31. Language Loss and Language Ideology in the Sepik Region of Papua New Guinea

This post is also available in: Catalan

William A. Foley
University of Sydney


A men’s cult House with marks of the totems from a village of the Sepik River basin


The Sepik region of Papua New Guinea is unquestionably the most complex and diversified area linguistically in the world. In an area of a hundred thousand square kilometres or roughly the size of the US state of Kentucky (ranked 37th by size in the United States) are spoken over a hundred languages belonging to sixteen different language families, and all of these distinct on the order of Indo-European or Austronesian. There is a number of causes of this stupendous diversity, which I cannot go into here (see Foley 2017), but what is relevant for our purposes is that this remarkable heterogeneity is now greatly in jeopardy, as nearly all of the languages spoken in the region are gravely threatened and many not being passed onto children. Transmission for many ceased as far back as the 1980s and for some, like Kopar spoken at the mouth of the Sepik River, even earlier, and this process has been accelerating, so that languages that seemed vital in the 1990s are now exhibiting tell tale signs of language shift. And is what perhaps even more remarkable is that these languages are not being threatened and supplanted by a major metropolitan or national language like English, Chinese, Indonesian, French, Spanish or Portuguese, which is what is happening elsewhere in the world, but by what was originally a trade language, a pidgin which has creolised, Tok Pisin, or New Guinea Pidgin English. What was a showcase of linguistic efflorescence in the world is rapidly become one of monolingualism and linguistic scarcity, and surprisingly, as we shall see, some of the same social forces that contributed to the rise of the former are at work now in the development of the latter.

To understand this we need to understand the sociocultural patterns of the peoples of the Sepik region and the beliefs about language, that is, linguistic ideology, that arise from those. Consider how Newton (1975:197) profiles the social structure of communities there: ‘the result of groups of people, under pressures of fighting, famine, or for other reasons, having moved about freely and regrouped as they could. Consequently no village is a monolithic unit. It is, rather, a more or less stable congeries of clans having closer ties with each other than with those of other places.’ Each Sepik village is the result of numerous immigrations, and its clan structure, a recapitulation, albeit, as we shall see, a typically mythologized one, of the migration history. Although each individual village community is a unique amalgam of immigrant groups, due to its individual migration history, the overall effect of this pervasive pattern of population migrations is a largely homogenized Sepik culture, as the cultural practices common to fissioning communities are dispersed to new sites through emigration. Forge (1972) estimated 200 to be the optimum population of a Sepik village before internal conflict led to fission and emigration of some clans to new villages, and the oral histories of Sepik communities amply attest to this as the major demographic process in the region.

Within the Sepik, the history of migration has been rendered cosmological (see Keesing (1981) on ‘celestialization’), and, as such, provides a social charter. The world is understood as being created by the activities of the ancestors as they migrated over land and rivers to individual village sites. These ancestors are totemic, creative spirit beings, and they are the focus of ritual and, ultimately, mundane interest, with any concern with actual distant historical figures tenuous at best. Because of the constant fission and reforming of villages along clan lines, the same basic cosmological beliefs and migratory schemas are shared across Sepik villages, regardless of the language spoken. Villages speaking Iatmul or Manambu of the Sepik language family and those speaking Yimas or Karawari of the Lower Sepik-Ramu family largely share the same cosmology and social structure regardless of the deep linguistic differences, a pattern repeated throughout the Sepik region. Within the region, there is no sense of intervillage solidarity based on shared linguistic allegiance; villages speaking the same language may have closer cultural and social links to neighboring villages speaking a different language than to each other. It is the shared cosmology and linked social structure that bind the region together in spite of the very high linguistic diversity.

This mythology has powerful implications for local conceptualizations of personhood The person sees himself as a member of a clan and bears names to which he is entitled as a member of a clan. The clans are not village-bound, but through the complex of myths, rituals and the activities of totemic ancestors, are shared with other villages, often speaking distinct languages and in the very nature of the cosmos itself. For example, the names a man bears are those of the totemic ancestors, and, indeed, through them, he is of the same nature as these totemic ancestors; the same name and totems for example are shared by villages speaking Chambri, a Lower Sepik-Ramu language and adjoining Iatmul of the Sepik family. A man’s place in the world and the kinds of relationships he enters into with others is largely ordained by his place in this wider cosmologically enjoined system. The kinds of exchange relationships, secular and ritual, he and his clan have with others has been set by the activities of the mythological ancestors, and, to a large extent, the advantages and disadvantages of these relationships and, hence, the relative rankings of his clan in ritual status, is predetermined. Because the clans are defined relationally in terms of exchange and trade potential, both in ongoing exchange activities and by the activities of their totemic ancestors as they passed through the lands where these exchange an trade partners now reside, exchange relationships, secular and ritual, current and primordial, are implicated integrally in the definition of the person. A person is partible, constructed by the elements of exchange built up over a lifetime, and these in turn are ordained by the person he was born into in terms of the wider totemic system.

