurable context, as they were executing particular actions and articulating empirically accessible opinions on relationships. He disqualified patterns of supra-individual social values as identified in a limited number of formal and comparable structures providing the ideological basis of the social process.

Relatedness Some fifteen years after the ‘critique’, Janet Carsten made an effort to revive ‘kinship’ studies on the basis of Schneiderian propositions. Her aim was to communicate ‘lived experience of relatedness in local contexts’ (Carsten 2000: 1) and accentuate the particular relationships as were exemplified by individuals in their respective social, material and emotional context, whether or not these relationships could be articulated in genealogical terms. She christened the area under discussion ‘New Kinship’, since it was aimed at discussing gender, the body and the person, along with other contemporary priorities of Western discourse, but not terminologies. In the introduction to Cultures of Relatedness, she described her approach as follows: Rather than beginning with the domain of kinship already marked out, the authors in this volume describe relatedness in terms of indigenous statements and practices – some of which may seem to fall quite outside what anthropologists have conventionally understood as kinship. (Carsten 2000: 3)

In other words, terminologies of kinship or relationship, conventionally associated with kinship or ‘kinship’ in anthropological research, are not included in Carsten’s concept of ‘Indigenous statements’. She took ‘kinship’ to be an equivalent of European kinship and avoided it.

108 • Lewis Henry Morgan’s Comparisons

In trying to exclude any kind of ideological bias, she instead decided to evaluate ethnographic facts beyond such a domain ‘that is already marked out’ by the (usually Western) researcher. What Carsten’s programme practically means may be illustrated by her monograph on the inhabitants of Langkawi, an island in the northwest of Malaysia near the Thai border, where men work as migrant labourers or engage in fishing to meet separately on the shore. Her ethnographic consultants were mostly women, who dominate the field of agricultural work. Kinship as ‘relatedness’ in Carsten’s words stands for their practices of feeding, hosting, exchanging or marrying and includes children, foster children and grandparents as actors. Such ‘New Kinship’ is recorded as a process that implies incorporation into the general practices of the islanders. Whenever other people enter their domestic scene, the locals ‘make’ kin by negating all differences caused by the external world. Women promote and execute this incorporation by emphasizing equality. They say ‘we are all relatives here, we are all the same’ (Carsten 1997: 14). The policy of erasing differences is bound to lead to ‘structural amnesia’, a phrase the author has adopted from J. Derek Freeman (1961), who had applied it to the Indigenous Iban in Sarawak. Freeman associated this concept with cognatic classification in general. For Carsten, it was not a result of ignorance, but the consequence of intertwining kinship and history. Like other cognatically organized societies, Malays lack sociocentric categories providing general rules of exogamy, but being Muslims, unlike the Iban and other Indigenous societies, they have nothing against ‘close’ marriages. ‘Through this notion of endogamy, people who marry each other become kin whether or not they were so previously’ (Carsten 1997: 16). The house, which may be enlarged by further material constructions whenever more space is required for newly married couples, serves as the formal framework of the domestic sphere. Outsiders moving in to stay become kin, irrespective of their earlier status. In fact, the idea of the entire society is conveyed by the image of a large house. The particular building with all of its extensions after new kin have been ‘made’ operates, as far as I can see, for Malay peasants in the same way as the Iban long-house outlined by Freeman (1961) provides the visible, and equally formal, design of their society. Such a material structure has been described for many Southeast Asian Indigenous societies (see Appell 1976), while relationship terminologies of the cognatic societal order have not attracted any notable attention. The house is known to have resulted from practical engagement involving its construction and further additions and yet, as a given omnipresent

Schneider, Relatedness, ‘Malayan’ and a General Comparison • 109

and immobile framework for the whole, it functions as an indispensable and fixed point of orientation, as does the ‘heat of the hearth’ (Carsten 1997) inside.

A ‘Malayan’ Terminology As mentioned above, Schneider conducted his early ethnographic research on Truk, where the islanders used a terminology called ‘Malayan’ by Morgan and Hawaiian or ‘Generational’ by later anthropologists. Subsequently, this critic gave a cultural account of ‘American Kinship’ (1968). His work included some attention to the meaning of one or another of the descriptive terms for kin, as used in the English language or in intrafamily slang. However, as shown by Morgan, the descriptive English kinship terms do not form a system. On Langkawi, Carsten’s consultants applied a ‘Malayan’ terminology and she too, like Schneider when discussing Truk and the United States, failed to find significant aspects in its formal structure to be analysed or mentioned at all. These ethnographic preconditions could suggest that certain aspects of ‘Malayan’ systems, and the commonalities they share with descriptive terminologies, may have influenced Schneider and Carsten not just to disregard the formal structure of these terms, but to reject terminological studies altogether. Undoubtedly, both anthropologists must have been introduced to classificatory systems and the refined debates on their relevance, but reading about a Dakota terminology, for example, is a rather different experience compared to living over an extended period of time among people who constantly and significantly employ the categories of such a system.7 In short, the particular terminological structures observed during their ethnographic fieldwork may not have been half as exciting compared to the specific sociocultural ideas, such as those involving house and hearth or substance and code, and the resulting relations. The general social order of all peoples applying ‘Malayan’ terminologies seems to coincide with an absence of unilaterally ascribed exogamous social categories (i.e. clans) or a sociocentric structure determined by affinity. Marriage choice is the major issue for Malay peasants and also for Indigenous people in a Southeast Asian longhouse. Individuals who are somehow involved decide on a marriage. They choose who is to join their paternal or maternal relatives, their siblings, their spouses or others belonging to the house who are not necessarily related in a genealogical sense. They are not bound by

110 • Lewis Henry Morgan’s Comparisons

some inescapable decree to avoid some and choose other categories as spouses; they may also be able to revise their choices after some time has passed. Though the kindred and other such cognatic institutions had been a much-neglected subject in anthropology until Freeman’s exemplary work on Iban social structure, considerable research followed that pointed to many societies operating without the unilaterally assigned clans as is usual among the Seneca, the Omaha, and most other Native North Americans, or the Indigenous people of the Middle Indian highlands and the peasants of the subcontinent. Clans enforce supraindividual affinal relationships and promote these reciprocal ties as the dominant concept of the social, or as the highest form of implementing the ‘spirit of the gift’. Sociocentric affinal categories provide the fundamental ideology of a society and thus go together with those Ego-centric classificatory systems that are predominantly outlined by the principle of affinity. In Langkawi and other parts of Southeast Asia, such affinal sociocentric categories are missing, as are Ego-centric systems determined by the principle of affinity.

Terminological Details To elaborate these differences, the following will consider the structure of the ‘Malayan’ terminology. As an example, I have selected the contribution by David J. Banks (1974), who, like many others, is rightly critical of Morgan’s evolutionist ideas on such a ‘Malayan’ system. Banks conducted extended fieldwork on the Malay Peninsula in the northwestern province of Kedah situated just across the sea in the east of Langkawi. His dissertation at the University of Chicago was supervised by Schneider. In introducing his Malay terms, I adopt his Table 1 (Banks 1974: 50) with only slight technical alterations.8 Table 6.1  List of Malay consanguineal terms (Kedah). datok (tok) bapak (pak) emak (mak) abang kakak adek anak chuchu chuchit (chitchit)

G+2 G+1 (male) G+1 (female) G0(e) (male) G0(e) G0(y) G-1 G-2 G-3

Schneider, Relatedness, ‘Malayan’ and a General Comparison • 111

A look at this list offers the impression of a ‘one-line terminology’. There being no affinal terms, terminological lines (plural), which owe their existence to formal documents of reciprocal affinal relationships, are nowhere to be seen. The list contains terms discriminating generations and, only in G0, relative age. Parents are distinguished by sex and so is the elder brother (abang) as a sole indication of gender inequality. Like the affinal dimension, the distinction between lineals and collaterals is conspicuous by its absence. Thus, the question may arise as to whether it is useful to speak of a Malay structure at all. I think one should, provided that the sex-distinctions are taken as given irregularities of a different kind compared to those of other well-known terminologies. In Malay, the emphasis lies definitively on the distance between generations. Only G0 discriminates seniors from juniors internally. Apart from the exceptions mentioned, a single category is supplied for all members in one of the above-listed generations. But this is not all. In a very elaborate diagram,9 Banks illustrated how these consanguineal terms are supplied with ‘affixes indicating degrees of genealogical relatedness’ (1974: 49) in all generations according to a style that is similar to Europeans speaking of ‘second cousins’ and the like. Thus, the terms in Table 6.1 are supplemented by additions to point out that the person referred to is one, two or three steps removed from the lineal relative. I suggest that the effect of these affixes is not so different from the European circumscriptions of this kind or from that of composite terms in Erse, Morgan’s favourite example of a thoroughly descriptive vocabulary. In Malay, these affixes substitute the missing collateral dimension in the linear assemblage of basic terms. Furthermore, Malays make frequent use of marker-affixes for their terms, as given in Banks’ Table 2 (1974: 52), which I have adopted as my second list in Table 6.2 below. The marker ‘pupu’ in this list indicates the collateral removal from the term to which the affix for the first, second or third degree may be added, whereas in generations other than G0, the addition of ‘penakan’ stands for ablineal in general. In the same way, a marker-affix indicates lineal relatives of one generation above the third ascending generation and another one points to a step-relative. Of some importance (as in several Indian languages) is the addition marking the birth order within a sibling group. As general terms, the list given in Table 6.2 includes one for ‘lineal consanguineal relative’ and another for ‘supragenealogical kinsman’. These might also be found in other types of vocabularies, but the marker-affixes for four affinal relationships appear to be unique, i.e.: (1) that for an in-marrying person in a relatively lower generation

112 • Lewis Henry Morgan’s Comparisons

Table 6.2  List of common marker-affixes for Malay terms (Kedah). panggilan pangkat pupu penakan nenek; moyang betul tiri menantu mentua ipar angkat besan

title indicating order of birth within a sibling group degree of collateral removal from a consanguineal ablineal; applied in a generation other than Ego’s higher than third ascending generation consanguineal relative lineal consanguineal relative step-kinsman, through the marriage of a lineal kinsman to someone other than a lineal kinsman a person married to a blood kinsman of a lower generation compared to Ego’s a person married to a blood kinsman of a higher generation compared to Ego’s a person married to a blood kinsman of the same generation; the blood kinsman of a spouse a supragenealogical kinsman the relationship of a person to the parents of the spouse of his child

(menantu); (2) that for one marrying into a relatively higher generation (mentua); (3) the addition of ipar to identify an in-marrying person of the same generation or a consanguineal relative of a spouse; and finally (4) besan to point to the parents of a child’s spouse. Instead of affinal terms, marker-affixes on the consanguineal terms (distinguished by generation) show that those persons who are connected by marriage are the same as the lineal relatives, although they are slightly different. The results that Banks (1974: 47) draws from these data initially state that the terms meet Morgan’s requirements for a ‘Malayan’ terminology in that they are generational, allowing for sex-distinctions in one or more generations and for distinctions of relative age in Ego’s own generation. However, Malays use these terms in three semantically distinct senses: • First, they are part of an ideology of consanguinity, or biogenetic substance, which recognizes generational, sex and relative age differences between consanguines. • Second, they prescribe a code of conduct that applies regardless of the consanguineal relationship. Banks calls them extra-genealogical. • Third, the Malay terms denote the closest relationships in Malay society, being defined by reciprocity, giving and all other manifestations of affection. The author identifies them as supergenealogical. As a result of the study and by reference to many further details, Banks is suggesting that the third sense is the primary sense and that the others are extensions of this meaning as required by moral law.

Schneider, Relatedness, ‘Malayan’ and a General Comparison • 113

In a coherent manner, Banks presents these basic terms as well as their affixes with the intention to demonstrate, how they are applied in particular situations or for specific purposes between the bearers of certain social roles. The aim of his admirable ethnographic work is to show that the terms do have a genealogical meaning, but are, above all, used to express a general ideology, prescribe a code of conduct and communicate specific central values. Banks’ position on the supergenealogical meaning of the terminology, which is not elaborated upon here, is forwarded in a convincing way, and even if it were not, my lack of ethnographic competence would disqualify any further comment. And yet the question remains as to whether the pattern as such carries a meaning, i.e. one contained in the structure of the terminology that could be compared with the ethnographic observations of the practical applications of the vocabulary. Such a comparison could indicate a fit or unearth major contradictions, or ways in which documents of ideas on the social whole are more or less translated into observed social behaviour.

Structural Features The ‘messages’ transmitted by a particular classificatory terminology can best be discerned in contrast to others. In the following, I will try to indicate the deviations of ‘Malayan’ by comparing its characteristic features to the well-known ‘Dravidian’ terminology (Trautmann 1981; Dumont 1983) and those of Seneca, Omaha and HMI to illustrate the different modes of classification that point to different classificatory priorities or an emphasis on certain general social values. ‘Messages’ on affinal relationships characterize the other terminologies, but not ‘Malayan’. Continuous affinal relations, or those being repeated from one generation to the next, are categorically excluded by Seneca, Omaha and HMI. By different techniques, these three terminologies prescribe the dispersal of such links, while ‘Dravidian’ stands for diachronic or uninterrupted affinity. However, holistic schemes of affinal classification always suffer from a contradiction which is covered up more or less elegantly. The eternal dissonance in any kind of affinal classification involving Ego as a part of a whole is caused by the very idea of the social as a creative domain reserved for anthropos. Humans, not animals, organize their social entities on their own in a wide variety of ways. Elementary sociality implies becoming engaged in exchanges with other humans. The primary and the ultimate achievement of such an engagement is the exchange of humans themselves

114 • Lewis Henry Morgan’s Comparisons

Table 6.3 Features of classificatory terminologies compared.

‘Dravidian’ Seneca Omaha HMI

‘Malayan’

Continuity/ dispersal of affinal relations

Affinally opposed pairs

Terminological lines (number)

Main principles

one continuous reciprocal relationship dispersal of reciprocal relationships dispersal of reciprocal relationships dispersal of reciprocal relationships

spouses in G+1, siblings in G0 and G-1 spouses in G+1, siblings in G0 and G-1 siblings throughout

consanguineal (1) and affinal (1)

spouses in G+1, siblings in G0 and G-1

consanguineal (2) and affinal (4)

no affinal relationships

none

consanguineal (1)

symmetric affinity; sex; generation; age symmetric affinity; sex; generation; age symmetric affinity; skewing; sex symmetric affinity; sex; age; generation; collaterality generation

consanguineal (1) and affinal (3) consanguineal (2) and affinal (3)

to procreate new generations and thus ensure a continued existence. However, classifying such a process ‒ i.e. the exchange of humans by and for humans ‒ has to cope with the central limitation articulated in several ways: • siblings cannot be spouses or vice versa; • only spouses can be parents of the humans in subsequent generations; • only siblings can be immediate descendants of their parents; • however, all-inclusive diachronic schemes must classify later generations as a continuation, i.e. in the same manner as the earlier ones. The cover-up of the contradiction in the ‘Dravidian’ system is probably the most elegant one, since it is so simple. The formal structure of G+1 has it that the members of the parental pair of spouses are both viewed as consanguines being affinally opposed by FZ and MB. Neither M and F nor FZ and MB are siblings, while their respective children are, as are all other pairs, involved in affinal oppositions of the three medial generations. The Ego-centrism of the scheme disallows a formal purity of the kind observed in sociocentric section systems. Seneca and HMI, because they imply dispersal, are more complicated than ‘Dravidian’, but work with the same initial pretence, whereas Omaha does not. In the latter case, the siblings F and FZ are affinally opposed by the siblings MB and M, and thus only sibling-pairs appear in all reciprocal relationships. However, since F and FZ cannot

Schneider, Relatedness, ‘Malayan’ and a General Comparison • 115

have the same children, FZH must appear out of nowhere, remain least conspicuous and disappear without leaving a ‘track’ in the shape of generational skewing that could diachronically manifest his existence. Clearly, FZH is an affine. Thus, the term for him is not the same as that for consanguineal FZS. Because F and FZ form one pair of the initial reciprocal relationship and must both be antecedents of consanguineous children who, however, cannot be the same, two separate consanguineal lines must be classified. ‘Malayan’ has no problems on account of this ‘primary anthropological impact’ because it overcomes the contradiction that is caused through the exclusion of incest by an outright denial of the very existence of affinity as an intergenerational classificatory principle. Empirically, this is demonstrated by Carsten’s account of the inclusive house. ‘Dravidian’ is elegant because of its simplicity and it can remain simple because it is avoiding affinal dispersal, i.e. a pattern implying a greater elaboration of the social in formal language. Accordingly, Seneca, Omaha and HMI require multiple terminological lines for their diverse affinal relationships. This diversity indicates their more refined ways to represent the social. Omaha and HMI manage such differentiation by introducing a second consanguineal line, even if in effect they cannot provide the immediate transparency that goes with Australian systems (such as Aranda) containing more than one consanguineal line. The Seneca solution is to ‘consanguinize’ the descendants of former affines. ‘Malayan’, on the other hand, can carry on very well with a single line of consanguines, or by lacking affines, even if logically ‘impure’ spots do appear in the shape of the two categories (in G+1 and G0) that differentiate sex. Otherwise equality rules supreme within each generation, though the distance between one and the other genealogical level characterises the entire scheme. Regarding the principle of collaterality or the idea that consanguines, whether or not they are opposed to affines, are not necessarily bound to appear as singles or in pairs, HMI deviates from the other four structures by introducing collateral specifications in G+1, G0 and G-1, as well as altogether separate relationships of symmetric affinity for Ego in distinction from his or her siblings. Since ‘Malayan’ does without reciprocity and without pairs, collaterals must equally be lacking. The separate existence of M and F, and also the fact that Malays have only an eB (=abang) among the G0 terms that otherwise include any sex, ‘pollutes’ the logical coherence of the whole. The general ‘message’ is that one term stands for one category in each generation, but this obvious dictum pronouncing the spirit of equality falters somewhat in view of M and F and also of eB.

116 • Lewis Henry Morgan’s Comparisons

As a result of such comparisons, it is clear that ‘Malayan’ represents a transition between classificatory and descriptive systems. The generational terms are the classificatory or formal element. In sharp contrast to Omaha, they emphasize the distinction between forebears and descendants for the whole. They mark distance. Affinity is not given. The specifications of F and M are obvious exceptions from the general mode of providing only one category that joins males and females in each generation. The descriptive element supplies a way to come to terms with individuals empirically related to Ego through marriage or by sharing one or both parents. The various marker-affixes are auxiliary constructions of the ‘domestic institution, which serves to organize a family by the bond of consanguinity’ (Morgan 1871: 11). By such additions in the appropriate case, it becomes obvious that each ‘relationship is thus made independent and distinct from every other’ (1871: 13).

Conclusion Schneider, Carsten and Banks have all conducted their refined ethnographic fieldwork in societies lacking unilaterally assigned sociocentric categories and a rule of exogamy that is equivalent to a public order defined by affinal exchange. In such an environment, a ‘Malayan’ relationship terminology will emphasize intragenerational equality and mostly ignore the sex-difference. In short, ‘Malayan’ is not really very expressive or refined and few colleagues have worked on such systems or would want to. However, to conclude from such field experience that all terminological studies are futile may be popular, but is hardly appropriate. At the same time, the ethnographic work of Schneider, Carsten, Banks and others should in no way be discredited when they study how the terms of a language are applied in certain situations or for particular purposes and as a way of assigning particular social roles. Such research can demonstrate that, besides conveying a formal structure, these terms allow a practical use and a discourse to articulate a general ideology or communicate a code of conduct as the expression of specific social values. Research on ‘relatedness’ is frequently assumed to be an innovative exercise, as if something formerly unheard of were suggested by ‘New Kinship’. However, as far as I can recall, innumerable earlier projects have been intensely involved in the study of practice and discourse and I see no reason why such an anthropological project should exclude separately conducted structural analyses of relationship terminologies

Schneider, Relatedness, ‘Malayan’ and a General Comparison • 117

in the event that the latter are of the classificatory kind and, one might add, in case they are not as simplistic as a ‘Malayan’ system tends to be. Social anthropology lives on comparison. In the end, it is concerned with the social potential of anthropos and not with local and regional particularities. Morgan had taken a keen interest in the special ways in which the Seneca dealt with gender, religion and the community of the house. This did not prevent him from conducting surveys on a worldwide basis to point out striking similarities in different languages and communities of several continents. In his day, he came close to finding a rather limited number of ways to cope with the task of ordering the unified whole of social categories. He remained unaware of affinal prescription and proscription or of a classificatory mode that did include a distinction between lineals and collaterals, but he understood that such social patterning had been effectively employed by anthropos before the establishment of the state and had been retained by the larger part of humanity even after states had been introduced.

Notes 1. I put ‘kinship’ in inverted commas whenever Morgan’s ‘apparently arbitrary generalizations’ (1871: 11) are applied by terminologies. 2. With co-editor Kathleen Gough (Schneider and Gough 1961), he offered a worldwide comparison of ‘matrilineal kinship’, thus confirming the existence of such a phenomenon. It was related to the political economy of the respective society, though in the prominent South Indian cases of the book, I find that some settled cultivators emphasize their matrilineal kinship while they live – under the same conditions – side by side with others articulating patrilineal kinship (see Dumont 1983: 67‒68). 3. The thesis drew heavily upon Radcliffe-Brown’s earlier article of 1940. 4. Ego’s MBD is not normally the daughter of the physical brother of Ego’s physical mother. Social norms of some societies, such as the north Australian Gidjingali (see Hiatt 1965), prefer MBD marriage even though they explicitly exclude a woman who is physically related to Ego. 5. See Chapter 1 above. 6. Apparently, these general binary concepts are widely distributed wherever Indo-European or Semitic languages are found (i.e. those lacking classificatory systems), or where the given society has been greatly influenced by an Abrahamic religion. As will be shown below, ‘Malayan’ systems, lacking the general classificatory principle of affinity, also seem to produce such general formal features beyond the domain of a terminology. 7. This difference has influenced my own work considerably. 8. Instead of symbols, I add the words ‘male’ or ‘female’ in brackets beside a term. 9. Since it is so extremely regular, it need not be reproduced here.

7 Social Evolution and the Australian Anomaly

The Issue Systems of Consanguinity and Affinity of the Human Family finally came out in 1871. In a London library, shortly before he was due to sail back to America, Morgan was able to lay his hands on a copy. In the British capital, he had been able to arrange short meetings with Darwin as well as other prominent scholars of his time (Maine, Lubbock and McLennan) who had floated far-reaching theories on social evolution in the discipline that was to become sociocultural anthropology. During these initial meetings, the colleagues and competitors found friendly words for his work, but had their reservations. In subsequent years, some would attack it in no uncertain terms (Resek 1960: 126). By contrast, Lorimer Fison, a little-known Wesleyan minister working in Australia, took the opportunity of Morgan’s visit to London1 to initiate a lasting form of communication with him that involved rather sensational ethnographic data on the social order of the aborigines. Morgan urged him to publish these findings and Fison became Morgan’s first disciple (Resek 1960: 127). Earlier, Fison had been encouraged to supply the Tongan and Fijian terminologies for Systems (Morgan 1871: 521), since he had been posted on these islands before his final destination and thereafter continued to take an interest in Indigenous social structure, especially in the highly complex affinal patterns, becoming one of the prominent pioneers of Australian social anthropology. Fison’s ethnographic work on ‘marriage classes’ made up much of the first chapter (Part II) of Ancient Society (Morgan 1877: 49‒59), which was followed by what Morgan presumed to be the ‘more advanced’ Iroquois social order in another four chapters of this book. Amazingly, this overture of Part II discussing ‘growth and ideas of government’ contained next to no information on the economy of aboriginal Australia, even though technological and economic

Social Evolution and the Australian Anomaly • 119

­ evelopment was as important a part of Morgan’s theory on social evod lution from societas to civitas, as was the ‘gentile organization’ (Morgan 1877: 12f.) of the Iroquois compared to that based on territory. Morgan associated Native North Americans with ‘barbarism’ because of their adobe houses, their pottery and their maize cultivation (1877: 17). Later on in the book, he speculated at length on the issue of ‘property in the status of savagery’ (1877: 445f.) or on the ‘lower’ (447f.), ‘middle’ (451f.) and ‘upper status of barbarism’ (457f.). In a conspicuously short note on Australia, Morgan, basing his remarks on Fison, wrote: In the ethical position the aborigines of this great island are near the bottom of the scale. When discovered they were not only savages, but in a low condition of savagery. (1877: 317)

Such a harsh conclusion does raise the question of the motives held by the usually careful and open-minded Morgan. I am not aware of any previous arguments on this issue. After having considered Morgan’s ingenious analyses of the Native North American classificatory systems of relationships, I propose that his ‘downgrading’ of the aboriginal Australian social order was not just a result of what appeared to be their very simple technology; probably, it was due more to the author’s ignorance of the value of affinity, as elaborated above in chapter three on Seneca. Access to the Amerindian systems was possible for Morgan, since he understood them to be generalizations of the consanguineal terms only.2 However, the amazing complexity of the Indigenous Australian systems is clearly based on their relationships of affinity. The value embodied by the sociocentric categories of sections and subsections is such an aspect. Another one is its combination with Ego-centric categories prescribing, for example, either uninterrupted reciprocal affinal relations or those in alternating generations. Yet another aspect is the unusual distinction between a formal affinal prescription (e.g. of the cross-cousin category) and a behavioural norm that excludes the specific genealogical relative (e.g. the genealogical cross-cousin) from such an alliance.3 However, this impact of the value of affinity is only recognizable if such a relationship is not confused with Euro-American notions of in-laws. In English, a ‘sister-in-law’ is one of a very small number of individuals related to Ego by the latter’s marriage or that of one of Ego’s siblings. On the other hand, an affine in the classificatory context of many Australian aborigines will form – along with others of that category – half of the entire society, i.e. that opposed to Ego’s half. Morgan was unable to examine the issue of inherited affinity and Trautmann (1987: 244) has suggested a personal reason for this omission because Morgan had married his cross-cousin. During the nineteenth

120 • Lewis Henry Morgan’s Comparisons

century, cousin marriages were widely practised in the European and American upper middle classes4 (Sabean 1998: 81‒82). The convention seems to have prevented him from associating such marriages with either ‘savagery’ or ‘barbarism’. However, the ill-founded and absurd thesis of ‘group marriage’ (see chapter one above) as the ‘lowest’ form lingered on in Morgan’s mind later being associated with aboriginal Australians. If 140 years ago the ethnography of the southern continent was yet to be initiated so that it would be very arrogant to criticize early theoreticians on the basis of modern research results, the questions about the ‘Australian anomaly’ remain all the same. The aborigines are enriched by a complex religion and a refined social structure, though their technological ‘base’ is as simple as that of hunter-gatherers. Europeans, being taught to give priority to the economic domain, tend to associate Native Australians with other foragers. The following discussion will attempt to come to terms with this apparent discrepancy between, on the one hand, a highly sophisticated worldview and system of ‘kinship’ and, on the other hand, the extremely restricted physical existence of these Indigenous people. My argument will initially require a rather long review of some developments in anthropology with special reference to the idea of social evolution.

Trend and Tide From the 1920s to the 1980s, Morgan’s books had almost obtained the status of sacred texts in the Soviet Bloc, since Marx had been greatly impressed by Ancient Society (1877) and asked Engels to popularize it. The outcome was the latter’s work on The Origin of the Family, Private Property and the State, first published in 1884,5 which soon became an adequate present on the occasion of lifecycle rituals in a socialist family. In the United States, on the other hand, during much of the twentieth century, Boas and his disciples had found a good number of relevant ethnographic and theoretical reasons to reject Morgan’s evolutionary theory and exclude it from anthropology syllabi. Similarly, in most of Western Europe, Morgan’s grand designs were either ignored or ridiculed. However, the dominant discourse is a wavering companion. When, beginning with India in 1947 and continuing until the 1960s, the colonies of West European powers gained independence and Cuba joined the Soviet Bloc, the ongoing Cold War generated an altogether different dimension of U.S. social research in Africa, Asia and Latin

Social Evolution and the Australian Anomaly • 121

America. Directly or indirectly, Washington financed innumerable projects designed to study the social complexities of what was sometimes called ‘the Third World’. During this period, anthropology realized that the political economy in many of these countries depended upon the peasantry as the most important social class and one subjected to rich and influential owners of large estates, the respective elite that had formed the backbone of colonial rule. In many of these countries, the members of this landlord class continued to acquire their education, general outlook and attractive income opportunities by temporarily moving to metropolitan Europe or America. As the specific insight of this research phase, the sociocultural conditions of peasant proprietors and other cultivators or craftspeople were found to be fairly similar, irrespective of the specific regional history. Perhaps an article by Eric R. Wolf (1957) for the first time in anthropology compared the specifics of such peasant communities of different continents (in Mesoamerica and Central Java). Subsequently, it had a significant impact upon major theories on the peasantry. Many inquiries then focused on the – general rather than specific – differences of peasant societies compared to others that had preoccupied the discipline during earlier epochs, when most of U.S. cultural anthropology was engaged in research on Native North Americans, while British and Commonwealth social anthropology primarily examined what was then called Structure and Function in Primitive Society (Radcliffe-Brown 1952) as found in the many colonies of the United Kingdom. Before the United States conducted its massive military campaigns in Southeast Asia that greatly affected all other concerns in relation to ‘the Third World’, such general distinctions became the basis of rather influential theories. This trend was more or less accidentally supported by a specific occurrence during the 1950s, when a German named Paul Kirchhoff, who during the Nazi era had gone into exile in Mexico, spent some years in the United States. Being a Marxist associated with the camp of Leo Trotzky, he was not in the good books of Soviet scholarship, but was in a position to work as a visiting professor at the University of Washington (Fried 1968: 370) and also deliver lectures at Columbia. His rather impressed students and listeners included Morton H. Fried, Elman R. Service and Marshall D. Sahlins, though primarily these later luminaries must have been influenced by the historical materialism of Leslie A. White and the cultural ecology developed by Julian Steward, who had founded the new approach of evolutionism in the United States, a trend that was largely ignored by Commonwealth anthropologists and remained untouchable in both East and West Germany.

122 • Lewis Henry Morgan’s Comparisons

Though cooperating regularly, Fried, Service and Sahlins6 differed on several points. Thus, Fried (1967), impressed by Kirchhoff’s concept of the ‘conical clan’,7 gave considerable weight to the establishment of chiefdoms8 as a societal form while downplaying the relevance of Sahlins’ ‘tribal society’ as a transitional feature. This may have been due to Fried’s regional interest in Inner Asia and the Chinese borderland, an ethnographic zone characterized by the overwhelming importance of pedigrees. However, the common interests of the ‘neo-evolutionists’9 included (in Morgan’s tradition) studies of specific societies as well as theories on anthropos or human evolution. The scholars directed their historically oriented research towards political, economic and ideological issues, as found in the respective cultural context, and at the same time towards a formal worldwide comparison of such features. Inspired by prehistorian V. Gordon Childe (1951), they were out to identify general societal forms of the simpler or the more complex kind, irrespective of the question whether these forms, in all of their historical specifications, might be the most efficient adaptations within a particular natural environment or during a certain period of history, and irrespective of ethical postulates or the degree of enlightenment that might have been attributed to the people associated with one form or the other. In short, their version of evolutionism did not propose that humankind had passed from ‘lower’ to ‘higher’ periods, but observed that specific features of relative sociocultural simplicity or complexity in empirical social systems happened to be fairly widespread forms across the globe, though the particular historical events leading to them in the region or on the given continent were to be understood within their own context. Under the impact of postmodernism, the discipline has again turned its back on any kind of evolutionism. Thus, the following will go against the tide by retaining basic evolutionary considerations as one perspective among ‘the many other legitimate anthropological concerns’ (Sahlins and Service 1960: 3).

