UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Kewa reciprocity : cooperation and exchange in a New Guinea Highland culture LeRoy, John D. 1975

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


831-UBC_1975_A1 L47_6.pdf [ 22.94MB ]
JSON: 831-1.0093590.json
JSON-LD: 831-1.0093590-ld.json
RDF/XML (Pretty): 831-1.0093590-rdf.xml
RDF/JSON: 831-1.0093590-rdf.json
Turtle: 831-1.0093590-turtle.txt
N-Triples: 831-1.0093590-rdf-ntriples.txt
Original Record: 831-1.0093590-source.json
Full Text

Full Text

KEWA RECIPROCITY: COOPERATION AND EXCHANGE IN A NEW GUINEA HIGHLAND CULTURE by JOHN D. LEROY B.A., Oberlin College, 1966 M.A., University of Michigan, 1968 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in the Department of Anthropology and Sociology We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA Ap r i l , 1975 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission fo r extensive copying of t h i s thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It i s understood that copying or publication of t h i s thesis f o r f i n a n c i a l gain s h a l l not be allowed without my written permission. Department The University of B r i t i s h Columbia Vancouver 8, Canada Date i i ABSTRACT KEWA RECIPROCITY: COOPERATION AND EXCHANGE IN A NEW GUINEA HIGHLAND CULTURE Melanesian c u l t u r e s c h a r a c t e r i s t i c a l l y assign great importance to the transaction of objects, and many have h i g h l y complex systems of exchange; yet few anthropologists have sought to define s o c i a l structure i n terms of r e c i p r o c i t y . This thesis does so: i t i s a study of the pr a c t i c e and conceptualization of r e c i p r o c i t y among the Kewa of the South-ern Highlands D i s t r i c t of Papua., New Guinea. Two modes of r e c i p r o c i t y — cooperation and exchange — are seen as p r i n c i p l e s f u n c t i o n a l l y equiva-l e n t but methodologically p r i o r to descent and a f f i n i t y . The importance of r e c i p r o c i t y can be measured by the range of c u l t u r a l materials i t explains. I i n v e s t i g a t e four sets of data. Part I examines the Kewa moral order: the system of ideas through which r e c i -procal r e l a t i o n s are maintained or, i f broken, are r e i n s t a t e d . Through these ideas, i n p a r t i c u l a r the concepts "thought" and "ghost," non-r e c i p r o c a l persons are brought to an awareness of t h e i r o b l i g a t i o n s . The c e n t r a l importance of mediative objects i s manifested i n the i n s t i -t u tions of compensation and competitive r e c i p r o c i t y . Part II i n v e s t i -gates the organization of r e c i p r o c a l r e l a t i o n s i n kinship. Two chapters on r e l a t i o n between structure and praxis (action) bracket a d e t a i l e d study i n consanguinity, a f f i n i t y , and marriage preference. i i i Part III consists of a study of institutionalized reciprocity — the south Kewa pig k i l l . Three chapters examine exchange and cooper-ation in the transactions of shells and pork, in verbal opinions about the conduct of the ceremony, and in metaphorical songs about pig k i l l i n g . Part IV i s an analysis of fifteen Kewa myths of different armatures: male siblings, father and son, cross-sex siblings, and spouses. The interpretation, which attends to both form and content, brings out two aspects of myth: (1) myth reflects and defines Kewa moral and struc-tural relations; (2) myth i s a form of di a l e c t i c a l reasoning which en-deavours to understand the cultural t o t a l i t y in terms of i t s parts: f i l i a t i o n , siblingship, a f f i n i t y , cooperation and exchange. A number of theoretical approaches are debated and applied: B r i t i s h social anthropology, French structuralism, and phenomenological philosophy. Throughout I have adopted a method which attempts to be both structural and d i a l e c t i c a l ; structural because i t sees cultural def-inition in terms of oppositions, and dial e c t i c a l because i t sees these oppositions as experiential as well as lo g i c a l , contradictory as well as contrastive. Within this d i a l e c t i c a l perspective each of the parts of the thesis can be considered a particular aspect of reciprocity, each a mediation or determination of the general definition of reciprocity found in the introductory chapter. The four parts move the understanding of reciprocity from the general and the abstract to the particular and the mediated. The movement is from the conceptual system underlying practice (Part I) to the dyadic or triadic relations of structures (Part iv II), to the institutional setting (Part III), and f i n a l l y to a discourse in which these positive r e a l i t i e s are contrasted with putative ones (Part IV). A concluding chapter examines one singular instance of how Kewa ideas about reciprocity have modified themselves as a result of the European presence. V TABLE OF CONTENTS Chapter 1. The Problem and the People 1 Part I: Morality Chapter 2. Aspects of Being 23 Chapter 3. The Moral Context 44 Chapter 4. The Tawa 77 Part II: Social Structure Chapter 5. Structure and Praxis I 108 Chapter 6. Cooperation and Exchange in Consanguinity 124 Chapter 7. Cooperation and Exchange in A f f i n i t y 147 Chapter 8. Marriage Preference and Structure 166 Chapter 9. Structure and Praxis II 184 Part III: The Pig K i l l Chapter 10. Events 196 Chapter 11. Opinions 225 Chapter 12. Metaphors 243 Part IV: Narratives Chapter 13. Youth and Ribuali 275 Chapter 14. Two Brothers 320 Chapter 15. Father and Son 338 Chapter 16. Brother and Sister 361 Chapter 17. Brother and Wildman, Sister and Sky-Maiden 383 Chapter 18. Woman and Boy 418 v i Table of Contents (Continued) Chapter 19. Epilogue 438 Bibliography 467 Appendices Appendix 1. Dialectic and Scarcity 474 Appendix 2. The Repa 479 Appendix 3. The Boundaries of Classificatory Kinship 494 Appendix 4. Marriage Sample 499 Appendix 5. Counting the Shells 506 Appendix 6. Supporting Narratives 516 Appendix 7. Fieldwork Relations 531 Appendix 8. A Note on Sorcery 535 v i i LIST OF TABLES Table I Principal Cultivated Crops II Time Use III Contributors to a Tawa Faction IV Transactions of a Tawa Supporter V Kewa Kinship Classification Rules VI Kewa Kin Terms VII A Marriage Payment VIII Result Categories for Comparison of Marriage Pairs IX Frequency of Sibling Exchange X Pig K i l l Transactions XI Koiari 1971 Pig K i l l : Chronology X I I Rupale Content XIII Kinship Armatures in Narratives XIV Kinship Ties Between 56 Men of One D i s t r i c t XV Marriage Survey Results XVI Pearlshell Transactions XVII Affinal Exchanges XVIII Individual Variation Among Pig K i l l e r s XIX Performance of Two Pig K i l l e r s 15 21 93 93 130 132 150 171 172 204 218 259 288 495 500 507 509 512 513 v i i i LIST OF FIGURES Figure 1-1 The Concepts Reciprocity and Morality 7 1-2 Map 18 3-1 Participants in a Dispute 56 3-2 Compensation 66 3- 3 The Moral System 67 4- 1 Participants in a Tawa 82 4-2 The Tawa Sequence 87 6-1 Consanguineal Relations 127 6-2 Cooperation and Exchange 133 6-3 The Atom of Kinship: the Normative Version 134 6-4 The Atom of Kinship: the Minority Choice 136 6-5 Overriding the Cross Relationship 138 6-6 Case 1 139 6-7 Case 2 140 6- 8 Generation Conversion of Cross Kin into Parallel Kin 143 7- 1 The Reciprocals Pase, Suba, and Yake 148 7-2 Death Payment for a Wife 157 7-3 Exchange Cycles in the Main Marriage Payment 161 7-4 Exchange Cycles in the Return and WMB Payments 162 7- 5 Pigs and Uterine Ties 163 8- 1 Distribution of 112 Marriages Made by Yetipa Repa 169 8-2 Possible Types of Exchange Marriage 171 8-3 Marriage With True and Classificatory Sisters 174 8-4 The Reciprocal Kaleke 176 8- 5 Marriage of Classificatory and True Opposite-Sex Siblings 179 9- 1 Sharing at the Repa and Sub-repa Levels 188 9-2 Charter of a Ruru 190 ix L i s t of Figures (Continued) v Figure 9-3 Charter of a Ruru-cluster 192 10-1 Movement of She l l s and Pork 206 10-2 The Pig K i l l Arena 210 10-3 "Following the P e a r l s h e l l Feet" 214 12-1 The Men's House (from above) 245 12-2 Gardening and Pig K i l l i n g 267 12-3 Male and Female 271 13-1 Two Oppositions 305 13-2 Metaphorical Kin Relations 311 13-3 Metaphorical Kin Relations 318 14-1 Transformation of Armature 330 .14-2 Tida 1, 2, 5 and Part of 4 Compared 335 15-1 Si b l i n g s h i p and F i l i a t i o n 338 16-1 Elements i n Tida 361 16-2 Transformation c f Armature 363 16-3 Kin Relations i n Tida 9 378 16-4 Tida 5 and 9 Compared 380 17-1 Tida 9 and 10 Compared 401 17-2 Armature of Tida 10 and 11 Compared 407 17-3 Two Expressions of Incest 415 18-1 Transformation of Armature 420 18-2 Tida 14 and 15 Compared 436 19-1 S p a t i a l Coordinates 441 19-2 Tree and Karst Hole 442 19-3 So c i a l Coordinates 443 19-4 The Encirclement of I a p i 460 A-1 Repa Organization of a Hamlet and a V i l l a g e 481 A-2 Map of Iapi D i s t r i c t 485 A-3 Ruru Organization of a Grassland D i s t r i c t 486 X L i s t of Figures (Continued) Figure A-4 Ruru Organization, Respecting Relative Sizes 488 A-5 Inter-district Organization 489 A-7 Repeated Marriage and Affinal Usage 497 A-8 Wife-giving, Wife-taking and Af f i n a l Usage 498 A-9 Genealogy 1 501 A-10 Genealogy 2 502 A - l l Genealogy 3 503 A-12 Genealogy 4 504 x i ACKNOWLEDGEMENT The fieldwork on which t h i s t h e s i s i s based was c a r r i e d out during 1970-72 and was supported by doctoral fellowships from The Canada Council. I thank t h i s i n s t i t u t i o n f o r i t s support. My o r i g i n a l proposal focused on production and exchange r e l a t i o n s and t h e i r bearing on the conceptualization of s o c i a l organization i n Papua, New Guinea. Since much of the data on t h i s problem, and even the formulation of the problem i t s e l f , bear on Highlands populations, the Highlands seemed the l o g i c a l place for fieldwork. My choice of the Southern Highlands D i s t r i c t was prompted by the r e l a t i v e l a c k of research c a r r i e d out there. The neighbouring populations of the Mendi, the Wiru, and the F o i had been studied or were being studied at the time. The Kewa were r e c e i v -ing continuing i n v e s t i g a t i o n by two l i n g u i s t s , K a r l and Joyce F r a n k l i n . But with the exception of one a r t i c l e on s o c i a l organization (Franklin 1965) no intensive studies had been made of Kewa society and c u l t u r e . The F r a n k l i n s ' work proved valuable to my research, and I express my indebtedness to them. For a people who c l a s s i f y whitemen as e i t h e r government, mis-sion, or business, the anthropologist i s at f i r s t d i f f i c u l t to place. A l l the more so since he, unlike the others, makes demands which could e a s i l y be thought unreasonable: he wishes to l i v e among them, he s o l i -c i t s t h e i r help, and he asks for t h e i r knowledge. (I give an account of my fieldwork in Appendix 7.) My main indebtedness, undischarged by x i i whatever I managed to give in return, i s therefore to the Kewa them-selves. In particular I want to thank Uda, Yawi, Bilisapo, Malawe, and others of Yakopaita; Kabe, Wialiwada, Mabo, the memory of Yekipu, and others of Koiari; and Tema, Parea, Pobarame, Pusa, Robo (Walameara), Robo (Taara), and others of Iapi. A number of other persons have been helpful at various stages of the research. Drs. Ray Kelly and Anthony Forge counseled me on f i e l d -work prospects in other parts of Papua, New Guinea. Members of the anthropology departments at the Australian National University and at the University of Sydney freely offered their advice. For their aid in expediting my application for research and for their assistance in Port Moresby, I thank Dr. Marion Ward and the New Guinea Research Unit; for similar aid and for their generous hospitality, my thanks go to Drs. Ralph and Susan Bulmer, then of the University of Papua, New Guinea. I also express my gratitude to the Administration of Papua, New Guinea for allowing me to carry out the research. For their material aid and kindness, my thanks in particular to Mr. Simon Pearson, Officer-in-Charge at Erave during my research, and to Mr. Rick Creigh, then Assistant D i s t r i c t Commissioner at Kagua. A number of people assisted me similarly at Erave: Fr. Dunstan Jones and Fr. Maris Goetz of the O.F.M. Capuchin mission, Mr. and Mrs. Robert Campbell, and Mr. Fred Lipscomb. Finally, my special thanks to the members of the Anthropology and Sociology Department of the University of British Columbia who commented on my work from i t s inception to completion, Drs. C y r i l Belshaw, Kennelm Burridge, and E l l i Maranda. x i i i GLOSSARY OF KEWA TERMS This glossary includes terms mentioned more than once in the text. The letters S and W stand for South dialect and West dialect variants, when pertinent. agapukupi W ainya S ama W ame araame awa ayako W b a l l ewa W (kewa S) kabereke W kaberekale S kalado W kama S (ama W) kani W (kai S) kepa W kone lo merepa neada S oma W (koma S) pase pu raba raguna S person with a smelly mouth (an insult) opposite sex sibling (address) cf. kama; big-man brother, male-speaking synonym for repa mother's brother, sister's child compensation payment opposite-sex sibling; a brother and a s i s -ter people l i v i n g to the south the act of contracting supporters (through prestations of pork and shells) for a pig k i l l ; yae kabereke: contract for work on yaeada; yada kabereke: contract for war-fare alliance wildman, demon ceremonial ground cross-cousin ceremonial digging stick thought, behaviour; kone rogo : guil intestines people l i v i n g to the north accommodation huts built for pig k i l l e r s and v i s i t o r s to emote, fe e l , die same-generation affine, principally WB or ZH l i v e r to help hat, worm by raguna a l i (a leader in ex-change) xiv ratu W anger, resentment, frustration re remo repa S (ruru W) r i b u a l i rob a rogoma rome rupale sekere suba W tapada tawa tida S ( i t i W) to (or ro) wasa yaeada W yano W (yago S) base, origin, cause; sponsor, owner, organi-zer; base of tree; remo re: ghost finder; yaeada re: manager of yaeada construction; tawa re: a principal in a tawa ghost patrician hermit, "rubbish man" stomach ghoul exchange of valuables, primarily with affines and matrikin metaphorical song mother-of-pearl shell; sekere uni, sekere  alu (bone pearlshell, head pearlshell): valuable shells cross-generation affine, principally WF or DH men's house competitive exchange narrative body soul accommodation huts built for pig k i l l e r s and v i s i t o r s replacement, duplicate; sol i c i t o r y g i f t , return g i f t yasa metaphorical song X V Note on Pronunciation a father,chat (or a s ? ) e say, the i easy, pis o flow, eau u loot, toute b prenasalized mb d prenasalized nd g prenasalized ng_ (in S) r r, in i n i t i a l position l i k e _tr_ 1 CHAPTER 1 THE PEOPLE AND THE PROBLEM In my encounter with the Kewa I repeatedly found myself asking a question: why are the Kewa so preoccupied with their possessions? Why do they constantly plan, make, or dispute exchanges of shells, pigs, and money? This question, or a similar one, arises whenever anthropologists 1 attempt to understand Melanesian cultures. More generally, we Wester-ners ask them of Melanesians as part of a contact and mutual interrogation between our culture and theirs. They form what the Kewa would c a l l the re — the base, source, and fundamental aspect — of this thesis. In theoretical terms my question i s , why is the mediative ob-ject so important in almost every cultural practice (whether i t be cere-monial exchange, compensation, or sacrifice) wherein one human subject tries to modify another? "Why" questions are legitimate, but anthropologists do not, as a rule, deal with them. Philosophers do, but they have l i t t l e or no familiarity either with primitive cultures or with anthropological methods. Hegel (1967) had the concept of a "Desire" and Sartre (1953) that of a "lack of being" in the For-itself, the human consciousness. Consciousness reaches out and attempts to appropriate otherness (objects or subjects), hoping thereby to acquire the plenitude and recognition achieved only by the possessor or the Master. According to at least one anthropological view (Levi-Strauss 1966: 249, 254) by grounding the argument in a philosophical system I 2 would be acting li k e the informant who explains some configuration of events in terms of a myth told by his ancestors. Rather than place myself in this position, I have chosen to deal not with the "why" but with the "how," with the arrangement or "form" of cultural phenomena. One reaches an under-standing of phenomena when one sees how they coexist within a cultural to-t a l i t y in an orderly relation to other phenomena. Unlike the philosopher who is concerned to account for the possibility or the necessity of what i s , the anthropologist just takes what exists and rearranges i t to be more understandable (Levi-Strauss 1966: 248). This thesis w i l l relate various aspects of a New Guinea High-lands culture, the Kewa, using the notions of reciprocity, morality, structure and praxis. 