From Povos Indígenas no Brasil
Photo: Isaac Amorim Filho , 1985


Self-denomination Where they are How many Linguistic family
1.700 (CTI, 2016)
2.500 (CTI, 2016)

The Matsés, also known as the Mayoruna, inhabit the frontier region between Brazil and Peru. Their communities are located along the Javari river basin in the far west of Brazilian Amazonia. In Brazil, they live in the Vale do Javari Indigenous Territory (IT) along with other peoples from the Pano and Katukina language families. Wars undertaken by the Matsés in the past century and the resulting incorporation of captives from other indigenous groups meant they became the largest of the northern Pano peoples.

Skilled hunters, the Matsés trek across large areas during hunting and fishing expeditions and use their knowledge of the forest paths not only to defend their territory but also to manage resources. By alternating their hunting, fishing and swidden sites, the Matsés avoid exhausting the soils and animal populations, despite maintaining relatively fixed communities on the river shores, and simultaneously ensure the occupation and surveillance of their lands. Over recent years, the Matsés have suffered the loss of many people from diseases that have spread unchecked because of the poor level of healthcare provided in the region.

Names and language

Índios matsés praticando a respiração de rape, aldeia 31, Amazonas.
Índios matsés praticando a respiração de rape, aldeia 31, Amazonas.

The word matses is polysemic. It can denote the people we know by this name in opposition to others (who are called maiu or matses utsi), any indigenous people in opposition to non-indigenous (called chotac), or ‘people’ in opposition to other beings. Matses may also mean the group of co-residents or kin closest to the speaker, as well as the set of maternal uncles and paternal aunts’ husbands (in this case preceded by the particle cun, indicating possession). In Peru, the word is used as an ethnonym. In Brazil, the Matsés are also known as the Mayoruna.

Mayoruna is a term of Quechua origin (mayu= river; runa= people), used from the 17th century onwards by colonizers and missionaries to refer to groups inhabiting the region formed by the lower Ucayali, the upper Amazon (Solimões) and the Javari (Erikson 1992). From the first documents produced by early missionaries and others colonizing the region until the start of the 20th century, not much can be ascertained about these groups. Lexical comparisons between lists of ‘Mayoruna’ and ‘Maxuruna’ words collected by Castelnau (1851), Martius (1867) and Spix & Martius (1831) and the present-day Matsés indicate the proximity between these languages and the possibility of their belonging to the same subset within the Pano family (Fleck 2003). However this linguistic proximity is not enough for us to attribute any cultural or historical continuity between the Mayoruna of the 19th century travellers and today's Matsés/Mayoruna.

The Matsés are the largest of the current groups belonging to a subset within the Northern Pano (Erikson 1992). This subset also includes the Matis, Kulina-Pano, Maya, Korubo, and possibly other groups who still avoid any permanent contact with non-Indians. As well speaking mutually intelligible languages, these groups share cultural features such as the use of the blowgun, specific hunting techniques, patrilineal moieties and so forth. However beyond the markers of a supposed shared identity, what justifies treating these peoples as a group is the affirmation made by the Matsés themselves that they are closer because they “speak the same language” or recognize these groups as examples “of how our ancestors lived.”

With the exception of the Matsés and Matis, who have been the subject of ethnological studies, we know very little about these peoples, either in terms of their languages or in relation to their cosmologies and social organization. Almost all the population of the Kulina-Pano live in just one village on the Pedro Lopes river, an affluent of the Curuçá. There are also some Kulina-Pano married to Matsés and living among them, mostly in Trinta-e-Um village (on the upper Javari) and Nova Esperança village (on the Pardo river). A small group of 25 Korubo (adults and children) was contacted in 1996 at the confluence of the Ituí and Itacoaí rivers. Funai's Isolated Indians Coordination Team runs a surveillance post, the Vale do Javari Ethno-Environmental Protection Front, close to the site where this group of Korubo lives. The group named Maya (or ‘the isolated Indians of the Quixito’) is currently living in ‘voluntary isolation.'

The Matsés population is significantly more numerous than all the other Northern Pano peoples closest to them linguistically and culturally – a fact explained by the warfare activity undertaken by the Matsés in the last century.

Linguistic situation

The Matsés living in Brazil are mostly monolingual with children raised in the villages being educated exclusively in the indigenous language. Only those people who have worked or studied in the surrounding Peruvian or Brazilian towns speak Portuguese and Spanish fluently.

Location and population

Índios matsés, aldeia do Lameirão, Terra Indígena Vale do Javari, Amazonas.
Índios matsés, aldeia do Lameirão, Terra Indígena Vale do Javari, Amazonas.

In order to determine the area in which the Matsés live today, as well as the communities and the swiddens used on a daily basis, we need to include the vast territory traversed during hunting and fishing expeditions; the old swiddens and settlements, which may be reoccupied, and those visited to harvest peachpalm; the houses maintained by some families at the more remote swidden sites, where they live for some weeks or months each year; and even houses in Brazilian and Peruvian towns such as Angamos, Palmeiras do Javari, Atalaia do Norte and Tabatinga.

The Matsés communities are located in the basin of the Javari, a river forming the Brazilian-Peruvian border along its entire length, located in the far west of Brazilian Amazonia. In Brazil, the Matsés inhabit the Vale do Javari Indigenous Territory in the southwest of Amazonas state. Approved in 2001, it covers 8,519,800 hectares. Along with the Matsés, this territory is occupied by the Matis, Kulina-Pano, Korubo and Marubo (all Pano family) and the Kanamari (Katukina family) as well as isolated peoples (see the map of the Vale do Javari IT).

