CLANS AND THE JOKING-RELATIONSHIP AMONG THE PLATEAU TONGA OF NORTHERN RHODESIA E. Colson Formalized joking between clans has been reported from many tribes in British Central Africa(1)p Tew has attempted to place the joking be- tween clans within the wider context of the funeral friendship which ; exists in one form or another from the Zambezi River in the south to Lake Tanganyika in the north and from the east coast to the Luangwa river(2)* Radcliffe-Brown has also dealt with the subject in his analysis of the general nature of joking relationships(3),. The Tonga are thus not unique in practising such joking. This paper does not report any new phenomenon, nor shall I attempt within it to develop any new theory of the nature of Joking relationships. I am limit- ing myself to a description of the relationship as it exists among the Plateau Tonga. Despite Tew's suggestion that "funeral friendship" should be adopted as a descriptive term to cover the phenomenon as it occurs in this area, I shall use the old form "clan joking relationship" since among the Tonga the funeral is only one, and perhaps not the most important, situation in which the relationship operates. Indeed, one could cavil at either term, since neither joking nor funeral duties effectively define the relationship. In many situations clans perform reciprocal services for each other. One might therefore call the institution, "clan reci- procity," and those entering into it "clan reciprocals." The Plateau Tonga are a Bantu-speaking people living in what is now the Mazabuk- District of the Southern Province of Northern Rhodesia. To- day they number between 80,000 and 120,000 people. To the east of them live the We or Valley Tonga of the Gwenbe District in the Zambezi River Valley. They share with the Plateau people the same clan system and among them the institution of clan-joking also appears in the same form(4), Among both Plateau and Valley Tonga, descent and succession are in the matrilineal line. Today, they are organized into chiefdoms under Native Authorities instituted by the British Administration. Tradition- ally, they had neither chiefs nor other forms of instituted authorities to bind them into a tribe or some organized political body. Instead, though they recognized their common cultural and linguistic affinities, they were content to give their loyalty to much smaller groupings. These were of two types: small neighbourhoods composed of a few villages or- ganized about a common rain shrine and cult; and a large number of small groups based on kinship. The members of such a group were usually dis- persed throughout a number of neighbourhoods. These groups I have called matrilineal groups to distinguish them from the much larger clans, which ;,~ ~ ~ ~ ~ ~ ~ ~~4 are known by the same term, mukowa (5). The matrilineal groups were bodies which shared inheritances, bridewealth, and other privileges and responsibilities. Their members met together for common purposes, both ritual and se0ular. They acted as vengeance groups, and formed mutual assistance associations. Theoretically, all members of a matri- lineal group wore descended from a common anoobtress, a few generations back, though usually her name had been forgotten, and those within the group were not able to trace their genealogical links with each other. Though I have written in the past tense, the matrilineal groups remain important elements of Tonga social organization, and many of their functions remain intact. THE CLANS: The clan is a much less integrated body than the matrilinoal group, and its functions are not as clearly defined. True the Tonga, even today, will argue that the clan system is of the greatest impor- tance in their lives, but this is because they are usually thinking of the matrilineal group which is only situationally distinguished from the clan of which it forms a part. Still, though the clan is widely dispersed and owns no property, has no shrines, no common rites, no occasions on which all its members gather for some common purpose, the clan is the single unit in Tonge social life which has more than an ephemeral existence. Villages usually have short histories. Even neighbourhoods may shift their boundaries and composition within a generation. People move from one to another, and need not spend their lives within one village or neighbourhood (6). Matrilineal groups may also be disrupted, and reform themselves into new grounings, as people become geographically separated and no longer maintain the old relations with their kinsmen. The clans alone are thought to be eternal. Most Tonga take them as a part of the natural order of things left to them by their ancestors from long long ago. Over most of Tongn country there is no myth which purports to explain the origin of clans or people. Both have always existed, or were created simultaneously. In the western areas, such as the present ohiLfdom of Macha, some people havo a legend that tho first