The Canela (Eastern Timbira), I: An Ethnographic Introduction.
By William H.Crocker
Smithsonian Contributions to Anthropology,
Number 33, 487 pages, 11 tables, 51 figures, 78 plates, 1990.

<<back  table of contents  next>>
<<return to literature

Part III: Social Organization

The Canela have been compared to the Australian aborigines (Maybury-Lewis, 1979:303) because, like the Aborigines, they have developed a complex system of social organization at a relatively low level of technological development. Anthropologists who try to work out scales or ladders of unilinear or multilinear social evolution find the Canela social complexity to be uncharacteristic of most other tribes that are between the hunter-gatherer level and the fully agricultural one. (See Carneiro, 1967). Because this complexity makes the study of social organization a crucial one, and because most Central and Northern Gê specialists have concentrated on this topic, I have placed the main focus of the book on presentation of Canela social organization.

The first chapter, "Socialization and Related Adult Activities" is about the enculturation of the young through the various foci of socialization. It is also about the social forces that keep adults within traditional lines of behavior. This traditional human processing, as carried out by parents, was permissive until puberty when much of the control was passed to aunts and uncles, who restricted the behavior of their nieces and nephews through severe discipline. By the mid-1970's, these roles of the aunts and uncles had all but disappeared, with the roles of the parents strengthened somewhat.

This chapter is followed by what is partly the product of socialization, "Psychological Polarities, Values, and Behavioral Orientations." Aspects of this sort seemed best studied among the Canela through analyzing key polarities, such as individuality within solidarity.

The third chapter, "Socioceremonial Units," is informational, describing and defining for the rest of the monograph the principal social units involved in on-going daily living and in the festival system, whether political [III.D] or ceremonial [IV.1, 2].

The fourth chapter with sections on the chieftainship, council of elders, and judicial system emphasizes the political structures of Canela society. The tribal chief is in control of most of the political power in the tribe, while the members of the council of elders check and limit his political behavior and act in their own right with respect to ceremonial matters, governing festivals and bestowing prestigious awards on good performers. Canela informal law is well developed, especially with respect to keeping marriages with children together. Interfamily public hearings are frequently held to reduce excited or hurt feelings related to almost any disturbance of the peace. The chief of the tribe is the ultimate judge and arbiter in legal matters.

The terminological relationship systems discussed in the fifth chapter include consanguineal and affinal terminology, name transmission, Formal Friendship, Informal Friendship, teknonymy, the contributing-father system, mortuary terms, and ceremonial practices. These systems constitute most of the interpersonal social organization, in contrast to the political social organization, in a tribe at the Canela's position in social evolution. Marriage completes the discussion of interpersonal linkages. The kinship system is Crow III in Lounsbury's system (Nimuendajú, 1946:351–393), but "parallel transmission" equivalences published by Sheffler and Lounsbury (1971:110–112) provide more congruence with both the empirical data and the social structural correlates.


The chapter concentrates on the specific points in the life of a Canela at which the forces of socialization are most intense. "Forces of Socialization" [III.A.3] is divided into three age groups: children, adolescents, adults. In discussing each group, emphasis is placed on the "forces" (that is, the outside influences, pressures, and facilities) that (1) restrict or discipline (i.e., limit), (2) reward or positively motivate (i.e., attract), or (3) activate or help (i.e., equip) the individual to develop in the socially desirable patterns. Thus, I see the individual moving through life reacting to (1) the restricting and restraining forces, (2) the facilitating and rewarding pressures, and (3) the aids provided and her or his own personal abilities. Reality is more complex. This simplification was adopted for heuristic purposes.
The next to final section presents ethnotheories of growth [III.A.4], some Canela ideas about the social development of the individual, and a summary of my observations [III.A.5] about Canela socialization.

[III.A.1] Research Methods

It would be methodologically correct, but too time-consuming, for an ethnologist to move systematically throughout an entire village, observing socialization from house to house, day after day. To some extent I did just this in both villages of the Canela in the late 1950s, and in the Apanyekra village of Aguas Claras (Map 8), while involved in other kinds of research (census taking, for instance). At such times I did notice and record various kinds of socialization in most houses of the villages. My principal studies of socialization, however, were carried out through observation in my houses of residence and through extended discussions with research assistants, which systematically covered the field of socialization.


About 75 percent of my observations on socialization of children were made in the Canela houses of my adoptive sister Te?hôk and my adoptive brother Hàwmrõ and his wife, Mïï-khwèy. In Te?hôk's house, her several sisters were usually present with their children. In my brother's house I could watch the socialization practices of my brother's mother, and later, the practices of his three grown and married daughters with their babies and children. Four generations lived in that house. The socialization practices in the house of my adoptive Apanyekra sister Pootsen provided another source of observation.

In all three of these houses of adoption, the age range of children was from breast feeding through belt winning for girls and through Pepyê internment for boys. During my 22-year period with the Canela, I observed some children mature into adulthood and raise their own children. It was not uncommon to see mothers and daughters raising their similarly aged babies and older children all together.

Opportunities to observe casually what older people were doing for children were excellent. I simply sat on a mat or chair while eating three meals a day, day after day for years, and watched what was happening in large rooms where many activities were occurring. I was one of the many fixtures of such rooms, being inactive but eating and observing. It was possible to take notes right there, or to speak into a microphone in English without disturbing the family interaction. Moreover, I often traveled with whole families to farms, or to one of the other villages. These trips were slow, lasting three days instead of one, because women and children took a long time in preparation for moving and then walked slowly. Sometimes I observed the traveling group and wrote or spoke simple notes for later elaboration. Observations made at such times (while eating or traveling) were often discussed with research assistants who went and asked details from the adults involved to improve our understanding of the facts of the situation. The socialization incident was then elaborated upon and its possible variations and parameters discussed in my research assistant council.


Another way of approaching the study of socialization besides observing incidents was working with a group of research assistants on this topic. In 1958 and 1959, I studied with one or two research assistants [In.4.d]. Then, in the spring of 1960 and again in 1964, I worked intensively with a group of research assistants who used to debate points among themselves and then turn to me with their opinion in Portuguese, after I had been listening to their debate in Canela.

Although socialization was never again the principal topic of investigation after 1964, I reassessed this subject with research assistants from time to time during the 1970s to learn more about the more obvious trends and changes I was observing. It should be assumed, however, that the socialization process presented here comes from examples observed in 1959 and 1960 (i.e., "the late 1950s" [In.4.f]), unless otherwise stated, or unless some observation of the 1970s is specifically being contrasted with the earlier period.


Canela villages are relatively homogeneous in social training because of the openness of Canela life and the lack of significant intratribal ethnic differences. Nor are there class differences of any significance, although they were beginning to develop in the 1970s. Individual differences exceed any variation in family traits that might evolve because of relative status differences. However, child training in a Wè?tè house [III.D.3.e.(3)], or a political chief's house, could be somewhat different from socialization in an ordinary family house because the sociocultural atmosphere and traditional assumptions could be quite different. Significant differences could also exist between a house of self-sufficient individuals of high self-esteem [I.G.1,4,15] and a household of people who, rather than maintaining their own self-sufficiency through farm plots, depended on work at the houses of backlanders for support in September through December of each year [II.C.3.g]. This chapter, however, is not designed to cover such a possible range of variation in socialization, and focuses instead on a normative range of behavior.

I did not carry out intensive studies of Apanyekra socialization processes. Because their tradition is almost the same as the Canela's (both tribes being Timbira Indians of the eastern variety), it is assumed that their socialization processes are similar. I observed the Apanyekra, for a total of 10 months over 17 years. It must be remembered, however, that the slightly different values of the Apanyekra probably are reflected in their socialization processes as they are in their versions of certain myths.

[III.A.2] Foci Of Socialization

The foci of socialization presented here represent periods in a baby's, child's, or adolescent's life that are more intense and formally focused than other periods. The underlying principles that govern socialization in general operate more conspicuously in these periods, and comparisons between earlier (pre-pubertal) and later (post-pubertal) foci provide information on continuity or change.


An infant lives at the breast of its mother. In the late 1950s infants were rarely put down except to clean them. Whenever a baby cried it was given a breast on which to suck. It was believed that infants and babies should not cry very much, and that the moment any baby did, it should be pacified, or distracted. A breast was the usual means. The four or five women in a household took care of each other's babies [III.F.7], and almost always more than one of them could furnish milk. Thus, the very young babies, up to several weeks old, could be switched from breast to breast when one mother was absent, maybe fetching water or firewood.

Unrelated individuals who had fed at the same breast as babies were said to be "milk siblings," and sometimes in later life called each other by siblingship terms. (This is also a backland practice, so the Canela may have learned it from them [II.d].)

In the house of Te?hôk, her uterine sister (Hapôl) was usually present, and two more "sisters" (really parallel cousins), Tel?khwèy and the younger Khroytsen, lived next door. In the late 1950s and 1960, Te?hôk's daughter, Te?kurà, was part of the baby-tending team, and by the 1970s another daughter Píyapit had grown up to join it, but by then Te?kurà had died.

Ideally, even when a mother was sleeping, she held her baby in her arms. In the 1970s, however, babies were often left in hammocks as long as they were not crying; but if any crying started, they were picked up and nursed immediately. Sometimes they were carried around in commercial cloth slings instead of on the hip. Fathers sometimes hold babies, but this is not a regular practice.


Babies were not only fed on demand, but often were nursed for long periods. On a trip to Barra do Corda in 1960, Atsuu's baby was at her breast and nursed almost the whole time, even while they were in the back of a truck where riding was rough. This trip, through Leandro and Escondido (Map 3), must have taken a dozen hours, including rest periods. The breast was the place of security for a baby, and a mother encouraged it to nurse there during any long periods of stress.

Active attempts to distract a baby were a principal socialization technique and pattern of adult behavior. A baby became enculturated to put up with almost no confrontations and relatively few difficult aspects of living. When something unpleasant happened or when a baby was frightened, its attention was distracted from the disturbing object. Its mother gave it her breast, or it was put on the breast of some woman it knew.

However, by the time that the baby was five or six weeks old, it could distinguish and preferred its mother to other women and often did not accept the breast of another "mother." No attempt was made to force a baby to accept a woman it did not like. Sometimes in my brother Hàwmrõ's household, babies were given to their grandmother or great grandmother, the older Khroytsen (Figure 22), to hold and nurse, even though she had no milk.


At about 2 to 4 months, depending on a mother's attitude, she feeds her baby mashes of foods, such as bananas, manioc, sweet potatoes, rice, and even brown sugar. By 9 months, she might feed it soft, well-cooked meat. I never saw, and my research assistants denied, what Nimuendaju (1946:108) reports: that mothers gave babies pre-masticated food.

A baby is given only soft foods until it has teeth. In the late 1950s, babies were spoon and finger fed. No bottle feeding existed. Nothing was forced on a baby of that age. By the 1970s, however, some babies had plastic pacifiers in their mouths.


Babies first experience some degree of freedom in creeping and crawling across the earth or sand floors in their maternal houses. They are not restrained. They are encouraged to go where they wish and explore whatever they happen to find.

Sometimes older children put a toy a half meter ahead of the baby and coax it to crawl to the toy. When the baby almost has the toy in its hands, the child moves the toy a little bit further, not to frustrate the baby, but to encourage it to crawl a little more. Any signs of frustration, however, are appeased by immediately letting the baby have the toy. No crying for any length of time is allowed.

When my wife arrived for the first time in Sardinha (1963), she was disturbed to see little boys barely able to stand playing with large knives, wielding them and poking them at objects. While this happened quite often, I never saw or heard of a baby hurting itself with a knife in such a situation. Babies also wander close to fires, but more than one adult, including a parent, is usually watching. They let the baby wander and explore, but the moment it endangers itself with fire, or anything else, they call to it to draw its attention to move in a different direction. They may also pick the baby up, giving it the breast as a distraction so that it will not cry for what it has lost.

In the late 1950s, I witnessed an excellent example of the freedom to explore allowed toddlers. In Baixão Prêto, a large rooster, well tied with cords, was lying on the ground just inside the house. A little boy, A?prol, barely able to stand, approached closely and watched it. The rooster could swing its head slightly and could have pecked at the boy, but not with much force because of its tied position. With great patience, his parents, especially his father Hàwmrõ, were observing the situation, but nobody interfered. They could see the baby was not going closer to the rooster and therefore was safe.

This kind of freedom continues for boys of 7 or 8, when they go in groups into the cerrado to play [II.D.1.c]. They are not called to account for what they were doing during most of the day. This freedom to explore continues for men as adults when they go on trek to the large coastal cities [II.A.3.a.(3)]. Groups of men, sometimes with a woman or two, escape most restrictions of tribal life when they go out "in the world" (no mundo), as they call it [II.D.3.i.(1)].


At 11 to 15 months, adults encourage babies to stand and walk by holding them upright by their hands and encouraging them to take steps. A mother often holds her baby boy from behind with one hand while guiding its hand with her other hand into shaking a toy gourd rattle, hoping he might grow up to be a sing-dance leader [II.F.1.a]. I saw parents holding girls of that age by the ribcage, causing their bodies to go up and down, while their knees bent in the traditional rhythmic manner. Such a mother was hoping her daughter would grow up to be a great sing-dancer [II.D.2.e.(1)].

They encourage babies to walk at the beginning of their second year, though they do not make an ordeal of the learning experience. They still give it a breast if it cries, and relieve it of any activity that might cause crying. Nothing is forced. They believe each child will walk when she or he is ready to do so [III.A.4.a].


One of the oldest Canela men, Khà?po (Plate 70e) [I.G.7], who was in his early 80s in the late 1950s, reported that, in earlier times, weaning occasionally took place as late as 3 to 4 years. He demonstrated the age by pointing to two little girls, Kahuk and Píyapit, who were 2 and 4 years old. My research assistants identified several living people who had been weaned at this age (e.g., Khen-khwèy of Baixão Prêto and Yawè of Ponto), but this was not a usual practice by the late 1950s. By then, the weaning took place this late only when it especially suited the mother, they said. If nursed too long, they believe the child would grow up a weak person (irerek) [III.A.4.b].

My younger Canela research assistants said that the proper time for weaning is between teething and walking, depending on how the child appears to be maturing. Any time beyond standing and walking, up to 1½ or 2 years, was considered too late by the late 1950s. In any case, if the mother becomes pregnant, in spite of the practice of avoiding intercourse while actively nursing, they quickly wean the baby. They believe mother's milk is for the fetus rather than for the baby. They said a baby would become sick if it continued to nurse while a new pregnancy was developing.


The principal technique in weaning is to fool the baby (i?-hey: it /fool/trick/deceive: fool it: enganar êle: fool he [as is said ungrammatically in Maranhão backland Portuguese]). They attempt to trick a baby so as not to have to force it to do anything against its will. Thus, they encourage a baby to eat soft mashes of foods increasingly, especially when it is obviously hungry and more receptive to accept almost anything. Later on, even when it is less hungry, such soft foods are given more and more often in order to accustom it to them and train it to accept them in place of a breast. Any food that a baby does not like is quickly removed from its mouth to avoid an outburst of crying. If crying does occur, a breast is quickly given to the baby. They make attempts, however, usually through mild chatter and persuasive talk, to take a baby's attention away from a breast and focus it on the substituted food.


If weaning is absolutely necessary because a mother is pregnant, she sometimes applies babaçu nut oil or mild pepper (pimento) to her breasts to make them distasteful to her baby. This technique is a regular backland approach, according to research assistants, and therefore highly criticized. The Canela say that the backland women have no feeling for their children [III.B.1.b.(1)]. These women always wean their babies early with pepper, which makes them cry most of the day. The Canela disapprove of such treatment and only a few mothers practice it.


As in any culture, babies learn to talk by listening to their parents and imitating them. Parents attempt to have them repeat phrases, especially in joking relationships between kin. For a baby girl, if one of her joking uncles comes into the house, an appropriate adult begins the relationship by telling the girl what to say to joke with him. I heard extended conversations of this sort between uncles and nieces and aunts and nephews. (Naming-aunts and uncles, and those who have assumed disciplinary roles, seldom joke [II.D.1.b,c].) They even tried to teach me Canela in this manner. In old Ponto in the late 1950s, Pat-khwèy, an aunt next door, used to scold me, but my niece Te?kurà spoke my answers, so I merely had to repeat her phrases, at first scarcely understanding them. They were usually highly descriptive of personal sexual matters.


The urine of a baby in her arms does not seem to bother a Canela mother. If she is standing, she might hold the baby away from her to let it urinate more freely while she continues talking with somebody. With tiny boys this might be a problem, however, because of the direction of the spurting, but even here it caused women little concern. If a mother were sitting and busy working or talking to another person while holding her baby in her lap, she let the liquid wet her wrap-around cloth and pour down her legs, without giving it very much thought. Cloth dries quickly in the dry tropics and can be washed easily during any one of the several visits to the stream a woman is likely to make each day. Thus, they do not attempt to control urination in any way.

Female research assistants commented on how differently backland women viewed the urinating of their babies. They do not tolerate their clothing becoming wet this way.

In the late 1950s, Canela men urinated openly, anywhere, in the sight of Canela women. Research assistants asserted that if a grown woman walking along a trail in the late 1950s needed to urinate (even with a number of men present) she simply stepped somewhat off the trail, loosened her wrap-around skirt, and urinated while standing. By the 1970s, however, women no longer do this openly, but they still urinate standing in the dark of the early morning within 20 meters of their houses, when most people are asleep.

In the late 1950s, however, the Canela were concerned about being seen urinating by certain relatives. They were reticent in front of any individual they addressed using the personal pronoun [III.E.3.c], instead of ka. A man avoided being seen by his mother, sister, and daughter, and a woman by her father, brother, and son [III.E.2.b].

Traditionally, men went completely naked in their houses and villages. During the exile (1963–1968) in the village of Sardinha [II.B.2.g], they had to start wearing shorts or long pants because of the occasional, unannounced visits of Guajajara women, who lived in a village 500 meters away [II.A.2.g.(6)]. Thus, urinating in full view of the opposite sex tended to stop also. These changes with respect to clothing and urination were maintained and completed after the return to the cerrado in 1968. Since the late 1960s, men no longer go naked in Escalvado, and they are careful to hide themselves when urinating except in the presence of men.


The principal point to make about toilet training is that it is carried out completely and satisfactorily only when the child can understand and do it for her- or himself. The toilet training of a child occurs between the ages of 2 and 4 years. There is no punishing or shaming if the child makes a mistake. Proper behavior is rewarded simply by signs of approval and affection.

When a child of 2 to 3 years starts to defecate, the mother, father, or even some other person carries her or him by the armpits quickly to the bushes behind or beside the house. There the child is allowed to continue, and then is cleaned by the parent or another person who happens to be child tending at this time.

During one of my visits to Baixão Prêto, a certain little boy of three in my family was still making mistakes in the house. His parents were very patient with him. They did not punish or make him feel ashamed; they merely moved him outside and helped him take care of himself. They talked and explained while this process was going on but made no attempt to take him out to the bushes ahead of time to help him have the bowel movement there when they already knew he wanted to have one. They let him make the mistake rather than saving him from making it.

The expected result is that sooner or later any young person will, upon feeling the need for a bowel movement, go out to the bushes her- or himself. If the child still makes mistakes inside the house, no form of punishment or disapproval follows, even after the child knows well enough to carry out this function outside the house. They simply tell the child to go out and finish and hope that next time the child will not make the same mistake indoors.


The tradition for defecation, unlike that for urination, was for women and men to go separately into the cerrado far enough to completely hide themselves. This, of course, did not pertain to small children, who squatted in sight in bushes near the village and between houses with no concern on their part or on the part of older people who saw them.

The method of cleaning the body after defecation was not to wash the area or to wipe it with leaves or paper, but rather to scrape it clean with the strong, smooth central rib or edge of certain leaves. Thus, when adults looked for a hidden spot away from the village, they usually made sure they were near foliage with a firm and sharp but even rib or edge. The waste was left to dry in the sun, or until some animal, usually a pig, came to consume it. Such practices seemed to work in the cerrado countryside where the sun baked feces dry, and where they were scattered throughout a large area because of the long distances individuals had to go to find sufficient cover. (Most small trees are cleared for firewood around any village within two years, so the cover is lighter than generally found in the cerrado.) Thus, unburied feces, their traditional practice, were not a health hazard in the cerrado. Moreover, the Canela bathed themselves twice a day.

Although they carried out adult defecation in private, the Canela were not concerned about anybody knowing they were leaving the village to defecate. In the late 1950s, I often heard people of either sex say, when they passed me leaving the village, that they were going out to defecate: wa ikhwè (I defecate: I am going to defecate). In similar situations, they said they were going out to urinate: wa itu (I urinate: I am going to urinate). Such natural daily salutations, like wa yü: I sit; wa tsa: I stand; or wa nõ: I lie (I am lying down), are the way they say "hello." By the late 1970s, however, nobody referred to their going out to defecate any more as a salutation. They had become embarrassed.


Defecation practices was one of the Canela non-adaptations to the village of Sardinha (Map 3) in the dry forests [II.B.2.g.(5)]. They still followed the principles developed in the cerrado that they merely had to be out of sight and did not need to bury feces. While this often involved walking as many as 500 meters in the cerrado, in Sardinha the distance necessary for cover was 50 meters or less, especially when it was raining heavily. Consequently, feces deposited near the village produced unattractive odors in the bushes and shrubbery.

Accumulation of waste so close to the village contributed to sanitation and health problems. I never saw or smelled such conditions in the cerrado where the waste was more widely dispersed. As I understand it, the Guajajara Indian defecation practices in the dry forests around their Sardinha village were very different. They were more embarrassed about defecation, so they sought better cover and went greater distances. The Canela did not bury their feces in the dry forests, except when they came from a child who had left them in or around a house. Then, a parent removed the feces with a shovel to some place not far away and buried them.

After the missionary-linguist and his family arrived in Escalvado in 1968, they built an outhouse behind their house, which was on the village circle. For the visit of my second wife and three step-children in 1969, we followed the same practice. Possibly because of us, the Indian service personnel tried to teach these practices to the entire village. They required men of every house to build outhouses about 30 meters behind their houses. These outhouses were actually built but rarely used except for urinating, and just when somebody wanted to do this quickly.

The problem with outhouses was that they had to be built properly and maintained carefully or flies would swarm out of them into family kitchens. To prevent this, they had to dig the holes sufficiently deep, and they had to continually refurnish a pile of earth to throw over and thoroughly cover the feces below every time, so light would not reach them. The alternative was to make the hole lightproof. To do this, they had to make a usable cover in such a way that no light could penetrate the hole, and this was difficult but not impossible to do with just the woods and carpentry of the village. Another approach in some areas, though not the Canela one, was to lightproof the outhouse itself-impossible with just the available materials.

While most Canela were certainly capable of obtaining (not making) usable, lightproof covers or repairing the inevitably developing cracks in such a hole cover, not enough of them were sufficiently interested in the construction and maintenance processes to do these necessary tasks well enough. Therefore flies swarmed from their outhouses and invaded kitchens and sat on food. This practice was obviously not a healthy one to continue, as the Canela themselves liked to point out.

Some of my research assistants maintained outhouses to save time—to avoid long absences from our council sessions—but they nevertheless complained of smells that they believed would bring them diseases.


In a society where frequent extramarital sexual relations were the custom [IV.A.3.f], and where the purpose of certain festival acts was to help young women become accustomed to multiple sexual relations with men [II.D.2.e.(3)], sex training of the very young and of adolescents must be especially interesting and unusually important to them.

[III.A.2.j.(1)] Penis Play

Mothers occasionally twist or tweak the penises of their young baby boys. This is done, like the use of a breast, to distract a baby from whatever he might be doing that adults feel is inappropriate, untraditional, or dangerous. It is done to give a baby an alternative pleasure, or interest by indirectly diverting its attention away from what is not desirable [III.B.1.h.(4)].

More often than mothers, however, the infant's other wife category persons [III.E.3.a.(6).(a),(c)] are the most active in teasing or playing with his penis. In my sister's house, one childless young woman in her early thirties, who had been married for a while to one of the baby's older brothers in his twenties, spent a lot of time teasing the baby about his penis. (Since she had been a wife to his older brother, she was a classificatory or "other" wife to him.) Several years later when the baby was 4 or 5 years old, she threatened several times with a knife to cut off his penis. This was done in a spirit of fun, nevertheless he was scared, and sometimes cried. Then she rushed in to reassure him, becoming very supportive.

Research assistants said that sometimes much older brothers pull a baby's penis, and that women may also do this to a tiny nephew. In the latter case, this practice is consistent with the extensive joking relationship that traditionally exists between aunts and nephews [II.D.1.b, c]. Nothing similar was recorded for a baby girl.

Under no circumstances are any attempts ever made to pull back the baby's foreskin, which should remain intact until the time of his first sexual experience [II.D.3.c].

[III.A.2.j.(2)] Masturbation

In babies and young children masturbation is strictly disapproved of and not allowed. If a little boy of 1 or 2 years develops the habit of playing with his penis, his mother gently corrects him. If much older, his mother and sisters are embarrassed and ashamed to approach such matters. Thus, for a little boy of 5 or 6, a grandmother or aunt who has been summoned for the purpose tells him to stop the activity. Parents traditionally do not like to talk about the sexual matters of their children [II.D.3.c.(1)], although they would talk to their children in the absence of aunts, uncles, and grandparents. One unmarried female research assistant [In.4.e] reported that she had had to correct her young son in this way. The Canela can be expected to be flexible about almost any matter.


One reason given for such directness with both girls and boys is that either could lose their virginity payment, the significance of which is serious for girls [III.F.4.b,c.(1)] [II.D.2.a] but slight for boys [III.A.2.j.(5).(b)]. Apparently, some foreskins have to be torn somewhat to be retracted, they say. Sufficient stretching could occur through masturbation so that no tearing occurs during first intercourse.

Just like the girl, the boy is said to lose his payment for the loss of his virginity, if his foreskin can be retracted easily and does not tear when he has his first sexual relations with an older woman. Although this is the traditional position on male virginity, one male research assistant [In.4.e] said his foreskin did not tear upon retraction in intercourse with a classificatory wife at the age of puberty, though no masturbation preceded.

Whether this physical concept of virginity can apply to men of the world in general is an interesting ethnological question. An obstetrician in Washington, D.C., told me that this variation—a tight or loose foreskin—exists among human babies. More research is needed among tribes in which males are naked from birth through puberty, tribes in which no traditional alteration is performed, such as circumcision or subincision, or could have occurred through masturbation. However, it may be too late in the history of the world to find such comparative ethnological evidence.


Canela do not pull back the prepuce and expose the glans in order to clean or inspect it. This is not done generally, although some may do this in a hidden manner under stream waters, they say. A woman or man feels embarrassed to see a glans penis exposed, including men seeing those of other men, except for a man with a sexually contacted classificatory spouse or his wife.

Although both sexes used to go naked in their homelands most of the time, some circumstances were considered embarrassing regarding body exposure: the exposure of the glans penis for men [II.D.3.c.(3)], and the visibility of the inner genitalia for women.

[III.A.2.j.(3)] Opposite-Sex Siblings' Sex Play

Canela socialization is mild in disciplining children; but opposite-sex sibling sex play is one of the two occurrences about which they are quite non-permissive, although never cruel or abusive.

In the context of socialization, I asked research assistants what could be the worst possible occurrence imaginable (short of death or dismemberment) that could occur if a parent were returning home and heard a great commotion in her or his house from a distance. What would the parent fear? The answer eventually after much discussion, was either that two young sons had been fighting or that a young daughter and son had committed some form of incest, the latter being the worse by far of the two possibilities. Actually, incest (Glossary) between young cross-sex siblings was so unthinkable that it was not the first mentioned offense. Incest was not thought of as a possibility, but when I suggested it, the reaction of research assistants was one of extreme dismay.

A mother uses distraction to discourage the usual infractions of tradition, or just frowning disapproval when the misdeed involves a 2 to 4 year old. She would, however, be very severe and scold a daughter of this age who was playing with her little brother's penis or a boy who was exploring his little sister's genitalia, they said. At the ages of six to eight or older, however, such occurrences could precipitate the calling of aunts and uncles to administer more severe punishments [II.D.1.c].

It is considered worse for girls to be caught in opposite-sex sex play than for boys. It is very important for a girl not to lose her virginity so that she can receive her virginity-payment (ganho) [III.F.4.c.(1)] promptly from her first lover (by definition her first husband).

[III.A.2.j.(4)] Adolescent Or Adult Incest

With adolescents or adults, cross-sex sibling incest is punished, according to tradition, by a shortened life, or even by early death for uterine siblings. Another result of uterine sibling incest was that both participants soon become crazy.

The Canela have a story about sibling incest that occurred in 1937, after the time of Nimuendajù in 1936 but before the first residence of Indian service personnel near a Canela village in 1938. In this case, it is believed that the full siblings went crazy after having had sexual relations with each other. The woman died very soon after. The man survived but became so physically dangerous that they had to imprison him in what was called a "pig pen" (a small stockade). They constructed a cage a little larger than his standing body of strong poles made of saplings stripped of bark, put in the ground as posts, and tied together securely. Confined and continuously watched in this stockade, the man soon died. This occurred in one of the two old village sites that are close together on the Raposa stream just below the actual village site where Nimuendaju (1946:33) joined two parts of the tribe during his last visit with them in 1936.

This story is not a myth in the classical sense, although it may become one. In the late 1950s the Canela informants knew the names, time, and place of these events. They showed me the site of the stockade in 1960.

[III.A.2.j.(5)] Sexual Education of Males

Young boys first hear about sexual relations in stories told by adults. Adolescents or grown people do not modify their stories involving sexual relations just because of the presence of young children or pre-adolescents. Sometimes when my research assistant group was talking about sexual relations, small boys sat near us listening with interest. None of my research assistants seemed at all concerned about their presence.

Another way that young boys learn of sexual relations is by hearing the sounds of sexual intercourse coming from platform beds in the rafters [II.D.2.d]. A young wife, or a young woman who has lost her virginity but not retained a husband, is often sexually active at night in her high platform bed. Sexual relations are approved of by the family as long as they cannot be seen—even though everything can be heard. Thus, a little boy can add these sounds to the stories he has heard and understand something about what is going on.

If a young boy has a classificatory spouse of the right age (adolescence to 20s or even 30s), the joking relationship described earlier, which involved her threatening his penis with a knife, might take place [III.A.2.j.(1)]. In this context, particularly with their verbal exchanges rather than just the threats, he is likely to learn extensively about sexual relations.

A young boy also learns about sex from overhearing the ordinary joking between aunts and nephews and uncles and nieces. In my sister's house, when a certain uncle of one of my adolescent nieces came in, they invariably had a sex joking exchange [II.D.1.b.(3)]. One time he said she had a large vagina, like a mortar made from a tree trunk, into which it would please her to have a large wooden pestle grinding. She responded that his penis was twisted at a strange angle and had a black head. This descriptive joking went on between the two relatives, amusing everyone for about 15 minutes.

Inferences from Nimuendajù suggest that parents and siblings of such role performers were embarrassed to hear such verbal exchanges. However, my observations were that such one-link-away kin merely sat or stood by quietly, paying little attention, but nevertheless listening and appearing to be unamused but certainly not embarrassed.

The situation in which many boys learned most extensively about sexual relations was when they spied on couples having sex in the woods. Young boys of 6 to 9 years were often used as messengers between lovers. One research assistant told me he first saw sexual relations taking place when he was such a messenger boy. Since he knew where the tryst was going to take place, he went there, hid, and watched. Considering the freedom allowed boys between the ages of 6 and 12 [II.D.1.c], I have no doubt this kind of sexual learning often took place.


Research assistants reported that homosexual acts did not occur between pre-adolescent boys wandering alone through the cerrado. If discovered, this offense would have been cause for being struck by an uncle. Boys were warned thoroughly about such matters, and of the likelihood of losing their first sex payment if their foreskin became loose.

Turning to associated adult practices, only three men were thought of as being homosexuals during my period with the Canela. Two were in their 60s in the 1970s, and one was also identified for me in Nimuendaju's (1946) volume. Both wore wrap-around skirts like women, though the lower edge of the skirt was a little higher than the knees, instead of well below the knees as women wear them. One of them had effeminate mannerisms but the other one did not. Neither raced with logs when younger but tilled the soil and helped their female kin keep house. The more effeminate one was ridiculed occasionally, but not to his face, so research assistants said. He was married, but his wife required him to leave, even though they had children. I never heard the less apparent homosexual being ridiculed, even though his wearing a short wrap-around skirt made it clear to everybody that he did not intend to carry out certain male roles. He lived with his female kin. Both of these men belonged to the age-set of the Pró-khãmmã of the 1960s and 1970s. However, they seldom sat with the Pró-khãmmã in the plaza in the late afternoon, nor were they active talkers in the council of elders.

Research assistants said neither man was an active homosexual, but that the more effeminate one occasionally allowed Canela men to have anal intercourse with him when he was younger. No tradition existed among the Canela for homosexuals or transvestites to follow. The Canela have no berdache tradition, as do some North American Indians, and they have no myths or stories about homosexual practices. It is an important comment on Canela social relations that the individuality of these two men was respected, and that they did not receive extensive criticism. They were allowed to remain as they were [III.B.1.f.(2)].

One man of the age-set of the new Pró-khãmmã (Glossary) of the 1980s had obvious effeminate characteristics. At the time of my arrival in 1957, he was a late teenager just leaving his childless first marriage, with his family paying the girl's kin heavily [III.D.3.e.(5)]. Then, he went to live at the Ponto Indian service post (Plate 11a) and made dresses there on the new sewing machine, following the instructions of the Indian agent's wife. In 1963 and 1964, he lived away from the Canela community entirely, spending many months as a cook's assistant in hotels, first in Fortaleza and then in Sao Luis.

When he returned to the tribe in 1966, he owned and played one of the first radios. Later, between 1967 and 1969, he married again and had a first child in 1972. Research assistants commented that nobody was very sure that the child was actually his son. The child was more likely the product of several contributing-fathers [III.E.9], they thought, although as the social husband he might have contributed some small amount of semen, but they doubted this, considering him impotent. Nevertheless, this younger homosexual—if he was one—was not conspicuously oriented in this behavior in the late 1970s. He remained a married man, had several children, and did not wear a wrap-around skirt. Homosexuality was more disapproved of by the late 1970s than the late 1950s because of acculturation, a fact that may account for his marrying and not wearing a skirt.


An adolescent male has his first sexual experience [II.D.3.c] at about age 13 with a woman considerably older than himself who wants to initiate her young classificatory spouse, or who might simply like the young fellow.

Older research assistants said that the earlier age for initiating a young man into sexual relations was for the old "wife" to be 45 or 55, but that could not happen these days. Now, the woman would be in her late teens or 20s. Afterward, an aunt of the boy approaches the woman to receive a small payment from her.

One research assistant claimed that a woman's vagina is very hot, and therefore not good for the penis of an adolescent boy. He also said that if a woman took a boy who was too young, he might be so shocked by the experience he would become ill.

A belief supported by all research assistants was that the penis and testicles grew after his first sexual relations, and that these organs matured because of occasional sexual relations. Also, the first nocturnal emission occurred as a result of sexual relations.

[III.A.2.j.(6)] Sexual Education Of Females

Little girls learn about sexual matters by hearing detailed sex jokes, probably not from older "husbands" but rather from uncles. Of course, they also hear sounds coming from platform beds in the rafters (Plate 9b) [II.D.2.d] occupied by an older sister or another female relative.

By the time she is 6 years old, a little girl is closely segregated from boys and has to stay near her female relatives, especially her mother, doing small tasks [II.D.1.c]. The feared potential danger is that a gang of little boys might catch and experiment sexually with her, it was said.

In the late 1950s, a girl started wearing wrap-around cloth by 11 or 12, but by the late 1970s, she began at 7 or 8.

An uncle who is not carrying out a disciplinary role might joke with a niece extensively. One such uncle in his early 30s threw his 10-year-old niece on the ground in the boulevard, in front of all their relatives, and pretended to be having sexual relations with her, thrusting between her legs, to the merriment of everybody, but to the inexperienced girl's embarrassment. There is safety in the kinship relationship, in the protection of clothing, and in the presence of onlookers.


By the age of 10 or 11, a girl received inspections of her genitals from an aunt, or the person who had made it her responsibility to carry out the role of her disciplinary aunt. If it were found that she had lost her virginity without gaining a husband [II.D.2.a] [III.F.4.b] at an age ranging from 10 through 14, her aunt or uncle required her to reveal the name of the man (or men) who had taken it. If she refused to tell, her uncle might have slapped her. This is one of the very few extreme situations in which an aunt or an uncle might have resorted to physical punishment, a practice that was not continued into the 1970s.


Before serving as a girl associate or participating in the various semipublic extramarital situations, a girl learns to be sexually generous through individual experiences. The man who takes her virginity is her husband by definition, if he has fathered no children in an already existing marriage. He remains her husband unless his kin pays for him to leave her [III.F.4.b.(1)].

For several months after her marriage, the young girl is allowed to be exclusively with her husband. Then men, her "other husbands" [III.E.3.a.(6).(a)], begin to seek her out to ask for what they believe is their right, namely, to have sexual relations with her. If she refuses too often, so that female and male groups begin to talk about her lack of cooperation, the refused men organize themselves to teach her to be generous (hà?kayren) [III.B.1.a].

Usually, these men gain the cooperation of a female companion of the "stingy" girl. If the companion agrees that the girl has been stingy, the companion takes her stingy friend out to the woods or into the cerrado to collect fruits, having first told the men where to find them. The men may leave the companion to herself or enjoy her sexually, but a half dozen "other husbands" will force intercourse on the stingy girl, if necessary, holding her down and having intercourse with her in turn. The lesson for the stingy girl is that she must be generous in individual relationships with men. When men desire her, she must give her assets or suffer such group encounters again.

My research assistant groups reported on different occasions that if the "stingy" girl were hurt (bleeding or a bone broken) in such an encounter, her kin could not collect a payment [III.D.3.e.(5)] as a result of an interfamilial judicial hearing. Her uncles would be too ashamed of her [III.D.3.c.(3)] to bring her case to a hearing [III.D.3.a]. Her mother and her female kin would also be ashamed of her but would give her sympathy.

Comparing such group encounters with the Murphys' gang rape among the Mundurucú, research assistants of both sexes say that the Canela "stingy girl" group encounters mature an early adolescent girl into traditional sexual generosity, while the Mundurucú gang rape serves to maintain male supremacy over mature women. "The loose woman who seduces men takes the initiative from them," and if she persists, gang rape by 20 Mundurucú men or more, not by four or five rejected Canela lovers, follows as her punishment for stepping beyond the limits of male control (Murphy and Murphy, 1974:107). Older Canela women participate and take the initiative in sexual trysts as fully as men [III.F.8.b]. Three women giving a male work force "relaxation" in the early afternoon do it out of generosity and to cooperate with the chief [III.D.1.c.(2)], not to enable men to maintain their supremacy.

This kind of training for generous sexual behavior still existed in the late 1970s. It is relatively easy for traditional occurrences that are as inconspicuous and private as this "lesson in generosity" to continue into modern times, because it is not a socially visible activity carried out in the plaza, by a farm hut, or near some other public place. The traditional sexual practices being lost are those that are socially conspicuous, like the Ayrën ceremony [IV.A.3.f.(2)], where the sexes sit on either side of a fence and choose lovers sitting on the other side (Figure 47). The purpose of this ceremony was fairly obvious to an Indian service agent in the 1940s, exerting his "right" to roam around freely [I.A.1] [III.A.2.s.(2)]. Similarly, the festivities during the Wild Boar day [IV.A.3.f.(1)] can be hampered by the presence of backlanders or Indian service agents wandering through the village. A group of young men teaching a girl to be generous is carried out only where it can be seen by Canela individuals. Thus, it can be expected to remain part of the practices by which girls are made to understand what is expected of them as mature women.


By 12 or 14, if not earlier, a girl becomes a girl associate (Glossary) to a men's society, during which time of service she becomes accustomed to group sexual relations with men, depending on the character of the society into which she is inducted [II.D.2.e.(1)] [III.C.9].

A female is not considered a fully responsible adult until she has borne a child [II.D.2.h] [III.F.4.h]. Ideally, childbirth occurs after her belt painting by her mother-in-law and after she has passed some time in the free më nkrekre-re stage [II.D.2.g] [III.F.4.e.(2)]. Thus, before becoming pregnant, she should have been "made tame" (kapônu-re) in a group ambush session, if necessary, and she should have become experienced in sequential group sex through the fun of being a girl associate of a men's group. Through these group experiences she is well prepared for the free existence of the nkrekre-re woman who spends much of her time mixing with men and having free sexual relationships with them even though both may be married.


Socialization largely prevents Canela individuals from being aggressive in tribal life. In earlier times, aggression in times of war was highly valued, but such hostilities were expressed against the outsider [IV.C.1.c.(2),(14),(16)]. This need for extra-group aggression, however, has not existed for over 150 years. Consequently, training in this area cannot be expected to represent pre-pacification orientations.


Aggression between women rarely occurs. For little girls, I noted only a few examples. In the house of my sister, Te?hôk, one of her daughters who was just 2 years old was behaving in an aggressive manner to a smaller girl, a baby from Porquinhos. Her mother did not scold her; she merely put mild negative expressions on her face, and this was a sufficient indication of disapproval to check her daughter.


Second only to cross-sex sibling sex play, the worst possible behavior male research assistants could imagine was fighting between their sons. Fighting between boys is simply not permitted. As with incest taboos, the taboos against fighting are so strong that the parents might call in the disciplinary-uncle to take action. In an extreme case, the parents might even encourage the uncle to hit the boy if his lectures and scolding are insufficient to gain their son's respect [III.A.3.a.(2).(i)]. Some research assistants emphasized that "respect" and "fear" (hũũpa) are very close, so that if some considerable degree of fear is not instilled into the boy, he would not become a fully socialized adult [III.A.4.b]. These same research assistants lamented the loss of control of the uncles over the nephews, a loss that was apparent by the 1970s.

Research assistants pointed out that if the boys were the same size, the situation would not be as serious as if one of the boys was larger and had initiated the fight against the smaller one. They do not tolerate bullies. Causing scratches and bruises was not considered sufficient damage to require legal payments, but if blood were drawn, bones broken, or an eye severely injured, the aggressor's family had to pay a sizable amount after due process through a formal hearing [III.D.3.c.(3),(4)]. Such a hearing and consequent payment are a great family shame [III.A.3.c.(3).(a)].

The taboos against boys or adolescents fighting are so strong and effective that such offences may not have taken place at a serious level during the 22 years I was there, because I never was aware of such occurrences nor did my research assistants report any to me [III.D.3.c.(3)]. The fights reported occurred only between drunk adult men.


In my brother Hàwmrõ's house, A?prol (age 3) was unusually high spirited and willful. He did not get into fights with other boys but was aggressive and demanding of them. Research assistants said that A?prol's behavior reminded them of a traditional belief that if a mother hit her son on the back of the head with a dead bat, he would grow up to be fierce and daring. The only way this bat's spirit could be taken from him would be if a woman, presumably the mother of a boy he had aggressed, were to bite him lightly on the forearm. Then he would become tame and docile once more. Research assistants suggested that A?prol was like the boy who eventually matured into becoming a war leader (hààprãl) in past eras.


A?prol's behavior reminded me of the older Tààmi who perished in the July attack of 1963 [II.B.2.f.(3)]. Everybody knew that when Tààmi was a young man in his 20s traveling out "in the world," he had killed a Brazilian backlander in the state of Ceará. He managed to escape from the Brazilian authorities, however, and returned to his tribe by keeping to the countryside and traveling only at night. When he arrived in the Canela area and sent word in about what had happened, they required him to go through the traditional ceremony of the warrior who had killed an enemy tribesman before allowing him to rejoin daily life. This ceremony included leaping over a dead deer upon first entering the boulevard and a long seclusion outside the village and very severe food and sex restrictions [IV.D.3.d].

In the late 1950s, the older Tààmi also had displayed a fierce demeanor. He did not get into fights, but if he had, he would surely have won through his strong will and compulsion to prevail and through his innate ferocity. Men did not test the aggression-potential of this man, whose temper threshold seemed very low. They stayed out of his way. Research assistants convinced me that the older Tààmi would have been the ideal Canela warrior in earlier times, and that little A?prol in my brother's house behaved in a similar manner.

When boys are in their Khêêtúwayê internment, their uncles come to their two age-set cells (Plate 41a) [IV.A.3.c.(1).(a)] and tell stories about ancient warriors who were fierce when fighting the enemy but who curbed themselves while in the tribe citing Tààmi as a model.


During my time with the Canela, I never saw boys fighting. Fights probably occurred in the cerrado when boys were off in groups exploring and having a good time by themselves, but research assistants reported no fights of this sort to me, maybe because of their forbidden and shameful nature. I believe, however, that generally the research assistants reported almost everything to me.

The Canela orientation is toward non-confrontation in almost all the various foci of socialization. They resort to diversion, not punishment of babies and small children for avoiding unacceptable or dangerous activities. Their parents never confront them with an absolute "no," nor does anyone else who is socializing them, except their uncles at puberty [III.A.3.b.(1).(b)]. Moreover, the society provides many ways to channel feelings of hostility and aggression into other activities. Frustrated young people, depending on the sex, may sing (Plate 32d), dance, run, and participate in log races. These diversions provide many outlets for the young Canela's energies [II.F].


Children are allowed to eat anything they want with the exception of tripe, brains, and the eyes of animals, all of which are reserved for the aged. Other foods that are denied to older persons are not denied to children. They may eat meats that are considered "rich" (encarregado) and are included in the restrictions of anyone becoming a shaman [IV.D.1.c.(2)] or going through the period of their post-pubertal restrictions [IV.D.3.c]. They also may eat liver and kidneys which people at the age of making babies may not eat for fear such organs would make them infertile or sterile.

Children are not required to eat at regular intervals, and no forced or required eating exists. Little boys come and go as they want, and are not usually called to specific meals unless they happen to be present at the time of their serving. Little girls eat with their parents. In the late 1950s, meals were very irregularly served even for adults. People ate when meat was brought in by hunters, regardless of the time of day or night.

These especially permissive eating practices for children continue until puberty, when the lives of girls and especially boys are considerably changed. Then, aunts and uncles require food restrictions of several months for girls [IV.B.1.f] and of several years for boys [IV.A.3.c.(2).(a)].


Besides showing respect for the individuality of children [III.B.1.d.(1),(2)] during the times of weaning and toilet training, parents allow boys to roam quite freely in the cerrado between the ages of 7 and 12 [II.D.1.c]. They play games in the cerrado, such as tossing chips of manioc on top of other chips from a distance, diving into streams to see how much vegetation could be brought back at one time, and climbing trees together. Boys swim much of the day and amuse themselves. For instance, they tie tiny strands of buriti fiber to flying insects or lizards in order to lead them around, and they shoot birds with miniature bow-and-arrows and slingshots. In contrast, girls keep close to older women, do chores, and help their mothers and older sisters with household matters and babies [II.D.1.c].

At the age of 5 or 6, boys carry messages to other houses, fetch water when older people are thirsty, and bring burning sticks with coals for old men to light their cigarettes. Girls begin to help their mothers and sisters in almost any activity they are capable of carrying out. Little girls of seven often mind younger children and babies.

By the time a boy is 6, he might go on hunting expeditions with his older brothers; by age 10, he might go hunting by himself after his father or uncle train him. At 8 years of age, one of my nephews, Ta?pa [III.A.3.a.(1).(a)], went fishing alone and usually brought home a number of small fish for the family to eat. A father might take a boy of 10 or 11 to work in the family fields, depending on the time of year. This is not, however, a consistent practice. Boys are allowed considerably more freedom than girls, approximating the relative amount of freedom allowed adult men in contrast to adult women [II.C.3.d] [II.D.2.h.(2)] [III.F.4.e.(2)].

Continuing the contrast into adolescence and adulthood, boys are allowed to run wild in packs in the cerrado. They are interned (Glossary) in cells during festival situations [IV.A.3.c.(2).(a)], so their age-set leaders (Glossary) and uncles can enculturate them [II.D.3.d]. Girls are confined to the sides of their mothers and older sisters in service to the family household; but the moment their belts are painted by their female in-laws, they are free to have as much extramarital sex as they wish [III.F.8.a]. Men are free for most of their lives to move with their age-set [II.D.3.f], hunt for game, travel out in the world, and chat with each other each morning and evening in the plaza. Women, once children arrive, are bound by their duties to their home and farm. With acculturation, however, men's activities have shifted increasingly to extensive work on their farms for which their earlier psychological orientation provides little support [II.D.3.i.(6),(7)], while women's activities have shifted less—from extensive gathering to more farming.


A number of socialization factors come into play in the Khêêtúwayê (Glossary) festival for prepubescent boys (Plate 41) [IV.A.3.c.(1)]. First is the impact of internment for several months with members of the boy's age-set [III.C.3.a]. This is a boy's first time away from his parents, when he lives with a group of boys ranging in age from 4 to 10 years old [II.D.3.a]. Instead of being under the authority of parents, they are under the command of young men who are not their relatives. If the novices disobey, they can receive several whipping strokes with a thin, flexible wand (pĩ ?-hii-re: wood it-thin-dim.) carried by each of the commandant's assistants.


While learning the new aspects of formal group living, the boys file to the plaza several times a day to sing songs [II.F.1.c.(1)] that attract ghosts. The boys are in danger of losing their lives to these ghosts, except for the protection of their sisters or other female relatives who come to the plaza to hold them from behind by their ribcages as they sing. Thus, the influence of female relatives as protective agents is another socializing influence besides the impact of group living away from home. Uncles also stand behind the line of female relatives and sing. The message conveyed to the boy novices is that if people live in groups, they will be safe, and if they rely on female and male relatives, the latter will protect them from ghosts and presumably from other known and unknown dangers.


More socialization of young boys during the Khêêtúwayê festival takes place while they are interned in the two large cells (Plate 41a). Their being fixed in location, probably against their will, makes it possible for their uncles and certain older people to come and instruct them about the traditions and values of the ancestors, through telling stories. The period of group seclusion, known to the backlander as prisão (prison), stands in sharp contrast with their usual freedom to roam the cerrado.


It is difficult to assess the socialization impact of ear-piercing (Plates 24, 25) [II.D.3.b] [IV.B.1.e], because the Canela no longer practiced it in a regular manner by my time. In contrast to roaming the cerrado in small groups, or internment in the Khêêtúwayê festival in a large group under the command of an older man, the ear-piercing seclusion took place alone in the youth's maternal home in a dark corner separated from the rest of the house by mats. The boy had time to think about his life, and his advising-uncle visited him frequently, teaching him values through telling stories.

The expression for advising and counseling, or lecturing and warning, is to hapak khre (make/do ear hollowed-out-space: to open up someone's ear hole). I associate this kind of activity with creating in the boy socialized responses to hearing and understanding, namely, to receiving orders and obeying them. The Canela are a very order-oriented society [III.D.1.a.(1)] [III.B.1.k]. Attention to the ears and to receiving aural information epitomizes the compliance required in the socialization process.


When it is known that a youth has had his first sexual experience [II.D.3.c], his advising-uncle tells him that he may no longer spend nights in his mother's and sisters' house. After a week's seclusion in his mother's house, he has to sleep in the plaza [III.A.3.b.(1).(c)], or in a future wife's house. If sick, he may return temporarily to his mother's and sisters' house. The simplistic rationalization for this move out of his natal home is that his mother and sisters will be very embarrassed to see his erect penis at night while he was sleeping. Today young men wear shorts or even long pants and sleep clothed so that nothing embarrassing can occur in the house of their mother and sisters. The primary reason for the move to the plaza was more profound, however: the move effected the transfer of relative control over the boy from parents to uncles. Now that the move to the plaza does not take place, the transfer of authority to the uncles is not complete.


In contrast to the Khêêtúwayê festival, the Pepyê festival is post-pubertal in orientation [II.D.3.d] [IV.A.3.c.(2)]. The main focus of the festival is the internment of the novices in their maternal houses, where discipline is exerted through restrictions against the pollutions of certain foods and sexual relations [IV.D.3.c]. These restrictions are to develop and maintain strength for the youths to carry out the prized roles in life [IV.D.3.f]. Maintaining restrictions is a tool for developing one's ability in running, racing, hunting, log carrying, and endurance under the midday sun, and formerly, ability in warfare. When a man (or woman, to a lesser extent) does well in some activity—except for sing-dancing and in agriculture—it is said that he (or she) must have maintained very high restrictions against food and sexual relations during his (or her) post-pubertal period.

When the youths are released from their internment, they continue to associate with those in their age-set. They spend about 6 weeks together camped outside the village, following the same routine every day under the disciplinary leadership of their commandant until the terminal phase of the festival. Sometimes they carry out tasks for the community. Thus, in the Pepyê festival, socialization toward individuality takes place during the internment, but the emphasis toward group solidarity [III.B.1.d] occurs in the terminal period of the festival.


Traditionally, the older generations maintain considerable control over the younger ones. The older generations apply their authority through the aunt-niece, uncle-nephew, and uncle-niece relationships. That the uncles largely or even partly prevented adolescents from having sexual relations with each other is a measure of this authority and control. The public act through which they enforced much of this control was uncles disciplining nephews before the dancing line of women in the late afternoon [II.B.1.e].


One of the most dramatic disciplinary events I witnessed was the hazing that took place in the full ceremony of an afternoon dance [II.E.7.b] [II.D.3.c.(3)]. An uncle came into the plaza dressed as a warrior and summoned a nephew to stand before the line of sing-dancing women, who stopped to watch with impassive attention. The uncle then yanked the youth off the ground by his sideburns (Figures 14,15), which caused the youth much physical pain, as well as shame. Traditionally, not currently, this harassment was given the youth for two reasons. The first was to see if he could endure the shame and pain [III.B.1.e]. The second was to punish him for having had sexual relations with too many young women. This current disciplinary act, however, appeared to me to be perpetrated by the uncle more to satisfy his ego than to teach a youth that his actions were socially unacceptable.

Other disciplinary measures that have been taken on these occasions in the past were that an uncle might pull sharp-edged grass [II.A.3.b.(2).(b)] under the youth's armpits, thereby cutting him; or scrape the youth's back, legs, and chest with the teeth of a coati until blood was drawn. As many as six nephews were disciplined in this manner on any afternoon. Since 1938, however, such disciplinary actions constitute mock situations, because the authority of the uncles over youths has diminished within the society.


