Go to the Digital Library top page!

Language Arts

Lost Tongues and the Politics of Language Endangerment

by Salikoko S. Mufwene

Question: What do we mean by language endangerment?

Salikoko Mufwene: We talk about language endangerment more or less in the way that ecologists speak of species endangerment. There are parts of the world in which the local physical ecology has changed to the point where some indigenous species cannot survive or thrive in them. Likewise, the economic and social ecologies of some indigenous languages have changed to the point where those languages can no longer thrive. The speakers are shifting to other languages as vernaculars. I think this is a technical term that I should explain. A vernacular is the kind of language that people use for day-to-day interaction, as when they communicate with their friends or their relatives.

Question: Would the way that you and I are speaking now be called a vernacular?

Mufwene: Well, I don't know, and the reason why I don't know is because I'm a non-native speaker and my registers don't vary a lot. Whereas for native speakers, there are times when communication is not particularly personal, and that more formal variety of communication is not the vernacular. But when people are fully relaxed and can use basic terms, terms that could sometimes be prohibited from formal settings--that's the vernacular.

Question: So language endangerment refers both to migration from the vernacular as well as the disappearance of a vernacular entirely?

Mufwene: It can describe the fact that the vernacular is no longer used, where it would normally be used. But in some communities, it's only the vernacular that is used for anything, and the language itself is being abandoned. Look at Native American communities in the United States: A lot of Native Americans no longer speak their ancestral languages but use only English in the United States, French in Quebec, or Spanish or Portuguese in Latin America. That shift from Native American languages to the colonial languages produces language endangerment.

But in Africa you come across cases where one of the indigenous languages has expanded its position and is used in many parts where it wasn't spoken before. Swahili used to be spoken primarily on the coast of east Africa. Now it has become the national and official language of Tanzania. It has also become the national language of the eastern part of the Republic of the Congo, and even the national language in part of Kenya. In these places it functions as what we call a lingua franca. But in urban centers it is also spoken as a vernacular, especially by people who were born in the city. The people forget the traditional ethnic allegiances which were associated with specific ethnic languages and speak just Swahili, alternating it with French or English. In those particular kinds of settings, we say that the indigenous, ethnic languages--other than Swahili--are endangered. In other words, they are threatened by the expansion of Swahili.

Question: How serious a problem is this? Is it possible to estimate how many different spoken languages there are in the world and what percentage might be endangered?

Mufwene: The statistics that I have read recently say that there are about 6,700 languages spoken around the world. They estimated that within a century or so, there will be just about 3,000 languages still spoken. And within two centuries or so, there will be even fewer languages. The number might just drop to a couple of hundred.

That approximation is based on the speed and extent of globalization these days. The major colonial languages--English, French, and Spanish--are among those that will make it. Languages of big economic powers such as Japanese are likely to make it, as are languages spoken by millions of people, such as Chinese and Hindi. By the sheer number of speakers they stand a good chance of making it, although some of the languages associated with industrialization and globalization--and we refer here to English especially--are likely to prevail. Languages spoken by small ethnic communities, perhaps only by 2,000 people, or by a couple of hundred-- will die out very fast. But I have reservations about this assessment, because you cannot determine the vitality of a language based simply on the number of speakers.

Question: How can you evaluate the vitality of a language?

Mufwene: It's a sign of vitality if the community, be it big or small, continues to speak the language and transmit it from one generation to another. There are many parts of the world where communities have always been small and those communities have always lived in isolation from the rest of the world. Just because a community is small does not necessarily mean that its language is endangered as long as the community remains isolated. But once it is absorbed by a larger and more influential community, then you can say that its language is endangered.

Question: Is this what made Swahili so successful--its ability to absorb the languages around it?