people, already organized into clans, descended from the sky to light on a certain hill whore their imprints ma:? still be seen. Even in this myth, clans and people originate together. The only additional information they have about the develop- ment of the present clan system is that once two of the present clans were one, and then for some reason separated into the clans knltfas Baleya and Bantanga. This division again is said to have occurred long before the memory of livin: men. Since then the clans have continued unchanged. There is no myth which gives one priority over tho others or purports to relate its history. This division into clans is considered to be an invariable human institution. The Tonga find it practically impossible to conceive of any society in which people ere not divided into clans. Since most of the other peoples they have encountered also have a clan system, 46 belief is not surprising. Indeed, many of the same clans appear in .tribes, and foreigners who come to settle in Tonga country have o difficuilty in fitting themselves into the Tonga clan system. They fr belong to a clan which is also represented among the Tonga, or Mind some way of equating their own clan with one of the existing Xclans.. Thus, though Tonga country has always received influxes of c. from other tribes, and today contains many foreigners who came lto work in the European areas near the railway and then settled in onga reserves, the Tonga clan system remains unaffected. Strangers t introduce new clans into the country. They find their place with- * existing divisions, only fourteen in number. The clan system then gives a common basis of understanding with e of other tribes, and makes Tonga society into an open system. For h their membership in the same clans, foreigners can find their * within Tonga society, while the Tonga can also associate themselves other tribes if they leave their own territory. Clan membership gives rise to only a few limited obligations and a, which perhaps is why it is so easily extended to strangers. In 4ys before British Administration brought peace to the land with its uon vengeance and enslavement, the clans may have been more impor- in offering security to the man who ventured outside his immediate area, for members of tho same clan were oxpooted to otfor oach r hospitality and assistance. It was considered unethical to enslave, o hold in slavery, a fellow clansman, though a slave assumed his r's clan without affecting their relationship. Today this aspect of membership is no longer of any importance. But the clan continues otion in other fields. It still governs marriage, to the extent clan exogamy is rigidly maintained. I have recorded only three * of marriage within the clan where there was no question of slave nt to complicate the issue. For a slave or a person of slave-descent quasi-member of the clan with whom marriage is permitted. The Tonga as clan exogamy as the most important. aspect of the institution. at fifty years ago, when missionaries first went amonp! them and ied them about the nature of the clan, the Tonra said that the clan *an institution given to them by God so that people might marry pro- y (7). And so they still view the matter. The clan system also pro- a a mechanism for finding acceptable substitutes for certain rites hich matrilineal kinsmen should participate if no one of the proper tory is available. Finally, the clan system forms the basis for a em of joking relationship and reciprocal services. Before describing this, however, it is necessary to discuss clan a, and their association with certain animals. Twelve clans are very widely throughout Tonga country, from the Zambezi River to the rn borders. Two more seem to be found only in the western areas. he northwest, the system 6f names seems to be rather different, and ably it is affected by the system of the neimhboturinpt Ila. Most of the clan names cannot be translated today, though they are 47 assumed to refer to animals. Many of the clans have a number of names, f"Any one of which may be used. Informants do not know why these multiple names exist, and they certainly do not seem to designate divisions within a clan, or local groupings. The same person may sometimes say, for in- stance, that he is a Mukuli, again that he is a Mutenda, and again that he is Muunga. When queried, he will comment, `It is the same clan. It just has different names." Pach clan is associated with a number of ani- mals or natural phenomena. These are not totemic associations, since no one avoids or honours his clan animals in any way. The association betwee a particular animal and a particular clan is not invariable throughout Ton- ga country. In some areas, one clan will be associated with an animal which is attributed to quite a different clan in some other area. Occa- sionally even within a single area there is some disagreement between people as to the proper clan association for different animals. But