Societal Forms It goes without saying that the more modern approach of evolutionism replaced old ascriptions like ‘savagery’ and ‘barbarism’, which, in the general jargon of his epoch, had also been used by Morgan. Instead of the former label, modern conventions refer to ‘hunter-gatherers’, and in the following I will apply this term only to individuals who obtain their livelihood from ‘immediate return’ (Woodburn 1982), pri-

Social Evolution and the Australian Anomaly • 123

marily from foraging, so that sometimes hunting is relegated to the second position of that category. These contemporary people belong to mobile bands of fluctuating membership and have only very limited posessions. According to a prominent thesis by Sahlins (1968b), they accumulate hours of leisure rather than material property, though his critics do not think this to be a universal feature. While individual gatherers are strongly committed to sharing food and the other goods within a band, they only to a very limited and nonritual extent engage in exchanges with outsiders. They also avoid major ritual activities in favour of a superhuman power. The latter is seen as benevolent and taken for granted. The high degree of individual autonomy within these groups lacking institutionalized leadership is complemented by very little social pressure from other bands. A woman and a man may or may not choose to live together and will do so without ceremony or fear of group sanctions after a casual divorce. Only coresident family members are regarded as kin. In short, their generally low profile and fluctuating membership is the very strength of such ephemeral bands. The mobility, small size, lax organization and material moderation of their members provide ‘the security of the weak’, resulting from their extremely limited or nonexistent ambitions to gain prestige, property and power. Conflicts within the band or with outsiders are solved by withdrawal. However, such propositions, on account of their rather general scope, always include the possibility of empirical cases appearing in somewhat modified forms. In some cases, even the much-discussed differences between ‘immediate-return systems’ and those of hunter-gatherers’ operating systems of ‘delayed return’, as developed by James Woodburn (1982), can be articulated in relative terms only. By contrast, those called ‘tribal societies’ by Sahlins (1968a) imply a social order meaning significantly more than fluctuating assemblages of individuals. Their respective size is much larger, i.e. ranging between a few hundred and several million people who live in permanent homes and mostly on domesticated plants and animals. Even in cases, when they do not produce food, because a bountiful natural environment provides them with sustenance that does not require soil management or the raising of animals, such societies are known for their long-term investments in houses and storage rooms, boats and harpoons, traps, weirs, fences and many other technical devices to acquire, distribute and preserve food. Collectives of such people, though never ‘the tribe’, may unite in defence, or may attack other communities in open conflict in order to guard or capture women, land and material goods, or obtain the services of the defeated. Permanent local or village units, rather than

124 • Lewis Henry Morgan’s Comparisons

individual owners, control settlements and cultivated land, whereas privately owned articles are of use-value only, or else are exchanged and accumulated to be given away in ostentatious ceremonies. Formal marriage is an established feature of these anarchic societies, with sanctions being applied to deviant behaviour. In Highland Middle India I have seen families with elderly parents who never attained the status of ritual adults because they had not married in style. Members of such societies either belong to matrilaterally or to patrilaterally ascribed sociocentric categories (clans) formally designed for affinal exchange, or else only to the egocentric network that, at times, encourages certain unions and disallows others in order to promote ongoing formal shifts of individuals between local collectives. Within the boundaries marked by such social categories, empirical persons will in many cases be free to choose their spouses, though in others cases, marriages may be arranged or even enforced by elders, or else both alternatives are acceptable. The resultant affinal exchanges are frequently intertwined with far-reaching systems of formal gift-giving obligations on the many ritual occasions when the worship of multiple divinities is associated with diverse but local superhuman influence. Valuables then pass from one subgroup to another, maintaining intergenerationally relevant structures that provide the normative order and social orientation for individual members. Thus, they form a social cosmos beyond individual impulse and potential. At the same time, these small or large ‘tribal’ units, or their federations, are associations of people without a sovereign central authority or a staff of specialists to execute administrative policies. In short, members understand ‘kinship’ regulations to provide a naturally given (and thus legitimate) formal order applied to systems of reciprocity in the mutually inseparable domains of political, economic and ritual activities. Individual autonomy is limited by permanent and stringent obligations in favour of the local group, which in turn provides social security and fairly equal access to material goods, even if gender inequality and some ritual ranking of persons and social categories go together with immaterial privileges. These will be associated with males rather than females, or with mature householders as opposed to the unmarried youths and the aged. In peasant society, the sovereign centralized state wields supreme power and is frequently conceived of as the highest ranking owner of all immobile property and as redistributing its benefits. Land is not assigned to a single proprietor; ownership is instead conceived as a chain of ranked claims to the soil. These demands are enforced by the elite of noncultivating landlords who simultaneously reside in urban

Social Evolution and the Australian Anomaly • 125

centres. As the most privileged landowners, they are empowered to extract, using their specialized staff, a maximal share of the surplus product. Peasant proprietors do the work and deliver the ‘rent’ while supervising tenant labourers and rural craftsmen. The latter are again subdivided into different categories, such as those of permanent or temporary tenancy. In the event that the returns of the peasants surpass the amount required for their physical maintenance, landlords are likely to raise the so-called ‘rent’. On account of such privileges, cultivators display little interest in investments or in improving the techniques and the coordination of the labour force.10 Peasant society also implies the existence of a centralized administrative staff of scribes and a standing army of the ruler along with ritual specialists supplying many different kinds of services to maintain innumerable minor, major or gigantic temples or shrines for the worship of multiple saints or divinities. One god is, at least regionally, likely to be the ultimate authority in all matters, and the priestly staff are empowered to apply far-reaching sanctions against real or suspected rivals or alleged nonconformists. Such sacred centres will simultaneously hold lands as a result of the many endowments of the pious. The human sovereign is generally seen as the first servant or even as the physical representation of the supreme divine power. He (the overlord is mostly a ‘he’) will thus collect ‘rent’ from his subjects as commissioned by the divine power in order to channel these resources into the construction and the organization of large temples, their respective staff or their bombastic processions, and also for the erection of palaces, tombs and royal gardens for public display. Major irrigation works or canals and dams, and also highways connecting them, remind the subjects that the supreme power provides the fertility of the fields and protection from floods and droughts, and is thus legitimized as the absolute ruler. Under these conditions, the (relative) control of family and individual property is of great importance in peasant households, a piece of cultivable soil being the most coveted goal. Authoritarian male heads of a joint family try to preserve the unity of siblings and cousins, and to continue joint policies of the local descent group consisting of several such units, though the unavoidable rivalry between close relatives over land inheritance and marriage policies will sooner or later spoil such attempts and promote an aggressive atmosphere at the village level. To a major extent, marriage alliances determine peasant attempts to consolidate the respective power base of the family in question. Thus, sisters and wives are kept under tight control by older and even by younger men of the same local descent group. Many senior women and, as final authorities, men (fathers, uncles and brothers) see to it

126 • Lewis Henry Morgan’s Comparisons

that a virgin, together with a very considerable dowry, is transferred to the bridegroom’s family for political reasons and very often only after negotiations lasting for several years. On both sides, numerous closer and more distant kinspeople are involved in creating the alliance, as they are affected by it. At the climax of such a project, a major amount of savings, or borrowed funds, is spent on ostentatious weddings. Individual liberty is a feature of the highest class only. By contrast, industrial market society is in multiple ways and fields more complex than that peasant society. During the past five decades, formerly unimaginable degrees of interaction in time and space have differentiated the agencies of contemporary transnational capitalism, especially as a consequence of new techniques to transfer, in seconds, billions of euros to all corners of the globe. However, these modern or most recent forms will be left to the expertise of the specialists in several other social sciences, while anthropologists, in contradistinction to specialists of other subjects, are supposed to be trained in studying and comparing societies of the above-mentioned three forms characterized by sharing, reciprocity or redistribution.11 They resemble the ‘periods’ Morgan and contemporaries had described under the hideous titles of ‘savagery’, ‘barbarism’ and ‘civilization’. What is lacking in contemporary evolutionism is the idea of some universal ‘passing through’ one form after another in the course of history, though at different degrees of velocity. Also absent is any kind of judgement on the ethical, moral, technical or intellectual capacities and talents of the people involved. There is no ‘higher’ or ‘lower’ culture. As a reminder of the kind of ‘progress’ observed over time, it may be recalled that, during the last century, certain states defined by their highly complex economic, political, and religious order have been engaged in crimes against humanity of unimaginable dimensions. While these lines are written or read, the most ‘advanced’ social systems lead the general effort to destroy the world. In other words, reaching a state of social complexity is not necessarily considered a joyful achievement. Our modern individualism is, of course, averse to generalize societal forms and views them as the unified basis of human behaviour and ideas. Thus, most modern Westerners would be rather hesitant to conceive of the pressure of affinal obligations, resting on the shoulders of innumerable contemporaries, as a Durkheimian ‘social fact’. When at the start of the monsoon season (as was the case many centuries ago) hundreds of thousands, or even millions, of peasants are involved in the effort to pull the huge cart containing Jagannath, the ‘Lord of the Universe’, from his gigantic temple to his impressive palace for

Social Evolution and the Australian Anomaly • 127

‘retirement’ and when the relevance of this procession is underlined by the presence of the Prime Minister of the Indian Union and hundreds of priests directing the colossal ritual, most of us are unlikely to consider Marxian class relations as the basis for such an extremely complex and drawn-out performance. Perhaps the many individual decisions involving all aspects of life for foragers come closest to our modern understanding of social affairs, but then the general policy of ‘noninterference’ among hunter-gatherers may sooner or later become incomprehensible to Westerners. Much of contemporary anthropological research has simply withdrawn from any such general concepts. It focuses on the actions of individuals, but all the same, I will try to offer a comparative overview below.

Immediate Return Systems versus Delayed Return Systems Distinguishing hunter-gatherers into those of immediate and those of delayed return was initiated and later refined by James Woodburn (1982, 1988). Clearly, his distinction was motivated by the evident divide between a rather simple and a more complex organization of work and workers, even if both types of hunter-gatherers operated without domesticated plants and animals. The classification was necessary because those peoples living on delayed return organize themselves very much in the same way as other anarchic societies involved in long-term economic planning with regard to land and animal management, whereas this aspect is lacking in the immediate return kind of egalitarianism of individual assemblages due to the features listed above. The societies of the American Northwest Coast serve as a good example to illustrate this point, since they used to exist on fishing and a range of foods from hunted animals and wild plants, while at the same time maintaining complex systems of defensive and aggressive interrelations, social ranking, affinal exchange and spectacular gift-giving ceremonies (see Rosman and Rubel 1971). Like these Amerindians, Woodburn associated all Australian aborigines with systems of delayed return (1988: 35) in spite of their very simple technology. He came to this conclusion because prior to the European occupation and in some cases until the twentieth century, they displayed ‘activities … oriented to the past and the future as well as to the present’, but especially since they used to operate with ‘assets in the form of rights held by men over their female kin who are then bestowed in marriage to other men’ (1988: 32). Along with his coauthor, he emphasized:

Conflict  resolution

Transfer of  goods Gender status Individual  autonomy Group  autonomy Leadership

Inheritance Relative  equality Property  concepts

Size of system

Division of  labour

Food supply

from abundant nature or domesticated plants and animals by own labour

from foraging and hunting by own labour

Peasant societies

no institutionalized leadership individual or group withdrawal

high

absolute ruler and councils of kin in different classes village and camp elders as formal opinion state sovereign, landlords, peasant leaders leaders and councils feuding; local councils standing army and landlord decrees; intrapeasant autonomy in family and property affairs

medium

from plough-cultivation and/or cattle-holding by peasants for landlords none or minimal between women and men; sometimes far-reaching for both urban and rural, between cultivators and rural craftspeople manual and nonmanual labour of women and men small, mobile bands of settled in camps or villages; millions of large or small states with urban fluctuating individual members may be associated with a single administrative centres, towns and membership acephalous unit (‘tribe’) villages none by local collectives by individuals and joint families fairly high fairly high but status (not power) ranking social classes, i.e. privileged access to the means of production very little property and collective ownership and management of graded land claims linking the (possibly) plenty of leisure land; individually built houses sovereign, landlords, peasants and different statuses of tenants sharing of food and tools; systematic gift exchange between local redistributive centres of state and limited trade groups religious or secular elites slightly higher status of men higher status of adult male householders marked male dominance in all spheres high medium low

Anarchic societies

Hunter-gatherers

Table 7.1 Three societal forms: features compared.

few or no permanent sites

Places of  worship

no ritual specialists

Marriage  payments Concepts of  kinship Common  ideology

immediate family members residing together extreme individualism

none

Marriage choice individual choice to marry, divorce or remain single

Ritual  functionaries

Divine influence general notion of a ‘giving environment’13 Ritual activity low degree of ritual activity

low profile of individuals and groups

Exposure of   local units

spectacular state works and temple processions; display of dowries; grand family rituals numerous smaller and spectacular grand shrines/temples linked to the state multiple local divinities of diverse sovereign power ultimately vested in potential and domains centralized divine power standard annual and monthly village and ritual calendar filled with innumerable local rituals; ritual healing events and obligations; mass processions leadership assigned to knowledgeable multiple specializations of a large and laymen; male/female shamans diversified male professional priesthood universal marriage by individual choice elderly males and females control or arrangement; no divorce after intricate chastity and marriage childbirth policies; no individual choice bridewealth demands met by local dowry demands met by the bride’s descent group joint family extensive sociocentric and Ego-centric corporate descent units; joint families categories for affinal exchanges engaged in affinal exchanges consanguineal solidarity and affinal corporate class spirit reciprocity

spectacular gift-exchange events during lifecycle rituals exposing local groups in the process inconspicuous village-centred sites for multiple rituals

130 • Lewis Henry Morgan’s Comparisons

control of women by men, especially middle-aged men, and … control of young men by middle-aged men … [Thus] women and young men are dependent to a significant degree, ideologically or materially or both, on middle aged men … in the Australian case … with ideological domination well developed but with economic and political control restricted to a few limited contexts. (Woodburn and Barnard 1988: 28)

The particular anomaly of the Australian case lies in the fact that the aborigines did have: above all … a range of varied, committed, binding social relationships of kinship or of contract in which the participants have formal obligations to each other. (Woodburn and Barnard 1988: 28)

Riches (1995), on the other hand, tried to overcome the simple dichotomy, differentiating the particular stages of transformation and ending up with three different types of hunter-gatherer societies, the third being the ‘sociocentric structure’ type of the Australian aborigines. In similarly ambivalent terms, Layton classifies the latter as ‘part immediate, part delayed return societies’ (2005: 139). For him, technological complexity and territoriality constitute different axes of variation. His juxtapositions clearly indicate that the Inuit command a complex technological equipment, even though the social organization of this relatively small population in the arctic environment is, as can be expected, of a simple kind, whereas the Native Americans of the Northwest Coast have, along with their complex techniques of fishing, hunting, harvesting and storing, developed a highly differentiated social structure within a variant environment blessed by an abundance of natural resources. In Australia, however, the contradiction has no easy explanation. The question remains as to why the spiritual and the social worlds of the aborigines are so complex, while the technological base and the style of consumption are so simple. As far as I can see, classifying different types of hunter-gatherers, especially those of aboriginal Australia, is an important but a yet-to-becompleted task, since Woodburn’s main argument (‘control of women by men’) is simply another way of indicating that aborigines foster gender relations that may otherwise be found in food-producing societies, but not among hunter-gatherers of immediate return systems.

Before European Occupation Prior to 1606, when the first Europeans are known to have reached Australian soil, the life of the Indigenous peoples must have been rather

Social Evolution and the Australian Anomaly • 131

different from that of subsequent periods, after imported measles and smallpox epidemics and also tuberculosis decimated the population. Unrestrained racism reduced aboriginal life and liberty after 1770, the year when James Cook initiated British occupation. By 1900, only 93,200 native Australians remained.12 Recovering glimpses of the Australian past before the European onslaught seems to be facilitated by the fact that, as most experts agree, genetic research indicates only a single immigration wave comprising 1,000‒2,000 modern humans entering the continent some 50,000 years ago or earlier (Williams 2013). Subsequently, the southern continent remained isolated. Even if a second wave could be discerned, geographical preconditions ruled out the many movements, migrations and amalgamations of peoples observed in other parts of the world. The habitat in general must have changed several times compared to contemporary conditions, since during an early phase it had supported a comprehensive and diversified megafauna, i.e. large vertebrates with mature individuals weighing more than 40 kg, that disappeared some 13,500 years after the first arrival of humans on the island, though shorter periods of coexistence are indicated for some regions (Saltré et al. 2016). The extinction of these significant sources of food has been the subject of controversial debates on human or nonhuman causes, which is beyond the scope of this chapter. My subsequent data on greater prehistoric population changes are taken from Williams (2013) and indicate that 18,000‒21,000 years ago, the Last Glacial Maximum must have had a far-reaching impact, since population declines of up to 61 per cent are observed. Only 9,500 years ago, the pre-LGM numbers were reached once again. During the Early and Mid-Holocene, the largest population increase occurred, beginning 12,000 years ago and reaching its peak in the Late Holocene, which coincided with a general climate stabilization. Then evidence suggests an exponential growth as part of a long process of rising population, beginning in the Terminal Pleistocene. The start of this growth phase seems to have coincided with a rising sea level and a strengthening of the northern monsoon as continental shelves were flooded. At the time of initial European contact, the population size is estimated to have been in the order of 770,000 to 1.1 million. The highest estimates of about 1.2 million relate to the time some 500 years ago in the early sixteenth century . A regional study in Western Australia conducted by Slack et al. (2017) concludes that the permanent and prolonged use of the more marginal or ecologically less favourable foraging environments of the interior plateau is a comparatively recent development in the region’s long archaeological record. If this finding is generalized, it would mean that the ­population

132 • Lewis Henry Morgan’s Comparisons

of the preoccupation era was concentrated in the fertile parts of the continent, which were later monopolized by European immigrants. To me, it does not appear to be too far-fetched to assume that aborigines, when being pressured by Europeans into the inhospitable regions of Australia, were forced to drastically reduce the previously regular size of their groups and give up any technical equipment that would only be of use in a more fertile environment. At the same time, the aborigines retained their complex ritual and territorial organization, which turned out to be extremely important for survival, when, with changing phases of scarcity and sufficiency, groups could, on the basis of ‘kinship’ relations, disperse or unite in cooperation, as Mervyn J. Meggitt (1962: 49‒50) described in great detail for contemporary Walbiri. As to prehistoric rituals, on 26 February 1974, the discovery of the Mungo Man only 400 metres away from Mungo Lady (found in 1969) created a sensation. Archaeologist Jim Bowler found the skeleton remains of these two individuals in western New South Wales and could later report (Bowler 2014) that these fully modern human beings, eventually firmly dated as having lived 41,000 years ago, had a formerly unimagined cultural significance, since the 170 cm tall Mungo Man had been buried in a carefully prepared 80‒100 cm deep grave and, more importantly, since ochre had been painted on the body or sprinkled over it, a substance that could not have been found anywhere in the local or more distant vicinity of the site, but must have been imported from afar. Accordingly, regular group contacts over very long distances must be assumed to have existed at that time. Mungo Man must have been buried in ceremonial style, indicating a generally complex development of spiritual life.

Conclusion Relationship terminologies are not produced by some individuals for a specific purpose. They evolve and change over centuries or millennia and do not dissolve following the transformation of external conditions, because they are understood by their users to be based upon natural relationships. Due to this longue durée, they are valuable for students of anthropology and incomprehensible for others who reduce their efforts to the study of observable individuals in action. In Australia, such terminologies point out a significant social complexity. Dating from the late nineteenth century onwards, anthropological studies of Australian aborigines have found that small and mobile bands made use of only very simple hunting and gathering techniques

Social Evolution and the Australian Anomaly • 133

associated with immediate-return systems, but archaeological research reveals that before the epoch of occupation, several major changes to the natural environment must have occurred and that even under the most recent ecological conditions, Indigenous people were likely to have inhabited the fertile parts of the continent rather than the least hospitable regions into which they were later driven by the colonizers. When, for the first time, Europeans landed on Australia, the size of the Indigenous population was almost ten times as high as it was in 1900. Also, during the last 41,000 years, aborigines certainly led a complex ceremonial life. Given these data, the refined Indigenous social categories based on ‘kinship’ are likely to have been developed in periods witnessing the relatively easy acquisition of natural food resources and allowing (when compared to occupation time) numerically higher and more permanent assemblages of humans. If, under such conditions, people developed complex social and spiritual categories and preserved them even after the invaders had achieved dominance, such a history would hardly be extraordinary. It is similarly observed on other continents, when favourable ecological conditions led to Indigenous complexity before the European onslaught terminated such developments altogether. Given these circumstances, the existence of a highly differentiated system of aboriginal ‘kinship’ can be explained by the same arguments that apply to the situation of precontact Native Americans on the Northwest Coast. Classifying Australian systems as delayed return only on account of the fact that adult men physically and ideologically control women and young men does not appear to be justified. In other settled societies (e.g. Highland New Guinea), a similar kind of male dominance is observed, though the food supply is much greater compared to that in arid Australia. In the colonial period the historical transformation to conditions of extreme material hardship may even have aggravated the particular aspects of any given inequality between men and women. Briefly, the case of the Australian aborigines seems to suggest that historical research ‒ i.e. that into ‘specific evolution’ (Sahlins 1960: 19) ‒ offers a reasonable explanation for the assumed anomalies in the context of a ‘general evolution’ to ‘normalize’ the empirical finding.

Notes  1. Fison happened to be on leave in Britain at the time.  2. As far as I can see, my above interpretations of the Seneca and Omaha systems as being based on the value of affinity have not been mentioned before.

134 • Lewis Henry Morgan’s Comparisons

 3. A most impressive ethnographic monograph featuring all of these details is that by Lester Hiatt (1965).  4. My own German greatgrandfather, a life member of the Frankfurt am Main City Council, married his first cousin. Adam Kuper (2009) argues that such ‘close’ marriages played a crucial role in the rise of the English bourgeoisie.  5. The German original was ‘Der Ursprung der Familie, des Privateigentums und des Staats’.  6. See, for example, Service (1962); Fried (1967); and Sahlins (1972).  7. The ‘conical clan’ (Kirchhoff 1968: 371f.) is a form of classification based upon pedigree and the seniority principle, but without the rules of exogamy.  8. A chiefdom has an established and absolute political leadership, and thus lacks the usual criterion of Indigenous anarchy, though – unlike the state – it may very well exist without a clearly delimited territory.  9. The authors themselves rejected this appellation: ‘Our perspective is plain old evolutionary, not neo-evolutionary’ (Sahlins and Service 1960: 4). 10. In many parts of the world, the so-called ‘green revolution’ during the 1960s and 1970s meant that landlords removed peasants and tenants from ‘their’ respective estates in order to re-employ a small number of the cultivators as wage-labourers who handled modern machinery and other agricultural investments, with the result that owners obtained more funds for luxury goods or for investments in innovative industries and services. 11. Karl Polanyi (1957) is of course the major theoretician when it comes to the comparison of the systems of reciprocity, redistribution and market exchange. 12. Retrieved 4 February 2019 from creativespirits.info/aboriginalculture/ people/aboriginal-population-in-australia. 13. The concept was adopted from Nurit Bird-David (1990).

8 Order in Anarchy HMI ‘Gentile Organization’ Compared

Ancient Society As a lawyer, Morgan paid considerable attention to those Iroquois who had become chiefs due to their personal qualification, to the hereditary ‘office’ of sachem that was passed on within the gens (clan) and also to the councils of these dignitaries (1877: 68). He even chose the phrase jus gentilicum for some Iroquois social norms he had listed (1877: 67), but in Ancient Society, he primarily discussed conjectures of an evolutionary scheme rather than empirical conditions. He himself emphasized these objectives: A council of Indian chiefs is of little importance by itself; but as the germ of modern parliament, congress and legislature, it has an important bearing upon the history of mankind. (1877: 107)

Compared to Systems of Consanguinity and Affinity of the Human Family of 1871, Morgan’s most popular book did not reach the same level of analytical intensity and originality. However, as a significant difference, Ancient Society (1877) was less technical1 and thus more readable for Marx as well as a broad spectrum of educated nonspecialists, even if, as the basis of its major argument, it retained the insights derived from the terminological analyses of Systems. Nevertheless, its conjectures on the periods of human evolution ‘from savagery to civilization’ were bound to fade away with the demise of evolutionism as the dominant discourse. For several reasons, Morgan had positioned the Iroquois in the ‘Lower Status of barbarism’ (1877: 66), but in the due course of time, such ranking turned out to be an irrelevant exercise. However, he did discuss a society in which the ‘state did not exist’ (1877: 64) and named it a ‘gentile organization’ (1877: 60) because it was based upon an assemblage of gentes or clans. The gens (clan) was defined as ‘a body of consanguinei descended from the same common ancestor,

136 • Lewis Henry Morgan’s Comparisons

distinguished by a gentile name, and bound together by affinities of blood’ (1877: 61). Though none of the Iroquois clans introduced by Morgan presented such a ‘common ancestor’, I see the more influential misconception in the proposal of the gens (clan) as a ‘body’, i.e. a corporate unit. The disastrous demographic decline of the Amerindians after their contact with the European invaders must have significantly reduced the sheer number of clan members to a finite or measurable quantity of individ­uals, or one allowing the European descent bias to divert ethnographers from the main characteristic of the clan as a sociocentric category of affinity. These two aspects, i.e. the demographic catastrophe caused by Europeans and their Abrahamitic bias of ‘the common ancestor’, may have been responsible for Morgan’s rather bloodless description of his ‘gentile organization’ in Ancient Society, in which he dwelt on corporate groups of higher and lower orders, i.e. phratries and clans, but did not really explain how they operated in order to achieve social cohesion. Only after staying with the Indigenous peoples of contemporary Middle India was I able to obtain notions of empirical modes of ‘kinship’ relationships as the basis of a society without rulers and ruled. However, further fieldwork is a pressing necessity, since the available data on this region are of different standards and epochs in the discipline. Many existing details on the HMI social order are as yet unknown, just as many large anarchic societies of the region have never been discussed at all in anthropological accounts. In the following I will offer a kind of stocktaking of HMI social units and at the same time will try to show how these are able to tie together a people without a state or an administration.

Anarchy For contemporary Westerners, anarchy as an empirical feature is a rather unreal scenario and I am not sure whether it is discussed at all in modern political science. At present, people without a government are unthinkable ‒ or so it seems. The following will try to introduce an anarchic pattern of societas that has, during the past millennia, amounted to public order. To a great extent, this continues to be so in spite of the ever-increasing inroads by the state and the economic forces mobilized by civitas. In 1980‒81, during my initial phase of fieldwork in highland Odisha, ‘development’ had just been initiated. In the first thirty years after Indian

Order in Anarchy • 137

independence, a huge bureaucratic apparatus for the administration of the Scheduled Tribes (STs) had been created, but the people concerned had been largely left alone, whereas from the mid 1970s onwards, the ‘uplift’ of so-called Primitive Tribal Groups (PTGs) commenced in the form of major practical interventions. In Odisha, these twelve PTGs were at the top of the ‘development’ agenda. Their members had never been questioned about this process, nor were they ever consulted during its course. To this day, communication between the Indigenous people and their developers is of the indirect kind: roads and houses are being built; children may be herded into schools; and businesses are being set up in previously nonexistent towns. Primarily, the massive immigration of lowlanders is altogether changing the scene. At the same time, at least the members of the so-called PTG seem to continue village life as before. When the following sketches discuss relational patterns in their midst as are relevant for their public order, and also those of other STs in the culture area, it should always be remembered that the household is the navel of their social existence while ‘the tribe’ has never been a concrete unit or a formation involved in any kind of activity. The family is the most sensitive and organizationally relevant unit of the anarchic social order, followed by the local group in the village and its supralocal network. By degrees, the importance of a given unit diminishes with its increasing distance from the basic cell. However, my European bias will force me to invert this priority and initiate the description from the most exterior boundaries of the Indigenous world.

The Region Introducing the sociocultural conditions of more than 100 million aboriginal inhabitants living scattered across the vast highlands in the least hospitable parts of India is a difficult venture. Compared to mainstream caste society, the people called Adivasi in innumerable administrative documents in Indian languages and by the many anthropologists of the subcontinent are a separate cultural entity, the study of which is very much left to projects of provincial university departments and Tribal Research Institutes situated in the lowlands and staffed by caste Hindus. The very existence of the culture area is next to unknown to most non-Indian anthropologists.2 The largest among several zones of Indigenous South Asians can be demarcated by gross references to the highlands, and only these, between the Ganges/Yamunda valley in the north and that of the

138 • Lewis Henry Morgan’s Comparisons

river Godavara in the south. On the eastern boundary, the narrow and densely populated lowland coastal belt on the Bay of Bengal is inhabited by the Hindu peasantry, as are the valleys, carved out by major rivers flowing eastwards to the sea. With the sudden rise in altitude and decline in fertility, the land abruptly changes to the zone of the Indigenous people. Its western boundary is not comparably ­unequivocal. Adivasi classified under the generic name Bhil live in the Aravalli Hills and are also broadly scattered in small hamlets across the lowland deserts, even over the international border with Pakistan. With or without detailed data on Indian history, the malaria-infested hills, which until very recently lacked roads and towns, are easily identified as ‘retreat zones’. During past millennia, whoever was defeated, conquered or bullied by superior forces could withdraw into the forested highlands, as India had always been an underpopulated and partitioned land. Many states emerged and disappeared. From the eighteenth century onwards, British colonial powers entered even the remotest corners of the subcontinent to exert direct or indirect control, confiscate the forest and prohibit shifting cultivation. The idea was to introduce plough agriculture (meaning more revenue) everywhere. Newly instituted proprietors had to pay ‘rent’ to the crown and, with a considerable increment, collect it from those tilling the soil. European policies were believed to bring moral and material progress3 and yet, again and again throughout the nineteenth century, gatherings of injured and insulted Adivasi, armed with bows and arrows, rose in spontaneous rebellions4 to be crushed by colonial forces,5 while the peasantry of the lowlands continued in a state of submission to whomever was in power. Among other reasons, the highland rebellions probably motivated the colonial government to create special ‘reservation’ laws for Adivasi. The constitution of independent India and successive governments since 1947 continued such a policy of reserving seats in national and state assemblies6 and in the higher and lower administrative services for those who formally qualified, and also scholarships in educational institutions. Moreover, land owned by a member of the Scheduled Tribes (STs), as defined by the state, could not formally be alienated by non-Adivasi unless, of course, the state wished to exploit such land or pass it on to national or multinational companies. However, only some Indigenous people are classified as STs, while others rank among the many different Scheduled Castes (SCs) or Other Backward Classes (OBCs). They may also appear as STs in one province, but not across its boundary in another, where they may be defined as SCs or OBCs.

Order in Anarchy • 139

Accordingly, it is not really advisable for anthropologists to rely upon these official categories.