2 Since the term "praxis" has only recently appeared in Anglo-American anthropological discourse, and continues to have a rather p r i -vate meaning, I shall say a few words about i t s use in this thesis. Praxis opposes to "practices" as abstract to concrete and as universal to particular. The plural noun "practices" (as in "cultural practices" or "pi g - k i l l i n g practices") denotes those operations which the anthropologist encounters as the v i s i b l e or manifest face of structures; they constitute the overt culture. Praxis refers to the individual participation in or experience of practices; i t refers to the individual's action or project (whether undertaken singly or in a group) which — though always realized in a singular, mediated and pragmatic form — is potentially i n t e l l i g i b l e and communicable across cultures by reason of a shared humanity. Discov-3 ering praxis does not require a mystical participation in or intuition of other cultures; rather i t i s discovered because man is capable of see-ing the universal in the particular, by stripping away the concrete deter-minations within a culture and reaching a level at which he can under-stand the human object. Praxis differs from the "action" of the action theorists in stressing the constitutive aspect of the act over i t s rational (choice-making) aspect. Constitutive means here that the sensible world is synthetically organized by men through their praxis, and that the cul-tural practices into which men are born reflect their active experience and are continually recreated by experience. (It i s not a contradiction to maintain, along with Levi-Strauss, that praxis presupposes a language and a thought that are analytical and constituted.) As for "structure," I use this word to refer to any organized system of cultural elements, be i t at the level of practices ("empirical" structures) or of thought ("logical" structures). The f i r s t corresponds generally to structure as acted out and/or apprehended at the conscious lev-el by actors; the second corresponds generally to the structured r e a l i t y comprehended by the researcher. The second meaning is that of the struc-t u r a l i s t , for whom Structure refers to the determining (though invisible) r e l a -tions which account for v i s i b l e reality; i t is the accounting reason or explanatory support, the rule or principle which accounts for the vi s i b l e relations (Rossi 1974: 90). In my opinion this meaning of "structure" corresponds to but one stage or purpose in anthropological reasoning. I shall argue that "structure" 4 has a practical side as well; that i t "accounts for vis i b l e r e a l i t y " only because i t i s really encountered as a determinant of action; that "rule or principle" is the observer's way of talking about constraints which actors create for themselves in their praxis; and that structure is significant as an explanatory tool, whether in kinship or in myth, only i f i t i s accepted in both meanings. In Part II I shall argue that struc-ture should be understood both in the analytical meaning (where i t denotes a system of elements, synchronically related, which ideally can be re-duced to mathematical expression or can be mapped out in charts) and in a d i a l e c t i c a l meaning (where "free" praxis ossifies i t s e l f and takes on the appearance of being constituted by some rule or principle). Reciprocity is related to exchange, a concept which Malinowski gave a particularly Melanesian stamp. Exchange has been occupying an increasingly important place in the literature on New Guinea. Wagner (1967) put i t on a par with consanguinity in his discussion of Daribi social structure, and other fieldworkers have documented the very s i g n i -ficant place of exchange systems in Melanesian societies (Strathern 1971a; Young 1971). But aside from a few recent studies (among them Burridge 1969 and Schwimmer 1973), few have attempted to define social struc-ture and organization in terms of reciprocity. "Descent and residence" is the established problematic (cf. for example, A. Strathern 1972). I shall argue that descent and residence yield a part i a l understanding. The terms reciprocity and exchange have been used in different ways by anthropologists, sociologists, and philosophers. Perhaps I can 5 simply define how I intend to use these words. Reciprocity has the widest meaning; i t is a principle, or a "mental structure" which is "the most immediate form of integrating the opposition between the self and others (Levi-Strauss 1969: 84)." Reciprocity takes the form of "duality, a l -ternation, opposition and symmetry...[which] are not so much matters to be explained, as basic and immediate data of mental and social reality which should be the starting-point of any attempt at explanation (ibid.: 136)." Following accepted usages I define reciprocity as a state or  relationship and exchange and cooperation as acts or actions. The three words refer to different r e a l i t i e s . The one (reciprocity) has to do with an internal state or perspective, the others (exchange and cooperation) with a kind of performance or practice. The Shorter Oxford English Dic- tionary gives the following meanings (among others): Reciprocity: a state or relationship in which there is mutual action, influence, giving and taking, correspondence, etc., between two parties. (Onions 1959: 1672). Exchange: the action, or an act, of reciprocal giving and re-ceiving. (Ibid.: 647). Cooperation: the act of working in conjunction (with another person, or thing, Jto an end, or jln a work); joint operation. (Ibid.: 390). Exchange and cooperation are practices of reciprocity. They are ways of acting out a reciprocal attitude or perspective. But here is a problem: exchange and cooperation are forms of positive reciprocity, not of reciprocity-in-general. Reciprocity-in-general also includes struggle, competition, agonistic interaction. Hence one has to distinguish: 6 R e c i p r o c i t y - i n - g e n e r a l : a state or r e l a t i o n s h i p of mutual a c t i o n . P o s i t i v e r e c i p r o c i t y : an amicable r e l a t i o n , r e a l i z e d i n coopera-t i o n and exchange. Negative r e c i p r o c i t y : an antagonistic r e l a t i o n , r e a l i z e d i n 3 f i g h t i n g or competition. In other words the abstract r e l a t i o n (or praxis) of r e c i p r o c i t y opposes to "non-relation" (which e x i s t s only f o r an outside observer); praxis takes concrete form i n e i t h e r p o s i t i v e r e c i p r o c i t y (exchange and cooperation) or negative r e c i p r o c i t y (competition, deception, f i g h t ) . There i s a t h i r d opposition whose two terms are concrete r e l a t i o n s toward others. Things or r e l a t i o n s characterized by p o s i t i v e - r e c i p r o c i t y are " r e c i p r o c a l " ; things or r e l a t i o n s characterized by negative r e c i p r o c i t y are "non-reciprocal." "Non-reciprocal," too, i s the r e l a t i o n to persons or things which do not e n t e r t a i n e i t h e r p o s i t i v e - or n e g a t i v e - r e c i p r o c a l 4 p r a c t i c e or communication. (This w i l l allow us to speak of c e r t a i n nat-5 u r a l beings or phenomena — cassowaries and possums, thunder and f l o o d — as "non-reciprocal." Figure 1-1 diagrams these terms. (Cf. also Appen-dix 1 on D i a l e c t i c s and S c a r c i t y . ) ) S i m i l a r ambiguities a r i s e with the notion of morality. The term "moral" has two r e l a t e d meanings. (1) It r e f e r s to that which may have an e t h i c a l judgement applied to i t , quite apart from how i t i s judged. Moral opposes here to "amoral." (2) It also r e f e r s to that which i s good, accepted, conforming to a customary standard, or i s s o c i a l l y d e s i r a b l e . Here moral opposes to "immoral." The noun "morality" 7 abstract relation concrete relations Figure 1-1 The Concepts Reciprocity and Morality also carries this double meaning. On the one hand (and especially in i t s plural form, "moralities") i t refers to rules of conduct, to behaviour either good or bad, proper or improper. On the other hand i t defines pro-per or virtuous conduct. Here as well the semantic f l e x i b i l i t y reflects experience: the tendency to identify what jls_ with what ought to be. There is some a f f i n i t y between the moral (in the second sense) and the reciprocal. According to one philosopher, the raison d'etre of a morality " i s to yield reasons which overrule the reasons of self-interest in those cases where everyone's following self-interest would be harmful to everyone (Baier 1965: 150)." One form of moral wrong in Kewa is re-8 fusing to be held accountable for one's actions when they affect other people adversely. An example is the refusal to compensate or to expiate a debt. This kind of behaviour is wrong because i t is self-interested. While i t may be to a person's advantage to withhold a compensation or pay-ment, for this both augments his wealth and increases his power over the other ( i f the latter does not press his claims), this course of action would be destructive were everyone to follow i t . Moralities support the normative system by enjoining behaviour which maintains some equilibrium in social relations. Debtor and creditor are moral when their actions f u l f i l l the moral condition of r e v e r s i b i l i t y : the behaviour is acceptable to a person whether he i s at the "giving" or "receiving" end of i t (Baier 1970: 108). In the chapters of this thesis I hope to convey an understanding of reciprocity in Kewa culture. The argument begins with the abstract definition of reciprocity (just given) and then moves from abstract and non-mediated to concrete and mediated. The mediations, or "determinations" as they may be known within a dialectic, are the parts of Kewa culture: the moral-religious system, the kinship structures, the institution of the pig k i l l , and the narratives. If the thesis can produce a concrete under-standing of Kewa reciprocity, by means of these successive mediations, i t s intent w i l l be achieved; for the dia l e c t i c , as I use i t , aspires to do no more than offer a (phenomenological) description of what i s . Thus in my concluding chapter I w i l l no more be in a position to define reciprocity than I am at the outset, since a l l definitions are incomplete, partial, and abstract. 9 Although the chapters move the exposition from incomplete to more complete, no one part of Kewa culture i s , of course, more abstract, non-mediated, or in any sense prior to the others. Part IV is necessary to understand f u l l y Part I, and indeed the thesis could have begun with the narratives. Nevertheless, I begin with the moral system for two reasons. F i r s t , the ideas about thought, ghost, sickness, emotion, and so forth con-stitute the conditions of cultural practices, the conceptual system of the culture; they are thus of a most general importance and interest. Second, these ideas constitute a theory about the causality of misfortune. Like other Melanesians, Kewa look upon one another with suspicion: behind each misfortune there is a social cause: ghost attack, sorcery, or witchcraft. Warfare was endemic u n t i l recently. In short, I see many cultural practices revealing a fundamental, latent interpersonal adversity. (It has been called a "paranoia.") In one way or another to overcome this uncertainty about the other i s the role of structure^ and these are discussed next. Part II considers reciprocity and morality in their structural aspect. The f i r s t question I attempt to answer i s a theoretical one: how does action acquire those characteristics of persistence, regularity, and objectivity which allow us to c a l l i t "structural?" Then I consider two structured forms of reciprocity, consanguinity and a f f i n i t y . An under-standing of any structure entails some definition of i t s relation to i n -dividual choice. This i s the question that is raised by the presence of a "complex structure" (as opposed to "elementary structure") of marriage. F i -nally, the discussion turns very br i e f l y to a consideration of "native" 10 models of structure. Appendices 2, 3 and 4 contain ethnographic p a r t i -culars relating to Part II. In Part III, which concerns the Kewa pig k i l l , I shift perspec-tive. Here I consider morality and reciprocity as realized in a specific setting. I am interested not only in the events and practices of the pig k i l l (Chapter 10) but also in Kewa opinions about reciprocity expressed in that context (Chapter 11). Chapter 12 examines metaphorical expressions about pig k i l l i n g : i t shows how some ambiguities and conflicts in Kewa culture are manifested at a symbolic level. Part IV concerns Kewa narratives (tida). Here I rediscover fam-i l i a r r e a l i t i e s : cooperation, exchange, and f i l i a t i o n as they are shaped by contact with the "possible" or imaginary order. My objective in this part is to demonstrate how the narratives can be interpreted as a meaningful, structured discourse; how they reflect cultural relations and practices; and how they thereby engage both the Kewa and ourselves in an act of com-prehension of their culture. In the f i n a l chapter I discuss the Kewa experience of history. I attempt to depict reciprocity and morality in one place at one time (the Iapi d i s t r i c t , 1972), but in this very singular experience the broad themes of Kewa culture are recognizable. The Kewa are a population of 40,000 persons inhabiting about 750 square miles of the Southern Highlands District of Papua, New Guinea. Most of their land forms part of the Southern Highlands plateau at 3200 to 11 6000 feet elevation. Geomorphology and human habitation have produced two different environments, a "grasslands" environment to the north of the Sugu River Valley and a "forest" environment to the south. I did fieldwork in three different communities: Kerabi (Kerabi Census Division), Koiari (Fore-Tsimberigi CD.) and Iapi (West Sugu CD.), a l l three in the Kagua Subdistrict and in the southern part of Kewa t e r r i -tory (see Figure 1-2). The f i r s t two are situated in the forest, the third in the grasslands. Ecological differences associate with cultural ones, and in several places in this thesis I w i l l refer to specific d i f -ferences between the two cultural environments. The forest area i s rough and sparsely populated (about 10 per-sons per square mile). Limestone predominates. Rainfall has eroded the rock into the craggy pinnacles and sinkholes (hereafter referred to as karst holes) that form the "broken bottle country" extending from Mount Bosavi i n the west to Karimui in the east. Only near the settlements does the forest give way to secondary grass growth, usually the t a l l , wild " p i t p i t " (Miscanthus). From the ai r there is l i t t l e evidence of human presence. As one proceeds north the land rises from 3500' to 4500' or 5000'. The austere limestone gives way to rounder sandstone and mudstone ridges, and the forest progressively confines i t s e l f to the ridge-tops. Population density approaches 100 persons per square mile in some valleys. Gardens are conspicuous in the open valleys. Walking tracks j o i n wide government roads; bridges of steel and concrete span major creeks and rivers. 12 Rainfall throughout the Kewa area i s moderate, averaging 110-140 inches per year. Although there i s no marked seasonal contrast, February and September are often the wettest months, May and November the driest; these differences are associated with some seasonality in the vegetation. Even in the driest months, though, some rain f a l l s in as many as twenty days of the month. The area is subject to dry spells during the winter months (July, August). Frosts are rare but may be experienced at eleva-tions over 4000'. The diurnal temperature range i s moderate, about 20°F. Conditions for plant growth are optimum throughout the year (C.S.I.R.O. 1965: 101). There is sufficient l i n g u i s t i c evidence, according to Franklin (1968), to assume that Kewa speakers moved into their present area from the north, perhaps from Enga territory. Known history begins with Euro-pean contact. The presence of Europeans was f i r s t f e l t indirectly: at least in the south Kewa area steel axe blades were in use by the 1930's. The i n i t i a l direct contacts came during the Kikori patrols, explorations undertaken to extend the administration's knowledge of the land between the Purari and Fly Rivers. In 1910 Staniforth Smith reached the Samberigi Valley from Kikori, contacted the Sau population there, but probably did not reach the Kewa. In the 1920's other patrols pushed to the Erave River. Faithorne and Champion skirted the Kewa to the east when they passed through Polopa territory. The f i r s t expedition to cross Kewa country was led by Jack Hides in 1935. Having ascended the Fly River, Hides crossed the Highlands from the Sisa-Bosavi area and contacted the Huli people, the upper Wage groups, 13 and the Kewa. Hides approached the Kewa from the northwest: he was looking for the Purari River and eventual passage to Kikori. Following the Nembi River (Figure 1-2) downstream to the Erave River, he passed near the lower Sugu Valley settlements of Iapi and Taguere, crossed to the Erave River and rafted downstream to the beginning of a rapids. He must have walked through the present-day site of the Erave Patrol Post. Pacification, as the conquest i s called, came mainly after World War II. A patrol post was created at Lake Kutubu in the 1940's, and pat-rols to the south Kewa area resulted in the establishment of a post at Erave in 1953. Patrols to the Kagua and Sugu Valleys were made from Erave and from Mendi. Kagua patrol post was established under Erave within the then Lake Kutubu Subdistrict in 1957. Subdistrict headquarters were trans-ferred to Kagua in 1961. A l l areas came under government control (deres-triction) by the early 1960's, and nearly a l l communities are at present (1972) members of the Local Government Council system. The Kewa today are administered from the four government stations at Erave, Mendi, Kagua, and Ialibu. The grasslands Kewa, with a denser population and a more acces-sible terrain, has benefitted more than the forest Kewa from economic development, primarily road construction. There is some indigenous cultiva-tion of birds-eye c h i l l i e s and of biksa (used for dye) in the forest area. Coffee is grown in two expatriate plantations in Erave; indigenous production of this crop has not succeeded. Some men and women participate in markets held once or twice weekly at the government stations, but the most successful Kewa commodity has been labour power. 