In the Javari Valley, the Matsés are distributed into eight communities along the Javari river itself and the Lobo, Curuçá and Pardo rivers. There is also a community located outside the Indigenous Territory, close to the Palmeiras do Javari border platoon (where 90 people live). According to a census conducted by Funasa in 2007, the total Matsés population in Brazil was 1,143 people.

In Peru, according to Matlock (apud Fleck 2003), the population totalled 1,314 people in 1998, while Fleck (2003) reports 14 Matsés communities in Peru in 2003. Most of these were located in the indigenous reserve, which covers 452,732 ha. The territory was named ‘Comunidad Nativa Matsés’ in 1993 and comprises the triangle formed by the Gálvez where this flows into the Javari river and a southern border formed by a line linking the two rivers, passing through the headwaters of the Choba river located between them.

The families shift between the villages frequently, including crossing the frontier. Hence it is difficult to establish precise data for the Matsés populations in each country.

Earlier demographic data

The data on the Matsés population in Peru, earlier and more reliable, dates from 1976 (Romanoff 1984), when there were 599 people in total. Of these, 508 lived in the settlement close to the upper Choba river, next to the mission post established by SIL in 1969. Forty-four lived at a settlement by the mouth of the same river, while seven resided in a house in the village of Angamos, located at the mouth of the Gálvez on the Javari river.

In 1984, there were between 800 and 1,100 people: 55% lived on the shores of the Gálvez river and the upper and middle Javari, while the rest lived between the Gálvez and Choba rivers in communities situated on the upper Tapiche and Blanco rivers and at the villages of Requena and Jenaro Herrera on the Ucayali river (Calixto 1984:16).

In Brazil, the oldest published figures date from 1980, when there were 337 Matsés. Of these, 86 lived in 10 houses on the Lobo river; 127 at the Trinta-e-Um village (upper Javari) in 22 houses; 32 people in a maloca at the village on the Ituxi river; and 92 at the Lameirão village (lower Javari). Records for the same year also identify a group living on the upper Lobo but with no contact with Brazilians (Melatti 1981).

In 1985, the team from the Javari Campaign registered 483 Matsés in Brazil, divided between the settlements of Lameirão (113), Trinta-e-Um (173), Lobo (107), Santa Sofia (35) and Ituxi (42), with 13 people living outside these communities (Campanha Javari 1985:17).

In the Javari Valley identification report, Coutinho recorded, in 1995, 651 Matsés in Brazil: 308 at Trinta-e-Um; 286 at Lobo; 88 at Lameirão; 42 at São Raimundo, and 27 at Palmeiras do Javari (Coutinho Jr. 1998).

Formation of the Matsés people and contact history

War period

Like the Marubo, today’s Matsés population is the result of the merging of various peoples who had previously inhabited different malocas and did not always speak mutually intelligible languages. The formation of the ‘ethnic’ group defining itself as ‘Matsés’ mainly derived from the gradual incorporation of captives (principally women and children) from other groups in the region.

In the 20th century, up until the 1960s, the Matsés launched attacks on many different malocas and groups, very often speakers of Pano languages. In these attacks, the practice was to kill the men and capture women and children who were incorporated into the families of warriors as their wives and children.  

Escoamento de toras de madeira dos altos rios, rio Javari, Terra Indígena Vale do Javari, Amazonas.
Escoamento de toras de madeira dos altos rios, rio Javari, Terra Indígena Vale do Javari, Amazonas.

Polygamy was linked to the capture of women; as a result, the Matsés increasingly became formed of families composed of Matsés fathers and ‘foreign’ mothers. During his research at the upper Choba settlement (Peru) from 1974-76, Romanoff recorded the impressive statistic that the population contained "captives from at least 10 linguistic groups, including speakers of what seemed to be dialects of Matsés (mutually comprehensible but with some distinct words), speakers of other Amerindian languages, speakers of Spanish and Portuguese.” The anthropologist cites a total of 74 captives and claims that “of the ten groups, the Matsés had wiped out at least four” (Romanoff 1984:69).

Data obtained in Brazil in 1975 refer to two captive Brazilian women and one captive Peruvian woman among the Matsés of the Lobo river (Melatti & Montagner-Melatti 2005 [1975]). Information from 1980 also records five Kulina-Pano women at the Trinta-e-Um village, as well as a Brazilian woman and a Peruvian man and woman at the Lameirão village (Melatti 1981:65).

While the Matsés say that they are a ‘mixed’ people, they also proudly emphasize features that differentiate them from the other peoples they know, stressing what we could call a ‘Matsés way of life.’ The ‘cultural hegemony’ identified among them by researchers (Erikson 1992, 1994; Romanoff 1984) is indeed notable given the heightened exogamy and the warfare pattern sustained by the Matsés over the last century.

Notable but still typically Amazonian: the goods and cultural practices of these captives were absorbed by the Matsés and today form part of what distinguishes them from other local groups. For example, they attribute the substitution of the blowgun (Matis style) by the bow and arrow to a people associated with the contemporary Marubo. Similarly, their myths tell how other cultural goods were acquired from enemies or ‘foreign’ allies: agriculture from the curassow; names from the water people; forest remedies from a people who lived downriver.