The Unmentionable Neither the many British colonial ethnographers nor the three outstanding researchers during that pre-independence epoch (Roy, Elwin7 and von Fürer-Haimendorf) and not even contemporary accounts mention the regular village neighbours of the landowner-cultivators in the highlands. These so-called weavers should make up about a third of the Indigenous inhabitants of an Adivasi village, where they usually live in somewhat separate quarters, unless their hamlet is situated within walking distance of the cultivators’ houses. Before the Berlin team,8 only Bailey (1960) and Niggemeyer (1964a, 1964b) have discussed these weavers who, under one name or another, live in the south9 of the Indigenous zone as craftspeople, petty commercial agents, cattleherders and field-labourers, but never10 as owners of the soil. I have personally been with the Arsi Sora clients whom Suryanarayan (2009) lists as weavers having ST status along with other Sora who are landowning cultivators, but normally the Indian state classifies landless highland people SCs, thereby joining them in a very odd administrative assemblage with the lowest Hindu castes of the plains who belong to a very different society. F.G. Bailey referred to weavers of the Kond highlands as ‘untouchables’ (1960: 121) and ‘beggars’ (1960: 130), generally classifying them as ‘dependants’ of the landowning Kond (1960: 121f.). Following the materialist approach of the Manchester School, Bailey came up with the thesis that Kond inevitably require additional workers for their wet-rice cultivation during the very short and labour-intensive transplantation phase. For this reason, Bailey argued, the Kond retain the craftspeople all year long with other – basically unnecessary – assignments (1960: 133‒34). The thesis is a neat one but is not entirely convincing, since the same weavers also reside in many Kond villages practising only shifting cultivation, mostly of millet, or without being involved in transplantation work. The German ethnographer of the Kuttia Kond, Hermann Niggemeyer, in a rather different style describes the weavers as rich and ruthless moneylenders of the poor hard-working Kond, emphasizing the cultivators’ ‘pressing dependence’ (1964a: 17) on their alleged exploiters. Twenty-five years after his fieldwork, I have stayed in the same smallest administrative unit as this author, initially living in a Kond hamlet and

140 • Lewis Henry Morgan’s Comparisons

on a later occasion among weavers, to realize that both Bailey’s and Niggemeyer’s rating and prose are untenable exaggerations. However, the latter’s monograph (1964a) contains literal accounts of very important myths indicating that major Kond rituals cannot be conducted without essential functions being assigned to weavers. In a different area further south, I found these clients even among the most withdrawn, conservative and xenophobic Bondo shifting cultivators, a society known for its incredibly high homicide rate (Pfeffer 2008) and one that would be most unlikely to accept any kind of outsiders in its midst. Perhaps the ethnographic policy of simply ignoring the existence of these craftspeople in the highlands is related to traditional European concepts of a ‘tribal society’. The idea of some ‘pure’, ‘untouched’ and ‘self-sufficient’ people continues to linger on. In Middle India, however, the weavers and occasionally other noncultivators are regularly found among the aboriginal landowners. Of these, some also till the soil, but specialize in cattle-raising, though they consume neither the milk nor the meat of their cows (see Berger 2002). It seems highly plausible that the owner-cultivators in the highlands have never been isolated and on their own. I am not aware of any convincing hypothesis indicating a relatively ‘recent’ immigration of the weavers. The very fact that the Kond and the many other Indigenous masters of the soil could preserve their own local economy, religion, language and social structure seems to be a consequence of their symbiosis with equally Indigenous commercial agents. All highland cultivators do their own woodwork, construct their own houses and slaughter their animals to consume the meat, but they do not exchange manufactured or nonmanufactured goods for material gain and would lose status if they did. The weavers used to produce cotton blankets as well as equally coarse skirts for women and loincloths for men. During the 1980s, I could still buy such articles and find looms in two or three very remote villages,11 but even at that time almost all had been abandoned, since cheap and colourful mill-made textiles had been brought to the weekly markets by external hawkers or were offered by government agents at extremely low prices in a major effort to change the appearance of the ‘naked primitives’. However, the weavers continued to live and work in the same villages. Their essential role seems to be that of petty traders, though they supplement it by several crafts to earn a little from producing baskets, mats and the like by processing a variety of wild plants. Though their social status is much lower than that of the ownercultivators, whose houses they must not enter, all the same they remain indispensable and interact in many ways with the lords of the land ‒ for example, in the regular drinking rounds.

Order in Anarchy • 141

During the 1980s and in a rather remote area, I met a middle-aged Kond who mentioned that he had never touched money. Selling or buying products and similar commercial activities used to be inconceivable for a free cultivator, as were regular services for others. Today, market forces are gradually entering the hills. To a considerable degree, cash is being used in Chota Nagpur or other regions where outsiders have been in command for more than 150 years, and locals are familiar with employment facilities and salaries. At least in the area to the south of the Mahanadi River, weavers continue to act as the only brokers of goods later given in ritual exchanges and they also advance bridewealth payments. For metal tools and implements, they have ‘always’ moved to the lowlands and even into metropolitan India to obtain what they knew would be required and affordable in the households of hill cultivators. In return, and again for a commission, they continue to take grain and animals to the weekly markets on behalf of the owners, though cash is never involved in these inner-village dealings. Weavers are skilled in working with numbers, which the lords of the land are not. Probably the brokers have ‘always’ insulated the anarchic world from the brute forces of commerce that have, for millennia, been operative in the lowland economy. In the event of obviously exceeding their brief, members of the ferocious cultivator majority in a village physically punish their agents to make up with them in due course. Very few of the weavers I met were operating as rich bankers. A few others were destitute village servants. The vast majority of them I found to be as poor as the owner-cultivators of the land. A society based on class could not be discovered in the hills. The prime asset of these agents is a swift mind. In a lasting symbiosis, some villagers control the land while others handle commercial exchanges. Economically, the two sides depend heavily on each other, but people do not marry across this status line.12 Significantly, the Sora include endogamous craftspeople of several specializations as well as endogamous cultivators and both units are assigned ST status by the state. In southern Odisha, it goes without saying that weavers belong to the category desia, meaning ‘people of the land’, and not to that of odia, meaning ‘people of the coast’. In the northern highlands, I have never heard of them being included in either the category horo, meaning ‘humans’, or that of diku, which is applied to the feared and despised people of the mainstream caste society. Whereas lowland speakers of the Hindi, Bengali and Odia languages apply a descriptive kinship vocabulary, in the hills both cultivators and craftspeople operate with a classificatory terminology in their many different languages.

142 • Lewis Henry Morgan’s Comparisons

Indigenous Cultivators Throughout the culture area, people assigned to the status of a certain ‘tribe’ as such rarely live within a clearly delimited territory, though they do have their respective centres of gravity in some highland zones. Since most of my work was confined to the eastern province of Odisha and the adjoining areas13 in the states of West Bengal, Jharkhand, Chhattisgarh and Andhra Pradesh, the following discussion will mostly refer to the anarchic societies in these regions. The names I have given to the different STs are those used in official documents and almost all of the literature, whereas for themselves, people tend to apply the first person plural or one of many terms that could be translated as ‘humans’ in one language or another and that may or may not have become the administrative appellation. Many official names can also be translated as ‘Children of the Earth Goddess’, ‘Earth People’ and the like. I have chosen these labels of the Census of India to facilitate a cross-checking in the anthropological literature that would otherwise be impossible.

The Bhumij/Santal Complex The hills in the northeastern Mayurbhanj district of Odisha and the adjoining West Bengal are home to more than a million Bhumij. The Santal settle as their neighbours in the same area and in many others. Several millions more assigned to this ST live further north in the eastern part of Jharkhand. All Santal are affinally interrelated by twelve sociocentric exogamous clan categories (paris) carrying totemic labels of plants, birds, other animals, and planets. These labels seem to assign different environmental altitudes,14 though the people themselves are not aware of any kind of structural implications. At least in some areas,15 villages are simultaneously subdivided into local descent groups carrying the status (khūt) of sacred or secular leaders (Gausdal 1960: 51f.). Lea Schulte-Droesch, after extended research in southeast Jharkhand, mentions gusti as the agnatic descent group of a given locality. The members ‘sacrifice in the same sacrificial circle’ (Schulte-Droesch 2016: 124, 129) to assemble for their own specific rituals devoted to the deity of their land, since myths16 make them aware of a common past in a particular region. They conceive of each other as the ‘people of one house’ (2016: 131f.). Such a local descent group includes smaller agnatic units spoken of as ‘people of the house’ (2016: 132), with the latter living together to share work and

Order in Anarchy • 143

income. Like these locally based agnates, all affines (pera) belonging to villages in the vicinity regularly participate in many common rituals. In these external settlements, all members of the same community – but belonging to another clan – are understood to be affines. The word also means ‘guests’. But not all affines are empirically selected for marriages. In the local and regional context, some may be avoided due to the relatively junior status of their clan,17 while others may have been involved in mythical conflicts and are thus ruled out (2016: 111). Schulte-Droesch also clarifies that the egalitarianism of the Santal in matters relating to power goes together with the superior status of relative seniority and the respect owed by juniors (2016: 113). According to this notion, I have noticed that Bhumij18 are collectively regarded as the seniors of the Santal wherever members live in the same village or in neighbouring settlements. Though both of these communities equally comprise cultivators of a high status, wielding a dominant influence in the region, the Bhumij prefer cross-cousin marriage, while the Santal are horrified by the very idea and arrange alliances only with families that are known not to have been linked by marriage in previous generations. It should be obvious that Bhumij and Santal are not ‘downtrodden’,19 but are self-confident and influential settlers in their home districts, as long as they have retained a life as cultivators and are not forced into taking up employment in India’s heavy industry sector. The cultivators would hardly be distinguishable from Hindu peasants had they not retained the rituals of their own localized religion (including the very different dietary rules), their political egalitarianism and the many particulars of their complex ‘kinship’ organization.

The Bhuiyan/Juang Complex In the hilly Keonjhar district of Odisha, the Indigenous Bhuiyan are said to be the seniors of the Juang and the social organization of both seems to be rather similar, though the data on the latter, collected by Charles McDougal (1963), are by far more informative than those on the former. For example, S.C. Roy (1935) does not mention the important institution of their men’s house, but in 1981, I personally saw one and discussed it as a centre of community life with Bhuiyan consultants. I have also stayed in a majang of the Juang. Such a building, which is bigger and more impressive than an ordinary house, used to be the ritual and secular meeting point for men in every village before the Indian government in the late 1980s constructed an interstate highway right through the heart of Juang territory, with the result that shops

144 • Lewis Henry Morgan’s Comparisons

catering for Punjabi truck drivers20 have terminated much of the former community life. The Juang used to have their own territory subdivided into four units of neighbouring villages called pirh, in each of which the respective elders used to join together in councils to solve internal disputes and raise ‘rent’ for the colonial creation of a traditional Indian king21 representing Keonjhar state (McDougal 1963: 58‒59). It may or may not be possible that such a fourfold territorial division pre-dated the British imposition of a formal administrative centre. McDougal’s census records many agnatic local descent groups in each village and also 3‒11 clans as their inclusive units (1963: 67). Altogether, he mentions eighteen of these named totemic sociocentric categories, with each being present in several villages (1963: 3‒4) of a given locality so that a territorial organization is implied, even if individual clan members at times lived in ‘foreign’ villages. A single clan dominates a settlement and its respective elders fall into this category. ‘Genealogical reckoning is shallow’ (1963: 70) is an important statement by McDougal, even if, in the heyday of the Africanist theory on ‘segmentary societies’, he attempted to discover segmentation processes, but ultimately remained unsuccessful. Each village and clan is an exogamous unit and from an individual perspective, the entire Juang community is subdivided into categories called kutumb (consanguines) and bondhu (affines). An unrelated third party does not exist. All Juang are conceived as mutual relatives. Each local descent group is exclusively and distinctly classified by others as either consanguineal or affinal (1963: 75). From the same Ego-centric viewpoint, residents of all villages are either one or the other (1963: 130‒40). The other ‘highly developed’ aspect of Juang social structure is the ‘principle of the equivalence of alternate generations … All generations are divided into two groups’ (1963: 140) or ‘generation sets’. In an egocentric perspective, one refers to adjoining generations and others of an odd-numbered ‘distance’, while the second ‘set’ includes the alternating generations as well as those of an even-numbered ‘distance’ form Ego. This twofold division does not depend on an Ego-based reference: the two generation-sets have independent existence in relation to each other. In certain formal situations the two generation-sets contained in a village function as groups ‒ not categories ‒ to which different rights and obligations are allocated. (1963: 141)

Same-sex members of one generation-set entertain joking relationships with each other, as do other-sex members, as long as they are

Order in Anarchy • 145

affines of one another (McDougal 1963: 144‒46), while ‘opposition and conflict’ marks relations between adjacent generations, even though ‘the biological age composition of a generation-set is heterogeneous’ (1963: 146). This strict opposition is in the same way valid for marriage rules: a person must always marry someone belonging to an affinal (bondhu) village and one of the same generation-set (1963: 155). For a man, ‘there is a marked tendency to avoid those groups from which his close agnates of the adjacent ascending generation ‒ father, father’s brother, and father’s sister ‒ have obtained spouses’ (1963: 158). At the same time, ‘any village maintains relatively continuous alliances with only a small number of other local units’ (1963: 169‒70). McDougal offers elaborate statistical evidence for all of these statements. For the Hill Bhuiyan in the adjoining territory, Roy does not mention any clans at all, though Lakshman K. Mahapatra (1959: 176) does elaborate on them, and Bhuiyan consultants enumerated some to me. In no uncertain terms, Roy notes that ‘totemic organization is absent’ and instead of clan exogamy, village exogamy is the rule (1935: 80). In fact, both Roy and Mahapatra may have reported correctly what they had found. Different sections of the Bhuiyan may follow different rules. According to Roy, all of their hill settlements, like all those of the Juang, are either classified as consanguine (kutumb) or as affine (bondhu) for an individual speaker (1935: 81). The territorial unit bar seems to be central and the same name is also used by the Kisan,22 who settle not so far away to the west on the other side of the Brahmani River. For Roy, the bar concept is ‘superimposed on this village organization’. The entire territory of the Bhuiyan is subdivided into bar units, each ‘consisting of three to twelve or more units’ (1935: 93). The elders of the villages constitute the council or ‘bar-panchayat’ (1935: 95), apparently having several servants,23 officious procedures and the common tutelary deity of ‘some prominent hill or mountain’ (1935: 98) of each bar. As a point of comparison, Roy mentions cross-cousin marriage for the Bhuiyan (1935: 142) in a general fashion (‘as among most other tribes of Munda stock’), while Mahapatra states the opposite (‘a crosscousin is considered equivalent to a brother or a sister’) and notes that even second cousins are excluded from marriage (1959: 179). I see absolutely no reason to doubt either statement. It may very well be possible (as I have frequently experienced) that Bhuiyan in one corner of their territory have opposite marriage rules compared to those of another locality. In both cases, the point as such ‒ i.e. the maintenance of affinal alliances ‒ is discussed with great vigour.

146 • Lewis Henry Morgan’s Comparisons

The Chota Nagpur Complex The best-known, though not necessarily best-researched, Indigenous complex is that of the Chota Nagpur plateau in Jharkhand province and the northwest of Odisha. Though they tend to settle in villages of their own, all ‘tribes’ of this region live interspersed with each other and also with neighbouring ‘tribes’ (Santal, Gond) on the eastern and western fringes. The mutual status of these many units is classified by the concept of relative seniority in a narrower and a wider sense. Generally all Kharia like to mention their senior status compared to the ‘medial’ Ho and Munda or the junior Kisan and Oraon (see Table 8.1), and others seem to agree that, perhaps adopted from the Brahmanical model, the stereotype of ‘priestly’, ‘royal’ and ‘worldly’ characters are associated with the three major units. People of the region classify three main categories as Kharia, Munda and Oraon, although on a closer look, each one of these is further subdivided into two respectively endogamous segments of either senior or junior status, with the Kharia being split into the Dhelki and the Dudh branches, while the second category is split into that of Ho and Munda, with the pair of Kisan and Oraon completing the set. It is of utmost importance to note that such a hierarchy does not affect power relations. It may also be reversed in some regions and it is not in any way reflected in the population figures. Presumably, both Oraon and Munda have more than 1.5 million members, Kisan and Ho more than a million, and Dhelki as well as Dudh Kharia approximately 500,000 each. Kisan and Oraon speak two different Dravidian languages, while the common Kharia tongue and also that uniting Ho and Munda belong to the Munda language family. The Kharia are frequently confused24 with small hunter-gatherer bands known as Hill Kheria, but the latter have nothing in common with the distinguished plough-cultivators who, over the past 150 years, have sent many children to Christian missionary schools. In such cases, the alumni became prominent academics and administrators. Roy and Roy (1937: 143‒44) mention many of their clans, but my repeated and intensive inquiries show that eight agnatic totemic categories are ‘the Table 8.1 Indigenous seniority concepts of the Chota Nagpur complex.

Senior Junior

Senior

Medial

Junior

Dhelki Kharia Dudh Kharia

Ho Munda

Kisan Oraon

Order in Anarchy • 147

original ones’. They make up all Dhelki Kharia, while nine clans (some of which having the same names as Dhelki ones) constitute the Dudh Kharia (Pfeffer 1993). The two Roys rightly mention the tradition of original villages (buinhari) associated with each clan within the given subregion (1937: 133) and among the Dudh Kharia, I was able to meet several ‘kings’, who were said to be the formal heads of such clan territories. There should be twenty-two such lands. The so-called ‘kings’ were neither wealthier nor more powerful than their neighbours and personally cultivated their land, but their status advantage may not have made life more difficult. Two local descent groups of different status but the same clan must have been tied to the earth in every village, as the name buinhari conveys. One of them would carry the royal title of Raja, the other the priestly title of Pahan. Roy (1912: 64, 69) mentions the same kind of local units carrying different status among the Munda and also the Oraon, the separate burial ground of a local descent group (LDG), and also status barriers in matters of ritual commensality (Roy 1915: 107, 392). Apparently the clan and the subclan concepts are regularly confused by the author. ‘In the course of time’, Roy states, ‘as members of each kili [clan] increased … other villages were founded in the neighbourhood by … the same kili … In social and administrative matters, they continued to act as one body’ (Roy 1912: 235). ‘The villages by batches generally of twelve – but sometimes more and sometimes less – came to be grouped together as a patti with the strongest and most influential amongst the headmen of the villages as the Manki or patti-chief. The remaining village-headmen swore allegiance to the elected Manki’ (1912: 65). What Roy seems to have discovered is a subregional settlement pattern of the Munda clans and a specific unit of a finite number of villages, the name (patti) of which is well known everywhere as an administrative term for North Indian peasant territories, though (because of the many parallels in other highland societies) in the Munda case, it may refer to a relevant local unit from the days before the colonial state. Eva Reichel (2009, 2018) is an important authority, since she has conducted more ethnographic fieldwork among the Ho than anyone else in the entire Chota Nagpur complex. Speaking their language, she has also closely cooperated with the Jesuit scholar John Deeney, who lived and worked among the Ho for more than half a century. Her conclusion is that the Munda have many exogamous agnatic clans and, in view of a list prepared by Deeney, 133 of these relevant affinal categories are known among the Ho, though they are neither totemic nor corporate units. Several of them are frequently defined as ‘brother clans’ of each other and thereby exclude intermarriage between members.25

148 • Lewis Henry Morgan’s Comparisons

Categories of Ho subclans have totemic names and may have been split several times within the recorded history of their migration. It should be noted that a person of clan A belonging to a subclan carrying the totemic label of, say, heron is in no way prevented from marrying someone of clan B belonging to another subclan labelled as heron. Clans and subclans are sociocentric categories. Reichel also introduces an Ego-centric group called miyad mandi chaturenko (mmc) or ‘the people of one rice-pot’. Membership comprises four generations of Ego’s deceased forefathers and foremothers within an agnatic pedigree. As I understand it, all descendants of the four males above a male Ego belong to the mmc that includes the dead forefathers and foremothers within this generational span. The deceased continue to participate in ritual meals. After the death of any member, all cooking pots of all member households must be removed and replaced with new ones. Corresponding systems of this very important unit are likely to be found in many other HMI societies.26 Perhaps they have not been discovered so far because ethnographers have failed to look for them.27 Reichel mentions the Santal gusti (see above) in this connection because she is also quite familiar with Santal ethnography. Ethnographers are used to LDGs as ancestor-centric structures and are quite unfamiliar with a membership limit according to the genealogical distance to Ego. Reichel defines mmc as an ‘exclusive, specified, and definitive unit within a sub-clan or clan’ and also as ‘empirical, egocentric, transgenerational’ (2018: 113). It includes known cooperating agnatic individ­ uals, their recently deceased ancestors and all agnatic descendants of the latter. The names of all living and dead members are recited in ritual by genealogical reckoning up to four generations above Ego’s senior most forefather. Members need not reside in the same village because Reichel reports them as living in different settlements: ‘In the case of migration the recollected common last place of origin before migration – as a kind of unifying focus – will obtain special attention’ (2018: 114). The other important membership rule is that on the entry and exit of women. Due to their birth rites, female and male children of female and male members are included in the mmc. Thereby they obtain the right of receiving adequate funerary rites at the time of their respective deaths. However, upon their marriages, sisters and daughters leave the mmc, while incoming wives become agnates to substitute them. Only due to such a status transformation may a wife enter the sacred kitchen of the house that is closed to all nonmembers. Since such a group of cooperating individuals regularly changes the limits of its membership, it is a dynamic formation.

Order in Anarchy • 149

Locally, all Chota Nagpur communities conceive certain collectives of ‘brothers’ in opposition to those of ‘affines’, but the marriage rules differ. A preference for cross-cousin marriage is regularly mentioned in the literature and Reichel (2018) also reports such cases in great detail. However, I found strict norms proscribing cross-cousin marriage in some Oraon, Kisan and Kharia areas, but not in others, where alliances were frequently repeated from one generation to the next.

The Kond Complex More than a million Kond inhabit the Kondmals (Kond Hills) of central Odisha and the southern part of the province, while an equal number can be found in the adjoining districts of Andhra Pradesh. The complex is subdivided into several regional units, of which I found the Desia, Dongria, Kuttia, Kuvi, Mala, Pengo, Porja and Sita Kond speaking a Dravidian language of several mutually understandable dialects. The boundaries between these generally endogamous subunits may remain vague and are frequently crossed by conventions of intermarriage and ‘fraternal’ relations connecting several local descent groups. Each of these Kond units articulates the idea of intermarriage by a small number of agnatic exogamous clans, the names of which tend to partially overlap. When I stayed among the Kuttia, I also found a specific territory being assigned to each clan. This piece of land was not necessarily contiguous and ‘sisters’ children’, as a minority, were living in villages of the dominant clan. Regarding the northeastern Kond region, a prominent ethnographer stated the following: For any one clan, the rest are divided into two categories: those who are agnates (dada-koku: ‘elder brothers and father’s younger brothers’) with whom one may not marry, and the uterine and affinal kinsmen (seri ahpa loku: ‘bride-seizing-folk’). (Bailey 1960: 53)

Further west, near a place called Phiringia, I found another variety of affinal sociocentric constructions to show once again that what Morgan called ‘gentile society’ in fact implies a wide spectrum of highly differentiated and refined concepts of social order. In this case, ten different pairs of clans are distinguished. A single pair of two fraternal patriclans28 – one senior and one junior – will always share the control of a single territory under a double name. Members of any one of these two clans must not intermarry within their common territory, but if the same clan (with a different partner) is also represented in others of the nine pairs, a member of one clan may marry a member of the

150 • Lewis Henry Morgan’s Comparisons

other, provided the two clans involved do not both belong to either the senior or the junior category. As always, clans are not corporate descent groups, but agnatic sociocentric categories that are mutually connected through intermarriage. The most detailed and inspiring account of Kond clanship is given by Hardenberg (2018) for the Dongria and their neighbours, the Desia Kond. On several occasions during the course of his long-term fieldwork in the Niamgiri Hills of the Rayagada district, Hardenberg was able to witness and analyse their clan sacrifice for the Earth Goddess, a central ritual that all Kond seem to share. Each of the Dongria clans owns one or several clan territories (muta) consisting of a number of villages (Hardenberg 2018: 224‒25). Between these units, the sponsorship of the grand buffalo sacrifice, a ‘total social fact’, circulates every year along with multiple other ritual obligations assigned to separate villages. Hardenberg’s overview (2018: 586‒88) shows how the principles of affinity, descent, territory and residence are combined in the social organization of the Dongria Kond as follows: • A clan is the named agnatic exogamous category whose members must not kill the buffalo they offer as a sacrifice to promote the fertility of the muta land. • A muta is a localized sacrificial community of some clan members, an aggregation of villages, between which the sacrifice and the exchange of ancient29 sacrificial tools and functions are systematically circulated. • A nayu is a settlement of the dominant clan. When their turn comes, its inhabitants buy the sacrificial buffalos to sponsor the event. • A padari is a named village territory containing a stone arrangement, which on specific occasions is associated with the Earth Goddess, where the clan sacrifice is carried out. • A punja is one of up to eight intravillage status categories that are each associated with either sacred or secular functions and ranked by their relative seniority. • At the same time, a punja is an LDG worshipping the same stone of a village, i.e. a stone that is named according to the respective status category. For the temporal order of the sacrifice, the sponsoring village (of the ‘sacrifiers’) takes its turn within the same muta of ‘brother’ villages. Some of these latter villages will also perform other major ritual functions on this occasion. In organizational matters, the sponsors are supported by nonmarriageable ‘ritual friends’, whose status is hereditary, and also by

Order in Anarchy • 151

‘protectors’, who are classified as cross-cousins and as such are nonmarriageable residents of affinal villages because of their parents’ alliances with the sponsors. In relation to the final category, the ‘sacrificers’ will appear in the ‘total social fact’ to hack the animal into pieces. They are young men of affinal villages who are well suited for future marriages (Hardenberg 2016: 589). Thus, it is clear that the Dongria Kond absolutely prohibit cross-cousin marriage, whereas the Desia Kond in their immediate neighbourhood regularly practise such alliances, even though local descent groups of both Kond units frequently interact in the form of feuds or economic and ritual matters. Like the other formations, such an Indigenous social organization is highly structured in many respects and is somewhat different compared to the stereotype of a mere addition of gentes in a simplistic ‘ancient’ society.

The Peasant Complex During the short visit to the Bora Sambar region in the borderland of central Odisha and Chhattisgarh, I found fairly affluent peasants who were registered as belonging to STs named Gond, Binjhal, Kond and Sora. They were also well respected by the non-ST minority of mainstream Indians. The Binjhal30 were the biggest group of landowners in the area and had the highest status. I also found my Kond and Sora consultants to be completely unaware of major cultural features that characterize Kond and Sora communities elsewhere, though my Gond hosts vaguely knew the phratry they were assigned to by tradition, but not the meaning attached to it and certainly not the holistic order of Gond phratries. When compared to regular Hindu villages of the plains, the absence of temples in their settlements made the most significant difference. All inhabitants were school-educated individuals and many were locally active as representatives of their joint families. Some were also engaged in formal ST politics of their respective state assemblies. However, though their language was Hindi or Odia, they continued to apply a classificatory relationship terminology with the structure I refer to as HMI.

The Sora Complex In the southern Ganjam and Gajapati districts of Odisha and the adjoining parts of Andhra Pradesh lies the land that is home to more than

152 • Lewis Henry Morgan’s Comparisons

500,000 Sora. In March 1983, I spent some weeks with the Jara Sora settled around the Mahendragiri, the highest mountain of the region that figures prominently in Hindu myths. The Jara live on shifting cultivation, but have also constructed magnificent terrace systems for their wet rice. Suryanarayan mentions twenty-three different endogamous Sora subdivisions of the same kind (2009: 188), some of which are those of craftsmen such as the Arsi Sora weavers31 whom I came across on other occasions, so that the omnipresent commercial agents seem to be formally integrated as part of ‘the tribe’. Like all Sora I have met or read about, the Jara were quite unaware of clans. As all other ethnographers also certify, theirs could be called a gentile organization without Morgan’s gentes. However, Suryanarayan goes into some detail on their local descent groups32 called kulam marangi,33 which, like the above-mentioned Kond punja and similar units in Koraput, is always associated with status. The three categories stand for secular leadership (gamong), sacred leadership (buya) and ordinary people (parja). Members mostly seem to marry within their respective rank (Suryanarayan 2009: 190). From an egocentric viewpoint, any of these local descent groups is either consanguineal or affinal (2009: 191) and each is likely to be represented in several villages where it will be recognized by its menhir (an upright rock), which the members had erected at the time of the guar ceremony (2009: 193), the grand buffalo sacrifice. Like the Ego-centric group described above called miyad mandi chaturenko among the Ho, the Sora LDG also includes the deceased forefathers and foremothers (2009: 201) for whom the sacrifice is conducted. Behura (1997) gives some valuable information on this grand secondary funeral, though the entire pattern of its socioreligious relationships remains far from clear. To me, the Jara Sora mentioned a unit that was neither discussed by Elwin (1955) nor by Vitebsky (1993), the two most proficient ethnographers of the Sora. It was named khêja. Apparently, the expression had a double meaning. On the one hand, it was a territorial unit of several villages, though their number was not to exceed twelve. Consultants were quite concerned about this point. Whereas the above-mentioned Kond muta is part of an agnatic exogamous clan, such a khêja, they said, was an endogamous unit that included a good number of local descent groups. Residents were to marry within this circle of settlements, but had to avoid a cross-cousin marriage. For themselves and their neighbouring territories, my consultants were able to list the respective villages in great detail. On the other hand, each of the two rows of houses, separated by the only road or courtyard in the middle of a settlement, was also called

Order in Anarchy • 153

khêja. The southern khêja was the dole khêja and was associated with secular leadership, while the northern naike khêja pointed to sacred functions. Each of them should never contain more than twelve houses. In other areas, I visited Sora villages severely influenced by Hindu Vaishnavi or fundamentalist Christian U.S. missionaries and containing neither the binary rows of houses nor a concept of khêja. In yet another region of fairly affluent villages with scattered houses uphill near Parlakimidi, Sora Roman Catholics were unaware of the khêja order, but insisted that they could never entertain marriages with the same local descent group within the three generations that follow an earlier alliance. To summarize, the Sora lack the sociocentric affinal categories known as clans, but nonetheless order their socioreligious life according to the principle of affinity that connects local descent groups.

The Koraput Complex The highland plateau of the old Koraput district used to be viewed as the isolated southern subregion of Odisha and is somewhat cut off from the state administration in the lowlands on the shores of the Bay of Bengal. In order to travel from the provincial capital to the area, civil servants had to either undergo the strain of a rather rough two-day tour on the few narrow roads crossing the Kond hills or otherwise, on a longer route of about three days, take the coastal express into the ‘foreign’ or culturally rather different34 Telugu-speaking province of Andhra Pradesh to board the extremely slow narrow-gauge train westwards, which would eventually take them to the edge of the plateau. In colonial times, until 1936, Koraput had been the northernmost border region of the Madras Presidency and sold to the highest bidder who became the Zamindar, a pseudo-king with a petty court. Neither the British nor the Indian National Congress made their presence felt. Briefly, the region was very much left to itself. With about 90 per cent of the population being Indigenous,35 very little social change could be observed until 1992, when economic liberalization policies were introduced by the central government. The ‘Koraput complex’ should be understood as the milieu of this plateau, uniting what people call the twelve desia (local) categories. The latter probably include the Bhumia, Bondo, Boda Porja, Didayi, Dom, Dora, Gadaba, Joria, Mali, Matia, Parenga and Rona, each of which is counted separately by the Census of India. Others are culturally close to them, but are understood to be Kond. Some villagers speak Dravidian, others Munda languages and all communicate in the Indo-European

154 • Lewis Henry Morgan’s Comparisons

Desia, which some have adopted in its entirety. For coastal Hindus, this latter tongue ranks as a ‘corrupt’ version of Odia. Of the above-mentioned categories, the Rona, Boda Porja, Bhumia and Dora36 are understood to be the ‘seniors’ of the rest, with Gadaba, Mali, Parenga and Matia ranking as ‘middle brothers’ above the remaining juniors. Again, all Gadaba are slightly senior to the Parenga, though within the former unit, the more populous endogamous half called Ollari is regarded as junior to the Gutob Gadaba. Some of these communities (Bhumia, Bondo, Didayi and Matia) split into moieties connected by the formal rule of affinal exchange between such halves, a norm that is widely accepted. Empirically, however, full conformity to the rule is quite impossible due to demographic irregularities in Koraput and anywhere else in the world.37 Eight agnatic and exogamous affinal categories (clans or bonso) of the sociocentric kind are understood as such by all Indigenous people of the plateau, i.e.: (1) fish; (2) cobra; (3) cow; (4) bear; (5) leopard; (6) monkey; (7) vulture/ hawk; and (8) sun. Their meaning seems to imply relative space. The first totem is water-bound, the second earth-bound and the third landbound. I have seen a bear climb the lower branches of a tree which seem to mark the space of this fourth totem and also, on several occasions, I found a leopard resting on the upper branches which may have been assigned to this animal as the fifth totem. Monkeys, of course, habitually move from one tree-top to another and this may be regarded as their specific – or sixth – space, while vultures as totem no. seven are flying high but under the sun as the eighth totem. Thus, separate spheres of existence, defined by different altitudes, may be associated with the labels, and perhaps even an opposition of four meat-eaters and four abstaining, but no such notions are expressed by the people concerned. Whereas occasional inter-‘tribal’ marriages do occur, intraclan marriages are absolutely unheard of. All of these exogamous categories are known as such in the area, but not all are represented in each community. Thus, the affinal exchanges of the junior Gadaba utilize the eight categories, while those of the senior half operate with only four. The following text will be confined to the Gadaba, since the other Indigenous people are quite similar. Over twenty years, I have regularly visited their senior branch, and Berger (2015), after living for two years in one of their villages, has written an outstanding monograph that is the major source for the following account. The Gadaba pioneers who initially cleared the forest (the ‘earthmen’) for a village are conceived as members of its dominant clan (Berger 2015: 104) and are formally linked in a union of villages known as the ‘twelve brothers’ (usually of the same clan) and the ‘thirteen

Order in Anarchy • 155

affines’ or villages. The latter are regularly connected with the former through intermarriage (2015: 190f.). Again, the number ‘twelve’ stands for completeness rather than empirical units and the ‘thirteen’ means one more in addition to the others. The male elders of these villages meet in council whenever major breaches of ritual norms threaten communal life. The main ritual is the secondary funeral. As a grandiose Maussian ‘total social fact’, it systematically ties together such units by formal obligations of multiple ritual exchanges involving the deceased in the form of buffaloes and ancestors represented by menhirs (Pfeffer 2001; Berger 2010). Within a village, weavers and affinally related individuals of several clans will usually be residents, but all major decisions (mainly those of ritual relevance) are made by inhabitants belonging to the single dominant clan. These are formally subdivided into four LDGs (kuda) of differently named status, two of them being characterized by sacred functions and the other two by secular functions. All of these four local descent groups gather in council as ‘the four brothers’ whenever the members of the dominant clan will decide on village matters. I would like to emphasize that Gadaba belonging to the group named ‘sacrificers’ do not all perform this function, but let one or two of them execute it on occasion. Of the kuda named ‘great leaders’, only one or two elders might at some instance (if at all) speak to external authorities on behalf of the village community. In short, the formal membership in these titled local descent and status groups should not be understood as a regular and specialized empirical activity of the individuals known by such a name.38 Similarly, all four LDGs are hardly ever represented in each village, even though the council of those who are is named ‘the four brothers’. In larger villages, such an LDG or kuda is subdivided along genealogical lines. Either the smaller kutumb unit (Berger 2015: 576) or the larger kuda is concerned with both ritual and marriage matters and the respective financial arrangements involved. The regular preference for cross-cousin marriage means in practice that empirical affinal contacts of each group to several external groups of different villages can be observed over time. As far as I know, among the Koraput communities, only the notoriously noncommunicative and aggressive Bondo prohibit cross-cousin marriages.

