Hierarchy, symmetry, and the xavante spiritual life cycle

ARTIGOS

Hierarchy, symmetry, and the xavante spiritual life cycle^*

James R. Welch

Fundação Oswaldo Cruz - Brazil

ABSTRACT

Xavante society, well-known for its dual structural aspects, provides the best documented example of a formal age-group system in South America. Although the basic features of the Xavante secular age-group system were previously described, a second age-group system in the spiritual domain is presented here for the first time. In this paper, I describe the Xavante spiritual life cycle, including a structural dynamic whereby age sets pass through age grades in alternate fashion, allocating them between two moieties. That basic morphology is shared with the secular age-group system and attests to a Xavante logic of hierarchy and symmetry as mutually constructed and non-contradictory. Interactions between the spiritual and secular age-group systems manifest in the daily experience of spiritual participants in ways that suggest plurality and contingency are essential features of Xavante social organization.

Keywords: age-group systems, Brazil, social organization, Xavante.

RESUMO

A sociedade Xavante, bem conhecida na literatura antropológica pelo caráter dualista de sua estrutura social, constitui um dos exemplos melhor documentados de um sistema organizacional com base em grupos de idade na América do Sul. Apesar de as características fundamentais do sistema secular de grupos de idade Xavante terem sido descritas anteriormente, um segundo sistema de grupo de idade, de natureza espiritual, é aqui apresentado pela primeira vez. Neste artigo descrevo o ciclo espiritual Xavante, incluindo a dinâmica por meio da qual as classes de idade passam, de maneira alternada, através de categorias de idade, alocando-os entre as duas metades. Essa morfologia básica é compartilhada com o sistema secular de grupos de idade e atesta a existência de uma lógica hierárquica e simétrica mutuamente construída e compatível entre si. As interações entre os sistemas espiritual e secular de grupos de idade manifestam-se na experiência cotidiana dos participantes espirituais, revelando o caráter plural e condicional da organização social Xavante.

Palavras-chave: Brasil, organização social, sistemas de grupos de idade, Xavante.

Introduction

Xavante society, like other Gê-speaking societies, is well-known for its dual structural aspects. Through the influential work of anthropologist David Maybury-Lewis (1967, 1979), political factionalism, characterized by a fundamental ideological bifurcation of society between people considered proximate and distant, came to be recognized as a primary feature of the social system. Also through his work, the Xavante came to be amply recognized as a South American example of an indigenous society with a formal age-group system, alternatively known as an age-set system or age-class system. These systems, which up to that time were better known from Africa, are characterized by age sets (cohorts or classes) that pass through a series of formal age grades (ranks) by means of public ceremonial events or rites of passage (Bernardi, 1985; Prins, 1953; Stewart, 1977). Formal age grades in these relatively uncommon systems differ from the informal age grades that are present in all societies and mark stages of the human life cycle based on individualistic and subjective criteria. Evidence of age-group systems in Gê societies was first mentioned for the Canela and Xerente (Lowie, 1939, 1946; Nimuendaju, 1942, 1946; Nimuendaju; Lowie, 1937, 1938).

One aspect of the Xavante age-group system makes it particularly interesting for discussions regarding the place of dual social morphologies in Gê societies, a topic that has received considerable scholarly attention (e.g., Bamberger, 1974; Carneiro da Cunha, 1982; DaMatta, 1983; Lave, 1975; Lévi-Strauss, 1944). In addition to the conjunction of age sets and formal age grades that comprise the age-group system, the age sets are allocated in alternation between two age-set moieties. The basic features of this system were first reported by Maybury-Lewis (1967) and I undertook its reevaluation in my doctoral dissertation (Welch, 2009).

In this system, throughout a lifetime, male and female members of age sets pass through a series of formal age grades, which entails pre-initiation (wapté),¹ novitiate adulthood (ritei'wa), and mature adulthood (iprédu). Promotion between age grades occurs collectively, as an age set, according to a public ritual calendar involving initiation rites (danhono) that occur about every five years. At the conclusion of these rites, a new age set is inaugurated and parents assign to it their young sons and daughters of roughly similar ages. For boys, age-set membership entails residing away from their natal households in a pre-initiate house. For girls, age-set membership does not stipulate a change of residence. Each age set bears a name from a pre-determined sequence of eight, such that about every forty years each age-set name is recycled. In the Pimentel Barbosa/Etênhiritipá community, where I did my fieldwork,²,³ the sequence of age-set names is tirowa, êt½pá, airere, hötörã, anharowa, sada'ro, abare'u, and nozö'u.⁴ Consequently there is a need to distinguish between senior and junior age sets that carry the same name. That is accomplished by appending the suffix -'rada ("first" or "old") to the elder of two groups bearing the same name. As these age-set names are applied in turn, they are also allocated in alternation between two unnamed age-set moieties that are associated with the alternating location of the pre-initiate boys' house on opposite sides of the village. Thus, the four age sets inaugurated into the pre-initiate house on one side are, in perpetual sequence, tirowa, airere, anharowa, and abare'u. The four age sets that cycle on the other side are, in turn, êt½pá, hötörã, sada'ro, and nozö'u.

Essential to this division into age-set moieties is a social morality of sidedness in how their members relate to one another in certain contexts. It stems from a privileged relationship between members of alternate age sets, which are adjacent age sets within the same age-set moiety. Members of the older of these two age sets serve as mentors (danhohui'wa) to members of the younger, who are their protégés (hö'wa nõri). The relationship between them is characterized by a profound sense of allegiance involving indulgence and mutual trust. These relationships last a lifetime, such that every second group of pre-initiate coresidents joins a chain of mentors and protégés connected through intimate bonds of friendship and respect. Thus, all members of an age-set moiety share a deep sense of loyalty. They form a single group of mutual interest and concern, keeping one another's secrets from members of the other moiety. They are all on the same side and designate each other "our age-set side" or "people on our age-set side" (waza'runiw)mhã). In contrast, members of the other age-set moiety are their rivals and disconfidants. Members of the each moiety call the rival moiety "the other age-set side" or "people on the other age-set side" (hö'amoniw)mhã).

