Globalization and the Myth of Killer Languages: What’s Really Going on?

Posted on in Politics and languages 229 vedi

Salikoko S. Mufwene

Globalization and the Myth of Killer Languages: What’s Really Going on?



Globalization and colonization

Precursors of modern colonization and globalization

Will there be an English-only Europe?

The economics of language maintenance vs. shift






This paper is about some misgivings I harbor over some recent linguistics publications on language endangerment, i.e., the process whereby a language loses grounds to another either because it is spoken by fewer and fewer speakers and/or because it is heavily influenced by the competitor, which its speakers seem to command or like better. The targets of my criticisms include Crystal (2000, 2004), Dalby (2002), Nettle/Romaine (2000), Maffi (2001), Phillipson (2003), and Skutnabb-Kangas (2000), to whom I will often refer collectively, as they basically share the same positions, though there are minor differences in the specific ways they state them. Oversimplifying reality, they have generally reduced globalization to a process that makes the world more and more uniform by the world-wide diffusion of intellectual, linguistic, military, technological, and other cultural products associated with hegemonic regimes such as the USA, the United Kingdom, and Australia. In other words, they equate globalization with what would be the westernization of the world and talk about indigenous languages as if indigenity were no longer a relative notion and such languages were to be found only in the Americas, Asia, and Australia (Mufwene 2002). Capitalizing on the increasing attestations of Hollywood movies and McDonald restaurants around the world, they speak of McDonaldization and Americanization as if these phenomena either instantiated or produced globalization, whereas they should instead be treated as its (by-)products.

To be sure, these authors are rightly concerned with growing political and economic power inequities (Stiglitz 2002, Steger 2003) that have been in favor of especially North America, Australia, Japan, and of Western Europe and have led to the endangerment or loss of several vernaculars not associated with their dominant, global economic systems. However, their generalizations, which have projected the same situation all over the world, have not been equally sensitive to the fact that, like the European colonization of the rest of the world over the past half-millennium, economic and cultural globalization has not proceeded uniformly. Ironically, Nettle & Romaine (2000), for instance, make this world heterogeneity obvious in showing the differential impact of Western European languages on the indigenous ones in the Americas and Australia. The rate of language loss is much higher in North America than in Central and South America, while, I can add, the process is already complete on the Caribbean islands. I discuss the reasons for this differential evolution below. Overall, the critical connection between globalization and colonization has not been fully explored, a shortcoming that has led to the confusion of the economic process of globalization itself with the unprecedented insidious and pervasive way in which it has proceeded since the collapse of the Berlin Wall, or the disintegration of the Soviet bloc, in the late 1980s (Steger 2003). Rare are scholars of language endangerment who have provided a historical perspective (e.g., Hagège 2000, Dalby 2002) that should help us make better sense of what has been witnessed over the past few decades.

A special metalanguage has also emerged, in which languages are treated as agents and speakers of the endangered or dead languages only as victims. Not only has it been acceptable to speak of “language wars” (Calvet 1998) but also English has been portrayed as the “killer language” (Price 1984; Nettle & Romaine 2000) par excellence which by its own actions putatively has driven, and is still driving, many languages to extinction. False concerns have likewise arisen about English endangering continental European languages, because it is now the dominant lingua franca of the European Union (Phillipson 2003), the one that most bureaucrats and delegates are likely to speak in settings that do not require ideology-based use of language, for instance, in offices and in the corridor, outside parliamentary sessions and ministerial meetings (Swaan 2001). The distinction between the vernacular and lingua franca functions of languages has completely been overlooked in such cases of language competition, critical though it is to determining whether or not a language is endangered.

In the same vein, the rhetoric has been less about the rights of speakers than about the rights of languages to survive (Skutnabb-Kangas 2000) and, in much of the linguistics literature, about the benefits of linguistic diversity to the linguistics enterprise (especially the extent to which research on language universals and typology is negatively affected by the lost languages). The literature has generally underscored the cultural impoverishment of the affected population – not withstanding issues arising from such a static notion of “culture” – but little has been said about whether it has all been losses and no gains among the relevant populations who have somehow adapted to their changing socio-economic ecologies.

In this essay, I elaborate positions articulated in Mufwene(2001, 2002, 2003a, 2004) to show that the competition and selection following from the contact of languages made possible by colonization are a much more complex process, which proceeds differentially, consistent with the ecology-based approach to language evolution outlined in Mufwene (2001). I submit that globalization itself is an aspect, if not a product, of colonization, which can be traced back to the beginnings of agriculture and its impact on human societies (see also Cowen 2001). It is a non-uniform and non-unilinear process which has affected the linguistic landscape of the world differentially and calls for an account that should avoid oversimplification and unidimensionality.

Globalization and colonization

Globalization is not synonymous with what is immediately suggested by the French word mondialisation (roughly, “universalization” as covering the whole globe).1 It can be local or regional, having to do with interdependencies in a system or a network (such as in the industrial world). Global economic systems in this local sense are actually the unmarked structure in polities such as the Americas and Australia, where indigenous languages have massively been replaced by European colonial languages over the past couple of centuries. They have evolved from a European tradition in which numerous vernaculars have been replaced by national, official languages.2

Globalization at the world-wide level is a function of how faster and more reliable communication, both in terms of means of telecommunication and transportation of people and goods, has made it possible to establish interdependent industries and resources that are thousands of miles apart. It is also a function of how economic colonization by multinational corporations has replaced the political and economic colonization of some territories by specific metropolitan nations, such as in the former British Empire.3

Globalization cannot be associated with uniformity either, as many of the diffusions associated with globalization acquire local characters and therefore reflect some cultural hybridization (Pieterse 2003, Tomlinson 1999). The McDonald menu is partly adapted to the local diet, in the same way as the restaurant Chinese cuisine is produced differently in different parts of the world (e.g., Hong Kong vs. France vs. the USA). Like Chinese restaurants, McDonald eateries also serve their customers in the local language. Major prestigious hotel chains such as the Hilton and the Hyatt, which lure American businessmen abroad, are typically also decorated in ways that reflect local customs, up to the size of the room, although these are bigger by local standards. And they too operate in the local lingua franca and currency, although special accommodations are made to American guests.4 We should normally examine things beyond the epiphenomena that strike our initial perceptions and, as observed by Tomlinson (1999), resist overgeneralizing from the illusion of uniformity given by the “globalized [business] spaces” of air terminals and five-star international hotels to the larger cultural spaces from which these economic islands stand out. We should determine the extent to which these manifestations of globalization qua mondialisation are integrated in the local cultural ecologies, instead of assuming a priori that only the local cultural system is affected by the imported culture. And we should think beyond usage of English at a Hilton in Tokyo or in Paris to ask ourselves why this foreign language is being used in such a setting and whether the setting is representative of other domains of language use in the larger society (in this case Japan or France), such as in domestic communication, in the same cities. It is also informative to determine whether the lingua franca used by airline employees and by customs and immigration officers at airports also functions as a vernacular among themselves. As becomes obvious below, multilingualism does not automatically endanger those languages that are not associated with economic or political power.