The Apanyekra held a hazing ceremony in 1975 in connection with their Pepyê internment festival. Pepyê novices knelt in a row sitting on their heels in the center of the plaza facing their similarly kneeling sexual partners, whether wives or lovers (Plate 37b,c). A lover's Formal Friend may take the lover's place in this act [III.A.3.c.(2).(b)], sparing her or him this embarrassment. Chiefs and elders then gave the group severe lectures for breaking internment celibacy. I have never seen such a severe lecture among the Canela. My Canela research assistants claim their ancestors never practiced this ceremony.


With the arrival of the Indian service personnel to live by the tribal village in 1938 [II.B.2.b], the Canela had to abandon this hazing practice. It was too embarrassing for an agent or his wife to watch a man openly hazing his nephew. These Barra do Corda dwellers called such hazing acts "cruelty" and forbade their practice.

With the abandonment of this hazing ceremony around 1935, the uncles lost a social device that enabled them largely to control their nephews [II.B.1.e.(2)], and, more particularly, they could no longer enforce the prohibition against youths having sexual relations with girls their own age. Once their nephews no longer feared this kind of public punishment, the uncles lost their traditional influence.


The arrival of Indian service personnel in 1938 also coincides with the end of the practice of youths having sexual relations with women approximately 20 years older [II.B.1.e]. The age-set of the older Kaapêltùk graduated through their final Pepyê festival internment in 1933 during the time of Nimuendaju. Their members reported that, according to tradition, they had sexual relations during their post-pubertal period almost exclusively with older women. The graduation of the age-set of Chief Kaarà?khre in 1941 occurred after the arrival of Indian service personnel, and their effect on these puberty-aged youths was considerable. Members of Chief Kaarà?khre's age-set reported they had sexual relations with older women in their post-pubertal period only occasionally. In contrast, research assistants from the age-set of the younger Kaapêltùk reported that they never had sexual relations with older women during their post-pubertal period. His age-set graduated from their final Pepyê festival in about 1951 (W. Crocker,1984a:75).


Another practice that was lost because of the arrival of Indian service personnel was the custom of women without children sleeping with men in the plaza at night. It was too embarrassing to a Canela woman to be found in the plaza before dawn by an Indian agent who was wandering through the village at night. The Indian service personnel were ill at ease when they came to realize that "promiscuous activities" were taking place, and they revealed their puritanical orientation by shaming the Canela about such practices. The Indian agent at the Ponto village post between 1940 and 1947 told me he had seen women sleeping in the plaza and that they did have sexual relations there. The disapproving tone in his conversation with me was obvious.

[III.A.3] Forces of Socialization

Socialization can be seen in terms of many forces impinging upon, or influencing, the individual, molding his behavior and character. I heuristically divide these forces between those that I see to be restraining and those that I see to be rewarding or facilitating. In addition, for adolescents and adults there are forces that I believe to be enabling—namely, those resources or facilities that channel the individual impulses to move forward against the restraining forces or with the facilitating ones. I see enabling forces also to be "props" or instruments that the individual chooses to use, or not use, through her or his own volition, such as restrictions on food and sex to keep out pollutions [III.A.2.q] [IV.D.3.a].

The process of socialization will be divided diachronically into three parts for the purpose of discussion: (1) children, (2) adolescents, and (3) adults.


Both sexes from infancy to a few years before puberty are referred to by the same terms—a?khra?re (children [no ]) reflecting the fact that the stages (Table 9, Figure 16) they are passing through are relatively similar for males and females, as are the forces which are socializing them. A few years before puberty and always after puberty, however, the series of terms for their stages are different, reflecting the greater differences in the forces of socialization for the two sexes and the differences resulting between the sexes—më kuprè (the-pl. girl: girls) and më ntúwayê (the-pl. youth: youths [male]).

[III.A.3.a.(1)] Rewarding And Motivating Forces

Some forces are "positive," rewarding, and motivating (in the sense that they can be seen as moving the individual on through life), while other forces are "negative," in that they restrain and mold the individual. (No moral connotations are meant here.)


In the late 1950s, a Canela mother's primary goal was to take care of, gratify, and socialize her children. To accomplish this, her principal activities were to take care of the house, cultivate and bring food from the farm plot, and keep her family happy.

A particularly good example of this supportive role was demonstrated by my sister, Te?hôk, toward her son Ta?pa (trade: he trades for things). This motivated and energetic youth had a harelip, the only one in the tribe. Since the Canela usually attempted to hide rather than explain away such abnormalities or problems [III.B.1.f.(3)], Ta?pa was in a difficult situation, because he could not conceal the defect. He may have felt he had to do more than other boys to overcome such a handicap. In any case, at about the age of eight, Ta?pa began fishing independently for his family rather than roaming the cerrado with the other boys [II.D.1.c] [III.A.2.m]. He was successful in this enterprise, and frequently brought fish to supplement the family fare. This was greatly appreciated by his mother and father, and Ta?pa was highly rewarded with praise and extra food.

His mother served and apportioned the food at meal times to each of her children, to her husband, and then to herself [III.E.2.e.(1).(a)], usually from a large cast iron pot into small bowls made of gourd. (In later years, dishes or bowls of enameled metal were popular.) Quite visibly, so that all could see, his mother, Te?hôk used to ladle an extra portion into Ta?pa's bowl, expressing her great pleasure with this particular son and his fishing. Although it is unlikely that Ta?pa would have been treated badly because of his harelip by the other boys in the groups that roamed the cerrado, the extra support that Ta?pa received from his mother surely helped him to develop self confidence. Ta?pa later became the age-set deputy commandant [III.D.1.i.(1)] of his age-set.

The father, depending upon his age and particularly his personality, could be merely an adjunct to the family, or he could be quite strong, as was the case with my sister Te?hôk's husband, the older Krôôtô. There was no father-in-law in this family, so the older Krôôtô was the principal male, except when my sister Te?hôk's brothers and male cousins came to dominate a situation. The older Krôôtô played a large role in socializing and in pleasing the children, both in rewarding as well as in disciplining them.

In the house of my brother Hàwmrõ, he too was a significant influence. We were proud of his hunting ability; he often brought home game for everyone to eat. He was a skillful storyteller to the children when they behaved well, keeping their attention upon the narrative by a subtle rise and fall in the inflections of his voice and by the vivid narrative details that he presented. Most fathers are good narrators, whether of myths, stories of the ancestors, or of events that might interest the children.

Fathers make simple toys, often with wheels, including little airplanes. They also make little trucks that can carry small stones, pebbles and rocks. Particularly in the late 1950s, such trucks were in style, and were referred to as caminhão (truck). They are frequently seen in houses or pulled on a string along the boulevard by small boys [II.P.3.a].


Although parents may spend 10 times the amount of time with their children as the aunts and uncles, the aunts and uncles [II.D.1.b] still bring love and attention into the child's world. Except for the certain ones carrying out disciplinary roles, they behave in ways consistent with the Canela values of affection and generosity.

Praise and joking from the visiting aunt or uncle comprise the principal forms of reward for the niece or nephew. Moreover, the uncle frequently tells stories to the children during his visits in the evening, although these occasions are becoming less frequent. By the mid-1970s, the parents were taking the socialization roles over more completely [II.B.1.e]. The visit of uncles after the council meeting in the morning was still a frequent occurrence, however. The socialization uncle must be particularly aware of what is occurring in the lives of his nieces and nephews, and consequently usually appears for a few minutes to discuss matters with his sister [II.D.1.b.(2)]. This was the case in Te?hôk's household.

Children are kept in awe of the disciplining-uncle's presence. They see him coming and going, very often leading or taking a strong part in the tribal council meetings. For instance, Tep?hot rarely joked with his nieces and nephews because he was their disciplining-uncle. He came into the house with a serious expression on his face and was respected by everybody there.

The naming-uncle traditionally gives small bows and arrows (Plate 19a) to his named-nephew, and when these break, he supplies another set, or another appropriate toy [II.D.1.b]. Though they might be seen as toys, the weapons are capable of killing birds. The boy is encouraged to shoot some birds and bring them home as food for the family. Ideally, the naming-uncle takes his named-nephew hunting when he is about 10 years old and teaches him the art of tracking animals. He also tests his named-nephew to see if the lad has the talent and the ability to use a maraca gourd rattle, and to sing traditional songs to its rhythm.

Uncles other than the naming and principal socialization one joke extensively with their nieces and nephews and raise their morale.

A naming-aunt supplies her named-niece with a head carrying-basket and builds a caring relationship with her, as does the socializing aunt. While a naming-aunt's role for a girl is similar to a naming-uncle's for a boy, an advising-aunt has less of a disciplinary role, because much of the disciplining is carried out by the disciplining-uncle, even for a girl.

[III.A.3.a.(2)] Restraining Forces

Some forces are clearly restraining, like punishments, in that they prevent or inhibit individuals from carrying out certain kinds of behavior, while other forces manifest lack of restraint, like not using "bad" names or not restraining a very willful individual. Either case pertains to the use of control or the lack of control, so both are considered here.


Parents (the mother in particular, with the father in a supporting relationship) are supposed to teach children not to be stingy with food and toys, etc., not to become angry, not to hit other children (especially younger ones), not to break household items, and not to lie, though veracity is not highly valued. Simple lying is not a serious form of misbehavior [III.B.1.h.(4)], but telling falsehoods maligning other people is. The really objectionable offenses, however, are stealing, fighting, and sex play, especially with opposite-sex siblings, nor may hostility be shown to any pet animal. Parents are very quick to be severe about their children's eating clay, which practice is treated in Canela myths (see Khrúwapu in W. Crocker, 1984b:195).


When a child is less than 3 years old, and certainly less than 2, the parents are very mild in their verbal socialization. In this case I am particularly thinking of my sister Te?hôk, whose attention to her children seemed to be her continuous activity in life. Her nonstop crooning comforted the baby. By the time her child was 1½ to 2 years old her constant flow of mild words (directly related to what the child should or should not be doing) was taking on meaning for the child. This monolog was in the form of repetitive mild requests with a rising emphasis in the voice, to carry out certain behavior. Training is through verbal insistence with the expectation that the child will comply: an incessant stream of mild talk, never spoken in anger, but often slightly shaming. Situations such as the child's going too near the fire, slapping a younger baby, or getting into a bowl of poisonous manioc juice, are prevented by this kind of continuous talk.


Distraction of the child seems to be the principal Canela technique for channeling actions toward acceptable behavior. They explain this approach as "tricking" the child (i?-hey: it fool; enganar êle: fool it) in order to make it do what they want.

One time when the parents needed the Indian service agent, Hugo Ferreira Lima, to give their 2-year-old daughter, Kahuk, an injection, her father, the older Krôôtô, told Sr. Hugo to wait, saying he would trick the child. He played with her and caused her to smile and be happy, but whenever she saw Hugo's face she burst into tears with fear. Hugo's intention was simply to jab the needle into the little girl's buttocks and leave. The older Krôôtô's intention, in contrast, was to bring the girl into a happy mood and into a position where the injection could be carried out swiftly without her being aware of it. Finally little Kahuk was lying on her stomach on top of the table with her face towards the wall playing with something, and then the older Krôôtô signaled Hugo to perform his duty.

The injection was given in an instant, with the little Kahuk "fooled" into acquiescence. She cried in surprise, but in an instant was in her mother's arms and at the breast. Distraction of the child and then tricking her, while carrying out the act that she had perceived as threatening—all for the sake of not frightening Kahuk and not confronting her will—is the principal pattern of Canela socialization at that age.


In any discussion about socialization with research assistants, Canela parents or grandparents invariably made the point that their kind of child training was very different from the practices of the backlander [II.A.3.d.(1)] [II.B.4]. According to the Canela (and what I saw of Brazilian backland socialization supports them), backlanders call their children by negative names, such as cara feia (face ugly), cara inchada (face swollen), and você é ruim (you are bad). According to Canela thinking, parents or aunts and uncles should not speak negatively to their children or nieces and nephews. (Similarly, I could not express displeasure about a research assistant's work without great offense; but I could say she or he was hindering my research by arriving late.) A person may not even treat pets with verbal abuse. (Dogs are not pets.) An adult had to be positive and indirect in talking to another person. Loud accusing talk that other families in the village could overhear (i.e., public shaming of individuals) is not characteristic of Canela socialization, but some loud conversations, nevertheless, sometimes may be heard between adults.


Another technique that I observed often in the late 1950s amounted to mildly scaring the child (fazer medo: to make fear). One time, little Píyapit, age 4, was crying so hard that she could not stop, and her mother, Te?hôk, was trying to comfort her. In the meanwhile, her father, the older Krôôtô, attracted her attention by picking up a small hand broom and threatening her with it. There was an exchange of conversation between father and daughter, in which the older Krôôtô simply forced his stronger will on her through scary facial expressions, so that she stopped crying.

Parents may threaten that (1) a ghost will come and kill the child, (2) an animal will arrive to eat it, (3) the child will have to go out the next day to work on the farm, (4) strong pepper will be put in its mouth, (5) the child will have to stay in the house all day long and the following day too, or (6) the backlanders are coming. Parents, however, did not threaten so far as to say that (1) they will give their child away to the backlanders, (2) the Devil will come to get the child, or (3) the child has unattractive characteristics. Negative characteristics of the Devil were more absolute [V.A.5.c.(2)] and frightening than those of ghosts. It is thought these days that if such negative outcomes and traits are made clear to a child when, in a moment, God the Protector is not paying attention, then the Devil might come and give a disease to the child, or actually take the child away in death [II.B.4].


Frightening a child out of some form of behavior, or into cooperation, is carried out only to a certain point. For instance the same Píyapit, a year later, was crying so willfully and inconsolably that she could not be stopped. Her parents led her behind the house and left her there to cry as much as she could possibly want. The principle here was that if a parent could not control or dominate a child when it had become too willful, it was better to let it have its way and ignore it than to try to overcome and subdue it, i.e., break its will.

The fierce little 2-year-old A?prol, in my brother's family in Baixão Prêto, was certainly an example of this kind of willful behavior. I saw his parents give way to him many times. One day just after my arrival when I was giving toys to various children, he demanded a miniature truck in addition to another toy I had already given him. I denied his demand because otherwise there would not have been enough toys to go around to the other children in the family. He was so insistent in his temper tantrum that his parents could not comfort him and make him keep quiet. They came to me for another small truck, which they knew I had. I said I could not let them have it because it was due to be given to a child in another family. (Everything had to be rationed fairly, as in a Canela communal meat distribution [III.B.1.b.(2)].) The child nevertheless kept on crying and the parents asked me again for the truck which I finally gave. Thus, the little boy was appeased and calmed himself shortly, having gotten his way, his will not broken.


For certain extreme forms of misbehavior beyond the age of 5 or 6, such as fighting with other boys, especially in cases where the child was being a bully, or in cases of opposite-sex sibling sex play, mild forms of hitting and slapping by parents are permitted. This could include hitting the child on the palm of its hand with an object made of hardwood. This instrument presumably came from backland traditional sources because it only has a Portuguese name, called a palmatória (Table 8, item 92). One research assistant related that a mother sometimes said to a child that if his father had to hit him, he would not grow up quickly the way other children did, and when he had grown up, the father would be ashamed of him.


After the age of about 3 years, parents no longer force children to do anything against their wills. In the particular case of a girl of 4 who refused to allow an Indian service agent to give her an injection, her refusal was respected by her parents.

The question of how to administer medicine—forcing it on a child for its good or not giving the medicine to the child at all—seems to have placed many Canela parents in a position of ambivalence. In the late 1950s parents often refused to give their child medicines simply because the child did not want to take them. They could trick a child below the age of three into taking medicine, but it was not possible to do so without a fuss—even by threatening and overpowering the child—over that age. It was against their principles to even try to do so.

One time my sister, Te?hôk, and her husband, the older Krôôtô, wanted me to give the little Kahuk, 2 years old, de-worming pills. Kahuk did not want to take the pills. After several attempts, Te?hôk held the little girl's nose and caused her to swallow the pills with water, and in the scene that followed, she tried to comfort her daughter as much as possible. But Píyapit, 4 years old, simply refused to take the pills and was not forced to do so.

This issue of forcing children was particularly interesting to observe over the course of acculturation. In the late 1950s no force was used on children over three, but the ways of backlanders and Indian service personnel were influencing Canela behavior, particularly in the administration of medicine. By the late 1970s parents who would not have done so in the late 1950s were verbally forcing children up to 8 or 9 years of age to take medicines these children did not want to take. Consequently, while most forms of socialization were growing less harsh, giving medicine was becoming more direct and compulsory, especially when the Indian agent (Sr. Sebastião) insisted and the parent was convinced he was correct [II.B.2.i.(4).(a)].


Parents carry out the major socialization of their children, but call upon the aunts and uncles when traditionally more severe authorities are needed. I often heard a mother making the threat: "if you do not do this, your uncle will come." The uncles held positions of great authority and respect in the judicial hearings [III.D.3.b], which the little nieces and nephews could observe probably several times a month. Nevertheless, with all of their potential and attributed power, the uncle could speak harshly, or use physical punishment, only once or twice against a niece or nephew. If he were not obeyed, he had to let the matter return to the responsibility of the parents. This trend has been particularly true since the diminishing of power of the older generations during the late 1930s and early 1940s [II.B.1.e].

Canela research assistants made it clear that whereas parents could slap a child or pull its ear as an ultimate act of discipline, aunts and uncles could not touch their nieces and nephews except for the most extreme offenses; the Canela feel that the aunts and uncles do not have sufficient feeling for the child to be able to apply such discipline at the right time.

One of the ordinary occasions during which an uncle could actually strike a nephew was in the plaza before the female line of dancers in the late afternoon when he was performing his wild warrior act [II.D.3.c.(4)] [III.A.2.r.(1)]. The only other time that an aunt or uncle could conceivably strike a child is when it is older than 6 or 7 and involved in cases of sex play or in serious fighting between boys. This avuncular discipline is only resorted to when parents cannot manage the child. Such behavior would have to be extremely unusual for uncles to actually strike a child. Research assistants said such discipline was conceivable but almost unheard of either for current or earlier times. If a girl refused to tell who had taken her virginity [III.A.2.j.(6).(a)]—then the uncle slapped or hit her a few times.


Because restraining forces assume so much importance after puberty, I discuss them first. Then enabling forces, which are scarcely discernible for children but are quite apparent for adolescents, are presented. Finally, rewarding forces, which are more formalized for adolescents than for children, are described.

III.A.3.b.(1)] Restraining Forces

Parents in the late 1950s became less significant in their children's adolescent socialization, although they remained very close to their children. Aunts and uncles took more initiatives in the adolescent socialization of their nieces and nephews because of the nature of the required discipline.


Parents do not like to engage the sexual problems of their children. Considered embarrassing, such topics are not discussed between parents and children (or between siblings, or between Formal Friends), unless absolutely necessary. For instance, when a mother has no relatives to call upon while living away from the tribe, she may have to communicate with her children about their sexual concerns.

Parents also do not like to manage the food and sex restrictions (Glossary) that are imposed on their children, especially the serious and extreme restrictions their sons have to endure just after puberty [II.D.3.c.(1)] [III.A.2.p]. A mother is considered to have too much feeling (hapê) for her children to be firm or strong enough to manage the matters of food and sex restrictions. The father is considered only slightly less soft than the mother. An aunt, however, would have no problem talking about sexual questions with either her niece or nephew within the context of the joking relationships [II.D.1.b.(3)]. Carrying on from this joking role, she could easily manage the actual affairs of a young man with mental agility, humor, and little or no shame.

Besides teaching about restrictions against food and sexual relations, the uncle (Glossary) teaches the youth about various medicines that can remove pollutants from the body. These pollutants could enter the body in the polluted juices of "rich" meats [IV.D.3.a].

Uncles [In.4.i], unless they are the naming- or advising-uncle, have a joking relationship with their nephews; so any of these uncles could manage their nephews' sexual relations and affairs. The naming-uncle and the disciplinary-uncle [II.D.1.b.(1),(2)] are expected to be more formal than the other uncles; nevertheless, they too could manage such matters if necessary. Because the aunts and uncles in their various roles are two or more genealogical links away (Glossary: Further-link kin) from their nieces and nephews (Figure 20) [III.E.2.b], they are expected to take care of their matters more dispassionately.


The very special and characteristic message of the advising-uncle for his niece, and to a much greater extent for his nephew, is that she or he must learn to endure hardships and suffering (both physical and emotional) particularly with respect to family life: awkanã: agüenta (endure!: the imperative form). Although young people surely hear about such harsh values as they are growing up, they are not confronted with self-denial and endurance before puberty [II.D.1.b] [II.D.3.c.(1)].

The permissive socialization of a boy up to the time of puberty takes a sharp turn toward moderate confrontation, so that the youth will be hard on himself and learn to endure the difficulties of war, sports, hunting, and marriage. Thus, potentially, through this externally imposed device for learning moment-to-moment self-denial, he develops a high degree of self-control and may also learn shamanic abilities [IV.D.1.c.(2)]. (However, compare the confrontation above with the general value for adults of avoidance rather than conflict [III.B.1.h].)


Quite pertinent here is the story of strength through virginity of a youth called Pààtsêt. He lived during the last century in one of the earlier village sites of today's Canela area-the village of the first Canela chief, Kawkhre, just a few kilometers south of the old Ponto village near the headwaters of the Santo Estévao stream (Map 3). For some unexplained reason, Pààtsêt lived away from the tribe with his grandmother. Apparently, he maintained an excellent post-pubertal period of restrictions against polluting foods and sexual relations, so that when he returned to the tribe he was strong enough to outrun any of the other young men. He could also race with the heaviest logs, passing all of his competitors.

Research assistants thought he was about 16 but had not had his first sexual experience. However, when he experienced his first sex with a young girl, he lost his physical advantage over the other runners and racers of his tribe and became like any other youth.


In earlier times, the principal confrontation between uncles and post-pubertal nephews occurred when the uncle commanded his nephew to live in the plaza and (1) to visit his female kin only for food, (2) to eat very little food and no "bad" foods, and (3) to not have sexual relations with the young women in the plaza. However, living in the plaza at night, the youth came into contact with childless older women [III.F.4.e.(2).(a)] with whom sex would be possible if it were not for the post-pubertal restrictions against frequent sexual relations just imposed on him by his advising-uncle. The young man has moved from the protection of his mother's house into the permitted-but-restricted sexual life of adolescence. The nearly complete proscription on intercourse with young girls his own age was to avoid acquiring the girls characteristic weaknesses through sexual relations, whereas he could obtain the characteristic strength from an older woman in the plaza [II.B.1.e.(2)]. Even so, he was enjoined to have sexual relations only rarely, and then mostly with the same woman.

His uncles also ordered him to eat sparingly so as to grow large and strong and be able to chase down and kill deer in the noon-day sun. They also told him to avoid certain "bad" foods that have polluting effects [IV.D.3.d.(1)], such as most meats and certain vegetables. If the advising-uncle caught or heard that his nephew was violating these food or sex proscriptions, he would respectively confiscate the food or shame him before the late afternoon sing-dance line of women (Figures 14,15) [lI.D.3.c.(3)] [II.E.7.b] [III.A.2.r.(1)].


After the age of puberty young adolescents begin to rely on, and to have respect for, their Formal Friends [III.E.5]. If a young person commits a wrong upon someone, the Formal Friend of the wronged person may carry out a "game" (hààkrun-tsà: play-thing) against the injurer. Thus a ceremony is performed in which the injurer must repeat his offense against the injured person's Formal Friend. (For more about Formal Friendship games, see [II.F.4].) Fear of exposure through such games helps to restrain improper behavior against another.

[lII.A.3.b.(2)] Enabling Forces

Enabling forces become apparent to young people by the time they begin to do things for themselves, instead of having matters arranged for them by their parents or aunts and uncles. These forces are referred to by the Canela as "their leg-supporting devices" (më ?te-?kaypal tsà), which I have termed alternatively in English as "helping hands" (W. Crocker, 1982:154). The principal facilitating devices of people's roles of this kind are food and sex restrictions, "medicines" (certain herbal infusions), Formal Friends, and opposite-sex siblings.


The inculcation of self-control through food and sex restrictions (Glossary) provides a means [IV.D.3] by which the youth and young woman gain strength to succeed in any of their chosen life roles, except for sing-dancing and agriculture. They may become better runners, hunters, athletes, and log racers, and better at anything for which endurance is important, including marriage. The Canela refer to the food and sex restrictions as i?-te ?-kaypal tsà (his-leg its-supporting instrument: his supporting thing), one of the tools for achievement in life, which is figuratively equivalent to "his helping hand" (W. Crocker, 1982:154).


"Medicines," Formal Friends, and appropriate relatives may also be considered as categories of "enabling forces" to the adolescent. The purpose of some medicines [IV.D.4] is to rid the body of pollutants, and one of the purposes of a Formal Friend is to defend a person from her or his wrong-doer. To a considerable extent, siblings (i?khyê: sibling, leg: includes thigh, calf, and foot) also strongly support each other, especially in consanguineal and classificatory opposite-sex sibling relationships, as emphasized by teknonymy [III.E.8.a].

[III.A.3.b.(3)] Rewarding Forces

For both sexes, achievement in their chosen roles is a rewarding force because members of the tribe, and particularly individuals of the opposite sex, reward the achiever with high consideration and good opinion. The approval by the opposite sex becomes very important to post-pubescent individuals. As in any society, young men and women do many things to attract the interest and the high esteem of the opposite sex [II.F.1.c.(5)]. If a person's role is being carried out well, a person of the opposite sex in Canela society may reward the individual with sexual intercourse [IV.A.3.f]. Public recognition by the community at large and various relatives is also awarded both girls and boys in the form of belts, sashes, bracelets, etc. (Table 8, items 1-8) that they may wear on their persons for all to see.


A girl who performs unusually well in sing-dancing and general good citizenship may be awarded a ceremonial sing-dancing sash of honor (hahï) by her naming-aunt (Plate 58e,f) [II.G.3.a.(8)]. The girl has to continue to sing-dance and do well (Plate 32a) and be the first woman out in the early morning to start the female sing-dancing line and encourage others to come after her. Another award may be earned just after the Waytikpo ceremony [IV.A.3.b.(3)], which is the climax of any of the internment festivals. The Pró-khãmmã may award a girl who performed well a little gourd back pendant (krat-re) of high honor (Plate 59f] [II.G.3.a.(6)]. In other circumstances the Pró-khãmmã may award her a comb back pendant (khoykhe-re) of high honor (Plate 59b,g).[II.G.3.a.(7)]. Girls may wear these awards whenever they go out to sing-dance in the female line or in festivals.

Thus, there are many incentives to motivate young women to perform well, especially during the festival activities. These artifacts of honor are kept in little closed baskets (khaypo) (Table 8, item 62), perhaps for the rest of their lives, to help recall these achievements of high ceremony.


For a young man who carries out his festival roles and daily life roles very well, a naming-uncle may instruct the youth's sister to make a cotton ceremonial racing belt (akàà) (Plates 57c, 59h) [II.G.3.a.(5)]. The youth may then wear this while he is log racing to demonstrate to others that he is highly esteemed by his kin. A more highly prized award is a belt encircled with pendant tapir hoof points (tsù) (Plate 57d) [II.G.3.a.(3)]. This belt also demonstrates that he has performed well and is highly esteemed by his relatives, particularly by his naming-uncle, who presents it to him.

If a young man has done extremely well during certain festivals by sing-dancing and adding to the merriment and general joy, he could be awarded certain of the following ceremonial items by the Pró?khãmmã after the Waytikpo act: (1) cotton wristlets (Plate 60f) [II.G.3.b.(7)], (2) a ceremonial lance (Plate 56e) [II.G.3.a.(2)], (3) a feather bonnet (Plate 56e) [II.G.3.a.(1)]. These items are highly prized especially because women give themselves easily to such winners [III.A.3.c.(3).(j)].


Forces of socialization are usually seen to be operating on children and adolescents. Adults are assumed to have been socialized so that studies of socialization usually stop with the entry of the subjects into adulthood. The view taken here is that socialization, to some extent, continues for the entire life of a person.

[III.A.3.c.(1)] Rewarding Forces

The primary rewarding force for adults is the general high esteem of relatives and friends.


A very important reward in Canela life is the winning of partners for extramarital sexual relations. Since these activities are so widely practiced and were a traditional Canela custom (W. Crocker, 1974a) [IV.A.3.f], the popularity of women with men, and of men with women, is an extremely important rewarding factor in the Canela sociocultural system. When asked why they preferred Canela life to living out in the world, male research assistants consistently said they liked the young Canela women and longed for them when they were away. Women rarely travel out in the world, so the same opinion was not sought from them, although I expect it would be similar. Moreover, women initiate the choice of partners at least as often as men for extramarital sex, research assistants say.


One intergenerational adult bond is maintained by giving and sharing "advice." Members of the older generation advise the younger generation for as long as the old people live. Parents advise their children even when their children are adults. Similarly, aunts and uncles advise and counsel their nieces and nephews until the former are almost in the grave. This is not to say that the older generation kin give the younger generation advice on every little matter, but rather that the older people try to bring objectivity to the problems of their younger kin. Out of respect for the old, many younger people listen to them and do gain advantage from their counsel, even when these younger people are over 50 years old themselves. When the older Mĩĩkhrô [I.G.3] died at just over 80 years of age, his son, Rõõ-re-?hô (Plate 68d) [I.G.11], 45 to 50, lamented that his father was no longer alive to advise him. The extent of this intergenerational advice-giving is pervasive throughout the tribe.


Successful men in their late 30s and 40s are rewarded by the placement of their children into prominent festival positions by the Pró-khãmmã [III.D.2.c]. To have a daughter as a Wè?tè girl (Glossary) [IV.A.3.e.(1)], as a girl associate of the novice festivals [IV.A.3.c.(1).(c)], or as a Pepkahàk girl associate [III.C.3.f],is avery high honor; the girl's or father's kin, as well as the father's extramarital sex favorites, are very proud of such appointments. The younger Kaapêltùk boasted to me on tape about his being chief of the tribe [Ep.4.b.(2).(a)], father of a Wè?tè girl, and father of the Ceremonial-chief-of-the-whole-tribe [IV.A.3.c.(3).(e)]. These statuses facilitated his ascendency to the chieftainship, he implied. Mothers are also equally proud of their son's or daughter's winning awards and high-status ceremonial positions.


There are also prestigious roles for men, particularly that of master sing-dance leader [II.D.3.i.(2).(a)] [II.F.1.a]. A number of grand traditional positions in the festivals exist for male sing-dance leaders (Plate 41c). There are also high political positions, such as chief of the tribe or one of the assistant chiefs [III.D.1.a,f].There is membership in the council of elders for men once a certain age is reached [III.D.2]. If a man is a member of the Lower age-set moiety and therefore a member of the Pró-khãmmã [III.D.2.a.(3)], he is entitled to the honor of receiving and eating prestigious meat pies (Figure 19) from hàmren individuals when they present themselves in the plaza after a social absence [II.E.7.b.(1)]. These Pró-khãmmã councilors are also the persons who reward adolescents for excellent festival performances [II.D.2.f,3.e]. Such formal public high status roles are lacking for women, although there are a few women, like Tel-khèy (principal research assistant) and Katsêê-khwèy, aged 60 in 1970, who assume significant informal (unappointed) roles of leadership in the Festival of Oranges. Nevertheless, women dominate the day-to-day scene in households, because the control of their brothers and mother's brothers is decreasing with each decade. Traditionally, women own houses and farms, although backlanders and Indian service agents refer to such Canela ownerships as belonging to women's husbands.

While the public roles are gratifying, the most satisfying roles of all appear to be those of a successful parent, grandparent, aunt or uncle, who were looked up to by many young kin, although such attitudes are changing.

[III.A.3.c.(2)] Enabling Forces

Adults and adolescents receive the same social reinforcement in guiding their behavior: food and sex restrictions, "medicines," Formal Friends, siblings (especially cross-sex ones), and others, such as following orders. In a sense, an individual can make almost anything (sources of money, farms, etc.) into a "helping hand," but I am referring to the traditional, Canela termed, "leg-supporting devices" [III.A.3.b.(2)].


As with adolescents, a principal enabling force is the extensive use of food and sex restrictions, but adults tend to pay less attention to this helpful device because, presumably, they have already assumed most of their life roles. However, a sensitive shaman and a great hunter, maintain a certain level of restrictions to retain their special powers and vision.


While food and sex restrictions are more of an enabling factor for adolescent development, Formal Friends play a far more significant role for adult social control (see Pepkahàk festival [IV.A.3.c.(3).(f)].


Specific male roles, e.g., shaman and hunter, require specific enabling devices, e.g., ghosts and visions. The "powers" received from a ghost while becoming a shaman [IV.D.1.c.(5)] give him the power to heal. The "vision" acquired from the use of "medicines" enables a hunter to find and attract game [IV.D.4.a] [V.A.5.b.(3).(b)].


Another significant enabling force was the formal order a Canela waits to receive before carrying out an action. For me, this was an especially surprising and notable kind of behavior. No sing-dance leader sang for a group, for instance, unless he was instructed to do so. No festival act could be put on unless the Pró-khãmmã, or some other appropriate group or person, gave the order. Of course, no orders were needed from the outside to initiate personal or family undertakings. To a considerably greater extent than in the Western world, however, Canela individuals expect orders (halkhwa-?khôt: word-following) [III.B.1.k] to come from somewhere, and these are considered "leg-supporting devices" or "helping hands" (W. Crocker, 1982:154). After receiving such an order, a Canela felt well-supported in carrying it out, and he felt he needed this support.

[III.A.3.c.(3)] Restraining Forces For Adults

Restraining forces for adults are many, but only major examples are discussed here.


"Shame" (pahàm) is enculturated into a Canela at an early age. The Canela talk about shame (Plate 48c) as being the inhibiting factor preventing an individual from performing less traditional forms of activity. A person with a high level of shame ("loss of face") would not be caught in a direct lie or an obvious theft because he would be too ashamed. Research assistants spoke of the hàmren (Glossary) individual as being a person of great shame (restraint) [III.C.7] [IV.A.3.c.(3).(e)], that is, their behavior was inhibited by tradition. They spoke of the Clowns as having very little shame (restraint) (Plate 46d) [III.C.7.b] [IV.A.3.c.(4).(b)], meaning that the members of this society were very little inhibited by most traditional requirements. Of course, the degree of difference in restraint is relative because, whether hàmren or a Clown, an enculturated person is restrained by a large number of learned sociocultural patterns.

Formal Friends experience great shame with respect to each other. In the kinship system, shame between the generations inhibits much of the behavior of younger people toward older ones. Being of the opposite sex introduces a restraining influence in most situations, especially when people are affinally related unless they have come close to each other as classificatory spouses. (For further development of similar ideas, see Da Matta, 1982:161–164.)

Canela shame (pahàm) is a broader concept than the English one. In fieldwork, "shame" was discussed frequently, and it seemed to me that most restraints on individual behavior in public and private situations were attributed by Canela individuals to pahàm. Depending on its context, Canela shame includes the Asian concept of "loss of face," the inhibitions and pride of high status people in many cultures, the respect (often due to fear) frequently found between the generations, and the respect (sometimes due to fear of incest) often found between the sexes unless they are permitted to have sex. Canela shame is more of a socially external concept than an individually internal one in its enforcement on a person. It is not "guilt" in the sense of some Western subcultures in which an individual experiences severe suffering and self-recriminations for having committed a wrongdoing, even though friends and the public know nothing about the wrongdoing and are not making accusations. Nevertheless, the Canela individual, in my understanding, has internalized traditional values thoroughly, so that she or he significantly fears potential accusations, especially in the form of gossip and stories, if not in the form of social actions, such as pressure from female kin for women or coercion by his age-set or the council of elders for men. Thus conformity with traditions occurs in most situations.


Formal Friends exert a significant restraint upon the individual not to commit a wrong act against another person (including children and adolescents). When such an event occurs, the primary Formal Friend of the injured person stages a Formal Friendship "game." For instance, if a parent slaps a child too much, or if a child is stung by a bee and becomes very upset, the primary Formal Friend (khritswè mpey) of the injured individual requires that the damage be done to her- or himself in the same way that it was done to the injured person (Nimuendaju, 1946:101–102).

In Escalvado, for example, a baby (who should have been sleeping between its parents) rolled into the fire during the night and was somewhat burned. The baby's primary Formal Friend heard about the circumstances through village gossip and prepared to chastise the parents. She assembled some relatives and friends and they approached the house of the baby and its parents. While a small crowd gathered, the Formal Friend walked into the dwelling. All work in this house and the surrounding houses ceased, and everybody gathered to watch the evolving Formal Friendship "game" (khritswè yààkrun-tsà: Formal-Friend's play-thing).

After the parents had been well forewarned and had thought through their roles and what they were going to have to do, the primary Formal Friend of the baby proceeded with her act. She required that a fire be made in the sleeping area just as it had been burning the night before when the baby rolled into it. She then lay down on the ground on the parents' mats, and talking and joking all the time, required the parents to lie down beside her on either side just as they should have done for the baby. Then she squirmed out from between the parents, presumably as the baby had done, and rolled herself into the fire, actually burning herself slightly. The intent was to become burned exactly as the baby was, though precision is not required but extent of damage is. In wriggling away from the parents and rolling into the fire, the Formal Friend of the baby shamed the parents in a most ridiculous manner. The onlookers, and even the parents, exhausted themselves laughing. (The parents' names are not included here to protect them [In.4.e].)

Then, as part of the traditional proceedings, the mother paid the Formal Friend and her family followers one large bowl of food, a traditional "payment." This usually consists of pieces of meat on top of a considerable amount of cooked rice or manioc.

This Formal Friendship game [II.F.4], as played against the parents, operates as an effective social control device. The parents will be more careful in the future about how they position the fire and the baby before going to sleep. It exacts a payment, that is, a formal punishment [III.D.3.e.(5)], from the parents. Even though the game is carried out in a spirit of joviality, the offenders nevertheless dread this kind of shame "being passed on their faces."

The Formal Friendship games as a Canela social institution are assuming an increasingly greater role in Canela life since acculturation toward Brazilian backland ways of living began. The Formal Friendship role is usurping that of the disciplinary-uncle or aunt, which is greatly diminished today [III.A.2.r.(1)]. Formal Friendship shaming, as a controlling and restraining force, however, is not applied in the same social contexts as was the aunt and uncle's social pressure: The disciplinary-aunt or uncle role kept nieces and nephews from having sex with individuals their own age [II.B.1.e] and from eating "bad" foods so they could grow strong [III.A.2.q]. The Formal Friendship games are not necessarily intergenerational in their application, and they do not facilitate the extramarital sex system [IV.A.3.f] through the reduction of sex jealousy between spouses [III.F.4.e.(2).(b)]. Thus, these Formal Friendship roles may be seen as a shift in societal emphasis of acceptable behavior as new values are adopted from the more dominant culture of the backlanders.


There is a particular story for parents who exert too much pressure on their adolescents. Apparently, some time in the 19th century, a girl refused to marry a certain young man, her mother's choice. In the course of disciplining her for her rebellion, the mother slapped her daughter's hands. The daughter felt so shamed by the punishment that she quietly went out into the cerrado and hung herself from a tree with the cords of her ceremonial belt (Plate 59d) [II.G.3.c.(1)]. Although parents have lost their influence over their children's marriages [III.F.4.c.(2)] and many other aspects of living, this story still serves as a warning to parents not to be too severe in punishment. Similar stories and myths serve as restraints in other sociocultural sectors.


A particular Canela restraint is the fear of being accused of stinginess (sovina: hõõtsè) [III.B.1.a] i.e., antisocial. Not being generous and open is considered such an evil form of behavior that individuals with strong desires may use the accusation to force another person to submit to unreasonable demands. A Canela owner of an axe, for instance, would be sorry for the other Canela who, for some reason or other, strongly desired to use that axe. This kind of strong demand was respected—or at least, resisting the demand generated fear. The fear of being considered stingy, mean, or antisocial is a potent inhibitor. By the 1950s witchcraft scarcely existed, so it was not fear of witchcraft retaliation that made people "generous"; it was merely that they did not want to gain the reputation of being stingy. A person's "other spouses" also did not like a stingy person.


Most Canela, particularly women, are afraid of rumors or false stories being spread about them (tswa-?nã). Adult women could be heard saying that they would not do something for fear of generating rumors. The fear of negative small talk among the Canela is far more inhibiting than I have experienced similar rumors to be in the United States.


An important restraining force for a man is control by his age-set (më hakhrã) [III.C.3.a]. Boys grow to manhood belonging to the same age-set and carry out many of their activities with this group. The approval of the other men in the age-set is very important to the individual.

For instance, in the 1970s when Ku?khop was angry at his wife Píyapit and living away from her and their children for a number of reasons, Tep?hot, Píyapit's disciplining-uncle, but also Ku?khop's age-set mate, went to Ku?khop with several other age-set mates to discuss the matter with him directly. Ku?khop could not retain anger against his age-set mates, with whom he had passed through the internment festivals. Ignoring pleas of age-set mates is difficult and very antisocial (hõõtsè) and it constitutes cutting off a principal source of support for the future. Thus Ku?khop gave in, and after fully voicing his complaints to Tep?hot, returned to Píyapit.


Political pressure constitutes a great restraint on the individual. Although neither the chief of the tribe nor the council of elders has any sort of police force with which to enforce their instructions, individuals feel constrained or coerced to follow their orders. (For discussions on the high degree to which the Canela rely on orders, see [III.B.1.k] [III.D.1.a.(2)].)


The fear of illness or even death from the influence of sorcerers constrained the freedom of adults in the tribe 50 years ago. Curers, or shamans (kay), received powers from ghosts, and almost always used these powers beneficially to facilitate their own relationships in the tribe or to cure people of diseases [IV.D.1.e.(1)]. Sometimes, however, the shamans, became sorcerers and used their powers for antisocial purposes to have their way or to carry out revenge [IV.D.1.d.(2)]. In 1903, a woman sexually refused the shaman Francelino Kaawùy a number of times (Nimuendaju, 1946:240) [II.B.1.c.(4)]. According to the story, he subsequently cast spells of illness (huutsùù) upon her so that she eventually died. One reason Canela individuals were afraid to be stingy was that they were concerned about witchcraft spells being cast upon them if they were not generous, even when the refusal was reasonable. However, by the late 1950s, they feared only moderate sicknesses rather than death, because the existing shamans were not considered very powerful [IV.D.1.d.(1)].


In recent times, Satan, the Devil (Më?pa Yõõtswèn), has a significant role in restraining antisocial activities. The belief in the folk Catholicism (Glossary) that the Canela have been gradually adopting [II.B.3.(2)] [Ep.4.d.(1).(e)] is that God is watching almost all the time, and so is aware of and protecting almost everyone. The problem, however, is that God cannot watch and protect His people all the time. Therefore, when He is resting and not watching, Satan can take what is his due. If a person acts outside of the traditional behavioral patterns, Satan might enter the scene and take advantage of the person, causing injury or death. Thus, if a parent calls a child by a bad name, or if the parent threatens to give a child to the backlanders, Satan may easily enter and take the child for his own (in death) or make the child sick.


Probably one of the most significant factors in restraining behavior, as well as in rewarding it, is the socially sanctioned practice of extramarital sexual relations. Because extramarital sexual relations with persons in the "other spouse" category (më ?prõ ?nõ: plural wife other or më hõ ?camarad: plural possessive comrade: their comrade) (Figures 29, 30) are so easy and frequent, they have a great impact on Canela behavior patterns. A Canela does not do something very often that will risk her or his popularity with the opposite sex in her or his classificatory spouse category [III.E.3.a.(6).(a)].

It is clear that the human being throughout the world is motivated to a considerable extent by the opinion of available members of the opposite sex. For the Canela, the possibilities of having sexual relations outside of marriage far exceed such opportunities in most other cultures (W. Crocker, 1964, 1974a). Even in cultures where extramarital relations are condoned, they do not usually receive the public support of traditional festivals that they do among the Canela [IV.A.3.f].

With the Canela, both women and men talk to each other in their same-sex groups about their sexual experiences: women by the streams, men in plaza age-set positions (Figure 24). Consequently, much of what goes on between classificatory spouses becomes known to the entire village except to the married-to spouse, who does not want to hear and is not told. When discussing sexual techniques with female and male research assistants, they spoke about how relatively quickly certain men came to orgasm (number of strokes) and which women "cried" or were silent, and which became "wet" quickly.

Similarly, ungenerous, antisocial behavior becomes quickly known throughout the society and affects the behavior of individuals in the classificatory spouse category toward each other. Fear of losing popularity with members of a person's "other spouse" (classificatory spouse) category is the most effective Canela (not Apanyekra) cultural restraining device, according to my assessment. (The Apanyekra extramarital system is not as extensive.) However, acculturation has already reduced (1950s) and is diminishing (1980s) the viability of the extramarital system, because extramarital relations are slowly becoming abandoned as a festival-sanctioned system. The spread of folk Catholicism [Ep.4.b.(2).(e)] and SIL Protestantism [Ep.5.d] will eventually almost eliminate this formerly extensive sociocultural sector.

[III.A.4] Ethnotheories of Individual Development

The Canela were not very verbal about the underlying reasons for their behavior. Communication about ethnotheories of any sort was most fruitful between 1974 and 1979 because by those years the powers of analysis among research assistants had become more developed. Because almost all of my research on socialization was done in the late 1950s and 1960, it is largely lacking ethnotheories about socialization. Two very important viewpoints emerge, however, which seem to be in contradiction with each other. But to find oppositions in these kinds of materials is not unusual.

The first view is that a parent is a caretaker, or trustee, in bringing up children. The pertinent belief here is that children are born with their personalities already immanent, and that the parent's responsibility is to provide the proper environment within which the child is to grow up. Consequently, the parent must not take a very active role in the child's development, because the child matures through its own inner forces.

The other view is that if parents are not severe enough with their children, scolding and occasionally slapping them on the hands, the children will grow up soft and unable to endure the hardships of life, and will therefore be selfish and willful as adults.


To the extent that the future personality of the mature adult is inherent in the infant when it is born, the child's parents do not have to play a very active role in the socialization of the individual. This belief may be the reason why mothers are so permissive in feeding, weaning, toilet training, and various other foci of socialization. It may also account for parental behavior when a particular child is willful and persistent, eventually giving up and letting the child have its way.

The exceptions to this permissive training lay in the areas of sex play between opposite-sex siblings (and homosexuality) and fighting between siblings and among boys in general. In these two cases, the issues are confronted and totally controlled. No fighting is permitted in Canela life, that is, within the tribe. Sexual relations, even though extensive, are proscribed within certain limits, or the opposite-sex kin of an individual would be seriously reduced in number [III.F.3]. An accusation often found in joking relationships is that the other person has no relatives (i.e., has turned them into "other spouses"), meaning she or he has no kin of the opposite sex (a-yuukhyê naare: you-kin not: you have no kin). But with respect to other sectors of socialization, the Canela are permissive toward their children until puberty. Thus, prepubertal socialization appears to be consistent with the parent-as-trustee ethnotheory for socialization except for the expression of sexuality and fighting [III.A.2.j.(3),k.(2)].


"Respect for older people comes from the fear of being slapped or hit." This statement was made emphatically by one of my research assistants and later accepted by the whole research assistant group. At another time, an important Canela research assistant said that if you did not put fear into a child, it would not obey you and later would become selfish, self-centered, and willful: stingy and antisocial [III.B.1.a,b]. This second ethnotheory of socialization suggests not only that there is an active role for parents, but that if the parents, aunts, and uncles are not strong and severe in certain respects, the child and adolescent would grow up to be an adult with entirely unacceptable personality traits.

Consistent with this second ethnotheory is the general belief that firstborn children are likely to be soft, selfish, and willful because their parents have too much feeling and concern for them and let her or him become spoiled. Subsequent children, especially the last child, are believed to be the "best" because their strong desires have become blunted by older siblings and by the discipline of the parents who no longer feel so strongly for their children that they can not impose the necessary discipline. Research assistants presented a number of examples to prove their points. The most obvious one was a case in the house of Te?hôk. Her oldest child, an otherwise healthy adult male, was very weak (só andando: only walking: barely moving around), while her other son, the many years younger Mĩĩkhrô, was quite active and able as well as disciplined. This is often, though not always, the case in other families. In the house of Hàwmrõ in contrast, the oldest child, Hômyï-khwèy, a woman, was the most admirable of her many siblings, as evidenced by her reliability, imagination, and willing service to her family.

If a man behaves badly and hits his wife, it is said that his parents must have spoiled him; that they were not sufficiently harsh with him and did not hit him when they should have. These after-the-fact statements are typically Canela, and recall their classical statement that if a certain person succeeds well in activities that require endurance, he must have maintained very high food and sex restrictions just after the age of puberty. This relationship between food and sex restrictions and endurance or fortitude can be thought of as an ethnotheory of growth but also of the development of self-confidence, self-control, and willpower.


One way of finding out how a people see life is to ask them to draw its course in the sand of the ground, birth to death. Then, the investigator can notice the words they use in describing their drawings. In addition, if the drawer is able to do so, she or he should carefully define the key words used. I found the key expression më haalíya to mõ (they in-steps with move: they progress through life in steps) described the Canela concept of movement through life: in steps up and down. The sand drawer made the steps slope upward to mid- and to old age, and then let the steps slope downward. When asked about this "progress" for different human characteristics, the peak of life's rising and falling varied. For running, the peak is in the 20s or 30s, depending on the kind of running. For wisdom, it is in the 50s or 60s. All these progressions through life, however, follow the same pattern—sloping steps up and down—except one: shamanism. The kay is said to move up sharply with each visitation from a ghost but not to rise between visitations so that his steps of progress are flat. (For a portrayal of these stages of life, see Figure 16; for commonly used expressions identifying and describing these stages, see Table 9.)

[III.A.5] Summary

Canela children are relatively free of crying, temper tantrums, and the antisocial and destructive behavior found in many societies. Canela children obviously do cry, but childhood is relatively free from obvious expressions of hostility. Most children are active and vibrant, joyous and amused. Their faces are full of expression, both positive and negative. They are a delight to their parents.

In the late 1950s, if a baby or child cried continuously for a significant period of time, the parents thought of themselves as being a source of the problem by having broken current food and sex restrictions or certain taboos parents must keep until the child is over 2 or 3 years old. However, beliefs in the significance of food and sex restrictions in general, and in parental taboos for children in their first 2 or 3 years (Table 9, stage 11), were reduced by the late 1970s; thus, the causes for a baby's crying were viewed more directly. Mothers looked more for immediate causes, such as frustrations or willfulness.


The relative calm and joy among children and babies is probably due to the lack of frustration during breast feeding, weaning, and toilet training [III.A.2.b,f,i]. The Canela's tacit and overriding belief is that parents should do everything they can to avoid frustrating the child or causing it to cry. They attempt to make the child want to do what they want by talking to it with great patience, love, and consistency [III.A.3.a.(2).(b)]. They distract or "trick" the child [III.A.2.f.(1)] [III.A.3.a.(2).(c)] to remove it from a dangerous situation on its own volition.

Forcing a baby [III.A.2.c] and slapping the hand of a child [III.A.3.a.(2).(g)] are techniques of last resort, done only for its protection or because every other less serious attempt has failed. If this harsher course fails, the situation is turned over to aunts and uncles. However, if the aunts and uncles have no quick success, the case is returned to the parents [III.A.3.a.(2).(i)], where the parents may let the child have its way [III.A.3.a.(2).(f)]. I seldom saw children in temper tantrums, but when I did, parents let them alone to cry themselves out. Again, confrontation with the child is avoided so hostilities are less likely to build up inside the child.

The mother and infant are in very close contact almost all the time because a mother carries the baby around in her arms [III.A.2.b]. She is almost constantly sensing, checking, and rewarding her baby [III.A.3.a.(2).(b)]. She carries out this intimate training in small, subtle ways, especially before the child is given freedom to move around on the ground inside the house.

In weaning and toilet training, mothers do not frustrate their babies and small children. In toilet training, the child is restricted only after it can understand what is expected [III.A.2.i]. Then, little by little, the child is induced to take care of itself completely. Thus, the child is permitted to develop relative independence and self-reliance. This is particularly true for the small boy and later for the larger one who has great freedom to run in the cerrado with the other boys [III.A.2.m].


Overall, Canela socialization may be seen as encouraging relative dependency [III.A.2.n,o,q]. Babies and little boys, but especially little girls, are kept quite dependent on older people in authority. For the infant, the breast was given at any moment so that the baby becomes dependent on being rewarded at any time there is a problem instead of learning to wait and endure the frustration [III.A.2.b]. Little children are taught to "beg" things from their uncles and persons of authority in the tribe as well as from outsiders. These customs develop the expectation of having to depend on orders from other people before carrying out many activities [III.A.3.c.(2).(d)]. As a Khêêtúwayê novice and as a member of an age-set between festivals, little boys are thoroughly socialized into taking commands from the leaders and older boys in the age-set [III.A.2.q]. Thus, most Canela children grow up to be dependent on authority figures.

Girls receive more dependence training than boys [III.A.2.m], and the boys who are younger when joining an age-set receive more dependence training than those who happen to be older when they join an age-set [III.A.2.n].


Most Canela arrive at puberty having faced relatively few frustrations. They have strong positive feelings that need to be expressed and satisfied, great expectations of being cared for that need to be fulfilled, and relatively little value placed on self-control with which to contend with frustrating situations. At puberty, the aunts and uncles assume much of the responsibility for continued socialization [III.A.2.p] [III.A.3.b.(1).(b)]. The disciplinary-uncle is responsible for enforcing relatively harsh socialization measures, especially on the boy.

Canela research assistants were very clear about the changed atmosphere after puberty for the boy, though for the adolescent girl there is more continuity with her former patterns of socialization. The uncles' duties included keeping the young men from having sex with girls their own age almost entirely, allowing each sex to have limited sexual relations, mostly with considerably older individuals [III.A.2.s.(1)]. They also kept boys from eating most "bad" foods: rich gamey meats [III.A.2.q]. The directing of the sexual activities of young people was verified by the Canela research assistants, young and old who were alive at the time and who were very clear about the nature of extramarital relations prior to the arrival of Indian service personnel [III.A.2.s] and earlier. Research assistants emphatically stated that it was the aunts and uncles who suppressed the jealousies of young husbands, and even adolescent girls. However, after the first postpubertal period of near-abstinence from sexual relations, there was, almost as a reward, a second period (më nkrekre-re) [II.D.2.g] of free mixing with available members of the opposite sex whether or not a person was married.

The main theme of the Opening Wè?tè festival [IV.A.3.a] serves to place even older women and men than the më nkrekre-re stage into this mental and emotional orientation even today: during the festival summer period, everybody is going to amuse themselves with available members of the opposite sex who are not their own spouses (Figures 29, 30).