Scholars of Xavante social organization after Maybury-Lewis also mention another age-graded system in the domain of male spiritual life (Giaccaria; Heide, 1984; Müller, 1976; Welch, 2009). According to those accounts, males are introduced to the spiritual system during an initiation ritual (darini), which occurs about every 15 years, and subsequently pass through a sequence of four spiritual grades over the course of a lifetime. Different from descriptions by other scholars, my data indicate that this spiritual system also involves age sets according to a pattern of alternation, whereby successive spiritual age sets are allocated into two age-set moieties. It thus constitutes a formal age-group system, making the Xavante an exceptional case of a society with two independently operating formal age-group systems. This spiritual system marks social age in an analogous manner to the secular system, although in a categorically different domain of social relations. This finding deserves special note because it suggests that the pattern of moieties generated through alternation of age sets, previously documented only in the secular age-group system, is a characteristically Xavante social morphology in multiple fields of social relations.

The objective of this paper is to present the spiritual life cycle and discuss its relationship to other dimensions of Xavante social organization. In the first section, I describe the spiritual hierarchy, presenting evidence that it constitutes a formal age-group system. In the second section, I argue that Xavante society entails a logic of hierarchy and symmetry as mutually constructed and non-contradictory. In the third section, I discuss interactions between the spiritual and secular age-group systems in order to demonstrate that plurality and contingency are essential features of Xavante social organization. In the conclusion, I draw on scholarship regarding other Gê-speaking societies to suggest that the plurality and non-opposition of hierarchical and symmetrical social morphologies is not unique to the Xavante.

I base my arguments on both social structure and social practice.⁵ This approach derives principally from my field experience with the Xavante at Pimentel Barbosa/Etênhiritipá, in which I encountered "social structure" to be so overt and tangible, expressed abundantly in everyday speech and action, that I could not escape considering it to be a relevant factor in social reality. In the more informal moments of life, I found there to be rich expression of some of the more nuanced aspects of social life at the interface between structure and practice. For the Xavante, age hierarchies, in particular, are not abstractions; they are conspicuously part of how people see and interact with each other.

The spiritual age-group system

As a researcher at Pimentel Barbosa/Etênhiritipá, I was incorporated into the social system as a male member of a specific secular age set (êt½pá), which then occupied the novitiate adult age grade (ritei'wa), and as a member of the initiate spiritual age grade (wai'ãra). Much of my social interaction with the community was flavored by the collateral effects of assuming those social statuses. Consequently, those social bearings also inform this paper because my reading of Xavante social organization was affected by my particular position within it. When I visit the village, I am encouraged to participate in spiritual rituals (wai'a), not only to improve my understanding of Xavante society, but also because that is what is expected of me as a man. I participated in one spiritual ritual during my first visit to the field, and then many more times over the six years I have worked with the Xavante. On numerous occasions, including the ritual during my first visit, I was selected for the grueling task of carrying sacred cane arrows (ti'ipê) while dancing and singing from late afternoon until after sunrise the following day. I cannot say I enjoyed those long nights, but through them I came to appreciate in a personal manner the camaraderie and solidarity that exists between age peers.

A form of these spiritual rituals that was commonly performed during my fieldwork begins in a forest clearing between the village and nearby river. Some participants arrive in the early morning, before sunrise, while others arrive in the early afternoon. Women are not allowed in this clearing, even though the activities that occur there are not, strictly speaking, considered secret.⁶ Women do not witness what occurs there, but they may hear about it afterwards. The men spend a number of hours in the clearing, during which spiritual singers (zö'ra'si'wa) maintain a mesmerizing pulse of song while sitting in a circle at the center of the clearing. Meanwhile, spiritual initiates (wai'ãra) stand around the upper half of the clearing and spiritual guards (dama'ai'a'wa) make repeated passes in front of them, in which they swing their arms in stylized shows of aggression and sometimes stomp forcefully on initiates' feet. All the while, spiritual post-officiants (wai'a'rada) sit at the far periphery of the clearing talking amongst themselves and gleefully heckling the participants.

Subsequently, the entire contingent of male spiritual participants and elders return to the village. There, a select set of about six to eight initiates, each carrying a sacred cane arrow (ti'ipê) fletched with avian down and laced with poison pigments, run from the center of the plaza to a separate location in the forest, where secret activities are held away from eyes and ears of women and uninitiated males. Meanwhile, the other initiates, guards, and singers continue singing in a wide arc as they pass in front of the houses that face the plaza. As the sun begins to descend and the first round of public singing is complete, initiates return to their houses, where they collect food items prepared by their mothers, aunts and grandmothers, and deliver them to the guards in the center of the plaza. The guards retrieve these corn loafs, cooked wild tubers, bottles of soda, and other treats, from two communal piles and, sharing them with their elders, consume them amidst animated conversation.

The singing continues throughout the night, which often can be quite cold in the region of the Serra do Roncador, Mato Grosso. However, not everyone remains. Singers, guards, and post-officiants often retire to their homes to enjoy full nights of sleep. Even the majority of initiates go home, as they wish, to rest up for the continuation of the ritual the following morning. The only individuals required to remain are the six to eight initiates selected to carry the sacred arrows and sing throughout the night, which is a most difficult task during the early morning hours.

Although only those few initiates are obliged to remain all night, they are rarely left completely alone. There are usually several guards about to make sure they maintain the rhythm of their song, continue the seemingly endless rounds around the village with only minimal time to rest, and stay away from such comforts as coffee or a warming fire that burns in the middle of the plaza. Also often present are a few diehard post-officiants, who remain near the fire in the throes of happy nighttime conversations, and occasionally initiates who were not chosen to carry sacred arrows but wish to offer support to their age mates. At times, when the strains of cold and lack of sleep cause the initiates to drop the rhythm or sonority of their song, someone wrests himself from sleep, returning to the plaza to help them recuperate the performance. Eerie sounds, which only occur on occasions of spiritual rituals, sometimes emanate from the forest near the village and suggest the veiled presence of other entities in the vicinity.

At the first detectible signs of morning light, the small circle of initiates is gradually rejoined by the men who slept during the night. One-by-one, initiates, guards, and singers step back in and add their voices to the tired songs of those initiates who did not have the luxury of sleep. Once the sun is up, singing is concluded and the sacred cane arrows are retrieved from the select initiates who cared for them throughout the night. Often, in the cold early morning, arrow competitions mark the end of the ritual. In these, guards in hiding behind the ring of houses shoot special ceremonial arrows in tall arches into the village plaza. These arrows are longer from hardwood tip to bamboo base than an adult man is tall and sport showy feather fletching, interspersed with strips of snake skin, and two tufts of hair at the base. Initiates rush to grab them from the air before hitting the ground, a dangerous activity but one that carries the reward of a gratifying public cheer. Then, once again, initiates return home to fetch food items for their elders.