It is difficult to understand globalization as a form of either partnership or colonization without making sense of the notion of colonization itself. Outside population genetics, colonization has typically been associated with the political and economic domination of a territory by another in the interest of its own metropolitan industry and/or population. But the history of mankind suggests that we not overlook its meaning as the settlement of a territory by a population originating in another, such as in the development of some diaspora communities. One can also argue that the more common political and economic domination interpretation of “colonization” is a consequence of the resettlement interpretation. Both interpretations bear on the following discussion of the differential effects of globalization.

One way or another, colonization started the first time agriculturalists and pastoralists encroached on the territories of hunter-gatherers and either assimilated these residents or drove them away. The earliest instances of language competition must have started then, although we cannot imagine much of what happens until we think of, for instance, the Bantu populations dispersing into territories formerly inhabited by the Pygmies and the Khoisans, gradually driving the indigenous languages to extinction.

Linguists have seldom addressed the following relevant questions from the point of view of language evolution:5 Have the Pygmy and Khoisan languages not contributed to the speciation of the Bantu languages through their substrate influence? What can linguists tell us about the regional balance sheet from the contacts of Bantu, Pygmy, and Khoisan languages in Sub-Saharan Africa? That is, how many languages have died and how many new varieties have emerged? From the perspective of language birth and death, have things always evolved in the direction of fewer and fewer languages being spoken? Were there more languages spoken on earth when human populations were smaller? Is there paleontological evidence that supports such an assumption?

I argue below that colonization is related to globalization to the extent that both have led to the establishment of asymmetrical economic interdependencies between territories or parts of their industries, to unilateral control of resources, to attempts for uniformity in the modes of production and in the transportation of goods, and to inequities in the ways that costs and benefits are distributed.6 In both cases, they have increased the sphere of economic interaction.

Recent colonial history, that of European expansion since the 15th century, also suggests that at the macro-level we distinguish between three basic economy-driven colonization styles: trade colonization, settlement colonization, and exploitation colonization. It also shows that the latter two generally emerged from, or concurrently with, the first. One way or another, trade colonization did not last forever. The economically or militarily more powerful partner wound up subjugating the weaker or weakened one, thus imposing the terms of trade and/or of the exploitation of the coveted resources. While it lasted, trade colonization had no negative impact on the indigenous languages, unless one subscribes to a purist position that is opposed to lexical borrowings and other xenolectal structural influence on their language.

The typical balance sheet from these contacts is that trade colonization has often produced new languages, called pidgins. They are based on European languages in Africa, the Indian Ocean, and the Pacific but on indigenous languages in the Americas. Pidgins based on European languages include Nigerian and Cameroon Pidgin and Englishes, Babu English in India, Tai Boi in Vietnam (the only documented French pidgin), Tok Pisin and Bislama in, respectively, Papua New Guinea and Vanuatu.7 Trade pidgins based on Native American languages include the Mobilian Jargon in the Southeastern USA, Delaware Pidgin in the area from New Jersey and New York to Delaware, Chinook Jargon in the Northwestern part of the United States and British Columbia, and Lingua Geral or contact-based Tupi in Brazil. Overall, trade colonization has enriched the linguistic landscapes of the relevant territories. Things would not change until a trade colony evolved into a settlement or exploitation colony. I submit that from the point of view of language competition, the linguistic outcome of this contact evolution was different, depending on whether a territory was colonized on the settlement or exploitation style.

One important characteristic of European settlement colonies is that the European colonists typically maintained their respective languages as vernaculars, up to when one of them one gradually prevailed as the exclusive vernacular of the citizens of the new polity,8 such as English in the USA and Portuguese in Brazil. An important reason for this evolution is that, unless they were indentured laborers, the colonists lived in nationally segregated communities often until the early 20th century. Although languages such as Italian and German did not die out until the 20th century (and pockets of them continue to be spoken in Brazil and Argentina), settlement colonies have generally caused a demographic attrition of languages other than the dominant European languages. Eventually, the same process would affect several indigenous languages, being the most advanced in North America and in Australia (based on Nettle & Romaine 2000).9

Things have not evolved the same way in all settlement colonies. They must be subdivided into at least two categories: 1) those whose primary industry consisted of plantations of sugar cane (the dominant kind) or rice, relying on slave labor or non- European contract labor since the second quarter of the 19th century, and whose majority populations now consist of non-Europeans (e.g., Anglophone and Francophone Caribbean islands and coastal sates south of the Caribbean Sea); and 2) those whose industry was diversified and relied marginally or not at all on plantations, whose farming industry used mostly or a significant proportion of European indentured servants, and whose majority populations consist of descendants of Europeans (e.g., North America and the South American mainland). In both cases, we must understand that in shifting to the European languages of their polities Native Americans typically followed in the footsteps of African slaves and some European immigrants before them, as these immigrants were absorbed in the new economic world order before Native Americans could compete in it.

The Africans were typically first to give up their languages in part because they were initially integrated (though certainly discriminated against) in the homesteads, where they were minorities. However, another important reason is the gradual way in which the relevant colonies evolved into segregated plantation societies in which the creole slaves no longer spoke African languages. Yet, it is by them that the bozal slaves, those coming freshly from Africa, were “seasoned” (i.e., acculturated) to the new society. For the bozal slaves, who were societally multilingual, this acculturation process entailed learning the Creoles’ vernacular, especially since they (the Bozals) did not have a common group language to speak among themselves. Under the virtually Babelic conditions of the growing plantations, the slaves’ subordinate socio-economic position made it impossible for one African language to prevail as a lingua franca (future vernacular) on any particular plantation, least of all in any particular polity. Undoubtedly because the European colonial languages also enabled the slaves to function in the socio-economic systems in which they were exploited, the creole varieties into which they evolved would become their vernaculars. By the time African and East Indian contract laborers were brought, after Emancipation, to replace the former slave laborers on the plantations, African languages and cultural practices were too stigmatized to be maintained as vernaculars, even within groups that were relatively homogenous ethnolinguistically, such as the Yorubas of Trinidad (Warner-Lewis 1996).10

As noted above, the Europeans too gave up their heritage languages under the pressure of an emergent dominant national socio-economic system that functions only in one language. Subscribing to the same ideology of monolingualism as their counterparts in Europe,11 most of those in North America, for instance, who did not speak English as a vernacular gave up their heritage languages as they gradually assimilated to the dominant Anglo-Saxon regime in order to feel more integrated or to feel less discrimination. The process seems to have been faster when they were also stigmatized, or excluded by the 19th-century ideology of race from the category of (pure) White, as in the case of the Italians and the Irish (Fenton 2003, Guglielmo & Salerno (2003). The processes of language endangerment and loss in settlement colonies cannot be adequately understood without showing, as I have attempted here, to what extent the speakers of the affected languages themselves were involved in the socio-economic activities that produced these negative results. Languages were not warring each other. Speakers simply adapted to challenging and changing socio-economic ecologies, with the noticeable distinction that they have responded to changes imposed by forces outside them.