Non-Indians began to be an important part of this warfare dynamic around the 1920s when they intensified their expeditions and attacks on the headwater regions of the Gálvez, Choba, Javari and Curuçá rivers. One of the emblematic events of this period was the acquisition of firearms by the Matsés, who recount how they were taught how to use them by a woman captured from the Peruvians.

Relations with non-Indians

At the peak of the rubber boom on the Javari river, roughly between the 1870s and the 1920s, the Matsés lost their access to the river, frequented in the dry seasons to collect yellow-spotted turtle and giant Amazonian turtle eggs from the beaches. There is no mention of malocas or groups that could be Matsés in the documents produced by explorers of the Javari during this period (Melatti 1981:69). At this time, the Matsés were avoiding conflicts with whites, withdrawing to interfluvial areas, and maintained a pattern of dispersal that enabled them to keep away from the rubber extraction fronts as the latter moved up the rivers (see Mobility of local groups during the war period).

An episode narrated by the Matsés living in Brazil describes the first time their ancestors saw a rubber tapper camp. This narrative has a mythic structure marked by a succession of equivocations: the Matsés try placing gun cartridges in their ears, thinking they were ear pendants; matches were used as lip decorations; they drank the latex thinking it was manioc drink; and they scared themselves looking in mirrors and ran away. From this ‘first contact’ they took only a chicken, an animal they had never seen before.

Direct conflicts began to appear in the accounts from the 1920s onwards. In 1926, a Peruvian man working on the Gálvez river was interviewed by Romanoff (Romanoff 1984) and claimed that the rubber bosses were unable to set up on the Choba river due to the attacks by the Indians.

The response to these attacks were punitive expeditions – the so-called 'correrias’ – in which Matsés women and children were captured. Undoubtedly this only helped intensify Matsés warfare activities. Today some older people tell how their grandfathers launched a revenge raid against whites in the town of Requena. In these attacks they captured women and acquired firearms and metal tools.

Meanwhile the wars between the Matsés and other peoples continued. In Brazil, testimonies collected by Melatti and Montagner-Melatti (2005 [1975]) refer to a conflict occurring in 1933 between the Marubo of the Maronal community on the Curuçá and the Matsés living on the Pardo river (cited in the report as ‘Mayo’ and ‘Mayoruna,’ as they were called by their non-indigenous informants). During this episode, one Matsés man died and the Marubo took their women. According to this source, this group of Matsés had lived in the region since the 1920s (Melatti 1981:69).

From the 1950s onwards, the exploration of rubber in the frontier region was gradually replaced or supplemented by logging activity and the trade in forest game and skins, mainly to supply the towns of Peruvian Amazonia.

The expansion of commercial logging in the region coincided with the creation of the Peruvian Angamos border platoon (1947) and the Brazilian border platoons of Estirão do Equador (1958) and Palmeiras do Javari (1965) along the shores of the Javari river. Documents from this period confirm the involvement of the Peruvian and Brazilian military in punitive raids against the Matsés, accompanied by civilians who had lost relatives to the Indians.

In Brazil there is, for example, a record of punitive raids in response to the Matsés attack on loggers from the Sacudido river area on August 31st 1958. On this occasion, “3 civilians and 59 military personnel found and defeated 3 indigenous malocas, one in the region between the Flecheira river [affluent of the Curuçá] and Santana river [affluent of the middle Javari] and two on the Negro river [affluent of the Curuçá]” (Coutinho Jr. 1993). In 1963, a punitive expedition was conducted by the Brazilian army against the ‘Mayoruna,’ organized by the commander of the Frontier Unit (GEF) based in Manaus.

Around 1960, outbreaks of warfare were again recorded between the Matsés and Marubo: the Matsés captured three Marubo women living in the Maronal community on the Curuçá river. They fled along the Amburus creek, an affluent of this river, and were later pursued by the Marubo who killed fourteen Matsés (Melatti & Montagner-Melatti 2005 [1975]).

Also in 1960 the Matsés living on the Curuçá expelled rubber tappers working on the river. In the raid, they killed Indians called Kulina (today identified as Kulina-Pano) who were living on the Pedro Lopes river and had established peaceful relations with the rubber tappers, who fled (Melatti 1981:69).

In Peru, Romanoff (1976) cites the 1964 punitive expedition that left the town of Requena for the Gálvez-Choba interfluvial region, organized by the mayor and the town's ecclesiastical authorities. They reached a Matsés maloca were they were surprised and attacked. The injured were rescued by helicopters from the US Navy. After the rescue, Peruvian airplanes bombed the location. Some men and women who today live in Brazil, childhood survivors from the bombed maloca, have impressive recollections of this episode.

These events reveal how it is impossible to draw a neat dividing line between a traditional way of life and a way of life defined by the arrival of the whites during this period of intense warfare. Dispersal and warfare were occurring long before direct contact with non-Indians. At the same time, colonization indirectly made itself felt through epidemics and the territorial readjustments enforced on the region’s different native peoples.

From the 1970s onwards, though, the cessation in armed conflicts and the relative sedentarization of most of the Matsés – enabled through the gradual establishment of pacific relations with some whites – can be interpreted as factors that brought about profound changes, experienced as such by the Matsés themselves.