14 The staple crop throughout the Kewa area is sweet potato. Two kinds of taro, bananas, and a number of greens also have an important place in the diet. Forest Kewa follow the "lowland" pattern of cultivation (short use with long fallow of forest regrowth, no mounding or turning of soil) while grasslanders use a "highland" method (longer use with indefi-6 nite fallow of grass regrowth, turning of s o i l ) . Men and women spend an appreciable amount of time in the gardens. There may be l i t t l e else to do in those areas where the mission impact has been strong. Such is the case in the forest area. With fighting and many traditional rituals suppressed, l i f e has taken on a certain monoto-nous, almost somnolent quality. Many males choose to escape boredom by seeking work in the European population centres. Those who stay behind spend their time listening to village disputes over divorce and adultery, going hunting, working on the government road, or just talking idly in the men's houses. A noticeable change occurs when a pig k i l l i s being prepared. The village i s mobilized, a c t i v i t i e s and conversation are focused. There is excitement. Disputes drop off, not only because there are other things to be done but also because everyone seems to urge a detente. Such was 7 the ambience in Koiari village in late 1971 and in Iapi in early 1972. In a general way the Kewa resemble their neighbours. The nor-thern Kewa have much in common with the Enga, Mendi, Imbonggu and Wiru, while the southern Kewa correspond more closely to the Foi, Sau, Foraba 8 and Daribi. Kewa culture w i l l be the subject of the chapters to follow; 15 Table I Principal Cultivated Crops Kewa Name Common Name Botanical Name South Northwest Sweet potato Ipomoea batatas mondo sapi, saliba banana Musa spp. kai epele taro Colocasia sp. ma ma taro Xanthosoma sp. taro taro pandanus Pandanus spp. aga apare yam Discorea sp. bira bira sugar cane Saccharum officinarium wali wa Edible p i t p i t Saccharum edule kuni kuni Edible p i t p i t Seatria palmaefolia minya padi corn Zea mays kuni kuniga beans Psophycarpus tetragnolo-bus (?) pesa Sago Metraxylon sp. kawi kawi a resume here is unnecessary. But one element of Kewa social organization, referred to throughout the text, is best introduced immediately. This i s the Kewa repa or "clan." One of the f i r s t things I learned about Kewa social organization was the presence of named groups associated with land and with settlements. The term for these units is repa or ruru depending on the dialect. In the forests, where people l i v e in hamlets or villages, a settlement is associa-ted with one or more repa. In the grasslands, where residence is dispersed, a tract of community land containing several scattered homesteads is asso-ciated with a ruru. Even before he knows the structure of Kewa consanguin-16 eal kin terms, the outsider learns two additional things about repa (and ruru): f i r s t , they are not residential units so much as kin groups; second, a l l same-generation repa co-members are "siblings." Repa are p a t r i l i n e a l . Their members are related through male f i l i a t i o n ; they undertake cooperative a c t i v i t i e s , share a common mythic and h i s t o r i c a l background, and have a supra-lineage genealogical status (not a l l repa members can demonstrate actual interrelation). Every Kewa man, woman and child belongs to one repa, normatively his or her father's. Residence is normatively p a t r i - v i r i l o c a l , and children grow to maturity in the company of age-mates who share the same repa name and are therefore siblings. They form the core of the local repa. As siblings, the age-mates cannot intermarry; they find spouses in other repa. Repa sisters depart to l i v e with their husbands, brothers bring in wives from the outside. The stress on male f i l i a t i o n is evident in the alternatives araame ("father and brothers," male speaking), arabali ("father and brothers," female speaking) and aramaai ("father and father's brothers"). Actual repa organization is complex, geographically (repa are dispersed as a result of group fission) and organizationally (repa of different orders of inclusiveness are combined or opposed within a single community). Appendix 2 examines repa organization in more depth. For the purpose of this thesis, a repa is a local group of p a t r i l i n e a l l y related men who — as brothers, father's brothers and sons to each other — are solidary and cooperative. Where are the Kewa precisely? The map (Figure 1-2) locates them geographically. To locate them culturally is the task of the four parts of this thesis. As for their place within a more diffuse "human experience," that is l e f t unsaid. Perhaps the concluding chapter has an answer. 18 Figure 1-2 Map 19 Footnotes 1. When I say "Melanesians" I have in mind especially the men, since Melanesia is a culture area where men dominate the practical and sym-bolic domains. During my fieldwork women did not confide in me the way men did, and I almost always saw them through the eyes of their brothers or husbands. The following study therefore imparts a male, and to that extent a par t i a l , view. It would be interesting to ex-plore the nature of this p a r t i a l i t y at some length, but I shall make only a brief comment. Kewa culture is organized along male lines in that: (1) men are united through patriliny and residence, women not; (2) men are the main participants in the enactment and transmission of the important practical, moral, and symbolic relations in the cul-ture; and (3) women are largely confined to exercising indirect con-t r o l over cultural things through their husbands and brothers. Fo-cusing on the male side as I do, I undoubtedly emphasize (1) the nor-mative and symbolic elements of the culture, at the expense of the sometimes countervailing domestic p o l i t i c s initiated by women, and (2) the internal structural order of norms and symbols, at the expense of their possible function to ideologically conceal, or ju s t i f y , parts of Kewa culture. 2. The term occupies a central position in Sartre (1960); i t appears re-peatedly in Levi-Strauss' La Pensee sauvage, especially in the f i n a l chapter "Histoire et dialectique." As of 1966 (the year of the English translation, The Savage Mind) "praxis" has appeared recurrently in studies on the philosophical roots and practice of structuralism, positivism, or anthropology generally. Cf. for example Murphy 1971, Krader 1974, Scholte 1974, and Rosen 1974. It has had a longer usage in Marxist literature, where i t often i s used synonymously with "practice." 3. Within negative reciprocity one could, of course, also distinguish a 'negative cooperation' and a 'negative exchange' (cf. Kotarbinski 1970). As for 'reciprocity-in-general,' Sartre (1960) specifies that four conditions are f u l f i l l e d in a reciprocal relation. Reciprocity implies (1) that the other person i s a means to the same degree that I am a means, that i s that he i s a means to a transcendent end and not my end; (2) that I recognize the other as praxis, that is as an on-going totalization, at the same time that I integrate him as an object in my totalizing project; (3) that I recognize his movement toward his own ends in the same movement through which I project myself toward my own; (4) that I discover myself as object and instrument of his ends by the very act which constitutes him as objective instrument for my ends. In positive reciprocity each person makes himself a means in the other's project so that the other may make himself a means 20 in his own project, the two transcendent ends remaining separate. (This i s the case with exchange or the prestation of services.) Or the end is a common one and each makes himself the other's means so that their combined efforts may realize their single and transcen-dent objective. (This i s the case with cooperation.) In negative reciprocity the four conditions are f u l f i l l e d on the basis of a r e c i -procal refusal: each refuses to serve the ends of the other and, while yet recognizing his objective being as a means in the adversary's project, each puts to profit his own instrumentality (in the adver-sary's project) to turn the other in spite of himself into an instru-ment of his own ends. In the case of struggle or fight, each person reduces himself to his own materiality in order to act on that of the other; each through his feints, ruses, frauds, and manoeuvres, allows himself to be constituted by the other as a false object, as a delu-sive means (Sartre 1960: 192). 4. Levi-Strauss notes that: ...observers have often been struck by the impossibility for natives of conceiving a neutral relationship, or more eactly, no relationship. We have the feeling — which, moreover, i s illusory — that the absence of definite kinship gives rise to such a state in our consciousness. But the supposition that this might be the case in primitive thought does not stand up to examination. Every family relationship defines a certain group of rights and duties, while the lack of family relationship does not define anything; i t defines enmity (Levi-Strauss 1969: 482). He then cites Evans-Pritchard, speaking of the Nuer: Either a man is a kinsman, actually or by f i c t i o n , or he is a person to whom you have no reciprocal obligations and whom you treat as a potential enemy (quoted in idem). 5. The word "possum" w i l l be used instead of the more unwieldy term "marsupial." 6. Cf. Robbins 1963 and Brookfield 1970 for more description. 7. Table II gives the result of a time-use survey in Kerabi (a total of 2700 man-hours for both sexes) and in Koiari (1010 man-hours among men). The Kerabi survey was made between 25 June and 6 August 1971, the Koiari survey between 9 November and 6 December 1971. The dates were chosen mainly for my convenience. (I have noted that major sub-sistence a c t i v i t i e s do not vary seasonally.) The Koiari survey i n -tentionally covered a one-month period ending three weeks before the 21 1971 pig k i l l ; i t reflects a period of r i t u a l preparation and some ri t u a l activity (see Table XI in Chapter 10). I collected the data by (1) making a l i s t of a l l the married men and (in Kerabi only) the women in the village, then (2) following the f i r s t four or six persons (or, for three weeks, eight persons) for one week, then the next four or six for a week, and so on. "Following" means making direct obser-vations, occasionally making discrete inquiries, or recording infor-mation (contributed by an assistant or two delegated to do the same) about the a c t i v i t i e s of the surveyed persons. Certainly there was no ignorance on the villagers' part, nor was there any resistance, with regard to this study. In Kerabi some men and women would spend sev-eral days at a time in second residences, ostensibly to tend gardens and pigs but for other reasons as well. A short pilot survey of time-use in these residences enabled me to adjust the main sample. The noticeable contrast is between Kerabi men and Koiari men. In Koiari there is a reduction of act i v i t i e s e to h and a significant increase in activity _1. One could c a l l these two patterns "modern" and "traditional." Table II Time Use Kerabi Koiari Men Women Men a. Roadwork 12 12 14 b. Domestic-Leisure 27 23 33 c. Pig Care 3 5 3 d. Gardens, firewood 21 38 17 e. Disputes, court 7 3 1 f. Hunting 6 0 0 g. Visiting 4 2 1 h. Mission 6 8 1 i . Sick 4 4 4 j . Other government work 2 2 0.5 k. Entrepreneurial* 3 0 0.5 1. Social, ceremonial 3 1 23 m. Other 3 2 1 TOTAL 101 100 100 * Includes tending c h i l l i e s in Koiari and (two men) tending a new tradestore in Kerabi. 8. For the Enga cf. Meggitt 1965, Waddell 1972; for the Mendi cf. Ryan 1959; for the Wiru cf. A. Strathern 1968b, 1971b and forthcoming; for the Foi cf. Langlas 1968; for the Daribi cf. Wagner 1967, 1972. 2 2 In addition to these sources the reader can consult, on social organ-ization: Barnes 1967, Cook 1970, Franklin 1965, Glasse 1968, Glasse and Meggitt 1969, Langness 1964, van der Leeden 1960, de Lepervanche 1967-68, Pouwer 1960, 1964, 1967, A. Strathern 1968a, Watson 1970. On exchange and p o l i t i c s : Berndt and Lawrence 1971, Bulmer 1960, Elkin 1953, A. Strathern 1969. On religion: Lawrence and Meggitt 1965, Read 1955, M. Strathern 1968. On subsistence and ecology: Brookfield 1964, 1970, Clarke 1966, Rappaport 1967. 23 PART I MORALITY CHAPTER 2 ASPECTS OF BEING If, as I have said, the conceptual scheme governs and defines practices, i t is because these...are not to be confused with praxis which...constitutes the fundamental totality for the sciences of man....Without questioning the undoubted primacy of infrastructures, I believe that there is always a mediator be-tween praxis and practices, namely the conceptual scheme by the operation of which matter and form, neither with any independent existence, are realized as structures, that is as entities which are both empirical and i n t e l l i g i b l e (Levi-Strauss 1966: 130). In the introductory chapter I set out some general ideas about reciprocity. I defined i t as a "state" or "perspective" which is implemented in two practical forms, cooperation and exchange. Reciprocity is the more abstract term, describing an aspect of praxis; exchange and cooperation are more mediated, denoting types of cultural practices. The relation between non-mediated and mediated implies that i f one were to strip away a l l the concrete determinations in ceremonial exchanges, r i t u a l , warfare, domestic relations, disputes, and so forth, one would be l e f t with this generalized notion that we have, defined as "reciprocity." I w i l l proceed inversely. Having begun with abstract definitions, I hope to constitute the concrete by accumulating determinations. But in order to move from praxis to practices, Levi-Strauss asserts, one has f i r s t to know the mediating conceptual scheme. Unlike Levi-Strauss, who would see in this scheme analytical reason at work (that i s , an inert c l a s s i f i c a -24 tion of things on the basis of categories of understanding) I shall stress i t s fundamentally dia l e c t i c a l character (cf. Krader 1974: 339). The main opposition, or antithesis, is familiar to philosophical and r e l i -gious phenomenology: mind and body. Or, in Kewa, kone (thought) and to (body). These categories and their interaction (in oma, the act of awareness of an organic state such as sickness or emotion) are the sub-ject matter of this Part. In calling attention to a "dial e c t i c " I want to point out that these categories are an object of Kewa understanding. They are en-tertained not just passively as a kind of f i l t e r for experience, but actively as well. They are less a conceptual "grid" than devices through which a person may choose to understand or dissemble himself. In this chapter I shall discuss, f i r s t , concepts pertaining to the body; second, the concept "thought" or "behaviour"(kone); third, the concepts "ghost" (remo) and " s p i r i t " (wasa); and, fourth, other Kewa entities (wildman,demiurge). The f i r s t three conceptual classes are re-quired for the arguments of Chapter 3; the fourth class w i l l be needed for Part IV. 1. Body: to, lo, and pu. Kewa recognize the role of both father and mother in conception. They consider the child to be "bu i l t " from the mother's blood (we, kupa) and the man's semen (tene page). The foetus is b u i l t gradually through repeated intercourse. Parturition i s designated by the verb mata, which means "bear" or "carry in a net bag." A woman recognizes preganancy when she no longer "bears" blood, when the 25 1 blood goes to make the child that she w i l l bear. Men also "bear" (mata) children, for the term refers to social as well as physiological parentage; but they may also be said to "bear" (ria) children in the sense of "carry on the shoulder." To refers to the "body," the physiological aspect of one's 2 being resulting from the combination of blood and semen. In a narrower sense _to may refer to the trunk or abdomen. Within this area the l i v e r (pu) i s of central importance, for i t i s most commonly thought to be the site of emotion and impulse. Of a-person who acts without deliberation i t may be said that he has "acted from the l i v e r . " The stomach (roba) is mentioned in a similar context, either alone or in association with the l i v e r (pu roba); the intestines (lo) are also referred to as the locus of emotion. When consuming these parts of a pig, Kewa say that the pu is stronger than roba or 1<3. Liver and intestines are associated with male and female qualities, with blood and with the umbilical cord and 3 afterbirth respectively. 