The difficult task of carrying sacred arrows while singing all night is only assigned to senior members of the initiate age grade (wai'ãra ipredumrini). This is because it is considered as somber a responsibility as it is a challenging one because it must be carried out to completion without fail. Staving off sleep, enduring the cold, maintaining vocal syntax and cadence, and carrying sacred cane arrows (ti'ipê) throughout the entire night require extreme force of will. It is a grueling chore bemoaned by most initiates. Yet, performing that chore is an essential part of their spiritual training. It is a test of their resilience, a means of acquiring spiritual strength, and a prerequisite to advancing to the grade of spiritual guards.

The stark differences between the ceremonial roles of initiates, guards, singers, and post-officiants, described above, are evidence of a highly formalized system of age grades that informs all aspects of Xavante spirituality. External to that system is the category of boys who have not yet been ritually initiated into the spiritual system. They are considered spiritual pre-initiates (wautop'tu). In my experience, spiritual pre-initiates do not in any way participate in spiritual rituals, except as spectators, when allowed. I do not consider spiritual pre-initiation to be a formal spiritual grade because it denotes those individuals who are excluded from the spiritual system. However, that distinction is arbitrary.

Spiritual initiates (wai'ãra) are those young males who were initiated during the most recent 15-yearly initiation rites and are the youngest participants in spiritual ceremonies. Spiritual initiates are expected to attend spiritual rituals on a regular basis in order to improve their stamina as a condition for spiritual receptivity. Although spiritual rituals may be held for numerous reasons and follow several different formats, just one of which was described above, they all include an element of regimentation for initiates, including dietary restrictions with public chastisement for transgressions and ceremonial responsibility for singing for extended periods under physical stresses, such as lack of sleep and extreme temperatures.

The next spiritual grade in ascending order is that of "guards" or "soldiers" (dama'ai'a'wa), those translations being the ones typically used by the Xavante (guardas and soldados, in Portuguese), are who are promoted from the initiate grade in the same 15-yearly rites that marks the entrance of initiates into the system. In the context of spiritual rituals, guards watch over the initiates as enforcers. They punish dietary transgressions and maintain a presence during spiritual rituals to ensure that initiates carry out their duties without indulging in sleep, drinking coffee or alcohol, or seeking heat during the coldest hours of the night. Indicative of their authority over initiates are bows and arrows that they carry and not infrequently use with jabbing motions against misbehaving initiates. Thus, guards, consistent with the militaristic nomenclature of their status, assume an overtly antagonistic stance towards initiates but arguably do so to encourage their spiritual betterment. Like initiates, guards are not responsible for conducting the spiritual ceremonies per se, but direct certain aspects, including those that specifically have to do with maintaining vigilance over initiate behavior.

Actively conducting the spiritual ceremonies are "singers" (zö'ra'si'wa). No longer responsible for presiding antagonistically over initiates, singers do the core work of officiating spiritual rituals. For several hours from early dawn before each spiritual ritual, singers gather in a small clearing between the village and the river. Already painted, adorned, and with palpably focused states of mind, they greet the early morning light sitting in a small circle, often around a smoldering fire. At their backs, planted in the ground, stand sturdy ceremonial clubs, from which are suspended small bundles of personal items. They shake rattles vigorously, incessantly, while untiringly repeating a song that was unveiled for this one and only occasion. Despite the impressive visual and auditory potency of their performances, it would seem that the primary work of singers is introspective. They resume that work later in the afternoon and continue it as the entire progression of guards threateningly dances past the line of initiates, who outstretch their feet so the guards might stomp on them as punishment for eventual dietary transgressions. Later in the afternoon and again the next morning, back in the village, the singers lead members of the younger spiritual grades in singing around the village, reinvigorating tired initiates and guards with their determined and authoritative bearings.

The most senior spiritual grade is that of post-officiants or elders (wai'a'rada), who were initiated into the spiritual system prior to the third most recent spiritual initiation rites. Having already served as initiates, guards, and singers, post-officiants have no formal role to play in spiritual rituals. Their attendance is optional and their participation often limited to the sidelines. They are considered something like spiritual retirees, having left that work in the hands of well-prepared younger individuals. Yet, post-officiants are not superfluous to spiritual endeavors. Besides watching spiritual rituals, they often help in their organization and may, should they choose, speak up to counsel or give advice to participants. Also, although post-officiants watch spiritual ceremonies from the sidelines, they do so vocally, heckling poor performances to motivate improvement. Such critique is not reserved for initiates or even guards, as even singers may be ridiculed by post-officiants should opportunity arise. Accordingly, post-officiants remain senior authorities in spiritual rituals despite being formally excluded as participants. With an initiation interval of about 15 years, the entire succession from initiate to post-officiant status may take approximately 45 years. Depending on the age at which a young male is first initiated into the spiritual hierarchy, attaining post-officiant status may occur relatively late in life.

Notions of age, seniority, authority, and responsibility are all implicated in the spiritual hierarchy. Whether one is a pre-initiate, initiate, guard, singer, or post-officiant, his level of spiritual seniority has a bearing on how he interacts with others during spiritual rituals and in daily social life. That dynamic is especially apparent in the special antagonistic relationship between guards and initiates, whereby guards have explicit juridical authority to enforce upon initiates certain rules of behavior. Specifically, they prohibit initiates from eating certain common foods, most notably the meat of numerous species of game animals. Guards must liberate those foods one-by-one for individual initiates before they may be freely eaten without fear of reprise. Should guards not liberate a food, the prohibition does not terminate when one ceases to be an initiate. One 47-year old post-officiant told me he was not yet authorized to eat marsh deer (Blastocerus dichotomus), known as cervo do pantanal in Portuguese, or traira fish (Hoplias sp.). Such hierarchical authority, expressed through ceaseless surveillance and punitive retribution, impresses upon younger males the necessity of behavioral obedience.

The oppositional stance of guards towards initiates finds stark contrast in the very different sort of relationship that exists between singers and initiates. Whereas guards administer to initiates antagonistically, singers do not publicly attend to them but are, in fact, their closest spiritual allies and private spiritual mentors. They are considered both responsible for and together with (dasiré) members of the initiate age grade. In private spaces and in hushed voices, singers counsel initiates in secret aspects of Xavante spirituality. Shared between them is a sense of shared mutual concern and absolute trust. Notably, singers indulge initiates in keeping the secret of their violations of the dietary restrictions imposed by guards. Thus, singers' interest in initiates' spiritual development is of a categorically different nature than that of guards.

