On Biolinguistic Diversity

On Biolinguistic Diversity -

linking language, culture and (traditional) ecological knowledge

Tove Skutnabb-Kangas[1]

Invited plenary lecture at the interdisciplinary seminar "At the limits of language", organised by Department of Biology and Department of Linguistics and Philosophy, Universidad Autónoma de Madrid and Cosmocaixa (March 2004).

In press, in conference book edited by Angel Delgado Buscalioni, Luisa Martin Rojo, and Isabel Fuentes.

“Cultural diversity is closely linked to biodiversity. Humanity’s collective knowledge of biodiversity and its use and management rests in cultural diversity; conversely conserving biodiversity often helps strengthen cultural integrity and values” (World Resources Institute, World Conservation Union, and United Nations Environment Programme, 1992: 21).

“Biodiversity is not an object to be conserved. It is an integral part of human existence, in which utilization is part of the celebration of life” (Posey, 1999: 7). Or, mutatis mutandi, maybe: “Linguistic diversity is not an object to be maintained. It is an integral part of human existence, in which using the languages is part of the celebration of life”

Each Contracting Party shall “subject to its national legislation, respect, preserve and maintain knowledge, innovations and practices of indigenous and local communities embodying traditional life-styles relevant for the conservation and sustainable use of biological diversity and promote the wiser application with the approval and involvement of the holders of such knowledge, innovations and practices and encourage the equitable sharing of the benefits arising from the utilization of such knowledge, innovations and practices”.

(from Article 8(j) of the Convention on Biological Diversity)

1. Introduction

The following description of biolinguistic/biocultural diversity is on Terralingua's[2]. home page (

Terralingua supports the integrated protection, maintenance and restoration of the world's biological, cultural, and linguistic diversity –– known as biocultural diversity –– by creating innovative tools to analyze the links between these diversities and demonstrate their global significance.

Language, knowledge, and the environment have been intimately related throughout human history. This relationship is still apparent especially in indigenous, minority, and local societies that maintain close material and spiritual ties with their environments. Over generations, these peoples have accumulated a wealth of wisdom about their environment and its functions, management, and sustainable use. Traditional ecological knowledge and practices often make indigenous peoples, minorities, and local communities highly skilled and respectful stewards of the ecosystems in greatest need of protection. Local, minority, and indigenous languages are repositories and means of transmission of this knowledge and the related social behaviors, practices, and innovations.

As with biological species, languages and cultures naturally evolve and change over time. But just as with species, the world is now undergoing a massive human-made extinction crisis of languages and cultures. External forces are dispossessing traditional peoples of their lands, resources, and lifestyles; forcing them to [migrate or] subsist in highly degraded environments; crushing their cultural traditions or ability to maintain them; or coercing them into linguistic assimilation and abandonment of ancestral languages. People who lose their linguistic and cultural identity may lose an essential element in a social process that commonly teaches respect for nature and understanding of the natural environment and its processes. Forcing this cultural and linguistic conversion on indigenous and other traditional peoples not only violates their human rights, but also undermines the health of the world's ecosystems and the goals of nature conservation.

This description is a background for my paper. I will be asking questions suggested by the organizers: "What connections may exist between the loss of our biological heritage and that of our linguistic heritage?" and "With the gradual loss of some languages, will we be losing at the same time priceless knowledge of other ways of inhabiting nature and of resources for sustainable development?"

Most people realize the importance of maintaining biodiversity whereas relatively few know about or lament the much faster loss of linguistic diversity. A very short version of one of the important reasons for why linguistic diversities should be maintained is as follows: Linguistic diversity and biodiversity are correlationally and probably also causally related. Much of the knowledge about (necessary) elements of integrated ecosystems and the relations between these elements and about how to maintain biodiversity is encoded in small local languages. Their speakers live in the world's biologically and often also linguistically most diverse areas. Through killing these languages (or letting them die), we thus kill many of the prerequisites for maintaining biodiversity. In a short article, it is possible only to give a crude overview of some aspects of our present knowledge about the relationships between biodiversity and linguistic diversity, in the form of some of the classificatory tools used to approach the area, and point to questions and gaps in that knowledge. This includes some reflections on methodological challenges in relation to how to "prove" the causal relationship.

I would also like to mention some of the accusations, misconceptions, misunderstandings and critique of some aspects of ecolinguistics. Some of these refer to both the biodiversity - linguistic diversity connection and to work with linguistic human rights as one of the tools for supporting the maintenance of linguistic diversity. Just to use one researcher's misunderstanding as an example, Kibbee (2003) accuses ecolinguists of making use of the strong form of the Sapir-Whorf hypothesis (e.g. Kibbee, 2003: 47), of Social Darwinism (ibid.: 47), of neo-colonialism (ibid.: 55), paternalism (ibid.: 55), idealism (ibid.: 56), of wanting to artificially preserve languages against the will of their speakers (ibid.: 53), of advising people not to learn international languages like English (ibid.: 55), and so on. Unfortunately there is no time to discuss these but some clarification and caveats in relation to definitions of concepts can also be used to deconstruct the misconceptions (and see Skutnabb-Kangas, in press, a, b). Finally, some lessons from the present knowledge will be mentioned.