in the main this difference of opinion seems to apply chiefly to what one might call secondary associates. Each clan is usually referred to as havinF one particular associate, and then informants remember to add that it also has others. These secondary associations are remembered chiefly in the praise names and slogans belonging to the clan and in clan joking situations. Withy the Tonga should have such a varied array of animal and other associates for their clans, I do not know. Possibly it results from the amalgamation of the various foreigners into the common clan system. Whatever the cause the associations persist and appear in the joking situati ons. The clans, and their most common associates are given below. Where an animal has been attributed by different informants to different clans, I have shown this by placing it within brackets. CLAN ASSOCIACTES 1. Bahyamb , hyena, rhinoceros, pig, ant, fish 2. Batenda elephant, sheep, lechwe, (hippopotamus) 3. Paleya goat, tortoise, black vulture 4. Pansaka leopard, bee 5. Bakonka (eland), jackal, rain, zibra 6. Bafumu pigeon, frog, (bipponotamus), cattle 7. Bansanje hare, honey guide 8. Bayuni bird 9. Pacindu lion, train 10. Peetwa crocodile, monitor lizard 11. Bantanga white vulture 12. Balonoo babboon, (buffalo), scavenger hird 13. Pancanga bush-haby 14. Bankombwe (buffalo), (eland) Clan names are used constantly in daily life throughout Tonra country. People are commonlyr addressed bar their clan names, or they may be honoured by being addressed by the clan names of their fathers. Everyone, includ- ing young children, knows his clan affiliation. 48 oh clan has a formal joking relationship with a certain number of glans. This is known as bu wanyina. A clan-joking relative is iuJwama (my follow joker), mujwanyoko (your follow joker), na his or her fellow joker). The term seems to be derived from or sister's child, mujwa, to *whioh is. addd.theQpo8seesivo.suf- ). Despite this derivation, the relationship is one which exists .clans, and it has no implication of kinship. Most people, indeed, e where actual kinship exists, the joking-relationship goed into o. Others claim that clanb-joking relationships are of two types: * forpl unchanging relationship between paired clans, -the other loular relationship between a person and the clans of his mother's and father's father. The latter type, they then go on to say, is true buJwanyina relationship but is only like it. Thus a woman who ak3a said that she had a clan-joking relationship with the Baloya, O qualified this statement withr `But they are not real I .ana. it because I am a grandchild *of the Baleya. My father was born (i.e., her father's father was Muleya), and so at a Baleya funeral mand*.thatz.meaatbe giv'en.:ttb me. But my children won't have n with the Baleya. Their will follow only the real bajwnyina o Bansanje, Bahyamba, and Balongo." o confusion may occur because there is also a formal joking rela- ip between kinsmen of certain categories. This differs in kind at of the clan-joking. Such joking exists between crass-cousins, affines of the same generation, and between grandparents and ldren. In the last category, the joking is extended to include *bers of the grandfathers' clans. This type of joking is known by ku-sobasyana ("to cause to play"). It involves a good deal of gB Botween grandparents and grandchildren it is confined largely exchange of pleasantries and sexual innuendos. Those indulging Pall out, "You are my wife," "My wife get water for me," "You are #dparents and so you must give me a wife (or husband)." Cross- ,nd affines of the same generation have the same type of verbal but they may also indulge in rough horse play, practical jokes, lan-joking is quite different, and the bystander easily recognizes differentiates it from the joking appropriate between kinsmen. ing involves the play upon the presumed antagonism between the imals. It permits obscenities and rough words# It permits those Frelationship to call to each other: "You mother is dead," "Your a brother is dead." Small children when theyr hear these statements h home weeping, thinking that a relative has actually died. Older laugh and retort: "Your mother is also dead." Clan-joking also a accusations of sorcery: "You are a sorcerer[ You are killing pleo" Such joking is called ku-tukila ("to use abusive language"). a permitted only between those in a clan-jokingf relationship. If olse used these words to a porson, it would be an insult. He would liod to account and made to pay damages before the matter was permit- t~drop. A - ~~~~~~~49 Some people may therefore assimilate, the joking appropriate to the grandfathers' clans to that of the olan-joking proper, but it is the relationship between paired clans which is the backbono of the system, and the ono which most Tonga stress. Each clan is paired with a number of others for this purpose, and these in turn are paired withn a number of others. This results in a web