The contrastive social dynamics between guards and initiates, on the one hand, and singers and initiates, on the other hand, entails a structural logic of alternation similar to that described above for the secular age-group system. As in the secular age-group system, each of the spiritual grades is occupied by spiritual age sets, which are life-long associations defined at the time of one's first initiation into the spiritual system. Through the collective fifteen-yearly spiritual initiation rituals (darini) their members are recruited and they comprise social units for the greater part of adult life. Members of spiritual sets learn spiritual knowledge together through the mutual experience of spiritual trials and the common exercise of age-determined spiritual prerogatives. The unity of their membership is expressed corporally through body painting and performance roles specific to their spiritual grades and socially through an explicit morality of secrecy and solidarity regarding spiritual matters.

Unlike secular age sets, spiritual age sets are not named.⁷ Consequently, the cyclical aspect of eight rotating secular age-set names has no parallel in the spiritual system. Also, men tend to discuss spiritual matters in oblique terms and beyond the auditory range of other people, with the result that spiritual age sets are somewhat invisible in the context of everyday speech. Despite being unnamed and infrequently mentioned in public, spiritual age sets are explicitly recognized social groups in the same sense as secular age sets. I make that assertion based not only on observation, but also on conversations with spiritual participants in which they explicitly described the system in terms of cohorts that pass between age grades and are allocated in alternation between two opposing sides. In this context, Xavante individuals used specific terms to refer to one's own spiritual age-set moiety (wasiré wai'a) and its complement (wai'a amo). They also spoke of the structural parallel between the spiritual and secular age-grade systems, noting that the operation of alternation is similar in both, despite the content of the spiritual domain not being comparable to that of the secular domain.

The social solidarity between singers and initiates also extends to alternate spiritual sets such that every second spiritual set in the initiation sequence is linked by a sense of mutual identity and a bond of sociospiritual unity. That unity establishes them as spiritual companions and couples them in a pedagogical mentoring relationship whereby singers are responsible for the indulgent cultivation of initiates' spiritual capacities. They are the initiates' sources of spiritual knowledge and guidance. Between them, spiritual subjects may be discussed openly, forbidden words may be voiced, and spiritual transgressions may be acknowledged without fear of recrimination. Alternate spiritual sets are allies, in that specific context, and are bound by a code of mutual duty. Members of the same spiritual moieties help one another, share with one another, and trust one another.

Whereas members of alternate spiritual sets are comrades, members of adjacent sets are adversaries, as illustrated most explicitly by the social dynamic between guards and initiates. Through the repetitive allocation of age sets between opposing moieties, alternate age sets are joined in separate and opposed chains of solidarity. The two groups are implicated in a total bifurcation of spiritual information rooted in an ethos of distrust and constraint. Members of opposite spiritual moieties should not, and in my experience do not, exchange spiritual knowledge. This restriction is a somber one, as spiritual information is so carefully guarded precisely to protect it from theft by those to whom it does not rightfully belong. Theft of knowledge is thought to threaten its potency for its original owners and grave sanctions are said to be levied against those who threaten its security. Operating in secrecy from one another, opposing spiritual moieties maintain their own, independent, bodies of spiritual knowledge and agendas of action. When I inquired into the subject, members of both sides asserted in no uncertain terms that neither they nor anyone else had any idea what was being taught in the other moiety. In effect, they function as two separate but equal spiritual traditions.

Hierarchy and symmetry

Among the explanations offered for age-group systems is that they facilitate militaristic organization. Some scholars argue that they serve as alternate forms of political integration in societies that lack centralized authority while others propose that they provide a standard mode of military recruitment between otherwise unrelated societies (Bernardi, 1952; Eisenstadt, 1954; Hanson, 1988; LeVine; Sangree, 1962). Based on a global cross-cultural examination of age-group organization, Ritter (1980) concluded that age-group organization serves to integrate men in societies that face constant warfare and in which local group composition fluctuates due to ecological conditions. In order to substantiate her hypothesis ethnographically and in South American context, Ritter presented Xavante age organization as a case example. Common to each of those proposals is the idea that age sets have a sociopolitically integrative function in warfare settings. However, they differ regarding the nature of its integrative role. Eisenstadt (1954) and Bernardi (1952) suggest it organizes people vertically through authority while LeVine and Sangree (1962) suggest it organizes people horizontally through recruitment.

My contemporary ethnographic data do not permit historical reconstruction of the origins of Xavante age-group systems and therefore do not allow me to confirm or refute these hypotheses. However, the apparent disagreement regarding the nature of age-set integration reveals a scholarly debate about whether age hierarchies construct vertical or horizontal social relationships and illustrates the complex nature of the relationship between social proximity and distance in age organization. Responding to dilemma, I submit that Xavante age organization involves an integration of these ostensibly contrastive principles. In both age-group systems, hierarchical and differentiating dimensions simultaneously implicate principles of symmetry and similarity. That is because both the secular and spiritual age-set moieties are generated through the passage of age sets through age grades. Said another way, symmetrical moieties and age-set ranks exist through and define one another. In this sense, in these two age-group systems, dual symmetrical features (moieties) are inseparable from hierarchical features (age grades and age sets). Thus, in those contexts, Xavante culture deemphasizes the apparent antithesis of each pair in favor of a formulation that prioritizes their mutuality.

My proposal echoes other scholars' characterizations of Gê social dualism as congruent with or presupposing asymmetrical structures (e.g., DaMatta, 1983; Lévi-Strauss, 1944, 1956; Seeger, 1989; Turner, 1984). However, such discussions generally refer to asymmetry between pairs or in relation to an external point of reference. In the examples presented in this paper, the form of dualism is categorical, based on membership in explicit social groups (moieties), each of which is inherently symmetrical with reference to the other. This symmetry derives from the diachronic aspect of the age-set alternation, whereby each phase of the repetitive dynamic is structurally equivalent to all others and each age set will, through time, occupy all positions in the system. It is reflected in the terminological reciprocality of the two sides as an opposition between "our side" and "their side," with no absolute terminological reference for either moiety. At any given time the arrangement of age sets in age grades will lend a temporary asymmetrical aspect to the moieties, but the cyclical nature of that arrangement means that all age sets and both moieties will in their turn assume all of the available positions.