2. Kone. The term kone refers to Awide range of interior states centring on the idea of "thought." As a noun i t s usage includes behaviour, act, knowledge, memory, intention, opinion, w i l l , and so forth. Kone is both intentional consciousness and the overt behaviour resulting from intentions. As a verb the term convers the verbal equivalents of the nouns li s t e d above. The verbal expression i s "to put thought" (kone sa, kone wia). There are two contexts in which the notion of kone is used in speech: reflexive and non-reflexive. The reflexive use of kone suggests 26 reason and judgement controlled by the awareness of alternatives. The expression designates the speaker's opinion, and i t occurs in statements of the type "I think that...." For example: (1) "My thought i s : we should not burn (the bristles) and give the pig to them, I think." (Menada na kiru atama kone wialo, na konemare.) (2) "But i f we burn and give pig to the men who come to k i l l our pigs, that's a l l right. I w i l l s i t down and think 'good'." (Pare nainya mena kiala ipulumi alinu repara kiru atamadare, ale meda dia, nimiri epelea kone wia pitua.) The non-reflexive use of the word kone indicates that the actor's thoughts were not aware of themselves as such. Rather someone else (or the same person at a different time) turned his attention to them. These are statements of the type, "he thought that...," "his opinion was...," "I thought...." Kone is mentioned in this context when i t contains an uninformed opinion or an opinion contrary to fact. (For instance in (3) and (4) the implication i s "a third person was actually accompanying the two" and " i t was not really the brother who shot the arrow.") (3) The two of them thought they were alone. (Ipu kome poloame ipu kone wia pirapeda.) (4) My brother shot (an arrow), she thought and... (balimi pia kone wiawa.) Finally, kone may function as a noun or as an intransitive verb to express a moral judgement about a person or a situation (examples 5-7) or to imply that moral interests are questioned (examples 8-10) : 27 (5) We do not have good thoughts (actions) , truly we have bad thoughts. (Epe kone na wima, ora koi kone waru wima.) (6) You gave only one thought and then came. (Nere kome kone u mea go epa pimida.) (7) I have this thought of dislike. (Gi pia kone wia pi.) (8) Since I have this opinion, I am speaking out. (Ni konema wiatada, goina apo laloda.) (9) This thought stays in my intestines and stomach. (Na lo roba para go kone wia ayo.) (10) I want to see his behaviour. (Ipinya kone adalua.) Kone is the means of discriminating between good (epe) and bad (koi), between things done rightly or properly (ora) and those not so done. It i s the faculty of reason which makes a man responsible for his actions. A being without kone acts from impulse and instinct, l i k e the pig which spoils a sweet potato garden without knowing i t has done something wrong. It follows that one's moral sense can also be deemed good or bad. Hence epe kone and koi kone, good thoughts and bad thoughts. Applying either to some specific action or to a person's behaviour in general, kone carries a strong element of moral evaluation. When a per-son is told he "puts bad thoughts" the implication i s not just that he has done things badly or ineptly, but that he has acted improperly be-cause he is an improper person. A Kewa who witnesses a social infrac-tion might say "bad behaviour," but i f he wants to phrase his criticism moiE strongly he can imply that the bad conduct might have been expected; 28 then he w i l l say "koi a l i ! " (bad man, immoral person!). But each implies the other. Wrong actions may result from the absence of thought as well as from bad thoughts. The only occasion when actions might not reflect kone is during possession madness (ema) or after an attack by a s p i r i t ( s i a k i l i S) or ghost (remo). Then the person becomes uncontrollable, immoral, exceptionally strong, and accomplishes feats of which no memory 4 is retained. Loss of consciousness occurs when a person no longer "holds" kone. Kone is acquired not from birth but from the age of social awareness. The shifting, unfocussed, purposeless activity of children is an index of their lack of kone. Birds and marsupials do not have kone, nor do pigs (except in narratives); small children are like them. The growth of kone is a gradual process involving the implantation of s k i l l s and knowledge by the parents and other close relatives. Prompting the child in the proper use of kin terms, urging him to give to others, imparting s k i l l s — these develop kone. There is no set age at which kone i s fu l l y acquired, for one continues to accumulate i t u n t i l s e n i l i t y or death. Some children develop i t much earlier than others. Early acquisition may arise from some childhood trauma such as witnessing the violent death of a father. A boy who has seen his father's death i s quickly inspired by an awareness of his responsibility as a social being and group member; he is awakened to the reali t i e s of l i f e , the presence of non-reciprocal and unilateral acts; he becomes less dallying and 29 thinks more. (As far as I know, a father's kone does not actually enter  into his son at death; such a possibility seems inconsistent with the way Kewa talk about kone.) 3. Remo and wasa. The word remo refers to those aspects of a man's being which are encountered after his death, his "ghost." A livin g person does not possess a remo, but in addition to his _t£ and his kone (his physiological and his rational functions) he does possess a wasa, or "soul." One's own wasa i s never the object of awareness, but i t is the subjective centre for a certain kind of experience. When one's active analytical faculties (kone) are engaged, these faculties along with the senses dominate a person's grip on reality. When a person "puts" kone he is in contact with a reality in which there i s reason and awareness. When the kone recedes (as prior to a sickness), one's hold on the kone-perceived dimension of reality also diminishes. The f i r s t sensation may be, for instance, that of walking without the feet touching the ground. In a later phase, when a person is possessed, he may believe he walks up in the tree branches or jumps over houses. This is the dimen-sion of reality with which the wasa is normally in contact. During sleep the kone i s inactive and i t i s then that the wasa informs the person about reality through his dreams (upa). Kewa say that the dream is what the wasa sees when i t leaves the body during sleep. The reality seen is not that of everyday awareness (for the latter i s , I have suggested, the experience of kone-dominated perception and reason) 30 but is a kind of symbolic and metaphorical transformation of that reality. Furthermore, the reality is not everyday circumstance but rather has some 5 connections with "power." Men dream of things which are thought to reveal some unilateral or non-reciprocal relation such as death, sorcery, hunting, or s t e a l i n g — a l l of which are inherently "powerful." Symbol (or metaphor) and power imply each other. No dream is ever insignificant, and one does not interpret a dream l i t e r a l l y . Dream interpretation requires kone: since the wasa is the subject only of immediate experience and can never reflect on what i t perceives, the dreamer must awaken before he can discover dream meaning. Like divina-tions and oracles, dreams never t e l l the whole story. Dreaming about a cassowary sacrifice would immediately suggest a death, and the identity of the sa c r i f i c e r would reveal the repa (clan) of the deceased. But the actual identity of the one who had died, or was about to die, may remain unknown. Shortly after death the wasa leaves the body and inhabits the vi c i n i t y of the corpse. It may then be referred to as "wasa remo" (or "soul ghost") or simply as remo. Usually invisible, i t s presence around the corpse may be revealed by a whitish glow, by a low whistle, or by a rustle in the undergrowth such as might be made by a rat. "We who l i v e eat good things like marita pandanus and pork, but to the man who has died we (Kewa) give the name remo." This observa-tion w i l l carry us a long way toward an understanding of the notion of remo. Marita and pork are not ordinary foods; they associate with communal 31 feasts, ceremonial exchanges, in brief with moral and reciprocal conduct. Eating marita and pork signifies discharging old obligations and creating new ones through the exchange of objects, services, or hospitality. So the dead person not only does not eat marita and pork, he i s also the worse for i t , and as remo he is characteristically resentful and envious of humans and human commerce. Once a l i v i n g man has become a remo he may be disposed to attack those against whom he bears a grudge. (I witnessed a white-haired old man remind his debtors of this fact by opening his mouth wide, baring his teeth, and then saying that in just such a way would he "bite" certain people after he has died.) But we shall see that remo also intervene on the behalf of li v i n g men. Invisible like the wasa, remo constitute an ambiguous form of experience, for rarely i s i t pos-sible to be certain about the identity of ghost attackers. Divinations and oracles can only limit the number of possible interpretations, and even they are often silent. Only two things are certain; f i r s t , men act non-reciprocally, incurring the resentment of others; second, men f a l l i l l , get lost in the bush, hear things which are not seen, see strange glows in the forest, even get tossed up to the tops of casuarina trees. Themselves immaterial, remo may stay inside small, curiously shaped pieces of stone or f o s s i l : in artifacts of quartz and obsidian; in f o s s i l shell, bone, beak or claw; in mortars, pestles and hammerstones belonging to an earlier Highlands culture. (These objects are grouped 32 by the r i t u a l process in which they are used: for instance, adalu ribu uses a set of mortars, rudu ribu the paleolithic objects and f o s s i l s , opayo ada the pestle, and so forth.) A remo exists either in a mobile, immaterial form in which case i t is uncontrolled and vengeful; or i n an immobilized material form, controllable and passive. These two forms are also the remo's nocturnal and diurnal beings. The sacrifice and r i t u a l in the cult house enjoins the ghost — by intoning spells, by applying pig grease and tree o i l to the stones, and by sharing pork — to remain in the stone even at night. Control of a ghost is contingent on the discovery of i t s stone. Once the stone is found and possessed i t is possible to bring some measure of control and s t a b i l i t y into one's affairs — at least as far as one's immediate ancestors are con-cerned. Ancestral stones are found in the gardens or in the forest. They are carefully guarded either in the cult house or in a cl e f t of limestone or tree base in the forest. Those who entertain moral, cooper-ative relations (who hence do not intermarry) keep their ancestral stones in the same place. Helping in the building of a cult house and sharing in the r i t u a l sacrifices both express and promote reciprocal relations. Two different repa may exchange their stones to express their moral equivalence. Alternatively a group developing an internal dispute w i l l at some stage separate i t s stones along the prevailing factional lines; henceforth they w i l l make separate r i t u a l s . The fate of ancestral stones mirrors that of l i v i n g men: where the collection of stones is large, 33 there the group is stable and powerful, fortunate in male progeny; where 6 the stones are lost in the forest or scattered in the gardens i t i s be-cause the men have died without sons to replace them or because the group is overrun in warfare. Where the repa is united, so are the stones, and the several constituent sub-repa (cf. Appendix 2) make their rituals together. Where two repa are closely a l l i e d without merging, they may exchange their stones, and repa A w i l l make a ri t u a l with the ancestral stones of repa B and vice-versa. 7 Remo associate with the wild and the non-reciprocal. Men who cooperate to control them r i t u a l l y are necessarily morally related, whereas the remo themselves are negations of this equivalence. Kewa say that they fear the stones. Traditionally (Kewa insist) the stones were seen only when they were being used r i t u a l l y ; doing otherwise brought on sickness. Women and children were not allowed to see the stones, and the bachelors who officiated in the rituals were the only ones actually to manipulate them. (Today under mission influence these ideas have lost some of their force.) But although remo are, by their very nature, perverse and pes-tile n t , they are less a threat for some men than they are for others. Some men go much further than others in successfully protecting themselves from either human or non-human incursions on their persons. These are men of property and substance, big-men (ama, amonai, kainya'ali) who, sons of big-men, have always occupied an advantageous position in ex-change relations with their fellows. Less apprehensive than most about the effect their actions have on others, such men are often thought to be 34 "bad" because of the way they turn rumour and suspicion to their advan-tage. They do not take seriously accusations of kone attack or sorcery, and they good-naturedly defend themselves against charges of hoarding, manipulation, and abuse of power. Big-men thus build up reputations of immunity from the effects of guilt or witchcraft (cf. Chapter 3). For this they are grudgingly admired. Such men may become remo re (source, origin, or base of remo), persons who have the a b i l i t y to discover the stone inhabited by a wasa  remo s t i l l "at large" after a death. The remo re discovers the stone and brings i t to the brothers or sons of the deceased, who purchase i t for as much as several pearlshells. Alternatively, the remo re may capture a remo which has caused a death and bring this to the deceased's rela-tives who, eager to protect themselves from further onslaughts by the ghost, give even higher compensation. The remo re finds the stone near the corpse, in the grass or undergrowth, in the roof thatch, etc., d i -vining i t s presence by means of spells. In the following account Poiolo Robo t e l l s how he received his remo re s k i l l s from his dead father. I and several others went into the forest to find a domestic pig that had run off. We separated. Without my knowing of i t the others discovered and k i l l e d the pig. Later on and by a different track, I came up to the place where the pig had been k i l l e d . The others had cut the pig with an ax and there was blood a l l about. I looked at this place and became mad (ema); I was walking up among the tree branches, not on the ground. It was from up there that I saw one ekamu possum on the ground by the pig blood. I wanted to k i l l the possum, but I could not come down from the branches where I was fl o a t -35 ing. Then my father came up. (Robo's father, named Yapa or Possum, was dead at the time.) In one hand my father was hold-ing the possum, which was crying "e-e-e!" and in the other hand he had a stone object. My father went into a karst g hole and I followed and took the possum. Then my father gave me some kapipi (tree-fern) leaves, eka (fern) leaves, kikala leaves (or pakena, used in some potions), a wabi (paper wasp nest) and some cane rope. He told me to put the possum into the opening of the wabi and to fasten the leaves around the outside, then to k i l l one rufous pig and to give the blood to this tied-up (rogo) thing. I did a l l this. The possum ate the blood. My father told me that i f I should find an ekamu possum in the bush, this w i l l "change into a rudu ribu, an ancestral stone." ( I n i t i a l l y the wasa remo i s a possum moving in the grass; when k i l l e d i t becomes the stone.) If I k i l l a rat, i t w i l l change into an adalu ribu stone. I have received compensation for finding these stones for groups li v i n g twenty miles distant. I may be called when a man is sick; then I find the appropriate stone, apply the blood, and say the spells which bind the ghost inside the stone. Robo's method would begin with s i t t i n g in the invalid's house and going into a kind of trance. Suddenly he would jump up and seize something in the corner or in the ashes. He would trap the object with his hands as i f he had caught a mouse. Others would not be able to see anything, for remo would be invisible to them. Once caught in the f i s t , the "mouse" changed into a stone, an adalu ribu. (If the remo escaped from the £e's grasp, i t was thought that the invalid would die.) 36 From this account ( i t should be apparent I have never seen a remo re in action) one can conclude that a certain amount of manipulation goes into the remo re's profession. Nowadays a few Kewa have come to perceive this. But i t would be imprecise to say that he plays on their fears, "cons" them. In reality, the presence of the non-reciprocal among men presupposes a human agent to control i t . The remo re renders a ser-vice which requires compensation. This is because relationships between liv i n g and dead, like those among the l i v i n g , involve exchanges. Pig blood, pork, t r e e - o i l , ashes and so forth are given to keep away the dead. When the remo stay away (staying away is the best they can do for men, except i n cases such as Robo's) a man is active, pursues his gar-dening, his pig husbandry and his exchanges; he profits from his good health and good fortune. Hence r i t u a l procedure, spells, and stones a l l have exchange values. This is why the remo re i s a man of substance. His wealth commands respect; he appears to have enjoyed a relative im-munity from the non-reciprocal. In Robo's case being a big-man j u s t i f i e d his divinatory power while "profits" from his services further increased his p o l i t i c a l stature. So, while there are big-men who are not specialists in divining and healing, just as there are such specialists who are not big-men, one can see that the two roles have much in common. 4. Other beings. There are a number of other non-reciprocal beings which enter into Kewa experience. They w i l l reappear only much later on in the thesis (Part IV and Chapter 19) but this is the proper place to make their acquaintance. These beings I c a l l "wildman," "ghoul," 37 and "demiurge." As with my translations of remo and wasa, "ghost" and "soul," these English terms are pis a l l e r s : what matters is the type of experience to which the beings pertain. I shall f i r s t describe the at-tributes of these beings and then define them in contrast to wasa and remo. "Wildman" is my translation for kalado (W), alomogiali (S), koropuali (S) and tapo (SE). The wildman is a wild counterpart of cul-tural man: he resides in the deep forest, i s solitary, and acts non-reciprocally. Though wildmen may be of either sex, be child or adult, people generally encounter the adult male. The wildman i s of strong physique; he wears his hair long, f a l l i n g down to the shoulders or to the waist ("like a European woman"); his eyes are bloodshot, his fingers long, his teeth tusk-like. In contrast to his h o r r i f i c features,his ornaments are exceedingly good; he wears a nose-piece, earrings, armbands, bailer shell, cane-ringed bark belt, and a good net or bark-cloth apron. Though they may be encountered abroad or in the forest, wild-men prefer watery places. Some say their abode is i n the deep pools where rivers bend or where tributaries join. They may also be encountered near streams, swamps, or karst holes. Some say that wild men cannot be seen by men, others that they are vis i b l e but choose not to show themselves. More simply: some men are known to see them, others not. A l l agree on the wildman's disposition: he is capricious, attacking some men but not others seemingly without reason. Like marsu-pials they are abroad mainly at night. Men encounter them while hunting 38 in the forest by moonlight, when on an overnight fishing expedition, or when k i l l i n g pigs. If wildmen travel by day, the weather is foul, forest and village laced by heavy downpours. Wildmen are said to be attracted by the smell of pig fat and by the odour of fluids secreted at sexual intercourse. Twins are born when a wildman mounts a woman after the husband; each twin bears the mark of his natural or cultural genitor. (The wild twin is k i l l e d ; the mother herself may die.) Forest Kewa say that some wildmen k i l l by throwing a barb which, unless extracted by an expert, causes a slow death. Others constrict the victim's chest and stomach with cords. A specialist may remove the barb by saying spells and f l a i l i n g the victim with nettles. If a death results the corpse is thrown into a river, for only in this way w i l l the deceased himself, or his remo, not become a wildman. "Ghoul" translates rogoma or rakoma (W). The rogoma may be of either sex but is generally female. She is a woman who has died but then come to l i f e again, saying she was not really dead. Thereafter she lives an apparently normal l i f e during the day — looking after children, pigs, and garden — but at night she does not sleep in the house. Instead she ranges abroad, seeking out recent deaths. When she comes across a recent corpse, wrapped up in pandanus leaves and suspended above ground, she cuts off a piece of flesh (or heart or l i v e r ) , puts i t in her net bag, and then goes off to her garden to eat. Corpses of deceased males seem to be preferred. 