In the late 1970s, the Canela continue to socialize their children in very much the same permissive way, though with somewhat less emotional closeness and patience. During puberty, however, they cannot discipline the adolescents with the same strength and harshness as was once the custom.

The traditional uncle-to-nephew hazing ceremony in the late afternoon before the line of sing-dancing women became impossible to carry out once Indian service personnel had arrived in 1938 to live beside the village [III.A.2.s]. Consequently, the relative transfer of authority from parents to aunts and uncles at the time of puberty has been reduced and aunts and uncles have lost their power, relatively, to influence young people significantly.

Another factor that has limited the partial transfer of authority from parents to aunts and uncles at puberty has been the fact that since about 1947 [II.B.2.b.(1)], the Canela have experienced difficult economic times. They have not produced enough food to be self-sufficient in meat and farm products. The result has been that between the months of September and December, or even August and January, many families have had to leave the village to settle temporarily first in one backland community and then in another, living beside the houses of backland Brazilians [II.C.3.g]. In such a social setting, which consisted of from 20 to 50 percent of the year depending on the family, aunts and uncles were not present to continue or assume the socialization of their nieces and nephews when they reached puberty. Aunts and uncles had to be with their own children in another backland community rather than with the children of their opposite-sex siblings.

This factor of geography (the distances between the various backland farm locations and the Canela village) prevented aunts and uncles from tending to and enforcing the traditional food and sex restrictions on their nieces and nephews [III.A.3.b.(1).(a)]. Parents traditionally were not used to taking care of such responsibilities, being too embarrassed to carry them out. Thus, food and sex restrictions often were inconsistently maintained by adolescents, many of whom believed that as a result they would not be able to run and hunt effectively. In my opinion, such adolescent beliefs easily became generalized to other activities, such as farm work and festival participation.


The Canela of the 1970s are a people with strongly developed feelings who characterize themselves as having insufficient self-control. In my opinion, Canela adolescents and young men in their late teens and early twenties range from being full of life and activity to being listless and detached. They are not controlled by their chiefs or by the older generation to the extent that they had been formerly [III.D.1.a.(4)]. It was through the use of food and sex restrictions that the ability to overcome hardships and the stresses of daily life (boredom, hunger, strains of racing, problems of marriage) had been developed [III.A.2.p]. Backlanders and Indian service personnel, however, do not believe in the efficacy of these social practices and have helped to undermine the Canela's use of them.

With the change in power relationships between the older and younger generations, frustration-tolerance training has almost disappeared, especially after puberty. This accounts for much of the disorganization in the tribe (especially concerning authority) during the 1950s and 1960s in particular.

Comparing the Canela with another Northern Gê tribe may be helpful. Traditional individual frustrations among the Canela appear to be less than among the Kayapó. The Kayapó's rop-króre kam aibãn state [II.E.9.b] and its consequent hostile and uncontrolled running around the village are inconceivable forms of behavior for the Canela. When in this state, Kayapo individuals become wild, damaging themselves and others (Moreira, 1965). The Canela may not need this emotional outlet—a catharsis—because they are less frustrated in their training before puberty and because their recreational outlets are so numerous and well developed [II.F]

While there is a balance between dependence and independence on the part of many individuals, it is clear that most Canela are relatively dependent and non-bellicose. It appears to be the willful children who, if socialized strongly after puberty, become the strong leaders, but these are few [III.A.2.k.(4),(5)]. The willful babies and children too often grow up to be dissidents, instead, in the sense that they are listless and idle most of the time rather than working with any sense of direction. This is quite understandable when their acculturative history is considered.

The Canela have a highly structured system of recreational outlets [II.F], which may be viewed as one of the stabilizing forces within the society. These outlets give the young a sense of identity and self-discipline, so that the ravages of acculturation may not be as devastating for them as it has been for other Brazilian tribes.



The analysis of certain Canela values, personality traits, and social behavior is necessarily a subjective approach. There will be no attempt to make distinctions between the theoretical orientations provided by various schools of psychology or psychological anthropology, or to define terms like "values," "polarities," and "dichotomies," because there is no intention to treat this subject matter in depth or to formulate novel theories. As with the chicken and the egg, these orientations may be seen as being either sources or products of the socialization process.

[III.B.1] Valued Orientations

In doing studies on valued orientations and personality characteristics that contribute to forming behavior, I found that the research assistants preferred to answer questions in terms of "paired" opposites. It was easier for them to explain one side of the issue by studying the other side at the same time, an aspect of their characteristic dualism [V.A.2]. My research assistants and I also found that besides conspicuous paired oppositions, there are also conspicuous paired polarities that are "complementary" in their relationship to each other. The following set of valued orientations with their complementary or oppositional polarities will be taken in the order of the openness with which the Canela can talk about them.


When Canela individuals are asked what qualities a good Canela should have, their first response usually is: a person must be generous (hà?kayren). The worst characteristic—its opposite—to display is almost always "stinginess" (hõõtsè).


If asked for examples of these traits, the Canela are most likely to say that the backlanders are stingy and that the Canela are generous. This means that the backlanders keep their possessions and will not give them even to people who need them badly. The backlanders, they point out, store a number of months' worth of staples. Then when strangers pass by, or even friends or relatives, they will say that they have no food even when the other person is very hungry. This kind of behavior is anathema to the Canela, who are particularly anxious about food [II.C.3.g].

In the late 1950s there were stories about a backlander in the community of Jenipapo do Resplandes who was so stingy that he would not obtain the proper medicines for his ailing grandparents. Although in the 1970s the Canela had more understanding of the backland position on the storage of foods and expenses for relatives, in the late 1950s they simply believed that if a backlander had food and did not give some to people in need, he was evil [II.B.2.b.(2)].


In the late 1950s, the Canela were almost always "generous" when confronted with a situation that called for a display of generosity. For instance, when a family man had killed a deer (Plate 15c) and hungry people came asking for some venison, a piece was quickly given to them. When asked, the man responded by being generous. Nevertheless, he usually tried to hide the fact that he had deer meat to avoid the requests of others. The generosity of the Canela family man, therefore, was more practical than altruistic. If he could, he would keep as much as possible for his family. The Canela accusation against any people of the region who say they have nothing to give is usually that such people must be hiding what they have. After all, this is what everybody does except the unusually generous person.


A female or male of high standards would place ample provisions at the disposal of others when a festival situation called for this kind of behavior. The person who believed himself to be generous would under such circumstances supply more than was generally expected. Thus self-image, how a person wants other people to perceive him, is very much a part of the Canela practice of generosity. An individual who is stingy and demanding may be suspected of becoming an evil shaman capable of casting spells of fatal illness. Thus the fear of the accusation of being a witch encourages generous behavior [III.A.3.c.(3).(h)] [IV.D.1.d].


The implication of generosity that seems extreme to outsiders is that both women and men must be generous with sex. The socialization for sexual generosity is extreme [II.D.2.e.(3)] [III.A.2.j.(6).(a),(b)]. It could be said that in the Canela world a person's sexuality is not her or his own to dispose of as she or he chooses, and that those in the tribe who desire to have access to this person's body are permitted to do so.

Other advantages that an individual might possess must also be shared. The sing-dance master [II.D.3.i.(2).(a)], the great hunter [II.D.3.i.(6)], the woman with a fine voice [II.E.4.a], and the visiting anthropologist who has a lot of paper for rolling cigarettes, all must be generous with the things that other people want. The institution of "begging," which is in keeping with this thinking, surprises and alarms outsiders. This "begging" is traditionally sanctioned, so that the word for it does not have negative connotations. Begging serves as a social leveler by redistributing material wealth, and it is an important part of the food distribution system. Canela individuals approach and simply ask for items that the outsider possesses, fully expecting to receive a portion of them as their right. (This trait is also evident among the Shavante, Maybury-Lewis, 1965:172.)The very definition of a good person is one who gives many things to the Canela when they ask for them. The institution of "begging" (a-?nã wè: something-for ask) is supported by a number of festival acts, found particularly in the festival of Masks [IV.A.3.c.(5).(c)].

In theory, women are allowed to be more stingy than men, because women must conserve, ration, and distribute family possessions and must think about the welfare of their children [II.D.2.i.(1).(5)].


The next most esteemed and expressed Canela attribute is the ability to feel and care for (hapê) other people. Its opposite is më kumã i?yapê naare (them in I-care not: I have no feelings for them), of course, but negatives in general fail to add to our knowledge of a concept. Another opposite, the expression yum kuuni (a-person whole) tells us more. This is a person who is whole (without openings) so that her or his feelings are contained or blocked. Thus, they do not let their feelings go out to others because they are inhibited.


In the late 1950s, favor for this human attribute (caring for others) was conspicuously manifested in a number of ways. The Canela said the backlanders were bad because they did not feel strongly or care enough for other people—including relatives and even their own children. Nuclear families were concerned only about their members. There were many stories about the backlanders and how unfeeling they were with respect to friends and relatives not living with them. The Canela told stories of this sort in the plaza at night to amuse each other [II.E.8], and probably to raise their self-confidence.

In 1964, when my wife Mary Jean saw that the six-year old daughter [In.4.e] of my Canela sister was holding and hiding her semi-paralyzed left arm, my wife cried quite visibly and the Canela noticed this. Here was a city person who cared and had feelings, they said. Nothing else really mattered as much; in their eyes Mary Jean was a wonderful person. Not to have feelings or not to care was akin to being stingy, which is part of that complex of negative characteristics the Canela express in the term hõõtsè.


Another aspect of caring among the Canela is the leaders' concern that everybody receive a fair portion in any distribution. The apparent sense of fairness and justice is supported by feelings of concern for the person who does not obtain her or his portion or who does not receive anything at all. For example, it is easy to cut and apportion meat and to divide rice or manioc flour into as many piles as necessary for each individual or family to receive its due (Plate 15b,d). However, some shared items are not so easily apportioned. When I was trading [I.C] with iron implements, it was not possible to divide a machete or an axe among several people who might want it. There could not be an iron tool for everybody, only one or two for each family, as was agreed to in a tribal council meeting. However, I was considered unfeeling if I did not have items of this sort for certain individuals who felt they must have them, even if this exceeded their family's quota as I had set it. If Canela individuals presented themselves to me and wanted an item strongly enough, fairness was not of primary importance; concern and feeling for other people came first. Rules previously agreed-upon between the council of elders and myself had to be broken because iron items (ferramentas) could not be supplied to everybody; the degree in intensity of the requester's feelings would require that he be given the axe he wanted [III.A.5.a].


Similarly, care or lack of care for the feelings of other people overrides the concern for "telling the truth." If the Canela feel strongly for the other person, they may not tell the truth if it would hurt their feelings, even slightly [I.G.4]. Truth is not a value of great concern among the Canela. Indirection is their usual means of communication [III.B.1.h.(4)], because it is more important to be concerned about people's feelings than to tell them the truth (Epigraph).

In the late 1950s in Barra do Corda, I sometimes bought meat for Guajajara Indians [II.B.2.g.(6)] who were staying at the Indian service agency, as well as for Canela individuals who happened to be in the city [II.B.4]. The Canela typically divided the meat into as many portions as there were Canela in the city and reserved each portion for the intended recipient's return to the agency for as long as a day. The Guajajara, in contrast, divided the meat among only the individuals, or families, who happened to be at the agency at the time I gave them the meat. The Canela saw this tribal difference, which was well recognized, in terms of the care Canela had for one another, and they spoke against the Guajajara for their lack of feeling for each other.

[III.B.1.c] Joy And Fun Versus Sadness And Introspection

The third most valued orientation most openly spoken about is joy, or having fun (amyi-?khin: self-liking: euphoria) [II.E.1.b] [IV.A.1] and a sense of humor (hapak-tu tsà: ear-swollen activity). The opposite of euphoria is weakness: më ?pek (those weak: the people who are weak and sad). Such people are motivated by little self-liking or self-esteem.


Fun is seen most obviously in the performances of the several joking roles: between a classificatory mother's brother and his "sister's" daughter, between a classificatory father's sister and her "brother's" son, between Informal Friends, and between classificatory spouses. (Joking between immediate consanguineals and married spouses is muted.) People in these classificatory roles are almost required to joke when they meet each other, unless they happen to live in the same house. Such joking is most extensive and enthusiastically expressed between the sexes and usually involves descriptive, explicit sexual references, the exception being between Formal Friends. For discussion and examples, see [II.D.1.b.(3)] [III.A.2.g] [III.A.3.a.(1).(b)] [III.E.3.a.(6).(c)].


Such a spirit of fun is also expressed by an individual for her or his Formal Friend when the latter is honored on some ceremonial occasion. The individual dances and goes into unusual comic behavior (i-?khay-nã: it-off-condition)(Plate 39c,d,e) for the Formal Friend. Then again, when an individual's Formal Friend has been accidentally injured—being stung by a wasp or having rolled into the fire at night—the individual stages the Formal Friend game [II.F.4] [III.A.3.c.(3).(b)]. The Canela predisposition to enjoyment makes them seem like children to the Brazilian backlander.


Sadness and introspection is the opposite of amusement. An individual is not allowed to sit alone and worry about her- or himself or think introspectively. If anyone is brooding in this way, some person in an appropriate role is likely to say ka ay-khãm pa: (you self-in listen) to bring the individual out of the inwardly-centered state. The Canela value a constructive, outgoing disposition that is involved with others. To be inward-thinking is associated with being stingy, angry (in-krùk: she/he/it-angry), and selfish (amyi-á-?khôt: self-superlative-following) [III.B.1.k], all antisocial characteristics from the Canela point of view. Moreover, such an individually inward orientation is associated with revenge, which they consider equally despicable.


A similar psychological orientation to that of enjoyment is the Canela love of activity and movement (hà?khritkhrit). Ideally, something should be happening all the time [III.D.1.a.(3)]. Somebody should be singing around the boulevard or people should be actively working at productive tasks. Activity and movement are the opposite of sitting alone and thinking of problems. Most activities should involve group participation [II.D.3.d.(1)] rather than an individual pursuit; a lone individual singing around the boulevard is directing his efforts toward the people in the houses [II.F.1.c.(5)] (Plate 32d).

For most of the tribe, the Canela way of life is full of activity and movement, with little time to rest (Table 7). Potentially, in a given day there could be three daily dances, two tribal council meetings, two age-set meetings, a judicial hearing, a log race from outside the village, two or three log races around the boulevard, and a number of footraces and athletic sporting events, besides all the daily work a person must do to keep a family properly maintained. This is a considerable amount of activity. This balance of various activities partly explains why the Canela consider "work" or "labor" [II.E.5.g] only one part of their social responsibilities.


Besides enjoying themselves and being active, Canela individuals are supposed to function and work in groups whenever possible: më kutom to mõ (they massed in go: they go along massed in a group). The opposite is simply yum ayte-á nã mõ (someone alone-superlative state goes: someone who lives always alone). Research assistants frequently raised the point of the need to always move in groups when we were studying the Pepyê festival in 1957.

A Canela moves from one group to another all day long, rarely working alone. At a young age, small boys are "captured" and placed in a festival situation (Khêêtúwayê) where they are interned in two large groups away from their family homes [III.D.3.a] (Plate 41a). They are said to be very similar to wild boar during their training period, running together, keeping themselves apart as a group, and always maintaining activity, except when in their cells [III.A.2.n] [IV.A.3.c.(1)].


In some Western thinking, we believe that in group involvement the individuality of a person is submerged for the betterment of the group. With the Canela, the individuality of a person, even though he must be involved in groups most of the time, is seriously considered and respected. This may be because of their emphasis on feeling and caring for others. If any particular individual in the group is being hurt too much by group activities, the members of the group stop their activity. (The only film edited on the Canela in the 1970s is on this topic [Ap.3.h].) Nobody wants to hurt anybody else too much as long as that person is behaving within acceptable Canela cultural norms.


Extreme individual tendencies, though joked about at first, are largely ignored by others if a person persists in carrying them out. Thus, over-aggressiveness in war-like leaders is largely accepted and ignored as long as certain traditional limits are not crossed, such as ingroup fighting [III.A.2.k]. It is generally believed that people are born with certain traits and that the parents or the group can do very little to change them [III.A.4.a]. Consequently, though the Canela force younger people to live in groups, they nevertheless do not interfere with the individual who persists over a considerable period of time in her or his own divergences from group behavior such as the rare manifestations of homosexuality [III.A.2.j.(5).(a)].

One of the worst forms of behavior is to say an offensive word directly to a physically disabled person about her or his handicap. (In the early 1970s, there were several disabled people: a harelip, a woman with an atrophied right arm, and a dwarf [In.4.e]). Fundamentally, this would be an action against the individuality of the person, and would be hurting or damaging this person because of a trait for which she or he is not responsible [III.A.4.a].


The Clowns epitomize themselves in the Fish festival where they assume control of the ceremonial events from the Pró-khãmmã elders [III.D.2.c.(1)]. As one of the daily events of the Fish festival, the Clowns sing the same set of very dignified songs that the Pepkahàk troop sings in the Pepkahàk festival (Plate 45a). The Clowns sing them in the same dignified manner as the Pepkahàk, suggesting that they too are as noble as the wet-headed Pepkahàk and capable of ruling like the Pró-khãmmã, whom the Pepkahàk ceremonially represent. However, between each song the Clowns perform in a "shameless" (më pahàm naare: they shame not) manner assuming extremely deviant behavior. Individual Clowns screech curses or obscenities in a high voice (one at a time) for all to hear [IV.A.3.c.(4).(b)]. Thus they remind the listening tribe that the Clowns, after all, are not the "shame"-restrained Pepkahàk or Pró-khãmmã [III.A.3.c.(3)], but rather the independent, individually acting, and even defiant Clowns (Plate 46d). Nevertheless, the Clowns respect the establishment-oriented Pepkahàk, spending most of the time singing their songs in a traditional manner but expressing their individuality only within (i.e., between songs) these confines.

In this way the Clowns can be seen as the "somewhat bad ones" or the "little bit bad ones" [III.B.1.g], not the "very bad ones" [V.A.5.c]. They are not "beyond the pale" of the governing Pró-khãmmã. Correspondingly, individual behavior is not anathema to the Canela; it is just not preferred behavior, as group behavior is. Nevertheless, individual behavior, and taking initiatives as the Clowns do, are acceptable forms of behavior and even enshrined in a festival for all to see as long as such "deviant" behavior is held within traditional limits (Plate 46c). Because it occurs only within a festival, the Clowns are "allowed" to win completely—"eating" all the Fish members—not having to compromise or conform entirely.

Research assistants had no direct equivalents for the concepts of "individuality" and "solidarity." However, taking one's own initiative paired with always needing to receive an order [III.B.1.k] before carrying out an objective is similar to individuality within solidarity. Research assistants usually found this "polarity" "complementary" [V.A.2] rather than "oppositional" (Glossary).


The Canela conspicuously admire endurance (awkanã: endure!) and dislike its opposite "weakness" (i?-pekpek to mõ: she/he-weakness with goes), especially in men.


The ideal male is supposed to be able to hunt in the noonday sun and not mind it. He should endure going without food and drink for long periods of time. He must be long-suffering with his wife and family and put up with considerable trouble and even abuse from women. It often seems to the observer that the husband is the passive element in the family because the wife is so active and can protest so effectively in public interfamilial hearings [III.D.3.c.(1)]. The demure attitude for sexually active women depicted by Murphy and Murphy (1974:132) for the Mundurucú is not characteristic of Canela women who look firmly back at men, challenging them any time depending on their relationship roles. The man is supposed to be impassive and bear all marital problems patiently. The domestic world is not the man's sphere of activity, except economically [II.D.3.h], unless he is playing the role of the advising-uncle.


According to female and male research assistants of my council group, women are expected to have less endurance and to be more frail, less patient, and more expressive of their feelings than men. Men discuss this difference with each other, saying that a woman is soft and does not know how to endure physical difficulties. Men expect women to give vent to their anger, to show pain and physical suffering outwardly, and to lose patience long before men in female-male relationships. Nevertheless, to me, Canela women are conspicuous in the effective way they manage domestic chores and care for their children, demonstrating considerable endurance and strength [III.D.3.b].


In keeping with this orientation toward endurance, the Canela admire permanence and items that last "forever." The expression nõ?nü-?ti mã (forever in) describes objects such as stones that cannot be shattered, glass that resists fire, and matrilines that survive time. Even more than matrilines, a corporate festival matrilineal unit (haakhat) [III.C.8.a] can be expected to survive as long as the tribe lives in a circular village. Canela research assistants described and talked about this kind of endurance through time with great admiration and pleasure. While their villages and farm plots move, and even matrilines come to an end with a lack of female offspring, the corporate festival units can last nõ?nü-?ti mã because they can be transferred from one matriline to another by the Pró-khãmmã.

The Canela admiration for endurance, and therefore for survival, is manifested in what may be their only form of prayer [II.F.1.c.(3)] [IV.D.5]: singing about enduring objects such as stones, the armadillo that lives through the heat of the midday sun, and the locust tree (ku?tàà) that survives cerrado fires.


The Canela do not comment or say very much about things that are beautiful (mpey-ti) versus those that are very ugly (?khêán-re). Questioning old research assistants about this elicits very little information. Nevertheless, it is easy to find out from listening to discussions whether people, artifacts, human behavior, or the transformation of the countryside is beautiful or ugly. The Canela do highly value beauty.


The Canela enjoy a straight line. They like tall, straight, reasonably thin human beings. Hair should be black and straight, and when cut and combed it should fall in straight lines.

They love the cerrado because it is relatively open (kaarã) and because the various trees of the cerrado are beautiful to them. Moreover, they enjoy seeing great distances from hillside to hilltop. They longed for these vistas while exiled at Sardinha (Map 3) for 5 years in the dry forests [II.B.2.g] (Figures 5, 6).

When roads are made through the cerrado the Canela take care that they are as straight as possible, deviating only because of features of the landscape that require turns. For instance, when the Pepkahàk troops construct connecting roads from their shelter through the cerrado to the village and to their swimming place, they make these pathways absolutely straight, with well?defined lines on either edge (Map 5) (Plate 5a).

When the Kô?khre log is erected in front of the Wè?tè house at the closing of the Wè?tè summer season [IV.A.3.e.(3)],every care is taken to see that this section of tree trunk is standing absolutely perpendicular and exactly between the midpoint of the plaza and the center of the door of the Wè?tè house (Figure 45).

When the women sing-dance in a line in the plaza, their torsos must be completely vertical with only their knees bending as their bodies go up and down in time with the music (Plate 33).


The Canela have a particular point of view of what is beautiful, and the range away from this traditional standard is not very great. On the other hand, if a person has a darker skin and the hair is somewhat kinky, and she or he is short in stature, and somewhat rounded beyond cultural expectations (Moon's descendants) [IV.C.1.b.(1)], this is not a disaster for the individual. She or he is merely less within the ideal as far as the beholder is concerned. There is no question of prejudice or ostracism. The concern or compassion [III.B.1.b] for the individual who is not ideal is foremost in their feelings.


Usually, the Canela attempt to hide any bodily defects if they can. Of course the dwarf could not conceal her size, the boy his harelip, or the old widow her atrophied right arm and hand. But the young girl whose arm was temporarily paralyzed in 1964 [III.B.1.b.(1)] attempted to carry it in such a manner that she concealed her semi-paralysis. Similarly, a man who in 1966 lost his leg because of a snake bite and subsequent gangrene, moved everywhere on crutches but tried not to call attention to his condition.


It is interesting that while "begging" (Glossary) was practiced by more than half the tribe in the late 1950s in order to have enough to eat [II.C.3.g], it was done with dignity and poise. With the exception of a few individuals, there was little self-consciousness about begging.

In the late 1950s the dignity and poise with which the old Canela women and men used to carry themselves was obvious and striking. This balance and self?respect has been slowly eroded in the younger generations.


The Canela consider acceptable most behavior of other Canela, ranging between "a little good and a little bad." The word i?-khên-re (it bad little) in contrast to i?-kêán-re ("á" being the superlative of very bad—beyond the pale) expresses this difference well. Most things are said to be i?-khên-re, meaning that they are only somewhat bad-maybe "naughty" if pertaining to children, or "nasty" if about adults (the Clowns). Words equivalent to "wicked" or "evil" [V.A.5.d.(4)] would not be used directly or indirectly except in the most extreme cases, which fall beyond this range, such as when a shaman becomes a sorcerer, killing other people with his spells [IV.D. l.d.(2)].


When a woman or a man acts in an untraditional manner, it is said that the person thinks she or he is greater than everybody else (më-hirô-á-pe kati: them-over-[superlative]-ly great: very over everybody in pretensions). Next to being stingy [III.B.1.a] (the greatest evil) trying to give the impression of being more important or better than other people is considered the second greatest evil. Canela research assistants did not volunteer information about this particular trait so that it had to be learned through observation and overhearing conversations.

For many years one Canela male [In.4.e] who spent most of his time away from the tribe used to come back to the village periodically, drinking too much alcohol and dressed in the gaudy style of the large city. He used to give away beads, cloth, knives, and other presents, and when he had finished his elaborate philanthropy, he expected to live off everybody else for at least six or eight months doing no work, which he considered was beneath his dignity. Besides being enormously fat, his behavior was aggressive and arrogant, as well as patronizing to most other Canela. He was generally disliked; nevertheless, the forbearance and tolerance of the people toward him was amazing to me, although he greatly departed from Canela standards of proper social behavior.


A woman who had a congenitally atrophied arm, and who was very ill-natured (according to the research assistants), might have been expected to provoke negative judgments. However, references to her were always mild and were never made to her face. Any negative words [III.A.3.a.(2).(d)] said overtly to her would have been a severe violation of Canela ethics.


Along these lines, the Clown society (Glossary), the Më?khên, might be called the "malicious ones." (The term i-khêán-re is not used for them.) The Clown society in the Fish festival, the members of which are supposed to represent the somewhat lesser (the more relaxed and individualistic) aspects of life, deliberately behave in an untraditional, inadequate fashion, transgressing basic rules of behavior. By doing what is wrong they emphasize what is right. For instance, the hut the Clown society builds in the center of the plaza is necessarily off line, with none of its angles being 90 degrees (Nimuendaju, 1946:226) (Plate 46c). Moreover, when they finish properly singing the great Pepkahàk songs as a group, they disband and walk away separately to their houses as individuals, singing individually and consequently out of rhythm and off tune with each other. In this drama, the Clowns epitomize the Canela "little bit bad" orientation-wrong but not evil. (See the "little bad" Mask [i?-hô-?kên-re: its straw little-bad dim.] in Plate 48d.)

The Clown girl associates wear long pigtails of hair that are quite different from the traditional style, although not grotesque (Plate 46c). They break most of the rules regarding clothing, adornments, postures, child care in performing their public dramas, including jokingly committing incest with one of their supposed brothers behind the small hut erected for them in the northern part of the plaza. What the male Clowns do to their girl associates in this act is carried out in a spirit of fun and joking which mitigates and neutralizes the breaking of the traditions and the maltreatment of the girls (Nimuendaju, 1946:228).

The Clowns, by definition, are not hàmren (Glossary). In contrast, the Pepkahàk troop (Plate 44c) is made up of both hàmren and non-hàmren individuals, and the Visiting Chief society members (Tàmhàk) [III.C.7.a] are all hàmren by definition, so they represent extreme formality (Plate 44d). These Visiting Chief society members, while sensitive as hàmren individuals, portray roles as a group that imply that they are better and greater than everybody else. This may be why they appear in only one act (of 1/2 hour) in the festival; their aloofness and arrogance could not be tolerated longer. In contrast, the Pepkahàk, (Glossary) have some hàmren and some non-hàmren individuals, which represent a better balance; therefore, the Pepkahàk can be the principal group of the whole festival. The dramatic epitome of the "somewhat good" people of the Pepkahàk festival balance the "somewhat bad" Clowns, which is the principal group of the Fish festival, in characteristic Canela off-setting fashion.


When the Pepkahàk (Glossary) race against their opposites the Ducks (Glossary) [IV.A.3.c.(3).(b)], the Clowns help the Pepkahàk win their race as well as the Falcons and Visiting Chiefs. Thus, the somewhat good Pepkahàk and the somewhat bad Clowns are on the same side in a complementary pairing as they move competitively against their common opponent, the "somewhat bad" Ducks, in an oppositional pairing. The structure in this ceremonial situation depicts complementary and oppositional pairings (Glossary) operating at the same time—the Pepkahàk and Clowns being paired in complementarity but not being extreme and the Visiting Chiefs and Ducks being paired in opposition and being extreme.

[III.B.1.h] Avoidance Versus Conflict

The Canela avoid direct face-to-face confrontations, as do the backlanders and Brazilians in general. When there is an unavoidable conflict or confrontation, one person simply removes her- or himself from the scene. These are not orientations that research assistants can describe to the visiting anthropologist; they are related behaviors that become obvious to an outside observer who does not share this orientation. These patterns are repeated over and over again in festivals.

I did not inquire into opposition or complementarity for these contrasting orientations, nor do I know even if they could be "paired" [V.A.1,2] (Glossary) by research assistants. I have included them here because, to me, they significantly characterize the Canela behavior patterns.


While this kind of behavior is characteristic of personal life between individuals, it is also a pattern for public situations. For instance, according to the 1984 communication from the Canela, Chief Kaarà?khre resigned as the chief of the tribe because he was denounced by one Pró-khãmmã in the plaza [Ep.2]. The normal pattern is for chiefs to die in office [III.D.1.h]. Apparently, he did not present any resistance, but simply removed himself as the political tribal chief.

Another example of confrontational avoidance in the public sector was that of the older Kaapêltùk (Figure 50) and Chief Kaarà?khre (Figure 18). These two chiefs were very hostile to each other as leaders of their two villages of Baixão Prêto and Ponto [III.D.1.g.(1).(b)]. Nevertheless, they were very civil whenever they were by chance in each other's presence. It was conspicuous, however, that whenever Chief Kaarà?khre was leading the council meetings in Escalvado (long after the schism between the two villages had been healed), the older Kaapêltùk was rarely present in the plaza. When Chief Kaarà?khre was away, the older Kaapêltùk often ran the council meetings. Because Kaapêltùk had been the deputy commandant of the graduating age-set of the Pepyê of the 1930s (Nimuendajú, 1946:182) (the Pró-khãmmã of the 1960s and 1970s), it was his role to preside over the council meetings. He was, however, the loser of the political succession and competition from 1951 to 1968 and, therefore, accepted his lesser position with equanimity and non-confrontation.


Non-confrontation in the ceremonial arena may be seen in the high honor Apikrawkraw-re act [IV.A.3.c.(3).(e)] of the Pepkahàk festival. During the act, the Pepkahàk troops and the Falcon society troops (Plate 44b) almost confront each other. However, a line of ceremonially high-honor individuals is placed between the two sets of symbolically fighting warriors. Even though the warriors are no more than 10 meters apart they may not breach the file of honor to confront each other [V.A.5.c.(2)].


In the kinship system, the roles that are most likely to bring individuals into conflict are the opposite-sex adjacent-generation roles: the mother-in-law/son-in-law and the father-in-law/daughter-in-law [III.E.3.a.(5)] [V.A.5.c]. These roles traditionally, and even today, call for complete avoidance. Nothing may be said between these two sets of people. They are not even supposed to look each other in the face. This is also true of opposite-sex, high-priority Formal Friendship relationships [III.E.5].

One relationship I knew of between Informal Friends [III.E.6], who are traditionally quite direct with each other, evolved into hostility and therefore indirection. They continued to call each other by the Informal Friendship terms of address (ikhwè?nõ), and superficially behaved according to this relationship's roles, though joking between them ceased. They avoided each other whenever possible, but if they could not, they completely controlled their inner hostility when in each other's presence.


The Canela avoid potential confrontations by a number of subterfuges such as lying, reporting false information, not appearing on the scene, and showing expressionless faces. In this way the Canela are often able to live through potentially dangerous situations without aggravating the conditions between the parties. They are experts in carrying out indirect behavior [V.A.5.c.(2)].


When research assistants are asked to describe the ideal Canela individual, it becomes apparent that they are talking about young people in their late teens or early twenties, with the women being somewhat younger than the men. These are the ideal human beings according to Canela norms of beauty.

When a young girl dies before she has become pregnant, the mourning throughout the village is intense. The emotion is also strong when a handsome young man with great abilities in running and singing dies. While older people manage the festivals and politically govern the tribe, youth is the epitome of the ideal Canela.

Research assistants were very cognizant and found it easy to say that the old (më ?wey: those old) and the young (më ntúwa: those young) are in paired opposition with each other.


An example of this extreme idealization of the young occurred in 1970. In the middle of the terminal phase of the Pepkahàk festival, the beautiful daughter of the younger Tààmi, Kuwrè (Plate 8d), died quite suddenly; she was sick only from 6 in the evening to about 3 in the morning. Both the suddenness of her death and her being in the prime of life caused great shock to the assembled tribe. The festival was postponed for her funeral and burial (Plates 30, 31e), and one day for mourning, a procedure that I have never known to happen during any other mourning period.


It is notable that in most of the festival performances the festival is either principally for young people or the main roles are performed by these young ones. The Khêêtúwayê and Pepyê festivals, especially, serve to enculturate young people. In the Pepkahàk festival, most of the Visiting Chiefs (Plate 44d) are young (10 to 30 years old), because it is believed that their shamanic powers are stronger then. The Pepkahàk troops themselves and the Clown society members are likely to be older, though there could be young men in their late teens involved in each membership. The girl associates of all of these groups are also quite young (6 to 14 years old).

When a man entitled to a certain festival role lets a named-nephew take his place, he has the youth perform beside him (Plate 52a) for one or two years, and after this apprenticeship, the young person (usually between 15–25 years old) takes charge of and performs the role himself.


The Canela, in many respects, live for the young people. The early morning and evening sing-dances [II.F.1.b.(2).(a)] (Plate 33) are definitely for the younger people, not really for anyone beyond the age of 25 or 30. In contrast, the late afternoon sing-dance takes place for the benefit of older as well as younger people. Marriages and family life, through the hearth group [III.E.2.e.(1),F.7], are adapted more for the requirements of children and adolescents than for the parents.


As is probably true in most tribal societies, consanguineal relationships are held to be more important than affinal ones. This is conspicuously the case among the Canela. Canela "blood" concepts and genealogical linkages strongly emphasize this preeminence [III.F.11,12]. The marriage link is usually less important to the individuals concerned than certain consanguineal ones, though marriages are unbreakable while children are growing up [III.F.9]. Family members focus not so much on the affinal tie between the mother and father as on the consanguineal relationships that grow strong between the parents and children. An important affinal bond, however, is the work-related tie between a man and his son-in-law [II.D.3.f]. The son-in-law is caught by his feelings for his wife and children into a relationship with his father-in-law, which leads him into working for his in-laws with great forbearance. This relationship is ceremonially recognized in the final phase of the Pepyê festival, when the women hold their sons-in-law by a cord symbolizing this tie (Plate 40a,c,e). In contrast, kin are connected to each other by unbreakable bonds of blood (Figures 38, 39, 41, 44). The Canela feel these bonds very strongly-the Apanyekra to a considerably lesser extent.

While a man does not feel fully at home and at ease in his wife's house until his children are partly grown, the tension between affines seems less among the Canela than among the Shavante, as the latter are depicted by Maybury-Lewis (1965:263): "Indeed Shavante appeared to regard a state of permanent tension between the dominant faction and the rest, between kin and affine as the natural order of things." The Canela regard internal peace and harmony as the natural order of things: the way things are supposed to be [III.D.3.e.(1)]. Affinal relationships are structured in a way that produces some tension, especially between brothers-in-law. Nevertheless, this tension is modified and reduced by the general belief and acceptance that harmony is the most important state for all personal relationships. The Apanyekra appear to be less compulsively concerned about quickly resolving even the minimal threat to interfamilial peace.


I did not appreciate before 1979 the extent to which the Canela prefer to carry out almost all activities only when orders are issued for their execution [III.D.1.a.(2)]. This is one of the most important and far-reaching Canela orientations, and it is notable that it is so highly valued by them. However, research assistants cannot explain or define this behavioral pattern. While working with "oppositions" in 1979, the significance between halkhwa-?khôt (order-following) and amyi-á-?khôt (self-superlative-following) became apparent and understandable to me. While all agreed these two concepts were "paired," some said they were opposites and others said they were paired in a complementary manner [V.A.2], depending on whether "self-following" was completely wrong or whether it was an acceptable form of behavior under certain conditions.


In the winter of 1979, for example, when cattle entered some of the farm plots of the Canela, even though they were strongly fenced on all sides, the strong delegado of the Indian service in Barra do Corda declared that the Canela had permission to kill such invading cattle. However, the Canela were very reluctant to do so, because the killing of cattle in 1963 resulted in the death of five Canela [II.B.2.f.(3)].

In 1979, it was reported that a young man in his 20s [In.4.e] had given orders to another man of the same age to kill a head of cattle that had repeatedly invaded the latter's farm plot and had destroyed much of his crops. Thus, the youth who had suffered the loss shot the cattle with impunity. I was told that as long as one Canela had given the order, even though he was of low political rank, the other Canela was blameless in carrying out the command. Evidently the responsibility for an act may be shifted from the doer to the authorizing person.


Further investigation revealed, however, that it is only the chief of the tribe, and in his absence certain of the sub-chiefs, who regularly give orders to large segments of the tribe. The Pró-khãmmã issue orders at certain times, especially in festival situations [III.D.2.c.(1)]. Similarly, heads of households, whether women or men, and in certain cases advising-uncles, issue orders about family matters and expect to have them obeyed. The shaman, on the other hand, proceeds on his own authority, receiving orders from no one. In most situations, unless the individual is clearly at the apex of authority in her or his particular sphere, she or he waits for the person in authority to give the orders. There is little competitive challenging for leadership.

Even today, orders have to be given for almost anything significant to occur in Canela life. The sing-dance leader must be selected and ordered to perform for every session. The Pró-khãmmã have to order that certain festivals be put on, or the special performers have to receive requests from other traditionally correct parties that a particular festival begin. There are dozens of roles in each festival, and if the Pró-khãmmã do not give orders and name the person who should carry out each role, most of these activities are not carried out.

For instance, the same man, or a member of his close kin, cuts the logs for the annual Pàlrà race every year [IV.A.5.e.(2)] because he inherited the right matrilineally [III.C.8.a] to do so. But if the council of elders does not give him the order to cut the logs, he simply will not do it. Similarly, even though the Corn Harvest festival ritual [IV.A.5.d] is held in the possession of a certain family line matrilineally, this family will not make the traditional preparations for the festival unless the Pró-khãmmã have called one of them to the plaza and asked her or him to do so.

As a result of this tradition for following orders, the Canela reacted favorably to the comparatively severe and extensive commands issued for their welfare by the strong Indian service delegado in Barra do Corda in 1978–1979. They became more cooperative with the Indian service agent, worked harder, and had fewer disputes than they had before his arrival.


I had long discussions with my research assistants about the meaning and application of the expression amyiá-?khôt (self-following: giving oneself orders to carry out certain activities). While research assistants agreed there were cases in which an individual had to give her- or himself orders and then carry out these orders, they said it was nevertheless preferable for the ordinary person to be acting on the orders of somebody else. Some thought that acting amyiá-?khot was evil.

These days, many Canela speak about orders from God as rules that everybody should obey rather than their own orders to themselves. In earlier times, however, as I pointed out to my research assistants in 1979, there could not have been any orders from God, because for the Canela—and research assistants totally agreed—God did not exist in the Christian (Católico) backland sense, only ghosts and culture heroes [IV.C.1.c.(17).(b)]. The implication is that the traditional Canela orientation was to follow orders for most undertakings.


While Canela research assistants were not easily verbal about the concepts of we and they, or inner and outer, these polarities, complementary or oppositional, are embedded in their language, festival groupings, and in almost everything they do.


In the language this pattern is manifested in their use of a dual pronoun form with both the singular and plural persons. For instance, a Canela can say i-khra, meaning "my child." However, if he is speaking about the child born to himself and his wife, and addressing just his wife, he can use the expression pa-?khra, meaning "our" (just the two us) "child." He can also say, if he is speaking to his three co-contributing-fathers [III.E.9] and his wife, mëpa-?khra, meaning "our" (all five of us) "child." Then in still another situation, when he is speaking for himself, his wife, and his three co-contributing-fathers but addressing some other person or group, he can say mei-khra, meaning "our child" (but implying "not your child"). We do not make these distinctions in English, presumably, because we do not need to the same extent to make these dualistic distinctions—to distinguish between just two groups.

The dual forms are useful when the speaker needs to indicate grammatically whether she or he is included or excluded from the group being addressed. It is easier to picture what is occurring by thinking of one age-set moiety sitting on the eastern side of the plaza, and the other age-set moiety sitting on the western side (Figure 24). Then, if the leader of the age-set moiety on the eastern side of the plaza stands up and speaks to his own people sitting there directly below him, using the dual plural pronoun më pa: (plural us [two persons]: we [more than two]), he is talking to all of the people in his group on the eastern side of the plaza including himself. Whereas if he turns around 180 degrees and talks to the people on the other (western) side of the plaza and uses the first person plural pronoun më-i: (plural-I: we) instead of mëpa, he is addressing the western age-set, giving them information about the eastern age-set, including himself.


From the above model, it should be easy to see why the language is so well adapted to speaking within the group (excluding outsiders), and to speaking of the group to outsiders (including the outsiders). These distinctions are important for communication within and between the several moieties [III.C.3,4,5,6]. (For other dichotomies or polarities, see [V.A].)

In their thinking and behavior, the Canela almost always divide their activities, perceptions, preferences, hopes, tastes, etc., into categories such as we and they, mine and yours; but they also attempt quite characteristically to bridge these dichotomies so that when discrepancies are pointed out, they prefer to hear about the similarities rather than the differences.

I interpret J. Melatti's (1979a:46–50) "principle of opposition between oppositions" in this context:

A notable feature of Krahó dualism is that every opposition which states a difference between two elements is counterbalanced by another one which insists on the identity of those same elements. The two oppositions thus cancel each other out.... At the same time it should be noted that, whenever a statement of difference between two elements is opposed by another affirmation of their identity, the first proposition carries more weight, for it represents the rule while the second indicates certain institutionalized exceptions to the rule. In other words, these exceptions act in such a way as to produce a double result: at the same time that they emphasize the rule, they also deprive it of its absolute character.

J. Melatti then describes the opposition between men and women, pointing out that while men are opposed to women (difference), two women are nevertheless included (similarity) in most men's ceremonial groups. Thus, "during [Krahó] initiation rites, two girls always accompany the boys who are taking part in the ritual" (J. Melatti, 1979a:49). Similarly, during the Canela Khêêtúwayê and Pepyê initiation festivals [IV.A.3.c.(1),(2)], two girl associates [II.D.2.e] always accompany the male novices. In daily life, the Canela like to bridge differences (women against men), making them similarities (women with men) [V.A.5.c.(1)], depriving the major rule (women against men) of its absolute nature. J. Melatti's principle is important and can be applied almost everywhere among the Canela.

[III.B.2] Observations

It is necessary to analyze Canela psychological characteristics in terms of polarities because research assistants see things this way and use such pairings (complementary or oppositional) (Glossary) in their explanations. Research assistants recalled more information if the researcher used this context with them in group meetings [Pr.2]. Moreover, they felt better if oppositions in real life were resolved [III.D.3.e.(1)]. Whenever I returned to the tribe after an absence, they would ask me in the evening council meeting about the world scene and sometimes about the Russians and the Americans. The old people, in their conversation with me, seemed to be visibly pleased when I reported that peace continued between the two superpowers. Many Canela individuals obtain personal gratification from resolving existing differences.


This chapter adds to data presented in "The Eastern Timbira" in which Nimuendajú (1946:77) outlined the various social groups among the Canela in a relatively complete manner, and on my articles of 1977 and 1979.

The objective here is to describe in depth each traditional social group (whether political or ceremonial): its membership, its methods and principles of recruitment, its operation in the society, and its changes or trends that are either incipient or have been in progress for several decades. (For the roles of social groups in festivals, see [IV.A].)

[III.C.1] Defining the Units

The social groups of highest visibility are the Upper and Lower age-set moieties [III.C.3] referred to as the eastern and western age-class moieties by Nimuendajú (1946:91). Nimuendajú (1946:79–82) also states that the Canela had exogamous moieties of the same names as his "age class" moieties, but the existing evidence suggests that this was not the case (W. Crocker, 1979:237–240; Lave, 1971:341–344; J. Melatti, 1967:64). Matriliny is limited to certain festival rituals [III.C.8.a] and is not related to marriage. (For a general discussion in which I concur on why Nimuendajú made the serious ethnographic errors he did in several Gê tribes, see J. Melatti,1985:19–20.)

These age-set moieties exist both in ceremonial and daily life [II.E.1.b], as do Nimuendajú's rainy season moieties. However, I call his "rainy-season" moieties the "Red and Black Regeneration" season moieties to keep them consistent with the events of the season during which they take place and consistent with the meaning of their term of reference in the Canela language, Më-ipimràk (they renewed-and-renewed) [IV.A.4] (Figure 25).

The rest of the numerous social groups operate only in their festival (i.e., ceremonial) contexts, although the behavior required of the members of some groups is often carried over into daily living, as with the Clowns and the Visiting Chiefs (Tàmhàk: Nimuendajú's king vultures or his courtesy chiefs) [IV.A.3.c.(3).(e)]. "Visiting Chiefs" is more descriptive of their prepacification intertribal role [IV.C.1.d.(1).(a)] than "courtesy chiefs."

The third named set of moieties, the Upper and Lower plaza moieties, are each divided into three plaza groups, their names depending on the festival [V.A.5.a.(1)] (Figure 17). There are also two unnamed sets of moieties which can be found as opposing "men's societies" (Nimuendajú's term, 1946:95) in three festivals. In the Pepkahàk festival the Falcons (east) oppose the Ducks (west), and in the Closing Wè?tè and the Masks' festivals the Jaguars and Masks (east) oppose the Agouti (west) (Glossary).

Nimuendajú (1946:95) includes the Clowns in his category of men's societies; I prefer to see them as a special society because their rules of recruitment (nonnaming) and their traditional behavior are too different and specific for inclusion. Moreover, the Clowns are paired with the Visiting Chiefs. These two societies represent high (traditional: Visiting Chiefs) (Plate 44d) and low (idiosyncratic: Clowns) (Plate 46d) ceremonial behavior which transfers into daily life to some extent. On the other hand, the plaza groups and men's societies (both name-set transmitted; Glossary) have little significance for daily behavior.

The remaining social groups, though still numerous, are more difficult to define. Some exist only momentarily in festivals during one 30-second act and others maintain their roles for several days. Moreover, it is not always clear whether such a "group" consists of one person or many. Groups of this sort can be found only in what I call the "river-oriented" ceremonies: the Fish, Closing Wè?tè, Mask, Sweet Potato, Corn Harvest, and Pàlrà (Table 4)—and will be designated as the "ritual" (Glossary) societies or groups [III.C.8].

[III.C.2] Recruitment Principles

One way of organizing and understanding social groupings, or societies within the overall sociocultural system, is according to how members are recruited. The principles of female recruitment are name-set affiliation [III.E.4], request by the societal membership, appointment by the Pró?khãmmã, matriline succession, and matrilineality. The most common principles of male recruitment are relative age and name-set affiliation. Four less common methods used are matrilineality, patriline succession, personal choice combined with group selection, and appointment by the Pró?khãmmã.

[III.C.3] Age-Set Moieties

The two most outstanding and noticeable Canela and Apanyekra groups are the Upper age-set moiety (Khèykatêyê) (Glossary) and the Lower age-set moiety (Harã?katêyê) (Glossary). Permanent members are males who are recruited on the basis of relative age and are grouped into age-sets (Glossary) [III.D.2.b.(1)] (Figure 24, Plate 40b,f ). When the tribe is functioning according to the age-set moiety system, its members can be said to be performing më hakhrã khãm (they groups in). (The morpheme -khrã means head, or something massed into a round shape, and the affix ha- is a generalizer.) Nimuendajú (1946:79) called this major tribal division the "eastern" and "western" exogamous or age-class moieties, but the words khèy and harã mean "upper" and "lower," respectively, though symbolically they can be associated with the east and the west. When I first arrived in 1957, I became aware immediately of the partido de cima and the partido de baixo (upper party and lower party) which were the obvious political and racing entities. I soon realized that these parties were Nimuendajú's (1946:90–95) eastern and western age-class moieties.

After identifying these Upper and Lower age-set moieties, I noticed that their names, if descriptive, seemed reversed. In old Ponto the geographic eastern rim of the circular village is at a lower elevation than the western rim, because the terrain slopes from west to east. However, in the village of Baixão Prêto, the eastern edge is located at the highest elevation; in Sardinha the western edge is the highest (Figure 5); and in Escalvado today, the western edge (Plate 6c) is higher than the eastern rim, but neither rim is the highest part of the village. The most elevated part of the rim lies to the south. Surely, if "Upper" and "Lower" once described relative elevations, they have since become only proper names without descriptive significance [In.4.g]. (See Ritter,1980, for a general discussion on age-sets, using the Canela in particular, and Vidal, 1977b, for an analysis of the changing role of age-grades among the Xikrin-Kayapó.)


New age-sets are formed about every 10 years. Boys who are born since the last age-set graduation in a Pepyê festival are interned (Plate 41a) during the next Khêêtúwayê festival (Nimuendajú, 1946:90–92). This internment is followed two or three years later by a similar "capturing" of the same boys (më-hapèn: them-captured;(Plate 42a-e) for a Pepyê festival, or for another Khêêtúwayê and then a Pepyê festival [IV.A.3.c.(1),(2)]. The process is characterized by a succession of ceremonial imprisonments (prisão, as the Canela say in Portuguese), so that the age-set is finally graduated into adulthood after having undergone four or five festival internments over a period of about 10 years. Adolescents and young men average in age from about 7 to 17 at mid- internment. Because they belong to a cohesive group, they become personally very close and know each other well. In earlier times, ages at graduation ranged from approximately 15 to 25, whereas today they range from approximately 12 to 22.

Apparent age at the time of the first Khêêtúwayê internment is clearly the principal factor in determining membership in one age-set or the next, though some manipulating occurs of age dividing lines, usually for political reasons. A young boy with potential abilities for leadership (and the right connections politically and among the Pró-khãmmã) is sometimes withdrawn from one age-set where he is one of the youngest and re-caught to be one of the oldest in the following age-set. Around 1913 this happened to Ropkhà (who was 85 in 1984 but dead by 1987; Plates 17b, 71e). He was first captured as a young member of the Lower age-set of the older Mĩĩkhrô [I.G.3], and later recaptured as one of the oldest members and the file leader of the following Upper moiety age-set (Figure 24). There are only a few cases of such deliberate overlap, but because the Canela do not know their exact ages, selections are made by apparent maturity rather than by chronological age.

An individual rarely changes his age-set membership, though this can occur as a result of hostilities and consequent personal mistreatment. Research assistants gave only two earlier examples, and no changes occurred during my time. However, one general exception exists: a sing-dance master [II.F.1.a] is chosen from his age-set to be the singing ceremonial chief (më-hõõpa?hi) (Glossary) of the following age-set [II.D.3.i.(2).(b)]. He is so honored by his selection that he usually changes age-sets to be with those who chose him. On the other hand, if a youth should carry out his Pepyê internment responsibilities very poorly, he could be removed from the age-set by being rolled in the ashes of a plaza fire site, making him ineligible for any age-set membership at all. Such a removal is said to have occurred once (mythically), changing the boy to a Pró-khãmmã (ashes-in) without membership in the Pró-khãmmã age-set (Glossary) [III.D.2.b]. He was then left out of age-set activities.

Fortunately for the Canela, the racing abilities and numbers of both moieties have been approximately even during the past few decades, a balance that helps preserve the vitality of the system. Among the Apanyekra this has not been the case, so that one side has won repeatedly, reducing the interest in the age-set system and, consequently, its viability.


Although size and racing ability between the two moieties has been relatively equal in the past, access to leadership roles has not been equal. The Lower age-set moiety members enjoy a permanent ascendancy over the Upper moiety in that they, and only they, assume the leadership as an age-set of the council of elders. This council consists of both moieties and meets in the center of the plaza every morning and evening [II.E.5.b,8] [III.D.2.a]. The expression, "Pró-khãmmã" (ash-in: in the ashes) refers to the Lower age-set that is in charge for 20 years and which will cede the leadership and control of the council of elders only to the next Lower age-set 20 years younger [III.D.2.c] (Figures 19, 24). The relationship between the age-set moieties is not symmetrical (cf. Nimuendajú, 1946:90–92).


Earlier, at the end of each initiation festival, each age-set, even the one being processed, received a different name from a pool of traditional names (Nimuendajú, 1946:91). No knowledge exists these days about traditional cycling of such names among successive age-sets, and the names themselves are all but forgotten. In 1958, the older Kaapêltùk gave me the following names: Pró-khãmmã (ashes-in), Kukhoy-khãmmã (monkey-in), Pàn-rã-khãmmã (macaw-flower-in), Ron-khãmmã (tucum-in), and Tsêp-ti-khãmmã (bat-large-in). The younger Kaapêltùk gave me one additional name: Hàk-ti-khãmmã (falcon-large-in). (In my notes, there is some question of validity concerning the last two names.)


The operating of an age-set alone and apart from its age-set moiety is now rare. Research assistants say that in earlier times considerably more independent age-set activities took place. The use of the age-sets as separate entities fell into disuse in the mid-1940s when the two leaders of the emerging age-sets were employed by the Indian service. The older Kaapêltùk (Figure 50) was employed by the Indian service upon Nimuendajú's recommendation in 1938, and Chief Kaarà-khre (Figure 18) was first employed in the mid-1950s. The employment of these two tribal leaders effectively removed both leaders from the daytime leadership of their age-sets. Neither chief could be present to lead their age-sets during the afternoon log races and the subsequent activities in the plaza and boulevard. This lack of understanding of Canela political structure on the part of a series of Indian service agents was a factor in causing a considerable degree of political disintegration.


Some specialists in Northern Gê ethnohistory (including D. Gross, personal communication) have thought that age-sets developed after contact with Brazilian settlers [II.A.3.a.(1)] [II.B.1.a] to build up warrior groups trained to fight together. Several of the words preserved in specific ceremonies suggest that these festival internments did indeed foster the training of young men as warriors. The name of the Pepyê festival itself means "warriors" (Nimuendajú, 1946:212). When "uncles" pull the Pepyê novices one by one from their cells to see if they have matured sufficiently, they ask harrassingly if the nephew is ready to fight the enemy should he appear outside the village [IV.A.3.c.(2).(a)].