In these systems, the sources of asymmetry lie in a bifurcation of hierarchy between ranked age sets within age-set moieties, which share a sense of social solidarity, and ranked age sets between moieties, which assume rivalrous stances towards one another. One possible reading of this form of hierarchy is that it contributes to a particularly Xavante sociology of knowledge transmission that involves contrastive methods of indulgence and tough-love. For example, on the one hand, spiritual mentors cultivate in their socially proximate protégés the knowledge and skills necessary for spiritual maturation through social proximity. On the other hand, spiritual guards help transform initiates into spiritually capable adults by keeping strict vigil and discouraging misbehavior. According to this view, although age asymmetrical relations between opposite-moiety adversaries differ starkly from those between same-moiety allies, both are examples of respect relationships, whereby individuals of different ages treat each other with socially sanctioned forms of deference or esteem (Murdock, 1949; Schusky, 1965). In the Xavante case, respect relationships are bifurcated into social solidarity and adversarial dynamics for the mutual goal of transforming youth into responsible and capable adults.

A plurality of age systems

The relevance of recognizing that Xavante social organization involves two distinct age-group systems lies not only in the logical and operational parallels that can be drawn between them, but also in their very plurality. Maybury-Lewis called attention to the plurality and apparent flexibility of Xavante age organization and other social means of construing social identity and difference, such as moiety systems (Maybury-Lewis, 1967). Silva similarly recognized the multiplicity of Xavante social structures and questioned what theoretical conclusions might be drawn from their fluid realities (Lopes da Silva, 1986). This plurality of age structures is apparent not only in the age-group systems discussed in this paper, but also in other dimensions of Xavante age organization. For example, the Xavante life cycle also includes informal age grades that do not pertain to the formal secular age-group system (Lopes da Silva, 1986; Maybury-Lewis, 1967; Müller, 1976; Welch, 2009). Also, evident in Maybury-Lewis's (1967) presentation of the Xavante kinship terminology are abundant terms that distinguish relative age and which suggest that seniority is a pervasive component of Xavante notions of relatedness. Furthermore, accounts of Xavante political dynamics attest the importance of seniority-based esteem for leadership and decision-making (Graham, 1995; Welch, 2009).

A similar observation may be made about Xavante dual organization. Maybury-Lewis (1967, p. 296) wrote that the "most immediately striking feature of Gê societies is their proliferation of moiety systems." Indeed, my data attest to at least Xavante five moiety systems (Welch, 2009). The first two stem from the spiritual and secular age-group systems, as described in this paper. A third also pertains to the spiritual complex and divides male participants into Wood-Owners (wedehöri'wa) and Rattle-Owners (umr½'tede'wa) according to their fathers' perceptions of their likely adult physical forms as either shorter and squatter or taller and thinner, respectively. As Maybury-Lewis (1962, p. 136) described, the difference between Rattle-Owners and Wood-Owners is they "involve intercession with two classes of spirits, one conferring generative power (life) and the other aggressive power (death)." Rattle-Owners appeal to a benevolent spirit (danhimi'te) that is associated with the sacred cane arrows carried by senior initiates (ti'ipê). Wood-Owners fight and dominate a malevolent spirit called (simi'hö'pãri) that is responsible for destructive winds. A fourth moiety system involves two ranked patrilineal segments, Tadpole (poreza'õno) and Big Water (öwawe), which are exogamous and therefore inform marriage arrangements (Maybury-Lewis, 1967). As one member of the Big Water moiety explained to me, "We are the servants of the Tadpole moiety. We defend them and sweep up after them." The Big Water moiety affords the Tadpole moiety ultimate decision-making authority, with the right to appoint official leaders from among their own ranks. The fifth is a division of the physical village into left (danhimi'e) and right (danhimire) sides, based on house location, which serves to organize village-wide distributions of food and goods (half of the goods are distributed within the left side of the village and half are distributed within the right side).

This multiplicity of systems attests to another feature of Xavante social organization, which is that it is extremely contingent. Xavante social reality entails such a diversity of simultaneous hierarchical and symmetrical morphologies that it is highly multidimensional. The spiritual and secular age-group systems are neatly illustrative of this plurality. In both cases, rank involves the presumptive seniority of maturity according to two contrastive configurations, one collegial and one authoritative. Considered together, they also contribute to a complex tapestry of social relations, in which there are no absolute power relationships, as between, for example, younger or older, junior or senior, and subordinate or dominant. Nor is there any absolute equivalence between age equals. Rather, plural age classifications unite and differentiate individuals in multiple and conditional ways that influence their modes of engagement in diverse social settings. With numerous age hierarchies and oppositions operating at the same time, the social significance of each is contingent upon the totality of any given social dynamic, including the full complement of other age-based relationships and how the individual actors involved make sense of them. Examples are in order.

The decision to initiate a boy into the spiritual system and thereby assign him to a spiritual age set lies with his father and is constrained by the relative infrequence of opportunities to do so, since spiritual initiation rituals (darini) tend to be held about every 15 years. Accordingly, each spiritual set includes members of diverse secular age sets, which are inaugurated about every five years, according to no formal system of correspondence between them. Thus, rank in one system may be equalized or inverted in the other and moiety opposition in one system may be transformed into moiety solidarity in the other.

At times, contrastive structural relationships between the same people are compartmentalized according to a logic of independent domains. For example, a moment of camaraderie and indulgence between a secular age-set mentor (danhohui'wa) and his protégé (hö'wa nõri) may be followed, a short time later, by a grave act of punishment between the same two people in a spiritual ritual (wai'a). The two events may be interpreted by those involved as having nothing to do with one another, with neither intimacy nor authority compromising the other in their respective settings.

At other times, seemingly contradictory relationships in different social domains are allowed to influence one another. For example, should a spiritual guard (dama'ai'a'wa) discover that a spiritual initiate (wai'ãra), his age-set mate in the secular system, ate a prohibited food, he may choose not to punish him in deference to their secular sense of comradeship. Conversely, that same spiritual initiate may refrain from eating a prohibited food in front of the spiritual guard, despite his being an otherwise trusted secular age-set mate, out of respect for his superior spiritual status. Secrecy, essential to the Xavante morality of age associations, is extremely useful in this respect because it may be employed to mitigate or compartmentalize age relations. Using the same example, the spiritual initiate may choose to hide his transgressions from his age-set mate to avoid being punished or the spiritual guard may choose to guard as secret the age-set mate's transgressions to protect him from punishment. Compartmentalization and accommodation of age relations between domains are central to their sustained practical and ideological compatibility.