The agency of the speakers of the relevant languages is more obvious in the case of English and other colonial European languages (notably French and Spanish) in former exploitation colonies, to which I return below. Those spreading them around the world are primarily not the European native speakers of the colonial languages but the Natives themselves, who have endorsed the hegemonic status that the European colonizers have conferred to their languages, having associated political power and to some extent also economic development with them, and having promoted them in the name of democracy (equal rights of the citizens to the dominant language) in their school systems, at least in the later stages of colonization. Where European languages have not faced much resistance – especially in former settlement colonies, where they have become popular vernaculars – indigenous languages have suffered attrition or death by the “neglect” of their own (would-be) speakers. I return to this question below.

As observed in Mufwene (2003b), the neglect or abandonment of indigenous languages of the Americas and Australia by Native Americans and the Australian Aborigines has little, or perhaps nothing, to do with loss of pride in one’s heritage. Loss of pride cannot be associated with language shift by the Irish, Italians, and other Europeans either – at least it would not be the primary reason. Several highly stigmatized varieties, such as basilectal African-American vernacular English, Gullah, Appalachian English, and Old Amish English continue to be spoken despite the stigma associated with them. Speakers stick loyally to the group identity that such languages assign them. Adaptation to changing socio-economic ecologies is the primary reason of language loss, which actually occurs through the cumulation of decisions taken individually by the members of the relevant population as they face the communicative challenges of their socio-economic environments. Language loss is noticed only retrospectively and language endangerment is noticed when the cumulation of such individual responses has progressed so far that proportionally fewer and fewer people find the heritage language useful to their communicative needs.

Among the questions that interested linguists still need to address is why the loss of Native American languages has been more extensive in North America than in Latin America. Could the geographical physical ecology of Central and South Americas hold part of the explanation? Shouldn’t we consider, for instance, the fact that the Amazon forest remained virtually impenetrable to European immigrants until the 1970s and its Native American residents and their languages became vulnerable to Europeans at the same time that its deforestation became intensive and out of control? Doesn’t it matter that Native American languages have survived the most in countries such as Paraguay and Bolivia, which are the least industrialized?

The above remarks lead us to consider again the case of exploitation colonies, in which, to adopt the language of Crosby (1986), the Europeans did not plan to build new Europes or homes but came to work on short terms for particular colonial companies or in the colonial administration, hoping to retire back in the metropole.12 The colonizers were not interested in sharing their vernaculars with the whole indigenous populations (Brutt-Griffler 2002, citing from the “Macaulay Minute of 1835” for India). They taught scholastic varieties to an elite class of colonial auxiliaries that would, and did, interface between them and the Natives. The scholastic varieties of their languages – which have evolved into what are now identified as indigenized Englishes, African French, and the like – have been associated with novel and special communicative functions introduced with the colonial regimes, such as formal settings or, more typically, higher education, higher echelons of the administration, and higher strata of the industrial sector, at which the local structure interfaces with the metropolitan and now also with other international structures. The indigenous elite interfaced between, on the one hand, the top level of the European colonial administration and of its economic system and, on the other hand, the lower level, in which almost only the indigenous populations were exclusively involved. The European official language functioned ensured communication with the former group whereas the new indigenous lingua francas now functioning as urban vernaculars permitted communication with the masses of the colonized.

The system was thus set up in such a way that the European languages would not directly endanger the indigenous languages, especially since the domains of usage of the European and indigenous languages were different and complementary. Colonial rule generally produced triglossias in which the indigenous lingua francas and urban vernaculars became associated with modernity. They have functioned as the languages of trade and salaried labor. In the city, they are what children acquire natively, often at the expense of the ethnic languages of their parents. They are also the languages that many adults who migrate from the rural areas adopt as their new vernaculars, especially if the city is not ethnically segregated. It is thus the urban vernaculars, not the colonial Europeans languages, that have competed with the ancestral, ethnic languages and endangered them, more so in the city than in the rural areas. Rural populations have largely kept to their traditional, pre-colonial ways. Many of them have learned the lingua francas only minimally for the occasional contacts that they have with outsiders.

One must also note in this context that most of the elite who are fluent in the European languages either do not speak these as vernaculars or have not given up their indigenous languages. They need them to communicate with their relatives who have not been equally successful or lucky with their formal education and/or economically, just as they need to maintain communication with the masses of their populations in the market place and in other domains of their social and political lives that operate only in the indigenous languages. They stick to their ethnic affiliations too, which the indigenous languages are associated with. They likewise make sure that their children remain fluent at least in the relevant urban vernacular, as it can be a liability to be (totally) uprooted from one’s parents’ cultural heritage. Where mixed marriages have made ethnic identification problematic, competence in the urban vernacular guarantees that one is not identified as uprooted from what can be claimed as indigenous.

As explained in Mufwene (2001), exploitation colonization has modified the linguistic landscape of Africa and Asia essentially in introducing European languages for novel communicative domains and causing the emergence and geographical expansion of new, contact-based, regional lingua francas that have become urban vernaculars. While these have endangered traditional, ethnic indigenous languages, they are not the only threat. Wars along ethnic lines and refugeeism are taking their toll too. In the latter case, the fate of the refugees’ languages depends largely on whether the relevant populations are integrated or not in the host territory, thus whether their children are under pressure to be assimilated within the local populations and find it unnecessary to speak their parents’ ethnic languages. This phenomenon is reminiscent of a factor that must have played an important role in the loss of Native American languages, viz., the relocation of indigenous populations escaping the ills brought by the Old World populations, which, according to Crosby (1992), put them at greater risk than the invaders’ weapons.

It is certainly misguided to attempt to account for all the above evolutions as unidimensionally as in much of the linguistics literature, simply invoking globalization and/or colonization, without ecological specificities, as if population contacts around the world had been uniform. As much as colonization and globalization are implicated in all these cases of the expansion of particular languages at the expense of others, the evolutions are primarily local, depending largely on how the relevant populations in contact interact with each other. It should be informative to find out how much is new in all this. I focus on this question in the next section.