Mufwene: It's a long history. The Swahili coastal people were already in contact with several other people when the Arabs decided to trade with east Africa. The Arabs adopted Swahili as the language of trade, and as trade expanded inland, toward the hinterland of Africa, they took Swahili along with them, and so a lot of people who traded with the Arabs used Swahili. When Europeans colonized Africa, they adopted Swahili as the language of colonization in interacting with the natives--I think the first were the Germans in Tanzania. Later on, the British and the Belgians also adopted Swahili as the lingua franca of colonization wherever they found it spoken.

The association of Swahili with Western colonization gave Swahili more power and prestige, and then little by little in the emergent new cities during the colonial period, children found it more and more practical to speak the language that everybody spoke, rather than to continue speaking their ethnic languages. Then the people who migrated to the city later, finding that Swahili was mostly spoken there, also learned Swahili. The association of Swahili with urban life has had an effect on rural populations, because whenever they have to interact with people in the city, they have to resort to Swahili too, rather than people from the city trying to speak their local indigenous languages.

So that explains the vitality of Swahili. It has acquired these connotations of urbanism, of modern life, next to the European languages that are at the top of the ladder. But a lot of Africans have to go to school to learn a European language, whereas people don't really have to go to school to speak most of the African languages. Just by interacting with other people you proceed by trial and error until you achieve a certain level of competence to communicate well with others.

Question: So people shift to the language that will afford them the greatest amount of opportunity, and in essence, this ties into power structures as well?

Mufwene: Yes, it is tied with power structure, and that is when I start thinking of globalization as an important factor in determining whether or not a language thrives. Globalization more or less reduces diversity and suggests homogenization. And it also suggests more or less speaking the same kind of language everywhere, just as it suggests using more or less the same kinds of products everywhere for the same kinds of things, or producing things in more or less the same way. There is this kind of invisible drive that controls everything everywhere, and people are being manipulated by that invisible but powerful drive.

Question: Do you think this causes people to change their language patterns in an unconscious way?

Mufwene: Yes, I do. People don't necessarily say "I'm no longer going to speak my native language." I think that when we start speaking of language endangerment, these are just consequences of the acts of speakers as we communicate on a daily basis and don't really think about the ultimate consequences of unconscious decisions that we make. We tend to speak language A rather than language B without necessarily realizing, until probably too late, that we have been favoring language A at the expense of B.

Question: To what extent are a language and its dialects or vernaculars defined by geography? Or would you say that's not really the case?

Mufwene: For some languages, yes; and for some other languages, no. One of the points that I made in an article on the evolution of English is that you can no longer define English as the language of England. Over a thousand years ago, that was the natural definition, the way you say Japanese is the language of Japan. Back then, we could have said English is the language of England. But now you cannot say that kind of thing, because English is also the language of North America, of Australia, and whether people accept it or not, it is among the languages that you find in Africa and Asia.

And still there are people who claim that you have to define English by who speaks it natively. That is absurd. When we count how many speakers of English there are in the world, the vast majority of those people are not native speakers. You just can no longer define English by the homeland of its original speakers. You could probably define several languages still spoken by very small groups geographically, better yet ethnically; but such identifications would apply inadequately to major languages, which have spread beyond the boundaries of their original homelands. One of the reasons why some scholars specialize in indigenized varieties of European languages spoken as lingua francas in former European exploitation colonies is the legitimacy and relative autonomy of the linguistic norms of these varieties. Their speakers, who are predominantly non-native speakers, feel free to innovate new structures and meanings in ways that suit their communicative needs and habits.

Question: So Gullah or the Ocracoke Brogue can be defined geographically and ethnically, but you think they have very different fates in front of them?

Mufwene: That's right. The factor in that case is integration in the larger population. Ocracoke Brogue is spoken by a white minority in a country with a white majority. When affluent people from the white majority migrate to Ocracoke Island, they can integrate the local whites. But when affluent whites migrate to Johns Island or Wadmalaw Island where the black traditional speakers of Gullah live, the island is structured on the model of the average American city, which is ethnically segregated. These islands have also become ethnically segregated, and the populations indigenous to them are not absorbed by the new white immigrants. They live on separate parts of the islands, and Gullah is not as much threatened by the new white population as the Ocracoke Brogue is. On Ocracoke, the indigenous whites marry the white newcomers and little by little the two communities become integrated. That is not happening on those islands where Gullah is spoken.