39 A woman becomes a rogoma when a rogoma inhabits her body at death, replacing the woman's wasa. In Kewa tida (myths) a rogoma and daugh-ter come up to a woman and child in the forest, bites them to death, and cuts the skins off; then rogoma and child don the skins; husband returns and eventually surmises that his wife and daughter have become rogoma. In l i f e rogoma are rumoured to live among neighbouring hostile or un-a l l i e d groups. The term "demiurge" denotes an autonomous creative force or decisive power. Demiurges bear proper names which vary from one place to 9 another. A demiurge seems either to be a unitary being or a male-female pair. It may be associated either with ce l e s t i a l or chthonic regions. The ce l e s t i a l deity in Iapi is known as Yeki, as Yekiuna (Yeki woman) or as both. The subterranean being bears the name Repanapada and inhabits the confluence of the Erave and Sugu Rivers. In the neigh-bouring Kanodoba d i s t r i c t (which does not border on a major river) the subterranean beings are known as Pari Ibimi and Pari Ibimiuna. These chthonic beings are thought to cause earth tremors either by digging in their underworld gardens or shifting their buttocks while s i t t i n g weaving net bags. The ce l e s t i a l powers cause natural phenomena such as thunder, lightning and heavy rains. Yeki's existence is inferred from the apparently purposive nature of these events. Lightning bolts are ways of designating persons who are, or who are about to be, marked by some non-reciprocal event (death, sickness, dispute or war). The marking process u t i l i z e s 40 metaphor and metonymy; i f a lightning bolt f a l l s on a particular tract of forest i t is thought that those who reside there w i l l be the cause of a war; i f i t f a l l s on a specific tree, a person who has taken that tree name for his own name w i l l sicken or die; i f a rock is broken the person in whose garden i t l i e s w i l l sicken or die, and so forth. The marks that lightning leaves on tree trunks are called Yeki's footprints. Wildman, ghoul and demiurge cannot be neatly distinguished from remo. I shall nevertheless attempt to separate them as types of experiences. Everyone has a wasa and eventually becomes a remo, but only a few unfortunate souls turn wild. That is the t e l l i n g fact. The question i s not: who (which remo) becomes a kalado or rogoma? But: is there a type of non-reciprocal experience where the actual identity of the being i s unimportant? If the experience of the non-reciprocal is imputed to a particular deceased person or to the dead of a particular group, then remo are invoked. (I should add "imputed to a particular person by_ a particular person or group.") That i s , a misfortune i s a t t r i -buted to a remo when i t appears to have originated somewhere within the reciprocal community; a human (or ghostly) agent i s readily discovered given the circumstances (timing, symptoms) of the misfortune. As we shall see in the next chapter, attributing a misfortune to a remo implies the likelihood of continuing reciprocities between victim and accused, since 10 the former is partly to blame for his misfortune. It is different with the three categories of being just discussed. They are associated with events which cannot be expiated for three reasons. 41 F i r s t , the identity of the being may be unknown: certain tracts of woods or watery places have long been associated with "a" ghoul or wildman. Second, the being may be associated with an enemy group with which there can be no positive reciprocities. Such beings are experienced as threa-tening or otherwise affecting the community as a whole. Third, the being i t s e l f is autonomous, unmotivated by the kone of the l i v i n g . Insane men and women are kalado- or rogoma-like in this respect. Or one could say that certain insane persons are thought to be kalado or rogoma by out-siders, especially hostile outsiders. Kalado and rogoma, wasa and remo, kone, to, lo and p_u — these are the dimensions within which Kewa experience themselves. In their en-counters with one another and with the non-reciprocal, though, Kewa do not experience these aspects of being as discrete categories. Exper-ience of one's organic being, of one's thoughts, and of the presence of ghosts combine in a single synthetic experience. Morality emerges when there is awareness of the way these categories of being pass into one another, when one sees the remo in the man. In Chapter 3 both kone and remo take on a more complete situational meaning. 42 Footnotes 1. One might object that "bear" is an inadequate translation. Perhaps. The English word does have, however, a parallel meaning with the Kewa: i t signifies both "to give birth" and "to carry." In Kewa the two meanings come together semantically, because the net bag is associated with the womb. 2. The pregnant woman may predetermine the sex of her foetus by having a small child of the desired sex sleep at her side. The remo (ghost) of an ancestor may v i s i t the father or mother in a dream, but I have not heard i t said that ghosts determine the sex of the unborn. Bodily differences between the sexes depend on cultural factors. Young males eat pork, game, and marita pandanus with their fathers, are physically active in the forest, purchase spells to make the skin secrete o i l , rub themselves with tree o i l , pig fat or marita juice — acts which produce the lean, muscular male physique. Young females on the other hand are enjoined not to eat meat; they spend their time in the com-pany of their mothers and sisters in the gardens, and accordingly their bodies retain the fatty softness which is the stamp of one's birth from a woman. But even though the skin of men may become tight over their bones and muscles, the very fact of having a body means men must continue to reckon with the fact of m a t r i f i l i a t i o n . Throughout their lives they must compensate the mother's group with valuables; they must give payment to the wife's group for the _t£ of the child. 3. It may be that l i v e r associates with "male" and deliberative emotions, such as frustrations over broken reciprocities, and that stomach and intestines associate with "female" or impulsive emotions, such as affairs of the heart. My data are insufficient here. 4. Acts of strength (such as jumping over a house) and immorality (such as beguiling a young woman) are associated. 5. The notion "power" — which I w i l l use sparingly — has basically the same meaning as "non-reciprocal." "Non-reciprocal" is used to desig-nate (a) unilateral or negative-reciprocal relations among men, and (b) the non-human world of wild and divine things. Both are charac-terized by non-moral behaviour, by excessive, insufficient, or uncon-trolled communication. Sorcerer, ghost, and marsupial are a l l non-reciprocal. 6. They are eventually rediscovered, perhaps only after several genera-tions of dormancy, and acquire new "identities." Some stones are kept by individual domestic units for private purposes (such as placating a 43 dead father or father's brother); others are kept by the repa as a whole and used to placate repa ancestors collectively in rituals such as adalu ribu, rudu ribu, kepeta ada, and others. Cult house r i t u a l is a topic which requires more than the passing reference given here. Because rituals were not being openly practiced at the time of my fieldwork, my data are incomplete. 7. Cf. footnote 5 . 8. A karst hole is a natural drainage tunnel descending more or less verti c a l l y through the rock to a subterranean drainage system. It is a common feature of limestone country. 9. I never managed to acquire more than a superficial knowledge of these beings. 10. This would separate ghost-attack from sorcery. Enemies, with whom there are no active reciprocities, practice sorcery. 44 CHAPTER 3 THE CONTEXT OF MORALITY To understand how these categories of being are part of a moral order, we have to know their interrelation. This implies looking at them in the social context. In this chapter I am not concerned with the "idea" of kone or remo so much as with the situations in which the mind reaches out toward these notions. Kewa say that kone has i t s locus between the brow ridges in the lower forehead (weno S, eno W) . But thoughts do not lodge there: impressions are confirmed or dispelled, ideas formed and quickly forgotten, doubts raised or resolved, and so forth. Yet some thoughts persist lon-ger, some seem to acquire a l i f e of their own. These are thoughts which bear on some loss, on some disturbing or intolerable situation, or on a sense of wrong. These emotional thoughts are the subject of this chapter. For the most part I w i l l be considering one emotion — frustra-tion or anger — and how i t mediates kone and remo. But I w i l l also look at other emotions, physical injuries, and the central role of compen-sation. Finally, I put forth some conclusions about the moral system. 1. Anger and "tied-up" thoughts. Anger, frustration, and 1 resentment — a l l these translate the Kewa word ratu (W). Ratu begins as kone, as rational thought centered in the forehead. Under certain circumstances — let us say that an exchange partner has refused to be reciprocal — kone undergoes a transformation. Thought becomes worried, 45 d i f f i c u l t and unmanageable, in other words emotional. When asked to ex-plain how thought modifies i t s e l f and becomes ratu, an informant (Ropa Parea) offered a hypothetical example. A man comes to my house in order to secure a pearlshell. He comes saying, "Give me one pearlshell," and I give i t to him. He takes i t and goes, and then he holds onto i t and does not give me something back. If, after the two days he has asked me to wait, he has s t i l l not given i t back to me, then there i s something in my intestines and 1 iver. I think to myself that I w i l l not give something to him another time, for he came, talked, and took the shell away. Or a man w i l l t e l l me, "Give me this thing of yours.and when the sow I keep has piglets I w i l l give you one of them in re-turn. If you give me three or four pounds ($6-8) that you hold, I w i l l give you a piglet later on." But when the sow carries piglets he does not think of me. In a situation like this i t does not help to be aware that the other's kone is bad. For one's own thoughts turn to the lost pearlshell and what might have been done with i t or with the piglet. The thought be-comes frustration. Kewa express this transformation by saying that the thought descends from the forehead to the l i v e r , stomach, or intestines, the place of the emotions. So while frustration or anger (ratu) originates as kone, i t ends up as something rather different. When they are frustrated men say that they "cook" their anger (as in an earth oven of hot stones and banana leaves, na ratu yawalo). Just as one covers up taro and bananas with banana leaves so they w i l l cook slowly, sometimes a l l day, between 46 the hot stones, so the frustration is slowly cooked; just as one sits i n -active and waits for the food to be done, so one broods passively, nursing one's anger. Ratu emerges whenever reciprocities are u n f i l f i l l e d and o b l i -gations ignored. For example a youth who does not help his father to collect firewood, to find game, or to carry trade pigs or bamboos of tree o i l , may yet expect to be given pork when the father k i l l s a pig. Such a youth is the source of his father's frustration. On the one hand the pork cannot be refused: "he is my son, so how can I deny him? And what is the importance of pork: i t is just something to be eaten." But on the other hand the pork is undeserved. There is a dilemma, the unful-f i l l e d obligations of son to father conflict with the values of recipro-city and sharing within the repa. So the father shares the pork with i n -difference, hands i t over with eyes averted. Ratu is to be taken seriously. If a moral balance between dis-putants is not restored through some form of public pressure (debates or courts), violence or sickness may occur. A violent act i s an immoral one, at least where reciprocities should be entertained. But Kewa recognize that when reason is thwarted a person may act provocatively, "from the l i v e r . " Reason says, for ex-ample, that those who are morally enjoined to cooperate in fact do so. Reason may also accept the fact that others may not always respect these obligations. What i s d i f f i c u l t to accept reasonably is this: i f the other i s allowed to remain unobliged, what reason is there for me to be obliged? Ratu emerges against the other because he questions one's own 47 observance of moralities. To i l l u s t r a t e : Nabayo and Punia were among those who shared kabereke pork (cf. Chapter 5) and thus contracted themselves to work on the construction of long houses for a pig k i l l . Punia did his part but Nabayo was lazy and never contributed any work. I saw Punia several months after he and Nabayo had eaten the pork; Punia was complaining about Nabayo, saying that he would take him to court. I asked him i f he f e l t i t was that important; why not just write off Nabayo as bad luck? Nabayo after a l l was known to be an indolent person. No, i t had to be pursued. Punia would not let i t drop; were he to do so "trouble" would come. Punia knew he would feel frustrated and would be prone to violence. Maybe he would attack Nabayo with an ax, he said. He thought i t therefore imperative that the issue be pursued. Rather than bringing balance between disputants, violence i s likely to widen the distance between them. Abusive argument may have the same effect. In Kewa culture there are no abstract ethics whose rules are i n t r i n s i c a l l y superior to personal or factional interests. One cannot be always "above" the other's impropriety, or turn the other cheek to i t , and yet have the material means to be reciprocal. The problem for morality is to adjust self-interest to other-interest espe-c i a l l y in those situations where the other is a disputant. If disputants are to reach equivalence there must be an admission that responsibility to the self is contingent on responsibility to the other. Kewa make this adjustment by acknowledging that kone (and here they mean "bad" kone) can make people sick. A person may " h i t " or "strike" another with his thoughts, much as he could strike the other 48 with an ax. (The two expressions are the same: raimi l i a , koneme l i a . ) There is an important difference between the two modes of attack, though. In the ax attack the actor is held accountable for his actions. The attack may have been ill-advised, but presumably i t was premeditated (except when there i s ema, madness). In the kone attack, on the con-trary, the person i s not responsible for the fact that he "puts" bad kone. He cannot help i t — i t is the situation i t s e l f , or the person against whom the kone is directed, that is responsible. One sees here the possibility of a moral solution to disputes. A person who wants his pearlshell returned i s well aware that striking back violently i s out of proportion to the offence and w i l l therefore re-quire i t s own compensation. But the same person cannot be held respon-sible for the fact that his kone involuntarily causes an il l n e s s in the debtor. Further on I shall underline the fact that a person cannot i n -tentionally attack another with his kone. Person A cannot be certain that Person B w i l l f a l l i l l just because he is frustrated with B, for the unintended result of frustration can also be his own (A's) i l l n e s s . For example: Tema went to the Sugu River and caught fourteen catfish. He gave four of them to the brother of the woman he was intending to marry, some to his future father-in-law, and some to a sister's husband named K. The rest he carried back home. He came back with a sore on his ankle where a sharp twig had cut through the skin while he was walking along the path. K had said he would give Tema money for the rest of the fish; he 49 told Tema to leave the fish in a house nearby and then to take one pearlshell from his own (K's) house in Porapala ground. Tema l e f t the fis h , but at Porapala K stalled and then told Tema the shell was in his wife's house on Iapi ground. But s t i l l Tema did not manage to secure the she l l . Then K said the shell was in the trade store owned by Komiti (government committee man), but the young man who ran Komiti's store said there was no pearlshell there. K later said that the shell was in his strong box, the key to which was held by K's father-in-law Malanaki, but Malanaki said that he did not have the key. By now weeks had passed. Because of this bad behaviour Tema had become annoyed. He had no pearlshell, he had no fis h , indeed a l l he had to show was a sore which, because of his frustration, would not heal. In this instance "causing" an illness means aggravating or pro-longing some di s a b i l i t y rather than originating i t . People suffer from tropical sores, attacks of chronic malaria, mild maladies of the digestive and respiratory systems, but these occurrences do not in themselves arouse suspicions. Only when the dis a b i l i t y lingers on does one suspect the i n -fluence of another person: sorcery,kone or remo attack. In the preceding example Tema explained the sore's refusal to heal as resulting from K's evasiveness. So far I have considered only the injured party's point of view. I have said that his anger and frustration can, as forms of "bad" kone, cause either the offender's il l n e s s or his own. As for the offender, his kone has already affected the other person adversely through his actions; however he, too, may f a l l victim to his own kone in the form of gui l t . 