2. Definitions

2.1. General definitions of biodiversity, linguistic diversity and biocultural diversity

I start with general definitions of some of the basic concepts. Biodiversity can be defined as "the total variability among genes, plant and animal species, and ecosystems found in nature" (Skutnabb-Kangas, Maffi & Harmon 2003, Glossary, 55).

The number of species has often been used as a proxy for biodiversity. In the same way, the number of languages can be used as a crude proxy for linguistic diversity (see David Harmon's discussion about the caveats in both proxies, 2002; see also Maffi 2001, Note 3). The parallel definition of linguistic diversity is thus "the variety and richness of languages in human societies[3]" (Skutnabb-Kangas, Maffi & Harmon 2003, Glossary, 56). The parallel does not mean that languages and biological species are similar in all or even most ways - they are manifestly not[4]. Just to mention an obvious one: languages are not living organisms.

A very general definition of biocultural diversity is "the diversity of life on Earth in both nature and culture" (ibid., 55).

If we want to see what the correlation is between biodiversity on the one hand and linguistic and cultural diversity on the other hand, some measures of linguistic and cultural diversity have to be found and various kinds of indicators of biodiversity have to be chosen. This is what I shall do next.

2.2. Definitions and problems in definitions pertaining to linguistic diversity: language, mother tongue speaker/user, native speaker/user, culture, ethnolinguistic groups

The whole concept of language is extremely vague, as we know and there is no other scientific way of defining it except analysing the power relations involved in whose definitions about the relative languageness or otherwise of various idioms prevail and why (see my discussion of what a language is, in Skutnabb-Kangas 2000, Chapter 1). This is a typical example of the borders of a concept being in the perceptions of the observer rather than in the characteristics of the observed (see later). One example is the latest edition of the Ethnologue ( the best global source for languages. It lists 6.800 languages, but some 41,000 names or labels for various languages.

Even if we knew what a language is, we certainly have extremely unreliable figures about the number of speakers for most of them, including the largest ones where the differences of estimates of the speakers of the same language may be tens of millions (see Skutnabb-Kangas, 2002).

Languages are in most cases both known best and transmitted to the next generation by native speakers/users[5] or mother tongue[6] speakers/users of those languages. But we are likewise using contested concepts here: distinguishing mother tongue speakers or native speakers from those who have learned some language only later and for whom it is not their primary means of communication in childhood (or one of them, in case of childhood bilinguals or multilinguals) is extremely tricky.

If we could define language and native speaker, we might then measure the relative linguistic diversity of geographical units, for instance countries, through the number of languages spoken natively in the country. The most linguistically diverse countries would then be the ones which have most languages. Papua New Guinea, with its over 850 languages would be the un-detested world champion (Table 1).

But thisway of measuring linguistic megadiversity has also been contested. Clinton Robinson argues, for instance, that the most diverse country is not the one which has the largest number of languages, but the one where the largest linguistic group represents the lowest percentage of all linguistic groups (Robinson; 1993). We get a very big difference in the list of the world's linguistically most diverse countries, depending on which measure we use (Tables 1 and 2; source: Tables 1.1 and 1.3 in Skutnabb-Kangas, 2000, pp. 36-37).

Table 1. Linguistic megadiversity: The countries with most languages in the world

Based on figures given in Krauss 1992: 6.

Table 2. The ten most linguistically diverse countries, according to Robinson 1993

LLG = Largest Language Group (adapted from Robinson, 1993: 55; based on figures in Ethnologue, 12th edition)

Measuring cultural diversity is even more difficult, regardless of how "culture" or "cultural traits" are defined (see, e.g., articles in Posey, ed., 1999, and Maffi, ed, 2001, for a good sampling). Since I concentrate on biolinguistic rather than biocultural diversity in this article, I will not discuss this further here. Suffice it to say that biolinguistic diversity is a more narrow concept than biocultural diversity; language is included in culture.

And putting languages and cultures together is even more risky, since there are many examples of non-convergence both ways. There are several cultural groups using the same language, or one cultural group using two or three different languages.

When ethnicity, another contested concept, is added, so that we get ethnocultural groups defined on the basis of languages, the measures become even more vague. Before I define ethnocultural groups, I present some of the caveats and challenges. All the concepts used, language and mother tongue, culture, and ethnicity, are social constructs, not inherited givens; they are hybrid and nomadic, dynamic and changing, not static; people may claim several of them at the same time and be multilingual and multicultural, and "multi-ethnial", or "bicountrial"[7]. All of them play ever-changing roles for people's multiple identities, and are variously focussed and emphasized in various situations and at various times; their salience is always variable. ALL identities are of course constructed to the extent that we are not born with identity genes. Even in cases where we are talking about phenotypically visible genotypical features like skin colour, very obviously the way these features are interpreted, are social constructions, not innate.