This conception of personhood has direct implications for beliefs about language and its usage, the local linguistic ideology. Sepik peoples conceive of language as action, as a way to get things done in ongoing exchange, both economic, commodities, and verbal. Language is not for making our internal thoughts known, expressing ourselves, but rather for maintaining ongoing sociability to achieve goals; indeed thoughts in these languages are commonly expressed as internalized speech, not the other way around, as in the Conduit Metaphor (Reddy 1993), which guides much Western theorizing about language, both lay and professional. These cultures commonly make a distinction between what can be loosely glossed as ‘understanding/care/heart’ (Yimas wampuŋ) and ‘will/image/spirit’ (Yimas aŋkaŋkaɲa) (Harrison 1990; Kulick 1992; Telban 1998). The Yimas word wampuŋ also means literally the bodily organ heart, the phonological form of the word an obvious onomatopoeic icon of the sound of its beating, so I will henceforth gloss it as ‘Heart’. It is also clearly linked to the idea of ‘insideness’, the word ‘inside’ being a derivative of it: wampuŋ-n < wampuŋ ‘Heart’ + -n ‘at’, and through this, to the pith of a tree and so softness: wampunŋ ‘sago flour’ (the soft, washed pith of the sago palm, the society’s staple food). The wampuŋ ‘Heart’ is the seat of desire and affect, and from the culture’s moral perspective this should be socially directed. One should be mindful of one’s social embeddedness and show concern for others, hence my alternative translation of ‘care’. Properly with wampuŋ one should ‘hear’ ant- the call of others’ needs. To heed the calls of others results in generosity, a very highly valued trait in Yimas culture; stinginess is, not surprisingly, deplored, and such behavior is described as kalck- ‘hard’ (like the hard outside bark of a tree), the opposite of the softness and ‘insideness’ associated with wampuŋ ‘Heart’.

The counterpart of wampuŋ ‘Heart’ in Yimas, as in other Sepik languages is ‘Will, spirit’ aŋkaŋkaɲa. Aŋkaŋkaɲa ‘Will’ is the powerful life force of a person, expressing itself in assertive claims to one’s prerogatives and staking claims to those of others. ‘Will’ can be increased in power by ritual, especially those involving the calling of secret totemic names and associated myths. In contrast to wampuŋ ‘Heart’, aŋkaŋkaɲa ‘Will’ is associated with the assertive, often aggressive individuality of the person; it is the image one sees reflected in a still pool, or nowadays, a mirror. A man in whom ‘Will’ is strong is described as kalck- ‘hard’, but in the context of aŋkaŋkaɲa ‘Will’ this is a term of admiration rather than the opprobrium attached to it when used in commenting upon wampuŋ ‘Heart’. Whereas hearing the calls of others through prior social responsibility is proper to wampuŋ ‘Heart’, aŋkaŋkaɲa ‘Will’ finds its articulation through assertive speaking. Wampuŋ ‘Heart’ is linked to the ear, but aŋkaŋkaɲa ‘Will’ is tied to the mouth. Speaking is metaphorically linked to hitting and striking in Sepik thought and hence copulation. Speaking can be penetration by the ‘Will’, and this is often resisted, leading to disputes.   Indeed, the Yimas word karm ‘spoken language’ is most commonly taken to mean ‘quarrel’: for example karm papk language there.is typically means ‘there’s a quarrel (going on)’.

