Embodied Speech in the Northwest Amazon.

Janet Chernela, a professor of anthropology at the University of Maryland, has a thought-provoking article “Language in an ontological register: Embodied speech in the Northwest Amazon of Colombia and Brazil” (Language & Communication 63 [Nov. 2018]: 23-32); here’s the Abstract:

Speakers of Eastern Tukanoan languages in Brazil and Colombia construe linguistic differences as indices of group identity, intrinsic to a complex ontology in which language is a consubstantial, metaphysical product—a ‘substance’ in the development of the person. Through speech, speakers of the same language signal a corporality based in theories of shared ancestry and mutual belonging while speakers of different languages signal difference. For Tukanoans, then, one creates one’s self in the act of speaking. These ontological beliefs underlie speech practices, influencing language maintenance and contributing to one of the most extreme examples of multilingualism reported in the literature.

It’s short (only ten pages); I’ll quote a few bits here and let you follow the link for the rest (I delete parenthetical references throughout):

The Vaupés River and its tributaries form the center of the Eastern Tukanoan Sprachbund, site of one of the greatest concentrations of linguistic diversity in the world. In this area of some 110,000 km² — a region larger than Denmark — about 40,000 residents speak languages from four indigenous language families: Eastern Tukanoan, Arawakanan, Tupi, and Nadahup. About a third of that number (20 or so language groups) are speakers of Eastern Tukanoan languages, including Desana, Piratapuia, Tuyuca, Barasana, Baré and Kotiria, whose numbers range from 10 (Yurutí) to 6151 (Tukano).

We are concerned here with speakers of Kotiria, a northern branch of Eastern Tukanoan, whose 1400 speakers are located along the banks of the Vaupés River from Santa Cruz in Colombia to Arara in Brazil. […] Kotiria villagers throughout this region speak indigenous languages in their everyday lives, with Portuguese and Spanish being the preferred languages with itinerant merchants, missionaries, and other outsiders. […]

The idealized, sociosemiotic arrangement of the ancestral founders is a template invoked to account for the establishment of situated descent groups along the river. Rather than using the often-misapplied terms from the anthropological literature to refer to these localized clusters of relatives (sib or patriclan), I use the Kotiria’s own concept, conveyed in the term koro (plural: koroa). Translated as ‘one-after-another,’ the term refers to a series of succeeding generations. Based upon recursive patrilineal principles, each koro is known by the name of its mythological founding ancestor plus the affix, -pona, meaning ‘children of.’

The members of a koro are understood to share a natural connectedness or, what Sahlins calls ‘a quality of inter-subjective belonging’ akin to Schneider’s notion of a shared ‘substance’ that imparts a transpersonal sense of being. One of those shared corporal substances is language. As an indicator of such ties, shared speech defines the community within which familial sentiments and moral commitments prevail. […]

For the Eastern Tukanoan Kotiria, for whom speaking and being are indivisible, the same phrase, Yʉ Kotiria hiha can be used to convey both ‘I am Kotiria’ and ‘I speak Kotiria.’ On one occasion I heard a Western researcher quizzing a speaker of Kotiria with the question, “How do you say ‘his father’s language?’” The respondent laughed politely before he explained what he regarded as plainly obvious: “His father’s language is his language.” The expectation is that a person who identifies as Kotiria should speak Kotiria at all times, unless comprehension is an issue.

By the same logic, when a Kotiria person is asked whether they speak their mother’s language, they are likely to respond by saying, “I am not X.” […] In yet another example illustrating the logic of the patrilect and the ontological foundation on which that logic is based, I report a conversation I overheard while visiting a village of Tukanoan Piratapuia on the Papurí River. There a villager admonished the child of a Piratapuia mother and an absent Colombian father, saying, “You shouldn’t speak Piratapuia, you should speak Colombian!” The speaker’s point was that the child should not be speaking his matrilect, Piratapuia; instead, he ought to speak his patrilect, ‘Colombian.’ The anecdote points to the conflation of language (speech), paternity, place, and identity. According to these equivalences, the critic’s expectation was a reasonable one, in spite of the fact that the child had never known his father, a Colombian trader. But it also illustrates that those who speak languages not their own — that is, those who speak in a language other than their patrilect — can become the targets of criticism. […]

In addition to patrilect and matrilect, which signal different sets of identities, and which are associated with different learning processes, the Kotiria recognize a third type of language learning, also defined in relation to speaker. In this form of language learning the speaker has no specified relationship with the social identities it indexes. According to the Kotiria with whom I spoke, acquiring a language that ‘belongs to’ others, described as khayo bu?ero, is understood to be a qualitatively different process than the natural process of learning one’s own language. Epps reports a similar phenomenon in her description of the Nadahup concept of ‘appropriating’ a language that is not one’s own. Together the two accounts suggest an area-wide phenomenon of some interest. According to the Kotiria speakers with whom I consulted, the process involves mimicry or copying (khayo), which closely glosses as ‘answering the same way he does’ (khayo, mimicry, or ‘answer the same way’). I have introduced the term alterlect to capture the sentiment of this category. […]