But with all these caveats, it is still the case that many of those groups who demand linguistic human rights do claim these concepts: they claim to know what their mother tongues are and which ethnic or ethnolinguistic or ethnocultural group(s) they belong to. They see their language as a "cultural core value" (Smolicz, 1979). Most indigenous peoples who have pronounced on their languages, share the attitudes from Canada, described by Mary Joy Elijah, herself Oneida. In her literature review (2002) she quotes from Resolution No. 9/90 Protection of First Nations' Languages, Special Chiefs Assembly, Ottawa, Ontario - December11, 1990, Georges Erasmus National Chief:

SUBJECT: Protection of First Nations' Languages

WHEREAS language is a direct gift from the Creator; and

WHEREAS First Nations languages are the cornerstone of who we are as a people; and

WHEREAS our culture cannot survive without our languages; and

WHEREAS the right to use and educate our children in our aboriginal languages is an inherent aboriginal and treaty right,

THEREFORE BE IT RESOLVED THAT, as aboriginal people of this country, First Nations languages must be protected and promoted as a fundamental element of aboriginal heritage and must be fully entrenched in the Constitution of Canada; and

FURTHER BE IT RESOLVED THAT the federal government has a moral and legal obligation, through (pre-Confederation) treaties and through legislation, to provide adequate resources that will enable First Nations languages to exercise this right.

There is in reality a very high degree of convergence between ethnicity, culture and mother tongue, regardless of how much liberal political scientists or post-post-modern sociolinguists want to denounce this, and "disinvent" the concepts. I have not seen more than a few dozens of examples of non-convergence, and they seem always to be the same ones. Even if there were several hundreds of them, they would still be exceptions rather than a rule, and even if exceptions are important as checks on theories, generalisations cannot build on exceptions but on what is more common. Several colleagues seem to try to raise exceptions to rules in recent debates in an eagerness not to be accused of essentialising (e.g. May, 2001, in press; Pennycook 2002). Likewise, the same few examples of loss of language, with the culture and identity still living on (the Irish, the Jews, and a few more) are always repeated and then used as proofs when claiming that there is little or no relationship between language and culture. It seems that several of the critics somehow automatically assume that if something is "constructed" (rather than "innate" or "inherited" or "primordial" or whatever one sees as the opposite of constructed) it is somehow a less valid concept. Knowing and accepting that a concept (like "mother tongue", "language", "ethnicity", "culture"), is socially constructed, does not in any way need to invalidate the concept. All science, for instance, is socially constructed - still our conclusion is not that we should stop talking about science or stop using it, or that we should somehow "disinvent" it. Critical analysis also means being open and not accepting vogue taboos.

All this knowledge, then, can be used to relativise the definition of ethnolinguisticgroup, used by WWF and Terralingua (Oviedo & Maffi, 2000: 1)

A human social unit that shares the same language and culture and uses the same criteria to differentiate itself from other social groups.

With this definition, WWF and Terralingua use the figure of an approximate total of 6,867 ethnolinguistic groups in the world. Terralingua has also used it for the map showing the distribution of biocultural diversity (Appendix to Skutnabb-Kangas, Maffi & Harmon, 2003).

If we want to relate the distribution of these groups to the distribution of biodiversity, we have to choose indicators for biodiversity and its worldwide distribution.

2.3. Definitions pertaining to biodiversity: Global 2000 Ecoregions, biodiversity hotspots, megadiversity countries

As already mentioned, the number of species has been used as a proxy for biodiversity. The first difficulty here is that we really have very little solid knowledge of these numbers; much less than about the number of languages. Figures between 5-15 million separate species are "considered reasonable" (Harmon, 2001: 63), with a "working figure" of about 12,5 million. But figures as low as 2 million and as high as 50 million(Maffi, 2001, Note 1) or even 100 million (Solé et al., 2003: 26) have been mentioned The highest figures are based on the estimate that most of the world's species (maybe up to 90%, Mishler, 2001: 71) have not yet been "discovered", i.e. named and described by (mostly Western) scientists; only some 1,5 million different species (from plants and animals to fungi, algae, bacteria and viruses) have so far been identified by natural scientists. Many may become extinct before having been studied at all.

A relatively simple global measure of ecological diversity which corresponds to the linguistic megadiversity list in Table 1 is megadiversity countries. These are "countries likely to contain the highest percentage of the global species richness" (the definition here is from Skutnabb-Kangas, Maffi & Harmon, 2003: 56; but see for an essay about them by Russell Mittermeier (et al.), the originator of the concept; see also Conservation International at

Trying to make the relative distribution of the world's known biodiversity more meaningful than just listing plain numbers of species in various countries, researchers have developed concepts covering larger units where there is a high concentration of species. The two I want to mention here are ecoregions and biodiversity hotspots.