of ties between clans, rather than a divi- sion of the clans into a number of segments, each of whi6h has ae.jQing relationship only with others- within its segment. The pairing of clans for this purpos6, howover, is not invariable throughout Tonga country. Even within the same small area, informants may differ whether certain clans are paired together. Nevertheloss, thore is sufficient consistency for people to be able to move from one section of the country to another without involving themselves in cases for joking with clans which do not recognize their right to enter into this relationship. The Tonga think that the pairing ought to be universal throughout the country, and they also maintain that it applies to all members of the appropriate clans and not to matrilineal groups or to local sections of a clan. They say that as soon as you learn that a man is a member of a paired olan you may begin to joke with him in the prescribed manner whether or not you have previously known each other. I have found the clans paired for jokinF in the following fiahion:.(10), JOKING RELATIVES C LAXT 0 0 | | 0 0 0 0 0~1 0 0 0 0 0 0 0 0 2 _ l | gq| ^ ?pp | m pp m pp pp m pp pp pp pp .Bayamba. X X X X X X X X X Batenda X X X Baley ;X Bansaka X X X X X Bakonka X X a X X X X Bafulau X _ X _ X x Bansanj= X X _ X XX X Badindu X XX_ XXX X X X X Beetwe3- X_ XjX X X X X __ X X BankoaXwe = = -X X - = Bankombwe j _ =-==_ X - 50D If the Tonga are asked why certain clans joke together, they either efer to some antagonism between the associated animals of the two clans, t.ier r(.s.oate a fol1k-tale which is said to describe the origin of the lationship. For unlike the clans themselves, the reciprocal relation- ip between them is thought to have come into existence after the o. iein people. A Muhyamba, for instance, will say, '"e have bujwanvina with e. Bansaka because they are bees who make honey, and we are ants who al their honey. That is what we always tease each other with. And we .e bujwanyina with the Baleya because they are roats, and we hyenas and steal them from the house. And we have bulwanvina with the oindu because they are lions who kill meat, and we hyenas come carefully .eat alongside the lion when it is satisfied. And we have bujnyi th. the Bansanje because they are hares, and both hare and hyena are 1iQlksters." When Joking partners meet thev tease each other with such *,!renoes, but among the Tonga the jesting does not turn on the counter- a.m of the two clans to superiority over each other because of the h4tionship between their animal associates. The two are assumed to be ,equal status. In this the Tonga differ from other tribes of the area - have the clan-joking. The jest may be elaborated into a folk-tale as follows: "We Bahyamba >we uJa with the Bansanje because we are hyena and they are hare. , day Hyena fell in love with a girl and asked her to marry him. She greed. Then along came Hare who also wanted to marry her. She refused At marry him because she had already accepted Hyena. Hare said, 'How can .u.Marry him? He is my mount. I ride upon him.' The girl then thought at perhaps Hare might make a better husband. Hyena came to see her. * refused to marry him, repeating that Hare had said he used Hyena as mount. Hyena went off in a rage. He found Hare lying in bed claiming ,be very ill. Hyena announced that-they had a case and Hare must come 1ih him to the elders to have it settled. Hare denied that he had ever aimed to ride upon Hyena, and said it was impossible for him to come now h )~e was ill. Hyena insisted. Hare finally agreed, and said he would h if Hyena would carry him on his back. Hyena set off with Hare on his ok. Hare suggested that Hyena should run so that they might get there fster. Thus Hare galloped up to the gathering, mounted on Hyena to whom he applied a switch from time to time. As they arrived, he shouted, 'Well, seeZ Do I ride him?' And he jumped- off. Since then Herena (Bahyamba) and e (Bansanje) have had ?uwnina. and that is what they joke about," A similar tale relates the origin of the butanyina relationship petween the Bahyamba, the Bafumu, and the Bansanje. "One day there was inooeros (Bahyamba), Hippopotamus (Bafumu), and Hare (Bansanje). Hippo- otamus always stayed in the water and never came out in the day time. hinooeros never went in the water. Hare went to Rhinoceros and said, Wy don't vou go into the water and bathe? You would feel fine. I'll oll you what I'll do. Tomorrow morning I will bring a rope, and you will ke one ond'andI will'take thb.otherl I will get in the water and we ill both pull. If I pull you into the water that will mean that you will eom tht6 the water'all'the tinei' Rhinoceros agreed. Then Hare went to ippopotamus and told him that he ought to come out of the water in the .