Precursors of modern colonization and globalization

There are well-documented cases of previous colonization that prompt us to speculate on parallelisms with recent and present cases of language endangerment and loss, so that we can make better sense of either the present or the past. We can start with ancient cases that the literature has hardly associated with colonization, viz., the domination of a large part of Europe and of North Africa by Rome, of England by Germanic populations from northwestern Europe, and of Sub-Saharan Africa by the Bantu populations originating in the Nigeria-Cameroon area. Two similarities deserve attention in the context of this volume: 1) the diversification of the prevailing language has been a concomitant of language dis- or replacement (i.e., language death); 2) the evolutions have been differential, calling for ecology-specific explanations for both language diversification and language loss. For instance, it is curious why England and the eastern part of the Roman Empire did not Romanize. The greater association of the Eastern Empire with Greek raises the question of why the Hellenic Empire did not produce a phenomenon similar to Romanization, especially when Greek carried even more prestige than Latin in the Roman Empire.

It is not clear whether the Romans colonized Europe in the exploitation or the settlement style. Trade colonization would apply only to the initial stages, just like in the recent colonization of Africa, Asia, and the Americas by Europeans. That the regime had an element of exploitation style is suggested by the fact that the Romans left their colonies when the area closer to Rome was threatened by their Germanic neighbors in the north. Yet, the fact that elite members of the colonies were not only granted Roman citizenship but were also allowed to become senators and even emperors suggests an important assimilation policy that is not typical of recent exploitation colonies. That retired generals were given land in the colonies, where they settled, is also evidence of some form of settlement colonization. On the other hand, the Roman empire was more of a network of colonial sources of wealth, amounting to specific interconnected towns (Polomé 1983), than a continuous mass of foreign land subjected to the economic, political, and military control of a metropole. Under this perspective, the phrase “all roads lead to Rome” makes a lot of sense, especially if analogized to today’s ramifications of world-wide global economic networks that benefit the powerful members more than the weaker ones.

Latin was introduced initially to the indigenous upper class, which the Romans supported militarily, advised to rule their territories in the Roman style, enticed to govern in the interest of Rome, and rewarded with powerful positions in the Roman political system. Thus, Latin became associated with power. Although interactions between the masses of the indigenous, Celtic populations and the Roman legionaries exposed the former to Vulgar Latin, the latter would not spread among them until after the Romans left the Empire in the 5th century. It is largely thanks to the retention of Latin by the local rulers as their vernacular, as the language of their administration and of the school system that the indigenous populations saw the powerful market value of their colonial language and gradually shifted to it. Just as in recent exploitation colonies where English and French have indigenized, so did Latin as more and more nonnative speakers used it to communicate among themselves. The more it served as a vernacular in the urban environment, the more it became associated with modern life, and the more ground the Celtic languages lost, although the whole process would be protracted over centuries. Usage of Latin in all sectors of the economic system must have been a critical factor, because it made Latin useful to the lives of the masses of the Celtic populations. The fact that the Romance languages evolved from Vulgar Latin, rather from Classical Latin, suggests that the school system must have been no more successful than the boarding schools of Ireland and the United States. Informal and naturalistic language transmission must have played an important role in the spread of Latin and its Pyrrhic victory over the indigenous languages.

We thus do not know whether European languages will have similar evolutions in Africa and Asia, because they are not being used yet in the lower sectors, where the masses of the populations can find jobs. On the other hand, it is not clear why England “Germanicized” instead. The Germanic populations colonized England, apparently in the settlement style, around the same time when the Franks and Vandals invaded the southwestern part of the former Roman Empire. Old English emerged within two centuries, about the same time Old Romance varieties emerged, and  before the colonization of Iberia by the Arabs (from the 7th to the 15th centuries). Subsequent colonization of England by the Danes, Norsemen, and the Normans had no similar linguistic consequences. Similar observations apply when we note that most significant evidence of language Romanization in the eastern Empire is to be found in Romania and in some parts of the Balkans (Friedman 2001), among populations that seem to have emigrated from Romania. It seems like the attitude of the colonized population to the colonizer’s language was a significant factor and that the agency of the indigenous population had a whole lot to do with the abandonment of their languages to death.

Thus, whether or not European languages will endanger indigenous languages of Africa and Asia is an open question. Right now it is the indigenous lingua francas that do; and their effect has been far less extensive than that of European languages in former settlement colonies. One would not even want to speculate on the effect of English, identified precociously by Crystal (1997, 2004) as a “global language,” because in most of the countries where its usage has increased over the past few decades, its strictly lingua-franca function is in competition with none of the indigenous vernaculars. The fact that a language acquires prestige from functioning as a world language does not necessarily situate it in the kind of ecological dynamic that would make it dangerous to indigenous vernaculars. Not all ecologies favor world languages over indigenous ones. I return to this question below.

Though it is only part of the Roman Empire that Romanized, it is, nonetheless, curious that no phenomenon similar to this occurred in the Hellenic Empire that preceded it. Yet, Greek had so much prestige that the cultivated Romans associated themselves with it and either retained it or allowed it to be used as the lingua franca of their administration and economic transactions in their Eastern Empire (e.g., Goodman 1991).13 How different was their colonization style from the Roman? Is there a correlation between the lack of linguistic Hellenization and the fact that Greek continued to be spoken just where Romance languages did not develop? Is it possible that Greek was contained in the elite class, without any association with the development of the local industry, in a way that made it unattractive to the masses of the population? After all, it is Vulgar Latin that prevailed, not Classical Latin, the variety learned by the intellectual and political elite. This development is not so unusual. For instance, we note in modern times that the association of Sanskrit with high culture has not kept it from dying. The development of Irish English seems to be more the result of the introduction of the potato farming and the participation in this new industry of the migrant workers who learned English informally than the consequence of teaching English in boarding schools (Odlin 2003). The term Irish English is associated more with the nonstandard vernacular than with standard Irish English, which is morphosyntactically closer to British standard English.

The fact that Indigenized Englishes have remained primarily lingua francas spoken only by a small elite fraction of the populations of former British exploitation colonies is indeed quite telling. They did not develop through the same kinds of ethnographic transformations that are associated with “native Englishes” of the British Isles, the Americas, and Australia. Indigenized Englishes are more comparable to usage of Latin at the Vatican, with its speakers putting into use knowledge acquired in school. The growing divergence of indigenized Englishes from “native Englishes” is a function of the xenolectal settings in which English is being spoken, settings in which the extent of divergence is a function both of the ethnographic dominance of the substrate languages – which most speakers acquire natively – and of the level of education of the speakers. All these combined histories suggest the resilience and vitality of colloquial and nonstandard varieties, especially when they are associated with economic usefulness to their speakers, compared to the exclusionary connotation of the standard varieties associated with the elite.

And we could also speculate whether their exclusionary connotation (Mazrui & Mazrui 1998) is not one of the reasons why the foreign official languages of former European exploitation colonies do not constitute a danger to the indigenous vernaculars. It can very well be the reason why Classical Latin had little influence on the development of the Romance languages and why Classical Latin had to yield to educated varieties of European national languages in the domain of scholarship. Less exclusionary languages would thus have more vitality than the more exclusionary ones. Although one may invoke some form of globalization – less or more sophisticated, depending on the relevant time in history and the relevant part of the world – a host of other ethnographic ecological factors must be considered to understand the dynamics of language vitality, endangerment, and loss.