Data supplied by the research assistants [Pr.2] while studying myths and war stories suggests that warring or raiding parties were led by unusually able warriors (hààprãl). These warriors, however, were followed by kin, affines, special friends, Formal Friends, Informal Friends, and admirers: a group that cuts across age-set lines [IV.C.1.d.(1).(c)] (W. Crocker, 1978:18). In fact, reference to only one age-set (Pró-khãmmã) appears in all the myths and war stories collected [Ap.3.b]. Perhaps hààprãl war leaders led followers against enemy tribes in the 17th century and earlier, while age-sets developed in the 18th century to fight settlers and other tribes in the new kind of tribal dislocations and intensified warfare caused by the settlers' pioneer movements.


In the 19th and early 20th centuries, the age-sets (and age-set moieties) played an important role in maintaining the agricultural level of the Canela and Apanyekra tribal economies. However, when leadership became relatively less effective [II.B.1.c.(4)], chiefs were not able to coerce their moiety members into the fields to complete the harvesting. It was particularly during this period that the tribes had fallen into difficult economic circumstances, thereby having to supplement their economies by working on backland farms [II.B.3.j.(1)] [II.C.3.g].

With power returning to Chief Kaarà?khre in the late 1970s [II.B.3.e], and through the leadership of certain Indian service agents [II.B.2.i.(4)], it became possible to summon the Canela age-set moieties to carry out regular work on the road to Barra do Corda and on the legal boundaries of the tribal area (Map 3). The younger Kaapêltùk's political ascendancy in 1963 [II.B.2.f.(4)], 1968 [II.B.2.g.(5)], and the mid-1980s [Ep.4.a,b.(2).(e)] was based on successfully leading his moiety into carrying out extensive agricultural projects.

The moieties usually compete to see which will do more work [II.E.5.g], and then they race home with logs in the afternoon [II.E.6.b]. They also compete in preparations for the termination of the great Wè?tè season festivals. The two age-set moieties go off separately to different parts of the reservation. There they hunt for two or three weeks to provide enough meat to support the festival over a period of ten days [IV.A.3.b.(2)]. At other times, instead of individual families going to their fields as usual, the moieties each go hunting separately [III.D.1.c.(2)], especially if the objective is to encircle and kill wild boar.


The age-set moiety system is still vital and active as a daily operating system. It is also the most notable and frequently used organizing principle for arranging social groups in festivals. In the Khêêtúwayê and Pepyê festivals (i.e., the initiation festivals) [IV.A.3.c.(1),(2)], the social groupings are almost always based on the age-set moiety system. In the Pepkahàk festival, the age-set moiety system emerges only at certain points during the festival, for example, when all the men race into the center of the plaza carrying the Awalwrêw-re poles [IV.A.7.c] (Nimuendajú,1946:220). The age-set moiety system does not function in the Masks' festival at all and rarely in the Fish festival (Nimuendajú, 1946:230). However, using age-set moieties instead of plaza group moieties is an alternative way (apparently rarely) of putting on the Fish festival. This alternative did not occur during my span of 22 years with the Canela.

When various elements of the Wè?tè season great festivals are considered in light of what elements appear to be underlying and what traits may be overlaid, the age-set moiety system appears to have been the last added element. The full analysis of this question is intended for inclusion in a later publication, but a brief overview is presented in [III.C.10].


The two girl associates [II.D.2.e] of the two Nkrel-re internment initiation festivals, the Khêêtúwayê and Pepyê, are appointed to their positions by the Pró-khãmmã. One set of girls is assigned for about five years to complement the Nkrel-re initiates during the first half of the initiates' 10-year training period. These girls are usually 6 to 8 years old when they are first chosen and have to be retired when the oldest one becomes pregnant, or older than 13 or 14. Then a new set of girl associates are selected by the Pró-khãmmã for the second half of the training period of the Nkrel-re initiates (novices). These girls are dismissed until the next festival the day before the end of each festival, and the second set is released for the last time the day before the graduation of the initiates as a fully formed age-set.

These initiation festival girl associates (Nkrel-re kuytswè),as well as the Pepkahàk girl associates and the Wè?tè girls [III.E.10, IV.A.3.e], are hàmren in status (wetheaded), but rather than being called hàmren, they are referred to as being pep-khwèy (warrior-women) [III.C.7.a].


Among the Eastern Timbira, the Krĩkatí still have their Upper and Lower age-set moieties, but their age-sets no longer exist. Initiation into each moiety is by "individual choice and arrangement" (Lave, 1971:344). Similarly, the Krahó still have ceremonially active Upper and Lower age-set moieties. While Krahó age-sets do exist, they are "decadent" with respect to use and membership, not every man knowing to which one he belongs. An experienced old man divides the youths into two groups, each of which is an age-set in one of the two opposing moieties. Thus, among the Krahó opposing age-sets are of the same age, unlike the Canela age-sets, which alternate ages about every 10 years (J. Melatti, 1971:348). In contrast, the Canela age-sets and age-set moieties are in full use in festivals and in daily life, and every man knows to which age-set and to which moiety he belongs.


The Canela are very clear about when they are in or are not in a festival or ceremonial state [II.E.1.b] [IV.A.1]. The expression amyi-?khin nã (self-esteeming in) means they are in a festival condition, but there is no specific word for not being in such a state, though they give më ?pek-pek nã (they weak-weak in: they go along feeling weak) as the opposite state.7

The Upper and Lower age-set moieties, and to a lesser extent the Red and Black Regeneration season moieties, operate both in festival situations and out of them, namely, in ceremonial situations and in daily life. In contrast, most other ceremonial groups operate only in their festival contexts.

Individual roles within a festival context may affect individual behavior in a non-festival context. For instance, a Wè?tè girl [III.E.10] behaves in a somewhat superior and reserved manner outside of festival performances as well as during them. However, the influence of festival roles on the behavior of individuals in daily life is largely limited to the Clowns (Glossary) [IV.A.3.c.(4)] of the Fish festival (who carry out practical jokes in daily life) and to the Ceremonial-chiefs-of-the-whole-tribe [IV.A.3.c.(3).(e)] of the Pepkahàk festival. Some other hàmren individuals (Visiting Chiefs), also behave with restraint in their daily lives because of the honor and sense of "shame" [III.A.3.c.(3).(a)] they assume in their life-long, seriously regarded festival roles. These are examples of individuals carrying the "personalities" of their festival roles over into ordinary living. In contrast, festival groups (societies) do not carry over their behavior into non-festival contexts, except for the Upper and Lower age-set moieties. For example, the daily log races [II.F.2.a] (Plates 34, 35) are almost always organized within the age-set moiety divisions. In addition, the commandant in a Khêêtúwayê or Pepyê festival summons his novices (initiates) out into the cerrado for practicing group solidarity and recreational activities between festivals, as well as during them. Moreover, the Pró-khãmmã Lower moiety age-set maintains its ascendancy over the Upper moiety age-sets in the council of elders' plaza meetings twice a day.

The great significance of the age-set moiety system lies in the fact that a considerable part of men's daily living is organized along its principles. Gê specialists who compartmentalize ceremonial and daily activities must realize that for the individual Canela (who knows which state she or he is in) some festival units and many behavioral patterns transcend ceremonial/daily living dichotomy.

[III.C.4] Red and Black Regeneration ("Rainy") Season Moieties

The Red and Black Regeneration (rainy) season moiety system (Mëipimràk) (Glossary) [II.C.4.a] [IV.A.4] (Table 4), in contrast to the age-set moiety system, is all but abandoned by the Canela but less so by the Apanyekra. This de-emphasis occurred so long ago that it is impossible to determine when it happened. Even Nimuendajú saw little of this system, so what he described is mostly a reconstruction from the memories of his old informants (Nimuendajú, 1946:168–170). I also spent much time on several visits trying to reconstruct the various performances of the Regeneration season moiety system. Several old and experienced research assistants helped in this task: the older Kaapêltùk [I.G.2] (Figure 50) and Tel-khwèy (Plate 73a), and the younger Kaapêltùk [I.G.4] (Figure 51), who facilitated communication as usual.

Since the importance of the Regeneration season moiety system is now minimal, it would be surprising if it occupied as much of the year and the same months of the annual cycle as used to be the case. The Regeneration season log racing, which is now the central focus of the system instead of the Regeneration ceremonies, currently takes place from October through the first part of January, if it occurs at all. It may be omitted. If the system was intended to become operative during the rainy season as the name of the moiety system implies (Nimuendajú, 1946:84), the ceremonies should take place in December through March or April, when most of the rainfall occurs [II.C.1.a] (Table 1). My reconstruction of the Regeneration season moiety festival and log racing system is very complex, and some episodes still have to be presented with alternative solutions. Thus, describing them and their analyses will have to wait for a later publication.


When some portion of the Regeneration season festivities is being performed by the Canela, they say they are in the state of më ipimràk nã (they renewing-and-renewing in: being in a state of changing again and again or regeneration in steps). The reference is to the practice of switching back and forth between the ascendancy of ("the time belonging to") the Red or the Black moiety. Moreover, the weather during this period oscillates between predominant dryness and occasional light rain, and the various fruits of the cerrado periodically ripen and fall in succession [II.C.2] (Table 2).

The people of the Red moiety are referred to as being the Kàà-mã-?khra (plaza at Indian-children: people of the plaza) while the people of the Black moiety are the A?tùk-mã-?khra (outside-the-houses at Indian-children: people of just outside the village). The expression a?tùk mã indicates the symbolically important area between the village circle of houses and the cerrado, where trash usually exists if women have not cleaned it for a festival. It begins immediately behind the rear wall of each house, so that the Pekahàk's trail around the village just behind the houses (Figure 24) and the houses in the second circle of houses being formed in the late 1970s are all a?tùkmã. The term i?tùk means "black," "very dark," or "it [is] black." The affix a generalizes ?tùk, and the second morpheme is only orthographically different (for esthetic purposes) from i?tùk. The term kàà does not denote "red" but can be symbolically associated with this color.


Recruitment to either moiety is by name-set affiliation [III.E.4.c] [IV.A.7.a.(1)]. The name, or rather the set of names, given to a name-receiver (usually a named-niece or named-nephew) by a name-giver (usually a naming-aunt or a naming-uncle [II.D.1.b]) belongs to the Red or the Black moiety by tradition [IV.B.1.c] (Figures 3537). It is the only social grouping into which females [III.C.9] are recruited by name-set affiliation. Otherwise, name-set affiliation applies only to men [III.C.5,6,8.c]. In the rare cases when a person changes her or his name-set [IV.A.5.e.(3)], the person might thereby also change her or his Regeneration season moiety affiliation (Nimuendajú, 1946:85), depending on the chance affiliation of the new name-giver.


The Regeneration season moiety system serves as the organizational principle for log racing and for their special meeting places in the plaza. The Red and Black moieties assemble on the edges of the plaza (Figure 25) for the usual morning and evening council meetings: the Reds meet on the eastern side and the Blacks on the western one. When their separate meetings are over, one prominent person of each moiety (the same men each time) meet in the center of the plaza to talk over any matters of consequence. I have seen this occurrence only among the Apanyekra. The Canela say that the apparent leaders of the Regeneration season moiety system, the i?kakhàl-katê, are not really leaders of the tribe in the political sense or as ancient war governors. They are just individuals who call their groups together and act as their spokesmen. Thus the Regeneration season moieties lack real "leaders" (cf. Lave, 1977:316–317) as the Canela currently use the terms pa?hi (Glossary) [II.B.1.b] and hààprãl [IV.C.1.d.(1).(c)], or even the more ceremonial më hõõpa?hi [II.D.3.i.(2).(b)].

[III.C.4.c.(1)] Alternate Ascendancy Status Of Moieties

The moieties alternate in holding the position of ascendancy [IV.A.4.b]. The leader of the moiety in ascendancy is called the i?kakhàl-katê (summoning-master) (cf. Nimuendajú, 1946:85) and the leader of the other moiety is referred to as the i?katsun-katê (meddling-adept) which, refers in part to the traditional behaviors occurring before, during, and after the log racing itself. When going out to the starting place of the afternoon log race, the members of the ascendant moiety suffer small indignities at the hands of the members of the "lesser" moiety exhibiting mock antagonism. For example, items in pockets or baskets are sometimes taken away. The attitude is one of leveling with respect to ownership, the lesser people taking items from those who are temporarily greater [IV.A.4.c].

The Blacks assume ascendancy after the erection of the Kô?khre log (Figure 45) of the Closing Wè?tè festival [IV.A.3.e,4.a] in September or October. The next day the Ayrën ceremony [IV.A.3.f.(2)], which includes the first log race typical of the Regeneration season, is held. All the racing logs of this Mëipimràk (i.e., Regeneration) season are made of cross-sections of buriti palm trunks, which have smooth gray bark. All the Black moiety logs are painted longitudinally with several heavy black charcoal stripes. They are cut long and thin (hapyê-shape) [V.A.5.a.(1).(b)]. The first Black moiety logs are less than an extended palm in length (about 22 cm) and perhaps only 5 centimeters wide (Katàm-re logs). Gradually, during the course of this first Black ascendancy period, the logs (exemplifying regeneration and growth) are cut larger, up to a little over a meter in length and a quarter of a meter in diameter (Katàm-ti logs).

After a few weeks, the Reds take over the supremacy and paint the two logs (called Wakmê-re) with a solid red urucu circle on both of the cut sides at the center point of their cross-sections on the somewhat spiny longitudinal strands of the buriti palm pulp. (The terms katàm and wakmê were not translatable.) These first Red team logs are small enough to be held in one hand and are shaped like wheels or discs, but they are made larger with time until they become large logs (Wakmê-?ti ones), though their length is still less than their diameter (i?po-shape) [V.A.5.a.(1).(b)]: the Red moiety characteristic. With the Reds in ascendancy, their leading man becomes the i?kakhàl-katê and the Black representative then is the i?katsun-katê. The Blacks harass the Reds in return when the latter attempt to reach the log racing site before the race. A few weeks later the Blacks take over the ascendancy again and go through similar procedures, with their logs made middle?size this time; still later the Reds assume the ascendancy to the end of the season.

[III.C.4.c.(2)] Ayrën Ceremony

The only Regeneration season moiety system ceremony that takes more than a few hours to perform is the Ayrën [IV.A.3.f.(2)]. This activity, lasting all day and ending in a log race, is performed the day after the Closing Wè?tè festival has ended. Here, women choose (Figure 47) men of the other moiety to go hunting for them and may or may not reward them with sexual favors if the men are successful. Then, the men race home to the village (Reds versus Blacks) with logs cut in the Regeneration season style and bearing a Regeneration season name (Katàm-ti).


During October, November and December, the Canela and Apanyekra (who traditionally raced according to the Regeneration season moiety patterns) now live scattered throughout the backlanders' region, sharecropping or working at odd jobs for the backlanders [II.B.3.j.(1)]. Thus, the economic deficiencies in the 1910s [II.B.1.d], and especially since 1947 [II.B.2.b.(1)], may be factors in the almost total loss of the Regeneration season moiety system. In more recent times, the Canela usually race once or twice in the Regeneration season moiety manner after the Ayrën ceremony, before returning to the age-set moiety system for log racing long before the traditional January termination of the Mëipimràk season.

The Apanyekra, in contrast, have preserved more of their Regeneration season moiety practices. They have several Ayrën ceremony performances, one of which I attended in December, 1974. After their Red and Black moieties met on the eastern and western sides of the plaza, their representatives met in the center of the plaza for a few minutes just to communicate decisions about the activities of the coming day (as the Canela had told me they used to do.)

A great deal will be learned about these matters when it becomes possible to compare the elements of the Regeneration season moiety system cross-culturally throughout the Eastern Timbira area, but especially with the Krahó and Melatti's monograph (1978) on the Krahó festival system.

Regeneration season moiety system activities appear infrequently as isolated acts in all the Wè?tè season festivals except the Masks' one, which is of Krahó origin. That is, there are moments in these festival-pageants when the groupings, marching formations, or chanted words are suddenly switched from being age-set moiety-oriented to being Regeneration season moiety-oriented. For instance, such a switch from age-set grouping to Regeneration division occurs when the Pepyê novices return to their cells after their commandant has scared and harassed them during his first appearance at the end of his internment (Nimuendajú, 1946:193). Just after this harassment, the novices return to their cells, the Reds filing clockwise around the outside of the circle of houses, while the Blacks march counterclockwise. It appears that at this point in an age-set oriented festival they needed to divide the novice troop (one unit in the age-set system) according to some principle involving halving and chose the Regeneration moiety possibility, still available from an earlier era.

[III.C.5] Plaza Group Moieties

The Canela and Apanyekra each have two moieties composed of six plaza groups, with three groups in each moiety [IV.A.7.b] [V.A.5.b.(1)] (Nimuendajú, 1946:87–90). Recruitment for all men to these plaza groups is by name-set affiliation, for girl associates by Pró-khãmmã appointment and group membership choice. These plaza group moieties appear throughout the Eastern Timbira tribes, the number and names of the plaza groups differing. The Krahó and Krĩkatí each have eight plaza groups, four in each moiety, their names varying. The Canela and Apanyekra have the same names and numbers for their Khêêtúwayê and Pepyê festival plaza groups, but the Canela have different names for their six Fish festival plaza groups (Figure 17).

The principal difference between the Canela and Apanyekra is that during the Pepyê festival, the Apanyekra plaza moieties act as the operating units, instead of the plaza groups. For the Canela, the plaza groups operate separately in the Pepyê and Fish festivals. However, among the Canela in the Khêêtúwayê, except during the final all night singing, the plaza groups merge into their two moiety divisions for the principal daily acts (Plate 41c).


A plaza group is characterized by (1) a special name (Figure 17), (2) certain name-sets for male membership, (3) a precise position in the plaza for meeting and chanting, and (4) membership in either the Upper or the Lower plaza moiety. These moieties face each other in the center of the plaza, three groups of one moiety on one side and three on the other [V.A.5.a.(1).(c)].8 In the Fish festival, the Upper plaza moiety (Glossary), the Khèy-rum-më-nkàà-tsà (Upper-side-they-plaza-placed: those located on the Upper side of the plaza) members construct three huts in a row on the eastern side of the plaza facing west. The Lower plaza moiety (Glossary), the Harã-rum-më-nkàà-tsà (Lower-side-they-plaza-placed: those located on the Lower side of the plaza) members construct huts on the western side facing east (Nimuendajú, 1946:87). In the Khêêtúwayê festival the Upper moiety files in from their cell of internment (Plate 41a) in a house on the eastern side of the village (Plate 41b) and forms a row facing north, while the Lower moiety marches in from their cell on the western side and forms a row facing south (Nimuendajú, 1946:173-1974, fig. 12) (Plate 41c). While symmetrical in number, the plaza moieties definitely oppose each other during the daily act [IV.A.3.b.(1)] of the Fish festival, the Otter group of the eastern side being ascendant, though the moieties join each other during the terminal part (in the weir for all Fish) while the Clowns oppose ("eat") them as they escape from the weir (Nimuendajú,1946:229) (Plate 46a).


The plaza groups occupy a central position in the plaza where they carry out daily act roles during the Fish and Khêêtúwayê festivals. They act as separate groups in the Fish festival, while in the Khêêtúwayê festival these groups act together within their moieties. Thus, in the Fish festival the plaza groups are principal units of the daily acts of the festival, while in the Khêêtúwayê festival the plaza moieties are the two halves of the age-set being initiated, and the age-set halves are the principal units of the daily acts [IV.A.3.b.(1)]. The casual observer may not be aware that the age-set halves are really the plaza moieties. In the Pepyê festival the plaza groups appear only on "great days" (Glossary) [IV.A.3.b.(1)], not as principals in the daily acts of the festival. In the Pepkahàk, Mask, and two Wè?tè festivals the plaza group moieties do not appear at all. The men's societies are used instead.

The plaza groups have different names in the Fish festival in contrast to the Khêêtúwayê and Pepyê initiation festivals (Figure 17), but the group membership is the same in all three festivals. Moreover, the Canela themselves sometimes confuse the plaza group names of the initiation festivals with the ones of the Fish festival, mentioning an initiation festival name instead of a Fish festival one when the latter is intended. Also, in the alternative,"variant" form (Nimuendajú,1946:230) of the Fish festival, the initiation festival plaza group names are actually used instead of the Fish festival ones.


The Fish festival plaza moieties have full-time girl associates (Glossary), but the plaza moieties of the Khêêtúwayê and Pepyê festivals have girl associates for only two events: the night of the all night singing in the Khêêtúwayê festival (Nimuendajú, 1946:176) and the day of the genipap painting [III.F.5.e] of the novices in the Pepyê festival (Nimuendajú, 1946:194–195) (Plates 27, 42g).

Recruitment of girl associates to the plaza groups may be by appointment by the Pró-khãmmã or by the membership of the groups with the approval of the Pró-khãmmã. The Pró-khãmmã appoint the prestigious Têt-re (Otter) girl associate without consulting the Otter membership. This is the only example of a male festival unit having just one girl associate instead of a pair, and this difference represents higher prestige. The most highly prestigious Visiting Chiefs have no girl associates [III.C.7.a.(2)].

For the middle-to-low prestige Fish plaza groups, the members choose girl associates with the approval of the Pró-khãmmã. The recruitment of the girl associates of the plaza groups of the initiation festivals follows the same pattern, with the more prestigious Khêt-re (Dwarf Parrot) (Glossary) and Tsêp-e (Bat) girl associates being chosen and appointed by the Pró-khãmmã.

Female membership in plaza groups is not the same from festival to festival, while male membership is, unless an individual has changed his name-set, a rare occurrence [IV.A.5.e.(3)]. Membership shifting is infrequent, temporary, and deliberate and done for a particular reason. One generic Fish (Tep) member, Kawkhre, who was an Armadillo (Awtsêt) by initiation festival reckoning, joined the Otters (Têt-re) so he would not see his Tep fellow members having sex with his sister, one of their two girl associates for that year.


Plaza groups seem to be basic to the Fish festival but to be "inclusions," "inserts," or devices, and not principal elements, in the Khêêtúwayê and Pepyê festivals: the two initiation festivals. If this is the case, and only hypothetically, the Fish festival might represent an earlier period in the evolution of the tribe's festival system than the two initiation festivals.

By "inclusions" or "inserts" I mean that the plaza moieties appear to act as relatively isolated units within the overall context of the Khêêtúwayê and Pepyê festivals, and that they are not as much a part of the whole festival-pageant (not single-acting, festival-length units). In the Fish festival, they are the principal units along with the Clowns. In organizing the festivals, the elders of some earlier period might have used the plaza group moiety concept of a still earlier period to divide the Khêêtúwayê novice age-set into two groups. In addition, such festival-designing elders might have used the plaza moiety concept for reviving the earlier more individual and family-oriented practice of "uncles" disciplining "nephews" [III.A.2.r.(1)]. The Pepyê festival is more impersonal and tribal-oriented wherein the commandant and his hierarchy of agents discipline the age-set initiates [II.D.3.d]. In any case, while the age-set moieties clearly provide the principal social contexts for the initiation festivals, the plaza group moieties seem to be "fitted in" to enhance the proceedings by providing special structures for them. Thus, it is plausible to conceive of plaza moieties as "left overs" from an earlier time, now used as devices for organizing specific festival situations.

[III.C.6] Men's Societies

The men's societies (Nimuendajú, 1946:95–97) participate in the Pepkahàk, Mask, and Closing Wè?tè festivals only, and so are exclusive with respect to the plaza groups that are found in the Khêêtúwayê, Pepyê, and Fish festivals. These men's societies have recruitment through name-set affiliation just like the plaza groups, but no correspondence between affiliation in the plaza groups and membership in the men's societies exists. The men's societies are the following (see Glossary): Falcons (Hàk) in the east versus Ducks (Kôkayu) in the west during the Pepkahàk festival (Plates 44, 45), and Masks (Ku?khrùt-re) with Jaguars (Rop) both in the east versus Agoutis (Kukhên) in the west during the Mask and Closing Wè?tè festivals, though the Masks sometimes help the Agoutis against the Jaguars [V.A.5.b.(1)] (Plates 48, 49, 52a,c,e).

Two similar groups exist called the Clowns and the Visiting Chiefs (Tàmhàk), which I do not consider "men's societies" (cf. Nimuendajú, 1946:96–97) because they are not name-set oriented and have unique ways of recruitment [III.C.7]. The Clowns take in new members through a combination of the group's selection and the individual's choice, the Pró-khãmmã having no input, while the Visiting Chiefs, at least in earlier times, had modified patriline succession.

Another associated festival group, the Pepkahàk troop, helps the Falcons compete (in log races) against the Ducks, though the troop is not a men's society, not being name-set affiliated. The Pró-khãmmã appoint each Pepkahàk member, who is a mature man, to membership in the Pepkahàk troop (the festival's principal group) for an adult reenactment of the food and sex restrictions of the initiations festivals [IV.A.3.c.(2).(a)] [IV.D.3.f].


In terms of their physical formations in ceremonial contexts, plaza groups and men's societies differ mainly in that the former take their positions in the center of the plaza and face each other in equal group numbers (i.e., 3 versus 3), while the latter maintain their positions on the edge of the plaza and face each other across the plaza in an asymmetrical relationship (i.e., 2 versus 1 [V.A.5.b.(1)]). The Pepkahàk troop races with the Falcons after their internment is over. Actually, in the terminal phase of the Pepkahàk festival, at least one log race takes place during which the Visiting Chiefs and even the Clowns race with the Falcons and Pepkahàk troop against the Ducks. These alliances are traditional. However, I was not able to resolve the question of overlapping memberships in this situation. Nevertheless, the Pepkahàk are inherently opposed to the Clowns, research assistants said, but in this one situation the Clowns help the Pepkahàk so that the analyst has to look at the context in each case before making interpretations about oppositions.

The plaza groups are referred to as më-nkààtsà, while the men's societies have no general term of reference. The men's societies have full-time girl associates like the plaza groups of the Fish festival. All of these social units (both plaza groups and men's societies) race with logs in their combined moiety formations, and rarely as individual societies. Plaza groups and men's societies also operate for the full duration of their festivals, except that in the Pepyê festival the plaza groups appear, disappear, and reappear sporadically. In the Khêêtúwayê festival the presence of the plaza moieties is disguised; they are actually the two operating halves of the age-set being initiated.


The recruitment of girl associates for men's societies is similar to their recruitment for plaza groups. The Falcon and Mask girl-associate positions are more prestigious and so are appointed by the Pró-khãmmã, while the others are selected by the membership of the groups with the approval of the Pró-khãmmã. Uncles can veto their nieces' selection (Nimuendaju, 1946:96).


J. Melatti (1979a:47–48) writes that among the Krahó, men might join a certain festival moiety by choice one year and the other moiety of the same festival the next year. The Canela were not yet doing this in the 1970s, though signs appeared suggesting that voluntary and temporary recruitment would eventually become the rule.

During the 1970 census taking, I asked men what festival groups they belonged to and found many young ones were doubtful of their memberships: they would have to ask their naming-uncles to know the next time around. Because such naming relationships were becoming tenuous, they might join the group they wanted to be with rather than the society they traditionally belonged to. Young men are more likely to forget their plaza group and men's society memberships than most other festival name-affiliated memberships because most of the latter are more individual and special. The plaza group and men's society memberships are more general and mass-oriented. When rights to name-set affiliated roles are also haakhat-related (Glossary) [III.C.8.a], their ownership is too prominent and prestigious to be forgotten, even by young men.

[III.C.7] "Wetheads" and "Dryheads"

All Canela males are either Wetheads (më ka-?khrã nkràà: they generalizer-head wet) or Dryheads (më ?ka-?khrã nkoo: they generalizer-head dry) (Glossary). The state of being wetheaded is called hàmren (Glossary), but no contrasting term exists for dryheadedness. Wetheads are more sensitive, conservative, serious, and inhibited by "shame" (Plates 27f, 68c), research assistants say, while Dryheads like to joke, play tricks, tell bad stories, lie, and are little restrained by "shame" [III.A.3.c.(3).(a)] (Plate 77b). The only women considered Wetheads are the girl associates of the three internment festivals and Wè?tè girls, and they keep this status for life. They are called pep-khwèy (Glossary) (warrior women) rather than hàmren, although this terminology is not used often today. Wetheads often have high ceremonial roles, while Dryheads have ceremonially less prestigious ones or none. Wetheaded roles and behavior are epitomized by the members of the Visiting Chief association (Tàmhàk), while the relative lack of inhibition by shame and the relative informality of Dryheads are epitomized by members of the Clown association. The Visiting Chiefs may number about 50 men and meet only twice during the Pepkahàk festival. They have no girl associates. In contrast, the Clowns, also numbering about 50, are the principal group of the Fish festival, vying in power even with the Pró-khãmmã for the duration of the festival. The Clowns have two girl associates, usually "infamous" ones (Plates 46b,c) [IV.A.3.c.(4).(b)].

Hàmren people do not constitute a formal group; they are individuals who are permanently "wetheaded" unless caught by the Clowns in a Fish festival. They become Wetheads when young through a number of traditional ways, but principally in one of three ways: (1) appointment to a certain festival position by the Pró-khãmmã, (2) transmission of a certain name-set from a name giver, or (3) inheritance of Visiting Chief membership through "patriline" transmission. The relative degree of ceremonial high status attributed to any particular hàmren individual varies with the way this status is gained, from the first (low) to the third (high) category.

The age-set file leader (mamkhyê-?ti:) is an example of the lowest level of hàmren status. He is appointed by the Pró-khãmmã and operates in this position almost all the time, in and out of initiation festivals. The chief [I.G.14] (Plate 68c) of the Visiting Chief membership is second highest in ceremonial prestige, his position being related to name-set transmission. The Ceremonial-chief-of-the-whole-tribe (Glossary), Kôyapàà [I.F.2.a] (Plate 70a), the son of the younger Kaapêltùk (Figure 51), has the highest prestige, along with Khrùt and his son. If this role cannot be passed on traditionally, due to the lack of a son, the Pró-khãmmã will choose another patriline, giving the role to their members. The Pró-khãmmã are the final arbiters for almost any ceremonial role determination when a traditional line terminates. However, they rarely have to exercise this authority [III.D.2.c.(1)], because candidates having traditionally correct kinship ties usually exist.

Although all Canela males are either wet or dryheaded, this dichotomy does not constitute moieties because the two groups come together as such only on two occasion during the Pepkahàk festival: in the late afternoon of the highest ceremonial day during the terminal phase of the festival to reaffirm and recognize their association so that each man knows where to go on the following day. The Wetheads go out to a farm with the wives of the Dryheads to carry out the usual events of the Wild Boar day [IV.A.3.f.(1)], including extramarital sex before a log race, while the Dryheads remain in the village with the wives of the Wetheads.

It is hard to be sure whether the Apanyekra share this tradition of wet-/dryheadedness with the Canela, because Apanyekra research assistants tended to always say yes when asked about such matters. This is because they did not want to appear to be lacking something that the more prestigious (to the Apanyekra) Canela had. Nevertheless, I assume the Apanyekra did have such a tradition, but have lost it in terms of still acting out associated ceremonies. The Krĩkatí also have remnants of this tradition (Lave, personal communication).


The Visiting Chief or Tàmhàk (urubu-rei: vulture king: Gypagus papa) membership is distinctly different from any other socioceremonial unit because of its traditionally patriline-oriented succession (cf. 1946:99, 217). The Visiting Chief membership, composed of Nimuendajú's (1946:97–100, 217–218, 223) "courtesy chiefs," is made up of intertribal male representatives. Eastern Timbira tribal male membership is from father to son; female tribal membership is from mother to daughter.

Representatives of five different tribes, which were formerly separate tribes, now live among the Canela. During the preparation for the Visiting Chief procession during the Pepkahàk festival, Canela men meet in five groups, the members of which are descendants of the five tribes. Traditionally, each member of the Visiting Chiefs was born in one remnant of an ancient tribe and is the honorary chief of another one. Thus, there should be only 20 Visiting Chiefs performing in the procession. There are, however, at least twice that number. I call this Tàmhàk membership "Visiting Chiefs," because each Tàmhàk acts as a resident receiver and ambassador for visitors from the tribe that made him such a chief when he was visiting them in their village (Nimuendajú, 1946:99–100). When members of one Eastern Timbira tribe visited another Eastern Timbira tribe in precontact times, they first communicated with their Visiting Chief in the other tribe to obtain his protection, sponsorship, and hospitality [IV.C.1.d.(1).(a)]. Otherwise, it was dangerous to go near the village of the other tribe [IV.C.1.c.(2)].

In 1958, the older Kaapêltùk gave me the names of the six ancient tribes from which the present Canela descend: Mõl-tum-re (going-along experienced dim.), Krôô-re-khãm-më-?khra (boar-dim.-location-Timbira-children), Tsoo-khãm-më-?khra (fox-location-Timbira-children), Karë-?katêyê (mud enemy-people), Apàn-yê?khra (piranha honorable-plural children), and Hô-?ti-khãm-më-?khra (hair/feathers-aug.-location-Timbira-children). Only descendants of the first five of these ancient tribes still existed. The younger Mĩĩkhrô said that the Mõltum-re and the Ràm-khô-khãm-më-?khra (almecega-tree stand-of location Timbira children) are the same people. The other tribes joined the Ramkokamekra-Canela, the central tribal group. Consequently, the Mõltum-re sit in the middle of the plaza on the evening before the procession of Visiting Chiefs, because they were the original tribe, while the other four tribes sit at the edges of the plaza in the direction of the former locations of the tribes from which they came. The name "Ramkokamekra" is not used today; "Mõltum-re" is the name of the tribe from which most Canela believe they are descended.

The Visiting Chiefs are hàmren by the simple fact of being Visiting Chief society members (like the "Honorable" ambassadors of the USA), and the required formal and conservative behavior follows into their daily lives to some extent. They are not the only hàmren members of the tribe, however. The holders of a number of other festival roles receive the honor of being considered hàmren. Nevertheless, Visiting Chiefs do characterize high ceremonial behavior, they say, and as such are all "wetheaded." For me, the noticeable restraining characteristics of "honorable" behavior are carried over into daily life more for the very prestigious people who have individual roles than for the many Visiting Chiefs who have only group roles in a procession.


Traditionally, a Visiting Chief's son, or his brother's son, or his parallel "brother's" son through all male links, assumed the role of Tàmhàk. This pattern of succession kept the role in the same remnant Eastern Timbira tribe, because Eastern Timbira tribal membership is from father to son and from mother to daughter. If the succession was given to a sister's son, or to a mother's sister's son, the right to hold a certain Visiting Chief's office might have passed from one remnant tribe to another one and not have remained in the same tribal line, the chief-chooser's and the chosen-chief's tribal line. Thus, passing the role to a nephew instead of to a son might have obviated the purpose of these visiting chieftainships, which was to provide continuous protection down the generations between the same two tribes, now both tribal remnants living in a larger tribe. In later times, especially with the current population expansion of the tribe and the diminished need for these individuals as peacemakers, a man's "sister's" son often takes the position, or in a house where no one else has the role, some young male assumes the Visiting Chief's role. This is consistent with the tendency for the members of every family household to want and demand a presence, as well as a material gain (usually food) for their performance in almost any ceremonial situation. Such open flaunting of traditions happened in the 1970s because the Pró-khãmmã were too weak and powerless to prevent it, as surely they were at other times during the 19th and 20th centuries near the end of the cycle of their 20 years of tenure [III.D.2.b.(2)].


The Visiting Chief membership assembles twice during the Pepkahàk festival: once in the evening before their formal procession to plan and prepare for this occasion (sitting in their traditional positions in the plaza; Nimuendajú, 1946:217–218), and again for their midafternoon procession to the plaza (continuing on to certain traditional houses, Plate 44d).

The Visiting Chiefs have several limitations that characterize their high ceremonial status: no girl associates and no "little boy" ceremonial members (më ?khra-re: pl. child dim.)9 (unlike most other social groups), and their infrequent appearance in the Pepkahàk festival, which is held only every 5 to 10 years. White is their associated color (chalk and falcon down) [II.F.5.a,e]. However, in 1958 red paints were mistakenly applied to their bodies during their ceremony. I have photographs showing that the Visiting Chiefs applied red urucu paint between the vertical lines of falcon down (Glossary) that year. Old research assistants explained in 1970 why this red paint was wrongly used, so that in 1970 and again in 1979 this mistake was not made.


Clown (Glossary) society individuals, the Më-?khên (they somewhat-bad: the mischievous and individualistic ones), have no shame (më pahàm naare: they shame not) (Glossary), research assistants say. They are less restrained by many of the traditions related to politeness, courtesy, and consideration for others [III.A.3.c.(3).(a)] [III.B.1.d.(3)]. They characteristically are brazen and do spontaneously what they want to do (Plate 46d).


Recruitment is by mutual agreement between an individual and members of the Clown society, as described in Nimuendajú (1946:96–97). When a serious person who is hàmren wants to become a person who jokes much of the time and sometimes plays tricks on others, he can join in the breaking of a wasp's nest (Plate 45b) at an early point in the Pepkahàk festival (Nimuendajú,1946:214). Then the Clowns know he is willing to join them and will induct him into their group near the beginning of the next Fish festival performance. Similarly, if a hàmren, overburdened with his "seriousness," begins to joke too much and to move away from his hàmren-type personality state, the Clowns may still try to "catch" him and bring him into their group during the next Fish festival, even if he has not struck the wasps' nest. As is usually the case in such matters, however, Clown members ask relatives of the lapsing hàmren whether he would acquiesce in his capture. Sometimes to have fun, the Clowns try to surprise such a person without warning, approaching him forcefully at first in the context of the festival to amuse the crowd. The Clowns approached my sister, Te?hôk, about capturing me in 1970, but she said no, preferring me as a hàmren, so they desisted.


The two girl associates of the Clowns (Plate 46b) are chosen by the Clown membership; the Pró-khãmmã have no say in the matter. Once the Fish festival is in session, the Clowns brook few interferences from the Pró-khãmmã. The Clowns often choose ex-Wè?tè girls, bringing them down from the sublime to the ridiculous. As Wè?tè girls, and therefore as ceremonial kin to the opposite age-set moiety from their fathers' [III.E.10], they are never really considered girl associates; they are not exposed to the same sexual training for extramarital purposes as girl associates experience in general [III.A.2.j.(6).(c)]. Thus, it is quite fitting for them to serve later as girl associates to the Clowns. Nevertheless, such a girl's mother and uncles can veto the Clowns' request. I remember well how my niece Te?kurà's father (the Clown's leader) had to absent himself from the family meeting over the Clowns' demands for Te?kurà as one of their girl associates in 1964. My family acquiesced; Te?kurà's father would not have, because he had "great caring" [III.B.1.b] for his daughter and was embarrassed to think of her having this kind of sexual experience, even though he, himself, was a Clown.

The Clown girl associates have a prominent festive role on one of the "great days" in the middle part of the Fish festival. Here they act out a number of Canela evils to the great amusement of the crowd assembled to enjoy this dramatic act (Nimuendajú, 1946:228) (Plate 46c). One of them drops her "baby" (a buriti pulp doll) to have mock incest with her "brother" while the doll, lying on the ground, cries outrageously—the voice of the other Clown girl. In their selection of girl associates and male members, Clowns like to seek those who already have, or who can learn to display, a flamboyant and independent personality.

[III.C.8] Ritual Societies

The haakhat-oriented ritual societies (Glossary) [IV.A.5] perform in the Fish, Mask, and Closing Wè?tè festivals and the Sweet Potato, Corn Harvest, and Pàlrà rituals. These "river-oriented" (W. Crocker, 1979:242) festivals allude to animals and plants in their ceremonies. The appearance of ritual societies in the Pepkahàk festival as well, however, undermines any one to one correlation between river-oriented festivals and haakhat-oriented societies. The ritual societies do not perform in the two initiation festivals (the Nkrel-re) or the Regeneration moiety system, although research assistants said that the institution of the two Wè?tè girls was once held by two families matrilineally (each in a haakhat); but this is not the case today.

The existence of the Canela haakhat, and Nimuendajú's mistaking the limited occurrence of the haakhat for overall tribal matrilineality (Glossary), are points I first raised and clarified in two articles (W. Crocker, 1977, 1979). Only about one-fifth of the Canela families possess a ceremonial haakhat. Some Apanyekra say they have the haakhat tradition also, but I am convinced after careful questioning that they were merely saying they had the tradition to appear to be more like the Canela, whom the Apanyekra consider more prestigious.

Members of ritual societies may appear momentarily as a group in a festival, as do the Turtles in the Fish festival (Nimuendajú, 1946:230); or they may be the principal individuals managing a whole festival, as are the members of the Corn ritual-holding family in the Corn Harvest festival (Nimuendajú, 1946:62–63) (Plate 53). Ritual societies may also be composed of many individuals, as in the case of the Sweet Potatoes of the Sweet Potato ritual (Nimuendajú, 1946:63–64) (Plate 47a,c), or they may consist of just one person (and therefore not be a society at all), as in the case of the cutter of the Pàlrà log in the Pàlrà festival [IV.A.5.e.(2)] (Nimuendajú,1946:163–165).


What the ritual societies have in common is that the right to carry out a ritual's roles may be possessed by an individual, by a family, or by one matrilaterally extended family, all of whom are based in one house, or by a series of adjacent houses along the village circle. In this sense, a haakhat (Glossary) may be a small part of a longhouse, a large part, or the whole longhouse [III.E.2.e.(2)] (Figures 24, 25), and the term haakhat denotes both a ritual and a physical house or set of houses.

In the term haa-khat, the morpheme haa-, or tsaa-, is a particular generic kind of vine. The term -khat is always a somewhat massive base with a "root" below it and something larger growing out above it. This three-part structure can be visualized horizontally as well as vertically and refers to an object as well as an idea.

For instance, night has a base (katswa-khat: night's-base): initial and full darkness. It also has a root: twilight, which is an extension back into earlier time from the base. Then, the major part of the concept, extending forward in time from the base, is the rest of the night. A tree is an obvious example with its roots, broad base, trunk, and branches, research assistants pointed out. In the case of the haakhat, family lines going back through time are the roots, ritual-holding families in their houses along the boulevard are the base, and future generations of descendants who are going to perform the ritual year after year are the vine or branching tree.

Within a haakhat, which is both a series of houses and a ritual, membership is passed on in one of three ways, which may change from generation to generation. The membership may be transmitted through name-set affiliation (e.g., the Kô?khre log hole-shapers [IV.A.3.e.(3)]), through matrilineality (e.g., the Corn Harvest festival family [IV.A.5.a,d]), or through a mixture of these two principles (e.g., the Capybara, see Nimuendajú, 1946:230). Because a male name-set goes out from the owner's natal house with a man's marriage and his move into his wife's house and returns to his natal set of houses when the man names his sister's son (or his "sister's" son), a male name-set can remain within a haakhat generation after generation (Figure 36). Female name-sets cannot do this; they circulate around the village (Figure 35), because a woman's names move into the families where her brother or "brother" has a daughter. A man usually has classificatory nephews outside [III.E.4.a] (Figure 37) his natal haakhat as well as in it. If he holds traditional rights to festival rituals, he must remember his nephews in his haakhat and pass on his ritual rights to one or several of them, depending on the tradition and nature of the particular ritual. However, if he passes the ritual to a "nephew" born outside the ritual's haakhat, he may be violating the ritual's tradition, and thereby contributing to the deculturation and changing of the system, an act which often happens these days.


Plaza groups and men's societies have name-set transmission but not matriliny. They do not have the haakhat. However, these groups and societies meet during a specific festival at traditional houses located along the village circle. These locations are called standing places (më ?ku?hê tsà: their standing place) or sitting places (më ?khrĩn tsà: their sitting place) (Glossary), depending on how much time they spend there—"standing" if brief and "sitting" if lengthy. Nimuendajú (1946:210) calls these ceremonially-designated family houses on the village circle "lodges." Each standing or sitting place remains in the same narrow sector of the village circle "forever," research assistants say. Thus, these lodges remain with certain families for generations; but individuals of these families are not necessarily members of the plaza group or the men's society who gather in these families' houses for festive purposes. If family members are lodge members, it would be by coincidence.

The festival society's members stand, sit, rest, eat, or decorate their bodies as a group in these houses, depending on the tradition of the festival act being carried out. They may perform their ceremony or ceremonies only in the traditional house, but they usually come out of it to perform activities in the boulevard, down the radial pathways, or in the plaza. The families who live in these houses (lodges) are proud to have the members of the plaza groups and men's societies temporarily under their roofs, and they are pleased to serve them water and perhaps some food, though they do not own or control the visiting groups or societies.

If the family that is living in the correct position to serve a ceremonial lodge no longer has female descendants to maintain the matrilines of the house, that house will cease to exist. Families of a neighboring house on either side will occupy the space of the terminated matriline and take over the role of serving the festival society that traditionally has its lodge in that immediate part of the village circle. The Pró-khãmmã decides which family will assume this responsibility. Some male research assistants point with nostalgia to houses not their own, or to spaces between houses, to indicate where on the village circle they were born and where their sisters' matriline ended.

In new villages, the position of a ceremonial lodge on the village circle in relation to the sun should always be the same. There can be no question of matrilineal succession to ownership of a festival lodge. The lodge is owned by the festival society members in the sense that they have the right to use the house in that particular position "forever." Hosting a festival lodge is a temporary right, held maybe for just a few generations, of a family matriline that happens to live "traditionally" in this part of the village circle. The festival lodge of a plaza group or a men's society (or in addition, of the Visiting Chiefs, Pepkahàk, or Clowns) is not a haakhat.

In contrast, a haakhat ritual is owned by a family or a matrilaterally related set of families, though one family or one person may take the leading role in performing the owned ritual. If the owner's family line dies out (i.e., cannot provide the women or men needed to perform the particular roles), the possession of the haakhat (taken as a ritual) passes to some other matriline including its related men within the haakhat (taken as a set of matrilaterally related houses). If no matrilines in the traditional haakhat (set of houses) can maintain the ritual, the Pró-khãmmã gives the ownership of the haakhat ritual to another family or set of families and their houses. This happened to the ownership of the Corn Harvest festival in early March of 1979. Thus, a haakhat has no traditional location on the village circle "forever." Consequently, it is quite unlike a festival standing or sitting place (a ceremonial lodge) which has a precise traditional location. Because a haakhat is corporate, it continues "forever" in the possession of one or another family in different parts of the village circle.


The general trend for membership transmission within a ritual society is away from matriliny and toward name-set transmission (Glossary). Although name-transmitted now, most ritual societies still are inherited within a haakhat. However, other rituals exist that are passed on through name-set transmission without being restricted to the confines of a haakhat (set of houses), going instead to any part of the village circle where a name-giver has a sister's or "sisters's" son.

In 1975, I compiled complete records of almost all ritual transmittals for as many generations as the oldest research assistants with good memories [Pr.2] could remember. The pattern of transmission of some rituals is consistently through matriliny or naming. The transmission patterns of others varies between matriliny and naming as individuals chose, or were forced to choose because of the lack of traditionally appropriate ritual-receiving personnel. Name transmission now is much more frequent throughout the society than matriliny, which occurs relatively rarely. Haakhat inheritance, as a process, is more appropriate for large matrilaterally extended families and villages with larger populations, while name transmission is more adaptable to survival in smaller populations, because the ritual may be passed on through this mechanism to any nephew in any place on the village circle.

It is easier for a young man these days (not knowing or caring about the traditions) to resort to name transmission rather than to matriliny (Glossary) when he has a role to pass on, a nephew to receive it, and no naming-uncle close at hand to tell him of the traditional procedures. Matriliny restricts his choice to a haakhat (one or more houses) and, bothered by such an “unimportant” matter, a young man is inclined to forget this limitation and give his name-set as quickly as possible to any “nephew” just born anywhere around the village circle.

This mixing of name-set transmission with matrilineality may explain Nimuendajú’s (1946:63) seemingly ambiguous statement about Sweet Potato role performance rights: “Membership is inherited matrilineally with the personal name.” Most probably, Nimuendajú’s (1946:79) belief in earlier exogamous matrilineality for the Canela came from his awareness of just a few examples of matrilineally inherited rights to possess, or to perform in, certain ritual societies, even though these rituals only operated infrequently and were possessed by only a few families.


Succession and membership in ritual societies that are solely haakhat-oriented (rare) follow strictly matrilineal principles for both women and men. In some cases the whole consanguineal family succeeds to membership and performance in a ritual (Turtles: kaprăn-pey-re: Platemis sp.) (Nimuendajú, 1946:230). In other cases only one person succeeds to ownership and performance (the foam girl: the Kôyamprô kuytswč) (Nimuendajú, 1946:226) of a particular ritual. In such cases a daughter should succeed her mother, but often it is a younger sister or some other closely related female who succeeds, such as a parallel cousin. The principal persons in a haakhat (female or male) look for an appropriate person in a small field of kin to carry out the role, considering age, voice, relationships, ability, personality, and the degree of beauty (mpey-nă) (Glossary) to be presented. Of course, in a haakhat that involves only name-set transmission (Tsů?katę-re) (Nimuendajú’s “hũyakrékate,” 1946:167) women cannot be role performers because female name-sets occur only in the Regeneration moiety system [III.C.4.b]. Nevertheless, a woman related to the male name-set receiver may consider herself the owner or controller of the ritual, especially if the male receiver is very young. In the Pŕlrŕ log cutter’s haakhat no name transmission exists and the one role is demanding enough for one man: Paapôl, who received it from his older brother some years ago.

The Khęętúwayę, Pepyę, and Pepkahŕk festival girl associates are hŕmren in ceremonial status, as are the two Wč?tč girls who are not really girl associates [III.A.2.j.(6).(c)]. All these young women are referred to as being pep-khwčy (Glossary) (warrior-women) rather than hŕmren, but they hold hŕmren-honor rank (Glossary) nevertheless; these are the only women who do. These young women of honor are all appointed to their positions by the Pró-khămmă. These appointments are the highest kind of designation for women, bringing great prestige to their families [III.C.3.f]. Other kinds of access to festival positions for women are less prestigious, including haakhat succession, and especially selection by group membership.

Until recent times, these high-status girl associates and the younger of the middle-status girl associates—Falcons (Hŕk), Masks (Ku?krůtre?hô), Jaguars (Rop), and the one Otter (Tęt-re) plaza group girl—had to be accepted in these men’s organizations as virgins (Nimuendajú, 1946:96). Parents worked to keep their daughters virgins so that they could be selected to fill these positions. In contrast, the older middle-status and the definitely low-status girl associates (Ducks (Kôkayu), Agoutis (Kukhęn), Clowns (Më?khęn), and certain plaza moiety girls) were traditionally inducted into the men’s groups as non-virgins, chosen precisely for their perceived abilities for rendering pleasing sexual services and female companionship (Nimuendajú, 1946:228).

  [III.C.10] Hypothetical Development through the Eras of Organizing Principles

Various organizing principles are found in the festival system and may reveal an evolutionary sequence for three hypothesized basic periods: early, riverine, and bellicose. The principles of recruitment to group membership for the three periods are the following: early (Regeneration: female and male name-set transmission), riverine (haakhat: matriliny and male name-set transmission), bellicose (age-set moieties: relative age of males).


My hypothesis is that the Regeneration moiety system was characteristic of the early period because name-set affiliation for women appears only here, while the general trend since then has been away from this principle. Moreover, name-transmission membership to socioceremonial units for both sexes exists broadly among the Northern Gę, while the other principles do not.


Many river-associated traits (mostly faunal) occur in certain ceremonials and myths, especially in the Fish festival, that are not prominent in the present Canela environment. Old research assistants said that the piranha (apan: Serrasalmo sp.), swamp deer (poo-kahŕk: suassuapara: Cervus paludosus), and capybara (kũũtém: Hydrochoerus) did not exist in the Canela area, that ducks and herons (garça: Tigrisoma sp.) only flew overhead, and that only the small varieties of turtles (kaprănpey-re: Platemis sp.), stingrays (tsęwtsęt-re: Trygon sp.), anacondas (ro-?ti: sucuruju: Eunectes murinus), otters (tęt-re: lontra, ariranha) (very rare), crabs (pay: Cancer uça?), and alligators (mĩĩ: jacaré: Caiman niger) are found in the Canela headwater streams. Research assistants knew about these non-local animals and about the larger-size varieties of similar animals living either down stream in the forests to the north, or down the Alpercatas or the Itapicuru rivers to the northeast. They had seen these absent, rare, or larger animals in their travels. (For Latin equivalents, see Nimuendajú, 1946:65, 229–230.)

To account for this water-oriented misrepresentation of the current Canela locality in their festivals I hypothesize a lengthy river-oriented (Glossary) period of Canela residence and development along a sizable river with swamps, lakes, and land for more settled horticulture, such as the lower Alpercatas (likely), the middle Itapicuru somewhat further to the northeast (possible) [IV.C.1.b.(7).(c)], or even the larger Parnaíba considerably further away (less likely) (Maps 1, 4). The haakhat, and evolutionary development toward matriliny, could have developed in such a riverine environment because of the greater possibility for reliance on sedentary living and its increased possessions, and because of the likely need for leaders to control larger numbers of people and their competitive problems with respect to limited resources. Here, incipient matriliny could have replaced female name-set transmission to socioceremonial unit membership but not necessarily have replaced male name-set transmission, with which matriliny could co-exist and still does today.

Insufficient evidence exists to hypothesize whether full tribally extensive matriliny (Glossary) with clans and exogamy evolved, so I prefer to rest on my claim of “incipient matriliny” (W. Crocker, 1977). Nevertheless, I suspect that if exogamous moieties had existed in full force for a number of generations, matriliny should have prevailed over female name-set transmission, which would therefore either not exist at all or be considerably less evident today. However, Nimuendajú (1946:90) holds the plaza moiety groups may be remnants of village-localized clans but makes no similar claim for the men’s societies, writing only that “there is no indication that they anciently represented the two exogamous moieties” [III.C.1] (Nimuendajú, 1946:97–98).

Similarly, the rituals in the annual festival cycle between the Regeneration and Wč?tč seasons [IV.A.5] (Table 4) are clearly haakhat-oriented and therefore riverine. The Closing Wč?tč, Mask, and Pepkahŕk festivals have mixed riverine and bellicose period representations, and the initiation festivals are fully of the bellicose period, although traits appear in them from all three hypothesized periods.