~~~~, day time, and that next morning he, Hare, would bring a rope and see if he could pull him out. So the pulling contest was arranged. Hare, out of sight of the contestants, so arranred the rope that Hippopotamus and Rhinoceros were pulling against each other. Then he ran away. Hippopo- tarmus and Rhinoceros pulled most of the day. Then each began to wonder what could be wrong and if Hare could really Dull this hard. They went to investigate and found each other. They said, 'Ah, Hare has tricked us " Since that day, these three clans-Behyamba, Bafumu, and Bansanje- have had the joking relationship. `t Another describes The origin of the joking relationship between Bakon. ka, BansanJe, and Bacindu. "One day Lion (Bacindu) fell into a pit trap. Jackal (Bakonka) came along and agreed to push a stick into the trap so that Lion could climb out. Lion was very hungry after his imprisonment. He announced that he was going to eat Jackal, who protested that this was no Juet' repayment for his assistance. Lion refused to listen. Before he could eat Jackal, Hare (Bansanje) came along and asked what was happen- ing. When he had heard the story, he asked Lion. to show him just what had happened. Lion jumped into the pit. Hare grabbed out the stick and told Jackal not to help Lion out of the trap again. So their went off leaving Lion in the trap. Since then Jackal (Bakonka), Hare (Bansanje), and Lion (Bnoindti) have had bunlyina.t" It would be pointless to oive any more of these tales, which are all of the same type. They are said to explain the origin of the joking rela., tionship, but they are certainly not essential to it. Many people deny any knowledge of the tales and say that they carry on the relationship with their paired clans because this is a matter of tradition: "Perhaps the old people knew how it started. We6 just know that it is the custom which they left for us to follow. So we follow it." Even the very old may not remember the tales attached to their own particular joking- relationships. One old man of about eighty told me that he as a M.uhyamba has a joking relationship with the Bansanje. when I asked him why, he cheerfully replied. "You must ask the old people. They never told me an I don't know." Certainlv the tales give no clioe to the importance that the pairing of clans has in Tonga life, for it enters into many situations, besides that of the formal exchange of insults and jests. Paired clans have re- ciprocal duties which they perform for each other. 'Whenever a man or woman has so misbehaved that he has brought upon himself the general con- d3mnation of his community, it is through the clan-joking partners that his shame is brought home to him. This can not be done through his own kinsmen-through the people of his own matrilineal group and the people of his father's matrilineal group-upon whom he relies for support and assistance. Nor may it be done through his affinal relatives. For them to shame him would injure their permanent relationship which is built upon mutual respect and support. Perhaps their exclusion is also based upon the assumption that if the offender wore amenable to the advice of his kinsmen~, he would have so patterned his life upon thoir advice that he could not have committed his offence in the first place. In any ovens 52. 1 his kinsmen are merely bystanders, and witnesses to his formal punish- Frnt, whlichi is in the hands of his clan-joking partners. If a man Fpoatedly takes and wastes the property of his kinsmen, his ioking- rtnaers chide him with his folly and mock him for his stupidity. Thav Ice general play with his shortcomings and give them full publicitye .a man and woman commit ince.t, either bv having sexual relations with jClansman, or with some other prohibited relative, the matter is not to rest. Their own kinsmen cannot proceed against them and force em to pay damages. But the man is forced to produce a goat which is lled. Then the women must cook porridge, while the man cooks the at of the goat. While they are thus at work, clan-joking partners tnd about jeering at them for their misdeeds. They also take the blood * the goat and smear it on the bodies of the offenders, saying, "Hore As your incest!" "Then everyone who is thern will be afraid and think, i ?XtI should ever do anything like this, they will give me the same punish- nt. I must never agree if someone who is my brother tells me that he nts mo. I must remind him of what would happen to us if wo did this one.' So this is for punishment. It is the clan-joking partners who st make them ashamed of what they have done." ,, The joking-partners also come to bring shame to any ono who has ,iempted to commit suicide. But in this, as wnll as in cases of incest, Ad in certain othor situations, the ritual serves not only as a punish- nt. It is also a means of averting the evil which would otherwise glow the kinsmen of those involved, or even the offenders themselves. 