The Gond/Halbi Complex Some seven to eight million Gond are found in at least six Indian provinces, so the homogeneity of their social organization and ideology is

156 • Lewis Henry Morgan’s Comparisons

an equally amazing and significant aspect. The name ‘Halbi’ stands for the large Indigenous population engaged in crafts and petty trades or menial occupations. On several occasions in several districts of Odisha, I came across Gond enclaves and, for longer periods, I visited villages of a unit comprising some 500,000 people officially called Koya. They are listed separately by the Census of India and inhabit the extreme south of the province as well as the adjoining districts of Andhra Pradesh. Only their name distinguishes them from the Gond units in their neighbourhood. As a result of some unknown incident in the past, some unknown British administrators, asking for their appellation, must have misunderstood the answer Koitor, meaning ‘Humans’ in Gondi. Thereafter, the odd version of this word has been retained as their official label. In every respect, the Koya are very much alike the so-called Bison Horn Maria Gond (Elwin 1943), their neighbours in the south of the adjoining Chhattisgarh province. For the present purposes, they need not be distinguished from the large unit of the junior Muria Gond further north (Elwin 1947; Gell 1992) in that state or from other Gond units in other provinces. In the Maikal Hills and further west in the Satpuria Hills, the Gond live interspersed with another large people called Baiga39 (Elwin 1939) or Bhumia (Fuchs 1960). Members of the latter are said to act as priests of the Gond and as such are ranked as seniors. The above-named people have been described in several major ethnographies, of which I will cover that of Grigson (1938), who was the chief administrative officer of the former Bastar state and was a very meticulous and critical ethnographer (e.g. 1938: 233‒34). At the same time, he acted as a compassionate advocate of Indigenous communities (1938: 236). In short, compared to the cliché of the frequently denounced ‘census-ethnographers’, he and his work represent the diametrically opposite quality. Koya and Bison Horn Maria, and also the Gond in Maharashtra (Buradkar 1947) recognize five exogamous affinal categories of the sociocentric kind that are assigned patrilaterally. The literature introduces them as ‘phratries’ and each of them includes several clans (katta). A clan is understood to contain living and deceased members. I have regularly come across several of the Koya phratries in the very concrete appearance on special memorial grounds where either menhirs (for some) or long, thick wooden poles (for others) represent the departed members. These material representations have been erected during grand secondary funeral ceremonies that are celebrated regularly. Each one of the five categories is represented by a totem that may differ

Order in Anarchy • 157

regionally and carry the following name: Marvi (goat or cobra; nine clans), Kuhrami (cuckoo; ten clans), Sodi (leopard or tiger; three clans), Markami (tortoise; twenty-one clans) and Kawasi (tortoise; eight clans). They have been carefully listed by Grigson (1938: 306‒7) who over a period of some years travelled in Bastar state ‘off the beaten track’ to discover them and their local conditions. In Bastar, the important difference compared to the Koya situation is that some phratries conceive of each other as fraternal in some territories, but not in others. At least in the Abujhmar Hills of southwestern Chhattisgarh, but also irregularly in other areas, clanship very much refers to territorial units, with members being attached to the respective earth (bhum) in a religious sense, so that bhumkal refers to their code of solidarity (Grigson 1938: 236). In the provinces further to the west and south of Chhattisgarh, the five-phratry system of totems is substituted by those referring to three gods; four gods, five gods, six gods and seven gods (Grigson 1938: 239), each of which is said to contain the same number of clans. In several publications (e.g. 1979), Christoph and Elizabeth von FürerHaimendorf have written at length about them and the specific clan cults that Stephen Fuchs (1960) associated with ancestral homelands. I came across this kind of numerical classification in a village of the Phulbani district in Odisha, where a former Zamindar (landlord) of colonial times had called in the Gond to execute the ‘rent’ collection for him among the local Kond. These ‘guests’ were diachronically connected with affines in the neighbouring Kalahandi district, where they continued to marry only into member clans of a single phratry belonging to an odd number compared to their own even number. They mentioned that the lowest number was assigned to the most senior phratry. Apparently the same kind of numerical system unites them in some way with their Pardhan weavers (see Hivale 1946).40 Grigson is one of the few ethnographers of the culture area who recognized the ‘poor memories’ (1938: 233) of the Indigenous people regarding their individual ancestors, or what I would call the ‘vertical’ classificatory dimension supplied by genealogies. Instead, he has elaborated on the ‘horizontal’ or affinal dimension of collective relationships. The Gond always consider all empirical contacts to other Gond people or units as those between either agnates or affines (1938: 237), i.e. dadabhai (literally FF and B) or akomama (literally MF and MB), known as those of sale and soira in the present Telangana province, where Christoph and Elizabeth von Fürer-Haimendorf (1979: 80) recorded the latter term for affines. These two words are also applied egocentrically to the respective phratries.41

158 • Lewis Henry Morgan’s Comparisons

Overview Table 8.2 below offers a summary of the institutionalized relations covering HMI ‘gentile organization’.

Conclusion During the years after the publication of Ancient Society (1877), a large number of outstanding anthropological monographs have demonstrated the finesse of intricate relational patterns assigned to anarchic societies based on what, following Morgan, one might call ‘gentile organization’. Morgan’s own design, being the pioneering effort, can hardly be criticized for the rather undifferentiated ethnography he associated with the Native North Americans in their alleged state of ‘barbarism’, since it only initiated empirical anthropology. Subsequent reports brought about the description and analysis of a wide variety of such anarchic systems as accomplished by excellently trained anthropologists in the British Commonwealth, the United States and other countries. Hopefully, this will continue. The above sketches were meant to encourage further work on ‘gentile organization’ in India and elsewhere. Apart from the Sora, all42 of the HMI societies seem to be organized by patrilaterally assigned exogamous clans, as well as in some cases phratries and moieties, but this observation as such says little about the variety of relational principles. HMI clanship may be of the territorial kind or not. In other cases, the purely territorial organization seems to be different, and more relevant, compared to that of clans. Several great Indigenous units of Chota Nagpur offer more than 100 clans with numerous subclans, while the five million Santal manage well with only twelve such categories. The huge Gond/Halbi complex, on the other hand, functions with such patrilaterally assigned clans being bundled together in a handful of exogamous phratries, while many other communities acknowledge a ‘fraternal’ relationship between some clans in one region and between others in a different zone not so far away. In short, clanship as such, or ‘gentile organization’, is little more than a general formula for a variety of intricate arrangements of affinal exchanges. The operational principles for relations ‘on the ground’ are equally simple and effective. In a village, they imply that a single societal unit defined as consanguineal is permanently engaged in intermarriage with

‘Tribe’

Santal Bhumij Bhuiyan/Juang Juang Bhuiyan Kharia Dhelki Dudh Ho/ Munda Munda Ho Kisan/ Oraon Oraon Kisan Several endogamous Kond  units Several endogamous Sora  units Koraput ‘tribes’ One of the ‘twelve’ Several endogamous Gond  units

Bhumij/Santal

‘Tribal’ federation

yes

7, 6, 5, 4 or 3 no in a phratry

yes

no no

no no yes yes yes yes yes no ? ? yes

? ? ? ? yes yes yes yes yes yes no

yes

yes

yes

yes ? ? ? yes yes yes no ? ? yes

yes

yes

yes

yes yes yes ? yes yes yes yes ? ? yes

?

yes

?

yes ? ? ? ? ? yes yes ? ? yes

yes

yes

?

? ? ? ? yes yes yes no ? ? no

yes

yes

no

no yes no no/yes no/yes no/yes no/yes no/yes no/yes no/yes no/yes

Subclan Territorial Local status Local descent LDG Villages of XC units groups groups sub-groups first settlers marriage

no

8, 4 or 2 of each ‘tribe’

12 20 18 not known 8 9 many many many many finite numbers no

Number of clans

Table 8.2 Overview of HMI ‘gentile organization’.

160 • Lewis Henry Morgan’s Comparisons

a limited number of other units of the same kind. Some, however, may be conceived of as the seniors of others and thus unmarriageable, and some marriage ties may not be repeated because the married seniors of the same groups seem to imply such a distinction, so that rules demand different external groups for each alliance in the subsequent generation. Even if cross-cousin marriage is the preference, the demographic factor will rule out an uninterrupted union over time between any two groups and will instead force them to enter into regular marriage exchanges with several different units. The normative demand cannot be implemented in practice. The formal cannot be equated with the empirical. The power relations of HMI ‘gentile organization’ always involve two fields of central importance, i.e. marriage and ritual. War may have been another sector in precolonial days, but nowadays feuding may be observed in the remote highlands only. At present, both consanguineal and affinal relations always involve ritual obligations of the group members, just as any major ritual classifies participants according to their status as consanguines or affines and further differentiations along these lines. The economic domain offers the dominant link between cultivators and weavers when intermarriage is ruled out, whereas the division of labour is mandatory. However, the commercial agents do participate in the grand rituals, though they are bound to perform these in subsidiary or ‘junior’ relations only. Usually several local descent groups of a village (but sometimes only one) are defined by a specific status linked to ritual activity. Male members in council also act as the decision-making body. Belonging to an LDG, or a subunit of one, means that they are beneficiaries of incoming ritual gifts and marriage prestations, just as all members – with the help of their commercial agents – will jointly meet the same kind of external obligations. I have seen the men belonging to the households of an LDG jointly slash and burn and later hoe a hill slope, while the women of the group jointly cut stalks of rice to bundle them up into one sheaf after another, which they carry through the hills to the threshing ground. The local decent group negotiates marriage matters and responds to abductions. In a unified effort, its men may take up arms to threaten others or meet such threats when others encroach upon their lands or fail to meet their respective obligations. These men, as a group, also carry slabs of rock to ritual partner villages in order to erect menhirs there, just as in return, they will take away the malevolent spirits of the deceased embodied by buffaloes (Pfeffer 2001). Morgan’s ‘gentile organization’ contains many complexities, even if nobody appears as to be an inventor, director, organizer or court

Order in Anarchy • 161

of appeal. Always of central concern are exchanges with the Earth Goddess involving a number of different consanguineal and affinal LDGs. She has provided the miracle of plant growth and the human local descent groups are in turn obliged to take and reciprocate this gift in the form of sacrificial animals.

Notes   1. Though Systems has been quoted very frequently, I am not sure if many anthropologists (other than ‘kinship’ specialists) have actually gone through its technical passages.   2. Some reasons for this general ignorance are given in Pfeffer (2014).   3. Annually (e.g. Hirtzel 1922), a serious ‘Statement Exhibiting the Moral and Material Progress and Condition of India was published by His Majesty’s Stationery Office.   4. On the Santal rebellion, see Rycroft (2012); on the movement of Birsa Munda, see Singh (1966).   5. For the literature on these movements, see Jena (1999).   6. I suggest that the anarchic background of ‘tribal’ people prevented any one of them from becoming a significant political figure in the country, since the very few known ST political luminaries have very exceptional private histories. Whereas members of the SCs (i.e. former ‘untouchables’) with their ‘isomorphic structures’ are serious political competitors of middleand upper-caste politicians, a ‘tribal leader’ is a contradiction in terms.   7. Here and there, Elwin’s many books contain a short derogatory remark on the people he ranks as immigrant Hindus.   8. Weavers also perform as musicians (playing only some or all instruments) in highland society. Guzy (2013) has studied these performances. The other major books of the Berlin team include those by Otten (2006), Skoda (2005), Strümpell (2006), Berger (2015) and Hardenberg (2018). The dissertation on the Santal by Berger’s student Schulte-Droesch (2016) was also fostered in the ‘Berlin environment’, as were the books on the Ho by Reichel (2009, 2018).   9. These regions have only been exposed to colonial influence much later and more superficially than those in the north. For the same reasons, the exponential influx of lowlanders (and increasing communal riots) in recent decades seems to be confined to the southern regions. 10. A few individuals may have been successful in crossing this barrier during the last five decades. 11. Many of Niggemeyer’s photographs, found in the Museum of the Frankfurt Frobenius Institut, show Pano at work with their looms or male Kond wearing cotton loin cloths. 12. Very rarely, I have found a powerful cultivator individual who had fallen in love with a weaver woman and taken her as a second wife. In other rare cases, a love affair of a weaver man and a cultivator woman made the couple flee to urban India for good.

162 • Lewis Henry Morgan’s Comparisons

13. Inner-Indian administrative boundaries have never reflected the settlement pattern of the Indigenous people. 14. The order of the totems (Pfeffer 1984) is swan, antelope, kingfisher, betelpalm, weed, Pleiades, owl, rice, falcon, lizard, pigeon and Bedea (?). 15. Norwegian missionaries must have found them in the Santal Parganas of today’s northeast Jharkhand. 16. The Norwegian missionary Paul Olaf Bodding (1942) published the Santal myths collected by his countryman and fellow missionary L.O. Skrefsrud in 1887. 17. The twelve clans are mutually distinguished as relative seniors and juniors (Pfeffer 1984). 18. Twenty exogamous clans are assigned to them. 19. This is the standard phrase attached to all Indigenous people by politicians and administrators in India. 20. Their particular manner of treating the tiny Juang women has contributed to the spreading of venereal diseases and AIDs. 21. Since the mid nineteenth century, the colonial powers appointed all sorts of adventurers as ‘kings’ of the ‘states’ they had carved out for indirect rule. Though such pseudo-monarchs did not interfere much in internal affairs of the local people, they did collect ‘rent’ in the neighbourhood of their palaces to construct and operate their respective ‘opera courts’ until their political privileges were removed by an act of government in 1949. 22. The information was personally communicated to me by the ethnographer of the Kisan, Deepak Kumar Behera. 23. As a lawyer who had studied in the early years of the twentieth century, Roy paid considerable attention to public offices in all of his books and probably emphasized this field of investigation, obliging his consultants to respond accordingly. Like Elwin, I always found elders to be inconspicuous and far from keen to assume any executive functions or emphasize their status. 24. For example, in Roy and Roy (1937). 25. Such fraternal ties between different clans are known throughout the culture area. I have observed that they are not generalized, but tend to vary within a larger region. 26. Demmer (2009: 271), writing on the Koya, recalls: ‘the dead are regarded as important and active members of the society. In fact, they are really a part of everyday life. In every hut of the Koya a part of the storage room (wijalon) is kept apart where the ’pot of the departed’ (hanal kunda) … is set up’. 27. My own research among some Kuttia Kond points in this direction. Every household in this group of people keeps a basketful of ten ball-shaped stones, each of which is associated by name with a deceased male forebear in each of the last ten generations. With the death of a senior man, his sons add his name to the ten and remove that of the most senior ancestor. Before Reichel’s work, I had failed to understand the social significance of these acts, since otherwise pedigrees are quite insignificant.

Order in Anarchy • 163

28. I was informed of the following senior/junior pairs: saiti/paba; pira/pagu, jura/ bongoti, sano/dauni; gutu/maetu, maetu/paetu; rabi/dauni, gani/kongeri, sura/ virsa. These names do not seem to carry a totemic or any other meaning. 29. They had been applied for human sacrifice during the nineteenth century. The British terminated such practices after an extended and bloody campaign. 30. Literally, the name means ‘without sweat’, implying a life without physical labour. 31. I have seen members of several low-status and landless Sora units working in the fields or as Kulis of Sora cultivators. Such jobs are unthinkable for members of any Indigenous cultivator category. 32. Unfortunately, he was writing about a ‘segmentary lineage system’. 33. This simply seems to mean ‘line of the seniors’. Elsewhere, such a nonexogamous LDC is known as birinda (Behura 1997: 323). 34. I experienced rather intensive mutual prejudices between caste Hindus on the Odisha/Andhra Pradesh border. 35. Again, the official classification of the state subdivides these Indigenous people into STs, SCs and OBCs according to very questionable criteria. 36. After long-term ethnographic fieldwork, Tina Otten (2006) has written the first and only monograph on the Rona with special reference to healing rituals. To my knowledge, the other three communities (each one having more than 100,000 members) have never been described in ethnographic literature. 37. In the very large Bhumia community, members of the majority cobra clan say they can marry a junior (or senior in the opposite case) cobra person if no leopard person can be found. 38. A person is named by such an agnatically transferred title plus a reference to the weekday of his or her birth. Accordingly, the number of names is altogether very limited. 39. Baiga norms rule out ploughing the earth, which is understood to be their Goddess, so the landowners lived on shifting (hoe) cultivation. When the British rather forcefully introduced the use of the plough in the hills in order to increase their revenue, many Baiga refused to comply and became impoverished. 40. Clearly, these musicians and genealogists (and weavers?) of a lower status are not immigrant lowlanders but have ‘always’ been associated with Gond cultivators. 41. For a fuller account, see Gregory (2009). 42. Probably the Maler and the Mal Paharia in the Raj Mahal Hills of Jharkhand also lack clans, but the ethnographic accounts are ambiguous on this point.

9 Bridewealth and Gender in Highland Middle India

The Issue When Morgan attempted to describe the ‘Property Career of the Human Race’ (1877: 457), he elaborated on several confusing details and suggestions that need not be discussed here. However, he rightly emphasized a major turning point in human history and in the development of any specific society: the introduction of ‘two forms of ownership, namely, by the state and by individuals’ (1877: 458). In modern anthropology, this relevance of private control over a piece of land to some or to a major extent and the interference of the state or other agents in property matters had been intensely discussed in the course of ambitious ‘peasant studies’ during the 1970s and 1980s, but later the issue seems to have been removed from regular teaching, as if all societies of the contemporary world had agreed on the Western form of individual ownership. This is certainly not the case in either peasant or anarchic India and many other regions of the world. The following will discuss HMI marriage payments since they reveal basically different sociocultural attitudes towards private property. Such transactions are important criteria indicating the distinct values and power relations in anarchic and peasant societies. Yet other concerns apply to the material transfers at marriages in pastoral communities, i.e. those accumulating ‘mobile wealth’, but I will leave aside the logic of the latter systems and will try to show how HMI bridewealth payments ‘merely’ concern affinity. In other words, they are tokens of ‘pure’ affinity, the outstanding value-idea containing ‘the spirit of the gift’, whereas peasant dowries determine the personal access to some social power via the control of some quantity of land and other property. I hope to illustrate and emphasize this point in a narrative form for readers unfamiliar with the highlands.

Bridewealth and Gender in Highland Middle India • 165

Settlement Duty Early in January 1981, my first excursion to the Gadaba people in southern Odisha ended in the small village of Lamptaput, which, during the subsequent thirty years, was to become a bustling provincial town home to some 30,000 immigrant lowlanders. The bus had dropped me a kilometre away from the few mud houses on a spot where the only other small road1 met the one leading to the boundary of the Koraput plateau to eventually descend to the coastal strip on the Bay of Bengal. At that time, the vehicle stopped near a thatched tea stall, a permanently empty Inspection Bungalow of the Public Works Department and two or three very recently built modest administrative quarters for the staff, the office and the family of Mr Hansda,2 Block Development Officer and thereby the representative of the authority of the state. To my good fortune, this Santal from the faraway north had become such a functionary on a Scheduled Tribes (STs) reservation ticket and also, during his first year of service, had ‘abducted’ his Gadaba sweetheart from a nearby village. The happily married couple were eager to show me their home and introduce me to Hansda’s many new affines in the surrounding highland, since he had an open mind and was equally sociable and determined. A year later, not just the Block Development Officer but also the entire outpost had changed. The bus stop was surrounded by four or five huge tents housing numerous teachers who were ‘on deputation’ from coastal secondary schools of the government to execute their ‘settlement duty’. For this purpose, they were asked to go to the villages of Lamptaput Block and register the respective land of the Indigenous people inhabiting the administrative unit. They would also provide a pota to the person they – according to whatever indicators – considered to be the individual owner of a field. The man3 could keep this certificate of the state as a formal proof of his private property, even if he could not read it. Such ‘settlement operations’ were taking place throughout the district. These extensive administrative efforts reminded me of the many thick volumes labelled Census of India of 19614 that I had found in the library of the Department of Anthropology at the university in the provincial capital. Not just the coverage of this southern ‘tribal’ district, but all maps of highland Odisha5 had simply contained place names written somewhere6 in the respective district without the village boundaries that were drawn for all places in the lowlands. Similarly, the maps were lacking the normal administrative data on villages. During

166 • Lewis Henry Morgan’s Comparisons

the second half of the nineteenth century, colonial census officers had standardized such information for state authorities, but in what was to become Odisha as late as 1936, nobody acted in this way in the highland areas under ‘indirect rule’. Before 1982, the modern administration in Koraput had simply left such territories aside. The fields had not been measured, ‘settled’ or taxed by the state. Such a ‘culture lag’ also implied that land ownership and inheritance had been of local concern and not the responsibility of the state administration. Similarly, marriage payments in the ‘settled’ areas on the coast were and are of a rather different kind compared to those among Indigenous highlanders, and these two issues are indicators of the same distinction, though the people involved are unlikely to have considered the matter at all. In the following, I hope to point to such a link between private property and marriage payments.

The Literature Marriage in India has received elaborate coverage7 in the discipline, including discussions on ancient or modern Hindu8 dowry obligations. In an even more general and comparative analysis, Jack Goody and Stanley Tambiah have meticulously examined such transfers in major contributions of 1973 that were later revised by Tambiah in 1989, after additional and more refined ethnographic data had become available. It goes without saying that none of these discussions has ever mentioned the bridewealth of the highlanders as the essential and invariable form of marriage payments in Indigenous Middle India. The omission is rather significant, since the two authors had explicitly devoted their book to a formal comparison of bridewealth and dowry. The former was a circulating fund among African hoe cultivators without permanent and fixed ownership of real estate, while the latter was a conjugal fund among Indian peasants and was very much a part of the intergenerational transfer of land or other professional capital. To my knowledge, HMI bridewealth had been referred to only superficially by nonprofessional writers of the colonial era. Hardenberg (2018: 283‒344) was the first anthropologist to examine all the relevant factors relating to this issue in great detail during extended fieldwork among the Dongria Kond of southern Odisha, about 70 km to the north of the Gadaba region. As to general debates on a worldwide basis, several contemporary scholars continue to study bridewealth obligations found on different continents. Their focus has shifted from its organizational relevance in social systems to the concerns of acting individuals

Bridewealth and Gender in Highland Middle India • 167

and the relative welfare of the latter. In applying a ‘narrow lens’ to gain such images, formal sociocentric relationships and their respective empirical realization are omitted by such research on the concerns of a bride. This shift in focus has significantly reduced professional competence. Though I see no relevant reason not to combine studies of gender with those of institutionalized social structure, the interest in bridewealth as an important construct of societal cohesion seems to have been dropped altogether. Details concerning marriage payments and the valuables involved used to be examined when scholars considered the issue as part of a general key to understanding the structure and evolution of certain societies. Regarding peasant India, Goody and Tambiah refer to parental property devolving vertically on both male and female offspring, with the dowry being a form of premortem inheritance of mobile wealth given to a woman. Upon her marriage, she will accept it to form her and her husband’s conjugal fund, whereas only her brothers will be heirs of agricultural land or other forms of the peasants’ infrastructure. Bridewealth in Sub-Saharan Africa, on the other hand, passes from the males of the bridegroom’s side to the bride’s male kin as a circulating fund, enabling the latter to obtain their own wives: Bridewealth passes from the kin of the groom to those of the bride; it forms a societal fund, a circulating pool of resources, the movement of which corresponds to the movement of rights over spouses, usually women. (Goody 1973: 17)

This observation offers the important initial insight that HMI bridewealth is not the reverse of dowry payments, but rather a kind of formal permit passing on from one male-dominated local descent group to the other as a regulation to obtain a wife for a member under socially agreed preconditions, whereas even after Tambiah’s subsequent modifications (1989), at least a significant part of a dowry is understood to become the private property of a married couple. Though the brothers of a peasant bride inherit all immobile family property, they and her father fulfil their sacred obligation to provide her with the adequate material basis in her postmarital existence by sending her out with a dowry. Among Hindu lowlanders, the husband and wife obtain a degree of decisionmaking power to invest this money and use the mobile material goods for their future welfare, but in HMI societies, the married couple do not receive any such benefits from bridewealth payments. Such valuables are jointly consumed or else they circulate between local descent groups as decided by their adult male members. In fact, bridewealth cannot be linked to any kind of premortem ­inheritance because land is

168 • Lewis Henry Morgan’s Comparisons

owned collectively and mobile goods that are to be inherited are hardly worth mentioning. This difference should be seen in the light of one of Morgan’s general arguments. Whereas his societas (i.e. anarchic society) is founded upon persons and purely personal relations, his civitas (i.e. state society), as a later development, is based upon territory, property and administration (1877: 13‒14). Morgan did conceive of such a development as progress, though certainly not as the ultimate goal of humanity, since he also wrote: ‘A mere property career is not the final destiny of mankind’ (1877: 467). At the start of the discussion, then, it should be clear that Indigenous bridewealth has nothing in common with ‘buying’ a wife or ways of obtaining private ownership of material goods, or investing valuables for future gains. None of the highland cultivators9 I met in Middle India displayed the standard peasant ambition of individually and permanently accumulating land and goods, though I did experience several cases of Indigenous collectives quarrelling or even fighting physically over the control of land. As to mobile property, I have experienced Gadaba who, over many years and through contributions within the local descent group, accumulated huge funds to buy many buffaloes along with plenty of rice, liquor and smaller meat-supplying animals in order to give all of this away within three or four days of a grand and complex secondary funeral (Pfeffer 2001). While peasants become prominent figures as a result of their permanent control of wealth, Indigenous local descent groups obtain prestige by giving it away. The following text will discuss the transfer of marriage goods from the bridegroom’s people to those of the bride and will thus exclude any consideration of material accumulation, even if it is obviously the first thought crossing a Western mind. Again in narrative form, the discussion will attempt to illustrate the Indigenous way to maintain respectability.

Conspicuous Kond Concerns Under the directions of a middle-aged man in the village of Burlabaru (Phulbani district), some thirty of his young Kuttia Kond neighbours are hiking through the steep hills to another village at a distance of about 20 km. All are in high spirits and continue to joke with each other and me. On such occasions, the male youths have made it a habit to cover their upper bodies with prefabricated shirts. As taught by the Gandhian village worker, the unmarried young women have put on

Bridewealth and Gender in Highland Middle India • 169

their saris rather than the short skirts and aprons they normally wear. The bachelors take turns in carrying the three or four large canisters filled with hard liquor and also several sacks full of rice. Even when crossing the ridge of a mountain, they laughingly pull along a water buffalo on a rope. The most expensive animal of the region is an obvious materialization of wealth. It has never been made to work in its owners’ village. These cultivators’ commission agents of the weaver category have bought and supplied the buffalo, and they have also distilled and supplied the liquor to increase the Kuttia debts, but they do not accompany the affinal party. In a general mood of relaxation and mutual frolicking, the village of the future bride is reached before sundown. On the next day, the animal is slaughtered and the body – but not the head – is cut into almost equal very small portions of meat that used to be part of the neck or rumen, belly or liver. After being cooked separately, these bits and pieces are thrown together again into the many leaf-cups of equal size that are assembled on the ground containing meticulously cut portions of equal size. They are served with some added broth. Before the eating is about to commence, the middle-aged leader of the visitor’s party and his local counterpart lift their metal cups filled with a kind of brandy and take turns to pour some of it into the other’s liquid. Then they simultaneously offer the first few drops of their cups to the Earth Goddess before gulping all of it down in a single action. Thereafter, for everyone partaking in the meal, one cup after another will be refilled immediately. While consuming meat, rice and liquor, hosts and guests sit on the ground crossing their legs and facing each other in two rows of males and, separately, two lines of women, while their spirits continue to rise. Another two days of almost uninterrupted dining, drinking and dancing will follow. Both male and female adults prefer the drinking part, while the unmarried youths primarily engage in dancing for fun and also, separately, as a sacred duty. As a point of difference compared to the Indigenous youths of Chota Nagpur or Koraput, the Kuttia Kond performance of rhythm and joy is not exceptionally artistic or imaginative, though they too, when dancing for fun, move shoulder to shoulder in a circular fashion like a kind of merry-go-round of humans only. The ritual dance is a somewhat solemn affair in which the young villagers and their visiting affines participate. In the wide corridor between the two rows of houses, where small rocks as the seat of the Earth Goddess form the centre, older girls and boys face each other while somewhat monotonously marching forward and backward in an east-west direction on this plaza. Here the first rays of the sun above the

170 • Lewis Henry Morgan’s Comparisons

Figure 9.1 Kuttia Kond sacred youth performance when meeting affines. Figure created by the author.

mountains will touch the earth and once in the course of several years will initiate the grand sacrificial killing of a buffalo.10 The ritual dance of the youngsters, or rather their march back and forth, continues for hours. Two boys beat huge kettle drums that must only be taken out when the presence of the Goddess is called for, and the other males continue to hit two bamboo sticks against each other (a long one and a short one), while the girls vary the sequence of their steps by following the beat of their counterparts. Such movement along the corridor of some thirty metres in length between the homes of the villagers (see Figure 9.1) should be understood as a fertility rite, as is also verbalized in the sacred texts sung continuously by both the male and female youths. At the end of these festivities, words and gestures of farewell by the two senior men in the middle of a crowd convey the essential messages. Again and again, the visiting father of the groom announces that he will continue to bring liquor and buffaloes, and at their final meeting a considerable sum in cash, while almost simultaneously his host, the bride’s father, will not stop shouting. Hardly taking time to draw breath, he continues to confirm the promise of giving away his daughter. Nobody else but the donor of the buffalo is to receive her. Having always honoured all previous commitments, so he says, the groom’s people will be allowed to take her with them in due course. The right-hand palms of the two men repeatedly cover each other before the visitors’ party hike back with the buffalo’s head tied beneath a long bamboo pole on the shoulders of two youngsters. In their home village, they will cook this symbol of the agreement for the Goddess in front of her altar and will subsequently distribute small pieces of its meat among all members of

Bridewealth and Gender in Highland Middle India • 171

the local descent group who had earlier contributed to the purchase of the animal and the provisions of rice and liquor, and who will contribute to the cash handed over at the final meeting with the bride’s local descent group in the affinal village.11

Other Conspicuous Kond Concerns On approaching the Kuttia Kond village Pikussi12 at about 11 am, I notice that the men are not out in the fields, as they would normally be. Instead, they linger around, having earlier taken a bath and put on clean sheets of cloth around their hips. After a while, the inhabitants of another fairly distant Kuttia Kond village enter Pikussi without a word. On walking through the small entrance between the palisades of the boundary, they form a long line starting with grim-faced women and then, marching behind them one after another, the adult men. The latter display the same mood after having walked on narrow paths through the forest. Instead of a salute, the women unmistakably articulate a single demand: ‘We want our daughter back!’ The response is more than polite. The incoming Kuttia are requested to settle down and accept hospitality, which they bluntly refuse. Repeating their order in no uncertain terms, the female visitors, and behind them their husbands, come to squat facing the male villagers on one side of the central corridor between the two rows of thatched mudhouses. More than anyone else, the female intruders are infuriated. They scream and shout. At times, one of them rushes up, crossing over to the other side to beat her lower arms on to the backs of one or two sheepish-looking male inhabitants of Pikussi. Several metal bracelets on each of the women’s wrists make sure that the anger is not understood as being of the symbolic kind only. Such loud exchanges continue for many hours, with the women in the front row directly facing the local men, who try to appease them with words, gestures and smiles signifying humility. While this heated debate is continuing, I find a grim-faced female youth of about 17 years squatting behind one row of houses, or out of sight for those quarrelling on the plaza. The backyard is the zone reserved for females of a village. The young person is being soothed and taken care of by the adult local women who, some years earlier, had married into the place. One or two of the Pikussi married women originally hail from the same settlement as their mute guest. All of them do not tire in their constantly unsuccessful attempts to make her eat and drink a little. She may, during the course of these hours, have to go

172 • Lewis Henry Morgan’s Comparisons

to the toilet in the nearby forest, but will return to remain stone-faced and stiff as a statue. The young men of the village, now nowhere to be found, had abducted her at night and for the time being, she is in no mood to speak to anyone. Significantly, she is wearing an elegant sari and fine ornaments rather than just the piece of old cloth that would usually hang round her waist day and night. Her hair has been oiled and arranged meticulously, and she must have taken a bath before the male youngsters grabbed her and carried her to their village. Be that as it may, after five or six hours of debate on the plaza ‒ in ordinary circumstances the male public space of the village and now filled with aggressive external women insulting defensive local men ‒ occasionally slight signs of fatigue become noticeable. In the end, an agreement is concluded and soon afterwards positive verbal responses are uttered on both sides, all the female and male adults begin to socialize in a relaxed mood, which improves every minute as considerable quantities of hard liquor are served with food that appears out of nowhere. All persons assembled know one another from many earlier occasions. During the spring months, they have met for the usual local rituals. The two villages – connected by inherited affinity – are likely to have earlier given brides to each other. If a local descent group is provided with one, it will be obliged to reciprocate only after one or more generations and cannot take another bride before that return gift has been completed for some time. Such exchanges with several descent groups of several villages imply a circle of major ritual encounters13 that include the dancing of the youngsters as well as feasting and drinking liquor excessively.