The Opening Wč?tč festival displays none of the organizing principles discussed above, except that the institution of the Wč?tč may have been haakhat-oriented, research assistants say, though it is not so oriented today. The elders may have initiated this festival as late as post-pacification times, to introduce and authorize the extramarital sex freedom of the entire Wč?tč season [IV.A.3], which in its extreme seems a special development of the Canela even in contrast to the Apanyekra. Research assistants report that the elders created one ceremony in the first decade of the 20th century: the Apikrawkraw-re of the Pepkahŕk festival [II.B.1.d.(1)] [IV.A.3.c.(3).(e)], which survives to this day. Thus, they may have created the two-day Opening Wč?tč festival (and the Festival of Oranges) earlier in the 19th century, perhaps not long alter their stabilization around 1840 [II.B.1.b], as an adaptation to peacetime living, to the demands of women in the absence of warfare, and to enhance tribal cohesion.


The age-set system may have evolved as a defensive device against the devastating threats of advancing Brazilian pioneer fronts in Piaui (Map 4) and along the Atlantic coast and the potentially closer contact with bandeira troops [II.A.3.a.(1)].

Visiting Chiefs and Clowns, and Wetheads and Dryheads are organizing principles and socioceremonial units that are hard to place on this evolutionary continuum. However, their inter-tribal emphasis suggests the bellicose period, but the origins of wetheadedness could be much earlier. The nobility implied in hŕmren status (as I have seen it) encourages me to think in terms of relative peace and plenty—the riverine period. Nevertheless, no haakhat that is solely matrilineal provides hŕmren status, and the Fish festival in which the haakhat is most prominent has few or no roles providing hŕmren status in themselves. Thus, I tend to associate the Visiting Chiefs, Clowns, and hŕmren-ness with the bellicose period. I also associate Eastern Timbira matriline (Glossary) and patriline tribal membership (cf. Nimuendajú, 1946:217) with the bellicose period because the patriline principle is consistent with succession for the Visiting Chiefs and for the two recently established Ceremonial-chief-of-the-whole-tribe lines in post-pacification times [II.B.1.d.(1)].

[III.C.11] Summary and Discussion of Canela Socioceremonial Units

The most evident social groups are the age-set moieties because they operate almost every day outside of festival situations, socialize the young boys and men of the tribe into maturity, and serve as political bases for training future leaders [III.D.1.j] and for the festival-governing Pró-khămmă. Recruitment is by relative age, and membership comprises all men starting from childhood. According to Canela dualism, these moieties are in diametric asymmetrical opposition to each other, the Lower moiety always being ascendant over the Upper one (Figure 24).

The exogamous moieties reported by Nimuendajú (1946:79, 82) are inoperative today and Gę specialists in general believe they never existed. Apparently, Nimuendajú mistook [III.C.8.a] occasionally occurring festival matriliny, held by a limited number of families, for earlier tribal matrilineality with exogamous moieties [III.C.1].

The Regeneration (“rainy”) season moieties formerly oper­ated on a daily basis in their season; now they are scarcely in evidence, having degenerated through internal devolution (my hypothesis) and culture contact. Their recruitment is by name-set transmission, and their membership comprises all women and men from the time they are named, which is within a few days of birth [IV.B.1.c]. According to research assistants, these moieties are in opposition with each other, alternating their traditional ascendancy and therefore their asymmetry over each other. The Regeneration moieties are the principal socioceremonial units organized along concentric principles and the only ones that include all women.

The plaza groups and the men’s societies are similar because they are (1) operative only in festivals, not in daily life, (2) characterized by name-set transmission for men and by Pró-khămmă appointment or choice by the membership for women, and (3) arranged diametrically in oppositional moieties [V.A.5.b.(1)]. Their memberships consist of all the men of the tribe from the time they are named. Each group or society has just two women assigned to it temporarily for the course of a particular festival, except for the Otter plaza group which, prestigiously, has only one girl associate.

Plaza groups differ from men’s societies in that plaza groups face each other in the center of the plaza in equal numbers (three versus three), while men’s societies face each other at the edge of the plaza in unequal numbers (two versus one) [V.A.5.b.(1)]. Moreover, the plaza group moieties are named (Upper and Lower), while the men’s society moieties are unnamed. Each man’s name-set associates him with a certain plaza moiety and a certain men’s society, but no traditional correspondence exists between the two memberships.

There are two special social groups that complement each other: the Visiting Chiefs (Tŕmhŕk: King Vultures) and the Clowns (Më?khęn). These two social groups epitomize wetheaded versus dryheaded behavior: formality/informality, seriousness/jocularity, inhibited traits/gross ones, high ceremonial honor/low ceremonial values, order-followers/ individualists [III.B.1.d.(3)]. The Visiting Chiefs may number 50 and appear on only two occasions in the Pepkahŕk festival. Their recruitment was traditionally from father to son, father to his brother’s son, or from father to his ‘brother’s” son; that is, they are “patriline”-oriented, an aspect of high status [III.C.7.a.(2)]. Whether they really are a society is questionable [III.C.6].

This patriline-oriented succession (cf. Nimuendajú, 1946:99) served to keep the Visiting Chief’s succession within the same intratribal remnant, because Eastern Timbira male tribal membership passes from father-to-son (cf. Nimuendajú, 1946:217). If the membership in the Visiting Chiefs were passed to a sister’s son (often the current practice), it could go to a member of another internal remnant of a formerly independent Eastern Timbira tribe.

A Visiting Chief in earlier times was the formal protector that members of tribe A had when they visited within the territory of tribe B. The Visiting Chief (a member of tribe B) was selected by tribe A when he had visited tribe A (Nimuendajú, 1946:99). Thus, when members of tribe A wanted to visit tribe B, they communicated with their Visiting Chief in tribe B to request his promise for their peaceful access, protection, and sponsorship.

The Clowns (numbering about 50) are the leading Fish festival society, performing every day during the Fish festival. They have two outrageously behaving girl associates (Plate 46b,c), who are selected by the Clown membership. The Clowns gain new members by waiting for hŕmren-status individuals to join them (which is indicated by breaking a wasps’ nest in the Pepkahŕk festival) or, at the beginning of the Fish festival, they simply try to capture certain hŕmren individuals, who conspicuously joke or lie in daily life, hoping for their cooperation in their internment. Visiting Chief and Clown society memberships are independent of plaza group and men’s society memberships.

The Visiting Chiefs and the Clowns, both as social groups and individuals, are considerably more distinctive and dramatic than their plaza group and men’s society counterparts, and the individual behavior of the first two groups is far more likely to be carried over into daily life. The Visiting Chiefs are likely to be restrained (Plate 68c), while the Clowns may be adamantly independent [III.B.1.g.(3)] (Plate 46d). The Visiting Chief and Clown societies have lodges on the village circle in traditional locations in relation to the sun, just like plaza groups and men’s societies.

Research assistants stress that in Canela dualism, the Visiting Chief and the Clown societies are in clear opposition to each other, especially in daily living. Nevertheless, in certain festival situations (e.g., log racing) the Visiting Chiefs are paired in a complementary manner with the Pepkahŕk, with the Falcons, and with the Clowns, and all four groups are paired in an oppositional manner against the few individuals who remain with the Ducks. Opposition and complementarity depend on context in festivals and on individual intention [III.B.1.d.(1),(2)] within a context in daily living. All men are either Wetheads or Dryheads, and as such they act as individuals as much as in groups; they do not act significantly as members of moieties. Thus, I have not considered this very definite tribal male dichotomy a moiety system. Nevertheless, the men divide along these lines in the terminal phase of the Pepkahŕk festival for the Wild Boar day [IV.A.3.f]. The majority of Wetheads are Visiting Chiefs, so young Canela tend to confuse the terms hŕmren (wet-headedness) and Tŕmhŕk (Visiting Chiefs).

In Canela dualism, Wetheads and Dryheads are in opposition and represent the extremes of “high” and “low” ceremonial honor [III.B.1.g.(4)], while the Pepkahŕk and the Clowns, though sometimes in opposition and sometimes in complementarity, represent intermediate positions, being somewhat high and somewhat low [III.B.1.g.(3)]. In daily life and in the context of the Fish festival, the Pró-khămmă councilors are in opposition to the Clowns but cooperate with them, and so are similar to the Pepkahŕk. Both the Pró-khămmă and the Pepkahŕk are composed of Wetheads and Dryheads. Political leaders can be either [III.D.1.i.(2)].

A “ritual” (Glossary) society is characterized by having a haakhat, which is a family- or extended family-owned right to carry out a certain traditional ritual. A haakhat (Glossary) also is a certain number of houses along the village boulevard defined by the occupation of these houses by the matrilines who own the ritual. (Young Canela confuse these two usages of the word haakhat.) Ritual societies are composed of one person or many individuals. Their festival performances may be completed in less than a minute or may take a number of days or years. Inheritance of membership is through matrilineal relationships, name-set transmission, or by a combination of these two principles [III.C.8.a]. If a family does not have the traditionally correct personnel to carry out its ritual, the right to possess this ritual may have to be transferred by the Pró-khămmă to another family that lives in any position on the village circle.

The festival haakhat lasts “forever,” research assistants say. Such rituals are found in all the Canela festivals except the Regeneration system and the two initiation festivals (the Nkrel-re), but are most frequent and conspicuous in the Fish, Sweet Potato, Corn Harvest, and Pŕlrŕ ceremonies: the river-oriented ceremonies, which emphasize fish, and water animal and bird representations.

Unlike the haakhat, festival “lodges” (Nimuendajú,1946:220) [III.C.8.b] are controlled by the festival groups that use these houses for their activities. The family serves the festival groups (e.g., plaza groups, men’s societies), giving the members water and maybe food, but the family does not “own” the unit or its ceremony, as they would if it were their haakhat. If the family matriline dies out, the Pró-khămmă appoint one of the neighboring families to continue the service. Festival lodges exist “forever” in the same locations.

In contrast to the age-set moiety system, including the institution of the two Wč?tč families [III.E.10], other moieties, groups, societies, and ceremonial memberships are quite inconspicuous and not easily recognized by the outsider. These less apparent socioceremonial units are the Regeneration (“rainy”) season moieties, plaza group moieties, men’s society moieties, Visiting Chief and Clown “societies,” Wetheads and Dryheads, and the ritual-holding families. These ritual family memberships do not form complementary or oppositional pairings [IV.A.2] in Canela dualistic conceptualization with the same facility as the other socioceremonial units, when matched and paired by old research assistants [Pr.2].

My hypothesized evolutionary sequence of organizing principles and resulting social institutions as perceived in the data of the Canela festival system is as follows. The Regeneration moiety system seems consistent with an “early” period during which naming for both sexes was important. A “riverine” period followed, including relocation and village settlements near the banks of a river system with marshes and lakes, allowing the development of more permanence in living styles and the evolution of limited matriliny in the form of ceremonially owned family rights to perform festival acts: the haakhat. Then, in defensive response to the threats of pioneer fronts and increased intertribal warfare, the age-set moiety system evolved with its “bellicose period” of tribal membership matrilines and patrilines, intertribal Visiting Chiefs, and wetheaded status. Finally, the “post-pacification” period includes the installment in the festival system of a special peace-maintaining ceremony (a remembered event) and maybe the piecing together of the Opening Wč?tč festival, as well as the Festival of Oranges, to involve the interests and concerns of women more completely, now that warfare is lost.


The political system is comprised of three parts (the chieftainship, the council of elders, and the judicial system), which form a relatively balanced political system. The chief is surprisingly powerful, considering the Amazonian traditions for chiefs (Kracke, 1978:2). He controls most tribal matters except where ceremonial questions are concerned, but his actions are checked by the council of elders. Members of one age-set of the council (the Pró-khămmă) manage all ceremo­nial matters, but when they fail to carry out their responsibilities, the chief will assume them. Similarly, the council members manage certain political matters for weak or new chiefs.  

[III.D.1]  The Chieftainship 

The Canela had one head chief (pa-?hi: [our{inclusive dual}]-bone: my and your [within the group being spoken to] bone: our collective strength), Kaarŕ?khre (Figure 18), who assumed the chieftainship in 1951. The Apanyekra, in contrast, had three chiefs in the 1960s, each for different purposes: for internal relations (Khenyawęn, age 59 in 1970), for external relations (Teynő, age 65 in 1970), and for festival activities (Kupaakhŕ, age 75 in 1970). More recently, however, the Apanyekra have had just one chief, Hŕ?hŕt, age 40 in 1975. During Nimuendajú’s (1946:161–162) stay (1929–1936), the Canela had three chiefs.

Old Teynő, who was considered the Apanyekra first chief by the Indian service, traveled frequently to Barra do Corda bringing back instructions and a small payment for himself. (No Apanyekra received salaries during my time (1957–1979), although six Canela received Indian service salaries [II.B.4]). The Apanyekra, however, considered old Kupaakhŕ as their number-one chief. When he was absent, Khen-yawęn (hill whale-back-like: small whale-back hill in the savanna), the very strong second chief, governed the tribe and ran the meetings of the council of elders. He was not Kupaakhŕ’s assistant in the sense that he was appointed by him. He was a chief in his own right, being capable and independent, but was handicapped because he was a Kenkateye descendant [II.B.1.d.(2)] rather than an Apanyekra. He nevertheless cooperated fully with Kupaakhŕ. Apanyekra research assistants cited this handicap pejoratively several times, but I did not receive a satisfactory explanation concerning why being born a Kenkateye placed Kenyawęn at such a disadvantage. I know he had one uterine brother among the Apanyekra but no female kin; yet he was married to my adoptive sister, Pootsen, who had numerous kin throughout the tribe.

The Apanyekra were very well governed by these three chiefs in the 1960s. There was less drunkenness and far more respect for and compliance with the tribal leadership than among the Canela. With the death of Kupaakhŕ of the Apanyekra in 1972, however, and the arrival of Sr. Sebastiăo [II.B.2.i.(4)] among the Canela, the situation reversed itself, with the Canela holding more respect for their leaders than the Apanyekra.

Traditionally a Canela or Apanyekra first chief has no insignia of rank and is addressed in no special way. Nor does he carry or sit on a special instrument of office. He is addressed and referred to through the regular relationship terms (consan­guineal and affinal, Formal Friend and Informal Friend, etc.) just like any other person. He maintains his position and authority by preserving the peace, resolving problems, acting in a chiefly manner (being authoritative, haranguing followers, maintaining a “pecking order”), and by keeping many of the available initiatives for himself. Rather than facing competition directly, he is more likely to remove himself from the competition and then assert his authority in some politically viable way.

The chief is the final arbiter of interfamilial hearings. Most of these judicial matters do not reach his level, because they are usually resolved at hearings between extended families. If no resolution can be found among the various uncles of the two families, however, the case is brought before the chief in a special hearing and his word is final. He decides the case and sets the payment that must be made by the family of the injurer to the family of the injured person. The system of payments is well developed. The chief’s power extends potentially into each individual family, including the families of the members of the council of elders.


The chieftainship among the Canela appears to have been surprisingly strong in the past when compared to other lowland South American tribes. In most tropical forest tribes the chief rules by consensus over several disparate groups of a few hundred people at the most. Evidence now exists that Eastern Timbira tribes ranged in population size between 1000 and 1500 and, therefore, may have needed stronger leadership.

While the leadership of the Canela’s Chief Kaarŕ?khre was not very authoritative during most of my 22-year period, it became quite strong and effective in 1979 under the influence of a capable Indian service delegado in Barra do Corda [II.B.3.e]. He coached Chief Kaarŕ?khre alone in his office in Barra do Corda for great lengths of time. The ease with which the Canela adjusted to this strength was impressive. Their greater morale under stronger leadership made me think that it must have been the norm in earlier times. The sudden ascendancy and considerable political strength of Khęę-khwčy as the prophetess of the 1963 messianic movement [II.B.2.f] also suggest the easy acceptance of rigorous leadership. Nimuendajú (1946:93) states that age-set leaders in the 1930s issued precise orders and implies that they were obeyed. (I can believe this in the context of the Pepyę festival group training I saw in 1957.)

It is they [class leaders] who actually govern the age classes, being possibly the only functionaries who literally issue orders among the Canela, a task for which they are trained from the beginning. Only they have the right to summon their class fellows, who are obliged to obey the call and may not assemble without their leaders. Anyone who has dealings with a class, including the chiefs, must turn to its mamkhyę?ti. These leaders are subject only to the council.
If age-set leaders behaved in this manner in the 1930s while in training, I believe that tribal chiefs may have behaved at least somewhat more authoritatively while leading their tribes in earlier times. Turning to symbols for what they are worth (suggestions, not proof), the Canela stress obedience (ear plugs) (Plates 24, 25) rather than haranguing (large lip plugs), and the reciprocal of strict obedience is the issuing of strong orders.


A number of Canela words cover broad ranges of meanings. One of these suggests aboriginal attitudes toward authority. The term hũũ pal pey (it heard/etc. well) means she or he has heard-understood-learned-obeyed-performed something. The implication may be that when a chief gives an order, his followers no sooner hear what he has said than they automatically carry it out. Semantic assumptions of this sort are dangerous to make, however, so I do not wish to stress the point. However, the terms are significant as applied to both the Canela and the Apanyekra in that they are consistent with relatively strong authority reciprocated by relatively strict obedience. Research assistants were emphatic about the instant connection between hearing and obeying implied in this expression. Even doubting the efficacy of an order is considered relative disobedience.


Terms for following orders appear very frequently in the Canela language. Anyone can give an order, and almost any activity seems to have been properly authorized if someone, even a person of little significance, has given an order to do it. Most orders, however, are given by leaders. A general expression is halkhwa-?khôt (word-following: following an order), and specific ones are “pa?hi khôt (chief following), “Pa?päm khôt (God following), “Krôôtô khôt” (Krôôtô[a person] following), and amyi-ŕ khôt (self-superlative following: acting under one’s own responsibility). Canela thinking is so full of such authoritative orientations that acting under some sort of order must have been very important in the past to any individual [III.B.1.k]. Acting under one’s own responsibility is possible but considered potentially dangerous, because it is associated with stinginess, self-centeredness, and antisocial shamans [IV.C.1.h].


The older Kaapęltůk (Figure 50)(the first chief of Baixăo Pręto, 1957-1963), liked to see things run with dispatch and precision. If he ordered a youth to go and fetch a gourd rattle and deliver it to the house of a certain sing-dance master to summon him to sing for the group, he liked to see the youth begin instantly and keep up the pace until his mission was accomplished. He then wanted to see the sing-dance leader come out to the plaza without delay. The older Kaapęltůk expected constant activity [III.B.1.c.(4)] on the part of the youths. They were never to be found sitting around and thinking about things (ay-khăm pa: self-in listening). A traditional leader never allowed such introspection, the older Kaapęltůk maintained.

The older Kaapęltůk expected that in the mid-afternoon when men dressed in their Wč?tč houses to go out to race with logs [II.E.6.b.(2)], they should decorate their bodies with fresh green palm leaf strips just above their ankles and wrists, just below their knees, and around their waists. They should have a headband of similar material, neatly fashioned with some traditional front piece above the forehead. During this period of preparation, the men should not be just sitting and thinking; they should be swiftly and effectively preparing themselves and painting their bodies with charcoal if appropriate. Then when they left the house and village for the log racing site, they should move with dispatch and certainty.

During this period some leader should be urging the men (më-hŕŕpôl: them urge-on: encourage them) to pay attention, not to fall asleep, not to daydream, but to remain focused on their activities. This attitude reminded me of my military officer-candidate training. Canela age-set leaders were sup­posed to lead by example, by urging their followers on in a certain tone of voice, and by telling them repeatedly what they ideally should be doing. The older Kaapęl told me several times in the late 1950s that earlier his people were like this, and that even in his youth (the time of Nimuendajú) the young men behaved with greater dispatch [III.B.1.f.(1)]. This is the way he himself behaved [I.G.2], but few other Canela demonstrate this sort of constant presence, awareness of others, and sharp focus of attention. This kind of behavior is consistent with more authoritative leadership than the Canela or Apanyekra have today.


The tribal chief is chosen partly for his anticipated ability to deal effectively with outsiders. In much earlier times, the principal person governing the tribe was a great warrior in his prime (hŕŕprăl) [IV.C.1.d.(1).(c)], but around 1835 a new kind of chief was imposed upon the Canela, a man chosen and supported by local backland Brazilian authorities [II.B.1.b.(2)]. This change marked the final step in the historical tribal transformation from warlike independence to peaceful dependency [IV.C.1.d.(1).(c)]. Backland community leaders would choose an outstanding tribal leader who spoke Portuguese fairly well and who could convince the tribe to comply with backlanders’ wishes. Research assistants said that in earlier times the council of elders tended to support the leader chosen by the backland authorities. These days the Indian service tends to encourage the council of elders to choose whom they want.


The parameters of the roles of a chief expand and contract depending on his abilities. While the roles of the elders, and especially those of the Pró-khămmă age-set, are more prescribed by tradition, the roles of the first chief are more variable. He is expected to meet all novel situations and to improvise. Thus, the extensions and limitations of the chieftainship will vary from era to era.

[III.D.1.c.(1)] Head Of Council Of Elders

The chief governs the council of elders (Glossary) during the daily evening and morning meetings. His leadership is more obvious in the morning with the planning (Figure 18) and the setting of daily work assignments than with the determination of the more general and often ceremonial topics discussed at evening council meetings.


Before such a meeting begins, the assembled members and adherents await the chief’s arrival as he slowly walks along the radial pathway from his house to the plaza. The elders sit in the center of the plaza with most of their membership facing downhill to the chief; that is, with few elders behind him except during unusually large meetings. When the chief is comfortably seated facing uphill toward the elders in the center of the plaza, he opens the meeting with significant tones, setting the formality of the meeting.The chief usually avoids becoming involved in the debate, but he provides crucial topics and enters the discussion near its end, having heard most of the significant points of view. Then he makes his comments and gives his opinion or decision with dignity and finality. To bring about conformity with traditions, he lectures extensively [IV.A.3.f] (Figure 18). (For more descriptive information, see [II.E.5.b,8].)


The Canela and Apanyekra tend to follow the agreements made in the presence of the chief in the plaza; however, such understandings are not binding. There are no agents to enforce the chief’s orders [III.A.3.c.(3).(g),(h)]. There is, however, a special and sacred quality about decisions made in the center of the plaza at a formal meeting, which commands respect and adherence [III.B.2]. Above all else, anger or aggression [III.A.2.k] must never be expressed in this central spot. Everybody should conduct himself with calmness and impartiality. I have rarely heard anger expressed in the center of the plaza; the preferred alternative would be to walk away from the plaza, back to one’s own house [III.B.1.h.(1)].


When the chief is not in the village or when he has not yet joined the council of elders in the plaza in the late afternoon, one of the deputy chiefs (appointed by the first chief or the Indian service) or a senior member of the council may preside.

Members of the Upper age-set moiety, whose age-sets are just older and just younger than the Pró-khămmă age-set, are not likely to take such an initiative. Among the Apanyekra the traditional dominance of one particular age-set in the council and in ceremonial life does not occur, so that such an initiative may occur.

In both tribes this leadership role of the chief or his deputy is carried out with great dignity. Archaic words are spoken in a formal tone which is heard as a very distinctive style. Today the young are not learning this ceremonial language.

Before the formal meeting begins, the Pró-khămmă and those old enough to sit with them chat in a casual way, telling jokes or talking about the latest hunting success. When a prestigious Pró-khămmă member, one of the deputy chiefs, some other elder, or the chief begins talking in the formal manner, the rest become quiet and listen respectfully. (This formal manner of speech is also used at interfamilial judicial hearings [III.D.3.a,b].)


The chief plans the events of the day with the help of the council of elders and the age-set leaders in the morning meeting. This planning often includes age-set moiety hunting in two different locations, harvesting on two separate family farms, or working on two parts of an access road or on keeping open the tribal boundary vistas through the cerrado [II.B.3.f]. It may involve the decision for each man to go separately to his own family farm with his family members, or to work with them in their village houses all day, as well as several other possibilities for the tribe as a whole. For instance, they might agree to visit the houses of backland Brazilians to earn money by working for them for several days [II.B.3.j.(1)] [II.C.3.g]. Furthermore, if they are about to terminate a festival, they may decide that the tribe will disband for two or three weeks to hunt and obtain the items needed in the final festival acts [IV.A.3.b.(2)].

The chief also appoints certain women without children to accompany the work groups [II.E.5.f, 6.a] [III.F.4.b.(2)] and young males to run as messengers when necessary. Through the messenger the tribe can request a backland rancher to come to the village with a certain number of cattle to sell them for a festival. Even though such messenger services are carried out on a rotating basis, it is becoming increasingly difficult to find runners who will make the trip for no immediate compensation [III.A.3.c.(3).(h),(j)] [III.A.5.a].

[III.D.1.c.(2)]  Age-Set Moiety Leader

Periodically, the Indian service requires work from Canela men on the service post’s farm. Chief Kaarŕ?khre in Ponto and the older Kaapęltůk as chief in Baixăo Pręto would make arrangements with their own age-set moieties to carry out work of this sort [Ep.3.a]. On group labor days, the Canela traditionally expect either the post personnel or a chief to furnish lunch for each worker. Thus, it is important for moiety leading chiefs to earn Indian service salaries [II.B.4] from which they can buy food for such lunches [Ep.4.b.(2)].

Such work is usually carried out in age-set moiety groups so that there could be two competing work forces. After a day’s work competing with each other to see who can do the most work, the two groups race back to the village carrying logs, continuing the traditional intermoiety competition [II.E.6.b]. Within each moiety the age-sets mix together. In the late 1950s, this arrangement required two leaders from each village. In Ponto, the younger Kaapęltůk led the Lower age-set moiety in opposition to Chief Kaarŕ?khre’s Upper age-set moiety, and in Baixăo Pręto the old Ikhč led his Upper age-set moiety in competition with the older Kaapęltůk’s Lower age-set moiety. Unfortunately, neither Ikhč nor the younger Kaapęltůk (both former Pepyę age-set commandants) were employed by the Indian service, so the source of merendas (light lunches) for their age-set moieties was always problematic. In spite of this lack of support, this system works well to supply labor for the post farms.

In the Canela village of Escalvado since 1971, the age-set moieties and their leaders clear the legal boundaries of the Canela reservation every year. These age-set moiety groups of workers also keep the roads open and in good condition (Maps 3,7).

[III.D.1.c.(3)] Tribal Representative

While the first chief’s governing of the council of elders may be his most socially visible function, his more important role is managing the tribe’s external relations: to backlanders and Indian service personnel. This duty requires daily visits to the service post, frequent trips to Barra do Corda, and possibly even travel to Brasília (formerly to Rio de Janeiro) several times a year. He may make reports for or against the local Indian service agents and school teachers, or request tribally desired goods or services. Chief Kaarŕ?khre has been quite successful in such missions [Ep.4.a]. For instance, in 1984 he obtained a truck from the federal Indian service personnel for his people even though he was no longer formally the chief of the tribe. Originally, the first chief was chosen by the elders and the Indian service, primarily for his ability to act as the tribal representative to various Indian service personnel [Ep.3.a] and other outside political authorities [II.B.1.b]. In the 1980s, the responsibility for selecting the first chief has shifted to the Pró-khămmă.

The chief manages relations with neighboring backland Brazilians (Plate 72), as well as with the Indian service. Backland merchants often visit the village to sell their products (Nimuendajú, 1946:216). While the chief cannot be concerned with each visitor, he does arbitrate problems that arise from this intermittent commerce. If a Canela causes problems in the backland area (Map 3), the chief deals with backland community representatives who come to the Canela village seeking retribution; or he may go to the backland community himself to settle the issue [V.B.3].


The Canela and Apanyekra distinguish among three kinds of outsiders that are not personnel of the Indian service. The first group consists of the backland farmers and ranchers for whom the Canela have contempt because of their general stinginess, their “small” personality traits, and their lack of gaiety and activity [III.B.1.a.(1),c.(4)]. Many of the backland families live alone by their cultivated fields and mix with others of their kind only during certain saint’s day festivals [II.d] [II.A.3.d.(1)]. Thus, they are often sad, quiet, and gauche in social situations. The Canela used to think these people were simply “mean” and “bad” (hőőtsč). When such Brazilians visit the village, the chief has little to do with them unless he needs to buy something. The chief does, however, expect the backlanders to stay near his house, occupying some part of it, or a lean-to near by. If such a visitor is well known, he might stay with a Canela family with whom he has a special relationship, such as being a compadre. In the mid and late 1970s, such visitors stayed overnight only in a backland-style house near the post (Map 4, F).

The second group consists of Brazilian visitors from the small city of Barra do Corda, 60 kilometers to the north. These people should, alter reporting their presence at the Indian service post, go straight to the chief to explain the purpose of their trip. The chief wants to know what is going on in his village; this awareness and sense of responsibility is part of his role. In the 1950s-1960s, the Barra do Corda town dwellers, however, felt they were superior to the Indians, so they usually did not pay their respects to the chief and were likely to burst into any village house at will, treating the Canela as if they were not people [I.A.1]. This sort of behavior by some Barra do Corda dwellers was particularly characteristic during the Canela’s stay in the dry forest village of Sardinha between 1963 and 1968 [II.B.2.g], when they lived only slightly more than an hour’s truck drive from Barra do Corda (Map 3).

The third group consists of visitors who come with permissions issued at the federal level, or as tourists who are guests of the Indian service or the SIL. These people come from Brasília, the big coastal cities, or even from other nations. They must see the chief to arrange for cooperation from the Indians. The chief has to accept them and treat them well because of their federal permissions. He has little choice with respect to their presence, because he is under orders from the immediate post agent, who is supported by the município, state, and federal service personnel [I.A.1]. Such visitors are, however, held in far higher esteem by the Canela and Apanyekra than individuals in the first two categories. People from the big cities who have top level permissions tend to treat the Indians with due respect and appreciation. One of the chief’s principal roles is to gain benefits for his people from outsiders. A chief’s demands, however, are limited by the fact that the chief knows the outsider may eventually report all events to the Indian service president. The chief wants to remain in the good graces of his president, because he knows he is the greatest potential source of benefits for his people [Ep.4.a].


A fourth category consists of Indian visitors from other tribes, such as the Krahó, Krĩkatí, and Pukobyé. The Apanyekra and Canela are familiar with each other (cf. Nimuendajú, 1946:155–156), and therefore go immediately to the house of particular relatives or friends, without seeking permission from the chief. However, a representative of the visiting group would be summoned by the chief to appear before the council of elders during a morning or evening meeting to explain their presence and its duration. Members of other Gę-speaking tribes are likely to go straight to the chief and would probably stay with him unless they had worked out a “kin” relationship with some family in the village.


The Apanyekra frequently receive other Eastern Timbira tribal members, maybe several times a year, whereas the Canela seldom do. This is partly because the Apanyekra are further west and closer geographically to the other tribes (Map 6). On the other hand, the Canela were hostile to the Apanyekra and their allies [IV.C.1.d.(1)], even as recently as the time of Nimuendajú (1946:155-156), and being a far larger group, had an attitude of self-sufficiency and superiority. While I was in Ponto in 1960, a group of Krahó passed from east to west along the Alpercatas River only 24 kilometers to our south without visiting.

[III.D.1.c.(4)] Chief Justice

The first chief is also the Canela’s “chief justice.” Problems involving consanguineal extended families, if not resolved at the interfamilial level [III.D.3.a,b], come before the chief, usually on Sundays when nobody is working. It usually takes several Sunday morning meetings before a settlement is reached at such hearings (Glossary) (audięncias: më aypën pa: they in-relation-to-each-other listen).

The chief makes his house large enough to hold such hearings (Map 4, J). Kin on both sides of the question and other interested people gather in a large room of his house or in a shed nearby―many more than at an interfamilial hearing. Kin and onlookers sit in the shade on mats, while the chief may recline in his hammock or sit on a modern chair. He calls on the accuser and the accused to speak and make their points, and then summons witnesses. The “uncles” of the parties are not as evident as at the interfamilial hearings. Although he directs the order of the proceedings, the chief says very little but listens carefully. When the chief renders his decision, it may involve a relatively large payment of shotguns, cast iron pots, or machetes from the family of the accused to that of the plaintiff. Sometimes a horse might be involved, but money is rarely a form of restitution. The principles involved are the same as for the interfamilial level [III.D.3.d], but the problem has to be a big one to arrive at the Chief’s house for settlement.

These payments are almost always made as ordered, even though the chief cannot force anyone to comply. If some family failed to make such payments, they and the individual concerned would suffer adverse public opinion and would be less favored by members of the opposite sex [III.A.3.c.(3)]. Moreover, the next time the chief had advantages to distribute, he most likely would not include the nonconformist among the recipients [III.D.1.i.(4)].


Some chiefs have been traditional shamans (Glossary) or have become involved in the use of backlander “magic.” When this is the case, greater compliance with the chief’s orders occurs for fear that he might use his special powers to enforce his pronouncements [III.A.3.c.(3).(h)]. Chief Hŕktookot of the Canela, who died in 1951, was said to have been an excellent curer, or traditional shaman (Nimuendajú, 1946:237), as was one of the several chiefs during my time among the Canela and Apanyekra. A former chief of the Apanyekra, Kupaakhŕ, who also was a shaman (now deceased), used his power for personal advantage by making nightly rounds of the village, expecting to be fed in most of the houses. No one dared refuse him or complain about this breach of tradition [IV.D.1.e.(2)]. Research assistants said that he considered this food to be payment for his services rendered as chief, but people were feeding him out of the mild fear that he might exercise his generally social powers antisocially.

From studying Dole’s (1973) presentation of chiefs and shamans among the Kuikúru, I can state that Canela and Apanyekra chiefs are clearly more prominent and dominate their tribes far more completely than their Kuikúru counterparts do. On the other hand, Canela and Apanyekra shamans are scarcely political figures at all in the Kuikúru sense [IV.D.1.b]. In my time, none of the significant Canela and Apanyekra shamans had political stature, and the Canela and Apanyekra political leaders did not govern mainly through their shamanic abilities, if they had such powers at all. The politically successful younger Kaapęltůk has no such powers [I.G.4]. Nevertheless, certain chiefs, whose shamanic powers were only moderate, did benefit considerably from these abilities because of the mild fear that they might use them.


A chief has the support of tradition to move into any power vacuum, should he be so inclined. He should have the ability to take advantage of this tradition if he is going to be a successful, strong chief. Thus, the chief, rather than the Pró-khămmă, is traditionally supposed to take the initiative when a situation calls for it, research assistants say. I have observed this initiative taken by both parties.

For instance in the 1970s, as the Pró-khămmă members were becoming quite old near the end of their 20-year period of tenure, the younger first chief occasionally usurped certain of their duties concerning the governing of festivals. In 1978-1979, Chief Kaarŕ?khre sometimes moved around the plaza, actively telling festival performers what to do next in the Khęętúwayę and Pepkahŕk festivals, much to the annoyance of the older Krôôtô, a Pró-khămmă, who should have been carrying out this role. However, these Pró-khămmă members were weak, having lost much of their personal strength, and so could not offer much resistance.

On the other hand, my 1986 communications from the Canela indicate that the young new chief (in his 30s), the youngest Mďďkhrô (Plate 76h), is thoroughly dominated by the strong, new Pró-khămmă in their mid-40s to mid-50s, led by the younger Kaapęltůk [Ep.3.b]. Thus, the strength and activities of the first chief vary with the abilities, attention, and concern of the Pró-khămmă.


There have always been several Canela and Apanyekra chiefs. Sometimes no one of them is the principal chief, and at other times one emerges as the first chief. In such a case, he may appoint deputy chiefs, or the Indian service may do this to help him. If no supreme leader emerges, “potential” leaders share the first chief’s power, and sometimes provide different services.

[III.D.1.f.(1)] Deputy Chiefs

Among the Canela, there have usually been one or two, and sometimes three, deputy chiefs. Rőő-re-?hô (63 in 1975) was such a deputy chief (Plate 68d) for the older Kaapęltůk in Baixăo Pręto during the late 1950s, as he was for Chief Kaarŕ?khre in Escalvado during the mid-1970s. Although he was really self-appointed, the two village first chiefs accepted him, because he was not a political threat to them and his abilities were somewhat limited. Rőő-re-?hô [I.G.11] acted in their place when they were absent from the plaza or the village. He had not been a age-set official in the days of his internment festival initiations [IV.A.3.c.(1).(c)]; rather, it was his strong ego and pride that drove him into assuming positions of leadership. Perhaps he felt he had something to live up to because he was the son of the older Mďďkhrô [I.G.3], the great historian of the tribe and his age-set’s file leader. But Rőő-re-?hô’s services were useful to the chiefs he served, as he was willing to do time-consuming tasks that others would have found onerous. He was also the town crier [II.D.3.i.(4)] [II.E.8].

[III.D.1.f.(2)] Self-Appointed Leaders

Neither the older Kaapęltůk nor Chief Kaarŕ?khre appointed any of the several "potential chiefs,” as I call them, to be their deputy chief of the tribe. These potential chiefs are individuals who have led factions during schisms [II.B.2.h.(2)], and they have always been former initiation festival officials [IV.A.3.c.(1).(c)]. Possibly, the first chiefs made no appointments of this sort for fear the potential chiefs would assume too much authority and usurp the chieftainship. In the mid-1970s this was why Chief Kaarŕ?khre (as leader of the Upper age-set moiety in its 50s) did not appoint the younger Kaapęltůk to be his deputy chief. The latter was the chief’s principal rival because he had been an age-set Pepyę commandant (Glossary), as well as being the unopposed leader of the Lower age-set moiety in its 40s [Ep.3.a].

[III.D.1.f.(3)] Indian Service Appointments

In 1975 the Indian service agent, Sr. Sebastiăo [II.B.2.i.(4)], appointed two deputy chiefs, one from each age-set moiety: Hŕktookot (age 45), from the Upper one, and Yőőkhęn (Plate 73b; age 38) from the Lower one. He appointed them to help Chief Kaarŕ?khre and to lead their moieties as work force managers. These two deputies, however, could not maintain authority over their people. Sebastiăo later replaced Yőőkhęn with the younger Tep-hot (age 36) [I.G.1] (Plate 70g,d), who was more able to take the initiative.


Schisms take place when potential leaders feel they want to be independent leaders themselves. When a chance appears for successfully taking a group composed of their kin, certain affines, and a number of friends to a different village location, they do so. Dissension rarely occurs within a village [III.B.1.h.(1)].

In 1955 the Canela tribe was still divided over the question of leadership succession after the old Chief Hŕktookot died in 1951 [II.B.2.c]. The older Kaapęltůk (Figure 50), who had been the Lower moiety age-set commandant at the time of his final Pepyę internment in 1933 (Nimuendajú, 1946: 182), and Ikhč, commandant of the next oldest Upper moiety age-set, both wanted to be chief. Though both were “potential chiefs,” neither was quick enough to head off the move made by young Kaarŕ?khre (Figure 18) [Ep.3.a].

[III.D.1.g.(1)] Succession To Chieftainship, 1951-1957

Traditionally, a first chief of the Canela governs the tribe until he dies (Nimuendajú, 1946:162), and then a subtle and quiet jockeying for position among the various candidates who would like to succeed him occurs. Those who have been deputy chiefs or recognized rivals to the late chief obviously have the advantage over others. Thus, in 1951 when Chief Hŕktookot (Doroteo) died, a number of possible successors existed. Kaarŕ?khre, who was a young deputy to the deceased chief, went to the Indian service headquarters in Rio de Janeiro and obtained a formal paper of instructions, which he claimed was his patente (commission). The tribe accepted his credential, and he was accepted as the new chief by the tribal council of elders as well as by the Indian service. (Olímpio Cruz (Figure 7) [II.B.2.b.(1)] told me that Kaarŕ?khre’s formal piece of paper was not a patente, but no Canela could read well enough to tell, so “what did it matter.”) Subsequently, Kaarŕ?khre acted as the succeeding chief of the tribe through the 1970s, even though he had considerable nonconfrontational, quiet competition from several of his rivals, mostly from the two Kaapęltůks.


Following the death of Chief Hŕktookot, a number of Canela farms were started and farm huts were built in the Baixăo Pręto area because its gallery forest soils were more fertile than the depleted stream edge soils of the village of Ponto (Map 7), research assistants said. In 1955, the owners of some of these farm structures, instead of living at Ponto and in these farm huts, formed a small traditional circular village of houses near their farms and largely moved to live there.

At some time during the following two years, Ikhč became the chief of this new village in the Baixăo Pręto area. As the village’s chief, it was he who adopted me into his household upon my first visit to Baixăo Pręto in August 1957. A month or two later, however, the Indian service put the older Kaapęltůk in charge as first chief. They replaced Ikhč and officially recognized the new village by sending an Indian service agent, Sr. Alcibiades Costa Resplandes, with his family from the backland community of Jenipapo do Resplandes (Map 3), to build a post house (Plate 11b) and live there. Thus a struggle for the loyalties of all tribesmen began to take place between the two first chiefs, each in his own village about 6 kilometers apart: about 265 persons in Ponto and 145 in Baixăo Pręto.


In my judgment, the older Kaapęltůk (Plate 70b) was somewhat restrictive and severe so that he was not as well liked as the more easy-going Kaarŕ?khre (Plate 75c). For this reason, some families drifted from Baixăo Pręto back to Ponto in the late 1950s and early 1960s. At the time of the tribe’s exile to Sardinha (Map 3) in 1963, the older Kaapęltůk still managed to retain a few families to live with him in Baixăo dos Peixes (Map 3) instead of Sardinha. Thus, the schism was continued for the duration of the 5-year stay in the dry forests [II.B.2.g] and was terminated by the founding of the present village in the Escalvado area in 1969.

The period of the schism was notable for the intensity of the negative rumors (tswa ?nă: sharpened/biting state) [III.A.3.c.(3).(e)] which spread between the two villages, vilifying both chiefs and certain of their adherents. In the 1970s, the Canela referred to this unpleasant period as the time of the "bad talk” (halkhwa ?kęn: speech bad) and gave this misery as the reason why no further schisms should ever take place. The elders did not meet daily to terminate such talk, they said.


The splitting of this tribe in 1955 into the villages of Baixăo Pręto (Map 3) and the new and smaller Ponto (initiated in 1955 and established in 1957) was a direct result of competition for tribal leadership [III.D.1.g.(1).(a)]. No Canela in the late 1950s, however, could verbalize this cause. I was given various other reasons for the divisive founding and maintaining of Baixăo Pręto, but never succession to the chieftainship or the leadership ambitions of individuals [III.B.1.g.(1)]. Presumably, such motivation was unspeakable if not unthinkable for my research assistants, including the older Mďďkhrô, the sophisti­cated age-set leader and cultural “librarian” [I.G.3] in his early 80s.

This apparent inability for overt expression of competition reminds me of my work with the “broad” term “listen-hear-understand-know-respond-obey-perform” (-pal-pal) [III.D.1.a.(1)] [V.B]. To doubt a chief’s order, even without voicing the “challenge,” was equivalent to disobeying it, though individuals nevertheless did do this in certain circumstances, research assistants admitted. Thus, they could not easily say openly that Ikhč or the older Kaapęltůk were acting against Chief Kaarŕ?khre and the existing social order because such behavior would constitute significant psychological confusion and emotional betrayal of the highest ultimate Canela value: maintaining the peace [III.B.1.h.(1)] [III.D.3.e.(1)].

Thus, according to my assessment, three factors contributed to the schism: (1) succession to the chieftainship; (2) ambitions with respect to personal power on the part of former age-set initiation festival officials; and (3) the strengthening of the schism by the formal recognition of Baixăo Pręto as a village, not just a farm settlement, by the Indian service. A fourth factor, according to the research assistants, was (4) the attraction of the far better soil in the Baixăo Pręto area. Gross’s analysis of village movements of the Canela and three other tribes in relation to their available resources should be considered in the context of these factors contributing to schisms (Gross, 1983:436–439).

[III.D.1.g.(2)] Separatist Movements, 1963-1968

In 1964, not long after the 1963 Canela exile to Sardinha about 45 kilometers away (Map 3) following the attack on their messianic movement [II.B.2.f.(3)], some Canela families began filtering back from Sardinha into a relatively hidden region (Campestre) (Map 3) in the cerrado homelands. By 1966 other groups had returned to the cerrado, and by 1968 four well-established settlements existed in different parts of the cerrado, each with its own chief: Ikhč in the Campestre, the older Kaapęltůk in the Baixăo Pręto area, the older Krôôtô near old Ponto, and Chief Kaarŕ?khre at Escalvado. All of these potential chiefs had actually tried to form separate villages before, some of them in the same areas. The older Krôôtô (Plate 77d) had not been an age-set commandant (Glossary) or an age-set file leader, but he had been a messenger boy (Glossary), because he was one of the youngest when his age-set was formed. The position of messenger boy is one of significance and recognition, if not leadership [IV.A.3.c.(1).(c)]. There would have been still another separate settlement if the younger Kaapęltůk had succeeded in staying in Sardinha on the Guajajara Indian reservation with a large agriculturally oriented group of adherents [II.B.2.g.(5),h.(2)].

[III.D.1.g.(3)] Reunification, 1968

To terminate these separatist movements, Chief Kaarŕ?khre put on a Khęętúwayę festival at Escalvado in 1968, summoning the whole tribe to attend and inviting everybody to participate by placing their daughters and sons in this initiation festival. In the Khęętúwayę, twelve positions for girls as associates of the Plaza groups exist, as well as two positions for associates of the novices.

This is why the Canela came together in Escalvado: they did not want their daughters or sons to miss the chance of being properly initiated: girls winning their belts and boys being introduced into an age-set. After the long four-month period of the festival, most Canela stayed in the Escalvado area and formed a new and much larger village in 1969, reuniting the tribe for the first time since 1955. Chief Kaarŕ?khre had demonstrated his great, behind-the-scenes leadership ability through the initiative he took in putting on the initiation festival in his settlement. Once together, temporarily, just for the festival, the potential leaders and their families forgot their rumored differences, or their unvoiced ambitions, in favor of a better future for their children [III.B.1.i.(2)]. That Kaarŕ?khre’s wife is a member of the longest longhouse [III.E.2.e.(2)] (Figure 24) surely helped end the schism.

The rivalry over the chieftainship, muted though it was, had ended, never to arise again among this set of potential chiefs. Ikhč died in the early 1970s, and the older Krôôtô and the older Kaapęltůk were in their 60s and 70s, respectively, in 1979, too old to compete for the leadership any more.


Pursuit of the tribal leadership among the Canela is subdued, to say the least. No one admits he wants to be the leader or that any one of the quasi-political steps he is taking is being carried out with that purpose in mind [III.B.1.g.(1),k.(3)] (Nimuendajú, 1946:162). Political succession is not a matter of general interest to the Canela. To my knowledge there are no myths or stories on this topic. Canela research assistants say that the chieftainship is passed on to a son, sister’s son, or “sister’s” son of the chief who has just died, always a male. The tribal council and outside regional “authorities” (autoridades) both play crucial roles. Since about 1835, outside regional leaders or the Indian service have played important roles in choosing chiefs, and the tribal council has tended to accept the outside selection [II.B.1.b]. As recently as 1957, the Indian service directly installed the older Kaapęltůk as first chief of Baixăo Pręto, and the Canela simply accepted this as part of the existing “tradition” and chain of command. The council’s alternative is to choose a second chief for internal purposes, as was done by the Apanyekra during the 1950s and 1960s [III.D.1].

In the late 1970s, Chief Kaarŕ?khre stated several times that his son, Kapręępręk would succeed him, and Kapręępręk did so in about 1981 but soon lost the position [Ep.2]. In 1935, Chief Hŕktookot succeeded Ropkhŕ, his older brother (Nimuendajú, 1946:162) who had died of small pox. Leadership ability is a more significant factor than kinship in helping the Pró-khămmă determine who becomes first chief, though “blood” is always a factor in who can develop a political following [Ep.]. Chief Kaarŕ?khre is not related to Chief Hŕktookot whom he succeeded in 1951. With Hŕktookot’s death the chiefly presence, which enables the development of a small but noticeable degree of elitism, passed from one longhouse to another (from matrilines in house ZZ to those in MM, Figure 24).

The younger Kaapęltůk, my special interpreter-assistant [1.G.4], never admitted (even in our most private and personal discussions) that his ambition was to succeed Kaarŕ?khre as tribal chief. I think that he wanted this more than anything else, but for several reasons he did not attain the position in the 1970s. First, the younger Kaapęltůk was too strong and capable a political rival to Chief Kaarŕ?khre in the 1970s in the same way that his name-giver, the older Kaapęltůk, had been a rival in the 1950s and 1960s. I believe that the younger Kaapęltůk’s ascendancy to chieftainship would have caused another schism. Villagers still remembered the tribal rift of the mid-1950s and 1960s and its damage in terms of harmful rumors between the two villages. Nobody wanted to suffer these painful experiences again.

Chief Kaarŕ?khre gained considerable political strength while the Indian service official (delegado) of 1978-1980 was in power in Barra do Corda [II.B.3.e]. Consequently, the position of the younger Kaapęltůk was subordinated. The delegado had long talks with these two political rivals and put them in their respective political places according to Canela tradition: Kaarŕ?khre as first chief and Kaapel as future head of the Pró-khămmă.

Moreover, the well-built Indian service post buildings, standing next to the village of Escalvado, were permanent and offered too many excellent and needed services (especially medical and educational ones) for potential Canela splinter groups to consider separatism. By the late 1970s, the Canela greatly appreciated the Indian service’s support [II.B.2.i.(2)] (Figure 1; Plates 5b, 11).


While chiefly characteristics vary from individual to individual, there are nevertheless some common denominators. It helps to be well liked, respected, trusted, and remembered for service to the community. A strong first chief maintains a tie of small personal debt (material in nature or from favors rendered) with each individual regardless of moiety or society affiliation. In theory, the chieftainship transcends any socioceremonial division, although kinship (the chief’s and his wife’s), achieved leadership positions, and age-set support facilitate the mainte­nance of the first chief’s authority. Individuals tend to support their kin and the leaders of the groups they have belonged to whom they have liked. Thus, leadership is both inherited and achieved, and is maintained partly through personal ties and through style of personal behavior. A respected chief behaves with a certain degree of self-confidence and command but never too much [III.B.1.g.(1)], and he provides the ultimate peace-keeping services through his governing of the council of elders and his settling of interfamilial problems as chief justice [III.D.1.c.(4)].


In festivals, political leaders follow their troops, as in the Pepyę festival where the commandant marches along side or behind his troop of novices instead of being out in front of them [II.D.3.d] [IV.A.3.c.(2)]. This is one of the training positions from which prospective chiefs emerge: the më-?kapőn-katę (them-sweep-master, the one who sweeps or dusts them off). There is nothing very war-like about this title. The semantics of the expression suggest that he takes care of them and develops them. Thus, a political chief of the tribe (pa?hi)—in contrast to sing-dance masters, ceremonial leaders in general (më hőőpa?hi) [II.D.3.i.(2).(b)], and war leaders—symbolically moves behind the people he is governing, not in front of them. This difference is very important. He must watch over them, look out for them, urge them on, think for them, give orders to them, and finally judge them in many matters.

In contrast, sing-dance masters (më nkrel-katę) [II.F.1.a] either dance facing their group in the plaza, or lead the way for their group as they go around the boulevard. The age-set “class leader” (mamkhye-?ti; Nimuendajú, 1946:93,172), whom I call a “file leader” (Glossary), leads in the Pepyę festival by marching at the head of the long file of Pepyę novices. He is a ceremonial leader [III.C.7.a] rather than a political one, so he must be less direct, having more “shame” [III.A.3.c.(3).(a)].

A war leader (hŕŕprăl) [IV.C.1.d.(1).(c)] is said to have led the way into battle, stalking through the countryside ahead of his file of men, and to have “governed” (governado) his tribe in times of peace. Still earlier in their history but before war became prominent, Canela group leaders are said to have been those who were more able and enthusiastic than other members of the group. Thus, the leader quite naturally gained the greatest esteem of his people but did not really govern them. Research assistants compared the similarity of such a leadership style to that of the foremost coatimundo (wakhőő). This animal moves at the head of his pack, providing no more leadership than his own example.


Nimuendajú (1946:97) thought that the political leader of the tribe was necessarily hŕmren (wetheaded) in status, probably because the principal chief of his time happened to be hŕmren (Glossary). Actually, the political chief of the Canela is most properly not wetheaded at all. This ceremonial rank and human quality is not inherent in the office of the chieftainship. There is a distinction between political leadership (non-hŕmren) and ceremonial leadership (hŕmren). The chief of the tribe has most correctly been the deputy commandant and then the final commandant of his graduating Pepyę festival age-set, and therefore the leader of his age-set for life; but commandants of age-sets are political leaders and not hŕmren in status in their own right.

In contrast, age-set file leaders are inherently hŕmren but nevertheless may become political chiefs of the tribe. These file leaders have been given their positions by appointment by the Pró-khămmă in one of the earlier initiation festival intern­ments. They are thereby hŕmren in status merely from having occupied this position. If an age-set file leader later becomes a politically viable person, he may be taken on by the principal chief as a subordinate chief and helper: a deputy chief. Then, when the principal chief dies, this young hŕmren age-set file leader is an obvious candidate to succeed the old chief. Chief Kaarŕ?khre of Escalvado followed this route to power. He was a Pepyę age-set file leader, who was taken by the old chief Hŕktookot as his deputy and assistant, so that he knew how to run the tribe. When Hŕktookot died in 1951, Kaarŕ?khre knew what to do and consequently succeeded him as first chief. However, a Pepyę file leader competes quietly over a life-time with his Pepyę age-set’s commandant for the eventual leadership of his age-set and maybe the tribe.

As in the case of old Chief Hŕktookot, three male roles were combined into one for Chief Kaarŕ?khre, and all three contributed to his political strength as chief of the tribe. Not only was he the political chief, he was also high in ceremonial prestige due to his being hŕmren. Moreover, people were somewhat afraid of him because he was a weak kay (shaman), but a “good” one, not an antisocial one.  


There can be no cult of personality among the Canela chiefs, deputy chiefs, and aspirants to the chieftainship. They cannot appear to be greater than others [III.B.1.g.(1)]. They wear no special clothing, nor do they carry distinctive staffs or weapons.

However, they are usually employed by the Indian service [II.B.4] and can provide lunches and work for their people on special ocasions III.D.1.c.(2)]. This advantage has not been necessary to being a good chief, although it is becoming so these days [Ep.4.a]. They sometimes build larger houses but so do some of the fathers of the Wč?tč girls [II.D.3.i.(2)]. The size of the house depends partly on the need for community service and partly on the owner’s need for social or political prestige.  