3the clan-joking partners act even in situations where there is no ght of shaming since no offence has been committed. If a granary 3glpses under the weight of the stored grain, the joking-partners come d curse the owner whilo they remove all the grain in their baskets. a is said to prevent the death of the owner foretold by the collapse the granary, for on a death mourners assemble ard the Granaries are ptiod to food them. Such occurronces-incest, suicido or attempted Oide, the collapse of a granary, and various others-are said to be wOza. This term is also used to refer to some subsequent misfortune s attributed to the original occurrence. For these, the clan bking-partners assist to ward off the threatened misfortune. 4t all funerals, jokiFp-partners have a role to play, but they Va, a particular duty of burying' the bodies of those who die gs suicides from leprosy, for these are considered to be malwoza (12). After most Mths, kinsmen and others living in the village of the deceased join bothor to bury the body, but they may not handle the bodies of lepers 4 suicides. Instead the -5okinlr-partners come to drag these out into o bush where they are thrust into an ant-hill and abandoned. For such ople there is no formal mourning, nor do the people assemble. Relatives 'thh dead will probably kill , -oat or a beast to foed the Joking- ntrers who have officiated, and thus the matter ends. At other funerals, ing-partners perform certain rites at the grave, such as pouring a #abash of water over the Rrave, and again after the burial they have M duty of crying out to begin the renewed wailing. They also act the 5~3 part of clowns, to release the tension of the mourners. `If a Muloya dies, thon we Bacindu go to laugh and joke. We say, 'Well, he has died. Don't care abbut this Perhaps you yourselves have killod him! You are soroorc:: ors. Then a Muloya will say to us, 1You too are going to die sometime. You also are sorcorersP" Wuhen cattlo are killod to fndod tho mourners, joking-partners go out and play about with the bodies of the boasts. If a rich man dies and many oattl5 are killed, the first one is for the jok- ing-partners, who may rush in and take the meat. They also speak out to tell the people that it is time to end this mourning and to disperse to their homes. If anyone else behaved in this fashion it would be a serious matter, which long ago might have led to his being enslaved by the offended, relatives of the dead. Joking-partners may also be considered to stand in a quasi-kinship relationship, but one which fits into none of the known categories of kins- men. For this reason they can be substituted whenever a kinsman of a par- ticular category is not available to carry out his appropriate task.,' At funerals and on other occasions when a large number of people are gaGthered, it is the duty of the affines who have married women of the matrilineal group involved to cook the food and perform much of the other work nooes- sary to the occasion. At ono funeral when I questioned why different men were working, I was told of one, "He is our joking-partner, but there are not enough of our affines here to do the work, so we told him that he must help us out and work like an affine. That is all right. He must help us because he is our joking-partnor." Again, when a man first obtains a now plough9 he must take it to the field of his father, or that of some member of his father' s matrilineal groups and work with-.it thoere before using it in his awa field. A young Mwrootwa said that whon ho got his plough ho took it to the field of a Muoindu. I asked why he had so departed from oustwn,. and. ho ropleod, "tBut ho is my joking-partner. I know I should tako it to my fathor's fieold, but. nono of my father's matrilinoal group^ and no mombor of his clan is living hare. So I took it to the field of our jokingw.partnor. That makos it all right for mo to use it now in.. awn fioldo" While the Tonga would argue that any joking-partner from any clan paired with your own might perform these services, in actual practice, of course, tho pnople in one neighbourhood havn worked out an informal arrangement by which only a few of the many possible joking-nartners participate. All joking-partners in the area need not undertake the ob- ligations. It is enough if only one appners, and no joking-partner may be sued for damages because ho has failed to put in an appearance or per- form his obligations. Other aspects of the joking-partnership brings a much wider range of participation, since the joking-partnership also governs the relationships between people of different clans on other occasions* Various informants have told me that long ago it was forbidden to take offence at anything a jdkinggpartnar did. "Long ago if my joking,.