Arrangements The two scenarios described above may offer a glance at two rather normal procedures to initiate a marriage among Kond or other Indigenous people of the culture area. The first encounter had witnessed one of many visits that substantiate the claim of a particular groom’s local descent group to formally take home a bride. The agreement is initiated and conducted by his father (or senior relative) as an endeavour of the most prestigious kind: a marriage contract envisaged soon after the birth of the potential bride and implying that year after year the flow of gifts to her village will ensure that she – some years after she reaches physical maturity – will be handed over to the groom’s people in an elaborate ceremony.

Bridewealth and Gender in Highland Middle India • 173

The second confrontation, as described above with the usual happy ending, apparently covered a case of abduction, even though, upon a closer look, some contradictory impressions seem to demand attention. Evidently the young woman was prepared for the experience of being carried away by the groom’s friends. In a similar case, I personally observed – before sunrise – the physical engagement of four or five young men with the bride screaming and shouting in resistance, while her captors triumphantly carried her into their village. However, I later learned that she had started making the noise only on approaching the abductor’s settlement and that in the months before this, she had been the groom’s sweetheart. In my own country, the public media have at length discussed the campaign of recent years regarding the motto ‘NO means NO!’ It is agreed to be extremely relevant for correct conduct in civil society. Some years ago, the German Penal Code was adjusted in the hope that such a change would result in more men respecting a woman’s decision to refuse intimate verbal or physical encounters. This conditioning, and even normal rules of behaviour that had been valid for decades, made it difficult for me to accept the idea that in Kond reality, an apparent abduction was a way to stylize an elopement of mutually consenting lovers. It was bound to include an impressive show of physical force and resistance, even though all women and men, be they old or young, were perfectly aware of the actual nature of the relationship. The bride, having earlier been initiated into the plans, was happy about the abduction and may have felt a sense of pride, especially if she herself had instigated it. Since such observations are not uncommon in spite of very different formal articulations, a more general picture of the ethnographic preconditions will be necessary to add weight to the claim that the enacted drama of a bitter conflict implied the public secret of a general consensus. The first scene of a marriage being arranged by the elders of the couple carries considerable prestige, because it is quite possible that it will end as an unsuccessful project. Impediments may appear very soon and may continue to arise over the course of time. Thus, either the bride or the groom – or the men who had concluded the deal – may die long before the year in which the marriage had been envisaged, since this is a society with a very high mortality rate, especially when it comes to children and infants. Also, harvests may fail, resulting in a reduction in the flow of gifts and the benevolence of the bride’s father. Moreover, over the course of many years, better opportunities may arise for him and his local descent group. The last but certainly not the least risk is that the bride herself will select a young man as her groom,

174 • Lewis Henry Morgan’s Comparisons

will arrange to be abducted by his friends and will do what must be done to achieve the desired result. Hardenberg’s long-term research in another Kond unit called the Dongria (2018: 407) discovered that prestigious arranged marriages connect influential persons of villages situated a considerable distance away from each other, while elopements unite couples of relatively nearer places, though still at some distance from each other. These important observations only make sense in the general context of these highland societies that Hardenberg elaborated on a statistical basis and by developing uniquely confidential ties with the young men who were regular visitors of youth dormitories. Such opportunities to observe ongoing practices as norms, or sanctions against deviant behaviour, had not been available to Morgan facing the sad remains of Iroquois society. Just as many of his contemporaries and successors, he had fantasized about ‘wives … acquired by purchase’ (1877: 367, 389), whereas in reality, rather different concepts of property apply in a classless society. Since such stereotypes have tended to reappear in very recent publications, empirical studies in a contemporary anarchic milieu seem to be more than justified.

Youth Dormitories In the past, everywhere in Highland Middle India, a house reserved for unmarried youths was a regular institution,14 but nowadays it is denounced as immoral by prudish state officials from the lowlands. As a result, the dormitory system has been abandoned, unless it is continued in an informal or hidden manner. In 1980, I stayed as a guest in the – then regular – boys’ dormitory of a Kuttia Kond village, and on a later excursion in 1982 and 1983, I also took note of the several low-roofed houses for the teenage girls in every Dongria Kond village. To this day, these buildings continue to function as described by Hardenberg (2018: 353‒82), i.e. as sites to be visited at night by unmarried boys coming from villages that are classified as affinally related. The point is that the exogamous Kond clans unite inhabitants of specific territories, though each of these need not be contiguous and may consist of separate lands. Amorous nocturnal visitors, be they female or male, must never belong to the clan of their hosts. When the Kuttia have completed their harvest and the spring season begins, teenagers in groups of either gender, with girls probably accompanied by a widow, will frequently move to affinal villages beyond the boundaries of their own clan territory to spend a night or two involving public dancing and

Bridewealth and Gender in Highland Middle India • 175

private love-making. They will then rest and sleep as guests in the dormitory15 of the local boys or girls. Thus, a person’s neighbouring villages of the same clan are out of bounds, whereas the affinal villages somewhat further away attract the youngsters for nocturnal visits and the more remote villages are, due to the distance, hardly ever approached and are thus less familiar. These distant settlements will be housing potential brides for the negotiated marriage as described in the first account. Given these formal preconditions, the slack season of the year will witness Kuttia groups moving to villages at a middle-range. The youths will dance on a threshing floor, while separate couples may occasionally disappear into the forest close by. However, among the Dongria, only the boys will hike to their affinal villages in the dark. Any two or three of them, or a lone lover, may enter one of several very low-roofed female dormitories of a village in which visitors can only move on their knees or in a horizontal position. Between these four walls, the girls are definitely in charge of affairs, whispering orders about what is to happen and to what extent this may go. Male disobedience will lead to noisy scenes and will thus alarm the married men and women sleeping in the bigger mud-houses only a few metres away. A male visitor or two of them may stay for a while before moving to another female dormitory either in the same or another village. It is clear that such adventurous excursions are expected by everybody and are certainly known to the elders or even encouraged by them. Casual young male visitors like to give cheap metal rings to their teenage female hosts and Dongria girls are proud to show these gifts in their hair in the daytime, since their prestige will grow with the number of such rings on display. Premarital sex involving consenting teenagers is a common feature in the culture area, while everybody agrees that adults should be excluded from these activities. As a result, the reasons behind abductions, as witnessed in caste India or in some West and Central Asian societies ‒ namely sexual motives or an aggressive male urge to fulfil a desire in adverse circumstances and in spite of ­resistance ‒ is out of the question in the highland societies of Middle India, since sex is not prohibited and is not a controversial issue among either seniors or unmarried youths. In a single night, one of the latter group may or may not choose to have intimate contact with several others. The cause of the kind of drama that I tried to sketch in the ‘abduction’ case is love rather than sex. Though Kond youths may be involved in numerous short-term adventures, it is also frequently the case that two individuals fall in love with each other and agree to spend their lives as a married couple rather than enjoy only fleeting moments together in the jungle or a low-roofed

176 • Lewis Henry Morgan’s Comparisons

hut. Formally, however, they will be unable to marry because the girl may have been promised to somebody else by her elders and, even if not, for these elders, it would be dishonourable and foolish to let their girl settle in another village without the lawful formalities that must include bridewealth transactions. As a result, the two lovers conspire to have the groom’s male friends abduct the bride, which also means that she must make a great show out of the nocturnal event. The drama will arouse everybody from their sleep and will subsequently lead to an encounter of the two village communities with the kind of open agreement that was the outcome described in the second scene detailed above. Briefly, the intention of the two youths is to transform an intimate affair that had been carried on quietly into a noisy public concern so that a legitimate marital union will be possible. At the end of the direct confrontation between the two villages, the consensus will imply that the bridewealth is higher than it would have been in a negotiated marriage. Observers are thus made to reconsider the rationality of events by highlighting the persons involved and the possible consequences of an abduction.

Types and Technicalities The standard work of Jane Fishburne Collier (1988) entitled Marriage and Inequality in Classless Societies is so well known that one may ask under which one of her three ideal types of kin-based nonstratified societies the Middle Indian examples should be classified. The author’s ‘Bride-Service Model’ must be ruled out, since only in rare cases may a highlander, who would be of a destitute family or happen to live without kinspeople and their support, decide to work for his parents-in-law. During the rest of his humiliating existence, he would have to stay in their village. I have only heard of such cases and have never observed one. Nor is the ‘Unequal Bridewealth Model’ fitting, because the kind of ranked societies described by Collier do not exist in Middle India. Her model involves: People in high-ranking families [who] are free to pursue culturally valued activities because they have slaves, outcasts, or debt bondsmen to perform necessary, but unprestigious tasks for them.

And also: People … tend to believe that no two people can be truly equal. Within kinship groups, full brothers are unequal, and between kinship groups, wife-givers outrank wife-takers. (Collier 1988: 143)

Bridewealth and Gender in Highland Middle India • 177

In fact, all Kond people work in the fields as rather poor cultivators and though special inequalities will have to be discussed, wife-givers never outrank wife-takers anywhere in Indigenous Middle India, since both, in the course of several symmetric exchanges, frequently take up the roles of the respective others, even if in a given situation, one side may have the upper hand. Generally an egalitarian spirit is noticeable throughout the culture area, in stark and lasting contrast to omnipresent hierarchical distinctions combined with power differences in caste India. Thus, only Collier’s ‘Equal Bridewealth Model’ remains, which, following Goody (1973), she calls a ‘circulating societal fund’, but this type also does not really seem to be applicable for the Kond and other local people of the region. Her chief concern, when distinguishing these ideal types, is the source of bridewealth, or the issue of who is gathering and transferring the valuables. Are they accumulated by: • the groom himself?; • slaves and debt bondsmen?; • the groom’s elder kinspeople who provide them for the relatives of the bride? In Indigenous Middle India, all of these potential sources of bridewealth must be considered and yet none of them serves as a satisfactory answer. The ethnographic data do not contradict Collier, but rather add another kind of constellation to those mentioned in her book. In all Kond villages about two thirds of the population belongs to the category of Kond cultivators and about one third consists of weavers called Pan or Dom, the commercial agents of the Kond. The cultivating landowners themselves collect the material to build their own houses and make their own ploughs or construct their own equipment for fishing and hunting, but crafts are left to the low-status neighbours who organize all market operations. Money is their medium. The Kuttia or Dongria Kond continue to live without much cash, even though they may take a little to the weekly market to buy a drink or the snuff they have come to like, and perhaps another cheap and fancy object. Their daily meals consist of millet porridge and vegetables grown in their own fields and gardens. The Pan family attached to their respective household supplies whatever else is required. The Kond have a reputation for violence and regard themselves as the masters of the country and also as protectors of their weavers. At the same time, the arrival of the lowlanders in the last three or four decades, along with the gradual introduction of police, courts, capitalism and corruption,

178 • Lewis Henry Morgan’s Comparisons

is likely to eventually change these old ties between cultivators and inherited middlemen. In relation to bridewealth, the family of a young man who has eloped with his bride will certainly require the help of the commercial agents of his village. After the pseudo-abduction described above and during the five or six hours of quarrelling on the village plaza, the Kond will deliver and receive the shouting and the beating, while simultaneously the weavers are intensely involved in the haggling on behalf of their respective cultivators. The Kond tempers will tire once the Pan have effectively concluded the deal according to which the groom’s weavers will advance the bridewealth to the agent of the bride’s family, with both acting on behalf of the two local descent groups involved in the marriage. Several aspects of this arrangement require further discussion. Initially it must be mentioned that the clan names of the commercial agents are the same16 as those of their patrons, which means that Pan of the groom’s village will be related to Pan of the bride’s village in some affinal way or another,17 and this may cause a conflux of a variety of interests among the commercial agents. The second aspect is related to the first. The weaver, when acting on behalf of the groom’s relatives, will provide the bride’s elders with bridewealth in cash, but these bride-people are likely to be indebted to their own village agent, so the money will actually be used to decrease this debt, meaning that it will go to him directly before it reaches the hands of the bride’s parents and their local descent group. However, this Pan receiver may have had earlier dealings with the groom’s agent, which may mean that the two simply clear an earlier debt and no cash actually flows down the chain, though it may also mean that the receiver will increase the earlier debt of the groom’s Pan, who, however, uses this opportunity to obtain a commission from his patron, the groom’s father, in order to improve his financial situation. The same is the case with regard to the bride’s parents and their agent. Another relatively common factor involves a chain reaction of elopements, since in a good number of cases, Kond women are not averse to choosing a new husband if they are not pregnant or do not have children. Similarly, young men may, after having concluded a first marriage, try to marry a second teenager. This may mean that husband A, or his father, will be indebted to his agent after his first wedding and will incur additional debts for his second union. It may also mean that his wife elopes with a man called B and, as a consequence, this new husband B (in technical terms his or his father’s agent) has to settle the debt of the earlier husband A. But before his most recent elopement

Bridewealth and Gender in Highland Middle India • 179

with a married woman, B himself may have lost his wife to a rival named C, which means that he will ask C to take over B’s new debts in relation to A as compensation. This, of course, is just an example. None of the Kond men will ever actually handle much cash. While A, B and C have to sort out their mutual relationships and those tying them to their respective commercial agents, these Pan, too, may quite separately have their own affinal and also business relations that include bridewealth obligations inducing them to be more or less active in the matter of advancing money to a certain Kond cultivator. In short, life is never dull in the Kond Hills. In their financial undertakings, the agents can be very clever and can gain considerable influence, while at the same time, the cultivators can lose track of the various links and amounts of indebtedness. Since only they can own land, they are not excessively worried about the debts. They let their weavers take an increased share of the harvest or one of several pieces of cattle and at the same time will be confident enough to rebuke any seemingly fraudulent or improper demands. Cultivators and weavers are likely to be involved in frequent quarrels that include mutual claims as well as the application of physical force, but they depend upon each other and are unlikely to terminate existing relationships.

The ‘Atom of Kinship’ In marriage matters, Kond women and teenage girls seem to be as active and outspoken as Kond males, though they are rarely concerned with the issue of who is organizing the bridewealth and how the transactions take place. In everyday life, a bride is certainly not mute, even if she fails to talk for hours after having been carried to her lover’s village. In the hills, girls and boys grow up with fairly equal rights and obligations, and also with the same concepts of a good life. Before she has had her first child, a young woman, besides being a most industrious and responsible worker in the fields and in her parental home, takes an active interest in social and erotic contacts with males of her age group among the inherited affines of her local descent group. Eventually she is also in search of a considerate partner as a spouse. If after marriage her young husband turns out to be disappointing and if she has no children, she will not hesitate to select another man with whom to elope. I should also mention that I have regularly observed that the romantic attachment between wife and husband continues through decades of marriage until old age and also in sickness and in health, or during times of material need or plenty.

180 • Lewis Henry Morgan’s Comparisons

The following observations have been made during my extended stays in Kuttia Kond and Gadaba villages and may apply to lesser extent or not at all to Indigenous people who, over a much longer time period, have been in close contact with governmental officers and immigrants from the lowlands.18 The fact that, at least in southern Odisha, male and female children over the age of seven or eight will voluntarily sleep on any veranda of the village, in the youth dormitory or in the home of a widow rather than in their parental house may have some effect on the relationship between spouses. During their childbearing years, women seem to be less active in public life, though they move at ease, on their own or in groups, through the forest and to other villages, contributing significantly to the sustenance of the family by collecting wild root fruits, tubers and other edible plants from the jungle. Among the elderly, men seem to lose relative status, while women maintain or even raise their ‘active voice’. In defence of the ‘wide lens’ applied by past anthropologists such as Goody and Tambiah, I suggest that this approach has helped to uncover social facts that were inaccessible to researchers who preferred the ‘narrow lens’ of their own individualistic priorities. For gender relations, an improved insight is likely to be gained when Lévi-Strauss’ thoughts on the ‘atom of kinship’ (1973) are applied to the standard attitudes of highlanders and compared to those belonging to the Indian caste society, be they Hindu, Muslim or Christian. It is well known that ordinary rich or poor people of the lowlands tend to maintain a very close and emotional relationship (or +) between brother and sister, while the husband‒wife relationship should always indicate a certain aloofness and distance (or ‒) in public, irrespective of the ‘real’ feelings of the couple. For example, acts of verbal or physical tenderness between spouses should be avoided altogether in the presence of third parties, whereas a woman may very well elaborate at length and in public through intensely emotional expressions how much she misses the understanding attitude and support of her brother. Similarly, again irrespective of his inner commitments, a man should maintain a clear distance in his relationship to his son or to his father by emphasizing the unquestionable prerogative of the senior person and the latter’s legal authority. After infanthood, a son should remain aloof (or ‒) from his father, and vice versa, rather than express love and an intimate understanding of the other’s personal idiosyncrasies. However, a maternal uncle should always and irrespective of other demands support, encourage and openly love his nephews and nieces (or +) in all situations, as indicated in Figure 9.2.

Bridewealth and Gender in Highland Middle India • 181

Figure 9.2 The ‘atom of kinship’ in caste India. Figure created by the author.

Figure 9.3  The ‘atom of kinship’ as experienced in HMI. Figure created by the author.

The stereotypical attitudes I have observed among the Gadaba and Kuttia Kond are rather different and indicate a structure that, to my knowledge, has not or has only rarely been discussed either by LéviStrauss himself or other authors. A woman should openly display a very close and emotional behaviour towards her brother and her husband (or +), and this attitude should always be equally reciprocated in public. However, she, her brother and her husband should display a definite reserve or distance, frequently amounting to open unfriendliness (or ‒), towards her seniors, whether they be parents or maternal uncles. Each age group should remain aloof from the other and should openly entertain close emotions with villagers of the same generation (see Figure 9.3). During the Berlin team’s19 research work in Koraput, more than one case of parricide came to be known to us and to villagers, though not to the state authorities.

182 • Lewis Henry Morgan’s Comparisons

This point could be exemplified by innumerable ethnographic examples and is also related to the issue of dormitories and bridewealth. Children are encouraged to be independent and to resist the commands or instructions of adults. Frequently youth groups cultivate and harvest fields of their own. In all kinds of situations, a youth responding with a ‘no’ to his or her senior is not an unusual reaction. I have always been rather astonished by this stark contrast compared to the all-out authoritarian social environment in lowland India. Under the abovedescribed conditions, bridewealth is sure to be a circulating fund and is certainly not a device to traffic in women or to discipline a female in one way or another. In caste India, it is quite common among the better-off, and also the poor, to let the entire unit of coresidents (i.e. the young, the old and the middle-aged females and males of a family) share a single room at night all year round. However, such an arrangement is quite exceptional among the Kuttia Kond and Gadaba, and is likely to be seen only when external guests have come to participate in major rituals. It implies an inner-family discipline as a matter of nightly routine, whereas independence is the norm in the highlands. Even if the dormitory is no longer visible in most contemporary villages, boys are made to sleep on verandas or even in field huts when protecting the crops against bears and elephants. Girls over the age of seven or so will be happy to spend the night in the house of a widow and the company of female age-mates. All of this implies that – as a matter of routine – only small children sleep in the same house as their parents. In the Kond Hills, all inhabitants have a short life expectancy compared to our own in the West. All must be hard-working and will own only a few material goods, which after their death will be distributed to close friends of the deceased in public. At the same time, women and men have many opportunities to celebrate and explore a considerable quality of life. I have never been given the impression that a young female or male was forced into marriage with an unwanted spouse. Giving and taking bridewealth is a way to formalize relationships between the two affinally related local descent groups. Even as a social framework of dramatic collective encounters in public, the obligation provides social predictability in an anarchic sociopolitical context. I simply cannot conceive of it as a source of exploitation or inequality applied between either women and men or cultivators and their commercial agents.

Bridewealth and Gender in Highland Middle India • 183

Private Property As to the ‘property career’ (Morgan 1877: 457) in the highlands, again the initiation of state interference seems to be a relevant factor in differentiating ideas of ownership. Apparently, in Chota Nagpur and some other areas, land has been ‘settled’ for a long time and assigned to individual owners to be inherited, while further south the ‘settlement duty’ is a recent phenomenon. However, collective control of village territories, assigned to the assumed descendants of the first settlers who cleared the forest, continues to be the normative concept. In the north of the culture area, villagers following the old religion reserve a sacred grove (jaher) of uncultivated land and uncut trees as their centre of worship, while the Kond and the inhabitants of Koraput further south keep rock arrangements or an earthen pot (hundi) as the ritual focus of a village. Specific forms of music and the formulas of the sacred specialist invite the Earth Goddess to come up, enter these sites and communicate with her devotees by accepting the gift of sacred food that is shared by those who are qualified to participate in the ritual. The Goddess is the land. As such, she provides the collective of villagers with their daily cereals, irrespective of what has been settled by the state. As return gifts, the cultivators sacrifice animals to their bloodthirsty mother. They raise chicken and pigs,20 goats and sheep, cows and buffaloes, but do not drink milk or eat eggs. In the Kond villages in which I stayed, the ruminants, if they are not to be sacrificed in the foreseeable future, have to pull the plough or are regularly taken out by an old weaver to graze in the woods, whereas all other domestic animals are mostly21 left untended22 until their time of sacrifice has come. They are not being worshipped,23 but are raised as future gifts to please multiple divinities. The largest of all, the buffalo, is periodically understood to be the Earth Goddess and, at the same time, is sacrificed to her.24 Unmarried young men who are affines of the sacrifiers hack the animal to pieces, because the crops will only grow after the blood of the Goddess has fertilized the fields. These Indigenous people do not ‘own’ land. As a village collective and subdivided into different local descent groups, they have instead, by virtue of their long-term mutual exchange relationship with the Goddess, acquired the privilege to cultivate her and harvest her fruits. Accumulated individual property of any kind makes no sense. Highland people entertain multiple reciprocal obligations in order to find their spouses and also enough to eat and to drink. Morgan’s

184 • Lewis Henry Morgan’s Comparisons

compassion for them was genuine, though like all other historical and contemporary ethnographers, he belonged to a different world.

Notes   1. A road covering some 30 km had been built in the 1940s to provide access to a hydroelectric project of the colonial government to supply the coastal cities with electricity. I cannot discuss the complex implications of this venture here.   2. This is a very common clan name.   3. The idea that land could belong to a women was out of the question for all parties involved.   4. Due to the Indo-Pakistan war of that year, the Census of India of 1971 had been reduced to minimal coverage and that of 1981 had not yet been published.   5. The situation may have been different in the outskirts of the Chota Nagpur plateau and other areas of mineral wealth, where the colonial agencies had been active since the beginning of the nineteenth century or even earlier.   6. I found many entrances of particular villages where they were not situated when I personally visited them.   7. An excellent introduction with many essential references is given in Uberoi (1994).   8. More than fifty years of regular visits and four long-term stays in South Asia have convinced me that mainstream Muslims and lowland Christians have retained the traditional dowry obligations of caste India.   9. By contrast, the weavers are certainly conscious of gains and investments, even if they only very rarely manage to accumulate a fortune worth mentioning. The Indigenous worldview excludes them from landownership. 10. Hardenberg (2018) has described and analysed this in great detail. This total social fact of the Kond used to be a human sacrifice. The British prohibited it and initiated the so-called ‘Meriah wars’, which led to many Kond deaths. Thereafter, buffaloes were substituted for humans as sacrificial objects. 11. The Gadaba bridewealth consists of a female calf and an adult cow along with the usual liquor, rice and money. The calf will be raised by the bride’s people, whereas the cow will be sacrificed to the village goddess and – from head to rump – will be cut into two halves. While most of this meat will be consumed by the two ‘sides’ jointly during the festivities, the right hind-leg of the cow is taken back by the bridegroom to his village, offered to his village goddess and consumed by the members of his local descent group. Thus, during the marriage season in spring, it is not at all unusual to meet young men surrounded by their respective parties and carrying such a leg on their shoulders. 12. This is about two hours’ walk from Burlabaru, which was mentioned earlier.

Bridewealth and Gender in Highland Middle India • 185

13. During the most important of these rituals, unmarried external affines of a village kill their sacrificial buffalo for the Earth Goddess (see Hardenberg 2018). 14. Gell (1992) has published a conscientious monograph on the character of a modern specific (i.e. ‘mixed’) dormitory that became famous by Elwin’s sympathetic account (1947), though many of his Western readers, misunderstanding the book, thought they could find their own version of ‘free love’ between individuals. My article (Pfeffer 2010) focuses on the details of dormitory organization as found in the literature on different highland societies of Middle India and as personally experienced in the Kuttia Kond and Gadaba villages. 15. Though I did not see this during my stay in the months before and at the beginning of the harvest, my hosts did not hesitate to inform me of it. 16. Hardenberg (2018: 79) has discovered this and has translated the names of the sociocentric categories, since the Kond words are of their Dravidian language Kui, whereas the Pan use expressions of the Indo-European lingua franca Desia for the same clan names. 17. The Kuttia and Dongria Kond do not allow cross-cousin marriage, whereas the Pan living in the same villages prefer such a special rule of diachronic affinity. 18. More than 200 years ago, the region of Chota Nagpur was subjected to colonial rule, with industrial and missionary interference beginning in the mid nineteenth century, whereas active social intervention by the state started in Koraput and the southern Kond mountains as late as the 1970s. 19. I refer to Guzy (2013) Otten (2006), Strümpell (2006), Berger (2015) and Hardenberg (2018). 20. During the 1970s, the Gadaba and some of their neighbours have, as a result of external influences, stopped raising pigs, while several of their collective ‘juniors’ continue to keep these animals. 21. Added to a Kuttia Kond house is a small pigsty for sows when they are delivering and feeding piglets. 22. Anyone who is interested can observe amazing forms of cooperation in a village between different species. 23. The sacrificial buffalo is worshipped extensively after assuming the identity of the Goddess. 24. The myth relates how she cuts herself into pieces.

 10 The Dark Side of the Moon

The Others Morgan was a family man, a bourgeois lawyer, a regular churchgoer and an eminent scholar. From an early age, he was also an actively engaged political person. During his lifetime, the remaining Native Americans were forcibly removed from their homes and driven to lands of different ecological conditions, where they were made to earn their living using agricultural techniques that were full of risks and unfamiliar to them. Most of the displaced Amerindians died or became destitute. For many years, Morgan tried to organize resistance against these policies of his government. His active interference saved some Iroquois from removal and death. Like Morgan and since colonial times, many mainstream citizens of India, both women and men, have opposed the social discrimination of the Adivasi, though these activists have succeeded only to a very limited extent. Since independence, successive governments have developed a large administrative body for the ‘uplift’1 of the Scheduled Tribes (STs), though such official efforts lacked a clearly comprehensible policy. An honest and far-reaching stocktaking of past achievements has been avoided and social discrimination continues as it did in the old days. At times it also develops new forms. Thus, in the late nineteenth century, a system of indentured labour was developed by the British, who deployed the Adivasi from the Koraput plateau and elsewhere to the tea gardens of northeastern India, where these male and female workers were made to carry baskets and pick the leaves. These kulis did not have rights in the past, and at present the regular Indian labour laws do not apply to them. Their indentured status has not changed during the last 120 years. Another example of a recently invented form of exploitation is a new commodity sold by clever businessmen. They have copied the beautifully drawn pic-

The Dark Side of the Moon • 187

tures of Sora shamans, i.e. their visions of another world that should always be erased soon after being created on the wall of a mud-house (see Vitebsky 1993). Instead, outsiders have printed these sensitive experiences onto t-shirts and matchboxes, and sell them to Western tourists on the beaches of Andhra Pradesh. Over many years, I have also observed officers of the ‘Dongria Kond Development Agency’ openly selling bronze statues meant for Kond rituals to foreign travellers who, at times, tried to sell on these objects of worship to museums in Europe at a much higher price. Many other forms of exploitation and discrimination can also be observed. No ‘colour bar’ separates the Indigenous highlanders from other inhabitants of the same province. Nobody would be able to identify the external appearance of a single individual as that of a person born and raised either in the western hills of the province or in coastal Odisha, but in the sociocultural context, the Indigenous people are immediately recognizable on account of their dress, their language and especially their general way of communicating. In some areas, such as the Kondmals and the Koraput plateau, lowlanders will mostly be taller and usually much heavier than the Adivasi. Briefly, an individual’s background is often quite clear before any form of communication. There are many social differences among these two categories. To me the most striking one is the independent social conduct of Indigenous women and men. In caste society, discrimination is an omnipresent feature and low-caste members will probably object to their status, but not to the authoritarian system as such. These isomorphic structures have been clearly elaborated by Michael Moffat (1979: 99‒153). The Adivasi, however, are not a part of such a system and will not accept a place in it. They may have to live in an urban slum or work as kulis in road construction, but this will not affect their self-esteem. Perhaps the much-lamented alcoholism2 is a suitable indicator of this difference. In the coastal towns of Odisha, I found hard liquor available everywhere in large quantities. Members of the peasant and working classes will find it even in the shortest lane of their respective ward. However, consumption is a secret activity. Mostly male drinkers seem to intoxicate themselves quietly and systematically indoors and without much company. By contrast, the men of the highlands and to a lesser extent the women are very outspoken in their desire to drink liquor. Nobody is ashamed of having been drunk in public and the most sacred rituals imply that at least a major part of the faithful do not remain sober. The ceremonies include ‘senior’ and ‘junior’ kinds of hard liquor according to different categories of participants. On such outstanding occasions, I have seen assemblies of more than a thousand

188 • Lewis Henry Morgan’s Comparisons

people in an advanced state of intoxication that could be interpreted as Durkheimian effervescence.3 Anthropologically, Morgan had committed himself to two rather different strands of activity: on the one hand, he immersed himself in the examination of Indigenous sociocultural relationships, formal as well as informal, and on the other hand, he fought against injustice and illegal social pressure committed by the dominant Euro-American majority against the Indigenous Americans. Regarding these issues, the following sections will discuss such experiences in several sectors.