Arrival of the mission (Peru)

The Matsés are unanimous in asserting that the process defined by themselves as ‘being tamed’ began in 1969 with the arrival of American missionaries from the Summer Institute of Linguistics (SIL). However, we should be careful not to be misled by this enemy-derived expression of ‘taming.’ Though the end of the armed conflicts coincided with the arrival of the missionaries, this was not because the Matsés adopted a Christian ‘pacific’ morality, quelling their potential conflicts with whites and other peoples. Without fully adopting the Christian ethos (evinced by their subsequent dispersal after an initial period of agglomeration around the mission), the Matsés succeeded in preventing attacks on their malocas by whites, which ceased after the arrival of SIL with the support of the Peruvian government. Given the size to which these attacks had grown (to the point, incredibly, of an aerial bombardment), the presence of missionaries in the area was undoubtedly important in terms of protecting the Indians. This fact enabled a gradual approximation of the group contacted by the missionaries (through the common practice of offering industrialized goods, left close to the Matsés houses).

Around 1963, before the decisive approximation, the two missionary women established the first contacts with speakers of the Matsés language: a Peruvian woman and her child who had escaped captivity, and a young man captured by another northern Pano group, called José (or Joe) by the women, who was found on the shores of the Javari, waving to a passing boat. The young man lived with the missionaries for a year and a half and subsequently tried to mediate with the rest of the Matsés group with whom he had formerly lived. He was killed by them, however. After this failed attempt to use an indigenous mediator, the missionaries sought direct contact. The women camped on the shores of the Javari river while airplanes transmitted messages to the malocas using loudspeakers. After initial friendly contact, Matsés families began to settle next to a landing strip located near to the upper course of the Choba. The strip was built by the Matsés under the direction of the missionaries (Romanoff 1984:54).

As is common during the first years of contact with non-Indians, a series of epidemics struck the population following the establishment of the missionary settlement. The medical care offered by SIL then became another lure for the more isolated groups who received news (and diseases) from visitors coming from the mission base. The Matsés say that some groups refused to make contact with the whites and still remain in voluntary isolation today.

During the period when anthropologist Romanoff stayed in the area, between 1974 and 1976, the SIL team (the pioneering missionaries and a couple that later joined the women) provided medical care and industrialized goods bought from the sale of craftwork produced by the Matsés. As well as the medical and commercial services, SIL’s main activities were: “linguistic research, Bible translation, evangelization (...), acting as intermediaries between the Matsés and outsiders, and teaching literacy.” The anthropologist also claims that the missionaries who worked among the Matsés had been contracted by the Peruvian Ministry of Education and therefore “also carried out administrative tasks” (1984:51).

A period of concentration in malocas near the airstrip, close to the Choba river, apparently gave way at the start of the 1980s to a period of dispersal as families began to leave the mission base and settle in houses on the Gálvez and upper Javari rivers. In these riverside communities, the extended patrilocal families (groups of brothers, their wives, sons and their wives, unmarried daughters and, sometimes, older dependent relatives) began to live in separate houses, each with a nuclear family, rather than in a single maloca (Calixto 1984:18).

Peaceful contact with Funai and the army (Brazil)

On the Brazilian side, Funai has been active in the Javari river region since 1971 following its implantation of a headquarters on the upper Amazon (the Upper Solimões Agency: Ajusol). The arrival of Funai coincided with the opening of the Perimetral Norte highway.

Around 1974 a post was installed on the Lobo river, which maintained contact with a group of Matsés. In 1975, a Funai attraction front was set up at the mouth of the Lobo, which maintained contact with three 'Mayoruna’ malocas (Melatti & Montagner-Melatti 2005 [1975]:10).

In 1978, the Palmeiras do Javari border platoon was involved in a conflict between various Matsés families living in the area covered by Funai’s local indigenous post. In 1977, a Matsés man who had left his community on the upper Lobo and lived in Atalaia do Norte, working as a manual labourer for Funai, tried to return and travel up the river, but was threatened by the group living in the malocas along its course. He assembled his relatives at the maloca on the upper Lobo and attacked the malocas on the middle course of the river. They were defeated, though. He then looked for reinforcements among the Peruvian Matsés and forced his enemies to flee to the Lobo's confluence with the upper Javari.

In January the following year, he returned to the Lobo to work but was attacked by residents from the settlement at the river's mouth. He retreated to his original village and later to the Peruvian side, once again bringing warriors back to attack the village at the mouth of the Lobo. Four men were confirmed as dying in this conflict, two on either side (Melatti 1981:71). Afraid of fresh attacks from Matsés living in Peru, the families left their settlement at the mouth of the Lobo and took refuge in Palmeiras do Javari. Although the attacks did not materialize (possibly because of the protection guaranteed by the Brazilian army), these families, unable to return to their homes, were transferred by the army and Funai to a location on the middle Javari called Lameirão (Melatti 1981:71).

During the 1970s, the logging and rubber extraction activities in the Brazilian region of the Javari basin (along with surveys conducted by Petrobrás) continued to threaten the integrity of the region’s indigenous populations, especially due to the danger of the transmission of non-indigenous diseases, in some cases fatal to the Indians.