A Canela chief cannot become a dictator because the council of elders would not follow him and the spirit of individuality epitomized in the Clown society could not be suppressed (Plate 46d) [III.B.1.d.(3)]. However, a chief can exert a considerable amount of authority. This power was probably stronger and carried out more directly in earlier times [III.D.1.a]. These days a chief has to exert his control over situations and individuals far more indirectly [Ep.4.b.(2)].

The political practice of using indirect techniques for individual rewards and punishments becomes evident even though the Canela will not talk about their tactics. They are afraid to do so. The effective rumor network rapidly spreads almost anything political anyone says [III.A.3.c.(3).(e)] to everyone else. I was careful not to talk about the political activities of individuals in my research assistant council. When I did ask questions about such matters with individual research assistants, I felt I had to be concerned about others overhearing what we were saying (such as unseen little boys beyond the thatched wall) and about what my research assistant would say to others about my perceived preference for one or another political leader.

Political power and efforts to exert indirect control over others are certainly the most secretly held information among the Canela. I have pondered the question of how much of such behavior (both initiatives and responses) is consciously planned. I suspect that a relatively large proportion of such behavior is unconscious, unplanned, and spontaneous at the interpersonal level but that it is quite well thought out, deliberate, and strongly motivated at the long term societal level.


A leader of one of the smaller uxorilocal Amazonian societies often builds his principal political base of power on his domestic group or cluster [II.D.3.h.(2)] (Kracke, 1978:35–37). He has his sons-in-law in his group because of the society’s uxorilocal structure, and their support is effective depending partly on how many years tradition requires them to stay with him. A leader may be able to keep his married sons in his domestic cluster as well, depending on tradition. Addition­ally, more distantly related, or even nonrelated, adherents may join a strong leader’s domestic faction. Among the Canela, the principle of uxorilocality is strong, and the position of a family’s matrilines on the village circle is almost equally binding [III.E.2.e]. Exceptions to these two principles exist, but individuals give compelling reasons for their family’s variation in every case, which are not arbitrary. Thus, within the Canela village, longhouses [III.E.2.e.(2)] do not form the principal base for building political power, although some power does come from the unity and influence of the leader’s female kin and his wife’s female kin.

Like everything else in the Canela world, the sources of power are divided and many, as Da Matta (1982) points out for the Apinayé in the title of his book. Probably, a leader’s strongest political base, however, comes from his leadership of an age-set (Glossary), whether or not he has survived politically after being its graduating file leader or its graduating commandant [IV.A.3.c.(1).(c)]. Nevertheless, it takes signifi­cant individual leadership ability to become an official of a graduating Pepyę festival age-set [II.D.3.i.(3)], as well as great political competence to survive and eventually become the first chief [III.D.1.f.(2)]. Thus, the leader’s personal qualities and abilities are ultimately very important, but so are his kinship ties and his wife’s kinship ties as well.

Chief Kaarŕ?khre’s natal kin are few (Figure 24, house P), but his lifelong wife is a member of the longest longhouse in the village (Figure 23, houses BB-NN). The older Kaapęltůk’s [I.G.2] natal kin are similarly limited to one house (V), and the longhouse he married into is only moderate in size (M-O). The younger Kaapęltůk’s [1.G.4] natal kin live in the second longest longhouse (TT-ZZ), and he married into the longest one (BB-NN). Thus, on the basis of blood and affinal ties alone, he should be the strongest leader. However, his becoming commandant of his graduating Pepyę age-set, followed by his continuous leadership of the Lower age-set moiety from 1957 onward [Ep.4.b.(1)], provides me with the most compelling reason for his current ascendancy. His name-set transmission ceremonial positions have next to nothing to do with his ascendancy. However, other reasons exist such as his long term advocacy of backland ways, seen in his emphasis on agriculture, folk Catholicism, and commercial trade [Ep.4.b.(2).(e)]. He also has a current program for his people. Furthermore, his personality and ceremonial status, carried over into daily living, contribute significantly to his present status [Ep.4.b.(2).(a)].

Similarities to the characteristic power base resting on domestic clusters mentioned by Kracke (1978:194) do not appear within the Canela village [III.B.1.h.(1)] and its sacred plaza [III.D.1.c.(1).(b)] but do become evident outside the village in its farm communities, which have grown numerous in the 1980s (Map 3). Each community with significant numbers constitutes the base for the source of power of a potential chief [III.D.1.g] or an ex-chief. Note that ex-Chief Kaarŕ?khre resides far to the west in the Two Rivers area while the younger Kaapętůk has the largest community and is putting in the biggest agricultural fields in the newly opened Pak-re area, the furthest one to the east. The older Kaapęltůk is too old to compete.

[III.D.2]  Council of Elders

The council of elders (Glossary) is the group of mature men who meet in the plaza regularly to discuss matters of the tribe. Today, this group consists of two to three age-sets as well as men who attend on their own from older and younger age-sets. In the late 1950s, the membership was narrower, including considerably fewer men and age-sets. In Ponto and Baixăo Pręto of the Canela, not more than 24 and 12 men, respectively, formed the council, while in Aguas Claras (Map 8) of the Apanyekra, less than six very old men met in the center of the plaza to debate matters. The Canela-Apanyekra contrast was notable.


Meetings held by the council of elders, whether in the evening or in the morning, have the same form. Generally the morning meetings [II.E.5.b] (Figure 18) are devoted to planning what is going to be done during the day (e.g., group work or group hunting, or individual farm work and individual hunting), while the evening meetings [II.E.3.a] are left open to debate more serious topics and to plan festival activities.

In both cases, members of the age-sets are likely to have gathered before the meeting in their traditional locations at the edge of the plaza (Nimuendajú, 1946:91) (Figure 24). The Pró-khămmă [III.D.2.b] is the first age-set to start the move toward the center of the plaza. If the chief wants to hurry the process, he might move to the center of the plaza, with the governing Pró-khămmă joining him almost immediately. Then, the members of the other age-sets that also comprise the council of elders, depending on what part of their 20-year cycle (earlier or later) the Pró-khămmă age-set is in, move in to surround them.

This movement into the center of the plaza is not likely to be seen by the observer as a movement of age-sets as units, but rather as individuals drifting in at different times from their age-set’s edge-of-plaza locations (Plate 40f). These movements are likely to be deliberate and hurried in the morning and more casual in the evening.

Once in the center of the plaza, the membership may chat informally for half an hour or more, especially if the chief is not present or if he does not choose to begin a formal meeting right away. Informal Friends might amuse each other by trying to throw sand in each other’s eyes, catching the other person off guard. When there are very few older men in the village, these meetings may commence even more informally. The men may forego meeting in groups on the edge of the plaza and just go to the plaza’s center immediately.


Sometimes the informal banter takes the form of swift but light exchanges, and at other times an individual tells a long story with everybody listening attentively. A favorite topic among men is hunting, and they go into the smallest details about the tracking, stalking, shooting, and finally, finding the wounded game and bringing it home.

A favorite informal topic is the latest difficulty or indignity suffered at the hands of some backlander. In the late 1950s the Canela liked to hear and speak evil of backlanders [II.B.2.e]. This enabled them to feel better about themselves [II.B.2.a]. Stories about backlanders included drunken brawls, killings, wife beatings, child harassment, and stinginess with respect to money and the care of old relatives [III.B.1.a.(1),b.(1)]. In the late 1970s there was less of this kind of talk, probably because there was a much better understanding of the Brazilian world and the interior backland culture [Ep.4.b.(2).(e)]. Moreover, the Canela lands had been demarcated, so encroachment on Canela land by the backlander was no longer to be feared [II.B.3.f] (Map 3).

The informal part of a council meeting is often devoted to investigating rumors: virginities lost, affairs, etc. Such stories come into existence every day and might be called the “spice” of Canela life. We have seen how the fear of negative rumors against oneself is a powerful force toward social conformity [III.A.3.c.(3).(e)]. Rumors may be discussed even in the formal part of the council meeting if they are sufficiently serious; such as the first report of a young man trying to leave his wife and child, or of cattle breaking into a farm, or a Canela being knifed while out in the world. Questions are then asked of relatives of the principals in the rumor and further facts are sought. The investigation usually leads to neutralization of the rumors [III.D.2.d.(1).(a),(b)].


When the chief, or the leading person present, feels it is time to proceed with the business of the meeting, he begins to talk in the old formal style of interfamilial judicial hearings [III.D.3.a] and council meetings. In ancient times, this formal style was almost a different language and was used more extensively in plaza meetings and judicial hearings, but little is remembered these days, except by certain specialists [I.G.5]. At this point, side discussions stop, and full attention is paid to the speaker.

The order in which speakers address the council is significant, because it is an indication of their individual political importance. In the 1970s, after Chief Kaarŕ?khre, Rőő-re-?hô (as deputy chief) often spoke first. When Kaarŕ?khre was absent, the older Kaapęltůk sometimes spoke first although he was seldom present [Ep.3.a], and then Rőő-re-?hô followed. In any sequence of this sort, the older Krôôtô was at least third or fourth, followed by the younger Mďďkhrô, a direct descendant of the late Chief Hŕktookot. These persons are all Pró-khămmă, except for Chief Kaarŕ?khre.

The length of each speech (3 to 10 minutes) is not as significant as its order, but after about the fourth or fifth discourse, the order no longer counts. The order is an informal matter (personally enforced by strength of personality each time) rather than a precise protocol.

After about 30 or 45 minutes, or even 90 minutes for difficult questions, the leader of the meeting begins to summarize the events and may pronounce some conclusions [III.D.1.c.(1).(a)]. Next, the terminal call, which rises slowly for about 4 seconds and then descends sharply for about 2 seconds, is voiced by all in unison. This call, a symbol of agreement, changes in form depending on whether the Wč?tč season is in session or not. In the latter case, the Clowns lead the terminal cry in a humorous, indecorous manner. Finally the town crier [II.D.3.i.(4)] sings out the news of the meeting for all the villagers to hear [II.E.8]. Some council members linger on in the plaza for more informal chatter, but others hurry home to their families. Soon the center of the plaza must be cleared so the evening sing-dance can take place.


The Lower moiety age-set whose members average 45 to 65 years old, depending on the part of their cycle they are in, is always called the Pró-khămmă among the Canela. These men dominate the council of elders for about 20 years.


As described in “Socioceremonial Units” [III.C], young Canela boys are formed into an age-set about every 10 years by being processed through four or five Nkrel-re (Glossary) initiation festival ceremonies over the 10-year span [IV.A.3.c.(1).(a)]. Individuals live as members of this age-set for the rest of their lives, only rarely changing the membership to one of the other moiety’s age-sets. Only the name, such as Kukhoy-khăm (monkey-in) or Pró-khămmă (ashes-in), of an age-set changes [III.C.3.c] (Nimuendajú, 1946:90-91). If one just-graduated age-set belongs to the Lower age-set moiety, then the next one processed about 10 years later belongs to the Upper age-set moiety, and so on [III.C.3.a]. Every Lower moiety age-set eventually becomes the Pró-khămmă when these men are in their middle to late 40s [III.C.3.b]. They tend to control the council of elders right from the start of their entry into this new position and for about 20 years thereafter [III.D.1.g.(1).(a),(b)] [Ep.3.a]. Members of the Upper moiety age-set never become the Pró-khămmă and are never dominant in the council of elders, although they speak up as individuals and sit with the Pró-khămmă in the council of elders.


When a younger Lower age-set, whose members are in their 40s, takes over from an older one, whose members by this time are in their 60s, the remaining members of this older and now retired Lower age-set continue to remain in the council of elders and help their younger moiety colleagues: their mëi yapal-re (our nephews). The reciprocal of this expression is mëi pa-?tum (our arms-experienced: our experienced supporters). (The term i?tum means “it dirtied,” “used,” “old,” “experienced,” and “ex-” in the sense of “former” as in ex-Wč?tč girl: Wč?tč-?tum.)

The council of elders at the time of the entry of a new Pró-khămmă age-set is composed of two Lower moiety age-sets with one Upper moiety age-set between them. To the extent that there are survivors from the oldest Upper moiety age-set in their late 70s or early 80s, members of two Upper moiety age-sets also sit in the council of elders. Ten years later, the situation is reversed: a new Upper moiety age-set enters the council of elders and the older Lower moiety age-set has died or become ineffective through age.

Considering the newness of their membership and their junior status, the younger Upper moiety members say very little at council meetings, but 10 years later when they are in the central position between two Lower age-sets, they may speak out frequently as individuals, though they are not in a dominating position because they are never the Pró-khămmă. These in-between Upper moiety members enter the discussions fully, especially when the matter in question has something to do with one of their relatives [III.D.2.d.(1)].


When I first arrived in 1957, the Pró-khămmă members sitting in the plaza in the Ponto village were the age-set of the older Mďďkhrô (late 50s to mid 60s). (See Nimuendajú, 1946:91 “rópkama (1913).”) To me, they were a truly noble group [III.B.1.f.(4)], walking to the plaza with great dignity and reigning with impressive equanimity and calm [II.B.2.g.(8).(a)]. They seemed to possess a high degree of impartiality and good will [I.G.2].

When I returned in 1963 just alter the messianic movement to find the Canela in the dry forest village of Sardinha, the old Pró-khămmă had given way to the new Pró-khămmă of the age and age-set of the older Kaapęltůk (early 40s to early 50s) (Figure 24). (See Nimuendajú, 1946:91 “pőhůtíkama (1933).”) This new Pró-khămmă seemed to lack the peace and calm of the earlier group but was more familiar with the ways of the backlanders. I sensed more generalized hostility in their personalities. The members of this age-set were the novices in the 1933 graduating Pepyę festival in which the older Kaapęltůk was outstanding for his remarkable performances and leadership abilities (Nimuendajú, 1946:182).

From correspondence, I understand that in the early 1980s the members of the Lower age-set of the younger Kaapęltůk took their turn as the Pró-khămmă. They were then (ca. 1981) in their early 40s to early 50s. This turnover was very much to be expected because during my last visit in 1979 the Pró-khămmă members who had not yet died were already very ineffective [III.D.1.c] (Figure 19).

In fact, the old Pró-khămmă of the late 1970s were so weak that a number of their roles were being assumed by the strengthened chief, Kaarŕ?khre [III.D.1.e]. Under the leader­ship of the younger Kaapęl, however, who can read, write, translate, and speak Portuguese very well, this new Pró-khămmă should be able to introduce a number of important innovations toward closer cooperation with the Indian service and the backlanders. For example, the Canela might possibly undertake some form of intensive agriculture to support their growing population in the face of the diminishing fertility of the local gallery forests of the reservation [Ep.7].


Because of the cyclic patterns in Pró-khămmă leadership, change among the Canela takes place in steps, rather than at an even rate. Change accelerates just after new age-sets of Pró-khămmă have assumed control (approximately: 1923, 1941, 1961, and 1981) (Nimuendajú, 1946:91). It also accelerates just after the death of strong chiefs (e.g., Ropkhŕ, 1935; and Hŕktookot, 1951 [II.B.2.c]).


Many members of the Lower age-set moiety, when they have become a member of the Pró-khămmă, seem to walk with more dignity and self-assurance than do members of the Upper age-set moiety, even if the latter are members of the council of elders. Even though only Lower moiety members can look forward to being Pró-khămmă, as they grow older, I have never heard of Upper moiety members making complaints against Lower moiety members because of this difference.

This asymmetry is a necessary part of the effective operation of a major Canela sociocultural system. It is important for peace and tranquility in the tribe, especially in this moiety system which operates on a daily basis. Tradition gives one moiety permanent and operative ascendancy over the other (cf. Nimuendajú, 1946:79), reducing the competition between the two. The age-set moieties reflect this competitive relationship, however, in their references to each other as mëi katęyę (our opposition-people/masters). (The expression -katęyę is also used for enemy tribes.) An older opposing age-set is referred to as, mëi hŕŕ-yę (our older-brother honorific-plural) and a younger opposing age-set is referred to as, mëi yő?hęw-yę (our younger-brother honorific-plural). This distinction between “older brother” and “younger brother” is a traditional mark of respect and authority, which they still carry out in address these days but not in behavior. It is notable that in this context older and younger brothers are more socially distant from each other than an uncle is from a nephew Table 11).


The identity of the formal leader of the Pró-khămmă remained a mystery for most of the time during my visits with the Canela. If a formal leader existed at all, he was never evident nor spoken about.

During the late 1950s no one was identified as leader of the Pró-khămmă in Ponto village, and in Baixăo Pręto the older Kaapęltůk so thoroughly exercised his power as village chief that none of the few Pró-khămmă there assumed this position of leadership. In the mid-1960s, the older Kaapęltůk became a Pró-khămmă but had his own village in Baixăo dos Peixes apart from the principal population of the tribe in Sardinha. As the former final Pepyę festival commandant of his age-set, however, he would have been the leader of the Pró-khămmă in Sardinha if he had been living there.

In the village of Escalvado during the 1970s, the same older Kaapęltůk had rejoined the main body of the Canela tribe together with his few remaining political adherents, kin, and affines. Sometimes he led the evening council meetings but only when Chief Kaarŕ?khre was absent. His intermittent ascendancy among the Pró-khămmă seemed natural enough, for he had the personality of a chiefly individual who would always assume responsibility whenever he could. However, no one identified him as the leader of the Pró-khămmă.

From my 1984 communications, the younger Kaapęltůk sees himself as the leader of the new Pró-khămmă. This ascendancy is the outcome of his original role as final commandant of his Pepyę age-set at the time of their graduating festival in 1951. He refers to himself as the më-?kapőn-katę (“commandant” in Nimuendajú, 1946:182) of the new and current Pró-khămmă, which is the same expression used for the person who is the commandant of the novices.

Thus, it appears that the Pró-khămmă have a formally recognized leader to the extent that the earlier Pepyę commandant-leader survives and is able to maintain his personal and political leadership. While the older Kaapęltůk did not succeed in retaining his leadership, the younger Kaapęltůk has done well so far. This flexibility and subtlety of the Canela system in providing, withdrawing, but retaining crucial leadership roles is typical, and is one of the most important factors contributing to their ability to adjust and survive.

Research assistants referred to the age-set of the Pró-khămmă in power in the late 1950s as the older Mĩĩkhrô’s age-set, and they said that Mĩĩkhrô was the file leader [III.D.1.i.(1)] of his graduating Pepyę age-set. Whether this would make him the “commandant” of the Pró-khămmă later on if its final Pepyę “commandant” did not survive politically or otherwise, I do not know. In any case, there appears to be at least two obvious spring-boards for reaching the political heights, whether the chieftainship or the Pró-khămmă: being assigned either the position of the commandant or that of the file leader of an age-set’s final Pepyę festival performance [IV.A.3.c.(1).(c)].


The Apanyekra do not use the expression “Pró-khămmă” to represent their corresponding Lower moiety age-set, or any age-set. When asked about this, they give the term më-?khŕ (elders), in its place, which the Canela also use in a similar way [II.D.3.j] but not as the name of an age-set. They use it to describe people of age and strength, people who have become strong and resistant because they have survived beyond about 55 years of age (Table 9).

While the Apanyekra have the same expressions for their age-set moieties, namely, Khčy-katęyę (Upper opposing-people) and Hară-?katęyę (Lower opposing-people), they do not have the asymmetry in power relationships between the age-sets in their council of elders. They form this body the same way the Canela do, namely, through Nkrel-re (Khęętúwayę and Pepyę) initiation festivals over a period of approximately 10 years, but their më?khŕ are composed of either moiety, and the distinction between the më?khŕ and the council of elders is not clearly made. Moreover, the Apanyekra performance of these festivals has been quite irregular for at least the last two or three decades, so a direct relationship between graduated age-sets and their më?khŕ may not exist.

It is clear that the Apanyekra have lost much more of their festival system than the Canela, and that they are less concerned about maintaining and practicing it. I have seen two Apanyekra Pepyę festivals (1958 and 1975) (Plates 36c,d, 37b,c) and one of their Khęętúwayę festivals (1966).


Sometimes it is hard to distinguish the roles of the Pró-khămmă from those of the council of elders as a whole. The individuals who comprise the Pró-khămmă age-set sit among the elders in general, and no formal distinction is made between them in the plaza. However, the Pró-khămmă do sit separately as an age-set on the southwestern edge of the plaza (Nimuendajú, 1946:91) (Figure 24) before joining the elders in the center of the plaza for a meeting.

The Pró-khămmă, as a group apart from the elders, are traditionally in charge of (1) coordinating the festival-pageants, (2) choosing the several young people who are to receive the honor awards after a festival, (3) receiving honor meat pies in the plaza, and (4) seeing that some hŕmren person of maturity tastes the first fruits of the season to test them for ripeness. Several other minor roles exist that can be attributed traditionally to the Pró-khămmă as a special age-set among the rest of the age-sets comprising the council of elders.

[III.D.2.c.(1)] Govern Festivals, Rituals, And Other Ceremonies

In contrast to the roles of the chief, it is the principal role of the Pró-khămmă to govern the ceremonial system of the Canela. Most of the ceremonials occur during the great summer Wč?tč season festivals, though they also appear at other times of the year.


For instance, the Pró-khămmă decide when it is the right time to put on the Opening Wč?tč festival. After it is over, the lowest graduated age-set may formally ask the Pró-khămmă if the great festival of the Wč?tč season can be an Nkrel-re initiation festival, or the Clowns may do so if they want to put on the Fish festival. However, the Pró-khămmă take the initiative with the Pepkahŕk festival, surprising the tribe. The Pró-khămmă decide when the Corn ritual is to commence and summon the correct family member of the Corn haakhat to the plaza to ask that this person proceed to undertake the ritual activities. The procedure is similar for the Pŕlrŕ ritual: at the proper time the Pró-khămmă summon the traditional Pŕlrŕ log cutter, the Pŕl-yitep-katę ([Pŕl-style-log]-cut-master), and ask him to go out the next day and cut down the kind of tree that can be used for a Pŕlrŕ ceremonial log. The traditional procedure for initiating any Canela ceremonial is quite precise.


Once the great festival has been selected, the Pró-khămmă choose most of the principal female and male performers, except for the Fish festival [III.C.7.b.(1)]. To make their decision, they leave in the morning for the seclusion of a farm hut or a shed well outside the village near the buildings of the Indian service (Map 5). A relative of a prospective principal actor is usually invited to attend if he would not ordinarily be present, and he passes the invitation on to the chosen performer or performers, who invariably accept.

For haakhat or name-set role owners, or the traditional performers of certain other roles, the Pró-khămmă summon them to the plaza during a late afternoon council meeting to ask them if they will carry out their particular role. It is at this time that the Pró-khămmă find out if succession in a particular line is intact. If it is not, they designate a new line to carry out the role [III.C.8.b].

At evening council meetings during the performance of a festival, certain Pró-khămmă members volunteer or are assigned to oversee and direct various acts of the festival-pageant taking place on the following day to be sure that the performers know just what to do. Performances are not very spontaneous. The performers, especially in group situations, simply do not know what they are to do next and so have to be told in this manner.


The Pepkahŕk festival has been performed about every 10 years (1958, 1970, 1979) and the Masks’ festival even less frequently (1960, 1970 only partially). For these festivals the Pró-khămmă have to discuss very carefully just what went on during remembered, previous performances. They want to carry out the traditions; innovation is avoided if possible. A number of days before each festival, they debate very carefully what should take place on each festival day at evening council meetings. Then each evening during the festival, they discuss exactly what is to be done the next day, and sometimes they even repeat this effort of reconstruction of traditions during the council meeting the following morning.

Except for the annual festivals, no one person among the Pró-khămmă remembers the exact sequence of acts from the beginning to the end of a festival and just how each role is carried out traditionally. At the evening and morning meetings, they have to piece the performance together with contributions coming from many individuals, including the Upper age-set moiety council members. Decisions are sometimes the conse­quence of strong personalities or political favoritism rather than good memories. Much later, during the preparation of the next performance some years later, the good memory might prevail and cause the act to be performed in a manner that is consistent with an earlier performance rather than the last one [III.C.7.a.(2)]. The Canela genius for “getting along” well with each other and their compulsion to arrive at a peaceful consensus (often at the expense of objective truth [III.B.1.b.(3)]) is very apparent at such meetings.

For instance, the younger Kaapęltůk complained in 1988 that the selected Pepkahŕk troop members were mere adolescents instead of mature men. He says on tape that the older Krôôtô claims this is the traditional way but that he, Kaapęl, knows otherwise. Apparently, the older Krôôtô—even though of the older, retired Pró-khămmă—won this point with his forceful personality (which I know so well) over the younger Kaapęltůk, even though the latter is first chief of the tribe and commandant of the sitting Pró-khămmă. Surely, when the older Krôôtô dies, the younger Kaapęltůk will still remember the Pepkahŕk performance of 1970 and will take his tribe back to the earlier “tradition” of mature internees. It is also possible that the younger Kaapęltůk will allow the performance of 1988 to become tradition because it has become so difficult to coerce mature men into going through a Pepkahŕk internment. The Canela are pragmatic about cultural change.

[III.D.2.c.(2) ] Bestow Honor Awards To Youths

At the peak of a great festival’s terminal period during the summer Wč?tč season [IV.A.3.b.(3)], the Pró-khămmă award prize ceremo­nial objects to several of the minor performers [III.A.3.b.(3)]. These treasured items are awarded on the basis of good singing and dancing and are given to both young women and young men [II.D.2.f,3.e].

In the late afternoon just at sunset [II.E.7.a], the Pró-khămmă summon the winners to appear in the plaza and bestow upon them these awards for civic merit [II.G.3.a.(1),(2),(6),(7)]. This is a sight I have always enjoyed watching. The Pró-khămmă receive reflected prestige as they present the awards [II.D.3.k]. The Upper age-set moiety members of the council of elders take no part in this ceremony.

[III.D.2.c.(3)] Accept Honor Meat Pies

One of the great privileges of being a Pró-khămmă is receiving meat pies, or bowls of food, and eating these gifts of honor in their traditional position on the southwestern edge of the plaza (Figure 19). When a person of hŕmren status returns to the Canela social world after having been out of communication with the plaza (away in the outer Brazilian world or away because of sickness, mourning, or couvade) she or he is presented to the Pró-khămmă in the plaza in the late afternoon [II.E.7.c]. The person is adorned with falcon down and urucu paint and is accompanied by female relatives carrying a large meat pie. This meat pie presentation to the Pró-khămmă is called the Hŕŕkwél ceremony.

Younger individuals than the Pró-khămmă do not eat meat pies in the plaza in the late afternoon. If meat pie distributions take place at this time, the pies are brought to the houses for eating, except for the Pró-khămmă who eat their pies in their southwestern plaza position.

[III.D.2.c.(4)] Receive First Crops

From January through April certain traditional crops become ripe and are ready for eating for the first time that year [II.C.2]. These include varieties of sweet potatoes, yams, corn, squash, peanuts, bitter and sweet manioc and others (Nimuendajú, 1946:58). Before anyone eats them, they are supposed to be presented to the Pró-khămmă, who have assembled in the morning in their southwestern spot at the edge of the plaza before the council meeting is held (Figure 24).

It is not necessarily a Pró-khămmă member who tests these first crops, but the ceremony is managed by them. A hŕmren person, probably from the Pró-khămmă because such a person is close by, is selected by the Pró-khămmă to test the new crops. If he chokes or if the passage of the food down to the stomach is in some way untoward, then the people may not yet eat the crop. The same food must be presented to the Pró-khămmă again in a few days.

If the food passes the test, one of the strongest and healthiest men of the Pró-khămmă blows his breath over the offering. It is believed that in this way his strength is spread to all tribesmen who later partake of the crop during that season. This ceremony is seldom practiced these days, and if a family member does send a bowl of food to the Pró-khămmă for testing, it is almost certain that a number of individuals of other families have already tasted it ahead of time.

This ceremony is sometimes practiced for new wild fruits that appear for the first time in the months of September through December, such as buriti, buritirana, bacaba, and bacuri (Table 2).


Like the chieftainship, the council of elders has a number of loosely defined roles. As an extension of the judicial system, or as an adjunct to it, the council of elders is a forum in which problems between families can be aired before, during, or after they are taken up at interfamilial hearings [III.D.3.a]. Then, the elders of the council discuss the festivals and help the Pró-khămmă reconstruct the sequence of events that tradition­ally should take place during the coming week or day. Also, in times of a weak chief or when the chief might be about to do something extreme, the council stands ready to advise him or to offer alternative solutions. Since the council meets twice a day when the tribe is in the village, it constitutes an excellent forum at which any older member of the tribe can air his views or simply discuss what has been going on during the past 24 hours. In this way rumors are often dispelled before they travel very far, and solidarity is maintained among the participants.


Everyone in the tribe is likely to be represented by at least a “mothers’ brother” or a “grandfather” (an “uncle” [In.4.i]) in the council of elders. When topics concerning any of his younger kin arise, this uncle is likely to say something about them. The Canela prefer to have such two- or further-link-away kin [III.E.2.b] representing them because a father cannot be impartial, they say. He has too much feeling and cares too much for his children. Traditionally, one-link-away kin must always support each other. The term for sibling, i?-khyé, also means “it pulls” (i.e., supports) or “her/his/its thigh/leg,” the limb that supports its body.


Because the council meets twice a day, discussions about problems between families are likely to arise in the council before they can be taken up in an arranged hearing between the families. Such meetings between extended families do not occur regularly and take place during the part of the day that is ordinarily devoted to work or hunting. Thus, it sometimes may be hard to schedule them to suit the various participants, or to have them take place in time to be of much use. However, the problem is often resolved so that a formal hearing is not necessary in the relatively impartial atmosphere of the plaza, with further-link-away kin discussing the matter for the two families.


We have seen that one of the worst evils of Canela life, according research assistants, is the negative rumor (tswa-?nă) [III.A.3.c.(3).(e)]. While the tribe was split between Ponto (Chief Kaarŕ?khre) and Baixăo Pręto (the older Kaapęltůk) and then between Sardinha and Baixăo dos Peixes for a total of 13 years, malicious rumors frequently caused hurtful feelings between individuals of the two segments of the tribe [III.D.1.g.(l).(b)]. I used to hear many complaints about such rumors; there were no joint morning and evening council meetings to dispel them. Research assistants inevitably said that the cause of the rumors was the schism in the tribe, and that if the villages were joined, the rumors would cease.

Research in 1979 on key words and concepts indicated that the Canela do not understand the Western concept of chance, and that although they have words for “maybe” (mălmă) and “almost” (aymăl), these concepts are very approximate and are easily lost as qualifiers. Thus, a certain lack of concern for reality testing in casual verbal communication facilitates the spread of rumors, which are then seriously addressed in tribal and interfamilial meetings.


During most of my time with the Canela, it seemed that the chief had more power than the council of elders, particularly in the late 1970s when the Pró-khămmă of the age-set of the older Kaapęltůk was reduced in number by deaths, and the power of Chief Kaarŕ?khre was rising through the help of the strong Indian service official in Barra do Corda [II.B.3.e]. During my visits, there were few obvious examples of the council of elders curbing the chief; but this was happening constantly in the informal discussions and the resulting interplay between the members of the two institutions. The chief usually discussed matters with the council at least once a day. In small societies in which individuals face each other directly, a chief realizes he has to live with his people for the rest of his life. Thus, he must remain in relative agreement with them or lose their support, so that considerable and significant checks on his behavior and consequent balances take place at the level of simple plaza meeting verbal communication.


In my 1984 communication from the tribe, I was surprised to learn that in about 1981 Chief Kaarŕ?khre was actually called upon to resign, and that he did so immediately. In 1979, I would not have thought this possible, since chiefs have traditionally died in office. Kaarŕ?khre’s overthrow was characteristically Canela. When a prominent member of the young new Pró-khămmă, Tep-hot, denounced him in the council of elders, Kaarŕ?khre promptly withdrew from the chieftainship without any fuss. When challenged, Canela leaders simply turn their backs and leave [III.B.1.h.(1)]. This is why challenges are so serious and so rare. Tep-hot must have had many of the new Pró-khămmă strongly behind him or he would not have denounced his chief. (For more information, see Epilogue, [Ep.2].)


According to the 1984 reports from the tribe, the council “voted” four new chiefs into office during the two to three years after the resignation of Chief Kaarŕ?khre, and they were all very much younger than him. The last one, the youngest Mďďkhrô, was selected in 1984 at age 35. Two sources report that he is not doing well, that leadership has gone to his head, and that Kaarŕ?khre is trying to regain his former position.

Four chiefs in so few years is unprecedented in modern Canela history [Ep.3.b,4]. Since about 1981, the council of elders, led by the new Pró-khămmă of the younger Kaapęltůk, has been governing the tribe. Thus, the council of the elders, including the Pró-khămmă, checks and balances the power of the first chief when the chieftainship is weak [III.D.1.e].


The high degree of solidarity found among the Canela is reinforced by the twice daily council meetings. They usually start very informally with some of the members gossiping about the latest game killed, a girl’s virginity loss, or a backland encounter. Much of the communication in these meetings constitutes emotional bonding for the speakers, and this is especially the case when the chatting takes place in the center of the plaza, where overt hostility must not be shown.

When the formal meeting starts and the tone of the speakers is radically altered, serious topics come up. Arriving at their solutions serves to enhance the solidarity of the participants: they have come through these “battles” together and worked out of them successfully, by compromise and consensus, techniques in group dynamics that have demonstrated sacrifice and good will [III.B.1.d].

Such strong reliance on tribal council meetings is not reported for most other Indian tribes. The Canela and Apanyekra institution of the tribal council meeting may be especially necessary for these tribes to maintain the desired level of peace, order, and individual satisfaction. Other tribal communities may be less concerned with peace and harmony [III.B.1.h.(2)]. The institution of the tribal council of elders contributes significantly to the smooth operation of the Canela sociocultural system.

[III.D.3]  Judicial System

The effectiveness of the Canela judicial system is remark­able. Considerably more than any other Timbira tribe in the last half century, the Canela have been able to stay together in large numbers in one community. I believe that one of the principal factors contributing to this societal cohesiveness is the effectiveness of their judicial system. It helps to keep most Canela individuals relatively satisfied with their lot by reducing their frustrations to a tolerable level.


The Canela judicial system is characterized by inter-extended family hearings (Glossary) (audięncias: më aypën pa), non-punitive restitution, payments by the extended family as a whole for one of their individual members, and searches for “objective” evidence from witnesses on which to resolve the case. The hearings can be convened and held very rapidly on almost any subject that is thought to threaten the peace and tranquility of the society [III.A.3.c.(3).(h)].

An uncle [In.4.i] of the offended person makes the complaint informally to an uncle of the person who is alleged to have caused the problem, and these persons and other members of the two families have informal discussions about the circumstances. Then the uncles agree when a formal hearing is to take place, usually within a few days and not long after the morning council meeting in the plaza. Morning is a time when men go hunting and fishing and families go off to work in their gardens if they have not left before sunrise. It is also when women become involved in their more time-consuming domestic activities, such as preparing hot rock ovens for meat pies, or when men start weaving large mats or carving required objects for a festival. It is a time when people undertake substantial activities [II.E.5.e].

If the problems of the hearing cannot be resolved in one day, during two or three of the morning hours, the meeting is adjourned and held again on the following day, or during the following week, depending on various circumstances. Usually the problem is resolved at this interfamilial level, but if no agreement is reached, then the matter has to be taken up before the chief of the tribe. At the end of the deliberations the chief pronounces his decision, which is final.


A judicial hearing takes place in the prime work period of the day and is not just squeezed in during a time when there is nothing important to do. At about nine in the morning with breakfast over and some of the household chores done, members of the two concerned families gather. The meeting takes place on whichever property of any concerned party is the most convenient and appropriate. Participants sit on mats or benches around an open area.

The atmosphere of a hearing is less formal than that of a council meeting. People arrive or leave; children play nearby. Older uncles or grandfathers of either party take turns conducting the meeting. Deputy chiefs often come to provide their considerable leadership abilities. Sometimes they call on individuals for testimony and at other times people speak up voluntarily. Any listener, however (even an outsider) can tell that a formal meeting is taking place. The tone of voices is different and the cadence slower and more precise than in casual speech. More archaic language is employed. Such hearings last 20 minutes to several hours, until noon or one.

While small children attending such hearings are not required to be absolutely silent, they are nevertheless taken out if they cry or create a disturbance. The atmosphere is quiet and serious. There is an air of dignity and respect towards all that is said; and there is poise in behavior and clear articulation in speech [III.B.1.f.(4)].

This drama is impressive to see. The Canela have a particular genius for resolving interpersonal problems. I used to wonder how a female plaintiff in a marital hearing could be allowed to go on railing in serious but angry tones against her husband, asking him to leave. How could the audience have the patience to sit through such stories of personal abuse when everybody knew divorce was not possible and never would be allowed because the unfortunate couple had children? Nevertheless, the people would be attentively listening with excessive forbear­ance [III.B.1.e] while she ridded herself of her hostility.


While any problem that amounts to disturbing the peace can be the subject of a hearing, the likely problems fall into definite and limited categories. By far the largest category is marital difficulties, maybe 90 percent of all cases. The frequency of marital cases in their judicial hearings suggests that the greatest area of societal stress among the Canela may be found in the institution of marriage. After marital problems comes theft, followed by physical abuse. There may be also some questions about farm plot rights, but these are rare. In earlier times, suspicion of witchcraft was a more frequent issue. I was aware of only one hearing on witchcraft during any of my visits.

[III.D.3.c.(1)] In Marriage

Marital difficulties that come up before hearings are diverse in their characteristics but can be put into four categories. The first problem that comes up in a woman’s life is the question of virginity: who took it and do her and her lover’s claims differ. This is very important, because the act of taking a girl’s virginity, done by an “unattached” man (i.e., one with no biological child of his own in a marriage [III.F.4.b.(1)]), means that he is now married to her. Some­times, but rarely, girls are caught by several male friends and willingly lose their virginity in this way. In this case none of the men are her husbands. She is “on the street” (na rua), and has lost the price of her virginity (i?-pore piktol: her-money lost) [III.F.4.c.(2)]).

Secondly, there is the problem of a young man not wanting to stay in his recent marriage. The hearing must determine the items in the payment to be made by his extended family if he leaves his wife. A woman usually does not want a man to leave her; but if she does, she denies that her motive is to get the payment. On the other hand, if a woman without children wants to leave a man, she takes up with another man and no payment is involved. The husband simply leaves and there is no trial. Hearings come up only when there is a disagreement over whether a payment should be made and over how much it should be. Men pay to leave women; women do not pay to have men leave their house (W. Crocker, 1984a:67).

Thirdly, for couples with children, the possibility of divorce almost does not exist―or did not, before 1970. There is still the question of the wife shaming the husband by an overly public extramarital act. Although traditional extramarital sexual relations are sanctioned by many of the festivals [IV.A.3.f] and enjoyed by all in practice [III.F.8], an insecure male may make a fuss when his wife has had an affair, the details of which come back to him in too public a way. He may feel he must take some action to save face and may leave his wife and children to return to the house of his sister or mother. From there he may make a public claim for a payment from her family to erase the shame “from his face” before he can be expected to return, thereby obliging his extended family to support his protest.

Finally, there is the typical case of a woman whose husband has been indiscreet in conducting his affairs and has also perpetrated many other unkindnesses, such as giving large presents to other women, providing insufficient food (espe­cially meat) to her family, and maybe even physically abusing her. Although she knows she cannot get rid of him while they have growing children [III.F.9], she may nevertheless act as if she thinks she can achieve divorce and may even insist that he temporarily leave her house and their children. Subsequently, the two extended families must hold hearings to convince her that she has to let him return. She usually will yield to the pressure from of her female kin but not until or after having received a considerable payment from his relatives. The important point may be, however, that she has aired her problems about him publicly and let him know what she thinks of him directly at the hearings or, almost worse, indirectly through friends (male or female) [III.A.3.c.(3).(e)].

[III.D.3.c.(2)] Theft

It is often difficult to decide what was stolen and what was freely taken, because personal items are so easily exchanged between good friends of the same sex, especially men. Moreover, since generosity is the highest value [III.B.1.a], it often behooves certain more prestigious individuals to look the other way while others take what they want. Taking other people’s food is hardly considered theft since nourishment is a basic necessity. Certain festivals even sanction “begging” [IV.A.3.c.(5).(c)], making food a free commodity.

Most traditional items are freely borrowed and kept, but a few objects like tapir hoof-tips, are too valuable, either in terms of rarity of original resource or time consumed in fabrication (woman’s sing-dancing sash of woven cotton [II.G.3.a.(8)]; Plate 58e,f), or of ceremonial honor (back-pendant comb [II.G.3.a.(7)]; Plate 59b,g), so that the owner could not just let other people take them. Acculturation is spreading the sense of possession to most other traditional objects, even to those of little value.

Certain traditional objects and equipment from the cities are of sufficient value to the Canela, so that taking them without compensation or an understanding with the owner constitutes theft. One of the most valuable traditional items is a girl’s necklace (hő?khre-tsęę: throat attachment) of many beaded strands [II.G.3.c.(3)] (Plate 57b). The ceramic beads have come from eastern Europe since the last century, but the style is traditional. Similarly, a man’s ceremonial belt (tsů) is valuable, both in the tribal world because it may have tapir hoof-tip pendants [II.G.3.a.(3)] (Plate 56d) and in the urban world because these pendants may be attached to the woven cotton belt by strands of European beads.

Some other items of value in the traditional category are, for women: cotton singing sash (hahĩ), comb back pendant (khoykhe-re), gourd back pendant (krat-re), a full belt or girdle (i?pre), forehead carrying basket (khay); and for men:ceremonial lance (khrúwatswa), war bonnet (hŕkyara), cattle horn (hő?hi), cotton singing bracelets (patsęę), earplugs (khuy), certain wooden staffs (khô-po). Any of these items (Table 8; [II.G.3]), would be considered stolen (ha?khĩya) if taken from their owners.

It is mostly urban-made items, however, for which individ­ual possession seems to be inalienable, following the backlan­der custom. Machetes, axes, shotguns, hoes, pots, caldrons, cups, plates, spoons, knifes, cloth, tailored clothing, suitcases, beads and many other similar items, when taken and kept, are considered to have been stolen. Some of these items would not be sufficiently valuable to bring up before a hearing; however, machetes, axes, shotguns, caldrons, and especially young girls’ bead necklaces would be sufficiently valuable. The bead necklace may represent the accumulation over years of bead collecting by a number of women, all bestowing their beads on a favorite daughter, niece, or granddaughter.

The apprehension of the thief is a particular problem. Apparently shamans are good at this. They are always watchful as part of their shamanistic role, and research assistants say that a ghost can tell them the location of the stolen items [IV.D.1.e]. It is particularly hard for the culprit to hide his tracks in a sandy region; such telltale traces often give the thief away.

The surest way of obtaining money for the stolen item is to take the goods to specific backlanders who buy them for a fraction of their real value and do not give away the thief’s secret. In the late 1950s and mid-1960s young men were occasionally culpable of such acts of desperation in order to buy gifts to please their lovers.

The mere restitution of an axe for an axe, or a pot for a pot, as is the custom, does not seem commensurate with the trouble the thief causes the whole community. Many people are inconvenienced and spend much time over the case. Everyone seems to have plenty of time to spend on such matters, however, and are troubled very little by losing time. The thief on the other hand, loses a great deal more if apprehended, because he loses the favor of many tribal members. Stealing has never been sanctioned by the Canela, and anyone apprehended in such an act is remembered for that act for many years. The chief is less likely to favor him, lenders would find it harder to give him credit, and potential lovers are less likely to cooperate [III.A.3.(j)].

[III.D.3.c.(3)] Physical Abuse

Fighting between men (outside of warfare) was and is very much against the Canela cultural tradition [III.A.2.k.(2),(4)]; physical abuse is even more against their tradition. Elements of masochism, however, especially in male competition for women, are entering the Canela scene with acculturation. When men speak during interfamilial hearings, they sometimes switch into Portuguese saying sou homem (I am a man) to excuse their nontraditional tendencies toward jealousy and retaliation against other men.

Physical abuse does occur, usually when men are returning from backlander homes or from Barra do Corda. The Canela believe that in such cases it is the cachaça (raw distilled cane alcohol) that is to blame rather than the person. On the rare occasions of physical abuse, it is usually the wife who has been the victim rather than other men. Thus, the case comes up before a formal hearing.

Cases of physical abuse occurred more often in the late 1950s and in the 1960s than in the 1970s. There was greater instability in these two decades; in the 1970s the Canela political establishment had largely disapproved of alcohol and such indulgences were less frequent [II.B.2.i.(5)].

One type of physical abuse that would never come up before a court occurs when a young girl has been stingy with her sexual favors [II.D.2.e.(3)] [III.B.1.a.(4)]. Then young men have to “tame” her by catching her off picking fruit with a girlfriend and forcibly having intercourse with her, each taking his turn (W. Crocker, 1974a:187, 1984a:65). If she should be hurt in such an encounter, her family would be ashamed that the youths, with the cooperation of the girlfriend, had had to resort to such measures to make her become more generous [IIII.A.2.j.(6).(b)]. In fact, they would be so ashamed that they would not think of bringing the case to trial on a question of physical damages.

In other cases of physical abuse, compensatory payments are awarded the plaintiff if the skin is broken and bleeding or if bones are broken. Bruises do not warrant payments.

[III.D.3.c.(4) ] Violence

While violence is not as characteristic of the Luso-Brazilian as it is of the Hispanic or North American, it is more characteristic of the regional backlanders than of the Canela who are meek and tame by comparison [III.B.1.h.(4)], these days; formerly they may have been fierce [III.A.2.k.(3),(4)]. Thus, it is to be expected that with acculturation, violence will increase among the Canela.

One of the most violent individuals [In.4.e], 40 years old in the 1970s, had spent much time away from the tribe in large cities. He even spent a year or more in the Indian service detention center in the state of Minas Gerais for committing homicide while drunk in Brasília. If he does not stay away from the tribe, more violence for Canela males can be predicted as others follow his example. He visited Escalvado during my stays there in 1975 and 1979, and caused much trouble for the leaders of the tribe and for the Indian service agent, declaring that the Indian service in Brasília was going to give him both positions.

In contrast, the younger Tsŕŕtu (Plate 45d, on left), in his 20s in the 1970s, who has spent far more time away from the tribe in Rio de Janeiro, has returned to the Canela in the 1970s and 1980s and is not given to violence at all. In 1984, he had a Indian service salary as the teacher’s assistant to help the students understand Portuguese by translating it into Canela [Ep.5.a].

[III.D.3.c.(5)] Suspicion Of Witchcraft

In earlier times, suspicion of witchcraft used to be a prime cause of dissension and schisms among the Canela and other Timbira tribes. Currently, however, witches have become weak in their psychic abilities [III.A.3.c.(3).(h)], and physical retribution against witchcraft is no longer possible. In the past, the Canela executed witches, but now the Indian service personnel would intervene, most likely by taking away the executioners as murderers, as they did in the last known execution for witchcraft in 1903 [II.B.1.c.(4)].

One morning in Escalvado when I was getting ready to receive my research assistant council, I was invited to attend a hearing on a witchcraft accusation. The hearing appeared to be the same as any other one. A deputy chief was in charge, the plaintiff and the accuser were called on to speak, witnesses were summoned and heard, and the uncles of both sides discussed the matter thoroughly. I had heard this rumor of witchcraft the night before but was nevertheless surprised and impressed by the rapidity with which the chief had called together the meeting to resolve the issue [III.D.2.e]. The accused immediately denied the accusation, and finally the accuser withdrew the accusation in the face of the evidence. The case was dismissed after two hours, and I did not hear of the matter again.


Interfamilial problems first arise, quite casually, in the council of elders [III.D.2.d.(1).(a)]. If not informally resolved there, they are taken up at a formal interfamilial meeting in a house on the village circle. Currently, if matters are not resolved at the interfamilial level, they are taken up again in the chief’s house behind the village on a Sunday or holiday. This occurs at the same time of day, about eight or nine o’clock in the morning. Thus, there are two levels for the formal resolution of problems and sometimes there has been a third.

When I arrived for the first time in 1957, a formal trial was in progress in the plaza among the elders. They met about once a week a number of times over the failure of one family to pay sufficient compensation to another family. A youth of the first family, Măăkhrô, had left a girl of the second family, Wakhőő, whom he had married when she was a virgin. Thus, there have been times when there were three levels at which problems could be solved formally: between family houses, in the plaza, and at the chief’s house. It is not clear from my research just how many levels were in operation in the past. Nimuendajú (1946:159) concurs with the plaza resolution of problems, although current research assistants do not believe formal plaza trials were the practice of their ancestors. When population levels were much higher in prepacification times, however, it might have been advantageous to have had certain problems solved at three levels. In addition, nonresolution at the interfamilial level might have led to resolution in the plaza when the chief was weak (i.e., new) and resolution in the chief’s house when he was strong.


A number of traditional principles exist according to which problems can be resolved when they are brought before a formal hearing. These principles have been alluded to before in the above materials but will be addressed here more directly.

[III.D.3.e.(1)] Maintain Peace

The first and foremost principle for the Canela must always be keeping the peace [III.D.1.g.(1).(c)]. Older men abhorred any occurrence, or even an idea, that could lead to divisiveness.

The older Canela (those in their 60s when I first arrived in 1957 [II.B.2.g.(8).(a)]) used to dislike questions that appeared to separate people into different categories. If I asked what portion (few or many) of the people did such and such a thing, they liked to respond, if it were a “good” activity, that all the people did such a thing. When I asked another similar question, the response was always “the whole.” Even though it would have been inconsistent with the first question, they nevertheless answered that all people did this second activity too, not recognizing the inconsistency. They enjoyed the idea that everybody did everything the same way without exception.

The need of these old men for uniformity in their answers disturbed me at first. I soon discovered that the basis for their desire to live with 100 percent uniformity was really their fear that categorization into separate parts—portions, “percentages”—might lead to divisiveness. In contrast, the present Pró-khămmă (1980s) have adapted to thinking in terms of social parts (and market-related prices) quite effectively [Ep.5.d].

[III.D.3.e.(2)] Consensus

The Canela genius lies in their ability to resolve interpersonal and interfamilial problems by compromise and arriving at a consensus—by overlooking what does not “fit.” Of course, acculturation could change this orientation.

It is relatively easy for the Canela in tribal council meetings and judicial hearings to come to a consensus which they then claim is a unanimous decision. They do not like to remember or recognize that differing points of view were expressed during the course of the debate. The Apanyekra may differ here. The research assistants who had worked with me in groups for years still had the same basic orientation, even though they knew that Pčp (W. Crocker) “wanted only the truth” [III.B.1.h.(4)] (Epigraph). It was difficult to convince them that legitimate different points of view were worthwhile for me; they wanted to give me only what was "right.”

In 1975 and 1976, I compared the Canela ideas about the folk Catholic story of the creation and God’s development of the world into more modern times with the more aboriginal points of view of the shamans. My faithful old research assistants kept wanting to arrive at a consensus in our group of four to six people [Pr.2]. I tried to convince them, with the very considerable help of the younger Kaapęltůk, that there could be two or more legitimate points of view, and that I wanted all points of view separately. Since the people on the folk Catholic side in my meetings were more assertive, I had to continually bring out and preserve the points of view of those on the shamanic side in the discussion.

These research assistants, true to the Canela spirit, wanted to develop a consensus and turn it into an absolute statement. This was a potentially weak point in my field research into memories about the past against which I always had to take precautions. The younger Kaapęl could help me watch out for this difficulty because he understood my thinking and preferred it for himself [Ep.7].

Similarly, debates in the formal interfamilial judicial hearings, which could be heated at times, were usually resolved into a consensus before the meeting was over, enabling everyone to part with a feeling of good will. The consensus emerged partly because the weaker individuals gave into the stronger, but also because the stronger individuals—if they were more prestigious or if they had less feeling invested in the case―gave in to the weak with strong feelings to preserve the peace.

[III.D.3.e.(3)] Compromise And Social Leveling

Several ways exist to be more prestigious among the Canela. A person or a family can have ceremonial prestige, political prestige, and also, in current times, prestige gained through acquired relative wealth (Gross et al., 1979:1049), largely through Indian service salaries [II.B.4]. In any of these cases it behooves the more prestigious, in the setting of a judicial hearing, to let the weaker persons have their satisfaction when it seems appropriate to do so.


I vividly remember a case that took place in 1958 in my adopted family. My sister’s oldest daughter, Te?kurŕ, had been the Wč?tč girl for a number of years until the year before, and so our family was ceremonially prestigious. (My presence was also a factor.)

One time when Te?kurŕ went down to the stream in Ponto village during the middle of the day to get water, she met a male friend there who flirted verbally with her. Someone saw the two together and told Te?kurŕ’s husband who was immediately consumed with jealousy.

According to Canela custom, Te?kurŕ had not done anything wrong except that the particular bathing spot was a public place. If she had been more thoughtful, she might have avoided lingering there alone for a conversation with someone not her husband. If she had had sex with the same man in the bushes further away from the village, this too would have been all right, even if someone had seen her and reported it to her husband. (Such reporting was quite untraditional.) In any case, the husband and his family made a fuss, and he left his wife’s house with all of his possessions and returned to his mother’s dwelling (next door). They had no children. Te?kurŕ’s husband came from a ceremonially less prestigious family and may have been looking for some excuse to redress the balance.

The husband demanded a payment from my family to erase “the shame passed onto his face” (passou vergonha na cara). Certain uncles must have told him he had no case [II.B.1.e], and they surely lectured him that he must suffer such jealousy and suppress it [III.A.3.b.(1).(b)] [III.B.1.e.(1)]. His immediate family (one-linkers [III.E.2.b]), however, persisted in the suit.

What I remember most about the trial was that the grandfather representing my adopted family, who happened to be my naming-uncle’s naming-uncle, the older Mĩĩkhrô [I.G.3], said that according to tradition no one had to make a payment on the basis of any wrong committed in this case. He said times were changing, however, and therefore our family should pay Te?kurŕ’s husband’s family something in order to keep up with the times.

A research assistant later told me that Mďďkhrô had given in to the opposing side because the husband’s family was poor and because they were for some unknown reason very angry and hostile. The payment had been made to keep the peace; less fortunate families were to be appeased so that they too could be happy and satisfied with the social status quo.

A solution of this sort reminded me of the pattern of the Regeneration season moiety log racing practices [III.C.4.c.(1)] [IV.A.4.c]. During their ascendancy, the prestigious side gives way to the harassments of the lesser side, a relationship that is reversed when the period of ascendancy changes sides. This social leveling is one more way in which the Canela preserve social cohesion and peace [III.D.2.e].