Following the same principles, contrastive age status in one domain may serve to reinforce seniority or juniority of shared age status in another domain at the same time that shared status in one domain is not overtly thought to affect the integrity of rank in another. For example, although there is no explicit coordination between the spiritual and secular age-group systems, members of specific secular age sets dominate the older and younger segments of spiritual age sets. Thus, when I did my primary research, members of the secular êt½pá age set exclusively enjoyed status in the spiritual system as senior initiates (wai'ãra ipredumrini). As a result, their spiritual activities were often attributed to them as members of a secular rather than a spiritual age-set. However, several years later, some members of the next secular age set (tirowa) had also become senior initiates, causing that spiritual stage to be occupied by members of two different secular age sets, which in that system were adjacent and in opposite moieties. Although the members of those two secular age sets treated each other as rivals in some other contexts, during spiritual rituals they were joined in a project of mutual concern and a shared trial of the highest order. There they stood side-by-side in equal subordination to spiritual guards, who had policing and punishment authorities over them all. Thus, age status in one domain may fragment the solidarity of age status in another or, alternatively, unify members of otherwise antithetic age sets.

These examples of multiplex hierarchical and oppositional relationships that variously unite and separate individuals within society point to a pervasive feature of Xavante sociality - there is no single point of reference for constructing inclusive and exclusive identity categories. Every outsider is also simultaneously an insider. Every equal is also a senior or a junior. Status is not absolute and is not fixed; it is contingent, transitory, and circumstantial. The Xavante reality is that identity is multiple, that each formulation of identity has its place and time, and that individuals of all ages have the autonomy to construe those formulations as they see fit.

The notions of rank and sidedness developed here in the contexts of the secular and spiritual age-grade systems are just two among many dimensions of sociality that texture how people view and interact with one another. Abundant other systems of social organization are similarly ubiquitous aspects of the Xavante experience, providing abundant means to unify certain sets of people as equals or similars and order others as unequals or dissimilars. Some of those assume hierarchical or dual morphologies while others do not. This plurality of social relations is not unique to the Xavante; the same point might be made about any social network in which people share multiple relationships that bear on their interactions differently in different social contexts. Nevertheless, it is an especially important point in relation to the Xavante life cycle because it entails such a notable plurality of highly formalized forms of age organization.

Conclusion

I have argued that the Xavante spiritual age-group system involves a logic of hierarchy and symmetry as mutually constituted that mirrors analogous organizational principles in the secular age-group system. I also discussed how the interaction of those two systems manifests in daily social engagement in order to illustrate my assertion that important aspects of Xavante social organization are its plurality and the resultant contingency that permeates the social experience. In the Xavante social world, multiplex systems of equality and difference are essential to how people understand the human life cycle.

The theme of plural age and moiety organization is also apparent in ethnographic literature regarding other Gê groups. Some scholars respond to the ethnographic fact of plurality by asserting that some single structure is dominant or causal, thus essentially denying its theoretical relevance (e.g., DaMatta, 1973, 1979, 1983; Carneiro da Cunha, 1978; Coelho de Souza, 2002). Nevertheless, multidimensionality is apparent in many Gê social systems. For example, although da Matta attributed multiple Apinayé social structures to a single underlying opposition, he simultaneously made the assessment that, "The social world of the Northern Gê is disjointed and its dual ideology permits multiple readings of social reality" (DaMatta, 1976, p. 247). Similarly, although Turner (1965) explained Kayapó social structures in terms of a single underlying socioeconomic and ideological dynamic, he also ascertained the abundance of contingent human relationships and recognized that social organization simultaneously unites and divides society. These examples attest to a theoretical difficulty in reconciling structural plurality with singularity.

There may be a tendency on the part of some scholars to assume that plurality and heterogeneity implies contradiction. That possibility is apparent in scholarship that asserts the impossibility of simultaneous social roles. For example, Turner (1965, p. 412) wrote that in Kayapó society, for a man to become a husband and father he must cease to be a brother. Similarly, Jackson (1975, p. 318) identified one way of understanding the variability of Gê social alignments as a "surplus" of memberships, whereby for one to be activated others necessarily must be inactivated. Those characterizations emphasize a dynamic whereby one dimension of social reality is emphasized at the expense of another.

My research suggests that such compartmentalization is not the only possible strategy for reconciling the multiplicity of social relations that exist between individuals. Another involves their simultaneous integration by allowing status in one domain to affect status in another. That dynamic is more closely aligned with the theoretical position that social organization is thoroughly plural and simultaneous. Crocker and Crocker (2004) illustrated that position when they accounted for multiple Canela age hierarchies and dual structures as alternative systems that serve to counterbalance one another for the sake of greater overall social solidarity. Melatti (1970, 1978, 1979) took a similar position in arguing that in Krahô society, multiple social and ideological configurations negate one another, thereby denying social contrast and increasing the overall equality of individuals. An important insight in Melatti's work is his description of the availability of multiple social perspectives. As he wrote, "Krahô rites appear to give individuals the possibility to view social relations and the relations between the elements of the Universe, as they imagine them, from different points of view" (Melatti, 1978, p. 357). These theoretical positions are important because they recognize the irreducibility of multiplex social organization.

Fisher (1991) took these ideas even further in describing Kayapó social structure as fundamentally contingent. According to his evaluation, the plurality of social structure is both simultaneous and mutually exclusive such that in the kinship realm "there is an internal relation between hierarchy, equality, and identity" (Fisher, 1991, p. 480). According to his view, one can simultaneously assume multiple roles and can switch between them in turn, a paradox that contributes to the conditionality of social status and the unpredictability of social action. Such is also the case in Xavante society, where multiple systems for reckoning age in absolute and relative terms contribute to a complex terrain of social unity and differentiation. It is a social landscape that stimulates and denies the distinctions between such heterogeneous oppositions as hierarchy and equality, separation and integration, individuality and collectivity, and similarity and difference.