The Party Prices on the New Delhi real estate market are comparable to, or higher than, those in Paris or Rome. In 2006, I was most impressed on entering the park-like piece of property consisting of lush lawns, banyan trees and beautiful flowerbeds between the three or four most generous mansions in the very centre of the capital. At the gate, some thirty uniformed security men were waiting to drive the incoming cars and lead about 150 guests of the party to various terraces, where several bands offered excellent Western music while attentive young waiters swiftly moved around to serve the finest strictly vegetarian food along with the most exquisite French wines. My host, who was aged about seventy-five, represented a family that had, over much of the past century, been close to the country’s ruling dynasty. He also mentioned his friendship with former German Chancellor Helmut Kohl and recalled several visits to my country, during the course of which he had met its most prominent business leaders. Naturally, he was highly educated and so was his wife of about twenty-five years who had just graduated from a leading university in the United States, which was the cause of the celebration that day. After hearing of my research in Odisha, he mentioned his lasting interest in the development of this province. Decades earlier, he had contributed his share by constructing and running several paper mills in the valley of the Nagavali River as it flows through Rayagada town in the central highlands. There the railway line directly connected the mills with metropolitan north or south India. From my ethnographic work in the Kond Hills surrounding the Nagavali, I was familiar with the site. As might be expected, all of the many trees that used to cover the highlands around Rayagada disappeared decades ago, but the mills own several hill-sites where they have planted rapidly growing Australian varieties to meet the ever-growing demand for paper in

The Dark Side of the Moon • 189

an emerging economy. Whatever the major substance making up the contemporary river is, it is of the unmentionable kind, but the rather large gated community attached to the paper company is supplied with filtered water from the drilled wells. It houses the families of immigrant engineers, technicians, administrators and skilled workers who can use the services of a small hospital free of charge, a cinema and, of course, all the necessary educational facilities. In their neat uniforms, younger and older students walk across the many lawns or board the buses on the company-owned roads. The English language schools and colleges employ female and male teachers from all states in the country. Sports facilities like playing grounds appear to be in excellent condition. However, the Indigenous population has not received any kind of benefits from this. Many other central highlands contain almost the entire coal reserves as well as most of the country’s minerals and rare earths, all very much in demand on the global market. The most prominent industrial complex was founded in 1907 by the Tata family’s company in a village of formerly independent Munda cultivators. Now the place is simply known as Tata or the city of Jamshedpur in Jharkhand. Being unable to bear the sight of the many slums in a thoroughly polluted environment, I have always tried to avoid the many legal and illegal mining sites and activities in the industrial zone, but had to visit Rourkela occasionally. Until 1956, this used to be a Kisan village on the northern border of Odisha, but then the well-known German trusts of Krupp, Demag, Mannesmann and Siemens developed it into an integrated steel plant. Today, the modern city with its many slums is permanently embraced by a thick grey cloud of extremely polluted air ‒ pure poison, one could say. In the south of the same province, on the Koraput plateau, I passed through the town of Damanjodi and had to observe how the huge plant of the National Aluminium Company (NALCO) had turned the neighbouring rivers into smelly yellow sewers.4 Like the gated community of the Rayagada paper mills, the employees in this company can also draw filtered water from its wells, whereas the villages in the nearer or more distant vicinity of the two industrial complexes must ‒ as has been the practice for millennia ‒ take their water supply from the rivers. In Rayagada, even for small quantities, the standard colour of the water is brown and in the villages surrounding Damanjodi it is yellow with a tendency towards orange. Perhaps I should also mention that the Indigenous villagers cannot simply move to another region after experiencing such forms of industrialization or after their settlements have been flooded by a

190 • Lewis Henry Morgan’s Comparisons

hydroelectric project which has frequently happened during the last six decades. Their entire spiritual and social life, and not just their economic condition, depends upon their location, since their village is a part of an extensively interacting socioreligious network of affinal and intra-agnatic exchanges. The villagers central place of worship is the local shrine of the Earth Goddess. Their earth is their life.

Ashram Education After independence, Gandhian concepts were supported by a major section of the Indian population. For the Indigenous areas, they envisaged a gradual or ‘soft’ transformation to modernity and the major Gandhian foundation deliberately tried to implement such ideas in a ‘slow’ or inoffensive manner. During the 1980s, I was able to meet five female Gandhian volunteers in different villages in several very remote regions in the Odishan highlands. For decades, these social workers had shared local life. As young women, they had vowed to remain celibate and serve the less fortunate people of their country. Each of them was living in a very ordinary thatched mud-house that they called the ‘ashram’ of a village, because any visitor was always greeted and treated as a guest. Vegetables from their kitchen gardens provided their meals, along with cereals supplied by the cultivators. The unmarried female teenagers of these settlements seemed to accept the authority of these senior outsiders in relation to matters of hygiene and cooking, gardening and, on special occasions, dressing in saris. Literacy classes and the like were never intensively promoted and had failed. The men of the villages, even the Indigenous Bondo known to be ‘fierce’, respected these women. Most adult male Bondo regularly go to jail, because these men are notorious for their homicide record which is one of the highest in the world. None of the Gandhian women, who by the time I first met them had become elderly, ever complained about these ‘fierce people’. Some social workers in any of the different communities of Odisha were even reported to have occasional disputes when they vainly tried to make villagers stop consuming liquor. Two or three of the ‘paternal aunts’ (as they were called) had also been involved in village disputes and become rivals of the male elders, though the latter generally avoided any public display of authority, while the Gandhians did not hesitate to be affirmative and openly announce their dissent. By the 1990s, and after long periods of service, all of these dedicated volunteers had retired without being replaced. I remained in touch with some, who mentioned their miserable pensions (previously the

The Dark Side of the Moon • 191

salary had been extremely low) and that their respective families had to help them to meet their modest expenses. The word ‘ashram’ is also used for state-run boarding schools in special ‘development agencies’ that were institutionalized during the 1970s for ten to twenty administrative blocks, where most of the inhabitants were so-called Primitive Tribal Groups (PTGs). In one case, reliable data on such a development can be supplied from the period beginning in the mid 1950s. The German ethnographer Niggemeyer (1964a), then welcomed as an honoured guest of the administration and the academic circles of the province, reported how his wife and he stayed for about half a year right in the middle of the jungle inhabiting a standard wooden house constructed by the Forestry Department. Two or three rangers were the only nonIndigenous humans in their neighbourhood. About twenty-five years later, after some six hours’ jeep ride on narrow forest tracks across hills and right through several creeks, I could deposit my luggage in the same wooden building and spend my first night there, before moving on to a Kuttia Kond village. Meanwhile, development had begun. A veterinary doctor idled away his time at his station and the general physician looked after his friends among various external specialists, i.e. foresters, administrators and policemen, but mostly teachers of the ‘ashram’ school who had all been housed in several new buildings. Some years earlier, a dirt road had been built to the railways station about 30 kilometres away.5 When the monsoon was over, trucks could supply the construction materials for solid stone buildings and carry off the precious local timber to be delivered in urban India by train. In a small self-made mud-house, an unofficial immigrant sold tea and offered the usual roadside meals for the less privileged staff members of the new governmental institutions. Discipline was essential for the teachers of this ashram school.6 Each morning, the pupils had to stand straight to salute the hoisting of the national flag. On these occasions, they were also made to sing the national anthem. The gardening classes I was able to watch from a distance resembled military exercises. Not just the rows of cabbage and the irrigation creeks had to be absolutely linear, the boys too had to stand straight in a line for most of the time. Two or three times I observed physical punishment being meted out. Away from this development centre in the villages of the Kuttia Kond cultivators, I learned that none of their boys had ever stayed longer than three days at this school. Sometimes I could take a photograph of a pupil’s green shirt serving as a scarecrow in a field. However, the boys (not the girls) of the weaver (Pan) category remained in the ashram institution and adjusted to the conditions there. By the time they left they were able to read and

192 • Lewis Henry Morgan’s Comparisons

write. The most important lessons, they told me, were those on ‘clean food habits’. These instructions condemned beef consumption and reenforced the drinking of milk, or the inverse of what was practised in the Indigenous communities. In the same vein, religious teaching introduced the great Hindu epics and in particular the worship of Hanuman, who seemed to be a general favourite. One can imagine how some bureaucrat somewhere must have decided to assign the monkey god to the Indigenous population. Yet another twenty years later, an interstate motorway had been built only three kilometres away from the place, so the bus took only fifteen minutes to carry me from the stop on the old road to the site that used to be the ‘tribal development agency’. It had changed into a market town of 10,000‒15,000 lowlanders. During my visits to the neighbouring Kond villages higher up, I could not detect any changes in their socioeconomic conditions worth mentioning. The inhabitants were not really eager to mix with immigrants and vice versa. One innovation, however, was rather noticeable. At major festivals, or when they could somehow afford to, the Kuttia villagers hired electricity generators, record players and large loudspeakers from immigrant businessmen to listen all-night long to Hindi film music.

Health In the Bay of Bengal, the city of Puri has an imperfectly organized general hospital, as I was able to find out personally. Due to the high risk of infection, it is also a rather dangerous place for patients. In the highlands of the same province, hospitals or even dispensaries are almost entirely absent so that the rare appearance of a hospital came as a major surprise. When, in 2002, I fell seriously ill while staying on the Koraput plateau, I was lucky to find competent treatment at a nearby health station. The single-storey building alone was most remarkable. Outstanding architects must have designed the many inexpensive, low-ceiling rooms, each of which had broad windows and a door to an adjoining part of the garden. The patients slept on the floor, as if on the terrace of a local mud-house, and almost in the open, while equally simple and effective toilet arrangements attached to every room conformed to the strict rules of hygiene. Admission was swift and involved almost no bureaucratic concerns. Male and female nurses moved about in the same kind of clothes that the villagers of the local area were wearing. For the entire treatment period, patients coming from the rural neighbourhood were only charged an insignificant amount, a

The Dark Side of the Moon • 193

sum equivalent to the price of a meal at the weekly market. The idea of the institution was to make it easy to enter the building for those in need of treatment. The Indigenous people of the surrounding hills have an extreme fear of dying away from their home, because then their respective soul (or person) would not be recycled into the body of a newly born child in the same village; they would be dissolved or lost for good. To counter such fears, the dominant concept of the hospital in their midst was to convince patients that they could be healed and walk out of the place. Amazingly, the female administrator and the doctors were equally qualified and friendly. The medical practitioners included several specialists, and all held degrees from renowned universities in the northern or southern provinces of the Indian Union. From conversations I learnt that all were planning to spend the rest of their lives in this place on the plateau. They were fluent in the lingua franca of Koraput and had a very attentive and polite manner when treating any patient. Since the hospital did not suffer from the usual embezzlements by the staff, the supply of medicines was never a problem. Patients or their relatives were not, as is normally the case, asked to buy their pills and liquids in a nearby town. As an obvious difference in comparison to other health stations, this one was not surrounded by innumerable shacks of petty businesspeople, nor was there evidence of the usual heaps of waste material. Within a few days, I recovered and was allowed to spend another two days as a guest with access to the different wards. The only cloud of disenchantment above this otherwise inspiring institution arose from the hospital’s dealings with the district administration. It had taken the organization five years to obtain the necessary permits required to construct and run the institution. Whenever in subsequent years the addition of another room was planned, power or water supplies became a problem, a vehicle was to be registered or any other official contact arose, seemingly endless delays resulted in major headaches for the organizers. In Indian English, the expression ‘to harass’ is applied to such situations. Over the years, the district administration ‘harassed’ the hospital staff through many major and minor interventions, restrictions and delays. Though there seemed to be no obvious reason for this kind of bureaucratic sabotage, probably the apparently irrational and obviously damaging behaviour displayed by state officers was due to the fact that the medical staff of the hospital (but not the other employees) consisted of Christians hired by a respectable all-Indian interdenominational foundation. No visible signs of any religion could be detected in the buildings assigned to patients, no religious services were held in their rooms or in public, and

194 • Lewis Henry Morgan’s Comparisons

I have not met or heard of an Indigenous Christian within a radius of 30 kilometres of the hospital. And yet no explanation other than ordinary ‘communalism’ seems to make sense.

Revolutionaries During the 1960s, Naxalbari, a village in West Bengal, witnessed a peasant revolt that eventually led to the foundation of what is officially called the Communist Party of India (Marxist-Leninist), a militant division which was called ‘Maoist’ in the media. Soon afterwards, members formed an underground army against the state that, over the past few decades, has continued to grow. From the Bihar provincial border in the north to the southern districts of Indigenous people in Andhra Pradesh, Telangana and Maharashtra, militant operations are conducted throughout the highlands and especially in states and districts with a large Indigenous population.7 Since 2010, foreign nationals have no longer been allowed to enter the most affected areas of Odisha. The Inspector General of the provincial police patiently explained to me the many different ‘Naxal’ attacks, ambushes and kidnappings, and their centres of gravity. In response, the state has sent its own forces and also, for about fifteen years in Chattisgarh, encouraged private armies of local strongmen to engage in battles with the ‘Maoist menace’, but none of these measures has produced the expected results. Internationally this guerrilla war in the vast highland territory essentially remains unnoticed. On two separate occasions, I was able to have extended conversations with individual revolutionaries who happened to be ‘on leave’ in a coastal district. Following Chairman Mao’s call to hide in the midst of the ordinary people ‘like fish in the water’, they used to move in their parental neighbourhood before being called back to action in the hills or after this had been completed. Both were young men who had earlier been employed as schoolteachers, but who had not received their monthly salary for a year or two, after refusing to cooperate in the corrupt practices of their superiors in the education department. They were not familiar with the Communist Manifesto or other such literature, but rather were devout Hindus. As they were not only able to name certain highland villages that I knew rather well, but could also recall some individual villagers and access routes, they had most probably been there in person and given correct accounts of past events and implications. It goes without saying that they were passionate proponents of social equality.

The Dark Side of the Moon • 195

Naxalites are not Indigenous highlanders. They have chosen these territories of east-central India for action because there the agencies of the state are weak or altogether absent. I have stayed in villages where earlier I was told that first a band of revolutionaries and then, hot on their heels, a pursuing police party had demanded food and shelter. The armed female and male rebels had been content with the ordinary millet gruel, while police officers insisted on chicken and hard liquor. Alcohol was out of the question for these particular Naxalites. On the contrary, the militants took their time to preach temperance, as well as ‘clean food habits’. They too seem to have been eager to apply their educational ideals.

Religion The Indigenous religion of the highlands is indissolubly intertwined with the socioeconomic structure. The layout of a house or village is defined by metaphysical criteria and the environment is understood to be inhabited by countless named deities. All villages are interconnected with others by strong ritual obligations to be executed by the respective local descent groups. The cults link the domestic sphere with rituals serving the major noniconic deity of the settlement. Ancestors are recalled in ceremonial exchanges that are formalized within greater local units. No rituals assemble a unit understood to be ‘the tribe’, just as members of such an entity never appear to be unified in any social, political or economic action. All cultivators of the culture area locally worship the Earth Goddess as the most powerful deity and conceive the very remote and passive male God in the sky as of supreme status. The Indigenous commercial agents are always somehow involved in this cult. In one way or another, Indigenous rituals recall and re-present the miracle of edible plants growing from the earth and ripening as they are exposed to rain and sun from above. For the gift of their vegetarian food as the principle source of energy, rituals demand nonvegetarian gifts in return. After independence, the decennial Census of India decided to position the Indigenous population of Middle India in the ‘Hindu’ column. Such a measure is plausible in the event that this term is understood as an equivalent of ‘Indian’. As is generally known, ‘Hinduism’ differs from Abrahamic religions by being nondogmatic ‒ a ‘Hindu heretic’ is a contradiction in terms. Logically speaking, Hindus can do without any kind of -ism. I have met many members of the urban middle classes who openly denied that they believed in any divine being or ritual.

196 • Lewis Henry Morgan’s Comparisons

They were able to unquestionably continue as respectable colleagues and neighbours because they conducted their daily lives in the Hindu way. However, a major segment of the majority community in contemporary India seems to have adopted an Abrahamic understanding of religion. The proponents of this segment seem to be demanding a certain Hindu confession of faith and the adherence to dogmatically imposed norms. For example, they form or support militant ‘cow vigilance’ units. This new type of religious understanding is especially explicit with regard to India’s Indigenous people who are not familiar with the cults of the lowlands and fail to worship iconic deities. Highlanders do not visit temples, nor are they acquainted with or extraordinarily impressed by the great Hindu epics. Also, between them, these people have their own status differences as well as their particular rules of purity and pollution that differ considerably from Brahmanical concepts. In short, they ignore caste. In past millennia, such deviations were of little concern for the Indian mainstream, but the colonial science of demography has introduced the idea of a relative communal representation in all walks of civic life. Since independence, this concept is coupled with the conviction that the respective religion is being threatened, if at times the respective creed does not matter in questions of appointments and the like. Thus, in the eyes of their respective followers, the religions of all major and well-known communities seem to be ‘threatened’.8 Many historians have shown how quickly such numerical competition has become popular on the subcontinent and how far-reaching the political tussles over religious majorities and minorities have become at every level of interaction. Numbers were transformed into slogans to agitate crowds. If, since the late nineteenth century, Christian missionaries baptized Adivasi in Chhota Nagpur, such action ‘threatened’ Hinduism and the Hindu community as such. Over the course of recent decades, proponents of militant organizations demanding and enacting ‘defence measures’ seem to have gained significant influence. The research undertaken by my colleague Heinrich von Stietencron has shown that from 1998 to 2004, when for the first time the Hindu nationalist Bharatiya Janata Party headed the central government, more Hindu temples were constructed in highland Odisha than in the entire history before that short period.9 In other words, the Censusborn ‘numerical Hindus’ of the hills were meant to be converted into ‘real’ Hindus practising the rituals of the lowlanders according to welldefined confessions comparable to those demanded by the Abrahamic religions. Since this party of religious nationalists has captured the

The Dark Side of the Moon • 197

central government once again, it is obvious that Hinduism is to be developed into a movement similar to that of Judaism, Christianity or Islam, or into an assemblage of certain fundamental and dogmatic demands. In ordinary matters of daily life, the new policies have had only a very limited effect. Though I have been a guest in at least ten new cement temples in different highland regions, none of them has attracted the local people of the neighbourhood for religious reasons. But in a very tangible manner, the massive inflow of lowlanders has all the same had considerable repercussions in ritual matters. For example, the spring festival of Holi may now be observed in formerly remote areas. Elsewhere it is conceived of as an occasion when all bars of ‘caste and creed’ are removed in playful and peaceful colour throwing. Anyone who has witnessed this kind of joy in ordinary circumstances in a lowland town might not entirely agree with such a definition, since the experience of being hit by plastic bags full of smelly, liquid and acid chemicals may not count as playful entertainment for all parties involved, but that is incidental. On the Koraput plateau, the young male lowlanders employed by a hydroelectric project have, over the last thirty years, developed their own Holi delights by stopping the buses on their branch road leading to the main road. They then force all non-Indigenous passengers to alight from the buses. The remaining female10 passengers, who are usually on their way back from the weekly market and are unaware of the Hindu calendar, are subsequently asked to rise. Each of the young men will face each woman while rubbing chemical colours with a nasty smell over each part of her body above the legs. Younger and older female passengers may be forced to go through this ordeal two or three times during a single bus journey. Later complaints are never followed up because the police authorities refuse to intervene in these matters, which, they say, involve religious freedom. The young lowlanders, who are usually physically taller and much heavier than the Indigenous population, are also selective in their actions. They would not molest Bondo women, knowing very well that otherwise they would be killed within days by rather tiny Bondo men who never leave their villages without a bow and arrow and a small sword attached to their hips. These exceptionally aggressive highlanders keep a distance from all other communities. Because of their special devotion to the Goddess, Bondo women have vowed neither to grow hair on their skulls nor wear any other clothes than a self-made fibre skirt with a breadth of about 40 centimetres. They also refuse to marry a man who is not several years their junior. These shifting cultivators

198 • Lewis Henry Morgan’s Comparisons

rank among the societies with the highest homicide rate worldwide. Men kill other men and then report to the police, ending up in jail for five to ten years. But their women have always remained untouched. Accordingly, the sex ratio is unbalanced. Many Bondo men have been killed or sent to jail. Women organize the domestic economy and carry out the hoeing and the harvesting in the fields. In 1994, I discussed these norms with the District Magistrate of Malkangiri, where several large villages of the Bondo are located. Since the days of the British Raj, the Collector, as the position is also called, has wielded almost unlimited powers in his district. The man I met was a highly educated liberal and was energetically out to improve local conditions by enlightened and nonconventional measures. Soon after he took office, he promised to hand over the sum of 200 rupees, at that time more money than an ordinary Bondo had ever seen in his lifetime,11 to every woman who would marry a relatively older man, cover her body in a ‘decent’ manner and stop shaving her head. This had been half a year before my time there, so on arrival, I found about a third of the Bondo women I encountered were wearing their hair just above the ears and were covering their body and thighs with a kind of compromise cloth of a length between that of a skirt and a sari. I could not collect data on their marriages. On my return two years later in 1996, the Collector had been transferred, the prize money had been abolished and the Bondo hairstyle had returned to the original convention. The ‘enlightened’ attempt to alter the local religious practices had failed. Adivasi serve as ‘Hindus’ in the controversies between the politicians of India. The major political parties of the country are keen to ‘protect’ the Indigenous population from being ‘enticed’ to change their religion and become Christians or Muslims. Apart from this priority, most Hindus of the lowlands do not usually show any intention of sharing religious and social sentiments, or anything else, with the people of the highlands.

Leadership Whereas the Scheduled Castes (SCs) of the mainstream society, who are understood to be the lowest in a hierarchy of isomorphic status categories, assemble behind powerful female and male leaders who for a long time have played major roles in Indian political life,12 the STs of Middle India lack any kind of leadership and have not really participated in Indian politics during the past seven decades, even though all

The Dark Side of the Moon • 199

parties field ST puppet candidates in the reserved constituencies and occasionally appoint an equally dependent Adivasi as chief minister of a province.13 To some extent, the Indigenous leadership may be associated with village elders, although the latter, as ordinary cultivators, will go out of their way to avoid any display of superiority. They may influence opinions, but are not expecting ‘obedience’ from fellow villagers and do not articulate ‘orders’. Above the village level, it is very difficult to find men of influence, not to mention women, since such a position would be of the informal kind and not associated with extraordinary power. By definition, a ‘tribe’ as such, or a confederation of ‘tribes’, lacks any kind of institutional or noninstitutional dominant body. Spontaneous rebellions in the past, initiated after prolonged suffering, have seen charismatic leaders who were killed by state authorities and continue to be remembered by contemporary Adivasi. They are even present in general textbooks of history courses, though these accounts do not seem to influence attitudes. Highland society is characterized by its local and supralocal relationships of affinal and agnatic exchanges. It is thus no match for organized intervention at a higher level.

The Dark Side of the Moon Historians of ancient Rome are familiar with innumerable details of this complex society because of the many written documents left behind by its administration and the more or less prominent literati of the epoch. At the same time, due to a lack of such sources, scholars know very little about the ‘tribal’ societies on the borders of the empire such as the Berbers in the south and the Germans in the north, the Huns in the northeast and the Arabs in the southeast. Accordingly, these social systems in the vicinity of the Roman state are known as ‘the dark side of the moon’ (Bicknell 1984) in the historical disciplines. The same expression could be applied to contemporary Highland Middle India. Between 1920 and 1960, the great pioneers of professional British social anthropology conducted intensive research in most colonies of the modern empire, but India was left out completely. Perhaps they were unwilling to compete with the many colonial officers who had been ‘in the field’ for years or decades and who had operated as ‘census ethnographers’. After Indian independence, very little primary anthropological research has been conducted in the vast Middle Indian region. Because

200 • Lewis Henry Morgan’s Comparisons

of this deficiency, the theoretical literature of the discipline has almost entirely ignored the inhabitants. At the same time, Indian political authorities have formally adopted a policy of ‘protective discrimination’, while many of their officers, through innumerable practical conventions, convey a spirit of neglect and social discrimination. Very few formally educated people in India or other countries appear to take a serious interest in the culture area at all. Morgan had other inclinations. Being fully embedded in the Yankee cosmos, he nevertheless made every effort to understand the Native Americans in his neighbourhood and support the cause of the Indigenous people anywhere in the world.

Notes   1. Nityananda Patnaik’s work (2005) is a good example of such a very influential officer’s approach.   2. In quantitative terms, it is comparable with German alcoholism.   3. See Durkheim 1995 [1912]: 217‒18.   4. A detailed study on the aluminium industry in the Indian areas populated by Indigenous people is given in Padel and Das (2010).   5. This was from the west, whereas I had approached the place from the eastern direction.   6. In other regions, I found ashram schools with starving pupils in very poor health, because the teachers had embezzled most of the money that was to be spent on provisions. It goes without saying that teaching had ceased altogether.   7. For a detailed study, see Shah (2013, 2018).   8. For a detailed study of the politics of conversion, see Sundar (2016a).   9. WebGIS-Interface for the databank ‘Temples in Orissa’, compiled by H. von Stietencron (currently a work in progress). 10. Far more often than men, women would attend markets to buy textiles or small household articles. 11. It was then about a fifth of what a primary school teacher would officially earn per month. 12. The most powerful political leader of the SC was probably Jagjivan Ram, Deputy Prime Minister (1977‒79), who almost became Prime Minister. Several representatives of the SC have served in the less powerful post of President of the Indian Union. 13. A former northeast-Indian Adivasi Speaker of the Lower House had been born and raised within an affluent and educated urban context.

Conclusion For the Record

This book identifies symmetric affinal relationships as the basic constitutional principle of two Native North American societies ­ (Seneca and Omaha) studied by Lewis Henry Morgan. The same concepts also order many large and small anarchic peoples in contemporary Highland Middle India. The principle is formally implied by relationship terminologies that differ markedly from others commonly known as ‘Dravidian’, since Seneca, Omaha and HMI formally exclude the repetition of the same affinal relationships in subsequent generations. In the eighteenth century, the first colonial troops entered parts of Middle India to introduce some agencies of a state dominated by Europeans. These institutions were later perpetuated and extended in independent India. However, in many highlands the holistic social structures of the Indigenous inhabitants have remained intact, while externally imposed governmental institutions are successfully rejected, tolerated or ignored. Wherever the anarchic order of past millennia has been completely abolished, most of the surviving inhabitants have ended up living in slums. The affinal constitution of these highland societies is almost always articulated by patrilaterally assigned exogamous clan categories and sometimes also by phratries and moieties, all of which emphasize the concepts of inclusiveness, collectivity and mutuality that govern the idea of marriage. The most relevant, formally refined and empirically indisputable document of the affinal order is the unique semantic pattern common to all relationship terminologies of the given Dravidian, Indo-European or Munda languages. For the first time, this book identifies this pattern as a whole and discusses its details. The given example of the Kharia system demonstrates the relations of this structure in a single matrix. Minor deviations and differences in some other terminologies are also presented. The pattern is applied constantly by many million inhabitants of the culture area.

202 • Lewis Henry Morgan’s Comparisons

About 150 years ago, Morgan (1871) discovered the constitutional quality of such relationship terminologies, which he eventually called ‘classificatory systems’. The thesis of this book depends upon Morgan’s unique achievement, i.e. his recognition of the ideological holism contained in such structures and manifested by their ‘apparently arbitrary generalizations’. In several systems, particular terminological equations or specifications may differ considerably from each other to articulate different generalizations. Thus, HMI terminologies are clearly ‘classificatory’, though, as an obvious difference compared to Seneca, Omaha and many others, they do discriminate M from MZ as well as F from FB as lineal and collateral terms. Accordingly, the usual textbook criteria of classificatory systems highlighting the merging of these terms must be reviewed. My point is that such equations or nonequations should be rated as secondary features when compared to the primary distinction between ‘classificatory’ terminologies and the European ‘descriptive’ nomenclatures. The latter lack a structure since, as Morgan recognized, their vocabulary refers to concrete or individual relationships and not to classes defined by mutual oppositions. Morgan’s ingenious discovery cannot be overestimated, even if he and his successors were unable to recognize how the affinal principle governed the above-mentioned North American terminological patterns.1 In his elaborate ethnographic efforts, he made it a point to collect all relationship terms of a given language rather than confining himself, by questionable criteria, to certain selections, as was widely practised during later epochs. Trautmann (1987: 52) has pointed out that Morgan’s comprehensive method of presenting and comparing his data followed the procedure of the prominent British legal scholar William Blackstone and was thus biased in favour of the concept of descent lines that govern European family law. However, the voluminous documentation in Systems of Consanguinity and Affinity of the Human Family has still proved to be invaluable because it has allowed – and continues to allow – generations of anthropologists to adopt, reconsider and ­criticize his findings in a meticulous manner. Morgan not only compared Amerindian and many other past and present relationship terminologies, but also associated these systems with the dominant concepts of power and property in the respective societies. His explicit general research strategy was the comparative method. In one form or another, it became the primary foundation of the new discipline during the last 150 years and continues to produce impressive results in anthropological research. The method enabled Morgan, and subsequently many others, to break through the cognitive walls of Eurocentric classifications and certainties in order to

Conclusion • 203

­ ypothesize about specific ‘plans’ of different societal constitutions. h Through their ‘apparently arbitrary generalizations’, the above-­ mentioned Amerindian terminologies were found to contain an overall structure of social relationships. They were not simple designations of family members, i.e. a ‘domestic institution’ (Morgan 1871: 11) in the European descriptive sense. Because all parts of such a structure must be interrelated, their respective generalizations are bound to express the given societal values. The discovery that classificatory terminologies and other primary social relationships were frequently governed by principles of affinity must be attributed to Lévi-Strauss (1949). Subsequently, Dumont (1957, 1983) offered an equally disciplined and refined analysis of the specific affinal order governing South Indian peasants. He distinguished the formal from the empirical, i.e. the three separate analytical levels referring to social categories, normative regulations and observable practices. A comparison of the results of the respective analyses may indicate contradictions that are likely to have been consequences of historical changes. These changes must have been irreversible in the sense that norms tended to deviate from structures of categories and categories would eventually break up, as Needham (1967) proposed after offering comparative examples from different continents. Like Lévi-Strauss and Dumont he had been inspired by their teacher Mauss and The Gift (1923–24). This work had pointed to the holistic social obligation to give, take and reciprocate ‘in archaic societies’. Here I assume that an overall social order of affinity is characterized by ‘the spirit of the gift’. At the same time, it should be clear that this ‘spirit’ does not imply some general altruism. In fact, I have witnessed how people quarrelled over the nature and the quantities of some prestations, and how ritual and marriage matters were often accompanied by controversies, as is common everywhere in the world. The ‘spirit of the gift’ amounts to the formal framework of sociality contained in the relational pattern as such rather than a form of kindness of actors involved in social intercourse. The chapters in this volume have also drawn considerable inspiration from Needham’s extended comparative efforts to further refine and systematize the contributions of Lévi-Strauss and Dumont by his concept of ‘prescription’ (1973) that is applied to terminologies articulating a positive imperative. Such formal structures invariantly prescribe specific affinal relationships over time either from one generation to the next or by alternating specified ties in a certain manner. Though neither Morgan’s Amerindian systems nor those of Highland Middle India are prescriptive, this book proposes that their respective social

204 • Lewis Henry Morgan’s Comparisons

Figure 11.1 Indicators of Seneca and Omaha symmetry. Figure created by the author.

order is governed by equally holistic principles of affinity, a point that obviously requires further explanation. At the core of both the Seneca and the Omaha terminology, the equations BW=HZ and ZH=WB indisputably indicate the systems’ affinal symmetry (see Figure 11.1 above). Formally, the gift of a sibling in marriage must be equally and directly reciprocated, though no links to an upper generation allow Ego’s spouse to be identified by the terminology in any manner. Since Ego’s spouse formally remains unconnected to any previous generation, these two systems do not prescribe a diachronic exchange, but, on the contrary, formally exclude a repetition of intermarriage. However, this negative implication, or proscription, can easily be articulated by a positive phrasing. Though the ‘spirit of the gift’ in these systems disallows an affinal exchange relationship between only two lines in a constant ‘give and take’, it positively implies multiple alliances within the societal whole, or a formal network to link all members systematically in reciprocal relationships rather than separate ‘pairs’ of allies. Thus, the imperative is identical, with the formally institutionalized affinal exchanges involving all the different clan categories of a people. The HMI structure contains the same imperative to vary affinal relationships, though, compared to Seneca and Omaha, it duplicates the central or encompassed relationships of affinal symmetry by distinguishing BW from HZ as well as ZH from WB. At the same time, the structure is able to avoid asymmetry by introducing a second consanguineal line represented by HZH and WBW respectively. Formally, HZH and WZW are distinguished from B and Z, and yet they must be understood to be consanguines (see Figure 11.2). Since BW is not equated with HZ and ZH is not merged with WB, these distinctions at first sight appear to contradict the notions of affinal symmetry. However, HZH and WBW are given as the representatives of a second consanguineal line2 so that the central affinal relationships

Conclusion • 205

Figure 11.2 Indicators of HMI symmetry. Figure created by the author.

are invariantly defined by symmetry. I am not aware of other formally symmetric systems specifying BW≠HZ and ZH≠WB. The fact that my comparison between, on the one hand, these or other Native North American terminologies and, on the other hand, HMI results in a common and very obvious affinal imperative – i.e. the formal obligation to entertain multiple symmetric affinal contacts over time – may be evaluated as an ironic reminder of Morgan’s thesis that had mistakenly equated the semantic pattern of the Seneca and the Tamil terminology. The ‘Dravidian’ systems of South India contain the affinal imperative to continuously ally only two terminological lines, whereas such a diachronic implication is formally out of the question for the affinal relationships of Seneca, Omaha and HMI, all of which demand new alliances with the generational change, even if such a formal implication is contradicted by empirical rules of cross-cousin marriage in perhaps half of all HMI societies. Affinal ties remain at the heart of the organization. By contrast, I have tried to show in Chapter 6 that Morgan’s ‘Malayan’ systems, like descriptive vocabularies elsewhere, altogether disregard affinity and the associated complications in the above-described collective sense. Thus, the ‘spirit’ of these ‘generational’ systems can hardly be compared with that of ‘Dravidian’, Seneca, Omaha and HMI. The latter three terminologies would be ‘semi-complex’ in LéviStraussian phrasing, but I reject the evolutionary tinge in the comparison of Lévi-Strauss’ ‘elementary’, ‘semi-complex’ and ‘complex’ systems. Instead, I identify the first two structures (‘elementary’ and ‘semicomplex’) as holistic ones in the sense of Morgan’s classificatory terminologies, whereas the kinship vocabulary of European languages is no system at all, but rather an assemblage of terms. Accordingly, structural analysis may be applied to prescriptive as well as the proscriptive terminologies, such as those discussed in Chapters 3–5, but is unsuitable for descriptive vocabularies. Prescriptive and proscriptive systems equally constitute holistic societal orders based upon the value of affinity.