In 1975, there were eleven rubber extraction camps on the Jaquirana (upper Javari), one on the middle Javari, also called Lameirão, and two on the Pardo river (Melatti 1981:80). In 1985, a document produced by the Javari Campaign recorded more than 170 loggers and more than 70 rubber tappers, as well as 11 river traders working regularly on the Jaquirana river and its affluents and on the Pardo and Grande rivers (the area occupied by the Lameirão community). The document also records that Matsés men were working in rubber and timber extraction (Campanha Javari 1986).

The document led to a national and international campaign in favour of demarcating an area of protection for the region's Indians. However, the physical demarcaion of the Vale do Javari Indigenous Territory was only completed in 2000 and officially approved on May 2nd 2001. Non-Indians living in the area were successfully removed, but the lack of adequate border inspection still allows loggers, hunters and fishermen to invade the area in certain regions.

Following demarcation, the Lameirão region ended up outside the boundaries of the Indigenous Territory. The families living there moved to communities on the shores of the Curuçá and Pardo rivers, where they still live today. Most of the group that remained on the Lobo river after the incident in 1978 continue to live on its shores and those of the upper Javari. The last large relocation occurred in 2006 when more than half the population of the Trinta-e-Um village, situated on the upper Javari close to the mouth of the Lobo, moved to the Pardo river due to a number of deaths caused by hepatitis B.

Mobility of local groups during the war period

During the period of intense conflicts, the Matsés maintained the pattern of territorial occupation and socio-spatial organization described by Romanoff (1984) as ‘fission-fusion.’ This pattern enabled a degree of success in the war period, given that they were not exterminated or enticed into debt-peonage by the whites as occurred with other peoples from the region during the 20th century.

According to statements from middle-aged men who heard the stories of these feats from their parents, the territory traversed by Matsés warriors at this time spanned from the area bordered in the west by the lower Ucayali and the eastern affluents of the Tapiche river, crossing the Brazil-Peru frontier (the Javari river) to as far as the Ituí river in the east.

The area of dispersal and concentration of families where they built their malocas was centered on the Gálvez river region (a left-shore affluent of the upper Javari in Peru) as far as the Curuçá (a right-shore affluent of the Javari in Brazil).

Each local group, living in a single maloca or a cluster of malocas, was composed by groups of brothers, whose sons, after a period of ‘bride-service,’ brought their wives from other malocas. In contrast to other Pano groups such as the Kaxinawá, among whom a maloca or village was ideally composed around two brothers-in-law whose children married with each other (Kensinger 1995), the Matsés patrilocal group depended on matrimonial alliances with others (or the capture of women).

Each maloca was constructed in the centre of the communal swidden. The local groups maintained several of these houses/swiddens, typically separated by one or half a day’s journey, which they used alternately throughout the year as they accompanied the seasonal production cycles. The main habitation was the house/swidden currently at its peak of production when the banana and manioc crops had ripened (around one year and seven months after planting, respectively). The houses/swiddens were occupied during the maize planting period and later during harvesting (Romanoff 1984:182). While the maize grew, the group would leave the new plantation and return to the main maloca. The men from the jaguar moiety (see Tsasibo x macubo dualism) were banned from seeing the maize before it ripened, since simply looking at the crop could jeopardize its growth.

Hence a single local group possessed various habitable settlements over the course of the year; this allowed not only seasonal dispersal but also readjustments within the group itself. The families making up a settlement did not always migrate together and sometimes during these movements they would regroup with other previously distant families. This was the basis of the ‘fission-fusion’ pattern. The composition and recomposition of the local groups favoured dispersal, which protected the women, children and older people from retaliatory raids by their enemies (Romanoff 1984:147).

The older swiddens or those with a low level of production were also visited, principally to harvest palm fruits (such as peachpalm and ungurahua palm) and their timber, used in the fabrication of spears and houses.

Formation of the villages or communities

The Matsés residence pattern began to change after the missionaries settled on the upper Choba in 1969. What essentially altered the situation was the introduction of a fixed point of attraction formed by the mission and the airstrip. This new element in the landscape was a source of new items desired by the Matsés, but required them to remain at the same location in order to attend the missionary meetings and, only then, receive the goods and medications they began to need and obtain protection from attacks by the regional non-indigenous population.

The cycles of fission and fusion described earlier (see Mobility of local groups) gave way, then, to a relatively stable settlement pattern of malocas built around the missionary base. However, the Matsés did not abandon their pattern of dispersal. While maintaining more permanent malocas on the upper Choba, the families continued to spend parts of the year in alternative malocas before returning later to the missionary settlement. Romanoff also mentions a ’centre/periphery’ residence pattern (1984:180).

This pattern, which alternates between residence in more permanent and populous communities or villages and in dispersed family swiddens, is still active today. In Brazil and Peru, the Matsés families live in villages or communities on the shores of rivers and creeks, allowing them to combine relations and trade with whites and the use of dispersed swiddens where they spend much of the year tending their crops and organizing hunting and fishing trips – which produce abundant game and fish in these isolated areas.

For the Matsés living in the Indigenous Territory on the Brazilian side, another factor influencing the choice of location for the dispersed houses/swiddens is the desire to defend their territory. The group of brothers considered the ‘chiefs’ of the Lobo village, for example, maintain swiddens in important points along the frontier and organize their own inspection patrols. They report that they have already had to expel invaders and that one of their support houses was deliberately burnt down (Matos 2006).


Matsés malocas are hexagonal in shape with a rectangular body formed by two longer opposite sides. The straw roof covers the entire structure aside from two openings about 1.25 metres in height. These 'doorways' are located at the front and rear of the maloca at opposite ends of a central corridor that divides the house into two parts. Each half is divided, in turn, into small compartments separated by straw screens that serve as 'walls.'