Recebido em: 26/02/2010

Aprovado em: 03/06/2010

• BAMBERGER, J. Naming and the transmission of status in a Central Brazilian society. Ethnology, v. 13, n. 4, p. 363-378, 1974.
• BERNARDI, B. The age-system of the Nilo-Hamitic peoples: a critical evaluation. Africa: Journal of the International African Institute, v. 22, n. 4, p. 316-332, 1952.
• BERNARDI, B. Age class systems: social institutions and politics based on age. Cambridge: Cambridge University Press, 1985.
• BOURDIEU, P. Outline of a theory of practice Cambridge: Cambridge University Press, 1972.
• CARNEIRO DA CUNHA, M. Os mortos e os outros: uma análise do sistema funerário e da noção de pessoa entre os índios Krahó. São Paulo: Hucitec, 1978.
• CARNEIRO DA CUNHA, M. Eschatology among the Krahó: reflection upon society, free field of fabulation. In: KING, H.; HUMPHREYS, S. C. Mortality and immortality: the anthropology and archaeology of death. London: Academic Press, 1982. p. 161-174.
• COELHO DE SOUZA, M. O traço e o círculo: o conceito de parentesco entre os Jê e seus antropólogos. Dissertation (Ph.D. in Social Anthropology)-Museu Nacional/Universidade Federal do Rio de Janeiro, Rio de Janeiro, 2002.
• CROCKER, W. H.; CROCKER, J. G. The Canela: kinship, ritual, and sex in an Amazonian tribe. Belmont: Wadsworth, 2004.
• DAMATTA, R. A reconsideration of Apinayé social morphology. In: GROSS, D. R. Peoples and cultures of native South America Garden City: Doubleday, 1973. p. 277-291.
• DAMATTA, R. Um mundo dividido: a estrutura social dos índios Apinayé. Petrópolis: Vozes, 1976.
• DAMATTA, R. The Apinayé relationship system: terminology and ideology. In: MAYBURY-LEWIS, D. Dialectical societies: the Gê and Bororo of Central Brazil. Cambridge: Harvard University Press, 1979. p. 83-127.
• DAMATTA, R. A divided world: Apinaye social structure. Cambridge: Harvard University Press, 1983.
• EISENSTADT, S. N. African age groups: a comparative study. Africa: Journal of the International African Institute, v. 24, n. 2, p. 100-113, 1954.
• FISHER, W. H. Dualism and its discontents: social organization and village fissioning among the Xikrin-Kayapo of Central Brazil. Dissertation. (Ph.D. in Anthropology)-Cornell University, Ithaca, 1991.
• FUCHS, S. Beyond agency. Sociological Theory, v. 19, n. 1, p. 24-40, 2001.
• GIACCARIA, B.; HEIDE, A. Xavante: auw½ uptabi, povo autêntico. São Paulo: Editora Salesiana Dom Bosco, 1984.
• GIDDENS, A. Central problems in social theory: action, structure and contradiction in social analysis. Cambridge: Cambridge University Press, 1979.
• GRAHAM, L. R. Performing dreams: discourses of immortality among the Xavante of Central Brazil. Austin: University of Texas Press, 1995.
• HALL, J.; McLEOD, R. A; MITCHELL, V. Pequeno dicionário Xavante-português português-Xavante Brasília: Summer Institute of Linguistics, 1987.
• HANSON, J. R. Age-set theory and Plains Indian age-grading: a critical review and revision. American Ethnologist, v. 15, n. 2, p. 349-364, 1988.
• HARRIS, G. G. Concepts of individual, self, and person in description and analysis. American Anthropologist, v. 91, n. 3, p. 599-612, 1989.
• JACKSON, J. E. Recent ethnography of indigenous northern lowland South America. Annual Review of Anthropology, v. 4, p. 307-340, 1975.
• LACHNITT, J. Dicionário Xavante/português: romnhitsi'ubumro a'uw½ mreme - waradzu mreme. Campo Grande: Missão Salesiana de Mato Grosso/Universidade Católica Dom Bosco, 2003.
• LAVE, J. C. Inter-moiety systems: a structural explanation of the proliferation of Ramkokamekra ceremonial institutions, Social Science Working Papers 69a and 69b. Irvine: University of California, Irvine, 1975.
• LÉVI-STRAUSS, C. On dual organization in South America. América Indígena, v. 4, p. 37-47, 1944.
• LÉVI-STRAUSS, C. Les organisations dualistes, existent-elles? Bijdragen tot de Taal-, Land- en Volkenkunde, v. 1122, p. 99-128, 1956.
• LÉVI-STRAUSS, C. Structural Anthropology New York: Basic Books, 1963.
• LEVINE, R. A.; SANGREE, W. H. The diffusion of age group organization in East Africa: a controlled comparison. Africa: Journal of the International African Institute, v. 32, p. 97-110, 1962.
• LOPES DA SILVA, A. Nomes e amigos: da prática xavante a uma reflexão sobre os Jê. São Paulo: Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, 1986.
• LOWIE, R. H. The associations of the erénte. American Anthropologist, v. 41, n. 3, p. 408-415, 1939.
• LOWIE, R. H. The northwestern and central Gê. In: STEWARD, J. H. Handbook of South American Indians, vol. 1: the marginal tribes. Washington D.C.: United States Government Printing Office, 1946. p. 477-517.
• MAYBURY-LEWIS, D. The Akw½-Shavante: a test case of 'dual organization' in Central Brazil. In: XXXV CONGRESO Internacional de Americanistas, Mexico, 1962: actas y memorias, v. 3. México: Editorial Libros de México, 1962. p. 135-136.
• MAYBURY-LEWIS, D. Akw½-Shavante society Oxford: Clarendon Press, 1967.
• MAYBURY-LEWIS, D. Cultural categories of the central Gê. In: MAYBURY-LEWIS, D. Dialectical societies: the Gê and Bororo of Central Brazil. Cambridge: Harvard University Press, 1979. p. 218-248.
• MELATTI, J. C. O sistema social krahó Dissertation (Doctorate in Anthropology)-Universidade de São Paulo, São Paulo, 1970.
• MELATTI, J. C. Ritos de uma tribo timbira São Paulo: Ática, 1978.
• MELATTI, J. C. The relationship system of the Krahó. In: MAYBURY-LEWIS, D. Dialectical societies: the Gê and Bororo of Central Brazil. Cambridge: Harvard University Press, 1979. p. 46-82.
• MÜLLER, R. A. P. A pintura do corpo e os ornamentos xavante: arte visual e comunicação social. Thesis (Master's in Anthropology)-Universidade Estadual de Campinas, Campinas, 1976.
• MURDOCK, G. P. Social structure New York: The Free Press, 1949.
• NIMUENDAJU, C. The Serente Los Angeles: Frederick Webb Hodge Anniversary Publication Fund, 1942.
• NIMUENDAJU, C. The Eastern Timbira Berkeley: University of California Press, 1946.
• NIMUENDAJU, C.; LOWIE, R. H. The dual organization of the Ramkokamekra (Canella) of northern Brazil. American Anthropologist, v. 39, p. 565-582, 1937.
• NIMUENDAJU, C.; LOWIE, R. H. The social structure of the Ramko'kamekra (Canella). American Anthropologist, v. 40, n. 1, p. 51-74, 1938.
• ORTNER, S. B. Theory in anthropology since the sixties. Comparative Studies in Society and History, v. 26, n. 1, p. 126-166, 1984.
• PARSONS, T. The social system New York: The Free Press, 1951.
• PRINS, A. H. J. East African age-class systems Groningen: Djakarta, 1953.
• RADCLIFFE-BROWN, A. R. On social structure. The Journal of the Royal Anthropological Institute of Great Britain and Ireland, v. 70, n. 1, p. 1-12, 1940.
• RADCLIFFE-BROWN, A. R. Structure and function in primitive society London: Cohen & West, 1952.
• RITTER, M. L. The conditions favoring age-set organization. Journal of Anthropological Research, v. 36, n. 1, p. 87-104, 1980.
• SCHUSKY, E. L. Manual for kinship analysis New York: Holt, Rinehart and Winston, 1965.
• SEEGER, A. Dualism: fuzzy thinking or fuzzy sets? In: MAYBURY-LEWIS, D. The attraction of opposites: thought and society in the dualistic mode. Ann Arbor: University of Michigan Press, 1989. p. 191-208.
• STEWART, F. H. Fundamentals of age-group systems New York: Academic, 1977.
• TURNER, T. S. Social structure and political organization among the Northern Cayapo Dissertation (Ph.D. in Anthropology)-Harvard University, Cambridge, 1965.
• TURNER, T. S. Dual opposition, hierarchy, and value: moiety structure and symbolic polarity in Central Brazil and elsewhere. In: GALEY, J. C. Différences valeurs, hiérarchie, textes offerts à Louis Dumont Paris: École des Hautes Études en Sciences Sociales, 1984. p. 335-370.
• WELCH, J. R. Age and social identity among the Xavante of Central Brazil Dissertation (Ph.D. in Anthropology)-Tulane University, New Orleans, 2009.