-partnor was wearing a coat, T could go and spit on it, and he worlld have to give me the coat because I am his joking-partner. He would give it to me even if it was a new coat.,| 54 id do this any time and not just at funorals." Again, I have been hat theft was not recognized between joking-partners. Some said you impregnated a girl from a paired clan that you were not charged damages. Her relatives sinmply cursed you up and down, and you paid * Others said that this was not true, that you had to pay AL11 da- Today, certainly, no one pleads a joking-partnership as lessening sponsibility in such a matter. Informants also said that previously l11ed a joking-partner, you paid loss than if you killed a momber other clan. In times of famine, you could go to a villagn whore d Joking-partners and beg for food and for seeds for planting. In- you might go to the granary of a joking-partner and take the grain t permission. "If I saw him, I couldn't give him a case because I know that we joked together." Old men say that they still observe custom, but that tho younper people do not. It was of considerable anon in the old uncertain days when famines wore of frequent occur- and thern was no Administrative Authority to prevent starvation by 4aportation of grain. For if you begged food or sood in a village S-you had no kinsmen or joking-partners, you were liable to nnalavomont. rg-partners, like clansmen, wore not nxpected to enslave oach other. In ordinary life, ono would not urge a case against a joking-partner. sign and seal of the relationship was the refusal to take offence, olised by the right to use crude and abusive language to each other. kinship relationships, nither oonsanguineal or affinal, one also should clow to take offenoe, but this was because there were practical advan- a from the continuation of the relationship which acted as a check anger and intemperate dispute. Even if anger boiled up, it could be d so that the relationship with its practical advantages could con- ueIn the Joking-partnorship, offence was outlawed from the beginning. one of the partners did take offenco and started to fight, thon damages to be paid. But this implied the and of the relationship. 'It is for- on to fight with thom. If you fight, it means that the joking-partnor- 1p is ended. You will be afraid to joke with them again because you will ink #These people only want to fight.t" Nevortheless, informants main- nnd that if a man found that his joking-partner was having an adultor- affair with his wife he would demand full damages without regard to o affoot that thin might have upon their relationship (13). Marriage between peoplo belonging to paired clans was not only acoop- d It waenan approved form of marriage. "That is our special place to rry." Full bride wealth was paid, but marriage into a paired clan was oridered to be a safeguard for the spouse who wont to live amongst his her affines, for though actual kinship ties, whether consanguineal or inal, placed the joking-partnorship in nboyance, such people had an dod reason to protect his or her interests. The existence of the joking-partnership, or clan reciprocity, there- ore gave the Tonga added security in a world made up of small opposed oups, a world in which oh found himself easily at odds with others and osoed to their vengeance. Against this he had only the security which ould be offered to him by the willingness of his kinsmen to take up his L S ~~~~~~~~55 OUarrels as their own. But the iokinr'ppartnership meant that offence could neither be riPron nor recoenized in dealings with a larco number of people with whom he cnme into contact. Parom the tnble showing clan pairs, the wide range of joking-partnorships is apparent. If I had collected information more systematically I suspect that all the clans wtould have emerged as paired with at least half the Rvnilable number of other clans. Thus the Tonga could move through a wide circle of rolation- shlps. with security.-. Ho had to be certein of the clans with which he joked, or he embroiled himself in difficulties, but if he followed the simplo rule that he joked where his mother joked, and not where his father ,j6ked. he was safe (14). This I think is the true significance of the paired clan arrange- mont among the Tongs, and perhaps elsewhere in Northern Rhodesia. It also has the affect of mobilizing and expressing public opinion through the mouths of joking-partners who by definition are not kinsmen and who reo protected in the exercise of this function by the outlawing of retaliation against anvthinT they may say. The jokinp-partnorship is brought into the context of the funeral, because in the funeral all those who have responsibilities to a person during his lifetime are Given particular roles in the ritual which surrounds his death. The funeral friendship is therefore only one aspect of the wider problem of how people may bo organized into groups which may then be effectively related to each other, to ensure the woll-boinp of a community. 