There is no particular reason why colonization must always be discussed in relation to European expansion. We can redress this common bias, at least slightly, in directing attention to black Africa, where the Pygmy and Khoisan populations were colonized, assimilated, or and/or displaced by Bantu populations and have lost most of their languages. About 8,000 years ago, the latter dispersed from their homeland in the eastern Nigeria-western Cameroon area. Initially, they dispersed southward along the Atlantic Ocean, eastward on the fringes of the Equatorial forest, and then southward along the Indian Ocean (Vansina 1990, Newman 1995). From these outlying areas of territories where the hunter-gathering Pygmies and Khoisans preceded them, the invading Bantu, mostly agriculturalists, also penetrated the interior. Becoming majorities, they imposed their economic systems and other cultural ways, including their languages, which accounts for the near-total loss of the Pygmies’ languages and the ongoing extinction of the Khoisan vernaculars in places such as Namibia, South Africa, and Botswana.

Technological superiority is not a prerequisite for a culture, including its language, to replace another, especially in an endogenous setting. Changes in socio-economic ecologies will suffice, especially when the populations in contact maintain relatively egalitarian relations. This is precisely what Tosco (1998) reports about language loss among the Elmolo and the Yaaku in Kenya. The Elmolo are a very small fishing population on the southeastern shores of Lake Turkana. They spoke a Cushitic language related to Dhaasanac and Arbore near southwest Ethiopia. In the late 19th century, the drought, famine, and bovine pleuro-pneumonia which hit the Samburu population, their pastoral neighbors, very hard also intensified neighborly contacts between the two groups. The Elmolo were gradually assimilated by the majority and shifted to their language, Maa. The Yaaku, also a Cushitic population, were bee-hive keepers. They had very little contact with the Maasai before the second half of the 19th century. Absorption within the majority population, whose language, Maa, functioned also as a lingua franca with hunter-gatherers of the region, and exogamy led to the loss of the Cushitic language. During the same drought and famine period, it is the pastoralists who depended on the hunter-gatherers for their survival, but it is the latter, demographically much smaller, who gave up their languages.

The situation is somewhat different but just as informative in Botswana, where the Khoisan have been giving up their language even in places where they are the majority. According to Batibo (1998),Yeyi (Khoisan) has been losing grounds to Tswana (a dialect of Setswana) in Ngamiland simply because the Khoisans have been subjected to some sort of serf status relative to the Bantu. What is equally informative about language shift is that in all these cases, English, the official language in both Kenya and Botswana, hardly figures as a contender, despite its political and economic prestige in these countries. Language competition is more local than linguists have shown it to be, and the fact that English does not function as a vernacular has certainly kept it out of the arena in which these indigenous vernaculars have been competing. This supports my position in Mufwene (2003a, 2004) that European colonial languages are generally no threat to the indigenous languages of Africa and Asia. In some cases, the expansion of the European official languages is hampered, instead, by the indigenous urban vernaculars, which are associated with an aspect of modernity that is more tangible, being closer to indigenous cultures.

The evolution may also reveal something else that seems to have been elusive to Western scholars, viz., that non-Europeans, especially in former exploitation colonies, do not really want to be (exactly) like Europeans, though colonization has led them to aspire at new socio-economic values. Invocations of the mass media, for instance, by Crystal (2004) to account for the spread of European languages in Asia and Africa and therefore for the potential threat the spread constitutes to indigenous language reveals ignorance of the fact that in the western world, where they are more accessible, the same mass media have failed to drive out scores of nonstandard varieties that are highly stigmatized. For instance, American White Southern English and African American English have proved to be very resilient. Even Gullah, which is far more stigmatized, has “hung in there” and is probably there to stay for another while (Mufwene 1997). It is endangered more by the exodus of its speakers from its poverty-stricken “habitat” of coastal South Carolina and Georgia than by the media and or the equally-often invoked influx of more affluent mainlanders who speak less stigmatized or more prestigious varieties.

Overall, bilingualism and bidialectalism do not entail language or dialect shift, and the ecologies of language death must be more complex and diversified around the world than Nettle & Romaine (2000), Crystal (2000, 2004), Skutnabb-Kangas (2000), among a host of others, would like us to believe. And lest the whole issue of language endangerment is also trivialized, it helps to distinguish, as does Calvet (2002), between language competition at the national level and competition at the imperial level, though there is no particular reason to assume his gravitational model, reinterpreted pointlessly by Swaan (2001) in terms of constellations.14

The reason why French is losing imperial grounds to English as a world language has to do more with differences in, inter alia, the international economic and military practices of the most important nations associated with these languages (viz., the USA, the United Kingdom, Australia, France, and Canada/Quebec) – as well as their cultural diffusion strategies – than with a more aggressive or assertive introduction of English to non-Anglophone populations. Technological eminence alone is not enough lure. Otherwise, German would have played, and could still be playing, a greater role as a  lingua franca in Europe than French does. Spanish has emerged as a world language more because of the legacy of its settlement colonization in densely populated parts of the world, a large part of Latin America, than because of the economic prestige of Spain.15 France performed poorly in the imperial venture of settlement colonization and would have to include French creole speakers to inflate the number of French speakers.

Overall, speakers of English and French as official or foreign languages in territories other than former settlement colonies are not particularly preoccupied by whether France, the United Kingdom, or the United States is economically or militarily more powerful. They are more concerned by the tangible, mostly local opportunities that their ability to speak one and/or the other language offers them to improve their professional profiles or increase their job prospects. Although many speakers capitalize on their competence in any or all of these languages in deciding one day whether or not they can travel to any of these countries or others, being able to use this competence to such an advantage was not a primary consideration, at least not for the majority of foreign language learners, when they decided to take classes in a particular language. For a large proportion, if not the vast majority, of such speakers, it was simply a matter of what foreign language was accessible or required during their formal education.

More relevant to the imperial competition among the world languages that function as lingua francas is how they were introduced to a particular territory. There is a certain irony here, because La Francophonie has been quite aggressive in developing all sorts of strategies to promote French around the world, but it also looks like the harder they try the more it is evident that French cannot keep up with English. On the other hand – if I may use a phrase favorite to Francophones when they refer collectively to the Britons and Americans – “Anglo-Saxon” companies do not make English a requirement in developing branches outside the UK, the USA, and Australia. Except for the highest management levels of their business ventures, the language of the relevant industry is the local lingua franca (especially at the lowest level of the work structure) or official language.16   Invocations  of  McDonaldization  to  illustrate  the  cultural  (including linguistic) Americanization of the world have typically overlooked the fact that McDonald eateries function in the local lingua franca, if not the local vernacular, wherever they are in the world. In keeping with the globalized business ambition of opening or keeping up the widest markets for one’s products, Hollywood movies are now released in several world languages, so that only countries that either prefer them in the original language or do not have enough economic clout receive them in English. In either case, the economic principle of supply and demand applies. Major, multinational corporations that shape regional world-wide economic globalization do not make themselves missionaries of the world languages, because they believe in using the most profitable strategies for the most benefits. Very often this approach entails using the language of the local workforce and the local market, except in (most) senior managerial positions.