Sepik peoples therefore harbor a deep ambivalence about language and its uses. Language can be used to help and support through hearing its call and attending to social concerns with wampuŋ ‘Heart’, but it can also be used to harm and exploit by the aggressive assertion of the individual’s wants with aŋkaŋkaɲa ‘Will’. Furthermore, language allegiance is and has been inherently unstable, as quarrels lead to fights and fissions in villages, leading to migrations to new areas and the taking up of other languages as fresh trading relationships are formed and marriages leading to the adoption of new totemic ancestors are entered into. This ongoing process of social fission has contributed much to the great linguistic diversity of the region.   In effect, a village language is very much a negotiable currency; when, because of altered social and trading relationships, it becomes devalued, people will switch to a new one more highly prized. This has happened over and over again in Sepik villages.   A language is esteemed for its value in trade and exchange relationships and little else; in other words, it is its wider sociocultural usefulness that matters. There is no sense of allegiance to a village language for identity, because identity does not come from the village language; rather it comes from totemic myths and exchange links, which are commonly crosslinguistic.  Sepik villagers do not trust language to reveal truth; yes, it can be used to solicit wampuŋ ‘Heart’, but is just as likely to enforce aŋkaŋkaɲa ‘Will’. Sepik villagers are consummate pragmatists in their understanding of language: language is conceptualized in terms of the effects it has, and these are taken to be largely negative; such a distrust of language has been reported as a widespread feature of Papua New Guinean communities (Robbins 2001). The constant shifting of language due to these factors, the ambivalence over its use, but at the same time its essential value in trade and exchange, has resulted over the centuries in the remarkable linguistic diversity mentioned earlier.

But unfortunately this is now a large part of the cause of its demise, for Tok Pisin is the exchange and trading relationship language par excellence. By its use trade relationships are available with others on a scale unimaginable before.   Globalization in its own local way has come to Papua New Guinea and the Sepik region, and the language of the region’s absorption into more globalized networks and integration is Tok Pisin. Given the local linguistic ideology it is obvious why language shift is happening there so rapidly and on such a massive scale. Whatever we as linguists may feel about this process, it is clear that Sepik villagers have a different view and that needs to be understood and fully respected.   The linguistic world as we knew it in the last century was still an effect of cultural, social and geographical isolation.   That isolation is now ending and is not going to return, regardless of the anti-globalist nationalist demagogues now on the contemporary political scene. The unstoppable forces of globalization are bringing all of us into closer and closer integration and interchange, and this inevitably will affect and supplant many if not most of the languages that developed in the long periods of isolation. New Guinea was the last area of the world to be colonized, and most of the island’s peoples developed in isolation from the rest of the world for millennia, the Sepik region particularly noteworthy in this respect. But that has now ended and its people are now becoming increasingly integrated into the modern world, a development they, with their strong emphasis on trade and exchange, largely welcome. The loss of some of their heritage from their period of isolation, their linguistic diversity, is a price they generally regard as well worth paying in order to facilitate their access to the wider world. And given their traditional linguistic ideology it is easy to see why they might look upon such a loss in this way. We may hold a different view, as linguists from our professional interests, or laypeople in Euro-American cultures due to our own Herderian (Foley 2005) linguistic ideology, but certainly no one can deny them that right to make their own choices, albeit to a large extent clearly guided, as we have seen, by their own linguistic ideology.


Foley, W. 2005. Personhood and language identity, variation and purism. Language Description and Documentation 3.9-38.

Foley, W. 2017. The languages of the Sepik-Ramu basin and environs. In Palmer, B., ed., The Languages and Linguistics of the New Guinea Area, 177-411. Berlin: de Gruyter-Mouton.

Forge, A. 1972. Normative factors in the settlement size of Neolithic cultivators (New Guinea). In Ucko, P., Tringham, P., and Dimbleby, G., eds, Man, Settlement and Urbanism, 363-376. London: Duckworths.

Harrison, S. 1990. Concepts of the person in Avatip religious thought. In Lutkehaus, N., Kaufmann, C., Mitchell, W., Newton, D., Osmundsen, L., and Schuster, M., eds., Sepik Heritage: Tradition and Change in Papua New Guinea, 351-363. Durham, NC: Carolina Academic Press.

Keesing, R. 1981. Cultural Anthropology: A Contemporary Perspective. Second edition. New York: Holt, Rinehart, Winston.

Kulick, D. 1992. Language Shift and Cultural Reproduction: Socialization, Self and Syncretism in a Papua New Guinea Village. Cambridge: Cambridge University Press.

Newton, D. 1975. Kubru shields: style and history. In Egan, P., ed, Art Studies for an Editor: 25 Essays in Memory of Milton S. Fox, 191-225. New York: Museum of Primitive Art.

Reddy, M. 1993. The Conduit Metaphor. In Ortony, A., ed., Metaphor and Thought, 164-201. Second edition. Cambridge: Cambridge University Press

Robbins, J. 2001. Ritual communication and linguistic ideology. Current Anthropology 42.589-612.

Telban, B. 1998. Dancing through Time: A Sepik Cosmology. Oxford: Clarendon Press.