50 "Guilt" is a translation for the Kewa notion of kone rogo, which 2 l i t e r a l l y means "thought that i s bound up." The idea of confession is implied. When a person has engaged in some non-reciprocal act, he may feel guilt i f the act is undiscovered and unconfessed. His thoughts re-main "tied up" unt i l he has "told a l l " (apo l a ) . Kone rogo thus arises where acts are not witnessed and judged by others. I l l i c i t sexual en-counters are the most serious offence of this type. Here i t is the adul-terous relationship rather than the casual a f f a i r which causes guilt. The adulterous person experiences anxiety and eventually, when some injury happens to him (or to his child), he (or the child) does not recover without confession. According to his fight partners, this i s what happened to Poiolo Kata. He was wounded in the buttock while fighting as an a l l y of another ruru. The wound was not serious and he would have recovered (people think) were i t not for the fact that while he was being sequestered in a house he "pulled" (yola) a woman and had intercourse with her. He did not d i -vulge this fact to others. He kept his thought and actions "tied up," knotted behind his sealed l i p s . Though the wound was the cause of the death (and warfare compensation required), the interval between the wound and the actual death was such that people say that he really died from kone rogo. Symptomatic of kone rogo are the clenched teeth of the unconscious invalid. Attempts to force the jaws apart f a i l . I shall offer another example in the next section. 51 2. Kone and remo. In Chapter 1 I said that the dead retain the memory of grievances they harboured as l i v i n g men: motivated thus, they may attack the l i v i n g . Here I have said that the bad kone of the li v i n g is also thought to produce illnesses or turn mishaps into serious af f a i r s . These two. explanations of sickness may or may not coincide. Misfortunes which affect an entire community or repa (such as failure of the gardens, death of many pigs or people) are caused by ances-t r a l ghosts who are angry with the l i v i n g . The community response is to build a cult house and make sacrifices. On the other hand, misfortunes which affect a single person are said to be caused either by remo or by 3 the kone of the l i v i n g , both explanations meaning much the same thing. The reason i s that a person's kone can affect the other (or himself) either directly as kone or indirectly through the mediation of remo. What the remo does i s attack not out of i t s own volition but in response to the kone of the l i v i n g . This seems to contradict my earlier remarks about remo as s e l f -willed. In reality there is no contradiction. The causes and interests of the dead coincide with those of the l i v i n g . Sons retain the memory of grievances suffered by their father, and old resentments re-emerge in new contexts. It is possible for ghosts to be both self-willed agents and passive instruments of the kone of the l i v i n g . An example w i l l i l l u s t r a t e the correspondence. There are two all i e d repa, Pari and Ropa. Let us suppose that Ropa was engaged in skirmishes with a hostile group. Ropa contracted several Pari men to help in the fight and one of the latter lost his l i f e . Some compensation 52 was given, but not enough. Ropa men think that sufficient wealth has been given and they also say they have no more to give. Pari men do not press their claims and several years go by.... A Ropa man f a l l s i l l and his brothers seek out the cause. Several p o s s i b i l i t i e s are suggested. Perhaps a divination is made, and they decide that a Pari fight ghost (yada remo) has attacked the Ropa man. The ghost is the Pari man who was k i l l e d in the fight; he has attacked the Ropa man because Ropa did not complete the compensation. The Ropa brothers make an additional payment to the Pari group. In this example the sickness was attributed to a ghost and to a past event, but i t is reasonable to assume that the outstanding debt was very much an active concern of the l i v i n g . Indeed the attribution of an illness to this cause rather than another reveals the contemporary concerns of the groups. Had the circumstances been somewhat different, Pari's concern i t s e l f (that i s , the kone) might have been held respon-sible for the i l l n e s s . Kewa affirm, then, that certain (but not a l l ) non-reciprocal states may be diagnosed as the act of ghosts or of bad kone. Attack of a ghost might be regarded as a stronger expression, attack of kone as an attenuated version, of a similar reality. The actual expression is that the remo may "steal the thought of the l i v i n g man." (Pa p i r i ana kone pake muara....) The verb pake  mealo, "steal or appropriate by unilateral act," reveals that the intent to harm is avowedly that of the ghost, and that the thinker is not held responsible for the sickness (in the way he would be responsible i f he 5 3 used sorcery). Indeed the fault and the amoral act is the ghost's. The li v i n g man cannot be reproached i f his thoughts are stolen from him. One can surmise that where a sickness i s divined to arise from the remo of another repa, the two repa entertain reciprocal relations with each other. On the other hand, an enemy repa, i f i t is discovered to be the source of an i l l n e s s , w i l l be accused of practicing some kind of sorcery. Ghosts usually attack those who default in exchange; indeed, failure to compensate a kinsman or unrelated exchange partner is the most common reason for ghost-caused i l l n e s s . Rome debts, warfare compensations, matrilateral compensations — any of these may be defaulted, with the re-sult that either the aggrieved party f a l l s sick himself (or i f already sick, does not improve) or the guilty party f a l l s i l l . When i t is the angered person who becomes i l l , Kewa say either that the kone i t s e l f i s the cause of the illness or that a ghost has 4 f e l t "sorrow" (oda) for the man and sent sickness to him. When i t is the offender who f a l l s i l l , his sickness can also be explained as resulting either from the injured party's kone or from the ancestral ghosts. Divinations discover the causes of i l l n e s s . When the sickness is found to result from some u n f i l f i l l e d compensation, the cure involves both making the necessary compensation and controlling the ghost which caused the sickness. The invalid gives the required amount of valuables to a representative of the repa whose ghost was found present. Then the diviner-healer (yainya kupaa, or "sickness blood man") leads the ghost out through an arch of grass stalks. He draws the ghost away from the 54 sick person by means of spells which urge the ghost to leave because com-pensation has now been made. The recipient of the compensation w i l l also c a l l upon his dead brother or father to leave the person alone, to 5 follow behind him as he leaves with his compensation. Whenever a person f a l l s i l l people begin to look around them for possible explanations. Even a small indisposition such as a skin ulcer i s sufficient to keep a man indoors away from his gardens and from other men. No longer free to pursue his projects — discussions, ex-changes, v i s i t s , or indeed any form of social practice — he leads an un-productive existence. He i s constrained to relinquish those reciproci-ties which make him a moral person. An immoral state in i t s e l f , sickness necessarily results from some immoral act. Airing suspicions of others or affirming their integrity, men find i n sickness an opportunity to assess the moral fibre of the community. The example which follows de-picts how a sickness becomes the object of multiple interpretations, each of which reflects different allegiances and oppositions. Sokele stayed until n i g h t f a l l talking with one of his wife's classificatory brothers, Parea, whose house was about a hun-dred yards from his own. Parea had cooked an evening meal in the earth oven and had invited Sokele to share i t . Before leaving, Sokele put some of this food in a tin to carry back to his wife, Ropa. He made his way back through the darkness. Upon arriving at his house he saw that the f i r e was alight but saw no wife. He went outside on a path to the garden, turned, and saw her beside the door of the house. She asked him angrily where he had been. He replied, and they exchanged some angry words. As he was about to enter the house Ropa h i t 55 him across the forehead with one of the doorway sticks and then ran off. Sokele followed and struck her. At this point Ropa collapsed in a kind of faint which lasted from that night, about nine o'clock, unt i l the following afternoon; she seemed to be unconscious but kept trembling, whimpering sli g h t l y , and her jaw was clenched shut. When morning broke she was s t i l l unchanged, and people ac-tively speculated about the causes. The most recurrent and plausible explanation was that some ghosts of Poiolo group had struck her. The reasoning was: the ghosts had attacked her because Sokele refused to work on the yaeada longhouses currently being constructed by the Kareva ruru-cluster (which includes Poiolo group) under the leadership of Sokele's close relatives, Poiolo Robo and Talapapu. Because of the way his parents' marriage linked two groups presently part of the Kareva ruru-cluster (Poiolo and Abasipa) Sokele is related to these yae leaders both as mother's brother/ sister's son (awa) and as father/son (ara/si), so i t would be most plausible for him to lend his support to these men. Poiolo men have repeatedly asked Sokele to s i t down and eat with them when the yaeada feasts were cooked; as a sister's son might do, he could show up and share food without obligations. But Sokele refused to lend his support and, out of discretion did not go to the yaeada si t e at a l l . It was therefore thought that perhaps these yae thoughts (yae kone) of resentment had been taken up by some Poiolo ghosts and used to strike Sokele's wife. Tero, Sokele's brother, considered that i t was bad form for Sokele not to work with the Kareva man: "When he dies i t w i l l be his wife who buries him and not his brothers!" At the Figure 3-1 Participants in a Dispute P010LO la la-papa Robo # Al o A . Akitu. E p e l e t a ' * ^aeada L . ABAS IPA ROPA Olanea fao.. Sokele T Tero Q R o p a Robo .yaeada KEY: r—»—i P0IOL0 M o © c l a s s i ft* c a t o r y s i bl i ng t i e Ruru name Persona l name Deceased Yaeada l e a d e r O N same time Tero thought that i t would not have been his own (Tero's) kone which could cause such an illness but those of older men, such as Alo or Talapapu. In Tero's opinion, the fact that the yae kone struck Sokele's wife and did not strike Sokele indicated perhaps that Ropa herself was somehow res-ponsible for Sokele's indecision with respect to the yaeada. The second possibility was that Ropa's sickness came from a dispute between Sokele and Robo, Ropa's true brother. For Robo, too, was a big-man engaged in the construction of a different yaeada for the next pig k i l l . It would be quite understandable for Robo to have his sister's husband, Sokele, working with him. (It was perhaps because Sokele had strong commitments to the two r i v a l factions — to the f i r s t through mother and father, and to the second through his wife — that he avoided making any commitment.) This explanation was that Robo's bad kone with Sokele might have struck Robo's s i s t e r . Mention of Robo's yaeada led to another line of reasoning. Sokele's adoptive father Yapi had been contracted to work for Robo by means of the g i f t of a pig. Yapi is Robo's suba, s i s -ter's husband's father. Sokele had not shared in the pig. There was, for a while, the poss i b i l i t y that Sokele would par-take, but he reneged. Was Yapi's kone possibly to blame? This was present in some people's minds as something to be taken into consideration; Yapi stressed that he had not put any pressure on Sokele to accept the pig, and that since then he and Sokele had gotten along well. Once (Yapi related) when Sokele and his wife did not come back from fishing i n the river, he himself gave food to Sokele's pig, and later Sokele returned the favour by bringing some firewood when he, Yapi, had none. Thus there was evidence for a l l to see that a posi-tive reciprocal relation prevailed between Yapi and Sokele. 58 Yapi no doubt saw the ambiguity of Sokele's position and of his own with respect to the competing yaeada groups. Another possibility mentioned was that Ropa had herself com-mitted some kind of social transgression and that her own thoughts had become kone rogo; her clenched jaw supported this theory. Perhaps she had been having some adulterous af f a i r . Someone else remarked that a certain diviner and spell ex-pert converted to Christianity had previously told her that i f she did not come to worship like other married women then sickness would be f a l l her. (Sokele, on the other hand, goes to the village worships regularly.) The same "Kristen man" had previously predicted that a certain young man who never came inside the mission house during worship would get into trouble, and that person subsequently went to j a i l for an aff a i r with an unmarried g i r l . Finally, Tero remarked that the woman Ropa behaves like a man: she is a strong, self-willed person who fights with her hands and her wits. Tero went so far as to suggest that her illness was a fake. While others maintained that a ghost must have been standing in back of her, judging from the way she lashed out at her husband, Tero hazarded that she was actually a man, capable of overcoming her brother Robo in physical strength and action. (For "her brother" Tero used the term ame, or same-sex sibling male speaking, rather than b a l i , cross-sex sibling female speaking.) Ropa's illness subsided after the second day, but while i t lasted i t gave rise to many interpretations, a l l possible, only one of them basi-7 cally incompatible with the others. Persons consider the possi b i l i t y that 59 they w i l l be implicated, and they bring their opinions into the open. If one's kone is exposed to view and found consistent with moral t i e s , then the cause must l i e elsewhere. 3. Other emotions. Ratu can be included with other forms of non-reciprocal intentions and conduct in the general category of "bad kone." Cupidity and self-centred behaviour would be other forms of bad kone. If ratu sometimes has i t s way in the Kewa moral system, i t is because i t may be an inevitable, and therefore an excusable, response to another's i n i t i a t i v e . Yet ratu is potentially the most destructive of the emotions, the one in which man reveals his non-reciprocal being most conspicuously. Sorrow and joy are, like anger, modifications of the normal, "practical" interest in the world. They diff e r from anger in being less harmful: they are not linked to remo attack. Yet they, too, contain an element of non-reciprocity and for this reason their presence among men is a moral problem. I shall consider the feeling of sorrow or sympathy (odo W, yara S) that a departing person may cause to those he has lived with. Let us say that the departing person came as an immigrant; now he is returning to l i v e with his natal group after having resided with his hosts for a number of years. Prior to his departure the migrant receives expressions of sorrow from his hosts. Either these expressions are direct statements that village people w i l l experience sorrow ("I w i l l be sorry," odoma omalua, yara komaloa) or the expressions may be about how they w i l l experience sorrow, how they w i l l find i t d i f f i c u l t to orient themselves 60 normally for a while. Their thoughts w i l l be with the departed person; they w i l l not be able to look at the empty house and may want to burn i t down; they w i l l not go outside but s i t indoors by a dead f i r e ; they w i l l not sleep well at nights, and so forth. This can be told in speech, perhaps with some exaggerated s e l f -debasement: "I w i l l eat my penis/vulva"; or i t can be communicated through some kind of strong metaphorical form, such as a rupale sung in a dirge-like cadence by a single individual or a group. For example: Ainya kana napu i n i madu tadelena, yeki tipa wanesima ama kea. Brother, many of your Kana ground casuarina seeds have car-ried and fallen striking; a l i t t l e Yakita daughter sweeps them in the ceremonial place. (The seeds which f a l l and strike the ground evoke wealth objects. The singer, a young woman, praises a "brother's" generosity and hints that com-pensation should be given her.) The proper response i s to make a compensation to that person, the quantity varying from a couple of shillings to several dollars or a 8 pearlshell depending on the closeness of the relation. If compensations are not forthcoming they can be solicited by means of yano (or yago, re-placement object) g i f t s , tendered with expressions of sympathy. Compensation is due because the emigrant causes sorrow in the other person the way one can cause an injury or an i l l n e s s . Sorrow is_ an i l l n e s s , for illnesses in Kewa do not concern the organic parts of the self only, not just the body or flesh (to), but involve the emotions and rationality as well. Indeed the expression oma (or koma) whose meaning 61 without specification is usually "die" or "be sick," (as one can say "oroma omalo," I am sick with a cold, I have a cold) also means "to emote" (as in odoma omala, yara komalo, I am sick with sorrow, I am sorry; and rana komalo, I am happy). Into this world coloured by sorrow-sickness, then, a compensation enters and causes joy (rana, pedo). Happiness, too, must be compensated. When, for instance, un-married g i r l s praise the way a young man stands in a dance or has decor-ated his face i t i s customary for him to make a small compensation. When, again, a small g i r l , who has been given a new grass skirt and perhaps a string of beads by her mother, receives praise from an adult, some compen-sation must be given. Praising one's sister's son or patrilateral cross-cousin, especially his thighs, his calves or his stature, can in the pro-per context (such as rupale, cf. Chapter 12) be a way of s o l i c i t i n g a maternal payment from him. (By referring to his body one draws attention to the consanguineal t i e between a cross-cousin and the mother's group which bore him.) The emotions of sorrow and happiness are therefore not dissimilar from that of frustration; a l l three may lead to the self being indisposed. (But I have not heard of sorrow and happiness causing indisposition in the other, as i s the case with anger and resentment.) In each case another person's acts (or kone) deprive one of one's moral orientation to the world. Once again, the "moral" in the Kewa context is the "normal" and "practical." A person burdened by emotion does not go to his garden, s i t down with his brother-in-law, think of his exchange obligations. Emo-tions loosen the strings tying him to this practical world. 