*

I thank the Xavante of Pimentel Barbosa/Etênhiritipá for participating in the research. I am especially grateful to members of my secular age set (

êt½pá) for their camaraderie and to members of my spiritual age-set moiety (

wasiré wai'a) for their insights. Valuable comments were provided at various stages of the writing process by William Balée, Ricardo Ventura Santos, Carlos E. A. Coimbra, Jr., and Nancy Flowers. This research was made possible by funds provided by the Fulbright Commission (Fulbright-Hays Doctoral Dissertation Research Abroad Fellowship, Award No. P022A040016) and the Tulane Anthropology Graduate Fund. Additional site visits were made during subsequent research projects funded by the Brazilian Research Council (Conselho Nacional de Desenvolvimento Científico e Tecnológico - CNPq, post-doc fellowship 500288/2009-7 and grant MCT-CNPq/MS-SCTIE-DECIT/CT 40.0944/2005-7).

1

Spelling of native terms follows the orthography currently in use by Xavante teachers at the Pimentel Barbosa Municipal School. It is an orthography in transition, originally developed by missionary linguists (Hall, McLeod; Mitchell, 1987; Lachnitt, 2003) based on local dialects and later modified through its application and transmission by literate Xavante individuals at Pimentel Barbosa/Etênhiritipá. The system, as it is currently applied, differs from versions in use in other Xavante communities (especially in a preference for using

s instead of

ts and

z instead of

dz) and even from previous representations in publications coordinated by members of the Pimentel Barbosa/Etênhiritipá community.

2

Fieldwork consisted of 12 months residency, from 2004 to 2005, and several visits each year thereafter. The study was approved by the Tulane University Institutional Review Board (IRB) and the National Commission on Research Ethics (Comissão Nacional de Ética em Pesquisa - CONEP). Research permits were issued by the Brazilian Ministry of Science and Technology (Ministério da Ciência e Tecnologia - MCT), directives no. 420 on August 25, 2004 and no. 604 on September 19, 2005. Permission to conduct research in the Pimentel Barbosa Indigenous Reserve was granted by the Brazilian National Indian Foundation (Fundação Nacional do Índio - FUNAI), authorizations no. 65/CGEP/04 and 68/ CGEP/05. In accordance with Xavante protocol, the aims and scope of the project were presented to community leaders and at the

warã, a meeting of mature adult males held daily at dawn and dusk in the center of the village. Although consent for the research was granted at the

warã by the mature men on behalf of the community, individuals were allowed to decline to participate in the project in whole or in part for any reason and without prejudice.

3

In May, 2004, when I began fieldwork, the subject community resided in a single village called Pimentel Barbosa or Etênhiritipá. In late 2006, that village divided for internal political reasons. Subsequently, my research included both villages. For simplicity, I refer to the pre-division village and both post-division villages as Pimentel Barbosa/Etênhiritipá.

4

The order for some other communities is

tirowa,

hötörã,

airere,

sada'ro,

anharowa,

nozö'u,

abare'u, and

êt½pá (Maybury-Lewis, 1967).

5

For the purposes of this study, I consider "structures" to be the relationship morphologies that link people and contribute to their social statuses in society. That formulation loosely follows Lévi-Strauss (1963) and is similar to what Radcliffe-Brown (1952) called "structural form". I apply the notion of social structure in such a way as to anticipate that it overlaps with and engages other dimensions such as social relations, patterns of social behavior, social roles, social values, social ideologies, and social institutions (Lévi-Strauss, 1963; Parsons, 1951; Radcliffe-Brown, 1940, 1952). Furthermore, I consider social structure to be inextricably linked to individual experience, and therefore to such anthropological concepts as practice, praxis, action, agency, and personhood (Bourdieu, 1972; Fuchs, 2001; Giddens, 1979; Harris, 1989; Ortner, 1984).

6

During my fieldwork at Pimentel Barbosa/Etênhiritipá, females did not participate in spiritual rituals. However, there is historical evidence from other Xavante communities that women were sometimes included as a form of disgrace in cases of spousal mistreatment (Giaccaria; Heide, 1984). My informants at Pimentel Barbosa/Etênhiritipá corroborate that account, but attributed female participation to the transgression of having engaged in indiscriminate sex. According to my sources, women have not participated in spiritual ceremonialism at Pimentel Barbosa/Etênhiritipá for many years. This account is supported by Nancy Flower's observations in the same community between 1976 and 1996 (Personal communication, November 7, 2007).

7

Although some definitions of age sets require that they have names, there is ample ethnographic evidence of age-group systems with unnamed age sets (Stewart, 1977).

Publication Dates

• Publication in this collection
  12 Nov 2010
• Date of issue
  Dec 2010

History

• Received
  26 Feb 2010
• Accepted
  03 June 2010