The compartments are called quënë and shelter a man, his wife (or two wives) and children. If the man has more wives, the other (generally older) wife may be located in the neighbouring quënë along with her children. Next are located the man’s brother and his own wives and children, and so on. Each quënë contains a fire where the women prepare most of the daily meals for their families, also warming the house at night. The main deemed to be the 'owner' (icbo) of the maloca (the one who exhorted his relatives to work on building the house) generally sleeps with his wife and children in compartments closer to the entry.  

Foto: Beatriz de Almeida Matos, julho de 2005
Foto: Beatriz de Almeida Matos, julho de 2005

In the main opening are placed the parallel benches made from whole tree-trunks. The space where the benches rest is called the nantan, where visiting men sit very formally and are immediately served whatever food is available by the women living in the maloca. Women, even visitors, never sit on these benches and instead eat in small circles on the floor along with the children.

The largest malocas observed by Romanoff in 1976, in Peru, were up to 35 metres in length and 10 metres in height, sheltering 100 people. In these buildings the anthropologist noted another two doors located in the middle of the side walls.

Today in Brazil, only the Lobo community located on the shores of the river of the same name has an inhabited maloca. This is occupied by a group of brothers identified as the village’s ‘chiefs,’ along with their wives, children and their widowed mother. Almost every night or late afternoon the men in this village meet in the nantan to talk and inhale snuff after the meal. They plan hunting trips and collective work, and chat about the day’s events.  

Maloca dos matsés, igarapé Lobo, Terra Indígena Vale do Javari, Amazonas.
Maloca dos matsés, igarapé Lobo, Terra Indígena Vale do Javari, Amazonas.

Until 2006 there was a large maloca at the Trinta-e-Um community. This was pulled down due to its age and the fact that its ‘owners’ were getting ready to move to Nova Esperança village. This maloca was not inhabited but served as a ‘men's house,’ especially for those of middle age who met their at the end of the day for collective meals followed by snuff sessions. This was also the place where chiefs welcomed visiting whites for meetings (representatives of NGOs, Funasa, Funai etc.).

The maloca possessed just the building infrastructure and two parallel benches running its entire length (made from tree trunks placed end-to-end).  

Foto: Beatriz de Almeida Matos, abril de 2008
Foto: Beatriz de Almeida Matos, abril de 2008

The vary majority of contemporary Matsés houses in Brazil have been built in the regional non-indigenous style. Each house shelters a conjugal family: a man, his wife and unmarried children. The preference for patrilocal residence has been maintained, meaning that when they marry almost all men build their own stilt house from paxiúba (stilt) palm near or next to his parents’ house and very often looks for his future wives in distant villages, including in Peru. It is also usual – forming part of the etiquette – for the groom, especially when young and marrying for the first time, to stay for a while in the house of his future wife’s parents before bringing her to their new home.

Hunting and fishing activities

Foto: Beatriz de Almeida Matos, julho de 2005
Foto: Beatriz de Almeida Matos, julho de 2005

Hunting activities are highly valued by the Matsés who consider themselves to be above all hunters of the interfluvial forest areas, despite living in villages on the shores of rivers. The favourite day-to-day prey are the spider monkeys and white-lipped peccaries (increasingly difficult to find), but the Matsés also hunt tapirs, woolly monkeys, collared peccaries, pacas, armadillos, deer, tortoises, alligators and sloths (the young of the latter are highly valued as pets). Additionally they hunt curassows, tinamous and various other forest birds. In the dry season they kill river turtles and gather their eggs from the beaches to eat. Collecting turtles eggs is an important activity, so much so that the Matsés call the period of one year seta, the same term used to designate the river turtle.

Hunters use bow and arrows, guns and hunting dogs. Very often, though, the animals are captured or killed before expending their arrows or ammunition thanks to a series of specialized traps and hunting techniques for each type of animal. For instance, the tapir can be killed with a trap, while armadillos are cornered in their burrows  and drowned with water. Pacas, with the help of dogs, are made to jump at the curve of a stream where they are surrounded; when they emerge, they are struck with sticks. Special cords are used to catch sloths by the neck and claws, removing them from the tree canopies to take them back to the village alive.

Foto: Beatriz de Almeida Matos, maio de 2006
Foto: Beatriz de Almeida Matos, maio de 2006

But the most notable feature of Matsés hunting practices is the active involvement of women in most of the hunts. Just as men very often accompany their wives to the swidden to fetch manioc, banana and other produce, women go hunting with their husbands. They help locate and corner the game, take part in the chase, recover arrows that miss their target, and attack the animals with sharpened sticks or axes. Children frequently accompany the couple and also take part at some points. From a young age, children are encouraged to look for animal young, both to eat and to serve as pets.

However, the presence of women in some hunts does not imply that the Matsés lack the idea commonly found in Amazonia that there is a certain incompatibility between women and hunting activity. A hunter’s lack of hunting success or skill, translated by the regional term panema, can be caused by excessive sexual relations. Some animals cannot tolerate the smell or presence of women; for example, when they set a trap for tapirs, men must abstain from sex with their wives. This may explain why older men ideally prepare these traps (Romanoff 1984:172).