206 • Lewis Henry Morgan’s Comparisons

In Highland Middle India, relative seniority is the second major principle of the terminology. It is also applied to innumerable other markers of social status, such as inter-‘tribal’ and interclan relations or those between different local descent groups. In a ritual context, not only actors but also specified foodstuffs and drink may be either ‘senior’ or ‘junior’. The HMI ideology of the affinal ‘gift’ constitutes the anarchic social order and also one lacking significant private property. Private wealth is given away in regular or extraordinary ceremonies and, as a result of such action, is the source of prestige. This major difference, when compared to peasant societies that focus on individual land control, must have relevant evolutionary consequences, since the complexity and sophistication of HMI terminologies can hardly be maintained as the constitutional order of a political economy marked by individual efforts to obtain and defend landed property. Thus, it could be argued that the complex peasant economy of South India must go together with a considerably simpler affinal system. A similar ‘reduction of affinal complexity’ could be achieved by the well-known descriptive terminology that goes together with the complex land tenure in North India. Thus, this book has tried to substitute Morgan’s concept of social evolution with more recent proposals comparing the relative simplicity or complexity of human social intercourse in several sectors and suggesting the threefold division of: (1) individualistic hunter-gatherers; (2) anarchic societies, or those only structured by ‘kinship’; and (3) peasant societies within centralized states. Several or many features of one of these ideal types may continue to exist encompassed by institutions of the more complex social order and no empirical people must have passed through ‘earlier periods’ to reach ‘later’ ones, just as any moral judgement about a ‘lower’ or a ‘higher’ stage is unwarranted. It should also be clear that the extremely complex forms of contemporary capitalism can only be analysed adequately by the specialists of other disciplines. Given these differences, the societies discussed in this book represent the anarchic form that for millennia was the most relevant one of anthropos. It includes peoples raising domestic animals, tilling land and also others who – like the noncultivating Amerindians on the Northwest Coast – had been endowed with abundant natural resources to develop settled communities and complex social and ideological systems. By referring to the available historical sources, I have tried to show that – before European occupation – Australian aborigines too must have belonged to this second societal form rather than to the hunter-gatherer category. Accordingly, their complex worldview or their sophisticated social order should not be rated as an ‘anomaly’.

Conclusion • 207

Under contemporary conditions, any research on the practical operations of an anarchic or ‘kinship’ society is bound to be difficult in view of the catastrophic global European expansion that has exterminated or significantly reduced the number of Indigenous societies. In North America, for example, only a small fraction of the native population has survived and continues to employ Indigenous concepts such as moiety, phratry and clan or to apply the Indigenous relationship terminology. These instruments of societal order had been designed for much larger demographic units. Accordingly, a contemporary people of a few hundred Native North Americans may conceive its eight constitutive clans as if they referred to corporate descent groups, whereas, by comparison, a contemporary HMI people comprising a few hundred thousand members, such as the almost unknown Indigenous Joria of Koraput/Odisha, would never mistake its eight constitutive clans for anything other than categories of affinal exchange. The book has tried to introduce the empirical building blocks of the old HMI social structure by initially generalizing two aspects: • All Indigenous owner-cultivators of the land cooperate with ritually inferior and landless clients who are as poor as the former. On behalf of the entire village, these so-called weavers work as musicians, craftspeople, cattle herders and labourers as well as commercial agents in transactions at the local weekly market or dealings involving the commodities of urban India. They also organize the transfer of valuables between their patron collectives, such as bridewealth or prestations of the main rituals. I estimate that these equally Indigenous clients comprise about a third of the population in a highland village. So far, very little relevant research has been conducted on these essential intermediaries, who are normally not mentioned at all in ethnographic accounts. • Following Sahlins (1968a), I must emphasize that the Indigenous order is primarily maintained at the local level. Belonging to a ‘tribe’ indicates a certain status, but is almost irrelevant for acts of internal administration, just as the ‘clan’ is an exogamous category and in some communities is associated with a larger clan territory, but never a corporate group. The organizational centre of all activities is the household, with its sacred kitchen being accessible only to closer agnatic relatives and women who after losing their agnatic status in their original village on marriage become close agnates of their husband’s people. Publicly operating corporate units with decision-making powers in matters of cultivation, ritual and even feuding are agnatic local descent

208 • Lewis Henry Morgan’s Comparisons

groups (LDGs) as defined by Leach (1961: 56). Two or more of these, and sometimes as many as eight, dominate village affairs. The fact that they are formally classified as relative seniors and juniors, or as bearing sacred or secular status, must not be misunderstood as an indication of power differences. In the major events of village life, either once in every generation or at least after a good number of years of a ritual cycle, these LDGs organize Maussian ‘total social facts’, or grand giveaway ceremonies on the occasion of collective megalithic secondary funerals (Pfeffer 2001) or at the Kond grand buffalo sacrifice for the Earth Goddess (Hardenberg 2018), when crowds driven by Durkheimian effervescence can be observed. Many other such major rituals that connect a considerable number of villages have never been studied. The interacting units are LDGs defined as such by their relative affinal as well as agnatic status. An LDG or one of its subunits will also organize and finance each one of the empirical marriages of its members as a major event of a village. Moreover, the particular term for marriage is also applied to other life-cycle ceremonies and to healing rituals which are bound to imply interactions between members of the respective LDG and external, affinally or agnatically related local groups. A considerable bridewealth must always be conceived as a circulating fund that formally ties together the involved LDGs of different villages, since these villages are frequently dominated by a single clan. Such a transaction ceremonially founds a marriage both in the case where the union has been arranged between the respective elders over many years to involve a constant flow of gifts, and also when the two principals have themselves decided to organize an apparent ‘abduction’ that is bound to be followed by elaborate disputes and acts of reconciliation between the respective villages. Whereas the HMI relationship terminology formally opposes categories of affines and consanguines and of seniors and juniors to imply multiple symmetric exchanges, the very same confrontations are empirically experienced in the economy, in ritual and in a normative order without rulers and ruled.

Notes 1. Morgan did not have access to HMI terminologies, which were only explored much later. 2. Two different consanguineal terminological lines have for a long time been identified in aboriginal Australian systems.

Glossary Indigenous people of India. Relationship of marriageability. Terminologically implied marriage relationship. Agnation Relationship through a male descent line that includes males and females. Clan Exogamous social category assigning a person to the maternal side in some societies and to the paternal side in others. Cognates Anybody sharing an ancestor with a person. Collaterality A person’s relationship to descendants of an ancestor’s siblings. Consanguinity Relationship that excludes the possibility of marriage. Cross-cousins Children of persons classified as sister and brother. Ego-centric Relationships from the perspective of an individual. Kindred Formally delimited social group containing a person’s collateral relatives. Local descent group (LDG) Local action group whose members are understood to share a known or unknown ancestor. Parallel cousins Children of persons classified either as two sisters or as two brothers. Skewing Relationship term equating members of different generations. Sociocentric Relationships from the perspective of a societal whole. Adivasi Affinity Affinal prescription

References Allen, N.J. 1998. ‘The Prehistory of Dravidian-Type Terminologies’, in M. Godelier, T.R. Trautmann and F.E. Zjon Sie Fat (eds), Transformations of Kinship. Washington, DC: Smithsonian Institution Press, pp. 314–31. Appell, G.N. 1976. The Societies of Borneo: Explorations in the Theory of Cognatic Social Structure. Washington, DC: American Anthropological Association. Bailey, F.G. 1960. Tribe, Caste, and Nation: A Study of Political Activity and Political Change in Highland Orissa. Manchester: Manchester University Press. Banks, D.J. 1974. ‘Malay Kinship Terms and Morgan’s Malayan Terminology: The Complexity of Simplicity’, Bijdragen tot de Taal-, Land- en Volkenkunde 130(1): 44–68. Barnard A., and A. Good 1984. Research Practices in the Study of Kinship. London: Academic Press. Barnard, A., and J. Woodburn 1988. ‘Introduction’, in T. Ingold, D. Riches and J. Woodburn (eds), Hunters and Gatherers, Vol. 2: Property, Power and Ideology. New York: Berg, pp. 4–32. Barnes, R.H. 1975. ‘Editors’ Introduction’, in J. Kohler, On the Prehistory of Marriage: Totemism, Group Marriage, Mother Right, R.H. Barnes and R. Barnes (trans.). Chicago: University of Chicago Press. ______. 1984. Two Crows Denies it: A History of Controversy in Omaha Sociology. Lincoln, NE: University of Nebraska Press. ______. 2012. ‘Omaha and “Omaha”’, in T.R. Trautmann and P.M. Whiteley (eds), Crow-Omaha: New Light on a Classic Problem of Kinship Analysis. Tucson: University of Arizona Press, pp. 69–82. Barnes, R.H., and T.R. Trautmann 1998. ‘“Dravidian”, “Iroquois”, and “CrowOmaha” in North American Perspective’, in M. Godelier, T.R. Trautmann and F.E. Tjon Sie Fat (eds), Transformations of Kinship. Washington, DC: Smithsonian Institution Press, pp. 27–58. Barraud, C. 2015. ‘Kinship, Equality and Hierarchy: Sex Distinction and Values in Comparative Perspective’. HAU Journal of Ethnographic Theory 5(1): 227–50. Behura, N.K. 1997. ‘Saora Guar Parab: A Secondary Mortuary Rite’, in G. Pfeffer and D.K. Behera (eds), Structure and Process, Contemporary Society: Tribal Studies, vol. 1. New Delhi: Concept, pp. 320–27. Berger, P. 2002. ‘The Gadaba and the “Non-ST” Desia in Koraput, Orissa’, in G. Pfeffer and D. K. Behera (eds), The Concept of Tribal Society. Contemporary Society: Tribal Studies, vol. 5. New Delhi: Concept, pp. 57–90. ______. 2010. ‘Who are You, Brother and Sister? The Theme of the “Own” and “Other” in the Go’ter Ritual of the Gadaba’, in P. Berger, R. Hardenberg,

References • 211

E. Kattner and M. Prager (eds), The Anthropology of Values: Essays in Honour of Georg Pfeffer. Delhi: Pearson, pp. 260–87. ______. 2015. Feeding, Sharing, and Devouring: Ritual and Society in Highland Odisha, India. Berlin: De Gruyter. Bicknell, P.J. 1984. ‘The Dark Side of the Moon’, Maistor: Classical, Byzantine and Renaissance Studies for Robert Browning. Byzantina Australiensa 5: 67–75. Bird-David, N., 1990. ‘The Giving Environment: Another Perspective on the Economic System of Gatherer-Hunters’. Current Anthropology, 31(2): 189–196. Bhattacharya, S. 1970. ‘Kinship Terms in the Munda Languages’. Anthropos 65(3–4): 444–65. Blackstone, W. 1893. Commentaries on the Laws of England in Four Books: Notes Selected from the Editions of Archibold, Christian, Coleridge, Chitty, Stewart, Kerr, and Others, Barron Field’s Analysis, and Additional Notes, and a Life of the Author by George Sharswood. In Two Volumes. Philadelphia: J.B. Lippincott Co. Blunt, E.A.H. 1931. The Caste System of Northern India: With Special References to the United Provinces of Agra and Oudh. London: Humphrey Milford. Bodding, P.O. 1942. Traditions and Institutions of the Santals. Oslo: A.W. Brǿggers. Bouez, S. 1975. ‘Parenté et Hierachie chez les Santal‘. L’Homme XV(3–4): 109–28. ______. 1985. Réciprocité et Hierarchie: l’Alliance chez les Ho et les Santal de l’Inde. Paris: Societé d’Ethnographie. ______. 1990. ‘L’Alliance Classificatoire chez les Ho: Préférence ou Prescription’, in F. Héritier-Augé (ed.), Les Complexités de L’alliance: Les Systems SemiComplexes. Paris: Editions des archives contemporaines, pp. 1–22. Bowler, J. 2014. ‘Mungo Man is a Physical Reminder of the Need for Indigenous Recognition’, The Guardian, 25 February. Buradkar, M.P. 1940. ‘Kinship among Gonds’, Nagpur University Journal 6: 147–80. ______. 1947. ‘The Clan Organization of the Gonds’, Man in India 27(2): 130–36. Carsten, J. 1997. The Heat of the Hearth: The Process of Kinship in a Malay Fishing Village. Oxford: Clarendon Press. ______. 2000. ‘Introduction: Cultures of Relatedness’, in J. Carsten (ed.), Cultures of Relatedness: New Approaches to the Study of Kinship. Cambridge: Cambridge University Press, pp. 1–36. Carsten, J., and S. Hugh-Jones (eds). 1995. About the House: Lévi-Strauss and Beyond. Cambridge: Cambridge University Press. Childe, V.G. 1951. Social Evolution. London: Watts. Collier, J.F. 1988. Marriage and Inequality in Classless Societies. Stanford: Stanford University Press. Darwin, C. 1859. On the Origin of Species by Means of Natural Selection. London: John Murray. Das S., and F. Padel (eds). 2010. Out of the Earth: East India Adivasis and the Aluminium Cartel. New Delhi: Orient Blackswan. Deeney, J. 1975. Ho Grammar and Vocabulary. Chaibasa: Xavier Ho Publications. Demmer, U. 2009. ‘Community-Society: Models of Social Life in Tribal India’, in G. Pfeffer and D.K. Behera (eds), Structure and Exchange in Tribal India and Beyond. Contemporary Society: Tribal Studies, vol. 8. New Delhi: Concept, pp. 266–80.

212 • References

Dorsey, J.O. 1884. ‘Omaha Sociology’, in Third Annual Report of the Bureau of Ethnology. Washington, DC: Smithsonian Institution, pp. 211–370. Dumont, L. 1953. ‘The Dravidian Kinship Terminology as an Expression of Marriage’, Man 53: 34–39. ______. 1957. ‘Hierarchy and Marriage Alliance in South Indian Kinship’, Occasional Papers of the Royal Anthropological Institute, 12. ______. 1961. ‘Marriage in India: The Present State of the Question. 1. Marriage Alliance in S.E. India and Ceylon’, Contributions to Indian Sociology 5: 75–95. ______. 1964. ‘Marriage in India: The Present State of the Question: Postscript to Part One; 2. Marriage and Status, Nayar and Newar’, Contributions to Indian Sociology 7: 77–98. ______. 1966. ‘Marriage in India. The Present State of the Question: III. North India in Relation to South India’, Contributions to Indian Sociology 9: 90–114. ______. 1980. ‘On Value’, Proceedings of the British Academy 66: 207–41. ______. 1983. Affinity as a Value: Marriage Alliance in South India, with Comparative Essays on Australia. Chicago: University of Chicago Press. Durkheim, E. 1995 [1912]. The Elementary Forms of Religious Life, Karin E. Fields (trans.). New York: The Free Press. Eggan, F. 1970a. ‘The Cheyenne and Arapaho Kinship System’, in F. Eggan (ed.), The Social Anthropology of the North American Tribes. Chicago: University of Chicago Press. pp. 35–95. ______. (ed.) 1970b. The Social Anthropology of the North American Tribes, 2nd edn. Chicago: University of Chicago Press. Elwin, V. 1939. The Baiga. Bombay: Oxford University Press. ______. 1943. Maria Murder and Suicide. Bombay: Oxford University Press. ______. 1947. The Muria and Their Ghotul. Bombay: Oxford University Press. ______. 1951. The Tribal Art of Middle India: A Personal Record. Bombay: Oxford University Press. ______. 1955. The Religion of an Indian Tribe. Bombay: Oxford University Press. Engels, F. 1884. Der Ursprung der Familie des Privateigentums und des Staats im Anschluss an Lewis Henry Morgans Forschungen. Hottingen-Zürich: Verlag der Schweizerischen Volksbuchhandlung. Evans-Pritchard, E.E. 1934. ‘The Nuer: Tribe and Clan’, Sudan Notes and Records 17(1): 1–57. ______. 1940. ‘The Nuer of Southern Sudan’, in M. Fortes and E.E. EvansPritchard (eds), African Political Systems. London: Oxford University Press, pp. 272–96. Feinberg, R., and M. Ottenheimer (eds). 2002. The Cultural Analysis of Kinship: The Legacy of David M. Schneider. Chicago: University of Chicago Press. Fletcher, A.C. and F. la Flesche 1911. ‘The Omaha Tribe’, in 27th Annual Report of the Bureau of American Ethnology. Smithsonian Institution, pp. 27–627. Fogelson, R.D. 2002. ‘Schneider Confronts Componential Analysis’, in R. Feinberg and M. Ottenheimer (eds), The Cultural Analysis of Kinship: The Legacy of David M. Schneider. Chicago: University of Chicago Press, pp. 33–45. Fortes, M. 1969. Kinship and the Social Order: The Legacy of Lewis Henry Morgan. Chicago: Aldine Press.

References • 213

Freeman, J.D. 1961. ‘On the Concept of Kindred’, Journal of the Royal Anthropological Society of Great Britain and Ireland 91(2): 192–220. Fried, M.H. 1967. The Evolution of Political Society: An Essay in Political Anthropology. New York: McGraw-Hill. ______. 1968. ‘Editor’s Note on Paul Kirchhoff’, in M.H. Fried (ed.), Readings in Anthropology, Vol. II: Cultural Anthropology. New York: Thomas Y. Crowell, pp. 370–71. Fuchs, S. 1960. The Gond and Bhumia of Eastern Mandla. Bombay: Asia Publishing House. Fürer-Haimendorf, C. and E. von. 1979. The Gonds of Andhra Pradesh: Tradition and Change in an Indian Tribe. New Delhi: Vikas. Gausdal, J. 1960. The Santal Khūts. Oslo: H. Aschehoug. Gell, S.M.S. 1992. The Ghotul in Muria Society. Chur: Harwood Academic Publishers. Godelier, M., T.M. Trautmann, and F.E. Tjon Sie Fat (eds) 1998a. Transformations of Kinship. Washington, DC and London: Smithsonian Institution Press. Godelier, M., T.M. Trautmann, and F.E. Tjon Sie Fat. 1998b. ‘Introduction’, in M. Godelier, T.M. Trautmann and F. E. Tjon Sie Fat (eds), Transformations of Kinship. Washington, DC and London: Smithsonian Institution Press pp. 1-26. Goodenough, W.H. 1951. Property, Kin and Community on Truk. New Haven: Yale University Press. Goody, J. 1973. ‘Bridewealth and Dowry in Africa and Eurasia’, in J. Goody and S.J. Tambiah (eds), Bridewealth and Dowry. Cambridge: Cambridge University Press, pp. 1–58. Goody, J., and S.J. Tambiah (eds). 1973. Bridewealth and Dowry. Cambridge: Cambridge University Press. Gregory, C. 2009. ‘Brotherhood and Otherhood in Bastar: On the Social Specificity of “Dual Organization” in Aboriginal India’, in G. Pfeffer and D.K. Behera (eds), Structure and Exchange in Tribal India and Beyond. Contemporary Society: Tribal Studies, vol. 8. New Delhi: Concept, pp. 67–82. Grignard, A. 1924. An Oraon-English Dictionary: In the Roman Character with Numerous Phrases Illustrative of Sense and Idiom, and Notes on Tribal Customs, Beliefs, etc. Mölding: Linguistische Bibliothek Anthropos. Grigson, V.W. 1938. The Maria Gond of Bastar. London: Oxford University Press. Guzy, L. 2013. Marginalised Music: Music, Religion and Politics from the Bora Sambar Region/Western Orissa. Berlin: Lit. Hardenberg, R. 2018. Children of the Earth Goddess: Society, Marriage, and Sacrifice in the Highlands of Odisha. Berlin: De Gruyter. Hiatt, L.H. 1965. Kinship and Conflict: A Study of an Aboriginal Community in Northern Arnhem Land. Canberra: Australian National University. Hirtzel, A. 1922. Statement Exhibiting the Moral and Material Progress and Condition of India 57. London: His Majesty’s Stationery Office. Hivale, S. 1946. The Pardhans of the Upper Narbadda Valley. Bombay: Oxford University Press. Hoebel, E.A. 1960. The Cheyennes: Indians of the Great Plains. New York: Holt, Rinehart & Winston.

214 • References

Homans, G.C., and D.M. Schneider 1955. Marriage, Authority, and Final Causes: A Study of Unilateral Cross-cousin Marriage. Glencoe, IL: The Free Press. Jena, D. 1999. ‘A Critical Analysis of a Tribal Uprising in Orissa: Case of Kandhamal’, in D.K. Behera and G. Pfeffer (eds), Social Realities. Contemporary Society: Tribal Studies, vol. 4. New Delhi: Concept, pp. 154–64. Karlsson, B.G. 2016. ‘Anthropology and the “Indigenous Slot”: Claims to and Debates about Indigenous People’s Status in India’, in N. Sundar (ed.), The Scheduled Tribes and Their India: Politics, Identities, Policies, and Work. New Delhi: Oxford University Press, pp. 109–33. Kirchhoff, P. 1932. ‘Verwandtschaftsbezeichnungen und Verwandtenheirat’ , Zeitschrift für Ethnologie 64: 41–71. ______. 1968. ‘The Principles of Clanship in Human Society’, in M.H. Fried (ed.), Readings in Anthropology, Vol. II: Cultural Anthropology. New York: Thomas Y. Crowell, pp. 371–81. Kobayashi, M., and B. Tirkey 2017. The Kurux Language: Grammar, Texts and Lexicon. Leiden: Brill. Kroeber, A.L. 1909. ‘Classificatory Systems of Relationship’, Journal of the Royal Anthropological Institute of Great Britain and Ireland 39(1): 11–22. Kronenfeld, D.B. 1989. ‘Morgan vs. Dorsey on the Omaha Cross/Parallel Contrast: Theoretical Implications’, L’Homme 29(1): 76–106. Kuper, A. 1999. Culture: The Anthropologists’ Account. Cambridge, MA: Harvard University Press. ______. 2009. Incest and Influence: The Private Life of Bourgeois England. Cambridge, MA: Harvard University Press. Layton, R. 2005. ‘Are Immediate-Return Strategies Adaptive?’, in T. Widlok and W.G. Tadesse (eds), Property and Equality: Vol.1. Ritualisation, Sharing, Egalitarianism. Oxford: Berghahn Books, pp. 130–50. Leach, E.R. 1961. Rethinking Anthropology. London: Athlone Press. Lévi-Strauss, C. 1949. Les Structures Élémentaires de la Parenté. Paris: Presses Universitaires de France. ______. 1964. Totemism, R. Needham (trans.). London: Merlin Press. ______. 1969. The Elementary Structures of Kinship, revised 2nd edn, trans. J.H. Bell, J.R. von Sturmer and ed. R. Needham. Boston: Beacon Press. ______. 1973. ‘Réflexions sur l’Atome de Parenté’. L’Homme 8(3): 5–30. Lounsbury, F.G. 1964a. ‘A Formal Account of the Crow- and Omaha-Type Kinship Terminologies’, in W. Goodenough (ed.), Explorations in Cultural Anthropology. New York: McGraw-Hill, pp. 351–93. ______. 1964b. ‘The Structural Analysis of Kinship Semantics’, in H.G. Lunt (ed.), Proceedings of the Ninth International Congress of Linguists. The Hague: Mouton, pp. 1073–93. Lowie, R.H. 1915. ‘Exogamy and the Classificatory System of Relationship’, American Anthropologist 17(2): 223–39. ______. 1928. ‘A Note on Relationship Terminologies’, American Anthropologist 30: 263–67. ______. 1929. ‘Relationship Terms’, Encyclopedia Britannica, 14th edn, vol. 19. New York: Encyclopedia Britannica Inc., pp. 84–89.

References • 215

Lyell, C. 1863. The Geological Evidence of the Antiquity of Man, with Remarks on the Theories of Origin of Species by Variation. London: John Murray. Mahapatra, L.K. 1959. ‘Ein Dorf des Berg-Bhuiyan Stammes. Eine empirischsozio-ökonomische Studie’, Ph.D. dissertation. Hamburg: Hamburg University. McDougal, C. 1963. ‘The Social Structure of the Hill Juang’, Ph.D. dissertation. Albuquerque: University of New Mexico. Mauss M. 1966 [1923–24]. The Gift; Forms and Functions of Exchange in Archaic Societies, I. Cunnison (trans.). London: Cohen and West. Meggitt, M.J. 1962. Desert People: A Study of the Walbiri Aborigines of Central Australia. Sydney: Angus & Robertson. Moffat, M. 1979. An Untouchable Community in South India; Structure and Consensus. Princeton: Princeton University Press. Morgan, L.H. 1851. League of the Ho-De-No-Sau-Nee, or Iroquois. Rochester, NY: Sage and Brother. ______. 1859. The Indian Journals, Leslie H. White (ed.). Ann Arbor: University of Michigan Press. ______. 1871. Systems of Consanguinity and Affinity of the Human Family: Contributions to Knowledge. Washington, DC: Smithsonian Institution. ______. 1877. Ancient Society. Chicago: Charles M. Kerr & Company. ______. 1881. Houses and House-Life of the American Aborigines. Washington, DC: U.S. Geological Survey/D.C. Government Printing Office. Murdock, G.P. 1949. Social Structure. New York: The Free Press. Naik, T.B. 1956. The Bhils: A Study. Delhi: Bharatiya Adimjati Sevak Sangh. Needham, R. 1962. Structure and Sentiment. Chicago: University of Chicago Press. ______. 1966. ‘Terminology and Alliance I’, Sociologus 16: 141–57. ______. 1967. ‘Terminology and Alliance II’, Sociologus 17: 39–53. ______. 1969. ‘Editor’s Note’, in C. Lévi-Strauss, The Elementary Structures of Kinship., revised 2nd edn., transl. J.H. Bell, J.R. von Sturmer and ed. R. Needham. Boston: Beacon Press, pp. xvii–xix. ______. 1971. ‘Introduction’, in R. Needham (ed.), Rethinking Kinship and Marriage. London: Tavistock, pp. xiii–cxvii. ______. 1973. ‘Prescription’, Oceania 43(3): 166–81. Niggemeyer, H. 1964a. Kuttia Kond: Dschungel-Bauern in Orissa. Munich: Renner. ______. 1964b.‘Kuttia Kond und Pano: Zur Stellung der verachteten Klassen in Indien’, in E. Haberland, M. Schuster and H. Straube (eds), Festschrift für Ad. E. Jensen (Teil 2). Munich: Renner. Otten, T. 2006. Heilung durch Rituale: Vom Umgang mit Krankheit bei den Rona im Hochland Orissas. Berlin: Lit. Overing Kaplan, J. 1975. The Piaroa, a People of the Orinoco Basin: A Study in Kinship and Marriage. Oxford: Clarendon Press. Padel, F., and S. Das (eds). 2010. Out of the Earth: East India Adivasis and the Aluminium Cartel. New Delhi: Orient Blackswan. Parkin, R. 1992. The Munda of Central India: An Account of their Social Organization. New Delhi: Oxford University Press. Parry, J.P. 1979. Caste and Kinship in Kangra. London: Routledge & Kegan Paul.

216 • References

Patnaik, N. 2005. Primitive Tribes of Orissa and Their Development Strategies. New Delhi: D.K. Printworld. Pfeffer, G. 1970. Pariagruppen des Pandschab. Munich: Renner. Pfeffer, G. 1983. ‘Generation and Marriage in Middle India: The Evolutionary Potential of “Restricted Exchange”’, Contributions to Indian Sociology 17(1): 87–121. ______. 1984. ‘Santal Totemism’, South Asian Anthropologist 5(1): 37–43. ______. 1993. ‘Kharia Totemism’, Journal of the Indian Anthropological Society 28: 221–27. ______. 1997. ‘Die Haardebatte. Gender, Glatzen und Gewalt der Bondo’, Zeitschrift für Ethnologie 122(2): 183–208. ______. 2001. ‘A Ritual of Revival among the Gadaba of Koraput’, in H. Kulke and B. Schnepel (eds), Jagannath Revisited: Studying Society, Religion and the State in Orissa. New Delhi: Manohar, pp. 123–48. ______. 2004. ‘Order in Tribal Middle Indian “Kinship”’, Anthropos 99: 381–409. ______. 2008. ‘Bondo Violence’, in M. Carrin and H. Tambs-Lyche (eds), People of the Jangal: Reformulating Identities and Adaptations in Crisis. New Delhi: Manohar, pp. 69–92. ______. 2010. ‘Jugend und Gender: Schlafhäuser der mittelindischen Stammesgesellschaften’, in D. Schulz and J. Seebode (eds), Spiegel und Prisma: Ethnologie zwischen postkolonialer Kritik und Deutung der eigenen Gesellschaft. Festschrift für Ute Luig. Hamburg: Argument, pp. 242–60. ______. 2014, ‘Ethnographies of States and Tribes in Highland Odisha’, Asian Ethnology 73(1–2): 259–80. ______. 2016. Verwandtschaft als Verfassung: Unbürokratische Muster öffentlicher Ordnung. Baden-Baden: Nomos. ______. 2017. ‘Engagement, Ethnographie und Gesellschaftstheorie: Das Vermächtnis von Lewis Henry Morgan (1818–1881)’, Paideuma, Mitteilungen zur Kulturkunde 63: 1–28. Pfeiffer, M. 2018. Kurux Historical Phonology Reconsidered: With a Reconstruction of Pre-Kurux-Malto Phonology and Copious Indo-Aryan and Munda Etymologies. Norderstedt: Academic Publishing. Polanyi, K. 1957. The Great Transformation: The Political and Economic Origins of Our Time. Boston: Beacon Press. Radcliffe-Brown, A.R. 1940. ‘On Joking Relationships’, Africa 13: 195–210. ______. 1941. ‘The Study of Kinship Systems’, Journal of the Royal Anthropological Institute of Great Britain and Ireland 71(1–2): 1–18. ______. 1952. Structure and Function in Primitive Society: Essays and Addresses. London: Cohen & West. Reichel, E. 2009. Notions of Life in Death and Dying: The Dead in Tribal Middle India. New Delhi: Manohar. ______. 2018. ‘The Ho: Living in a World of Plenty. Of Social Cohesion and Ritual Friendship on the Chota Nagpur Plateau’, Ph.D. dissertation. Frankfurt am Main: Goethe University. Reining, P. (ed.). 1972. Kinship Studies in the Morgan Centennial Year. Washington, DC: Anthropological Society of Washington.