The Kotiria with whom I spoke place a high value on linguistic purism. According to them, people who readily shift to other languages are likened to mu, the yellow-backed mocking bird (Cassicus persicus), that is said to ‘speak in all the languages of the world.’ The Kotiria deride the Kubeo, a Central Tukanoan group that includes an Arawakan subgroup, for speaking other languages.‘They sound like the mu,’ they comment, ‘imitating every kind of sound.’ The commentary illustrates the high value placed on linguistic loyalty and standardized performance by some Tukanoan groups, including the Kotiria; at the same time it points to the fact that not all groups place the same high evaluation on linguistic loyalty. […]

A high value is placed on the quality poo, a term that may be glossed as ‘internal discipline’ or ‘self-control.’ A bearer of this quality, pooriro (one with inner discipline), does not mix or confuse languages. I have been told, “I do not mix languages because I am a disciplined (self-controlled) person”: yʉ?ʉduruku doho me?nera, yʉ?ʉã duruku pooriro hiha. […]

An ontology of language is not at odds with reality: it is the actuality in which speakers live and speak. Where speaking-and-being are inseparable, as they are among Eastern Tukanoans like the Kotiria, the ‘one language equals one people’ maxim is a foundational presupposition. It is a simple truth.

Fascinating stuff; thanks, Jack!


  1. J.W. Brewer says

    So they have independently reinvented 19th-century Eastern European nationalism, complete with vaguely Hegelian metaphysics to go with it?

  2. David Eddyshaw says

    This seems to be just a somewhat breathless account of the well-known Vaupes tradition of linguistic exogamy, reinterpreted into Western ethnic categories.
    (There may be a more nuanced view in the actual paper, which I can’t access.)

  3. Well, rats. At least we now know the Kotiria have discovered Hegel.

  4. David Eddyshaw says

    There is a nice grammar of Kotiria by Kristine Stenzel.

    The key to understanding Hegel may well be to read him in the original Kotiria.

  5. Trond Engen says

    Well, it was new to me that the stability of the arrangement is ensured by an unbreakable moral obligation to speak the patrilect, understood as something that basically upholds the world.

    Phrases like “quality of inter-subjective belonging” is just put in there to make me stop reading.

  6. David Eddyshaw says

    an unbreakable moral obligation to speak the patrilect

    This is apparently not the case: the rule is that you speak to someone in their language, which is, by definition, their father’s language. So you speak to your parents in two different languages (your parents must have different linguistic affiliation from each other, because otherwise that would be incest.)

    Alexandra Aikhenvald says this explicitly in her Tariana grammar, describing it as the rule throughout the Vaupes linguistic-exogamy zone. (Though she also says that the system is beginning to break down now.)

  7. Then we can add “father tongue” to the already existing “mother tongue”-“native speaker” scheme.

    I suspect that they learn mother tongues as well, and use them in certain situations. Do they?

    Then it is diglossia…

  8. “in their language”

    Does this mean that in a symmetrical situation both speakers address each other in each other’s langauge? But what is the problem then with a child who does not speak Colombian? (this italic motivates new vocabulary).

  9. David Eddyshaw says

    I suspect that they learn mother tongues as well, and use them in certain situations

    Yes indeed; bilingualism* is the minimum possible (unless you never speak to your own mother.)
    Although it is/was also common to know more local languages, you could in principle get away with being “merely” bilingual, because marriage partners are typically taken from your mother’s kin, and speak your mother’s language (though not to you, of course. That would be Very Rude …)

    * Not “diglossia” in the technical linguistic sense: the classic example of that is Arabic, of course.

    Does this mean that in a symmetrical situation both speakers address each other in each other’s language?

    Yes, exactly that. So you address your mother in her father’s language, and she addresses you in your father’s language, which must be a different language. Similarly with husband-wife conversations.

    On the other hand, siblings and their father all talk the same language among themselves.

    There is/was a strong taboo against language mixture in domains that people are consciously aware of, especially vocabulary. The languages have however converged greatly in syntax and semantics (helped along by the fact that most of them are related anyway; Tariana is an interesting exception, as an Arawakan language fallen among Tucanoans, which highlights some of the convergences particularly well.)

  10. There is a nice grammar of Kotiria by Kristine Stenzel.
    aka Wanano. She wrote once about how after traveling to Brazil to find speakers of Wanano (a language of about 1500 speakers), she found a fluent speaker, interested in working with her, living a half-hour drive away from the University of Colorado where she was studying.

  11. Not “diglossia” in the technical linguistic sense: the classic example of that is Arabic, of course.

    I meant that if in a village A you speak A in the street and B with your mom it is somewhat similar to diglossia, because there is some sort of distribution in function. But I thought that maybe kids can be addressed in mom’s langauge too… You described a somewhat different scheme, and yet there is something diglossic about the situation. Even if your choice of langauge is determined by who you are adressing, B-speakers themselves belong to a certain sphere in your life/your society.