To improve their skill and success as a hunter, combating panema, dejection or laziness to go hunting, the Matsés use substances that make the body strong, harder and cleaner. All these substances are linked to the principle of bitterness (muca), which also regulates shamanic power (Erikson 1994).

Venom from the kampo toad, snuff made from tobacco leaves (nënë), bullet ant stings and the application of plant sap, which allow the hunters to sight their prey better, are all routinely used. Tobacco, in particular, is consumed daily by Matsés men.

Another set of rituals associated with hunting are the medicinal baths given by the Matsés to small infants to prevent the animals killed or eaten by their parents from harming their health. These plants, called neste, are mainly collected by older men and women who have the ability to identify them. Each leaf is associated with a particular type of animal. Only after their small children are bathed with an infusion from the correct plant can the parents kill or eat animals with strong meat, such as the tapir or spider monkey. After killing large snakes and jaguars, parents should also protect their children against the harm potentially caused by these spirits.

Matsés men, women and children fish in the rivers close to the villages on days when there is no game. The families also undertake collective fishing trips with poison (antinte) which they dissolve in the water of lakes and streams, especially in the dry season. Hunting and fishing expeditions that involve various days camped in the forest are also common. This movement is called capuec, trekking. The time spent in the forest is greatly enjoyed.

Tsasibo x Macubo dualism

Pintura corporal matsés, aldeia Lobo, rio Jaquirana, Terra Indígena Vale do Javari, Amazonas.
Pintura corporal matsés, aldeia Lobo, rio Jaquirana, Terra Indígena Vale do Javari, Amazonas.

The Matsés are divided between those who are tsasibo and those who are macubo. The tsasibo (‘tough’ + collectizing suffix) are also called bëdibo (jaguar + coll.) and shëctembo (peccary + coll.). The macubo (larva + coll.) are also called shëctenamëbo or aiabo (white-lipped peccary + coll.). Animals that wander in bands are typically macubo, while solitary and more ferocious animals (with dark, tough meat) are tsasibo.

The descriptions given by the Matsés of what it means to be macubo or tsasibo suggest that the moieties are much more ritual than political in function. Rather than denoting belonging to a class or subgroup that combines with others to form a coherent whole, the terms refer to ways of being and relating to other humans, spirits and animals.

Hence, for example, macubo men have privilieged relations with the larvae that feed off maize and that can destroy a crop if not removed. Only the macubo can undetake the task of removing the larvae since the latter are ‘their kin.’ If the tsasibo try to clear the plantation, the larvae multiply instead, destroying the maize crop.

The tsasibo, for their part, are not attacked by jaguars when they go hunting since the red annatto spots painted on their arrows, similar to those jaguars wear on their skin, warn the animals that those hunting are their 'kin’ and therefore persuade them not to attack the hunters. While the body painting most used by the tsasibo are the spots, the macubo use a pattern similar to the footprints of peccaries, stylized as triangles placed tip-to-tip, or parallel lines that refer to the macu (larvae).

From its formation in the mother’s belly, the child – whether female or male – will bear body markings inherited from the father. If the woman had sexual relations with both tsasibo and macubo men, the child may have markings from both moieties. The preferential marriage for a boy is with the daughter of his cucu, or mother’s brother, so that, without being an explicit rule, marriages are almost always between men and women from different moieties.

Mourning songs also differ according to whether the deceased was macubo or tsasibo. These songs are chanted by close kin, beginning late afternoon and lasting sometimes throughout the night. The songs are very beautiful and extremely sad, performed by one or two people who evoke the activities that they, the singers, and the dead person did together. A special vocabulary is used in these songs to refer to the deceased and his or her kin, to the hunted game and to the crops planted in the swiddens by the person.

Matsés dualism therefore echoes the dualism found among other Pano peoples, such as the Kaxinawá (Kensinger 1995). However, it does not function as a device capable of creating social groups, nor does it directly influence matrimonial choices (there is no moiety exogamy).


Like all the other peoples that live in the Vale do Javari Indigenous Territory, the Matsés fail to receive adequate healthcare from the official organs. Funasa’s Javari Special Indigenous Health District has been responsible for healthcare in the IT since 1999. Over this decade, there has been no reduction in diseases such as malaria, worms, tuberculosis, malnutrition and hepatitis: on the contrary, they seem to be out of control.

The lack of reliable statistics cannot hide the fact that the Matsés are suffering from diseases that could be controlled through adequate vaccination, medication and prevention. Malaria epidemics, for example, reappear each year with almost all the villages showing extremely high levels of infection. Meanwhile, the lack of drugs for treatment and microscopes for diagnosis is a basic fact of life in most villages.

However, the biggest concern for the Matsés today is the high level of hepatitis B and D infection, though again no reliable information exists. The Matsés do not know how many of them are infected, but the constant loss of young people, most of them under 30 years old, generates a pervasive mood of sadness and fear. Over 2005 and 2006, for example, more than 15 Matsés youths died from hepatic complications. In 2007, eight people died in the Trinta-e-Um village alone: the Matsés cannot be entirely sure that hepatitis was responsible, but report that the patients all suffered from liver problems. Sometimes the deaths occurred in tragic fashion since hepatitis D can kill in the space of a few days and in the final moments the patient suffers from unbearable pains and vomits blood. The number of deaths and reports of infection has led the Matsés and other peoples of the Javari Valley to question even the effectiveness of vaccines. Another fact causing widespread distrust among the Matsés in relation to the organ responsible for healthcare is that the kin of deceased patients never receive clear information on what killed them.