Though the corporations do provide opportunities through the basic fact of representing particular nations where they have developed branches, it is the aspirations of local (potential) employees that create the market for English (or any other language of power), which then takes a free piggy-back ride. Hopes to rise to high or top levels of the firms’ managements incite the locals to learn English (or any other relevant language). Still, we must remember that this language operates in domains where it competes not with the indigenous vernaculars or lingua francas but with other imperial languages. Hence, however “global” English becomes (in the way Crystal 1997, 2004 defines it or how Phillipson 2003 uses the term), it endangers only other European imperial languages with which it competes in the prestigious and potentially lucrative ethnographic domains allocated to them.17

The rapid and insidious spread of English thus owes a lot to the success of the “Anglo- Saxon” (mostly American) business and military interventions around the world than to the aggressive language-teaching policy underscored by Phillipson (2003). If language- teaching could do so alone, the vitality of Irish would have been restored by now. Unlike in the case of French in Quebec, the neglect of Irish from the local/national economic sector, which continues to function in English, has made the commitment of the Irish government and the school system significant only for the maintenance of knowledge not put to practice. Yet, the vitality of the language depends on those day-to-day practices in which speakers are enticed by the interactional conditions to speak a particular language rather than an alternative. The same explanation also accounts for why European colonial languages have not recruited more speakers than they could have in former exploitation colonies. They have been restricted to domains that do not involve, or exclude, the majority of speakers. Except to purists, languages do not need schools to spread nor to remain viable. Linguistics based on western practices has misled us into this illusion.


Will there be an English-only Europe?

Invocations of language practice in institutions such as the European Union and the United Nations to prove either the displacement of other languages by English or mere competition among major languages oversimplify the complexity of language endangerment issues. They ignore the fact that not all cases of individual multilingualism or executive monolingualism are dangerous to the lives of the less prestigious languages. Highly stigmatized minority varieties such as Gullah, African American vernacular English, and Old Amish English in the United States tell us that such languages can be quite resilient. Recent studies such as Swaan (2002) and Phillipson (2003) confuse usage of lingua francas in specialized domains bearing little on communication in vernaculars with competition among vernaculars in domains most likely to affect their vitality. Increased usage of English in the offices of the United Nations in New York and Geneva, or in the offices of the European Union in Brussels and Strassburg, do not in any way jeopardize the naturalistic transmission of French as a vernacular from generation to generation in France, in Francophone Belgium and Switzerland, or in Quebec.

From an ethnographic perspective, a distinction must be made between language use in a professional domain and vernacular language practice. Societally multilingual settings have naturally produced countless individuals, such as the white-collar elite in Africa and Asia, who routinely use languages that are not their vernaculars in their professional lives but continue to speak their ancestral languages in vernacular settings. The prestige associated with their professional languages has not at all deterred them from “transmitting” their indigenous languages (ethnic or urban vernacular) as mother tongues to their offspring. On the other hand, the recognition of Breton in the European Union as an ethnic or minority language will not necessarily revitalize it unless changes in language practice are adopted in Bretagne, where it faces the competition of French as a vernacular.

Competition between English and French as imperial languages affects their ethnographic function as lingua francas, not their function as vernaculars. It is not clear to me how French would be endangered by the fact that Africans in Francophone Africa are increasingly interested in learning English as a foreign language, especially when the vitality of French in France and in Quebec is not negatively affected by the acquisition of English as a second language by French citizens and by French Québécois. It is also debatable whether even the fact that English is used as a medium in higher education in Holland and Denmark is endangering Dutch and Danish. Though it is true that one can visit these countries without feeling the pressure to learn to say hello in the local vernacular, it is not clear that the more important role that English plays in their higher education will produce results more adverse to indigenous languages (Dutch and Danish) than the formal education of Africans and Asians in European languages. These have not been able to displace the indigenous vernaculars.

Monolingualism in most European countries is only indirectly a consequence of policies promoting one language per nation. It is the direct effect of economic systems that have operated only in some dialect of the official language, to the exclusion of all other language varieties. It is this utilitarian aspect of language practice, through the potato plantation rather than through the boarding school, that spread English in Ireland. It is also what spread English in America, making it more and more difficult for immigrants from continental Europe to sustain parochial education in their respective national languages. It is the same practice that has endangered Native American languages, not the Jesuit fathers’ Irish-style boarding school system. Africa has not had a similar experience with European languages because these have typically functioned as lingua francas for a small elite that is a legacy of the colonial auxiliary system and because the industry that would promote the spread of the European languages is very small. English is not about to endanger Dutch and Danish as long as the latter remain the vernaculars, the primary languages of the local governments, and the primary languages of the industry targeted by the vast majority of the populations, even if Dutch and Danish are losing grounds to English in some domains of higher education.

The claim that the spread of English as a world lingua franca is endangering the vitality of indigenous languages outside the United Kingdom, the United States, Australia, and New Zealand seems to ignore the naturalness of egalitarian multilingualism as a way of life in many parts of the world. Worse of all, it ignores the fact that in many of the former exploitation colonies where English is spreading, this imperial language, like the other European world languages it is endangering, is spoken only by no more than 20% of the local population. The remaining 80% of the local populations who are practically disenfranchised by the world-wide global economy are not at all affected by whether or not English or French is prevailing as a/the foreign language that interfaces their nations with the rest of the world.

The relevant literature has also forgotten some other basic things, including the following. Africans and Asians still stick to many of their traditions, despite the non- negligible cultural influence they have received from the West. Even the elite that speak European languages, as lingua francas and of course non-natively, still claim that indigenous identity, which has made them loyal to the indigenous languages. Indigenous lingua francas such as Lingala, Swahili, Wolof, Hausa, and Town Bemba, which are locally associated with urban life and modernity, are appealing to rural populations, whereas European languages are still treated as foreign and often also with distrust, especially by those with little or no formal education. The association of these European languages with the power of abusive and exploitative post-colonial governments has not at all helped their image to the masses of the disenfranchised populations, even among those who learned enough to make themselves understood in some “broken” form. Besides, the high rate of illiteracy in economic systems that are far from recovering from either stagnation or deterioration also constitutes a major obstacle to the spread of any language that relies on the school system for its transmission. Invocations of globalization by Crystal (20004) and other linguists to forecast the obsolescence and replacement of indigenous languages around the world by English have simply been too unidimensional and oversimplistic to help us understand the complexity of competition and selection in the coexistence of languages.