There is also a question of identity at work. Blacks are now in a situation where they feel more motivated to assert their identity by speaking in their own traditional way. This is more and more evident in places such as Hilton Head, South Carolina, where affluent whites have developed expensive estates that have increased real estate tax value on the island--a change that drives the indigenous poor, majority blacks, out. On the other hand, whites are more likely to speak in a way that is not going to confuse them with the blacks. It's the same kind of thing that we observe in the average American city.

Question: On the one hand, it seems as if social and economic forces are very powerful in deciding and informing people's linguistic traditions and what language they're going to speak. But you're also saying that identity and cultural factors can be a strong force as well.

Mufwene: Let me put it in a historical perspective, going back to the time of emancipation from slavery. When that happened and you start comparing the Caribbean and United States, you note these apparently contradictory linguistic developments. I mention the Caribbean because blacks have always been the majority there, and emancipation really produced social integration. Whites and blacks in the Caribbean speak alike, and if you notice any difference, it is socio-economically based, not ethnically based.

In the United States, whites have been the majority all along and emancipation actually produced more segregation. The Jim Crow laws were passed in the late 19th century in order to foster more segregation, which paved the way for black and white vernaculars to diverge, especially in the tobacco and cotton plantations of the South, where poor whites and descendants of Africans had spoken almost the same varieties until the late 19th century. Although that segregation de jure was abolished in the 1960s, segregation de facto has remained. By law, segregation is prohibited, but in practice it is all over. Whites and blacks socialize very little outside the work place and they speak differently; their vernaculars have little influence on each other's.

Question: Is that still the case?

Mufwene: Yes. Blacks do not intend to speak like whites, and whites do everything not to be confused with blacks in their speech except in small communities in the middle class where people are well integrated, and where you see children speaking alike. Still, when those children go and live in other communities that are more segregated, they are under pressure to assimilate to those communities. When black students from integrated neighborhoods go to predominantly black colleges, they are under pressure to adopt black speech patterns. Otherwise they are marginalized.

Question: Given the ability for language to speciate, to diversify, would it ever be possible to move to one global language?

Mufwene: I doubt it, because there will always be a way for whichever language prevails to speciate into other varieties. You can tell that today with English. American English, British English, Australian English, and South African English are all different. As noted above, in African and Asian former English-exploitation colonies, indigenized English varieties have developed which are spoken by the indigenous elite. Even though we continue to call all these varieties English and speak of them as the same language, we can no longer guarantee that two speakers of English will necessarily understand each other.

Question: How do you decide when something becomes a dialect?

Mufwene: That's a political decision. It's one of my favorite topics, in the sense that if you want to keep people in the franchise you claim that they all speak the same language, but if you want to disfranchise some speakers--or if the speakers themselves don't want to remain in the franchise--then it's a separate language or at least it is given a special label, such as Creole, pidgin, or indigenized English. When the speakers themselves claim this, it's like declaring independence and saying we speak our own separate language.

Question: Where, for example, do you see this?

Mufwene: Gullah on the coast of South Carolina. Speakers of Gullah say they speak English, but outsiders to the community disfranchise them and say they speak a separate language. And when you ask them why, they say, "We don't understand them." But ask those same people whether they understand speakers of Appalachian English, and they say no. They don't understand them, either. Do these people speak a separate language? Again, they say no, they speak Appalachian English. On the other hand, some of the elite from the same communities that speak such disfranchised varieties are often proud to declare their variety a separate language. This is the case of Ebonics, over which many Americans were conflicted in late 1996 and early 1997, when the Oakland School District Board decided to use it in the classroom.