62 This is a context which compensation can modify: the receipt of a compensatory g i f t reawakens interest in the world of things, pro-jects, exchanges. Thought turns away from the emotional situation to the wealth object received, and to the various ways i t might be employed. Apprised of the other's obligation to him by means of the g i f t , the per-son rediscovers his own obligations. 4. Injury and compensation. Injuries caused to another's body are v i s i b l e , and they cause discomfort or pain. To outsiders such as ourselves, the obligation to compensate them is therefore more vis i b l e than in the previous examples. If I have not discussed them f i r s t i t is because kone and ratu underlie their self-evidence. The term for injury compensation is ayako (W). Ayako may be given for anything from a small injury to a death, whether i t be (1) an injury or death sustained from a human agent, such as warfare, or (2) an injury or death sustained where a human cause is not manifest (e.g., "accidents"). In either case the terms of the compensation are arranged between the injured party, or his close relatives, and the r_e, a term which we can render as "origin", "cause" or "base." In warfare ayako the re man or men are generally the a l l i e s for whom one has fought, not the enemies. Enemies, at least the long-standing ones, remain non-reciprocal; I have not observed the Kewa to 9 compensate for the war dead of hostile groups. But a l l i e s have an ob-ligation to indemnify each other. This is because only one of the sev-eral a l l i e d groups who fight together is responsible for the war. 63 (Fighting erupts between groups of brothers or repa, but rarely is i t con-tained within the two groups originally involved. Each group enlists the aid of others by means of a "fight kabereke" system (similar to the yae  kabereke contracts used in the construction of the yaeada): the "fight cause men" (yada re aa) give a pig to a group of brothers to engage their support. If an ally is injured or dies in the fight, the "fight cause men" must make compensation.) Non-mortal injuries are compensated quickly, for the aim is to keep the person from dying. It i s through quick compensation that the injured person is spared the frustration and grief of having risked him-self with unfortunate consequences. For, as I have said, even a minor injury can become serious through frustration, worry and self pity. Ac-tual deaths incurred in war are another matter; here the ayako payment may take longer, occurring well after the conclusion of fighting. (They occur sporadically today in the grasslands area, where fighting ceased about a decade ago.) Death payments are large; they involve a quantity of shells and pigs which depends on the status of the victim, on the number of his brothers, and on the wealth of the re group. The f i r s t step in compen-sation is discussion and mutual agreement between the two groups that ayako w i l l be performed. Then the group of the dead man request compen-sation from the "fight cause" group by means of a solicitory g i f t (yano) of pigs, shells, tree-oil, and the l i k e . The "fight cause" group shares the valuables out. Eventually (several years may go by) these men put together a payment of valuables which should be twice the amount they 64 received. For every solicitory g i f t pig (for instance) there must be one pig to equalize (abula) i t and one pig to compensate for the death. The latter is called the awanopu pig, perhaps from aa (man) + wanopu (a variety of rattan cane), evoking the bindings of a corpse. The num-ber of pigs must be judged by their relative sizes, too, and this inevit-ably is the topic of long discussion. In the second situation requiring ayako there is no human agent manifestly responsible. Here is one example. Kareva _rep_a-cluster was constructing yaeada for a forthcoming pig k i l l . One Kareva youth injured himself while working. He had been cutting down a t a l l black palm when the ax glanced off the trunk and struck his foot at the instep, cutting i t quite badly. The mishap occurred on the day of a large feast-ing at Epeleta when many men, after working on the houses a l l morning, had gathered there for a meal. The boy was placed on a make-shift stretcher and was carried to the house site. Fortunately a government medical assistant happened to be in the village treating an epidemic of tropical ulcers, so there was l i t t l e doubt about what should be done for treatment. While men and boys crowded around the stretcher and talked about the event, one of the yaeada "base" men or leaders gave a pearlshell to the owner of the tree. Purchase of the tree had been arranged beforehand, but lest the non-payment compli-cate matters even more, that transaction was accomplished. A second leader gave a pearlshell to the unfortunate boy, who accepted i t indifferently and gave i t to his father. Other base men promised shells. When a l l this was finalized there was a lot of loud vocalizing (u la) which signified a return to normalcy. The boy, i t was said, now would not sicken. 65 Men admitted that i t was an inexperienced youth who had the accident; a grown man, more s k i l l e d , would not have l e t the ax s l i p . Yet mishaps of this type are not caused by inexperience so much as by temporary inadvertence, and ghosts are known to d i s t r a c t people thus. So the accident might well have betrayed some bad yae kone between the yae base men and the injured boy's group. But i t d i d not r e a l l y matter; the youth was working f o r the yae leaders and i t was up to them to respond. I t i s one of the marks of an important man, such as yae base men always are, to compensate quickly without complaint; by following a good s t y l e of behaviour, by having good kone, he gains immunity from the bad thoughts of others. Although the unfortunate boy did not seem at a l l solaced by the s h e l l s he received, the yodeling at the end of the business s i g n i f i e d that, as f a r as the community was concerned, the matter was dismissed. I f the boy got worse afterwards, one would have to think that the blame lay elsewhere, perhaps i n some kone rogo of the boy's. In another instance Ropa Parea gashed h i s hand while c u t t i n g cane for a bridge over the Erave River. Vine bridges were t r a d i t i o n a l l y b u i l t f o r the purpose of trade and t r a v e l , although today i t i s sometimes only a f t e r pressure from the government that they are r e b u i l t . Since men b u i l d bridges for t h e i r own communal use, there was no j^e and therefore no p o s s i b l i t y of compensation. But S a i Puluma, who had been working i n the f o r e s t with Parea, gave Parea a one d o l l a r note because he sympathized (odoma oma) with him. The appro-p r i a t e response for Parea here was to return t h i s amount, which was a yano, with an a d d i t i o n a l amount "given for the 66 body" (tona gia) because injury to i t has caused Puluma his sorrow. So Parea returned the dollar note with an additional five s h i l l i n g s . In this example the compensation occurs in a manner different from that of the previous case: i t i s the injured party who has to com-pensate the other for feeling sorry for him. The f i r s t prestation (the $1 note) communicates the sorrow of the other party, while the return pres-tation compensates i t . In the two examples there i s , however, a certain similarity: the yae leader is the cause of injury to the boy, and Parea is the cause of sorrow to Puluma. Compensation must go from the f i r s t to the second in either pair. It is evident that the two structures a r t i -culate, since the injured person can be simultaneously compensator and recipient of compensation. Thus even a minor injury may involve several persons who exchange small quantities of wealth. Figure 3-2 integrates the two ayako situations described above. It shows how in each case com-pensation is given to the physically or emotional affected party. Figure 3-2 Compensation Person A Causes Injury mjury compensation Person B Receives Injury, Causes Sorrow sorrow compensation Person C Experi-ences Sorrow 5. Conclusions. My intention in this chapter was to bring to-gether certain aspects of Kewa being in order to demonstrate the presence of a moral system. My conclusions aim to expose the dia l e c t i c a l a t t r i -butes of this system, f i r s t by underscoring the existential content of 67 these categories, and second by pointing to the p r a c t i c a l awareness and the manipulative use of the system. Allowing that s p e c i f i c thoughts can be e i t h e r bad or good, kone i s moral. I t stands for c o n t r o l , reason, and r e s p o n s i b i l i t y , and opposes to various "animal" aspects of man's being: J^o, lo-pu, and remo. The fundamental d i f f e r e n c e between kone and the other categories i s that kone i s awareness, whereas the others are objects of awareness. To i s ex-perienced through need, sickness, fatigue, and awareness of m o r t a l i t y . Pu e x i s t s through the emotions of anger, sorrow, joy, etc., while the presence of remo i s perceived through the experience of the powerful and the non-reciprocal. I t i s not the categories themselves but t h e i r i n t e r a c t i o n which i s the basis of the moral order. Suppose that there i s some r e l a t i o n be-tween persons A and B (Figure 3-3) and that A's kone had adversely a f f e c t e d B. B's kone turns bad, becomes f r u s t r a t e d ( c f . the double arrow between kone and lo-pu i n the diagram). Then A i s adversely a f f e c t e d i n h i s to, Figure 3-3 The Moral System to remo kone A l o , pu by i l l n e s s . A seeks an explanation with the help of d i v i n e r s , brothers, and other members of the community; he sees B as a possible reason for h i s 6 8 own plight. In his understanding A unifies the two non-reciprocal s i t u -ations: he infers the presence of a retributive remo. This awareness leads to a modification of his original action. Hence there is a kind of "feedback" cycle which maintains reciprocities. I have diagrammed the Kewa moral order as system-like structure (Figure 3-3). This raises a question of i t s i n t e l l i g i b i l i t y : what main-tains the structure in place? The other structures we w i l l be consider-ing — the tawa structure, kin and repa structures, p i g - k i l l structure, myth structure — maintain themselves because they rest on the moral order. On what does the moral order rest? The question is not an idle one, but i t risks taking us into psychology. Several suggestions offer themselves. The f i r s t i s that 10 kone depends on language, which is a form of interchange or exchange. The way language requires reciprocity i s most apparent i n the case of kone rogo. If kone rogo i s regarded as anti-social and harmful, a p a r t i a l reason i s that unconfessed thoughts cannot be reviewed and modified by others. That is how one can understand the Kewa " o b j e c t i f i -cation" of thought, the attribution of independent existence, as "things," to contingent relations. Once, for reasons of shame and guil t , thoughts are contained and thereby immobilized, they take on the appearance of 11 inert things. That, at any rate, must be the case i f one is to believe that kone have an instrumental existence for ghosts. Much the same occurs when there is ratu, except that whereas r kone rogo persists through the ignorance of others, ratu thrives on open conflict of interests. The question i s : under what conditions is i t pos-69 sible to believe that one's anger may cause sickness in the other or in oneself? Clearly, thought which descends from forehead to l i v e r , in other words becomes emotion, undergoes a physiological and affective modification. Less clear is the way this modification transforms aware-ness. Frustration i s an anxiety which arises when a l l the means to attaining some v i t a l end are barred. In the Kewa examples, a debtor does not respond to reason, or pleads i n a b i l i t y to compensate, or de-ceives; yet there can be no recourse to violence or to sorcery. The ina b i l i t y to conceive of some form of action engenders a sort of "para-l y s i s " of reason. Because there can be no practical or rational means of action, an irrational one i s chosen; the int e l l e c t gives way to affec-t i v i t y in order to make the conflict and the inactivity endurable. Affec-t i v i t y can, in its own way, dispose of the conflict i t s e l f both through the physiological phenomena associated anxiety (cf. Sartre 1948: 55ff; Levi-Strauss 1971: 588) and through the active emotional transformation of the self (or of the other) through contempt and hatred. The s e l f -contemptuous person sees himself as unworthy or undeserving, saying "The other does not return my loan because I do not count for much as far as he i s concerned." The resentful person, similarly, may magically trans-form the other into a person from whom non-reciprocal behaviour would be normal and expected; he i s given a l l kinds of despicable qualities (tattered clothing, dirty hair, thin beard, crooked teeth, no friends). One says, "He does not return my loan because he i s not a reciprocal or moral person." In both instances there is an emotional modification of the social world which makes the original conflict more livable. 70 A parallel emotional modification of the world takes place when there i s sorrow. It is because Kewa draw no boundary between sorrow and illness — for both are characterized by numbness, apathy, and inac-t i v i t y — that they desire a compensation from the departing person. Be-cause the other person has created for them a temporarily uninteresting and colourless world, they expect to be given an object whose acquisi-12 tion returns their interest in reciprocal a c t i v i t i e s with other men. My second concluding remark bears on the way Kewa may manipulate the structure. I have been ignoring a complicating factor: thought may attend to the experiencing of emotion, i.e., a person may be aware of his emotions. In so doing he may see the emotion in a larger context, he may transcend the immediacy and singularity of the experience. Be-cause an emotion i s regularly followed by compensation, consciousness of the emotion is simultaneously consciousness of the compensation which w i l l follow. Since the compensation is expected, i t s absence w i l l also produce frustration, resentment, or some kind of bad thoughts. In general, then, a person compensates another to whom he caused injury or sorrow in order to avoid the harm caused by the other; but to the extent that a com-pensation is structured and expected, the harm that one avoids by compen-sating arises less from the injury i t s e l f than from the resentment that would be caused by not compensating. For instance, when a departing per-son compensates those who say they are grieving for him, he does so mainly so that they w i l l not incur the ratu feelings which would arise i f he did not compensate their expressions of grief. Likewise when a young man distributes compensation to unmarried gi r l s who praise the way he 71 looks in a dance, he does so to avoid a situation in which they might find their expressions of joy uncompensated. This explains the fact that expressions of emotion, such as joy or grief, may be calculated; they are often strategems to secure a small quantity of wealth from the other who finds himself in a vulnerable situation. Mourning over death is often quite matter of fact (but of course may be also very sincere), expressions of admiration or sadness may be openly a pretense. (Since, after a l l , the compensation is almost inevitable, shows of sadness need not even be convincing.) They are, in brief, "false" emotions to the extent that they are acted out and sus-tained by the thought of the desired object. Although i t would be just as wrong to maintain that ah emotional display (say, over a departing friend) i s totally unfelt and contrived as to maintain i t was absolutely sincere, affective states among the Kewa are supported and modified by the expectation that they w i l l entail an exchange of wealth. In many situations i t is the expected exchange of wealth which shapes the emotional  content of the situation, just as simultaneously the situation evokes an emotion suggestive of exchange. One could therefore re-examine some of the examples cited in the discussion and find how a "secondary" emotion — the expectation of compensation — combines with the original, disinterested one to form the real interior relations between the persons. This is the second way in which ideas about kone support the moral view: not only does the kone-ratu-remo link oblige the other to be reciprocal, i t also returns the 72 emotional to the analytical by supplying a practical level of interac-tion for an affective one. It i s at this practical level — where emotions have, as i t were, "exchange values" — that the moral order maintains i t s e l f intact. Because they are open to compensation, emotions are always mediated, tem-pered with a reflexive awareness of themselves qua emotions. In this way the very emotions that are in some respects destructive of the re-ciprocal order become the means by which reciprocities are renewed. 