56 ENDNOTES )' Rioharda, 1937, (Bemba; Stefaniszyn, 1950 (Amnbo); Melland, 1923, pp. 251-252 (Kaonde); Doke, 1931, ppe 197-198 (Lamba) a 7. G, ' Cunnison informs me that it is also found amcng the peoples living in the Luapula Valley. Among the Southern Lunda, joking occurs nit between. clans, but between two sections of the tribe, aecording to information quoted in MoCulloch, 1951, pp. 21-22.. Clan joking has not been reported for the Ila, close neighbours of the Tenga with whom they share many customs. Instead joking is characteristic of ? the age grades. See, Smith and Dale, 1920, Vol. I, pp, 3Q8,3ICa [) Tew, 1951. She gives references to the literature describing f?neral friendships among the different tribes of the area. p5) Radcliffe-Brown, 1940; Radeliffe-Brown, 1949. Both articles are reprinted in Radeliffe-Brown, 1952. 4) Material for this pacer was gathered during the years September 1946- September 1947, July 1948-July. 1950, when I worked among the Tonga as a research officer of the Rhodes-Livingstone Institute. Most of i my time was spent in villages to the east of the Northern Rhodesian railway line which today cuts Tonga country into eastern and western sections, and my information is most reliable for this area. However, during my first tour, I also worked in villages to the west of the railway. In 1949, I spent a month in the Gwiembe District among the We. My thanks are due to the Trustees of the Rhodes-Livingstone Institute for their support of the work; to the Colonial Develppment and Welfare Fund which financed much of it; to Mr. Benjamin Sipopa, my clerk, who assisted in collecting the data; and to Professor M. * Gluckman, who has read and criticised this paper. 6) For further information on Tonga social organization, see Colson, 1951a. 6) Colson, 1951b. 7) Personal Cumunication from Father Moreau, S. J., who helped to found the Roman Catholic Mission in Tonga country in 1905. 8) The bu. prefix in Tbnga seems to carry any abstract sense. See Hopgood, 1940, p. 30. Torrend, 1931, p. 461 gives the meaning of mujwama, etc., as "my relative of a different tribe." By tribe, he means clan. But I have found the term restricted to those with whom the speaker has a clan-joking relationship. It is not used for those who are considered to be kin. 9) Age-mates may also joke together. .57 ('o) These pairs were collected from informants in Monze chiefdom west of. the railway, and in Mwansa, Chona and Ufwenuke chiefdoms to the east. In the northwest, in Mwanacingwala chiefdom, informants denied that bifjwanyina was connected ith clans, and I was never able to work out the system which governed it. These people are probably affected by their proximity to the Ila who do not seem to have clan-joking. In the Zambezi Valley, the We or Valley Tonga have the clan-joking pairs. During the month I spent in the Valley, I was working on other prob- loms, but I did collect the following clan pairst Batende are paired with Bafunu, Bahyamba, Balongo; Bantanga are paired with Bayuni, Baleya, Bnhyarnbn. Becindu; Bayuni are paired with Bansanje, Bahyamba, Bakonka, Batanoa, Bacindu, Batenda, Beetwa; Beetwa are caireG with Bacindu, Befurfi<, Bayulni; Bacindu are paired with Bayvuri, BRa'k'ahyamba, Balonge, B eetwa. Most of these pairings, but not all, were also found on the Plateau. (11) Banchanga are found only in the west, and I had no opportunity to work out their j oki ng-partner s. (12) Vralley Tonga informants denied that joking-partners had the task of burying suicides or lepers. This custom is therefore probably con- fined to the Plateau. (13) The Valley Tonga said that they ignored cases of adultery if a joking-partner was 'involved. (14) Dr. I. G. Cunnison informs me that on the Luapule River, people may assume the joking associations of both parents, though they belong only to their mothers' clans. Elsewhere in Northern Rhodesia, people seem to be limited to the joking-partners of their own clan. ,. . . .~~~~~5 B IBLIOGRAPHY 1951a "The Plateau Tonga of Northern Rhodesia," in E. Colson and M. Glucknan (eds.), Seven Tribes of British Central Africa, London, Oxford University Pres for the Rhodes-; Livingstone Institute. 1951b "Residence and Village Stability among the Plateau Tonga," Human Problems in British Central Africa, No. 12, pp. 41-67. 1931 The Lambas of Northern Rhodesia, London, Harrap. ode, C* Re *4O Tonga Grammar, London, Longmans, Green & Co. lloch. M., 795l The Southern Lunda and Related Peo London, Inter- national African Instcitu- Fd, F, H* 1923 In Witch-Bound Africa, London, Seeley, Service & Co. itfe-Brown, A. R* 1940 "On Joking Relationships," Africa, Vol. XIII, pp. 195-210. (1949 "A Further Note on Joking Relationships," Africa, Vol. XIX, pp. 133-140. 1952 Structure and Function in Primitive Society, London, Cohen & VWesi irds, A. 1947 "1Reciprocal Clan Relationships among the Bemba of N. E. Rhodesia," Man, Vol. XXXVII, pp. 188-193. 59 S.iithh, E and A. Dale 1920 The Ila-Speakig Peoples of Northern Rhodesia, London, Macmillan & Co. Stefaniszyn, B. 1950 "Funeral Friendship in Central Africa," Africa, Vol. II, pp, 290-306.' Tew, Mary 1951 "A Further Note on Fineral Friendship," Afric, Vol. XlI, pp. 122-124. Torrend, J, , 1931 An English-Vernacular Dictionary of the Bantu-Botatwe o Rhodesia, London, Kegan Paul. 60