The main reason for disfranchising Gullah is that its speakers are generally blacks, and it's very easy to invoke lack of intelligibility to claim that they speak a separate language, whereas, because speakers of Appalachian English are whites, the same "judges" don't want to disfranchise them that easily.

Here's another example: If you take the variety of Jamaican English that is called Creole, people say it's a separate language. But if you take a similar variety that has developed in Australia or New Zealand, nobody says that's a separate language, although for the average American it's very difficult to understand an average Australian or New Zealander vernacular speaker.

Question: If you ask them they would say they're speaking English.

Mufwene: Yes. They would say a New Zealander speaks English but a Jamaican doesn't speak English. So, not too long ago, HUD published a pamphlet in Jamaican speech, written in "eye" dialect, which is the representation of non-standard or non-native speech through a distortion of standard spelling in a way that suggests peculiarities of that speech. This suggested that West Indians must not understand standard American English and ignored that West Indians too are schooled in standard English. The sad part is that the targeted West Indians had a hard time reading the document, probably as much difficulty as the average American who is not familiar with eye dialect. Those things are already happening, and even in linguistics we are not quite aware of this. So by applying special labels to some varieties, we have already been playing that politics of disfranchising some people.

Question: Are there other ways emerging to analyze when something is a different language, and when it's not?

Mufwene: You cannot prove it by the structural features of a language because there are varieties that are not very different structurally, yet their speakers claim that they speak separate languages. Take Danish, for instance, and Norwegian. They are very similar, but the Danes think they speak a separate language from the Norwegians, and the Norwegians also think the same way. The same thing goes with Swedish.

Question: What do you think?

Mufwene: Well, I just have to accept what the speakers think. It's similar to the opposite situation with classical Arabic, which is used in the Koran, and Moroccan Arabic. They are completely different languages. If you speak Moroccan Arabic, you still have to learn Koranic Arabic as a separate language in order to be able to understand it. It's like asking a speaker of Spanish to learn Latin. But in the Arab world, despite big differences between Koranic Arabic and vernacular Arabic varieties in these countries, they all think it's the same language.

That's why one has to deal with politics and ideology in identifying languages. When we talk about language endangerment, and ask ourselves whether there will ever be a time when the whole world speaks the same language--one and the same language--those are other dimensions of the problem that we should think about. Because things are developing in such a way that whatever language prevails, that language is not going to remain the same. When I talk about this in my work, I speak of Pyrrhic victory: You win, but you lose at the same time because you are not the same anymore.

ABOUT THE AUTHOR | Salikoko S. Mufwene

MufweneSalikoko S. Mufwene is professor and chair of the department of linguistics at the University of Chicago, from which he earned his Ph.D. in 1979. He has published over 150 essays on the development of creoles and African-American Vernacular English (AAVE), on the morphosyntax of Jamaican Creole, Gullah, AAVE, and Bantu languages, and in lexical semantics. He served as columnist and associate editor of the Journal of Pidgin and Creole Languages (1988-1995) and is now on the editorial boards of several journals, including World Englishes, Language Variation and Change, and Langage et Societe. His books include The Ecology of Language Evolution, to be released June 2001 by Cambridge University Press. He argues in it that creoles have developed by the same evolutionary processes that have produced changes considered more "normal" or "ordinary" in language; the same competition-and-selection mechanisms that account for language change also account for language vitality and loss. In all cases of language evolution, the ecology rolls the dice and determines winners and losers, be they languages or structural features. His approach to language evolution is inspired by population genetics and macroecology.

More information on his research and work in progress is available at his web site. (http://humanities.uchicago.edu/faculty/mufwene/)

COPYRIGHT | Copyright 2001 The University of Chicago.

(c) 2004 The University of Chicago :: Please direct questions or comments to furlong@lib.uchicago.edu

Need help searching?
Search help

Search eCUIP:

Examples: or
Contact eCUIP!

Need help?

Return to the eCUIP top page!