The economics of language maintenance vs. shift

As noted above, linguists’ interest in language maintenance, often embarrassingly confused with language preservation,18 has been spurred largely by environmentalists’ concern with the preservation of biodiversity in the world’s ecologies. Advocacy on the subject matter has increased since, among other concerns, the deforestation of rain forests and the discovery that the relevant vegetation contained medicinal plants that can contribute significantly to modern pharmaceutics. The fact that the industrial exploitation of such medicinal plants starts with folk knowledge of the indigenous populations, which is most accessible in their own heritage languages, has revalorized the Sapir-Whorf hypothesis. According to this, every language structures the world, more accurately its speakers’ universe of experience, in its own peculiar way, sometimes storing knowledge that is foreign to speakers of other languages. This particular benefit has prompted language rights advocates to speak louder for the maintenance or preservation of indigenous languages in various former colonies. The advocates have argued that through the loss of their languages the affected populations are also losing cultural knowledge that had helped them survive in their traditional ecologies, an experience that endangers the future of the mankind in the same way that deforestation is. All this can be lumped together with numerous other cases of ecological mismanagement, which endanger the future of our planet, including our own as humans.

A global examination of the whole situation prompts several questions, the first of which is whether language advocates have a global understanding of the complexity of the issues. Are the language advocates and environmentalists approaching the subject matter of endangerment with the same or similar concerns?19

Two concerns have been the most prominent among language rights advocates: 1) as noted above, loss of languages entails loss of traditional knowledge that can be helpful to mankind (e.g. Nettle & Romaine 2000; Crystal 2000, 2004; Mafi (2001); and 2) loss of languages entails loss of linguistic diversity, which would be harmful to linguistics, because linguists would have fewer data for a more accurate understanding of the nature of typological variation and language universals (Krauss 1992, Hale 1992).

The second argument is a more vexing one to me but perhaps also the easiest to address. Bearing in mind that it is the indigenous populations themselves that gave up their own languages (understandably under new socio-economic pressures), it is not clear under what specific conditions they could have maintained them as vernaculars. In other words, what are the alternative ecologies that would have favored their commitment to their indigenous vernaculars? Likewise, because the affected populations’ life styles had changed or were changing, along with their socio-economic ecologies. Therefore it is not clear how useful the knowledge that is embodied in the endangered or lost languages would have been to their speakers , except in their memories of the past, because it could no longer apply. Since languages also serve markers of identity and can be used against its speakers, what’s the point of maintaining diversity if it appears to be adverse to one’s adaptation to the  new socio-economic ecology?  Aren’t linguists focusing  on an epiphenomenon and simply being paternalistic without making an effort to understand what has led speakers to give up their languages?

Regarding the loss of linguistic diversity – which the language rights advocates have made more useful to linguists than to mankind – genetic linguistics informs us that it would be too simplistic to assume that languages have always evolved in the direction of loss. The different xenolectal varieties that evolved out of the Celts’ gradual shift from their indigenous languages to Vulgar Latin have now become separate languages with their own numerous dialects. It is not clear what the future of creole and indigenized varieties of European languages will be, as they continue to diverge, contrary to the misguided literature on decreolization (Mufwene 1994). While creoles have already been disfranchised by linguists  as separate languages  (against the sentiments of their speakers), nothing precludes indigenized Englishes and the like from being identified some day as separate languages.20

As much as language rights advocates decry (rather inaccurately) the Anglicization of the world, it is also obvious that English is concurrently speciating into diverse varieties that may some day no longer be recognized as the same language. Besides, it is not likely that globalization will drive Chinese varieties and Japanese, among many other major or powerful vernaculars to extinction, at least not in the near future. So, as in the past, linguistic diversity will always be self-regenerating, although the patterns of typological variation will hardly always be the same – and why should they? To the extent that in affecting each other – even in causing some other populations to lose their languages – human populations are enacting natural processes of evolution, would it not be equally informative to try to understand the evolutionary mechanisms and conditioning factors that are involved? And wouldn’t the next step be that of finding out whether there are victim populations that are maladaptive, in the sense of not doing the right things in their struggle to survive in their changing socio-economic ecologies? To what extent are the solutions now proposed by language rights advocates really beneficial to populations whose languages are endangered? How can these solutions help the victims stop the endangerment when they have little, or no, control of the socio-economic changes that affect them and their languages?

Language rights advocates have dodged the above questions. To be sure, they have also advocated multilingualism but have not articulated the ecological conditions under which this practice is sustainable. Could it be maintained as easily in former European settlement colonies (where the European ideology of one language per nation has prevailed at the expense of ethnic identity since the late 18th century) as in Africa ( where the vitality of ethnic identity, which is primarily associated with language, has made difficult the implementation of this European ideology)? Can most of the indigenous languages be maintained without changing the current socio-economic world order among both the victims and those who control it? The answer to this latter question is obviously negative. The embarrassment is that language rights advocates have given little thought to the revolution that is entailed by their discourse. They have provided no answer to the implicit question of what alternative socio-economic world order must be recommended to the victims to meet their new material and spiritual aspirations, which depend in part on languages of the workforce.

Equally embarrassing is the fact that language rights advocates keep preaching to the victims without sensitizing the victimizers, those who run the socio-economic and political machineries that affect population structures. This machinery is not necessarily the government, which is also one of its products. It is true that funding has been raised to document and analyze endangered languages. In practical terms, one must ask whether linguists or would-be speakers of the relevant languages are not the ones benefitting (the most) from the exercise. If one had to rate the socio-economic needs of the relevant populations in a world that is admittedly materialistic – nonetheless one that has shaped their aspirations – the question is whether funding to linguists earning a living and building a career on the preservation of an endangered languages is helping create more jobs and improve living conditions for the relevant populations. Are linguists helping the humanity or helping linguistics? In what ways is the fact of documenting these languages as important as, if not more than, teaching literacy to several children in the language of the workforce so that, everything being equal, they may be more competitive in their socio-economic ecologies? The phrase “everything being equal” is quite critical here, and the question is whether what it so obviously refers to is not what should call for more advocacy. (See Brutt-Griffler 2004 for a similar argument.)