In keeping with the same predilection for social leveling, a number of complaints were resolved before they reached the stage of formal hearings. For instance in the early spring of 1975, the younger Kaapęltůk suffered a very significant financial loss. Many pineapples were stolen one night from one of his large farms. He told me that he was sure which family had done this, and his proof seemed quite convincing. I asked him why he did not press the case against that family in a hearing, and after some discussion he reminded me that he was the father of the Ceremonial-chief-of-the-whole-tribe [IV.A.3.c.(3).(e)]. He offered no further explanation; I knew the customs. He and his son, as the highest hŕmren persons in the tribe, could not press such a case against a much “lesser” family. He received pay from me, so he had financial prestige, whereas the other family had no source of income other than subsistence farming. Furthermore, Kaapęl was in charge of the whole Lower age-set moiety, a position that gave him great political prestige and power.

This kind of loss happened to Kaapęl a number of times during my visits. In his store, which he was trying to make profitable, it was hard to refuse giving goods on credit to certain individuals, even though he knew this amounted to giving the goods away freely [Ep.4.b.(2)]. In effect, his customers were taking advantage of his high hŕmren position (Glossary).

[III.D.3.e.(4)] Ease Shame, Save Face

As in the case of Te?kurŕ’s husband, payments are frequently made to save someone’s face or to ease her or his shame. The Canela do not have a guilt-oriented society basically; most problems are seen more in terms of the shame brought upon an offended party than the guilt of an offender. If no one has seen an offense, the perpetrator does not feel badly about having done it, though she or he knows the act was wrong. The offense is not felt as really being wrong until discovered and exposed. With acculturation, the Canela increasingly believe in the backlander’s folk Catholic concept that God is everywhere and sees everything, and that He will punish a person for an offense He has seen, even if no Canela has seen it [III.A.3.c.(3).(i)].

Regardless of this growing belief in God’s punishment (guilt-oriented), the Canela still live in a society where everybody knows everyone else. Even though the Canela are unusually quick to forgive and forget offenses [III.B.1.c.(3)], they nevertheless remember antisocial acts and tend to modify their behavior toward those who commit them. Thus, publicly saving face is still very important, giving the offended person a little more social viability in the future, and at the same time possibly restraining the offender in a similar future situation [III.A.3.c.(3).(a),(b),(e),(j)].

[III.D.3.e.(5)] Restitution, Not Punishment

Restitution in the form of a payment when something has been stolen is clear and easy to understand, but the principle among the Canela also holds for intangibles, such as easing shame, saving face, and making up for certain kinds of personal suffering experienced by the plaintiff. Restitution is sought not to punish the offending party as much as to improve the position of the plaintiff, as seen by the plaintiff. Thus, arguments and claims follow a restitutive rather than a punitive kind of thinking. Nevertheless, in extreme situations, such as fathers attempting to leave their children, punitive payments are demanded of the father [III.F.9.a].


Restitutive thinking is operative in setting the payment a youth has to make to leave a wife who has no children by him. The amount he must restore to her family varies according to the number of steps he has taken into the marriage, because her family has paid his at each step for “buying” him [III.F.4].

First, he and his natal family must usually pay for the loss of her virginity. Then, they may have to pay for the meat pies given by her family to his at the time of his “purchasing” rite.

Later still, there is the expense of the deer given to his female relatives just before they paint the belt and body of his bride with urucu.

The payments for disengagement must be considerably greater if his wife has had a child by him that died. After childbirth, research assistants speak of the ripping and tearing of the girl during this ordeal that has to be paid for by the young man’s family. Finally, there is the great expense of the postpartum rite—the contributing-father rite (hŕ?khrël)—and all the meat pies that were delivered to the young man’s family members at that time. Additional considerations would be the undue suffering he has caused her, such as shame and loss of face because of open infidelities, and hunger if he has not supplied her with sufficient meat.

Subtracted from these payments from his family to hers would be any public shaming she had caused him. A certain amount would also be subtracted if she had not provided him with good cooked meals and all the traditional comforts of home.

Sometimes these days, at the time of the son-in-law buying rite, the husband’s family sends the wife’s family a number of meat pies to balance those already received from the wife’s family. This would reduce the amount of the restitution that a young man’s family would have to pay his wife’s family upon his leaving her. (For marital balances from a somewhat different point of view, see [III.F.5.b].)


Thoughts of punishment do not enter the picture except in the very unusual situation when a young father is leaving his child or children as well as his wife—and they call this “children-leaving,” not “wife-leaving.” Then, the young man has to depart with “clean hands” (măo limpa), they say, retaining nothing except his clothing [IV.B.3.b]. This happened in 1958 when a man [In.4.e] in his late 20s with a wife and several children took the virginity of a young girl. He subsequently left his wife and started living with the young girl in her family setting. Under these circumstances, it was considered that he was married to two women at once, because taking a woman’s virginity and moving in to live with her usually constitutes marriage. However, this was not considered a proper marriage, because the man already had children in a marriage [III.F.4.b.(1)]. To the Canela, having two wives is absolutely forbidden, but at the same time very comical, and the jokes and rumors flew around the village. They likened him to a certain contemptible backlander, Tomas, who they believed had “wives” in two different backland communities. They even called him by that backlander’s name.

The girl’s family may have accepted his living in their house (which was unusual), because he was an excellent economic provider. The young husband left almost everything he had with his first wife and children, including a horse and saddle, many iron implements, a shotgun, and a farm plot he had prepared and cultivated. I returned in 1959 to find that this young man with two wives had rejoined his first wife and children, having eventually left the young girl but having paid her family a high amount. He did not regain his earlier payment to his first wife’s family. I have wondered what would have happened if his second wife had become pregnant and a baby born. That would have forced a Canela “divorce” [III.F.9] from one wife or the other because children would have been involved on both sides.


The Canela have always surprised me by the extent to which they seem to enjoy and to be satisfied by their way of living. This is manifested by the fact that so few Canela have left their society permanently to live in the outer world. Even though the Canela have had a tradition of traveling to the great coastal cities since some time in the last century [II.A.3.a.(3)], few have stayed away for long. The one period of two years when significant numbers did emigrate tends to demonstrate the generalization because of its unusual circumstances.

In 1963, just after the denouement of the messianic movement [II.B.2.g.(1)], and in 1964 and 1965 before news arrived in 1966 from the Indian service in Brasília that the Canela were to be allowed to return to their homelands during the next two years, some 40 young people had left the tribe to live in the great cities of the coast and in Brasília. All of these youths returned to their tribe by 1968, however, except for nine young men who were between the ages of 14 and 20 when they left.

These nine youths either had married away or had found comfortable situations that they must have felt were disadvan­tageous to leave. Three young men remained in Recife, two of whom had married city Brazilians. Another adolescent, the younger Tsŕŕtu [Ep.5.a], was taken in by a wealthy lady in the Cosme Velho part of Rio de Janeiro. She had lost her American husband just a year before and needed a handy man to help her around the house. Hôy went to Săo Paulo, became associated with a Protestant center, and came back to Escalvado sporadically in the 1970s. One Canela stayed in Brasília, and three more went to smaller cities. In any case, only nine young men left the tribe, apparently never to return, under the extreme conditions of their tribe’s difficulties in adjusting to a dry forest environment [II.B.2.g.(9)].

Since about 1880, only two or three other examples of emigration were reported in the history of the Canela. I take this general reluctance to leave their people to be a measure of high social cohesion, of the considerable satisfaction they derive from their tribal living, and of the sufficient degree to which individual frustrations are reduced. I see the effective operation of the judicial system as one of the most important factors contributing to lowering general frustrations to a tolerable level [III.A.5.a].



The expression “relationship system” is used by some anthropologists11 and not by others; thus it may be important to explain that I am referring to various terminological systems such as the consanguineal and affinal, Formal Friendship, and name-set transmission ones, which tend to pattern behavior even though terms of address and reference may be only occasionally employed.

Researchers of the Harvard-Central Brazil group have already described much of what is in this chapter for their particular Timbira tribes (Maybury-Lewis, 1979; Da Matta, 1973, 1976, 1982). Arnaud has made a significant contribution for the Gaviăo Indians (Arnaud, et al., 1976), and Gross’s (1979) approach should be considered. However, the materials here, though similar, vary from the tribal presentations of these other researchers, because all of these related tribes were somewhat different aboriginally and have become more different through acculturation and population loss (Wagley, 1973:154; 1974:375). Moreover, the Canela, through isolation and greater population numbers, have remained more traditional, and they have retained more of the Timbira traditions of 200 years ago.

[III.E.1]  The Nine Relationship Systems

The consanguineal and affinal terminological systems are the most obvious and frequently used of all the relationship systems. Next, there are the name-set transmission and Formal Friendship systems. Notably, Canela speakers give name-set and Formal Friendship terms a higher priority in choice and use than consanguineal and affinal terms when alternatives exist, except at the first-link-away from ego terminological range (Figure 23). Then in addition, there are the Informal Friendship and mortuary terminological systems, which are considerably less used.

The next three systems, teknonymy, contributing father relationships, and ceremonial relationships, are different in that they draw terms from the consanguineal and affinal systems and do not have terms of address and reference that are uniquely their own. The first of these, teknonymy, is terminological in the sense that a regular sequence of words is used in address and reference. For the next system, contributing father (co-father) relationships, terms from the consanguineal system are used, but the patterning of their application is somewhat different from the patterning in the consanguineal system. Finally, there is a ceremonial system found in the initiation festivals in which consanguineal terms are used. Away from these festivals settings, these consanguineal terms are still occasionally used to the extent that these festival behavioral patterns continue to be observed in daily life, but the use of these ceremonial terms is diminishing with acculturation.

[III.E.2]  Consanguineal Terminological System

The consanguineal system is similar to Crow (Glossary) and most similar to Lounsbury’s Crow-III (Lounsbury, 1964:375–377; Tables 10, 11; Figures 26, 27; see also Lave, 1979:22.) A similar arrangement of terminology for kintypes is found among the other Timbira tribes (Maybury-Lewis, 1979), with variations in the direction of Omaha in the west. There is evidence, however, that Scheffler and Lounsbury’s (1971) parallel transmission equivalences are somewhat more congruent with the terminological data and with the social structural correlates than Crow-III, or even a mixture of Crow-III and Omaha-III, with preference in conflict situations given to Crow (Keesing, ms). However, the Canela and all Timbira tribes are not matrilineal (Da Matta, 1967:142) [III.C.1.10.b]. There is no current evidence for clans having existed, though experts wonder from what structure the men’s societies and plaza groups derive (J. Melatti, 1979b). The Canela and Apanyekra have bilateral kindred with the matrilateral side emphasized and the patrilateral side evident mostly in the practice of individual rites. Nevertheless, the patrilateral kin are treated as relatives when such people happen to meet [III.E.2.e.(3)].

 [III.E.2.a] Terms FZ = FZD = FZDD = FZDDD = FM = MM; MB = MF = FF = FMB; F = FZS = FZDS = FZDDS

Mother’s sister and father’s brother are “mother” (nŕŕ) and “father” (päm), respectively, and mother’s sister’s children and father’s brother’s children are “siblings” (khyę) to female and male ego. There are no cousin terms. Father’s sister’s daughter is “father’s sister” (tůy) (Glossary) and father’s sister’s son is “father.” The reciprocals to father’s sister’s daughter are “niece” (tŕmtswč) and “nephew” (tŕmtswč), and to father’s sister’s son are “daughter” and “son,” that is, “child” (khra).

A special characteristic of this particular kinship system is that father’s sister is equivalent terminologically to father’s mother and mother’s mother; and mother’s brother (kęt) is equivalent terminologically to mother’s father and father’s father so that there are no special amital or avuncular terms. Female ego’s brother’s daughter is called the same as this female ego’s son’s daughter, or daughter’s daughter; similarly, male ego’s sister’s children are equivalent terminologically to male ego’s grandchildren―all being tŕmtswč, or “descendant,” regardless of sex and generation (Figure 20).

Father’s mother’s brother is especially interesting, since it varies between “mother’s brother” and “father.” Research assistants were clear that father’s mother’s brother should be “father” (Crow-III) when involved in name-set transmission but “mother’s brother” (“parallel transmission”) when not. (Compare Melatti, 1979a:53, 72.)

TABLE 10.―Some potential kintypes of the consanguineal and affinal categories.

TABLE 11.―Canela kinship terms (consanguineal and affinal, references and adddress, first through third persons, with affixes (left of line) and stems (right of line)).


The Canela distinguish between individuals who are one “blood” (Glossary) link away from ego (including ego’s parents, uterine full- and half-siblings, and children) and those who are two links or more away (i.e., further-link (Glossary) individuals) (Figure 20). (J. Melatti, 1971:351, and Da Matta, 1979:91, make this same division, though they do not express it in terms of “links away” from ego.) The Canela call these one-link (Glossary) individuals ego’s “restrictions maintaining people” (i-to më ipiyakri-tsŕ katęyę: me-for they restrictions-doing people) while all the rest (“further linkers”) do not carry out food and sex restrictions for ego [IV.D.3.a]. Except for ego’s parallel cousins (“siblings”) and their parents (“parents”) and for ego’s same-sex parallel cousins’ children (“children”) (Figure 23), all of the further-linked from ego individuals (except for cross-generational merging ones) fall into three categories: if they are in an ascending generation and female, they are called tůy and if they are male, they are kęt; if they are in a descending generation from ego, regardless of sex, they are called tŕmtswę.

Ego’s one-link kin and her or his further-link kin are additionally distinguished in that one-linkers are not included in the category khwč (relative/kin: hũũkhyę) while all further-linkers are. Furthermore, restrictions kin (one-linkers) are paired in a complementary manner with all other kin in Canela dualism [V.A.2] (Glossary). Apparently, the distinction between ego’s restrictions kin and all other kin is a major one.

In contrast, Melatti (1971:351) divides Krahó consanguineal kintypes into two groups: those involved in procreation (mother, father, child) and those involved in transmitting names (aunt, uncle, niece-nephew) with siblings belonging to either group. I find it more useful to distinguish the biological one-linkers from all other kin, as the Canela do with their concepts of “restrictions kin” and hũũkkyę.

FIGURE 20.―Model of one-link/further-links away from ego kin. Further-link kin terms of reference and address vary with sex and with ascendant or descendant generations in relation to ego (with kintypes due to Crow-like skewing and classificatory one-link kin (Figure 23) being exceptions). Thus, for ego’s further-link kin there are principally three terms: ascendant generations, tůy (female) and kęt (male), and descendant generations (tŕmtswč). Sex-distinguished terms for child and grandchild are limited (Table 11) and used infrequently.

Da Matta (1979:113–115, 1982:121–122) divides Apinayé (Western Timbira) consanguineal terms into three groups, instead of two, because ego’s mother’s and father’s same-sex siblings, and the latters’ parallel cousins (“siblings”) carry out the significant roles of adoptive parents for female and male ego, unlike among the Eastern Timbira.


From the above further-link delineations, there are the characteristic Crow-III cross-generation merging exceptions. Ego’s father’s sister’s son, who is “father,” has children, whom female and male ego call “sibling,” but this depends on what female and male ego happen to call this “sibling’s” mother, and there are several possibilities.

Ego’s father’s sister’s descending female line’s members are “father’s sisters” “forever” according to certain Canela research assistants. Similarly, all of these “father’s sisters” sons are “fathers,” forever. Actually; this cross-generational merging practice is often lost, especially among the Apanyekra with respect to the “father” terms more so than with the “father’s sister” ones.

Generally, ego’s father’s mother’s brother is “mother’s brother” (as in “parallel transmission” but not as in Crow-III) unless name-set transmission is involved, in which case he is “father.” (Of course, the reciprocals of the Crow-III-like cross-generation merging kintypes mentioned in this section are additional exceptions to the general further-link delineations mentioned in [III.E.2.b].)


When studying successions—who takes whose place—with research assistants, they are quick to say that a daughter takes her mother’s place and that a man takes his father-in-law’s place. They also say that a man takes his mother’s brother’s place, or his naming-uncle’s place. A man, however, takes his mother’s brother’s place principally when his sister takes their mother’s place. He takes it in tandem with her (Figure 21), as the most important male succession of all, they say. I pointed out to them that a man may take his father’s place in tribal membership in the Eastern Timbira world, in becoming the Ceremonial-chief-of-the-whole-tribe [IV.A.3.c.(3).(e)], and sometimes in becoming a Visiting Chief in the Pepkahŕk festival [III.C.7.a]. This kind of succession is not foremost in their minds these days, but they did apply the same terms to these son-to-father successions as to the others: ha-tsŕ yahęl tsŕ khăm (his-place’s filling instrument/person in: his father’s place he (the son) fills in). For the Canela, rather than talking about “succession,” we should talk about the junior person (as in the cases above) “taking the place of” the senior one.


These Crow-like kinship patterns are well expressed through the demographic arrangement around the village circle. Residence is uxorilocal with a husband coming from his natal home “on the other side of the plaza,” his khŕ-tsŕ (breast-place: the place of his mother’s breast), to live with his wife and her female relatives.

FIGURE 21.—Model of the two most important consanguineal successions: a woman takes the place of her mother when she dies, and the woman’s brother takes the place of her mother’s brother.

The cognates of the Canela term i?-khwč (its-piece: a piece of it), meaning some part of a whole, are found among all Timbira tribes. Lave (1979:17) refers to the multiple use of this term (mekwu) in kinship among the Krĩkatí. For her it meansvarying groups (smaller to larger) of close kin in relation to larger groups or to all kin. Mekwu may also mean all kin versus all other people in the village, or all people in the village versus all people in the tribe (several villages). Da Matta (1979:87–91, 1982:103–115) makes what is essentially the same point for the Apinagé term kwoya. (See J. Melatti’s, 1979a:61–62: meikhďa for the Krahó.) The Canela use khwč in a totally general way—for kinship and for material items. If a person wants a piece of brown sugar, she or he might say, “i-mă ?-khwč gő” (me-to its-piece you-give: give me a piece of it [brown sugar]). The more specific Canela term for kin is hũũkhyę. Other Timbira researchers do not mention cognates for this term. A Canela can also say i-kaprôô khwč (my blood portion [of the whole]: my kin), meaning female and male linkages equally.

Thus, the term khwč can be used in relation to any of the village circle or farm groups: mëi pul khwč: (our farm group: people who place their farms in the same area with us).

[III.E.2.e.(1)] The “Hearth”

Characteristically, two uterine sisters, married or not but with children, live in the same house as their mother and father (occasionally grandparents also) and maintain just one hearth (Glossary) location (hŕwmrő: hearth: cooking place) for cooking. They share all food, whether from the hunt, farms, or raised in the village (pigs, chickens, fruits, etc.). They are less likely to share small amounts of food bought from commercial sources, such as canned goods. The hearth itself may consist of 3 to 10 well-placed rocks to support 1 to 4 cast iron pots or caldrons at the same time over one spread fire. This “hearth group” (hŕwmrő) is the minimal and principal economic unit: almost all food collected or produced by this group is cooked by any of the sisters, by their mother, or by a growing daughter and is likely to be shared by all members of this living unit, including these women’s husbands and daughters, their temporarily resident married brothers, and their unmarried brothers and sons [III.F.7]. If some of these indicated close female kin die or were not born, other closely related female kin move in from adjacent houses—first or second cousins to the mother or her daughters. These hearth or domestic single house units are made up of from two to five active female kin—rarely one or more than five. These females usually, though not necessarily, cultivate adjacent farm plots. Melatti (1979a:50) calls this hearth unit a “domestic group.”


A still smaller unit than the hearth exists, the elementary family which is the eating unit. It has no name in the language but is spatially associated with a particular platform bed in the house. Spouses eat together, usually separately from her kin. A woman and her husband eat with “their” children (including step and adopted ones), apart from her sisters but maybe with her widowed, unremarried mother. If opposite-sex individuals eat in the same group, their proximity implies sexual relations are taking place between them, as does sitting on the same mat. Thus, adolescent and adult opposite-sex siblings are careful not to eat too closely together or to sit on the same mat.

FIGURE 22. Division of a Canela hearth group (hŕwmrő: hearth). In 1979, in Escalvado, Hômyĩ-khwčy, with her husband and children, set up a separate house (BB2) and hearth just east of her mother’s (Mďď-khwčy) house (BB1) along the village circle. Thus, Hômyĩ-khwčy’s BB2 household members no longer shared the same cooking location nor the same foods with her mother’s and sisters’ families. (Khroytsen, having no daughters, lives with her son, Hŕwmrő, who is in his wife’s (Mďď-khwčy’s) position on the village circle.)


As the older daughters attract husbands and produce a number of children, they usually put up a separate house on one side or behind their mother’s house and establish a new and economically independent hearth unit (Nimuendajú, 1946:83), but some informal but minimal economic cooperation contin­ues between the members of these two hearth groups.

Such a domestic and economic splitting off from her mother by a married daughter with several large children occurred in 1979 in one of the two Canela families into which I was adopted (Figure 22). (By coincidence, the name my adopted brother of this family used from his name-set was Hŕwmrő [hearth].) Homyĩ-khwčy (thorn-woman), age 32, Hŕwmrő’s oldest daughter, and her husband Rărăk (thunder), age 39, simply moved over a period of several days with all their possessions and four children about 15 meters to the northeast along the village circle (counter-clockwise) into a small house they had recently built in the nearest adjacent vacant place for a house. The distance was unusually great, however, because the rain waters had carved a channel between the two houses and made the closer area unfit for a building.12

By the mid-1970s, an outer circle of houses was forming around the principal circle of houses in Escalvado. In almost every case, female kin by all female genealogical linkages to the women living on the principal circle were constructing these houses. One temporary exception occurred in 1975 when an incompetent male in his thirties married a widow with no close female kin and put up a house behind his uterine sister’s. By 1978 they had moved to the location of her nearest but distant kin. All exceptions to uxorilocality were explainable. For instance, Jack Popjes’ [II.B.3.a] previously adopted "kin” moved their house position next to his during the construction of the new Escalvado village in 1969.

[III.E.2.e.(2)] Parallel-Cousin Matrilateral Arc

The neighboring houses on either side along the village circle next to the house of uterine sisters [III.E.2.e.(1)] are likely to be inhabited by these sisters’ first or second cousins, or by cousins who are genealogically even further away (Figure 23). In theory, and if name-set transmission and Formal Friendship systems (and others) do not alter the terminology, all the women of the same generation in these consanguineally related houses refer to each other as “sister” (tőy). They refer to the women of their first ascending generation as “mother” (nŕŕ), to the women of the second ascending generation or higher as “grandmother” (tůy), to kin of the first descending generation as “child” (khra), and to kin of the second descending generation or lower as “grandchild” (tŕmtswč). These sisters “brothers” descendants live elsewhere. All the female kin in these adjacent houses extended along an arc of the village circle are related through only female genealogical ancestral linkages. These women say they live in the same “longhouse” (ikhre-růů: house-long) (Glossary), their mothers’ longhouse (më tůltswč ikhre-růů: pl. mothers’ house-long), which is not one long house, but rather a row of a few or many separate houses. Apparently, the Kayapó use the same term kikrę ry (house long) for the same purpose (Verswijver, 1983:304). Married male relatives of these women do not sometimes come back with their families to live permanently in their sisters’ longhouse for political reasons, as they do among some other Eastern Timbira tribes (Lave, 1979:42-43), including the Apanyekra.


In 1971, the longest Escalvado village longhouse contained 13 houses, while the shortest longhouse consisted of one house. The most genealogically distant Asisters” in this long longhouse are fifth cousins, and they call each other "sister.” These longhouses have no names, and they are seldom referred to as longhouses. The long Escalvado village longhouse (houses BB through NN [house CC omitted] in Figure 24) is special, however, because the female ancestor, Amyi-yakhop, from whom all the women in the longhouse are descended in matrilines (Glossary) was an Apanyekra woman, so in traditional intertribal terms all the women of this longhouse are Apanyekra (Apŕn-khwčy-yę: piranha-woman-pl.). In the Rancharia village of the Apanyekra, there were 11 longhouses (Figure 25; see also Figures 26, 27).


Nimuendajú (1946:84) is right when he writes that a “house community” fails to “constitute a definite economic unit.” He was referring to several closely related hearth units, probably related through women no more distant than first or second cousins. Among the Krĩkatí such units have grown to sometimes include a woman's brother's family and other political adherents, so Lave (1979:40) calls them “domestic clusters.” I agree with her that these Krĩkatí units “are more integrated and solidary units than such groups were among the Timbira [Canela] of 1935.”

Canela longhouses are not necessarily economic, jural, ritual, or political units, though they are exogamous with very few exceptions, because their members recognize each other as kin. While never really political in purpose, a longhouse may be economic if identical with a hearth unit. A longhouse may be jural (if not too long) in the sense that relatively distant matrilateral kin may assist or take the lead in conducting inter‑extended family hearings (usually intra‑longhouse ones) and in making extended family payments resulting from such hearings [III.D.3.e.(5)] (Nimuendajú, 1946:84, 161). The making of payments from several hearth units away happens in the case of a man who wants to be prominent in the village or when a family lacks a close uncle to represent it in the plaza or at a hearing [III.D.2.d.(1).(a)]. A longhouse may be ritual when it is comprised of the same houses with a particular haakhat [III.C.8.a]). During the 1970s, Canela individuals were vague about the differences between a longhouse and a haakhat (Glossary). The latter is only ceremonial and basically matrilineal, and occurs discontinuously around the village circle. The former is matrilateral and distinguished by its strong tendency toward exogamy and by its traditionally fixed position on the village circle through time or from village to village in relation to the cardinal points. Continuing the contrast, arcs of longhouses are continuous around the village circle. (These distinctions and their complexities will be clarified in a later publication.) J. Melatti's (1979a:51) description of the “residential segments” among the Krahó is almost the same as for Canela longhouses. I am not sure, however, to what extent Krahó residential segments are formed just by female kin who are related through just female ancestral linkages. The Apanyekra have allowed exceptions to this rule to continue and maybe the Krahó do also.


Members of longhouses who live within two to four houses from each other along the village circle tend to place their farms in adjacent positions along one of the several gallery forest (Glossary) streams, but longhouse cohesion does not go much further. Although longhouse kinship is one factor in a member's choice of her farm plots location, other factors are important too: male concerns, such as personal age‑set loyalty, political alignment, and agricultural preferences. The man who is the leader of a farm community is important to that farm community's female members from several hearth units, so that a husband's adherence to a particular age‑set leader, or a tribal leader, is also a factor in choosing their farm's location. (Women own farms.) Potential chiefs, when the opportunity arises, tend to found different farm settlements, any of which may be dependent or independent of the principal tribal village or villages. This dispersal of leadership has occurred a number of times in Canela history, and the study of each community's composition provides clues to the sources of political power in the tribe. In 1987 the Canela had 14 separate farm communities (Map 3) dependent on one central Indian service supported village (Escalvado). One chief and three ex-chiefs led four of these farm communities, and I have to assume that the remaining ten communities were led by young and aspiring potential chiefs.

[III.E.2.e.(3)] Cross‑Cousin Across-The-Plaza Kin

Besides the matrilaterally related kin, there are also “across-the-plaza” (Glossary) (i.e., through a male ascendant) kin who live in longhouses “on the other side of the village” (khrĩ i?në rum: village over‑there side). These relatives can be descendants of a person's father's sister, mother's father's sister, or father's father's sister, though the latter case is rare. They live in a person's “grandmothers' longhouses” (më puptswč ikhre-rũũ: pl. grandmothers house-long). In a number of instances the terminology of address and reference used by individuals between the two across-the-plaza longhouses has been held consistent with Crow-III equivalence rules for three generations: father's sister/brother's daughter, father/son, mother's brother/sister's son, etc. (Reciprocal across-the-plaza relationships, of course, also exist.) These relatives live in a person's “grandchildrens' longhouses” (më tŕmtswč ikhre-rũũ) (Figure 43).

Such kin form the across-the-plaza (kŕŕ ?në rum: plaza, its-further side) part of ego's kindred, which is based on cross‑cousin instead of parallel-cousin relationships. Male ego should not marry into such relatives' longhouses or have sexual relations there, and female ego should not marry men from the longhouses of these relatives or have sexual relations with these men. Women in such across-the-plaza relationships, if they have not forgotten the relationship of the lower generations over there, often treat such men as a joking relative [II.D.1.b.(3)]. Such across-the-plaza (“patrilateral”) kin contribute food and services along with ego's matrilateral kin particularly in individual rites, though I found such across-the-plaza kin helping to wrap shuttlecocks in the Corn Harvest festival of 1979. This particular across-the-plaza cooperation, among kindred, though minimal, was notable because the Corn Harvest festival ownership is haakhat-oriented and therefore this Corn Harvest festival ownership is matrilineal and not bilateral [III.C.8.a]. Young Canela are becoming increasingly forgetful about such distinctions. I could not have discovered such refinements of the traditional system without the lengthy debates of much older women and men among my experienced research assistant council members [Pr.2.b].


There are many reasons why the Crow‑III (Glossary) (or “parallel transmission”) terminological patterns are so often broken by family members in the first, second, and third descending generations. Name‑set transmission, Formal Friendships, Informal Friendships, and contributing-fatherships are all factors in the alteration and disruption of the basic and underlying Crow III-like patterns. Among the Krĩkatí, Lave (1979:23) found that “naming relations take precedence over all but ceremonial trading partnerships and formal friendships as criteria for assigning customary terms of reference.” (The Canela and Apanyekra do not have ceremonial trading partnerships.) She pointed out that a man could call his uterine sister tuire (Canela tůy-re: “aunt”) if his father's sister had named her. This same precedence of the naming over the genealogical system takes place among the Canela and Apanyekra. Similarly, the older Kaapęltůk called one of his sons Hŕŕpin-re (Formal Friend, dim.), but his son called him intsũũ (my-father).

J. Melatti (1979a:57–58) writes about this same “confusion” among the Krahó: “...the application of the terminology gets more and more snarled as one gets farther away from ego's genealogically closest kin.”


This terminological disruption may have been considerably less in pre-pacification times two centuries ago, when some of the Timbira tribes may have lived in groups and even villages of between 1000 and 1500 in population. In those days, name-set and Formal Friendship relationships may have been made mostly with more distant relatives (name-set) and with nonrelatives (Formal Friendship) to extend the individual's personal support network throughout the larger groups.

[III.E.3]  The Affinal Terminological System

The Canela affinal system is less well known by ethnologists than the consanguineal one, because Nimuendajú (1946:105) covered it less completely. The Krahó affinal system is also less well known than the consanguineal one (Melatti, 1970:151–156, 1973, 1979:46–79). However, the Krahó affinal system should be very similar terminologically to the Canela one because the two tribes were so close geographically (Map 4) in earlier times and are so close linguistically now. The SIL considers their speech as one language [Ap.4]. Thus, kinship specialists will find new aspects in the affinal system presented here.

The Canela affinal system appears to be more extensively developed because the Canela were less deculturated and more traditional at the time they were studied than when the other Eastern Timbira tribes were studied. Moreover, the Canela were living in a larger single community. While the across-the-plaza consanguineal terminology of the Krahó was presented by J. Melatti (1970:151–156, 1973:14), the corresponding across-the-plaza fully extended affinal terminology probably no longer existed when he was there. It is also in this extended affinal terminology that Scheffler and Lounsbury's parallel transmission equivalence rules fit the data better than the Crow/Omaha-III systems. (I hope to present such materials in a later publication.)


Canela affinal terminology is predicated on four terminological distinctions and four kin category usages. Terminological distinctions are made between affines in the following relationships: (1) “in-house”/ “out-of-house”; (2) same-sex same-generation; (3) same-sex adjacent-generation, and (4) opposite-sex adjacent-generation. The kin category usages are that (1) avoidance terms can be used to replace classificatory spouse terms; (2) opposite-sex same-generation individuals are classificatory spouses, and they are often found in notably extended kintypes (Glossary); these relationships are widely spread (completely apart from the primary (-mpey) affinal system generated through marriage) in secondary (-kahŕk) affinal systems generated through affairs or courtships; (3) secondary consanguineal terms (-kahŕk ones) are used as affinal terms for the aged, and (4) secondary consanguineal terms are used to conform certain of ego's affinals to their spouses, who are ego's kin. These affinal terminological distinctions and kin category usages are used for both female and male ego's spouse's matrilateral longhouse kin and for ego's spouse's across‑the‑plaza grandmothers'/grandchildrens' longhouse kin. This affinal terminology follows the consanguineal terminological patterns of the circular uxorilocal village.

[III.E.3.a.(1) ]“In-House”/“Out-Of-House” Distinctions

The Canela make a clear distinction terminologically between individuals born into an extended consanguineal family and their affines married into the same family. The exception is male ego's wife's father, who by the time he has married children is terminologically “in” the house for the younger generations. Because the individuals born into a family come from “in” the uxorilocal longhouse while the individuals married into a family come from “outside” this longhouse, the expressions “in-house” and “out-of-house” are used to emphasize this distinction (Figure 28). The terminological basis for this distinction is that all of the out-of-house individuals can refer to their in-house affines with terms that include the basic morpheme pree. No reciprocals of these in-house terms, namely, the out-of-house terms, have pree (Figure 28). The term pree is associated with “tying” or “binding,” so that i-pree (I-bound) can be translated freely in this context as “the one I am bound to,” which would be quite appropriate for an “out” woman-in-law and an “out” man-in-law to say about their “in” in-laws. (Ego's mother's brothers' son's children are not included in this “in-house”/“out-of-house” context, because they are born into still another longhouse; thus they are too far away genealogically, as are many other kintypes.)

J. Melatti's (1979a:69-71) “marriage prestations” for the Krahó can be applied to the Canela; and when they are, they flow from out-of-house to in-house lines terminologically. Such voluntary marital gifts from a man's kin to his wife's kin (in contrast to formal payments [III.F.5]) from a woman's kin to her husband's kin are taken less seriously among the Canela.

[III.E.3.a.(2)] Same-Sex Same-Generation Relationships

Considering the “in-house”/“out-of-house” distinction between same-sex same generation affines (♂Z over ♂W and ♀B over ♀H), a woman's male relatives (in-house: born in her house) are referred to as pree (WB) by her husband, and in return, her male in-house kin refer to her husband as piyőyę (out-of-house: born out of their house, ♂ZM); and similarly, a man's female relatives (in-house: born in his natal house) are referred to as pree (HZ) by his wife; they in return refer to her as tswčyyę (out-of-house: born out of their house, ♀BW). All the out-to-in-house terms of reference of either gender, except avoidance terms and spouse expressions, have just one stem, pree [IV.B.1.h.(5)] as the principal part of the term. The reciprocals, in contrast, the in-to-out-of-house terms (except for the avoidance terms and spouse expressions) have two stems: one for each sex (Figure 28). This simplicity is impressive when the affinal reference terminology is viewed according to this dichotomy. Affinal out-to-in-house address terminology differs for each sex (Table 10), unlike the reference terminology.

This in-house/out-of-house distinction is important in the translation and use of the term “brother-in-law.” For instance, in the myth in which a Canela youth takes fire from the jaguar; or in the myth “Tetswa-re,” in which a man kills his brother-in-law with his fire-sharpened and pointed tibia (shinbone), it is the “out”-brother-in-law who is carrying out his revenge on his “in”-brother-in-law, his wife's relative. The “in”-brother-in-law has almost necessarily been arbitrary and inconsiderate to his “out”-brother-in-law as an outsider to his matrilateral kin group. Some translations of Northern Gę myths do not distinguish between these two different kintypes and terms for “brother-in-law.”

[III.E.3.a.(3)] Same-Sex Adjacent-Generation Authority Distinctions

An in-house woman of a senior generation (mother-in-law) is usually distinguished (reference) as pree-kęy by her out-of-house woman-in-law (or daughter-in-law) (tswéyyę), and an in-house man (including wife's father) is usually distinguished (reference) as pree-kęt by his out-of-house man (piyőyę) or son-in-law.

The above in-house terms of reference and address differ, while out-of-house terms of reference and address are the same as for same-sex same or adjacent generation affines (Figure 28). The terms kęy and kęt indicate seniority, and the suffix is honorific and not the plural in this case. The morpheme tswč has several referents; one of them means to be pierced or crippled. Research assistants gave no meaning for the out-man-in-law term, but I know that or - is the word for garden produce (manioc, rice, yams, etc.) and pi can indicate the past tense. The out-son-in-law (or out-brother-in-law) does a lot of work in the farmplots for his wife's family.

The “out”-in-law person (both sexes), especially the male, is held in a position of subservience and service by his “in”-in-laws. This “out”-in-law is responsible for much of the economic support (farming, hunting, contacts with backland merchants, etc.) of his wife's family. He works together with the other “out”-in-laws who are husbands of his wife's sisters and “sisters,” usually under the direction of his wife's father, their “in”-father-in-law. No special terms exist for wife's sister's (or “sister's”) husband, or for husband's brother's (or “brother's”) wife.

[III.E.3.a.(4)] Opposite- Sex Adjacent-Generation Avoidance Terms

Opposite-sex adjacent-generation terms both break and keep the in-house/out-of-house patterns, depending on the alternatives used (Figure 28). Where this pattern is kept, pree is always an in‑house term. Other in-house terms that imply avoidance, in which pree is not part of the term, are the most used terms among the Canela. In this alternative system, which implies avoidance, an “out” man-in-law refers to his “in” mother-in-law as hŕtswčyyę (wife's mother (in-house)), and a man (in-house) refers to his son's wife (out-of-house) as hŕtswčyyę. A woman (in-house) refers to her son-in-law (out-of-house) as wawč (woman's daughter's husband); a woman (out-of-house) refers to her father-in-law (in-house) as khră?tumyę (husband's father). These avoidance relationship terms were used traditionally for only adjacent generations, while the spouse terms were used for further generations. This was a realistic practice when young women and men in their mid-teens were supposed to have sex only with much older people [III.A.2.p], preferably of the postmenopausal age [II.B.I.e]. Now that they do not have such sex relationships, the avoidance terms may be used also for the second and even the third ascending generations, or the consanguineal grandparental term for age (tůy/kęt) may be used instead.

In the term hŕ-tswčy-yę, the morpheme or -tsŕ means to be “hurt” or “sick,” tswč has been discussed [III.E.3.a.(3)], y indicates the feminine gender, and has been discussed [III.E.3.a.(3)]. The term tswčy can be associated with tswčyyę, the “out”-woman-in-law term.

Research assistants said that the in wawč means “to beg” and that wa is the first person singular pronoun or adjective, “I” or “my.” The morpheme may also be the third person singular of -tswč: I-tswč (lst), a-tswč (2nd), (3rd).

The term khră-?-tum-yę (head-its-old-honorific: honorable old head) may be the most obvious and easiest translation of all the affinal terms, and thereby lends some encouragement to this kind of an investigation for the other less obviously translatable terms.

Ego's immediate avoidance affines carry out a full avoidance relationship with her or him, which is notable to the outsider in its completeness. More distant classificatory avoidance affines can be treated with considerably less formality.

[III.E.3.a.(5)] Avoidance Terms As Alternatives To Spouse Terms

These avoidance terms, hŕtswčyyę, khră?tumyę, and wawč, are alternatives to spouse terms when ego and alter are genealogically sufficiently removed from each other—sometimes three terminological links away from ego in a kintype (Glossary) (WMM), and surely four links away (WMBD). These avoidance terms are never alternatives for spouse terms in the kintypes “wife's mother” and “husband's father,” and their reciprocals, because ego and alter are too close (only two links away from each other; see Glossary: further-link kin). Similarly, since same-sex siblings are equivalent in Crow-III and in Canela ideology (Figure 38),”wife's mother's sister” and “husband's father's brother” are also too close. Avoidance terms must be used here as well.


Wife's mother's mother, however, can sometimes be a “spouse” now and was more frequently a“spouse” in earlier times (Figures 31, 32). In Canela theory wife's mother's mother's mother is also a “spouse” but is more likely a secondary grandmother through age. If, however, much joking occurred with her in earlier times, she could have been addressed by personal name, as may occur between “spouses.” Sexual relations were not necessary to create “spouses” between generations, just joking. Wife's mother being an avoidance woman in “parallel transmission” reductions, prevents all the women in a man's wife's ascending matriline from being reduced to husband's wife, or just “wife.” Thus, these affinal relationships are expressed more appropriately through “parallel transmission” (Scheffler and Lounsbury, 1971) than through Crow-/Omaha-III. However, I do not want to emphasize any formal system of reduction/extension equations as the Canela solution, because I basically agree with Maybury-Lewis (1979:308–311); the application of these formal systems of equations should, nevertheless, be mentioned.


Spouse terms used as alternatives for avoidance terms first became apparent to me in the terminological practices of the name-set transmission system. A name-giver has to decide whether to refer to his name-receiver's wife as “wife” or as “avoidance woman,” depending on whether he has sexual interests in her (traditionally he can) or just wishes to be on speaking and joking terms with her (“wife”), or whether he prefers to carry out a serious, responsible relationship with her (“avoidance woman”).

It is noteworthy that my naming‑uncle, the younger Mďďkhrô [I.G.10], chose to call my first wife, Mary Jean, by her Canela name (a “spouse”-address practice), while he chose to call my second wife, Roma, “avoidance-woman” (hŕtswčyyę) instead of Hô-mă, one of her Canela names. His choice of address reflected how he wanted to behave and actually did behave toward them, respectively: friendly and jocular or serious and respectful.


A similar decision (avoidance woman/man or classificatory spouse) has to be made for an across-the-plaza affine (Glossary) (e.g., HFZDS, HMFZDS, WMBD, WMBDD). For, instance, a man may call his wife's mother's brother's daughter, “wife” (primary); but he may also call this same woman's mother (his WMBW) “wife” (secondary) as well. Any woman in a man's wife's mother's brother's daughter's matriline (below her) is his classificatory wife (a primary one: -mpey) (Glossary) (Figures 29, 30, 33, 34). One Canela calls two such distant classificatory spouses (a woman and her daughter) “wife,” but this is usually not done. He “does wife” (to prő: make/do wife) with either one or the other. Once he “speaks” (kakhôk: a euphemism for sex) to one woman in such a matriline, her mother becomes a secondary (-kahŕk) (Glossary) avoidance woman and her daughter becomes a secondary daughter. (For clarity, in the matriline from top to bottom, male ego has an avoidance woman, a wife, and a daughter, all secondary ones.) But years later, if this secondary daughter becomes very attractive, he might have sex with her too. In such a case, he does not start calling her mother “avoidance woman” because once he has had sex or has joked with her extensively, though much earlier, he will always consider her a “spouse.” Thus, he continues to call her by her personal name implying the possibility of having sex with her as well as her daughter.

In summary, a woman's across-the-plaza “nieces” are her husband's “spouses” (WMBD, WMBDD, WMMBDD, etc.) as are her across-the-plaza “aunts” (WFZD, WFZDD, WMFZDD, etc.). A man's across-the-plaza “uncles” are often his wife's “husbands” (HMFZS, HMFZDS, etc.), as are his across-the-plaza “nephews” (HMMBDS, HMMBDDS, etc.). (See Figures 33 and 34.)

[III.E.3.a.(6)] Opposite-Sex Same-Generation Affines

These affines are “spouses”: “other spouses” (më ?prő ?nő: pl. wife other; më mpyęn ; pl. husband other), the Canela say. Moreover, they are considered to be primary (-mpey) spouses in both the married‑to spouse's longhouse and in the married-to spouse's across-the-plaza longhouses (Figures 29, 30, 33, 34). (The distinction between primary and secondary kin is very complex and will be discussed more fully in a later publication.)

It was helpful in learning affinal kinship to reckon from an outsider to the tribe as ego (e.g., a Krahó married to an Apanyekra), through this person's tribal spouse, and on through this spouse's one-linked and further-linked kin to any alter, since the outside person could not be related to such an alter except through her or his tribal spouse. Because I already knew the tribal spouse's consanguineal relationships, the outsider's affinal ones were easy to reckon through this technique, even though name transmission or Formal Friendships had caused changes in the system between the outsider's spouse and her or his alter. This use of an extratribal, Eastern Timbira person was a valuable technique for unraveling the several different relationship systems, which were so thoroughly interwoven among the Apanyekra and the Canela.


The anthropological expression “potential spouse” is not as accurate for the Canela situation as “classificatory spouse” or simply “other spouse,” as the Canela call it. This is because while sexual relations are traditionally allowable with all “other spouses,” although rare these days, most of the across-the-plaza “spouses” are not likely to become candidates for marriage even if the married-to spouse dies.


If a woman dies before all her children are grown up, a uterine sister, or a close longhouse “sister,” becomes a strong candidate to be the widower's next wife. The sororate is preferred. Then the widower can remain in the same house with his children. His deceased wife's further-away longhouse “sisters” (his “other wives”) and her across-the-plaza female kin (also his “other wives”) are unlikely candidates.


A man may have sexual relations with his wife's uterine sister and behave with her in the joking manner characteristic of classificatory spouses even in front of his wife; but the fact of actual sexual intercourse with her must be hidden these days. This is not because she could object to the liaison (such liaisons were the custom), but because her feelings and sense of pride would be hurt if the tryst became public knowledge. Exactly the same relationship exists for a woman with her husband's brothers and “brothers.” My bashful brother, Hŕwmrő, used to play sexual games with his wife's (Mĩĩ-khwčy's) unmarried uterine sister, Khęt-khwčy, forcefully tumbling her onto a mat and trying to suck her breasts in sight of everybody in the house, including his wife who laughed and urged him on. It was an unequal wrestling match, with Khęt-khwčy putting up a good fight, protesting and shouting bad names in fun.

One of my enjoyable diversions as a participant observer was the harassing (a joking relationship) given me by my “spouses” as I walked up the boulevard between houses BB (my brother's) and NN (Figure 24) when my wife Roma was away from the tribe. This long longhouse was Roma's longhouse by virtue of her being adopted into house NN in 1970. Sometimes, by the time I passed house JJ, the harassing was quite intense as many of her “sisters” came out of their houses and joined the chorus warning me not to “speak” to any woman or they would tell Hô-mă, their “sister,” when she returned.

For some reason that I never fully understood, this mock “harassing” was more intense and direct among the Apanyekra in whose villages my wife had never visited and was not adopted. The classificatory spouse harassing worked through my Apanyekra brothers' wife and “wives” and because any woman not a kin or a Formal Friend to me is an “other wife.”

[III.E.3.a.(7)] Secondary Consanguineal Terms

The Canela relationship system makes frequent use of the grandparental or grandchildren terms. This usage masks a number of other practices that would otherwise be easier to discern.

When a person has a much older affine, such as 40 to 50 years older, she or he usually refers to her or him as “grandmother”/“grandfather” (or “aunt”/“uncle”: tůy/kęt). These terms are said to represent secondary relationships (-kahŕk) rather than primary ones (-mpey), however, and there is no thought that these elderly affines have become kin. These days, the formerly extensive intergenerational (often two) extramarital relationships (prescribed for the mid‑teens) are no longer practiced. Thus, some affines that are two generations and most affines that are three generations above ego are referred to and addressed as “grandparents” though formerly they were sometimes “spouses.”

The affinal kintypes that are most resistant to this kind of acculturative transformation (from affinal to consanguineal terms) are the all feminine -pree ones. A woman will always call her husband's mother's mother toktůyyę (“in” sister-in-law) or propęękęy (“in” mother-in-law), and her husband's mother's mother's mother as well. (Four generation matrilines are often found in Escalvado, especially with the current practice of earlier marriage of women.)

Another application of consanguineal terms in affinal situations occurs when ego's affine of either sex is given the consanguineal term corresponding to her or his married‑to spouse, who is ego's kin. For example, ego may call her or his father's brother's wife “mother” because she is married to a man that ego calls “father,” and ego may call her or his father's sister's son's wife “mother” because ego calls her husband “father.” Affines being called kin terms because of their marriage is merely one terminological alternative, because ego may have good reasons for calling such affines by several other terms. She or he may already be related to them in other ways.

If a man calls his mother's father's sister's daughter's husband “grandfather,” this may be (1) because he recognizes a significant age difference, (2) because he calls this man's wife “grandmother” (his MFZD), or (3) because he knows this man is consanguineally his grandfather through some other genealogical reckoning.


There are essentially three kinds of affinal terms: (1) pree terms (primary and secondary, including their reciprocals); (2) spouse or avoidance terms (immediate and classificatory, primary and secondary); and (3) secondary consanguineal terms (for age and by marriage).

The understanding of why the Canela use a particular term (consanguineal, affinal, Formal Friendship, or otherwise) and the prediction of which term will appear as alter at the end of any particular kintype are mental processes that are made increasingly difficult for the anthropologist as the kintype becomes longer. These difficulties come from within the affinal system as well as from outside of it. On the inside, beyond two and certainly beyond three links away from ego, the individual speaker can make several legitimate choices, as are listed in the ideal affinal terminology presentations (Figures 31-34). From outside the affinal system, other relationship systems compete with the affinal one, again offering the individual speaker legitimate choices. The competition comes especially from the name-set transmission and Formal Friendship systems but also from the other terminological relationship systems.

A complicated and extensive terminological reference system is employed for affines of affines (using terms not presented here) so that when ego is speaking to an affine about this affine's affine, this other system of terms must be utilized.


The honorific pronoun is used instead of the regular ka and for the second and third persons, respectively, address and reference, singular and plural. Only spouses and “spouses” do not use . Instead of ka ha mő (you will go) and kę ha mő (he will go), these certain affine says yę ha mő (you, my affine, will go) and yę ha mő (he, my affine, will go). is used in the same way between all Formal Friends, and the honorific and avoidance behavior practiced between such affines and Formal Friends are similar, including a range in behavior from mild respect to complete avoidance.


Name-set transmission (♀F“Z”/♀“B”D, ♂M“B”/♂“Z”S) (Nimuendajú, 1946:77–79, 109–111) is complicated and can only be covered in a summary manner here (Figures 35,36). I collected over 200 recorded cases of name-set transmission, which will be discussed in a future publication.

A person may have from one to well over a dozen names in her or his name-set bestowed by one name-giver, but most young people today can remember only two or three of the names they are supposed to have. The names in a person's name-set are not semantically connected. Nicknames exist but do not have the emphasis placed on them that the Krĩkatí place on theirs (Lave 1979:19). Most names are sex-related, but a few are held by both sexes, like Ku?tŕŕ (locust [the tree]).

Following the principal model, a woman gives her set of names to a classificatory brother's daughter, and a man bestows his names to a classificatory sister's son [IV.B.1.c] (cf. Nimuendajú, 1946:78). Thus, women follow a pattern that is consistent with Omaha kinship, and men follow Crow. Actually, names are passed on to a variety of kintypes, but the principal ones are to the parallel first cousin's child (through all female links) and to the uterine sibling's child (Figure 37). (Compare J. Melatti,1979a:59.)


Canela name-givers call their name-receivers i-túwa-re (my young-one dim.: my young self) or itúwa regardless of sex, while a female name-receiver calls her name-giver tůy and a youth calls his name-giver kęt-ti or kęt-re, depending on the physical size of the addressed person. The female term tůy is unique to name-set transmission practices because in other situations (consanguineal or affinal) tůy-ti or tůy-re has to be used by ego, depending on the size of the addressed person. In contrast, the morpheme kęt- never appears alone (J. Melatti, 1979a:54). The Canela have not developed two terms related especially to name-transmission as have the Krĩkatí: inchigrunto (mother's name-receiver) and inchungrunto (father's name-receiver) (Lave, 1979:23).


The principal name-exchange ceremony does not take any particular form; the pair of opposite-sex “siblings” or uterine siblings simply agree to be name-exchangers with respect to one of their children well before the time of their birth [II.D.1.b.(l)] [IV.B.1.c]. If the contracted man does not arrive in the house of his “sister” within a few days after the birth, some other “brother” is likely to take his place. Completing the agreed-to exchange is important (Nimuendajú, 1946:78), but it is difficult for a woman to refuse another “brother” when he is insistent. The sense of compulsion for reciprocity between pairs of name-givers is clearly not felt as strongly as among the Krĩkatí (Lave, 1979:25).

There is only one public ceremonial act involving name-giving, which is not the more conspicuous and better known (by outsiders) name-changing ceremony [IV.A.5.e.(3)]. In this name-giving act, a male name-giver makes the high cry of his particular Regeneration moiety in the door of his “sister's” house at sunrise (Nimuendajú, 1946:85), the morning after she has accepted this “brother” to be the name-giver of her recently born male. The Canela are losing this practice, while the Apanyekra are retaining it. No similar rite exists for female infants.


In a relationship referred to as amyi‑půtŕl (self taking-over: taking each other over), a distant sister and brother agree to give a name-set to one of each other's children of the same sex as her- or himself. A man does this especially with his father's sister's son's daughter (♂FZSD=Z) or his father's sister's daughter's son's daughter (♂FZDSD=Z), or even with a distant female parallel cousin. A man's name-set exchanging with a uterine sibling, however, would not be referred to in this way because she would already be so close (personally and in terms of “blood”; Figure 38) that an amyi-půtŕl relationship would not make them any closer.

Research assistants said that a man has to choose between exchanging names (amyi-půtŕl) or committing incest (to aypré: do incest) with a distant sister (turning her into a “spouse”), such as his father's sister's son's daughter. She is two longhouses away genealogically, and therefore necessarily a secondary (-kahŕk [Glossary]) sister. A man also has to make this same choice with a “sister” related through some male links (♂FB“D”). His longhouse “sisters” (related through all female linkages) are considered primary (-mpey) ones, and are more conspicuously “siblings.” Thus they are more difficult to commit incest with. However, incest may eventually happen, which would convert the matrilines (mother-daughter-granddaughter) to other relationships and disrupt the continuity from generation to generation within the longhouses.

An unrelated outsider adopted into a tribal family is seen to be like an amyi-půtŕl sibling of some Canela, not like a uterine one. I put a name on my name-exchange sister's (Te?hôk's) son, Ku?tŕŕ-tčy (locust-tree tough) just after he was born in 1960. Then, Te?hôk put her own name (Te?hôk-re) on my step-daughter, Tara, when she visited the tribe in 1970. When I was leaving the Canela in 1971, my wife Roma was pregnant. One of Roma's “brothers” (the younger Tŕŕmi, a parallel first cousin of hers), told me in a simple verbal exchange of a few seconds to name his “sister's” son, if a boy, Ku?tŕŕ-khre (locust its-hollow). To honor that exchange, my son Myles was subsequently christened in 1972 in Washington, D.C., with that name as one of his middle names.