References • 217

Resek, C. 1960. Lewis Henry Morgan: American Scholar. Chicago: University of Chicago Press. Riches, D. 1995. ‘Hunter-Gatherer Structural Transformations’, Journal of the Royal Anthropological Institute 1: 679–701. Rivière, P. 1969. Marriage among the Trio: A Principle of Social Organization. Oxford: Clarendon Press. ______. 1984. Individual and Society in Guiana: A Comparative Study of Amerindian Social Organization. Cambridge: Cambridge University Press. Rosman, A., and P. Rubel. 1971. Feasting with Mine Enemy: Rank and Exchange among Northwest Coast Societies. New York: Columbia University Press. Roy, S.C. 1912. The Mundas and Their Country. Calcutta: City Book Society. ______. 1915. The Oraons of Chota Nagpur: Their History, Economic Life and Social Organization. Ranchi: Man in India Office. ______. 1935. The Hill Bhuinya of Orissa: With Comparative Notes on the Plain Bhuiyans. Ranchi: Man in India Office. Roy, S.C., and Roy, R.C. 1937. The Kharias. Ranchi: Man in India Office. Rycroft, D.J. 2012. ‘From History to Heritage: Adivasi Identity and Hul Sengal’, in M. Carrin and L. Guzy (eds), Voices from the Periphery: Subalternity and Empowerment in India. London: Routledge, pp. 48–74. Sabean, D.W. 1998. Kinship in Neckarhausen, 1700–1870. Cambridge: Cambridge University Press. Sahlins, M.D. 1960. ‘Evolution: Specific and General’, in M.D. Sahlins and E.R. Service (eds), Evolution and Culture. Ann Arbor: University of Michigan Press, pp. 12–44. ______. 1965. ‘On the Sociology of Primitive Exchange’, in M. Banton (ed.), On the Relevance of Models in Social Anthropology. Oxford: Routledge, pp. 139–236. ______. 1968a. Tribesmen. Englewood Cliffs, NJ: Prentice Hall. ______. 1968b. ‘Notes on the Original Affluent Society’, in R.B. Lee and I. DeVore (eds), Man the Hunter. New York: Aldine Publishing Company, pp. 85–89. ______. 1972. Stone Age Economics. Chicago: Aldine Atherton Publishing Company. Sahlins, M.D., and E.R. Service (eds) 1960. Evolution and Culture. Ann Arbor: University of Michigan Press. Saltré, F. et al. 2016. Climate Change Not to Blame for Late Quaternary Megafauna Extinctions in Australia’ Nature Communications 7:10511. Retrieved 4 February 2019 from https://www.nature.com/articles/ncomms 10511. Schneider, D.M. 1961. ‘Introduction: The Distinctive Features of Matrilineal Descent Groups’, in D.M. Schneider and K. Gough (eds), Matrilineal Kinship. Berkeley: University of California Press, pp. 1–33. ______. 1965a. ‘American Kin Terms and Terms for Kinsmen: A Critique of Goodenough’s Componential Analysis of Yankee Kinship Terminology’, Formal Semantic Analysis: Special Issue of American Anthropologist 67(5), part 2: 288–308. ______. 1965b. ‘Some Muddles in the Models: Or, How the System Really Works’, in M. Banton (ed.), The Relevance of Models for Social Anthropology. London: Tavistock, pp. 25–85.

218 • References

______. 1968. American Kinship. A Cultural Account. Englewood Cliffs, NJ: Prentice Hall. ______. 1972. ‘What is Kinship All about?’, in P. Reining (ed.), Kinship Studies in the Morgan Centennial Year. Washington, DC: Anthropological Society of Washington, pp. 32–63. ______. 1984. A Critique of the Study of Kinship. Ann Arbor: University of Michigan Press. Schneider, D.M., and G.C. Homans 1955. Marriage, Authority, and Final Causes: A Study of Unilateral Cross-cousin Marriage. Glencoe, IL: The Free Press. Schulte-Droesch, L. 2016. ‘Making Place through Ritual: Land, Environment and Religion among the Santal of Central India’, Ph.D. dissertation. Groningen: University of Groningen. Service, E. 1962. Primitive Social Organization: An Evolutionary Perspective. New York: Random House. Shah, A. 2013. ‘The Tensions over Liberal Citizenship in a Marxist Revolutionary Situation: The Maoists in India’, Critique of Anthropology 33(1): 91–109. ______. 2018. Night March: Among India’s Revolutionary Guerillas. London: Hurst. Singh, S. 1966. Dust Storm and Hanging Mist: Story of Birsa Munda and His Movement. Calcutta: Firma K.L. Mukhopadhyay. Skoda, U. 2005. The Aghria: A Peasant Caste on a Tribal Frontier. New Delhi: Manohar. Slack, M., et al. 2017. ‘Post-Last Glacial Maximum Settlement of the West Angelas Region in the Inland Hamersley Plateau, Western Australia’, Australian Archeology 83(3). doi: https://doi.org/10.1080/03122417.2017.1404 548. Spier, L. 1925. The Distribution of Kinship Systems in North America. University of Washington Publications in Anthropology. Seattle: University of Washington Press, pp. 69–88. Strümpell, C. 2006. ‘Wir arbeiten zusammen, Wir essen zusammen’: Konvivium und soziale Peripherie in einer indischen Werkssiedlung. Berlin: Lit. Sundar, N. 2016a. ‘Adivasi vs. Vanavasi: The Politics of Conversion in Central India’, in N. Sundar (ed.), The Scheduled Tribes and Their India: Politics, Identities, Policies, and Work. Delhi: Oxford University Press, pp. 134–67. ______. (ed.). 2016b. The Scheduled Tribes and Their India: Politics, Identities, Policies, and Work. Delhi: Oxford University Press. Suryanarayan, M. 2009. ‘The Segmentary Lineage System among the Saoras of Andhra Pradesh’, in G. Pfeffer and D.K. Behera (eds), Structure and Exchange in Tribal India and Beyond: Contemporary Society: Tribal Studies, vol. 8. New Delhi: Concept, pp. 185–203. Tambiah, S.J. 1973. ‘Dowry and Bridewealth and the Property Rights of Women in South Asia’, in J. Goody and S.J. Tambiah (eds), Bridewealth and Dowry. Cambridge: Cambridge University Press, pp. 59–169. ______. 1989. ‘Bridewealth and Dowry Revisited: The Position of Women in Sub-Saharan Africa and North India’, Current Anthropology 30(4): 413–35. Tiemann, G. 1970. ‘The Four-Got-Rule among the Jat of Haryana in Northern India’, Anthropos 65(1–2): 166–77.

References • 219

Tjon Sie Fat, F.E. 1998. ‘On the Formal Analysis of “Dravidian”, “Iroquois”, and “Generational” Varieties as Nearly Associative Combinations’, in M. Godelier, T.R. Trautmann and F.E. Tjon Sie Fat (eds), Transformations of Kinship. Washington, DC: Smithsonian Institution Press, pp. 59–93. Trautmann, T.R. 1981. Dravidian Kinship. Cambridge: Cambridge University Press, ______. 1987. Lewis Henry Morgan and the Invention of Kinship. Berkeley: University of California Press. ______. 2012. ‘Crossness and Crow-Omaha’, in T.R. Trautmann and P.M. Whiteley (eds), Crow-Omaha: New Light on a Classic Problem of Kinship Analysis. Tucson: University of Arizona Press, pp. 31–50. Trautmann, T.R., and R.H. Barnes 1998. ‘“Dravidian”, “Iroquois”, and “CrowOmaha” in North American Perspective’, in M. Godelier, T.R. Trautmann and F.E. Tjon Sie Fat (eds), Transformations of Kinship. Washington, DC: Smithsonian Institution Press, pp. 27–58. Trautmann, T.R., and P.M. Whiteley (eds). 2012. Crow-Omaha: New Light on a Classic Problem of Kinship Analysis. Tucson: University of Arizona Press. Uberoi, P. (ed.) 1994. Family, Kinship and Marriage in India. Delhi: Oxford University Press. Vitebsky, P. 1993. Dialogues with the Dead: The Discussion of Mortality among the Sora of Eastern India. Cambridge: Cambridge University Press. Viveiros de Castro, E. 2001. ‘GUT Feelings about Amazonia: Potential Affinity and the Construction of Sociality’, in L.M. Rival and N.L. Whitehead (eds), Beyond the Visible and the Material: The Amerindianization of Society in the Work of Peter Rivière. Oxford: Oxford University Press, pp. 19–44. Weber, M. 1956. Wirtschaft und Gesellschaft. Tübingen: J.C.B. Mohr. Whiteley, P.M. 2004. ‘Why Anthropology Needs More History’, Journal of Anthropological Research 60(4): 487–514. ______. 2012. ‘Crow-Omaha Kinship in North America: A Puebloan Perspective’, in T.R. Trautmann and P.M. Whiteley (eds), Crow-Omaha: New Light on a Classic Problem of Kinship Analysis. Tucson: University of Arizona Press, pp. 83–108. ______. 2016. ‘Why Do Crow/Omaha Kinship Systems Exist?’, unpublished manuscript of the Kinship, Cognition and Practice workshop. Halle/Saale: Max Planck-Institute for Social Anthropology, 14–16 September. Williams, A. 2013. ‘A New Population Curve for Prehistoric Australia’, Proceedings of the Royal Society B 280; 2013.0486. Wolf, E.R. 1957. ‘Closed Corporate Peasant Communities in Mesoamerica and Central Java’, Southwestern Journal of Anthropology 13(1): 1–14. Woodburn, J. 1982. ‘Egalitarian Societies’, Man (N.S.) 17(3): 431–51. ______. 1988. ‘African Hunter-Gatherer Social Organization: Is it Best Understood as a Product of Encapsulation?’, in T. Ingold, D. Riches and J. Woodburn (eds), Hunters and Gatherers: vol. 1. History, Evolution and Social Change. Oxford: Berg, pp. 31–64. Woodburn, J., and A. Barnard 1988. ‘Introduction’, in T. Ingold, D. Riches and J. Woodburn (eds), Hunters and Gatherers: Vol. 2. Property, Power and Ideology. Oxford: Berg, pp. 4–32.

Index

A abduction, 160, 173, 175–176, 178, 208 Abrahamic religions, 117, 195–196 affinal imperative, 205 affinal order, 4, 201, 203 affinal villages, 151, 174–175 affinity, 4, 9, 12, 20, 31, 35, 39, 40, 43, 45–46, 49, 52, 55, 61–62, 82, 93, 95, 98, 103, 115–116, 119, 133, 135, 153, 172, 203, 205 alcoholism, 187, 200 Allen, Nicklas J., 31, 35, 87, 90, 210 Amazonia, 24, 40, 43, 70, 219 anarchic society, 1, 2, 6, 28, 29, 47, 49, 60, 68–70, 79, 81, 86, 124, 127–128, 136–137, 142, 158, 161, 164–165, 168, 174, 182, 201, 206, 207 Ancient Society, 2, 6, 12, 44, 118, 120, 135–136, 158, 215 Andhra Pradesh, 142, 146, 149, 151, 153, 156, 163, 187, 194, 213, 218 Appell, George N., 108, 210 arbitrary generalizations, 29, 30–32, 38, 42, 52, 69, 117, 202–203 artificial systems, 16, 17 ashram education, 190–191, 200 assemblage of terms, 85, 205 atom of kinship, 179–180, 182 Australian aborigines, 6, 117, 127, 130, 132–133 authoritarian environment, 125, 182, 187

B Baiga, 96, 156, 163, 212 Bailey, Frederick G., 139–140, 149, 210 Banks, David J., 5, 110–113, 116, 210 Barnard, Alan, 4, 40, 130, 210 Barnes, Robert H., 3–4, 47–48, 60, 63–67, 71, 100–101, 210, 219 Barraud, Cecile, 102, 210 Bastar, 156–157 Behura, N.K., 152, 163, 210 Berger, Peter, 8, 140, 154–155, 161, 185, 210 Bhattacharya, S, 87, 211 Bhuiyan, 143, 145, 159, 215, 217 Bhumia, 153–4, 163 Bhumij, 95, 142–143, 159 Bicknell, Peter J., 199, 211 bifurcate collateral, 37 bifurcate merging, 37 Binjhal, 151 Blackstone, William, 41–42, 48, 211 Blunt, E.A.H, 101, 211 Boas, Franz, 18, 120 Boda Porja, 153–154 Bodding, P.0., 162, 211 Bondo, 95, 97, 140, 153–155, 190, 197, 198, 216 Bouez, Serge, 83, 87, 211 Bowler, Jim, 131, 211 bridewealth, 7, 129, 141, 164, 166–168, 171, 176–179, 182, 184, 207–208, 213, 218 bronze statues, 187 Buradkar, M.P., 86, 156, 211

Index • 221

C Carsten, Janet, 5, 12, 107–109, 115–116, 211 caste, 7, 27, 98, 101, 137–139, 141, 161, 163, 175, 177, 180–182, 184, 187, 197–199, 210–211, 215, 218 caste India, 175, 181 caste society, 137, 141, 180, 187 Cayuga, 13 Cheyenne, 68, 91, 212, 213 Chhattisgarh, 142, 156 Childe, V. Gordon, 122, 211 Chota Nagpur, 141, 146–7, 149, 158, 169, 183–5, 216 circulating fund, 7, 166–167, 177, 182, 208 civitas, 2, 119, 136, 168 clan, 6, 27–29, 43–44, 46, 49–50, 60, 64–65, 68, 79, 82, 98, 102, 109–110, 122, 124, 134–136, 142–150, 152–159, 162–163, 174–175, 178, 184–185, 201, 204, 206–208, 211–212, 214 classificatory reading, 41, 87–88 classificatory terminology, 3, 18, 20, 52, 68, 106, 114, 203, 205 Collier, Jane Fishburne, 176–177, 212 commercial agents, 96, 139–140, 152, 160, 177–179, 182, 195, 207 concrete relationships, 22, 31–32, 41 conjugal fund, 7, 166–167 Crow, 23, 37, 50, 63, 70, 83–84, 210, 214, 219 D Darwin, Charles, 15, 118, 211 Das, S., 200, 211 Deeney, James, 8, 88, 95, 101, 147, 211 Demmer, Ulrich, 162, 211 descriptive terminology, 3, 17, 19–23, 25, 29, 32, 38, 41, 52, 63, 99, 106, 109, 111, 116, 141, 202, 203, 205, 206 desia (Indigenous), 27, 30, 96, 141, 149–151, 153–154, 185, 210

Didayi, 95, 153–154 Dom, 153–154, 177 domestic institution, 20, 22, 25–6, 43, 116, 203 dominant clan, 130, 150, 154–155 dominant discourse, 17, 120 Dora, 153–154 Dorsey, James O., 65, 211, 214 dowry, 6, 7, 126, 129, 166–167, 169, 184, 213, 218 Dravidian, 5, 16, 19, 22–26, 42, 46, 55–56, 59, 61, 70, 84–85, 91, 94–96, 100–101, 113–115, 146, 149, 153, 185, 201, 205, 210, 212, 218–219 Dumont, Louis, 23, 26, 49, 54–55, 70, 82, 85–86, 91, 113, 117, 203, 212 Durkheim, Emíle, 64, 104, 126, 188, 200, 208, 212 E Earth Goddess, 142, 150, 161, 167, 169, 183, 185, 190, 195–196 earth men, 154 eclecticism, 19 Eggan, Fred, 60, 68, 212 elopement, 28, 173–174, 178 Elwin, Verrier, 101, 139, 152, 155, 161–162, 185, 212 Engels, Friedrich, 9, 120, 212 Erse, 41, 51–52, 111 Eskimo, 36–37 Evans-Pritchard, Edward E., 44, 98, 212 evolution, sociocultural, 2, 5–6, 8–9, 11, 17–20, 22, 24–26, 35, 118–122, 126, 133, 135, 167, 206, 211, 213, 216, 218–219 F Feinberg, Richard, 103, 212 fertility rite, 170 Fletcher, Alice C., 65, 212 Fogelson, Raymond D., 104–105, 212 formal analysis, 4, 5, 8, 11, 19, 20, 38, 104–105, 218 Fortes, Meyer, 9, 212

222 • Index

Freeman, J. Derek, 108, 110, 212 Fried, Morten H., 121–122, 124, 213 Fuchs, Stephen, 96, 156–157, 213 Fürer-Haimendorf von, Christoph and Elizabeth, 95, 101, 139, 157, 213 G Gadaba, 95, 101, 153–155, 165–166, 168, 180–182, 184–185, 210, 216 gated community, 189 Gausdal, Jan, 142, 213 Gell, Simran Man Singh, 156, 185, 213 genealogical links, 2, 32 generation sets, 144 gentile organization, 2, 16, 119, 135–136, 141, 148, 158–160 German trusts, 189 Godelier, Maurice, 213 Gond, 61, 90, 96, 102, 146, 151, 155–159, 163, 211, 213 Good, Anthony, 4, 40, 210 Goodenough, Ward H., 204, 213, 214, 217 Goody, Jack, 7, 166–167, 177, 180, 203, 218 Gregory, Chris, 96, 163, 213 Grignard, A., 102, 213 Grigson, V. Wilfred, 156–7, 213 Gutob, Senior Gadaba, 154 Guyanas, 40 Guzy, Lidia, 161, 185, 213, 217 H Halbi, 96, 102, 155–156 Hardenberg, Roland, 7, 24, 94, 101, 150–151, 166, 174, 184–185, 208, 210, 213 Hawaiian, 3, 36, 104, 109 Hiatt, Lester H., 117, 124, 213 Highland Middle India, 3, 5, 24, 85–87, 124, 136, 164, 168, 185, 195, 198, 201, 203 Hirzel, A., 161, 213 Hivale, Shamrao, 157, 213 Ho, 7, 97, 146–148, 151–155, 159, 161, 211 Hoebel, E. Adamson, 68, 213

Holi, 197 Homans, George C., 104, 213 Hopi, 23 hospitals, 189, 192–194 Hugh-Jones, Stephen, 12, 211 hunter-gatherers, 6, 120–121, 123, 127–128, 130, 206 I imperative to vary, 204 indentured labour, 186 individual, 2, 4, 6, 10, 19, 21–25, 27–32, 41, 43, 45, 49, 53, 57, 60, 63, 68, 75, 79, 86, 105, 107, 109–110, 116, 119, 122–129, 131–132, 144–145, 148, 151, 155–157, 162, 164–166, 168, 175, 180, 183, 185, 187, 194, 202, 206, 209, 217 Indo-European, 5, 18, 20, 22, 24, 96, 101, 117, 153, 185, 207 Iowa, 62, 82 Iroquois, 2–3, 11–14, 20, 23, 26, 29, 36–37, 42–43, 46, 53–54, 61, 84–85, 91–92, 97, 102, 118–119, 135–136, 174, 186, 210, 215, 219 J Jackson, Andrew, 13 Jena, D., 161, 214 Jharkhand, 5, 6, 86, 89, 96, 101, 142, 146, 162–163, 189 Joria, 153, 207 Juang, 96–97, 143–145, 159, 162, 215 K Kharia, 5, 87–90, 93–94, 96–97, 101, 146–147, 149, 201, 216 Kharia, Dhelki, 146–147 Kharia, Dudh, 146–147 Kheria, Hill, 101, 146 kin-type, 39, 46–47 kinship, 1–2, 4–5, 9–10, 16, 19, 20–21, 29, 30–31, 39–40, 42, 48, 50, 57, 62–63, 83, 87, 96, 100–10, 103–109, 117, 129–130, 141, 176, 205, 210–219

Index • 223

‘Kinship’, 1–3, 9, 21, 23, 25–26, 29–32, 35, 43, 45, 50, 63, 103–107, 133, 207 kinship terminology, 4, 20, 38, 101, 106, 212, 217 Kirchhoff, Paul, 47, 122, 134, 213, 214 Kisan, 146, 149, 162, 189 Kobayashi, M, 102, 214 Koitor, 156 Kond, 24, 27, 95, 102, 139–141, 149–152, 157, 159, 161–162, 166, 168–175, 177–185, 187–188, 191–192, 208, 215 Kond, Desia, 148, 150–151, 154 Kond, Dongria, 7, 95, 101, 149–151, 166, 174–175, 177, 185–186 Kond, Kuttia, 7, 102, 139, 149, 162, 168–171, 174–175, 177, 180–182, 185, 191–192, 215 Kond, Kuvi, 149 Kond, Mala, 149 Kond, Pengo, 149 Kond, Porja, 149 Kond, Sita, 149 Koraput, 152–155, 159, 165–166, 169, 181, 183, 185–187, 189, 192–193, 197, 207, 210, 216 Koya, 156–157, 162 Kroeber, Alfred L., 3, 18, 19, 106 Kronenfeld, David B., 65, 214 Kuper, Adam, 103, 134, 214 L La Flesche, Francis, 65, 212 land settlement, 165 Layton, Robert, 130, 214 Leach, Edmund R., 4, 208, 214 leadership, 123, 128–129, 134, 152–153, 198–199 legal-bureaucratic mode, 1 Lévi-Strauss, Claude, 9, 24, 44, 50–52, 61, 83, 99–100, 104–105, 180, 203, 205, 211, 214, 215 lineal, 37 local descent groups (LDG), 6, 49, 142, 144, 147, 151–152, 155, 159–161, 167–168, 178

Lounsbury, Floyd G., 4, 19–21, 24, 39–50, 59–60, 84, 214 Lowie, Robert H., 37, 214 Lubbock, John, 26, 118 Lyell, Charles, 15, 214 M Mahapatra, Lakshman K., 145, 215 Maine, Henry Sumner, 118 Malay terminology, 110–114 Mali, 153–154 marriage, 4–7, 11, 18, 23–27, 30–31, 34–35, 41–42, 44–45, 47–51, 59–61, 81–84, 86, 90, 94, 98–99, 102, 104–105, 109, 112, 116–120, 124–125, 127, 129, 134, 143, 145, 147–155, 158–160, 164, 166–168, 172–176, 178–179, 182, 184–185, 188, 201, 203– 205, 207–208, 210–213, 215–219 marriage payments, 164, 166–167 marriage rules, 7, 24, 27, 35, 81, 83, 86, 1145, 149 marriageability, 4, 31 Marx, Karl, 9, 120, 135 Matia, 153–154 matrix, 595, 32, 34, 53, 71, 73, 76, 87–88, 90–1 Mauss, Marcel, 104, 203, 215 McDougal, Charles W., 96–97, 143–145, 215 McLennan, John F., 118 Meggitt, Mervyn J., 132, 215 Melanesia, 24 missionaries, 87, 153, 162, 195 Moffat, Michael, 187, 215 Mohawk, 13 moiety, 6, 46, 49, 60, 65, 68, 81, 154 Munda, 5, 24, 61, 87, 98, 101, 137, 145–147, 153, 159, 166, 189, 201, 211, 215–218 Murdock, George P., 9, 37, 215 N Naik, T.B., 95, 215 National Aluminium Company, 189

224 • Index

Native North American, 2–3, 8, 23, 99, 119, 201, 205 natural systems, 16, 17 Needham, Rodney, 35, 37, 61–62, 64, 70, 99–100, 105, 203–204, 215 New Kinship, 5, 107–108, 116 Niggemeyer, Hermann, 139–140, 161, 191, 215 O Odisha, 5, 6, 86, 101, 142, 149, 151, 153, 164, 166, 190, 196 Ogden Land Company, 13 Ollari, junior Gadaba, 154 Omaha, 3–5, 23, 35–37, 47, 50–51, 62–85, 99, 100, 110, 113–116, 133, 201– 203, 205, 210–219 Oneida, 13 Onondaga, 13 Oraon, 98, 146 Other Backward Classes, 27, 101, 138 Otten, Tina, 161, 163, 185, 215 Ottenheimer, Martin, 103, 212 Overing Kaplan, Joanna, 32, 38, 40, 215 P Padel, Felix, 200, 211 Pan, 153, 177–178, 185 paper mill, 188–189 Parenga, 95, 153–154 Parker, Ely S, 13–14, 12 Parkin, Robert, 87, 215 parricide, 181 Parry, Jonathan, 98, 215 pastoral communities, 164 Patnaik, Nityananda, 203, 215 peasant society, 1–2, 121, 124–126, 128 peasantry, 6, 121, 138 person symbols, 31–2, 53, 56, 76 Pfeffer, Georg, 1, 8, 101, 140, 147, 155, 160, 161, 162, 168, 185, 208, 210–212, 214, 216, 218 Pfeiffer, Martin, 102, 216 phratry, 6, 60, 136, 151, 155–156, 166–168, 201 Piaroa, 32–36, 38, 70

plans, 2, 93, 203 Polanyi, Karl, 134, 216 preference, 7, 149, 155, 211 prescription, 33, 35, 51, 76, 81, 93, 99, 117, 119 prescriptive terminologies, 24, 51, 75 prestige, 168, 175, 178 primary relationships, 19 Primitive Tribal Groups, 137, 181 private property, 7, 120, 164–167, 183, 206 property, 2, 7, 119–120, 123–125, 128–129, 164–1168, 175, 183, 189, 203, 206 proscriptive terminologies, 24, 51, 75, 100, 205 public order, 2, 6, 116 R Radcliffe-Brown, Alfred R., 22, 106, 117, 121, 216 Reichel, Eva, 9–97, 101, 137–138, 148–149, 161–162, 216 Reining, Patricia, 9, 216, 218 relatedness, 103, 107–108, 111, 116, 211 relationship terminologies, 2–4, 8, 16, 81–82, 107–108, 116, 132, 201 removal, 3, 13, 14 repetition of intermarriage, 4, 23, 204 Resek, Carl, 9, 12, 26, 118, 216 revolutionaries, 13, 194–195 Riches, David, 130, 210, 217, 219 Rivière, Peter, 26, 40, 217, 219 Rona, 153–154, 163, 215 Rosman, Abraham, 127, 217 Roy, Sarat Chandra, 101, 125, 139, 143, 145–147, 162, 217 Rubel, Paula, 127, 217 Rycroft, David J., 161, 217 S Sabean, David W., 26, 110, 217 sacrificers, 92, 151, 155 sacrifiers, 92, 183 Sahlins, Marshall D., 6, 25, 86–87, 121–123, 133–134, 207, 217

Index • 225

Salish, 37 Saltré, F., 131, 217 same-sex-sibling merging rule, 19, 20 Santal, 27, 87, 142–143, 148, 158, 161–162, 164, 211, 213, 216, 218 Scheduled Castes, 27, 101, 138, 198 Scheduled Tribes, 27, 101, 137–138, 165, 186, 214, 218 Schulte-Droesch, Lea, 141–143, 161, 218 secondary relationships, 19 semi-complex structures, 61, 83–4, 90–100 Seneca, 3–4, 12–13, 22, 39, 41–55, 57, 59–62, 68–71, 73, 75–76, 78–79, 83–85, 94, 99, 114–115, 117, 133, 201, 204–205 seniority, 84, 86, 97–98, 134, 142, 146, 150, 206 Service, Elman, 121–122, 133–134, 218 Shah, Alpa, 200, 218 Singh, Suresh, 161, 218 skewing, 5, 57, 61–65, 70–71, 73–75, 78–79, 82–83, 101, 114–115, 209 Skoda, Uwe, 161, 218 Slack, M., 131, 218 social values, 7–8, 107 sociality, 113, 203 societas, 2, 119, 136, 168 Sora, 139, 141, 151–153, 163, 187, 219 South Indian systems, 24, 55, 60, 70, 84–85, 100, 117, 203, 205–6, 212 Spier, Leslie, 36–37, 218 state, 6–7, 11, 13, 25, 27–28, 86, 99, 102, 107, 120, 124, 126, 128–129, 134–136, 138–139, 141, 144, 147, 151, 153, 156–157, 162–166, 168, 174, 181, 183, 185, 191, 193–195, 199, 201, 206, 216 status, 2, 21, 27, 30, 39, 41, 43, 45, 46–47, 58, 73, 78, 81, 91–93, 96, 99, 101, 104, 108, 119–120, 124, 128, 135, 139–143, 146–148, 150–152, 155, 159–160, 162–163, 177, 180, 186–187, 195–196, 198, 206–208, 212, 214

Steward, Julian, 121 Strümpell, Christian, 161, 185, 218 subclan, 6, 65, 148 Sudanese, 36 Sundar, Nandini, 200, 214, 218 Suryanarayan, M., 138, 142, 152, 218 symmetric affinal exchanges, 5, 54 symmetric affinal relationships, 97, 201 Systems of Consanguinity and Affinity of the Human Family, 2–4, 9, 12, 39, 61, 93, 103, 118, 1135, 202, 215 T Tambiah, Stanley J., 166–167, 180, 213, 218 Tamil, 22, 42–43, 205 temples, 7, 37, 125–126, 129, 151, 196–197, 200 terminological line, 34, 36, 70–71, 73, 75, 79, 85, 88, 90, 100, 111, 114–115, 205, 206 terminology types, 3, 37, 38 territory, 2, 17, 144–145, 149–150, 174, 194 tetradic structure, 35, 94–95 the four brothers, 155 the people of one rice-pot (miyad mandi chaturenko), 148 the spirit of the gift, 51, 59, 64, 79, 164, 203 the thirteen affines, 154–5 the twelve brothers, 154 Tiemann, Günther, 101, 218 time revolution, 16 Tirkey, B., 102, 214 Tjon Sie Fat, Franklin E., 60, 84, 100–102, 210, 213, 218 total social fact, 7, 101, 150–1, 155, 184, 208 Trautmann, Thomas R., 3, 9, 10, 12, 1416, 19–22, 24–26, 39, 41, 43, 55, 60–61, 63, 65, 70, 83–85, 91, 101–102, 113, 119, 202, 210219 trend, 17–19, 22, 24, 120–1

226 • Index

‘Tribes’, 6, 27–29, 43–44, 46, 49, 86, 101–102, 123, 128, 137–138, 142, 145–146, 152, 159, 165, 186, 195, 199, 207, 210, 212–213, 215–216–218 Truk, 104, 109 Tuscarora, 13 U Uberoi, Patricia, 184, 219 V Vitebsky, Pierce, 152, 187, 219 Viveiros de Castro, Eduardo, 40, 43, 47, 219

W weavers, 138–141, 152, 155, 157, 160–161, 163, 177–179, 184, 207 Weber, Max, 1, 6, 219 West Bengal, 142, 194 White, Leslie A., 12, 121, 219 Whiteley, Peter, 10, 12, 14, 13, 26, 52, 61, 63, 70, 83, 91, 98, 210, 219 Williams, A., 131, 219 Wolf, Eric R., 121, 219 Woodburn, James, 122–123, 127, 130, 138, 210, 219 Y youth dormitories, 7, 174