73 Footnotes 1. Kone and ratu parallel the Medlpa notions noman and popokl, which are discussed at length in Marilyn Strathern (1968) . I read this paper for the f i r s t time only after returning from the f i e l d and therefore did not follow up, in their Kewa context, some of the i n -teresting ideas Strathern suggests. Also pertinent i s the accompany-ing paper by A. Strathern (1968b) . 2. From the verb roga, which means to wrap around, to tie up, or to en-close. The immediate associations (for me) are tying up tobacco in a cigarette leaf or wrapping up something, such as sorcery bundle, for concealment. 3. I do not consider sorcery here for the reasons explained in the pre-ceding chapter. I am concerned here with those misfortunes which are thought to be caused by a failure in positive reciprocities. 4. I have heard Kewa men express the second interpretation only in a couple of instances, and I do not ful l y understand i t . It need not be the offending person who i s affected by the ghost: i t may be a wife or child. Where the relationship between injured and offending parties is through some kind of agnatic links however remote (such as within a repa-pair or -cluster) the ghost may attack the offending party directly. Thus an ancestor of an injured Pari man may attack the offending Ropa man, since Pari and Ropa form a repa-pair. More generally, a ghost attacks any cognatic relative: a cross-cousin therefore can be affected. A f f i n a l links and ties between unrelated persons are different, however. A sister's husband who does not f u l -f i l l his a f f i n a l debts to wife's brother does not f a l l i l l from at-tacks by ghosts of his wife's group; i t is the wife's brother himself who becomes frustrated and i l l , and seeks to actively redress the wrong either by persuasion or by urging his married si s t e r to return home. Or i t may be the wife or the child who f a l l s sick through the action of the wife's group's ghosts. In brief, i t seems as though ghost attacks follow lines of f i l i a t i o n . If a person f a l l s i l l from an " a f f i n a l problem," i t is because he has ignored his mother's bro-ther, not his wife's brother. But i t also may be that a ghost may attack any person within the community. 5. He may intone something l i k e this: Brother, you who are here, father, you who are here Here is a shell replacement, here is a pig replacement, The shell from the south is given, the pig from the north, The salt from the Mendi is given. Take the shel l , the pig, and go. 74 Go stay there, at the tops of the casuarinas, at the mano tree, At the mountain ridge... . 6. The yaeada houses are accomodation huts built several years prior to a pig k i l l . The work is organized by sponsors called yae re aa, or "yae base men," by giving pigs to small groups of men. Sokele never wanted to be contracted in this way. A ruru-cluster is a collection of several small ruru or "clans" whose members do not intermarry and consider themselves to be siblings. 7 . Tero's second observation. It was my opinion, too, that there was a certain amount of deception here. If others did not question her authenticity i t was probably because a convincing deception must be based on a certain amount of truth. 8. Large compensations are given not only to offset sorrow but to com-pensate (a) the use of garden land and forest, the aid and support of men, and (b) the goods which were given to the destitute immigrant upon arrival and which involved him from the outset in reciprocal com-munity ties. 9. My discussion of warfare ayako must be considered provisional, for I am working with limited data. 10. This is because both language and exchange are forms of social praxis and share the same practical structures. Language, even when solitary, expresses some project. Most praxis is also language to the extent that projects involve the other person and to the extent that i t is through language that one's own projects are carried to the other per-son and his projects to oneself. When expressed, one's own thoughts take on an existence for others, and therefore an existence for oneself. Not because they are percep-tible through the senses but rather because the thoughts expressed in language are supported by the total l i n g u i s t i c system of s i g n i f i -cations — become part of the other's intentions and projects and are affected (supported, misinterpreted, disregarded, etc.) by them. One can conclude the following: as soon as a thought is given the form of speech i t becomes something external to the speaker and is transformed. It is transformed not only because i t may be misunder-stood or inscribed in a different set of intentions and meanings, but more generally because the meaning of a thought emerges with the speak-ing. Each utterance,in which a thought is turned into a syntagmatic chain of words within which each word modifies each other, modifies the thought by creating a f i e l d of discourse and meanings. As the rela-tionship of meanings within the f i e l d becomes ever more precise, there is a necessary reworking of the thought to further structure the 75 f i e l d (Sartre 1960: 172). But reciprocally i t is always only against this f i e l d that the modified thought emerges in order to further modi-fy the f i e l d , and so forth. It i s for this reason that one can speak of speech as a form of ex-change wherein two persons reciprocally modify each other's in t e r i o r i t y in view of some common end, such as passing on information, deciding on a mode of action, overcoming the other's views. Other modalities of exchange presuppose this relation while going beyond i t . 11. This i s the idea conveyed in Blake's "A Poison Tree": I was angry with my friend: I told my wrath, my wrath did end, I was angry with my foe: I told i t not, my wrath did grow. And I water'd i t in fears Night and morning with my tears; And I sunned i t with smiles And with soft deceitful wiles. And i t grew both day and night, T i l l i t bore an apple bright; And my foe beheld i t shine, And he knew that i t was mine. And into my garden stole When the night had veil'd the pole. In the morning glad I see My foe outstretch'd beneath the tree. 12. Cf. Sartre (1948): Passive sadness i s characterized...by a behavior of oppression; there is muscular resolution, pallor, coldness, at the extremi-ties; one turns toward a corner and remains seated, motionless, offering the least possible surface to the world. One prefers the shade to broad daylight, silence to noise, the solitude of a room to crowds in public places or streets (64). The reason Sartre gives for such a state i s : One of the ordinary conditions of our action having disappeared, the world requires that we act in i t and on i t without that con-dition . Most of the potentialities which throng i t (tasks to_ do, people to_ see, acts of daily l i f e J^ o carry out) have remained 76 the same. Only the means of realizing them...have changed.... In short, i t is a question of making the world an affectively neutral reality, a system in total affective equilibrium.... (Ibid., 64-65). This does not mean that Kewa beliefs about kone can be reduced to existential or psychological structures but only that they are a par-ticular form of consciousness of these structures as they articulate with the totality of the culture (in particular with the importance of exchange and with the existence of ghosts.) 77 CHAPTER 4 THE TAWA The preceding two chapters described the workings of the moral system, i.e., those ideas about mutual influence (or reciprocity) which judge certain kinds of thoughts or acts as good or bad. I did not i n -quire what kinds of acts were good or bad; that cannot be answered without reference to the social structure, the object of Part II. I asked rather what effect a disruptive act or thought had on the person. In this chapter I discuss a form of competitive giving. It w i l l soon become evident how this practice of reciprocity follows from and completes the ideas about reciprocity described above. This chapter articulates with the preceding one in two ways. Fi r s t , there is a basic difference between "sorrow" and "joy" on the one hand and "frustration" on the other. If sorrow and joy can be turned to material advantage, i t is because a reciprocal relation a l -ready exists between the parties involved. They can succeed i n applying subtle manipulations to one another because they are already in a rela-tion of positive-reciprocity. Frustration, on the other hand, arises because reciprocal relations have broken down; i t cannot be used in the same pragmatic way. In other words, an anticipated compensation may colour the emotions of sorrow and joy, such that they are experienced pragmatically or thematically as well as affectively, but ratu cannot expect automatic compensation. If A feels wronged by B, A's resentment of B does not auto-matically entail B's sickness, B's awareness of the wrong, and B's compen-78 sating A. That would be a contradiction since A's resentment arises because he has no means of control over B (other than force or sorcery). Moreover, Chapter 3 noted that ratu can also cause one's own debility (which increases frustration rather than dininishes i t ) and that ratu may equally well persist without any social consequences at a l l . The ratu-remo link discovers the meaning of a present situation in the past; i t is etiological, not prospective. There is a second source of ambiguity. It has been assumed that men have a moral sense, which in the Kewa context means that they accept responsibility for their actions, l i s t e n to their brothers and supporters, consult diviners, in short are subject to community control. Not a l l men share these qualities to the same degree. Some men appear to be guiltless and immune to the opinions of others. Such persons are often big-men (ama), assertive and domineering (puri, "strong"). Ratu does not seem to affect such men, nor do these men — who are men of action — seem to harbour ratu against others. Yet they, too, are i n -volved in disputes, often quite substantial ones (as measured in number of men and magnitude of property involved). For both reasons, then, a person cannot count on a wrong to be righted through the work of kone. He must act. The form which action may take is the tawa. In some respects resembling the Medlpa moka or the potlatch, in other respects closer to the reality of sa c r i f i c e , the tawa is the means whereby each party tries to reduce his opponent to poverty by obligating him to give away ever more property. 79 The structure of the tawa (which means "I w i l l talk") is that of reciprocity: once one party has begun the distribution-destruction of property the other side must accept the challenge and return i t . The two disputants alternately insult and taunt each other, each insult being solemnized by the k i l l i n g of a pig. The f i r s t disputant k i l l s a pig and t e l l s the antagonist to come and eat the pig blood; then he shares out the pork to his friends and supporters, giving none to his opponent. The latter does the same. This w i l l continue un t i l one party can no lon-ger find the pigs (or, today, cows) to k i l l . This chapter, which presents and interprets data on the tawa, continues the argument of the preceding two. It attempts to define the economic, p o l i t i c a l , and most of a l l the moral aspects of the i n s t i t u -tion. My approach w i l l be an hi s t o r i c a l one: I shall investigate in 1 considerable depth one particular large-scale tawa. The f i r s t part of the chapter recounts how the tawa was precipitated and conducted. The second section discusses the p o l i t i c a l , economic, and moral significance of the tawa. 1. Background and conduct. The principals (re) of the tawa that occurred in Iapi d i s t r i c t in 1971 were two classificatory brothers-of Ropa group: PUSA, of the Yekira sub-ruru (sub-clan), and ROBO of the Iapi sub-ruru. But the dispute really began a generation earlier with their fathers, MELEPA and SAPE, who were at odds with each other for rea-sons which remain somewhat unclear. It appears that the two men were participating along with several other Ropa men in the construction of a 80 cult house. Melepa was one of the organizers of the construction of the house; he had originated the idea and mobilized the services of the others. It was up to him to k i l l a pig when the house was finished, and he did this. But no one brought the bunch of bananas which was also required. After asking around in vain, Melepa went and, unseen, cut a bunch belong-ing to Sape. He brought i t back and told some others to cook the bananas along with the pig. When Sape saw the bunch of bananas he noticed that i t was of the same variety as one of his own bearing plants. Becoming suspicious, he went and checked behind his house. Indeed, the clump had been cut. Sape stormed back to the cult house site and began arguing heatedly. Melepa replied calmly, "What are you shouting about? Do you want to buy a pig 2 or what? Yes, I cut your bananas!" The two grappled, then separated; Melepa in disgust abandoned the house to other Ropa men. From then on there would be no reconciliation between them. The two sub-ruru, that of Melepa and that of Sape, remained unfriendly. Melepa symbolized the sp l i t by naming his children Poai ("cut" — because Sape wanted to cut him with an ax), Warakeame ("step aside" — i f they ever wanted help, they could be ignored), Rakeame ("break in two" — the two w i l l not be joined), and Kale ("hidden" — i f they c a l l for help, remain hidden). When internal fighting broke out between the Abasipa group and Sape's sub-ruru, Melepa moved up to Taguere, a neighbouring a l l i e d dis-t r i c t separated by a steep, forested ridge. In refusing to aid Sape, Melepa expressed the seriousness of the antagonism between them, for same-ruru brothers should join to fight with other ruru. 81 Melepa's brother, who had also moved up to Taguere at the same time, died soon after. Then Sape arrived in Taguere seeking sanctuary after being defeated by the Abasipa group. Melepa accused Sape of having had a hand in his brother's death through either sorcery or witchcraft. Taking a spear, he was at the point of running Sape through when he was forcibly restrained. Melepa then approached Sape's enemies, the Abasipa, and suggested that they might join forces to do Sape in , but the Abasipa men declined. When the fighting was over both men returned home with their children. Melepa died having instructed his three sons, Poai, Palea (Kale) and Pusa not to forget who Sape was. The children grew up and married. It was shortly after several of Pusa's pigs had died that the f i r s t real conflict between the second-generation disputants occurred. Sape's son Robo, like his father before him, was an important person. On one occasion he was looking around for help in acquiring a very large and valuable pearlshell (sekere uni "bone pearlshell"), the purchase of which would enhance his reputation. He asked for Pusa's support, but Pusa would not help him. One day when Poai's son T i s i went down to the main path he met Robo who said, "Your pigs have died and you have eaten them; now not one stays in your Yekita bush; you w i l l remain rubbish men, l i k e an e v i l s p i r i t , your mouth smelling, wearing ragged string aprons. As for me, I k i l l my pigs and cook them, and there are s t i l l many more where the p i t p i t clumps grow." T i s i went back and told 82 Figure 4-1 Participants in a Tawa Poai Ti s i YEKITA ROPA IAPI ROPA ROPA this to Pusa and Parea, but the two older men dismissed this; they did not think that Robo had practiced sorcery on their pigs. uni to Robo for bargaining, but Robo told them that he did not have enough pigs or shells to purchase i t with. The men returned to their homes. Shortly after, Pusa went over there with a shell of considerable value (a sekere alu, "head shell") and told them that with this shell he was going in with Robo; he promised, in addition to what Robo was offering, a cassowary, a large pig, and several shells. So the men made another long day's journey and arrived with their sekere uni at Robo's house for the second time. But of course Robo had nothing to give them and, embarrassed, had to t e l l them again they would have to go home empty-handed. Furious with Pusa for having deceived him, Robo dared Pusa to come forth with his cassowary, but Pusa only jeered back at him. Men from the other side of the Erave River brought the sekere 83 Another conflict developed between the two men. For his second wife Pusa had taken a woman of the Wai group, residing about a day's walk from Iapi. On one occasion his wife went home to v i s i t with her parents and then returned with an unmarried sister. The sister stayed with Pusa and later Robo married her. Sometime later she disappeared. Some be-lieved Robo had k i l l e d her — drowned her in the Erave River or thrown her 3 into a karst hole — but Robo always defended himself disbelievingly f