Despite deaths from hepatitis B being recorded among the Matsés since at least the end of the 1970s (Melatti 1981; Campanha Javari 1986) to the present, nothing effective has been done by the government organs to prevent its spread (Matos & Marubo 2006).

The CTI team (Centre of Indigenist Work) has produced press releases, texts and reports containing recent information on the seriousness of the health situation and the hepatitis infections among the peoples living in the Javari Valley, as well as analyses of the consequences ( There also exists a report produced by the anthropologist Walter Coutinho Jr. (2008) listing the available data on infection rates in the Vale do Javari IT.

School education

Índios matsés lendo o livro Povos Indígenas no Brasil, aldeia Lameirão, Terra Indígena Vale do Javari, Amazonas.
Índios matsés lendo o livro Povos Indígenas no Brasil, aldeia Lameirão, Terra Indígena Vale do Javari, Amazonas.

In Brazil, all the Matsés villages possess a teacher from the community itself, though none of them has yet completed their training. The State Education Secretary for Amazonas has been running indigenous teacher training courses among the peoples of the Javari Valley, but, due to problems in organization, the courses only take place sporadically. Despite complaints from the Matsés communities, only the Flores and Três José villages have had schools built by the Atalaia do Norte municipal council. One of the primary consequences of the low level of school teaching in the village is the exodus of students to nearby towns such as Atalaia do Norte, Benjamin Constant and Tabatinga. Many Matsés send their children while still young to the town hoping that by studying they will be able to obtain paid jobs.

With the aim of providing better working conditions for indigenous teachers, the Centre of Indigenist Work (CTI), a Brazilian NGO, has offered additional teacher training courses since 2003, along with village school monitoring programs and the production of teaching material. In 2008, the first school text book produced by Matsés teachers in partnership with the CTI was published.

Indigenous organization

The Matsés in Brazil do not have their own indigenous association or organization, but they have been part of Civaja (the Javari Valley Indigenous Council) since its inception during the campaign for demarcation of the IT. Recently the organization was renamed and is today called Univaja. Civaja, founded in 1991, was always directed by the Marubo of the upper Curuçá. Some Matsés leaders took part in all the Council’s assemblies, but they frequently complained that they have little say in the indigenous organization’s decisions and activities.

Sources of information

  • Calixto Méndez, L. Las practicas socio-culturales del uso de la tierra en el grupo etnico Matsés (manuscrito), 1985.
  • Cavuscens, S. & Neves, L. J. Povos Indígenas do Vale do Javari. Campanha Javari. CIMI e OPAN, Manaus, 1986.
  • Melatti, Júlio Cezar. Povos Indígenas no Brasil, volume 5 (Javari). CEDI, São Paulo, 1981.
  • Coutinho Jr., W. Brancos e Barbudos. Universidade de Brasília, 1993. (dissertação de mestrado)
  • Coutinho Jr., W. Relatório de Identificação e Delimitação da Terra Indígena Vale do Javari. FUNAI, Brasília, 1998.
  • Coutinho Jr., W. Hepatopatias No Vale Do Javari: Virulento Agravo À Saúde Indígena e Afronta aos Direitos Humanos. Ministério Público Federal, Procuradoria da República no Amazonas, Manaus, 2008.
  • CTI (Centro de Trabalho Indigenista). Diagnóstico sobre a Educação Escolar Indígena na T.I. Vale do Javari. 2008
  • Erikson, P. “Uma singular pluralidade: a etno-história Pano”. In: Historia dos Índios no Brasil. Manuela Carneiro da Cunha (org.), 1992, pp. 239-251.
  • Erikson, P. “Los Mayoruna”. In: Guía Etnográfica de la Alta Amazonía, Volumen II . Fernando Santos & Frederica Barclay (orgs.). Quito: FLACSO, 1994, pp. 1-127.
  • Fleck, D. A Grammar of Matsés. PhD Thesis. Rice University, 2003.
  • Kensinger, K. How real people ought to live. Illinois, Waveland Press, 1995.
  • Matos, Beatriz. “Os Mayoruna e a vigilância da fronteira.” In: Beto Ricardo & Fany Ricardo. (Orgs.). Povos Indígenas no Brasil, 2000-2005. São Paulo: Instituto Socioambiental, 2006, pp. 443-443.
  • Matos, Beatriz & Marubo, Jorge. “Colapso no Atendimento Provoca Retorno de Epidemias” In: Beto Ricardo & Fany Ricardo. (Orgs.). Povos Indígenas no Brasil, 2000-2005. São Paulo: Instituto Socioambiental, 2006, pp. 444 - 445.;_Javari-Sa%C3%BAde-Colapso%20no%20atendimento%20provoca%20retorno%20de%20epidemias.pdf
  • Melatti & Montagner-Melatti. “Relatório sobre os índios Marubo”. In: Série Antropologia 13. Universidade de Brasília. 2005 [1975].
  • Romanoff, S. “Informe sobre el uso de la tierra por los Matsés en la selva baja peruana”. In: Amazonia Peruana 1(1): 97-130. CAAP, Lima, 1976.
  • Romanoff, S. Matsés Adaptations in the Peruvian Amazon, Ph.D. dissertation, Columbia University, New York, 1984.