One declared purpose of a name-exchanging agreement is to maintain and even increase the number of a person's significant relatives, and therefore, to broaden the person's social support base (Ladeira, 1982), according to the research assistants. I believe that some of them were wrong when they said that it was mostly uterine siblings who exchanged names in earlier times. Maybe they were remembering their childhood days when the tribe was smaller or were recalling reports of the early 1800s when the tribe was in social disarray. Before contact with Western culture, however, Timbira tribes were much larger, maybe between 1000 and 1500 or more. Thus, even more so than now, it must have been advantageous to maintain as many kin as possible and to have some as far away as practical.

In a 1970 study of 198 uterine cross-sex siblings and cross-sex cousins who had agreed to exchange names for their children and had carried out at least one naming, close to 25 percent of the relationships were between uterine cross‑sex siblings and approximately 75 percent were between cross-sex cousins (“siblings”). Thus, I take name-exchange between cross-sex “siblings,” not cross-sex siblings, as the traditionally preferred form.


Although many of the names are traditional, it is possible to create new names for a particular name-receiver. The new name is usually made up from an event the name‑giver experienced just before the time when the name-set was to be given. For instance, a woman who had just chastised her husband by ripping up his side of their platform bed (so he could not sleep with her), named her name-receiver Pŕlrŕ-rę (platform-bed torn-up). This name probably will become one of the names of the name-receiver's name-set and so will be transmitted to a name-receiver a generation later and eventually will be known as being “traditional” (mam mënkętyę nkaakaa tsŕ khôt: early ancestors' breath thing following) The trend is toward creating these situationally derived names rather than choosing traditional ones.


A name-giver designates one name of the set of names she or he gives as the name to be used publicly by each name-receiver. A name-giver may have 3 to 10 name-receivers. The name-receiver, however, receives the whole set of names to pass on and all the potential ritual rights that go with the set, although any one of the name-giver's name-receivers (depending on the ritual) may be designated by the name-giver to carry out the roles of the ritual that are connected with the name-set.

Thus, for every name-set performer, there are other persons of the same name-set who are not involved in the performance of the name-set held ritual. (Ceremonial privileges are connected with certain name-sets rather than with certain names; cf. Nimuendajú, 1946:167.) If the ritual-designated name-receiver dies, the chief name-giver of the name-set's corporate group (usually the oldest) selects some other person from among his various name-receivers to carry out the roles of their ritual. In some cases, all name-receivers and all name-givers of the name-set holding group (there may be three or four generations involved) perform in the ritual together, as for the cutting out of the Kô?khre log's trough in the Closing Wč?tč festival [IV.A.3.e.(3)] (Nimuendajú, 1946:168). In the two Tsů?katę-re name-set transmission lines in the 1970s, the right to carry out the Tsů?katę-re roles in the various festivals and rituals (Pŕlrŕ, Closing Wč?tč, Pepyę) falls upon just one individual in each of the two name-sets: Kôham in Mďďkhrô’s line (Nimuendajú, 1946:165, fig. 10:6) and Rărăk in Pŕnhi's line (Nimuendajú, 1946:165, fig. 10:5). Such name-holding groups like the Kô?khre log's trough-cutting group are not formally split into older and younger halves as among the Krĩkatí (Lave, 1979:24-26), and their development is minimal among the Canela and their festival appearance infrequent.


A child's names are rarely changed, which, however, may occur only after a Pŕlrŕ log race [IV.A.5.e.(3)], which occurs just once or at the most twice a year. (This happens if a child has to be transferred to the care of a new name-giver.) A man who knows the traditional song (usually the town crier [the hŕŕpôl-katę: urging-on master]) sings the chant standing in the center of the plaza with his back to the name-giver who stands beside the name-receiver, all of them with their backs to the rising sun (around 7:00 to 8:00 AM) (Plate 51d). Special visitors from the cities may be named (honored) in a similar manner at any time, including in the afternoon or evening and without the benefit of a Pŕlrŕ log race, and the ceremony of being made a ceremonial chief (Glossary) is similar to the Pŕlrŕ name-changing one.

Nimuendajú was named this way and so was I, because we were outsiders. However, Canela and Apanyekra babies do not receive their name-sets in this ceremony, as Nimuendajú (1946:110) suggests, although some artifacts do.

[III.E.5]  Formal Friendship Terminology

Formal Friendship (Nimuendajú,1946:100–103) unites two people in a bond of honor, respect, and especially of mutual assistance and protection. If a person suffers in almost any sort of way, her or his Formal Friend (Glossary) has to help immediately, often undergoing the same suffering [II.F.4] [III.A.3.b.(1).(e)] [III.B.1.c.(2)]. This relationship, involves maintaining a respectful social distance and experiencing great shame (Glossary) when infractions against the relationship occur. Formal Friendships exist between two women, two men, and a woman and a man. A person has a number of Formal Friends, but only one of these Friendships is intense (including complete avoidance) while the other relationships vary in seriousness to a minimal point where mild joking may occur. Individuals united by Formal Friendship are referred to as being khritswč. A woman is addressed by her Formal Friends of either sex as pintswčy and a man as hŕŕpin.

(There are primary (-mpey) and secondary (-kahŕk) Formal Friendships, which will be discussed in a later publication, as will the description of the minor and limited system of terminology extending to the near kin of a person's Formal Friend.)

I use “Formal” rather than “Ceremonial” Friend to emphasize that the relationship exists in daily life outside of festivals and ceremonies (cf. Carneiro Da Cunha, 1978:74-94; Da Matta, 1982:87–99; Lave, 1979:17–18; J. Melatti, 1979a:74–76).


The honorific personal pronoun is used between Formal Friends instead of the regular ka and in the second and third persons, respectively, singular, dual, and plural. In addition, Formal Friendship terms apply when ego is referring to her or his Formal Friend, when ego is talking to a person about this person's Formal Friend, and when ego is addressing a Formal Friend but referring to her or his (ego's) other Formal Friend. The complexity of this system, which includes many special terms not presented here, is increased because of combinations in the terminology of the two kinds of -related persons: certain affinals and all Formal Friends. (These systems will be fully described in a later publication.)

[III.E.5.b] INITIATION OF PRINCIPAL FORMAL FRIENSHIPS Formal Friendship is initiated in a number of ways, but the most important manner is through the performance of the Ntęę ceremony in which a person chants a traditional song in front of the house of a pregnant woman and delivers a present to her relatives (Nimuendajú, 1946:100). When the baby is born, she or he becomes the Ntęę performer's Formal Friend. This latter person does most of the helping and honorific work for the child, such as body decoration with falcon down, until the child grows up. Then the relationship becomes reciprocal and may last until one of the pair dies.

A person's kin and spouse carry out her or his Ntęę ceremonial responsibilities. For instance, in the burial and funeral activities of a person's Ntęę Formal Friend, one or two dozen relatives of the deceased person's Formal Friend may offer their assistance [IV.B.3.a]. The Ntęę ceremony does not exist among the Apanyekra.

Formal Friendships are also initiated during the Pepyę festival proceedings, when youths jump into the water and emerge from it side‑by‑side, not looking at each other (Plate 39a) [III.A.3.b.(1).(e)]. The two girls serving the same male festival society as associates become Formal Friends for life, as do the commandant and the deputy commandant and the two file leaders of the novices in the Khęętúwayę festival.

Another source of Formal Friendship takes place when certain personal names are linked traditionally so that their bearers are necessarily khritswč (Nimuendajú, 1946:100). This practice is especially characteristic of the Apanyekra where a greater percentage of individuals than among the Canela are linked by Formal Friendship. Just after they gave me an Apanyekra name (Khenku?nă), one of the political chiefs, the old Teynő, began to use -related terms with me instead of ordinary ones. He explained that individuals bearing the names Khenku?nă and Teynő were always Formal Friends.

If there is an active crowd around a person's house and a great deal of noise, surely, a khritswč “game” (hŕŕkrun‑tsŕ: game-act) is in progress [III.A.3.b.(1).(e)]. A Formal Friend must come to the aid [III.A.3.b.(2).(b)] of her or his Formal Friend or the relationship may be felt violated by one of them.

These khritswč games [II.F.4] are increasing in popularity and frequency, maybe for the same reasons that the activities of Formal Friendship are spreading into ceremonial areas where Formal Friendship was not practiced traditionally [III.A.3.c.(3).(b)]. Formal Friendship spreads because the services of a Formal Friend cannot be refused for fear of violating the relationship. Also, the Formal Friend can earn food in hard times through these games and other services. For instance, a hŕmren person's female kin used to decorate her or his body just before this person went to the plaza with a meat pie (hŕŕkwčl) for the Pro-khămmă [III.D.2.c.(3)]. These days, however, the hŕmren person's principal Formal Friend with her or his kin and spouse usually do the decorating because they want to gain something. They can break the traditions these days because the Pro-khămmă are no longer strong enough to prevent it. (For individuals “rejoicing” [carrying out comic acts] [III.B.1.c.(2)] in the festival honoring of their Formal Friends, see Plate 39c-e.)

[III.E.6]  Informal Friendship Terminology

Individuals involved in Informal Friendships made between two men of the same age-set address each other as I-khwč-?nő (my-group-one: a person of my group; my age-set) and usually joke with each other when they meet. Informal Friendships (Glossary) are made almost entirely between men, although research assistants say it is possible to form such a relationship between the sexes. A novice and one of his age-set's girl associates may do this. There are far fewer Informal Friendships than formal ones. The former tend to become fewer in an age‑set as its members grow older, because the terminology of almost all the other relationship systems is held in higher priority by most Canela individuals.

Informal Friendships are only made during the Khęętůwayę and Pepyę festival performances, when a pair of novices jump into a stream together and come up facing each other (Nimuendajú, 1946:191). Informal Friendships are made only in this way in the two tribes (Plate 39b), and they last for a lifetime. Informal Friendships cannot be formed between women, because the two girl associates of the initiation festivals have to be Formal Friends by tradition. Both tribes make Informal Friendships with large city outsiders of either sex.

Ideally, Informal Friends may borrow or simply take things from each other in an exaggerated manner, expressing closeness and confidence in each other in this way. If, in reality and through time, it happens that they are no longer close personally and have become antagonistic toward each other, this joking behavior is largely curtailed, but not the associated terms of address and reference.

Informal Friendships provide one of the four joking relationships found among the Canela and the Apanyekra. Informal Friends often play both traditional and original tricks on each other. A characteristic trick is tossing sand in the face of one's Informal Friend who will wait for the right moment to carry out reciprocal action. This can be done even between members of the Pró-khămmă. From 1963 to 1979, I saw the older Kaapęltůk and the older Krôôtô periodically do this at some dull moment during a council meeting in the center of the plaza [III.D.1.c.(l).(b)].

Formerly, Informal Friends had sexual relations with each other's wives (J. Melatti, 1979a:74). In fact, the current practice is to call each other's wife “wife” and, in speaking to each other, to refer to any one of each other's children, as “our child,” using the dual inclusive form pa-?khra [III.B.1.1.(1)]. A secondary terminological system, quite limited in extent, spreads from ego through his Informal Friend to his Informal Friend's consanguines and affines.

[III.E.7]  Mortuary Terms

Special terms for referring to dead consanguines and affines were fully in use in the 1970s. These special terms are often similar, but with small added differences, to the ones used for the living. Sometimes, however, the basic morphemes used are entirely different. These expressions are very similar to those reported for the Krahó by J. Melatti (1973:14–15).

[III.E.8]  Teknonymy

Teknonymy is widely and consistently used among the Canela between opposite-sex siblings, married couples, and classificatory spouses who are having, or have had, a long term affair. It is used only in terms of address, not reference. There are no words that are unique to this system, but there are unique expressions that are used consistently and are associated with certain behaviors.

[III.E.8.a] Between Cross-Sex Siblings

Regardless of whether they have exchanged a name, a woman calls her brothers (uterine and most classificatory ones) i-túwa-re-më-hum (my young-one dim. her father: my name-receiver's father), and a man calls his sisters (uterine and most classificatory ones), i-túwa-re-më-ntsii (my young-one dim. his mother: my name-receiver's mother). (Itúwa-re is the term of address a name-giver uses for her or his name-receiver [III.E.4.c].)

This kind of address implies respect and caring, and is almost always maintained between adult uterine and amyi-půtŕl siblings. Other classificatory parallel-cousin and cross-cousin siblings (Figure 37) usually maintain this practice of teknonymy. When the relationship is weak and the individuals are young, they may not. If the more distant cross-sex siblings have sex, they are no longer considered siblings; they have become “other spouses.”


Married couples and unmarried “spouses” who have carried on long term affairs [III.F.8.a] call each other by the name of their first-born child, regardless of sex, whether living or dead. As with cross-sibling teknonymy, më-ntsii (possessive‑mother) and me-hum (possessive-father) are added, but in this case the addition is made to the name of the child. For example, Mďď-khwey and Hŕwmrő's first child was Homyĩ-khwčy, so they address each other as Hômyĩ më-ntsii and Hômyĩ mëhum (Hômyĩ her mother and Hômyĩ her father: Hômyĩ's mother and Hômyĩ's father). Sometimes, this form of address may be abbreviated to mëntsii/mehum or to just Hőmyĩ, when this daughter is not present.

In contrast to this marital or quasi-marital form of address, totally unrelated opposite‑sex individuals call each other by their own personal names whether or not they have had limited sexual relations, and most individuals who are classificatory spouses through the affinal system do the same [III.F.2].

[III.E.9]  Contributing-Father Terminology

The contributing-father (co-father) (Glossary) relationship system [III.F.11.a] does not have unique terms, or expressions of address or reference, but the system does determine behavior to a considerable extent. Moreover, its practice does occasionally change consanguineal and affinal patterns of address and reference [III.E.2.e.(3).(a)]. Consanguineal and affinal terms of address and reference are used, though the priority of such terms in relation to other relationship systems is relatively low.

I prefer “contributing-father” and “social father” to “genitor” and “pater” because no one in the tribe knows who the real genitor is [IV.B.1.b]. For the Canela, they are all genitors because they have contributed some semen to the formation of the fetus. A contributing-father, therefore, is a primary (-mpey) rather than a secondary (-kahŕk) father. I prefer the term “contributing-father” to “co-father” because “co-father” puts all the “fathers” in the same category, which is not the case. The role of the social father who is married to the mother is far more outstanding and distinct in its responsibilities [IV.B.2.d].

The contributing-father relationship system, which is comparatively extensive, is also based on the ethnobiological belief that any semen added by one or several men during the period of gestation contributes to forming the fetus. Thus, a person usually has several fathers in the “contributing-father” sense. These fathers are considered to be as much biologically the child's fathers as the social father.

A contributing-father is considered a one-link-away person [III.E.2.b] from the point of view of “blood” equivalence (Figure 39) with his contributed-to child [III.F.11.a]. This means he must maintain food and sex restrictions when his contributed-to child is ill because she or he has the “same blood” [IV.D.3.a,b]. Likewise, the child reciprocates when her or his contributing-father is sick. When the tribe is dispersed, messengers have to be sent to contributing-fathers as well as to non-classificatory first-link-away kin (Figure 20) in order to tell them to keep the necessary restrictions so that the sick person can recover.

I first became aware of the extensiveness (both in frequency of cases and in number of consanguineal links between ego and alter) of the contributing-father networks, when I was studying the ancestry of name-set-exchange siblings [III.E.4.a]. A considerable portion of these opposite-sex pairs were “siblings” because their parents, grandparents, or great grandparents had had sexual relations outside of marriage. For example, as an extreme case, a woman and her mother's mother's mother's lover's daughter's (in the lover's marriage) daughter's son (♀MMM“H”DDS) are distant “siblings” and so could exchange name-sets for one of each others' children.

Once the children of such an out-of-marriage union have recognized and maintained their “siblingship” for one generation (especially between two women) their same-sex children and, in turn, their parallel-sex-on-the-same-generation descendants (especially if all female) would have just as good a chance of continuing the two matrilines of connecting “siblingships” down the generations as if the origin of their “siblingship” had been in marriage (Figure 42). Often such “siblings” have bonds of great attachment, but in other cases the relationship is lost, especially if “siblings” on the same generation are “brothers.”

As with relationship systems with special terms, the contributing-father relationship system contributes significantly to the formation of behavior patterns. Contributing father-generated relationships—with obligations to provide food (the fathers) and maintain restrictions (all first-linkers)—continue for life for such “parents and children” and such first generation “siblings,” who more than likely are not biologically consanguineals according to Western science.

[III.E.10]  Ceremonial Relationship Terminology

The ceremonial relationship terminological systems have no unique terms; only consanguineal terms are used occasionally. These relationships, however, somewhat determine behavior in certain ceremonial settings. Moreover, these behavioral patterns are continued and carried out in daily, non-ceremonial life to a considerable extent, as reflections of positions held in festivals [II.E.1.b].

An example of this kind of behavior extended from festivals into daily life is exhibited between members of an age-set and their honored Wč?tč girl and her mother and father [IV.A.3.e.(1)]. Age-set members address her as “sister” and her parents as “mother” and “father” and behave correspondingly to some extent.



Marriage among the Canela and Apanyekra is monogamous, largely endogamous, and almost always uxorilocal. It can take place between any non-related persons of the opposite sex, but not, ideally speaking, between consanguineals, affinals (except in the sororate), and Formal Friends. Distant consanguines, affines, and Formal Friends sometimes marry or have sexual relations outside of marriage with each other. In so doing, however, they are committing incest (to ayprč). Matrilineality with exchanges between exogamous moieties (cf. Nimuendajú, 1946:79) does not exist and most probably never did [III.C.1] [III.C.10.b] (W. Crocker, 1977, 1984a).

[III.F.1]  Preferences and Restrictions

There are no marriage prescription or preference rules, except that if a woman dies, her widower is encouraged to marry her sister or close “sister” (Figure 23), so that he does not have to make mortuary payments and can remain in the same household with his children [III.E.3.a.(6).(b)]. Making mortuary payments would enable him to leave her kin (Glossary) early, well before being released formally by her family, and marry again into another consanguineal family after a period of mourning [IV.B.3.d.(2)].

A definite restriction is that a woman should not marry a man born in her own longhouse [III.E.2.e.(2)], although there are several exceptions in the one very long Canela longhouse (Figure 24, houses BB through NN, excepting CC). It has become so large that its members near the opposite ends scarcely think of themselves as being of the same longhouse.

Since there is no direct or indirect formal exchange of spouses between marriage alliance groups (Levi‑Strauss, 1963:149), among the Canela there is no question about whether it is wives or husbands who are “exchanged” in the classical sense. Quite clearly a son-in-law and his services to his wife's family are exchanged (“bought”) on several occasions for ceremonial foods and his wife's services to him [III.D.3.e.(5).(a)].

Exchanges are seen by the Canela as being between the extended families of the spouses. This interfamilial bond, however, is not strengthened generation after generation into a traditional alliance pattern, because brothers and their sons do not marry into the same families. They say there is no tradition against uterine brothers marrying into the same household, but in practice it is rarely done. Research assistants offer no reasons why uterine brothers, classificatory brothers, and fathers and sons avoid marrying into the same “hearth” household and generally even into the same longhouse if it is short [III.E.2.e.(1),(2)]. I suggest that brothers are not welcomed into the same matrilateral “hearth” (Glossary) group because the first-married wife's kin want to be able to control the later marrying-in sons-in-law, who, if the sons-in-law were kin, might offer too much resistance to being dominated.

[III.F.2]  Classificatory Spouses

Opposite-sex individuals who are neither kin, affines, nor Formal Friends, and certain opposite-sex affines [III.E.3.a.(6)], call each other by their personal names and behave toward each other as classificatory spouses (më ?prő-?nő: plural-wife-other: Timbira men's other wives, and më mpyę nő: plural husband other: Timbira women's other husbands). An unrelated person (i?ka?khrit) from another Eastern Timbira tribe is a classificatory spouse to every Canela or Apanyekra of the opposite sex. Tribal or village endogamy was and is the general practice, but there were and are a few exceptions, especially among the Apanyekra who have several intertribal marriages with the Canela, Krĩkatí, Pukobyé, and Krahó. In earlier times the fear of witchcraft, along with warfare, precluded most intertribal marriages, but such fears are minimal today [IV.C.1.c.(l)]. Women as well as men move to another tribe in marriage.

[III.F.3]  Incestuous Marriage

Marriage and sexual intercourse occasionally takes place between individuals who do not know they are related, especially if they are connected through all male linkages, because such bonds are easily forgotten (Figure 42). There can also be marriages between distantly related individuals such as fifth, fourth, or third cousins (through all-female links); between more closely linked persons, such as second cousins (rarely); or, as recorded only once, between first cousins once removed. (These individuals know they are related before having had sexual relations or before getting married.) In these cases (even for same-longhouse fifth cousins) their first act of sexual intercourse is considered to be committing incest (Glossary) (to ayprč: make transformed), which amounts to turning kin into affines. A man has to give a woman a small payment for having brought about this transformation.

When committed by distant kin, incest is thought to shorten the life span of the individuals involved, but when committed by uterine siblings, incest is thought to bring about madness and death within several years [I.A.1]. There was no talk about parent‑child incest.

[III.F.4]  Steps into Marriage

Unlike most Western marriages, which begin with a religious or civic official's pronouncement or the signing of a legal document, Canela marriage takes place through a series of events and becomes relatively more complete over time as certain steps are carried out.

Older women used to take the initiative in arranging traditional engagements between young people [IV.B.1.d]. They still play the more active role in searching for an able son-in-law when the young people leave it to them. A girl's mother approaches a prospective son-in-law's mother, and if agreement is reached, the two mothers turn to their male relatives for support. Whereas the ultimate control and command is theoretically held by each mother-in-law's male kin, it is the women who are more actively engaged in attracting and securing the young males to their matrilaterally extended families. These days, however, it is more likely the girl herself who selects her mate by presenting her mother and male relatives with a fait accompli through her loss of virginity and her identification of her partner.


In earlier times (Nimuendajú, 1946:118), marital engagements (ku ?te më to aypën tę: he past-tense plural relational restrict: they were restricted for each other) were made for potential spouses when the girl was 4 to 6 years old and the boy, 12 to 18. This relationship was usually initiated by the mother of the girl in her approach to the mother of the boy. Research assistants say that these engagements usually did not lead to marriage, even though both families carried out bride- and groom-service for years. Just prior to and after the girl's puberty these engagements could be broken if the boy's kin made a small payment for his release. Canela young people could not be forced into marriage. Usually, the young man found another woman before the girl grew up, or she simply did not like him. The Canela feel that while men can be coerced and disciplined [III.A.3.c.(3).(f)] [III.B.1.e.(l)], women's feelings are unchangeable. Today, such early engagements are not attempted.


When an “unattached” male takes the virginity of a girl, the two are said to be “married” in the usage of the Canela expression më hikhwa (they lie-down) and the Portuguese se casa. The act of sexual intercourse marries the couple in this limited sense [III.D.3.c.(1)], but this marriage is very weak and easily broken, and many steps still have to be carried out before the marriage is strong. “Unattached” for a man means he has no children by a formal wife.

Another marriage is created when an unattached man breaks the sexual mourning restrictions of a widow. Still another marriage is initiated when an unattached, unrelated older man starts to live openly in the house of an unmarried older woman (implying they are having sex). There are, however, a number of steps to be completed before the marriage becomes secure.


The expression “unattached” needs to be examined for its meaning in this context. If a married man has no children by his wife, he is “detachable” from her. Testing extremes in an hypothetical case, research assistants debated that if a married man takes a girl's virginity (or breaks a widow's mourning restrictions), he has married her too. Most likely, he would have to leave the wife with whom he was living, if he had not taken her virginity earlier, or make a large payment to the girl's family to leave her. The kin of the three concerned persons would call a meeting, and after testimony from the three principals and witnesses was heard, the man most probably would have to stay with the girl whose virginity he had taken (or the widow); that is, unless his first wife had just become pregnant. The likelihood of a prospective birth is more important for holding a marriage together than a lost virginity (or widowhood) [III.D.3.e.(5).(b)]. However, he or his kin would have to make a payment for the lost virginity (ku-?te ku-khën ya?pan tsŕ ?nă: he-past it-broke paying thing for: he broke it—its payment: virginity's fine), or for the lost widowhood. Such exercises with my research assistant council members were instructive but not definitive.


A “single” woman has no husband but she may have children. The Canela call this condition mpíyapit (Glossary), and this term applies to both sexes. Whereas men prefer their female kin not to be mpíyapit, these days, and speak with pity for a girl who, in the course of a hearing, may be about to be left in such a situation, they do not treat such women pejoratively. However, such women are expected to appear more often than others on the extramarital exchange days [IV.A.3.f] and to participate more readily in the early afternoons for male work groups [II.E.6.a] [III.D.1.c.(l).(d)]. Being a female mpíyapit is an accepted style of life (Nimuendajú's (1946:130–131) “wantons”) although it is becoming increasingly less desirable with acculturation [II.D.2.i.(6)] [II.E.5.f]. A man seldom remains a mpíyapit for long, often marrying a younger woman as soon as his period of mourning is over.

In the 1950s, the female më mpíyapit were held in the same esteem as married women, their actual level of esteem varying more with other factors, such as their personalities, their ability to work, or their number of kin. These single women often furnished their households with more money and food, given by lovers, than most husbands could provide theirs. Presents from lovers to single women are not “payments,” they are presents, because such women are not considered prostitutes. The Canela use the Brazilian Portuguese term rapariga (whore) for backland and city prostitutes but not for their më mpíyapit.

In return for her sexual favors, an mpíyapit (when a dedicated single woman) persuades men to work on her farm plot, particularly felling trees and clearing shrubbery [II.C.3.d]. Some single women, however, clear shrubbery and small trees themselves with machetes and axes, and do an excellent job. My principal female research assistant, Tel-khwčy, age 46 in 1970 (Plate 73a), did this year after year. Such women, however, usually persuade men to carry out a number of tasks, particularly in their fields. Tel-khwčy [II.B.2.e], although a mpíyapit in the late 1950s, was one of the two informal leaders of the women during the Festival of Oranges [IV.A.3.f.(5)] each year, a point which demonstrates her considerable prestige. By the 1970s, she was permanently married.


The Canela make a distinction between “contracting” marriage (më to aypën tę: they for each-other restricted) and “adjusting” the marriage (më aypën pa: they with-respect-to-each-other listen: they hold a council: an audięncia: a hearing (Glossary). Always occurring before sex, “contracting” a marriage in a small interfamilial meeting constitutes an enduring relationship, including a continuing exchange of services between families. “Adjusting” a marriage, occurring either before or after first sex, consists of some members of each family holding a larger meeting to consent to the marriage as well as to counsel the couple. During this meeting of the couple's kin, the uncles (Glossary) of each side do most of the talking but anyone may participate. The uncles try to find out if the young man intends to stay with the young woman. If he does, their uncles ask each of them, individually, if they really care for each other. If they both say they do, they are admonished and lectured about the roles and duties each should carry out in their marriage. This “adjustment” hearing [III.D.3.b] is the closest the Canela come to a wedding ceremony.


If the young man who has taken a young woman's virginity does not want to stay in the marriage, his extended kin must make a significant payment to her kin to release him. Such a payment might be several pieces of iron equipment (axes, machetes, hoes) and maybe even a horse, should they have one. The payment is relatively large in Canela terms, because both virginity and a first marriage are highly valued. Women make no payment if they do not want to stay with a man, but this rarely happens [III.D.3.c.(1)].


Sometimes, a hearing occurs just before the girl's virginity is taken when the couple obviously like each other (më aypën kin: they each-other like) and when the parents want to encourage their marriage. This is the sequence of events that parents prefer. If, however, her virginity is taken first, the marriage is referred to as having been “stolen” (ha?khĩya: roubado). If she subsequently fails to obtain the seducer's admission or to prove her case in identifying him during a series of hearings, she would be said to have “lost her money” (i?pore piktol: money lost), the payment due her if she loses her virginity without obtaining a husband. This loss depreciates her socially rather than constituting a significant financial loss to her extended family. The acculturative trend is moving from “contracting” or “adjusting” first toward “stealing” and “liking each other” and then “adjusting” afterward, but not toward the virgin's “losing her money.” These meetings bring both families into more closely recognizing and reinforcing the union of the young couple.


After a period of time has elapsed for the marriage to stabilize and appear sufficiently secure, the third step in strengthening the marriage is taken; the wife's female kin prepare to put on the purchase-of-the-son-in-law rite (mëpa wawč ?nă hŕmyől: our son-in-law for pay). In this small rite the wife's female kin make and carry large meat pies (Glossary) to the natal house of the husband [IV.B.1.i]. These days, the husband's kin sometimes send meat pies to the house of the wife so that the pies are exchanged. Then, if the bridegroom should want to leave the marriage later, there would be a lesser amount to pay the bride's family [III.D.3.e.(5).(a)].

The meat pie purchase of a son-in-law used to occur before sexual relations began, but now the rite almost always takes place afterwards. This is an individual matter, however, and apparently these alternatives existed at the turn of the century as well as now.


The fourth step toward securing a marriage is painting the young woman's belt. This takes place just after she has won her belt (i?pre) as a girl associate for a men's group in a great summer festival [II.D.2.e]. Her uncles (Glossary) hunt for a deer while she undergoes several days of restrictions in seclusion [II.D.2.f.(1)]. Then she carries the deer part of the way to her husband's house and her husband's female kin rush, running in a disorderly fashion (më ?prőt), to take the deer off her shoulders. Then they escort her to her mother-in-law's house and rework her belt, painting it red with urucu as well as covering almost her entire body with the red paint. (See [IV.B.1.h] for a full description of this sequence of rites.)


With this rite passed, the new wife has become more acceptable to her in-laws, and therefore is freer to involve herself in the more public extramarital festival days when sex occurs with “other husbands” (më mpyę-?no: pl. husband-other) [III.E.3.a.(6)]. Finally, she can be seen by female affines in a public place, about to sally forth in a group of women to meet a crowd of men in the cerrado for extramarital sex [IV.A.3.f.(1)]. Before the belt-painting ceremony, she could have become involved only in very hidden affairs, or her female affines would have criticized her and complained. After the belt-painting rite, however, she is encouraged to become involved in all the customary extramarital activities without these exploits causing embarrassment to her in-laws [II.D.2.e.(3)].


After the belt-painting ceremony, if she has not become pregnant, she has become one of the free and unburdened më nkrekre-re (pl. slippery-one dim.: little free/uncatchable ones). This is the period of great fun in a woman's life [II.D.2.g], because later she is hampered by raising children and household responsibilities. She is expected to make the most of her short time of freedom before children limit her activities. In contrast, men are freer during their later years as well as in their më nkrekre-re period [II.D.3.f].


In early times, the period of being an nkrekre-re (Glossary) woman was said to be a time of great turnover with respect to husbands. She was loosely married, but slept with men on the other side of the plaza from her husband and his age-set moiety members. (Men without children in marriage slept in the plaza [II.D.3.c.(2)].) As a result, she easily became attached to other men who, if she allowed them, showed her continuing attentions. Research assistants say that during this period, a woman might become the wife of several men in succession, merely by choosing to be with one and then another for a significant period of time, rather than with her original husband. This continued until she was caught by pregnancy with one of them who was thereby her permanent social husband until she was no longer encumbered with growing children. During the woman's free period her serial husbands did not have to pay to get out of their marriages if she left them first, and she did not have to pay anything herself either. Payments existed (and still exist) only to restrain men.


In modern times, serial marriages do not occur. This is because the husbands complain too much and require that their wives' families pay excessively “to erase the shame passed on their faces” (para apagar a vergona que elas passaram nas caras) that their wives' behavior has made them feel [III.D.3.e.(3).(a)]. The nkrekre-re period for a woman was possible because, in former times, the aunts and uncles worked very hard to suppress the sexual jealousy of their nieces and nephews, particularly the nephews [II.D.3.h.(1)]. Under such circumstances it was difficult for the young husbands to claim and win payments in formal hearings because of the “infidelity” of their wives [III.D.3.c.(1)]. The uncles did not side with them in the hearings.


On the first day of the Regeneration season log racing in the Ayrën (Glossary) ceremony [IV.A.3.f.(2)], a woman chooses an “other husband” to hunt game for her, and if he is successful, she usually rewards him with sexual relations. Then the following day the woman makes a small meat pie of the game and presents it, privately, to her mother-in-law. This rite the fifth step in marriage is like a personal answer to the belt-painting rite. Since the mother-in-law has shown her confidence in her daughter-in-law, the latter makes the former a present of the proof of her new freedom.


The sixth step is apparent conception. Up to this point in the life course of a couple, marriages are breakable by either partner. But conception catches the husband in a permanent union with his wife for the sake of the fetus and later for the welfare of the infant and future children. If there is a miscarriage, or if the child dies some time after birth but before the next conception, the marriage is no longer permanent; but it is more expensive for the husband to pay his way out of the union because of the suffering (the “ripping and tearing,” they say) his wife has undergone [III.D.3.e.(5).(a)].


Successful childbirth is the point at which a marriage becomes secure, and is the seventh step. In my 1970 Canela marriage study, there were no cases with ordinary circumstances of a man's leaving his wife and child (or children), and successfully staying away from them for more than a year. There were, however, numerous cases of 3- to 6-month separations. (No similar study was done of Apanyekra marriage.) A man's uncles, his age-set, and men in general all pressure him into returning to his wife and children [III.A.3.c.(3).(f)]. There were, however, seven cases in the 1970 marriage study with unusual circumstances: either a “crazy” man left both his wife and their children or a woman ordered him to leave and live elsewhere permanently, the equivalent of divorce [III.F.9].


Soon after the infant is born, both the wife and husband go into confinement in the same house, separated from each other by partitions. They carry out extensive food and sex restrictions for the sake of the health and survival of the baby. In my view, it is this confinement (the eighth step), as well as the birth of the child, that really cements the marital union. In earlier times, this occasion was the first time the husband came to live on a 24-hours-a-day basis in the house of his wife [IV.B.2.c].


About 40 days after the childbirth, the wife's family holds the postpartum restrictions termination rite (Më Hŕ-?khrël: they generalizer eat: they all eat), lasting most of a morning [IV.B.2.d]. At the end of this rite, during which the husband and the baby's contributing-fathers ritually avoid tasting one meat pie, the wife's family sends several large meat pies to the husband's longhouse kin. This is the ninth and last ceremonial step reinforcing the marriage, and occurs only for the first child. These various marriage reinforcing rites are less needed for succeeding births because the first child is sufficient to reinforce the marriage. Only the couvade is practiced for successive births, and then only to a lesser extent each time.

[III.F.5]  Marital Payments and Balance of Costs

The total of all payments made from the wife's to the husband's family to secure the marriage is large in Canela terms, but the husband's family has to pay this approximate amount back to retrieve him from the marriage should this become necessary. There are three big payments to the husband's family: (1) the meat pies (Glossary) of the buying-of-the-son-in-law rite, (2) the deer at the time of the belt-painting, and (3) the meat pies of the contributing-fathers' rite. In addition, there is the token payment of a small meat pie given just after Ayrën day. Occasional groom service (i.e., providing water and wood) takes place in the early stages of the relationship between the two extended families, as well as occasional bride service (providing meat).


Within the marriage relationship, the woman gives her virginity (kolmă ?kuuni naare: still whole not: no longer virginal), exposes herself to the risks and pain of childbirth, and renders many domestic services, including sexual relations. The debt of sexual relations between the sexes may be evaluated by noting how sex is exchanged outside of marriage: neither the woman nor the man makes a “payment” (hŕm-yől tsa: pagamento). The man, however, is supposed to supply a small “present” (agrado) just to please the woman every now and then. Research assistants are very clear that there is an important distinction between “payments” and “presents.” Payments are made to settle past “debts,” or they are future debts that may have to be paid. Presents are given to please people and no “return” for them is expected. Thus, sexual services given a man by a woman are not paid for, and they change the balance of intersex contributions only very slightly—maybe in her favor.


Concerning economic support and domestic services, men supply almost all the meat (the most highly valued food), and do most of the work in preparing, fencing, planting, and weeding the fields. This tends to balance the workload of the housework and child rearing done by women, who also work in the fields, planting, weeding, and doing more of the harvesting (except for rice) than the men [II.C.3.d]. Thus, these kinds of services from one sex to the other are relatively balanced. Nevertheless, when a man returns from trips out in the world [II.D.3.i.(1)], he gives his mother- and father-in-law significant presents, usually cloth, beads, iron farming tools, and possibly a shotgun (J. Melatti, 1971), which suggests he owes more than she. These contributions to the other sex are hard to “weigh” as being greater or lesser in relation to each other, and the relative amount in any category of a contribution varies in each case [III.D.3.e.(5).(a)]. In general, however, the balance of contributions can be represented as follows.

Wife to Husband

Husband to Wife

Occasional groom service (water and firewood) in early stages

Occasional bride service (meat) in early stages

Meat pies at time of son-in-law purchase rite

Continual supply of meat in middle and later stages

Deer at time of belt painting

Preparation of farm: clearing, felling, burning, fencing

Meat pies at time of contributing‑father rite

Cooking only when necessary

Preparation of food

Child tending only when necessary

Child rearing

Water and firewood when necessary

Supplying water and firewood

Building house

Maintaining house

Planting farm

Planting farm

Weeding farm

Weeding farm

Harvesting rice

Harvesting rice

Gifts upon return from trip

Harvesting roots and tubers when needed

Gifts to please wife

Loss of virginity


Pain and damage in child bearing


Sexual services


[III.F.6]  Purpose Of Payments

The purpose of payments that result from judicial hearings [III.D.3.e.(5)] is to keep youths and young men with their wives. Men are seen as being looser, freer, and hard to stabilize, but they are far easier than women to coerce and restrain socially [III.A.3.c.(3).(f),(g)]. Age-set training has served partly to establish this malleability, and men are part of an ongoing, active age-set for their whole lives, in both the festival context and in daily life. An age-set's force towards social conformity extends also to marriage [III.B.1.e.(l)]; men of the same age-set shame and ridicule each other in informal, humorous ways, thereby keeping each other in line.

Women, in contrast, are said to be stable, stubborn, and immutable. They are restrained by the fear of rumors (tswa-?nă: sharp-conditional) being spread against them, especially by their female relatives among whom they live and work [III.A.3.c.(3).(d),(e)]. In this sense, female kin are to a woman as an age-set is to a man: the prevailing and compelling social matrix. While this female matrix is focused principally on marriage and family, the male matrix is only partly focused in this way, so that additional persuasion is required to keep men in their marriages. In this context, payments serve to add stability to the male side of the marital equation.

It is not that the payments are economically so large that the difficulty in paying them restrains men; it is more the fear of being shamed in public hearings [III.D.3.b]. It is this personal shame, with his error publicly agreed to by his own kin, that hurts the individual male [III.A.3.c.(3).(a)], not the fine against his extended natal family.

[III.F.7]  Purpose of Marriage

Marriage exists for the purpose of raising children, that is, for the maintenance of a family group for socializing its young members. The small unit of the one-link-away (Glossary) family [III.E.2.b] (Figure 20) is held together by concepts of “same” blood (Glossary) identity; therefore, for example, all one-link-away family members need to observe restrictions when any one member is ill in order to help that person recover, which keeps the cohesion of the nuclear family high.

However, one mother and father pair is rarely the actual social unit for raising children, except when the tribe disperses to live in the farm plot huts or in the houses of regional bcklanders [II.B.3.j.(l)]. The one-hearth-for-all economic unit (Figure 22), which is usually composed of two to four families with or without husbands, is the domestic unit that is most instrumental in maintaining the cohesion of the larger group, for the ultimate purpose of developing children's behavioral patterns. The social forces of this basic food-sharing unit [III.E.2.e.(1)] also serve to hold the marriages in it together.

Joking relationships, to the extent that they are allowed to be carried out in this hearth group of several marriages, are considerably muted. For instance, if a man were living temporarily in a somewhat distant “sister's” household, this sister's children would be his “nieces” and “nephews,” with whom he would usually joke. Under the circumstances of living in the same hearth group, however, extensive joking is not appropriate [II.D.1.b.(3)] [III.E.6]. Marriages and hearth groups must be serious and viable social structures, because they are oriented toward dealing with two of the prerequisites for tribal survival: food preparation and raising children.

[III.F.8]  Lovemaking and Affairs

Lovemaking and affairs can be found easily outside of marriage, but this is not inconsistent with the fact that many Canela and Apanyekra married couples appear to maintain a very loving relationship with each other. Several festivals take place each year during which spouses go with different moieties and have sexual relations with their “other wives” and “other husbands” [IV.A.3.f]. Because these festivals are part of the ancient tradition of the tribe, extramarital relations are sanctioned within certain festival conditions.


Individual one‑time meetings, or even long term affairs, are entirely within the tradition. I used to hear of numerous examples of each. In fact, I believe that extramarital sex is (or was) the greatest source of fun and joy in Canela life, adding zest to the daily routine and making life considerably more worthwhile for the individual (W. Crocker, 1964a, 1974a.)

These extramarital arrangements are simple and entirely personal. A man feels a sexual attraction for an appropriate woman in the other spouse category who happens to be passing by. He makes this known to her with his eyes or gestures, and if she responds favorably, he soon finds an occasion to quickly say when and where they should meet, or she does. The assignation usually takes place in a concealed spot in the cerrado or by the streams, often in the late afternoon. When I was walking with Canela away from the village, they often pointed out with glee the tell‑tale marks of sex encounters in the sand. During these quick encounters, the man squats between the lying woman's legs, which rest on his thighs.

(For fascinating comparative materials on marital and extramarital sexual relationships and related activities, see Gregor (1973, 1985) for the Mehinaku and (Lizot, 1986) for the Yanomami.)


Women used to take the initiative in these matters at least as often as men, I was told. For example, when Western dancing with couples embracing was first becoming popular with the Canela in 1959, they danced indoors, on hardened floors, bare breasted, and to the rhythm of beaten metal barrels without other music. At that time, it was almost exclusively women who were choosing male partners. A year later, however, they were dancing well clothed to the accordion music Tsaahů had learned to play, and only men were taking the initiative of asking women to dance, in imitation of the backland style. There can be no doubt, however, that women often, if not usually take the initiative in extramarital sex affairs, just as research assistants say, because the context is traditional.

[III.F.9]  Divorce

In theory, divorce cannot occur while a couple's children are growing up, until all the children have left the home or become married. Consequently, no grounds for “divorce” (that is, leaving children) exist (cf. Nimuendajú, 1946:128–129). Nevertheless, most of these older couples remain married until death takes one of them. In my 1970 study of 96 Canela marriages, only 5 cases occurred in which a man succeeded in leaving his wife with his own children while they were still growing up. One mentally ill man left two wives, in succession, each with a child by him. One Apanyekra left a Canela wife to return to his tribe. One difficult woman was left twice, once by a man who later served time in the Indian service correctional detention center in Minas Gerais. There were, in addition, two cases of women who required their husbands (fathers of their children) to leave the marriage permanently. One of the fathers was at least a latent homosexual and scarcely supported his wife economically. Thus, there were only seven “divorces” in the study. The five cases of men leaving their children on their own initiative exhibited exceptional rather than ordinary circumstances.

Separations invariably involve numerous hearings called by the kin of both parties to mend the marriage. There are no myths, stories, or reports of such hearings taking place in earlier times that I know of, so this phenomenon (at least its extent) is possibly of 19th century origin.

A man who leaves a wife who was neither a virgin nor a widow at the time of his appearance in her life, nor the mother of his children, pays something to leave but not much, depending on how long he has been with her as well as on many other factors. She may or may not have children of her own, but this point makes little difference in the amount of the payment if he is not considered to be their father through marriage.


Between 1975 and 1979, two divorces took place in which men left women who were mothers of their children. One man had a wife, Pčp-khwčy, who was dying of tuberculosis over a period of years. I remember she obviously had the disease in 1964 and was wasting away by 1975. Because her husband had a salary from the Indian service and was willing to give her half his monthly pay, the council of elders and the Indian service allowed him to divorce her. The division of his salary was arranged through the Indian service. He then remarried. Another man, but one with no salary, left a wife and his children to marry a single woman, against the wishes of the council of elders and the Indian service. There were no exceptional circumstances in his case; he was merely a willful individual. Other unexceptional cases of divorce with children involved were in my 1984 communication from the Canela. Unfortunately, a trend in this direction may have been started.


The Canela speak of a man's divorcing his wife with his children as his “leaving his children” behind, not as his “leaving his wife.” It is surprising that there is so little “divorce from children” among the Canela in comparison with the Krahó and the Krĩkatí (personal communication with J. Melatti and Lave at the Harvard Pan-Gę Seminar, Cambridge, 1966). However, Canela men have many long separations from children, who always remain with their mothers. For one man, his separation lasted for almost a year in 1958–1959.

[III.F.10] Group Age-Set Marriage

Although marriage is associated with the loss of virginity, the Canela at the turn of the century held an age-set marriage ceremony in which every male in the age-set was “married” (or remarried) on the same day [II.B.1.e], according to my research assistant council and Nimuendajú (1946:123). The age range of the men was spread over a period of about 10 years; the age range of the women must have been similarly spread, but as much as ten years younger, so that some were already married and some were still virgins. Women who had babies, and their husbands, were excluded from the ceremony because of the inconsistency of their condition with the ideal model. Besides, the baby would ensure their staying together anyway, so they did not need the reinforcement of a group marriage ceremony.

Each age-set “graduate,” whose intended wife or wife was not already pregnant, was conducted in turn from the plaza by a Pro-khămmă, usually a grandfather of the graduate, to the house of his present or future parents-in-law where he was made to lie, limbs interlaced, with her. Then both were advised to care well for each other. The Pro-khămmă received a meat pie from the bride's kin and took it back to the plaza as a present to his age-set. The procession was repeated for each graduate, with the Pro-khămmă marching in file to escort one graduate at a time to his wife's bed. A ceremonial individual followed alone, considerably behind, singing a traditional song that was special for the ceremony. In 1923, however, the ceremony did not occur. The age-set file leader, Ropkhŕ, as on other occasions, was the first to undertake his age-set's activities. His wfe Yőőtsen had become pregnant before the ceremony was to be performed and the elders decided it would not be propitious to carry out the performance for any members of the age-set. File leaders of age-sets are hŕmren in ceremonial status and strong in psychic powers and prognostic sensitivity (W. Crocker, 1978:17), unlike the age-set commandant and the age-set deputy commandant (Nimuendajú, 1946: 193), who are just political figures [III.D.1.i.(1)]. Therefore, when an age-set's “seer,” with group-protective and danger-sensing abilities, cannot be the first participant in the ceremony to test its propitiousness [IV.A.5.e.(2)], it might well be safer and wiser not to carry out the performance at all. This same “seer,” Ropkhŕ [I.G.9], age 65 (Plate 17e), was a member of my research assistant council, when I did this research in 1964, and another member of the council, Pyę?khŕl, was a woman still older than him. They both remembered the circumstances around why this ceremony did not take place in 1923 (Nimuendajú, 1946:91) just after the graduation of his age-set. Moreover, until young men had been conducted to their spouses' homes, they were not supposed to visit them in the daytime. When most young husbands and even the age-set leader himself were seen to be ignoring this rule, as Ropkhŕ said, there was little reason for the elders to crry ot the group marriage ceremony.

[III.F.11] Ethno-ideology

Ethno-ideological concepts are generally used by preliterate societies to describe, support, or substantiate certain of their social structures. The Canela hold a number of these “truths” to be self-evident and enjoyed teaching them to me. The Apanyekra are assumed to hold similar views, but the data supporting these ethnoconcepts was collected mostly from the Canela.


The Canela believe in a concept of “blood” (kaprôô: blood) (Glossary) or some corporeal substance that is similar for and shared by certain relatives (W. Crocker, 1977:263). The degree of similarity of the blood of different relatives is a measure of how close they are as kin, as Da Matta (1982:51–52, 105–109, 160) maintains for the Apinayé and J. Crocker (1985:79) for the Bororo. Uterine siblings of both sexes have similar or almost “identical” (ipipën) blood, because they were born as a twig, or a branch, coming off the same umbilical cord. (Figure 38).

This ethnotheory maintains that all consanguineal first-link individuals (ego's parents, siblings, and children) [III.E.2.b] have “equivalent” blood (Figure 39), that is, their blood composition is very “similar” (i-pipën: it-balanced: literally, the items are the same in weight). Consequently, one-link-away kin have to carry out restrictions for each other when any one of them is ill in order to help each other recover (W. Crocker, 1971:325). One-link kin exist in a common blood “pool” which any one of them can pollute, hurting any weak member. Thus they must refrain from sex and from eating certain polluting foods when any one of their blood-pool members (i.e., one-link kin) is sick [IV.D.3.a,b].

After being married for several years, a couple also begins to have similarly equivalent blood. Their blood becomes equivalent through sharing certain of each other's body fluids—mostly through sexual intercourse and perspiring together. Thus, they have to carry out restrictions for each other when they are sick just as if they were consanguineal first-link relatives.

An expression for consanguineal relatives is kaprôô khwč (blood group). Thus, I thought that perhaps ego's spouse, after having spent some time living with ego, would become part of ego's kaprôô khwč, but research assistants say this is not the case. The spouse becomes one of ego's i-piyakhri katęyę (her/his-restrictions people/contenders) but not one of her or his blood kin (kaprôô khwč or hũũkhyę: consanguineal-relative).


In the Canela view of the flow of humanity through time (Figure 40), descendants (tŕmtswč: grandchildren) come horizontally towards (aypęn të: hither move) ego. Having passed ego, these former descendants are seen as ascendants (tůy/kęt: grandmothers/grandfathers) who are moving away (amu të; away move) linearly [V.B.1.b]. More precisely, what the Canela conceptualize are matrilines of women (mother, grandmother, great grandmother, etc.) moving away, with their married-to husbands (më nkętyę: pl. male-ascendants) attached to them. The female line is the more permanent one, and, as such, is seen as the structure that survives through time, with the male ancestors portrayed as marital appendages.


The Canela have different imagery for portraying a similar aspect of the flow of humanity through time. They observe the growth of a sweet potato vine on which additional potatoes appear further and further out along the vine, away from its central spot of origin and towards its growing tip. Each more distant potato is a female descendant further on in time. (Note that the imagery is reversed. Here, descendants, instead of ascendants, are moving away, if the observer is in the center of the potato patch.) The vine grows because a male from some other vine joins this furthest and newest potato in marriage so that eventually a still newer potato—their daughter—is formed even further out on the vine. Their son, born next to their daughter, traverses to another vine when he becomes adult, and finding its growing tip, marries the newest potato there to extend the vine and produce their own children. In this image the Canela see tribal “descent” as the horizontal spread of their people, in matrilines, away from a central point (Figure 41).


The Canela also use imagery to describe the transit of a youth moving from his maternal house across the plaza to the house of the mother of the young girl he is marrying. They say the youth moves like a shooting star crossing the great expanse of the sky (Nimuendajú, 1946:233). His crossing creates a kin bridge across the plaza (Glossary).

Two other images portray the congruence of the Canela social structure and the village structure. The first image is the ring of houses around the circumference of the village where a network of totally female kin (“sisters”: parallel cousins; Figure 42) forms the blood structure of the matrilaterally based longhouses [III.E.2.e.(2)]. The second image is the across-the-plaza network of kin (cross-cousins) initiated by the passage of many youths who traverse the plaza in marriage. (See Figure 43 for actual intermatriline ties in Escalvado.)

The Canela point to the form of their ceremonial meat pie (Plates 22e, 23b) as exemplifying these two structures. The rim of the circular meat pie, which is wrapped in wild banana leaves, is like the circular boulevard and its houses, and the bands of buriti straw crossing the meat pie are like the youths crossing the village in marriage.

The linkages, whether around the boulevard or across the plaza, are made through blood equivalences existing through time: a just-married youth's sister who is a one-link kin to him has blood that is equivalent to his. Soon his across-the-plaza wife has equivalent blood to his, and later, so does his daughter. Much later, his granddaughter has blood that is equivalent to his daughter's blood (but not to his own blood because his granddaughter's blood has become diluted by being two generations away; i.e., by two marriages; Figure 39). Thus, a bridge (hapŕŕ) [V.A.5.b.(1).(b)] of overlapping equivalent blood pairs (më-hapŕŕ: Timbira-speakers' bridge) spans the plaza (Figure 44).


This bridge of overlapping paired equivalent blood kin are all blood relatives (më kaprôô khwč). These kin are a unit, created through a marriage, that is held together over two to four generations (at the most) through the individual relative's recognition of shared blood. The individual recognizes her or his father's sister's, mother's father's sister's, and father's father's sister's kin and the reciprocal relationships [III.E.2.e]. Thus, an individual may have up to three across-the-plaza kin groups into which she or he cannot marry or have sex. In these cross-cousin kin groups, the individual recognizes up to second cousins and their children (the third descending generation), if the appropriate linkages are still intact. I know of no cases where fourth cousins, as calculated through this structure, mutually recognize each other. However, in the parallel-cousin matrilateral kin groups based on the longhouse, fourth and fifth cousins are sometimes recognized as “primary” (-mpey) “siblings.” Prepacification Eastern Timbira tribes, with their greater populations living in one village, possibly extended the longhouse and across-the-plaza kin structures to include more generations.

[III.F.12] Summary Of Village “Blood” Ethnostructure

While the social structure around the village circle of houses (Figure 42) is based on “sororally” related females and extended through time by marriage (Figure 39), the social structure across the plaza is based on an individual male linking two matrilines, through his siblingship to a member of one and through his marriage to a member of the other (Figure 43). Marriage not only extends matrilines generation by generation (Figure 41), it also holds diverse matrilines together for up to three or maybe four generations (Figure 42). Thus, marriage, in which blood of the couple becomes equivalent, is one of the two most important factors contributing to social cohesion. The other factor is genealogical relationships in which equivalent blood is also the basic concept. These two blood concepts, as well as others [III.D.2.e,3.f], hold the village and the tribe together. Thus, blood concepts, which are basic to both marriage and kinship, are also basic to village structure.

It is known that such blood ethnoconcepts support Timbira social structure (Da Matta, 1979:113), but the existence of genealogical blood ties beyond two-link-away individuals has not been stressed in studies of other Timbira tribes.

This relative lack of emphasis by other Timbira tribes on extended genealogical connections must be due largely to more extensive culture contact, depopulation, and consequent deculturation (Lave, 1979:36–44). Apparently, most other Timbira tribes (even the Apanyekra) base their terminological systems less on actual genealogical ties and more on name-set transmission and simple individual choices (fictitious ties) than do the Canela.

Probably blood ethnoconcepts are rarely exemplified as clearly in the demographic composition and the physical arrangement of the villages of other tribes in the world as they are among the Canela.

<<back  table of contents  next>>
[top of page]