    I think classic diglossia by Ferguson implies relatedness of langauges (partly fulfilled here) stability of the situation (Arabic has been like that for a long time) and diffrent function.

  12. John Emerson says

    I read that something about the New Guinea language plenum in which languages were not part of a group’s essence, but something that could be bought and sold.

    That’s all I remember, and perhaps it’s a little garbled.

  13. David Eddyshaw says

    There was something a bit along those lines in Don Kulick’s book on Tayap:


  14. I wonder, in the societies that practice linguistic exogamy in the Eastern Tukanoan Sprachbund, do different (non-human) spirits speak a certain language, or have a preference for a certain language?

    And are certain animals associated with certain languages? Do fish speak Bará, the language of the people whose autonym is waí mahá, literally “fish people”, I gather? (See here.) Do jaguars speak Barasana, the language of the people whose autonym is apparently Jebá-~baca “Jeba (the Jaguar’s) people”? (See page 2 here. I gather that in this instance, the symbol ~ indicates a nasal autosegment covering the whole root ~baca, which would be realised as [mãɲã], then?).

  15. Jen in Edinburgh says

    But what is the problem then with a child who does not speak Colombian?

    Maybe you’re required to use your ‘own’ (father’s) language to speak to yourself as you play?

    Is the impracticality intentional? In this case, the child does seem to have learnt the father’s language, although probably not from the father, but the possibility of a child being born from a one night stand with (or rape by) an outsider, and brought up in a community where no one speaks the father’s language at all, is there…

    It’s generally easier for a father to be absent, but even without that a child is presumably most likely to speak with (and learn to speak from) the mother, and yet it’s the mother and the child, not the father and the child, who are required to use two completely different languages to communicate with each other.

  16. One minute googling produced this quote from Aikhenvald (Language Contact along the Sepik River, Papua New Guinea):

    While there are minor dialectal differences between the three villages Avatip, Malu, and Yuanab, the last being especially divergent (see below), there are no linguistic differences between representatives of different clans and sub clans. Each subclan shares totemic terms and especially names; this ownership of totemic names is part of the patrimony of each subclan. Some can be shared by the whole clan group or by some of the subclans, and the knowledge of totems and names is tantamount to material riches (see Harrison 1990a; Penn and Lipset 1991). This feature is also characteristic of the Iatmul (Bateson 1958: 226-29), the Chambri, and a few other groups (I return to this in section 6.1).

    The Manambu, like other groups of River-dwellers along the Sepik, have an overwhelmingly “importing culture,” with an emphasis on exchange and value assigned to outside goods, both material and nonmaterial (Mead 1938). In many Sepik societies, linguistic items — words, totemic names, and personal names — were traditionally considered on a par with material goods, with spells, incantations, and names being traded and purchased (see Harrison 1990a:20-23, and further references therein).


    There had been long-term contact with the Iatmul involving trading spells and incantations in rituals, and also trading personal names. In Harrison’s words,

    from an historical perspective, the circulation of ritual forms in the regional trading system seems to have been a key formative influence on Manambu society [. . .], because the most valued scarce resources among the Manambu, and the items of strategic prestige value in the political system of their villages, were rights in ritual property, much of which the Manambu acquired from the Iatmul. Manambu ritual and cosmology seem, in fact, to be not only a kind of patchwork of the ritual and cosmological traditions of neighboring societies, but a largely bought patchwork, acquired piecemeal through trade. [1990a:20]

    Trading ownership of names and cults is a feature of numerous Sepik cul tures, including the Kwoma (Bowden 1983:67), the Abelam, and the Iatmul (Bateson 1958). This explains why names that appear to be Iatmul or Manambu are found pervasively among the Kwoma, the Chambri, and the Yessan-Mayo.[40]


    [40]. I do not discuss the origins of such personal names here, since the issue of name ownership in the area is politically very sensitive among the Manambu, the Kwoma, and the Iatmul (see especially Harrison 1990a; for discussion of a more recent name owner ship debate in Avatip, see Aikhenvald 2008a:14).

    Harrison 1990 is Stealing people’s names : history and politics in a Sepik River cosmology, it is both on Archive and in library Genesis.

    But it is mostly a featur of Indo-European langauges: all those schools and now web-sites where you pay for talking to a native speaker. I ever read somewhere about a man who wanted to export definite articles to China!
    P.S. Speaking seriously, though, it would be interesting to discover a langauge school in one of such societies.

  17. John Cowan says

    I think classic diglossia by Ferguson implies relatedness of langauges

    So it did. But within less than ten years Fishman showed that diglossia and (personal) bilingualism were orthogonal: Spanish/Guarani, Hebrew/Yiddish in the shtetl, and French/Darja represent both diglossia and bilingualism; French/Russian in the Empire was diglossia without bilingualism (in principle, peasants spoke only Russian, aristos only French); the immigrant experience from the 19C onward is bilingualism without diglossia, and of course neither bilingualism nor diglossia is experience by isolated groups who hardly talk to anyone outside their immediate community. Note that none of these cases necessarily involve languages that are closely related.

Speak Your Mind