Regarding losing heritage knowledge concurrently with the loss of languages, even Edward Sapir, who has been associated with the linguistic relativity hypothesis, has made it clear that language and culture are not wedded like two sides of a coin. One can change without engaging the other in the change (Sapir 1921). The facts presented above about deforestation and language endangerment also suggest that the only knowledge that a language would preserve about a destroyed ecology would be archival, available to experts. That knowledge would not be any more helpful to the average speaker than knowledge of medieval English cultural practices is to an average speaker of modern English. On the other hand, the preservation of the ecology and of the cultural practices it has traditionally sustained would entail transferring the relevant conceptual distinctions to the language that came to replace the indigenous one. This is made obvious particularly by the indigenization of European languages in former European exploitation colonies. Even in settlement colonies, where the impact of European languages has been generally catastrophic, these have been enriched with indigenous terminologies in domains that have been assimilated into European cultures. European cultures and languages of the Americas and Australia are new phenomena influenced by the local physical and cultural ecologies, despite the heavy heritage from Europe. The most annoying part is perhaps the static notion of culture that the language rights advocacy literature has suggested, as if traditional culture had to survive in its pristine form and as if European cultures were unaffected by the cultures of the populations that adopted them.

Another aspect of the economics of language shift is what opposes linguists to economists. The latter have found the multiplicity of languages to be inhibitive to economic development. For instance, Alesina et al. (2003) argue that, although there are exceptions, societally or regionally multilingual nations have a higher ethnicity index than monolingual ones.21 Countries in the latter category, especially those of the Third World, tend to have too many conflicting ethnic interests, which inhibit the implementation of efficient national economic development programs. One can of course argue that poor political leadership – rather than multilingualism or multiethnicity – is the real source of the problem. However, limited financial resources and socio-economic systems often inherited from poor colonial rules that were insensitive to local traditions have made it almost impossible for uncorrupted political regimes to evolve that would not make ethnicity and the ensuing multilingualism non-issues. Post-colonial governments committed to solving their development problems have thus shown a strong interest in the European ideology of one language per nation, especially under the pressure of western lending institutions which stipulate unilaterally the terms of their loans.

Ironically, most of the multiethnic nations concerned by this discussion are former exploitation colonies, precisely where there indigenous languages are generally less endangered, least at all by European languages. Whether or not the economists’ advice is sound, it has the merit of treating languages as factors that affect human economic development, therefore making the adaptability of people to the current economic world order central to the debate. To be sure, language rights advocates have always claimed to protect the interests of the speakers of endangered languages. However, they have hardly explained how their solutions should help the relevant populations adapt (more) competitively to their ever changing socio-economic ecologies.

While it is clear that past socio-economic world orders in which the indigenous languages used to thrive cannot be reconstructed, especially in former settlement colonies, language rights advocates have not articulated the alternative ecologies in which the endangered languages can be revitalized. Nor have they articulated, as noted above, in what ways competence in the endangered languages would make them competitive with those speaking languages to which they have been shifting. Because socio-economic integration has been an important factor in the process of language loss, one must wonder whether the advocates are not promoting an ethnolinguistic segregation that can be disadvantageous to the affected populations.

To be fair, what linguists have advocated is bilingualism that should enable the affected populations to remain loyal to their ethnolinguistic traditions while being fluent in the language of the workforce. My question is whether this is still possible when the ambient socio-economic ecologies have typically led to the disintegration of the affected communities, which have become minorities in their own homelands and whose members have increasingly been led to abandon them to join the majority populations. Noticing that a language is endangered is largely a retrospective outlook on evolution, particularly because language loss is the cumulative result of individual decisions that have been made independent of each other. Few individuals if any have the foresight of anticipating that their communicative practice will amount to population-wide language shift and loss. It is not clear whether such individuals react with guilt to the cumulative outcomes of their communicative acts when asked to think retrospectively about them, especially why they did not “transmit” them to their offspring. Chances are that in the latter case the conditions were not favorable to, or supportive of, the “transmission” of languages other than the non-indigenous vernaculars. In this case, it is also worth asking whether indigenous populations in former settlement colonies have shifted languages under conditions different from some of the immigrants, such the Italians and Germans in North America.

Would the relevant Natives want to go back to what is left of their isolated traditional communities? Perhaps language rights advocates should find out why most of them have not. Yet, it is likely that those left behind in the traditional communities who have given up their indigenous languages are following the example of those who have moved out. Even if the latter come back to visit every now and then, their competence in the heritage language may be in such a bad shape that they would prefer not to speak it poorly, or they may prefer to show their success by speaking the vernacular of the majority community. Language practices are easier to copy from speakers one can associate with than from others. Could language attrition in the affected communities reflect evolution toward a solution of the lesser evil? If so, then language rights advocates, who typically happen to be of European descent, should really explain why they advocate imposing on indigenous populations solutions that they would not impose on descendants of European immigrants, most of whom are indeed monolingual in a language other than those of their ancestors.

Overall, things boil down to the following: people react adaptively to changes in their socio-economic ecologies. Some of these involve giving up their languages, and sometimes their cultures, for those of the prevailing group while also preserving their identities through the ways they adapt the new languages and cultures to their communicative needs. From the point of view of culture, this is a process of indigenization that winds up producing a new and hybrid culture that continues to define their ever-adaptive identities. Although it is useful to document the past, it seems equally important to ask whether from an evolutionary perspective and from the point of view of populations which define their own identities language rights advocates should bemoan normal adjustments that people continue to make to changing ecologies. They should ask themselves whether they are approaching the question of language endangerment in the same ways, or with same objectivity, as environmentalists are addressing that of species endangerment.



Language endangerment is certainly not a recent process, not any more than globalization is. Globalization itself cannot be dissociated from colonization as a process whereby one population not only relocates to some other place but also dominates the latter’s indigenous population and imposes on the latter its own style of economic and political organization. The recent form of globalization differs from earlier ones by the scope of the world involved and by the speed of communication and transportation between “nodes” in its networks of interdependencies, as well as by the complexity of its organization. Globalization need not be of world-wide scope, and in fact most of its forms are local and regional, amounting simply to complex networks of interdependencies. This is in fact its basic form in Western Europe, in its former settlement colonies, and in Japan. It has developed the most in those nations that have espoused the ideology of one language per nation, a language that in its standard and vernacular varieties has emerged as the language of the industry and of the workforce. These are countries whose dominant populations have adopted largely monolingual practices and have marginalized other languages and/or reduced them to the status of minority ones.22 These are also countries where socio-economic pressures have most successfully driven speakers of the minority or marginalized languages to shift vernaculars in favor of the dominant one. Some of these territories, such as Europe and Japan are ironically among the most densely populated places in the world but have the fewest languages. Europe hosts only 3% of the world’s languages (Mayor & Bindé 2001).

It is not evident that modern local or regional globalization accounts for all countries with tendencies toward societal monolingualism. Settlement colonization history and the geographical size of the nation state can account for societal vernacular monolingualism in countries such as Burundi, Rwanda, Botswana, Thailand, and Korea.23 A long succession of imperial regimes may account for the situation in China, although it remains an open question whether several dialects of one single Chinese language or a cluster of Chinese languages are spoken in China. World-wide globalization since the late 1980s may be considered as a new form of exploitation colonization, one in which multinational corporations headquartered mostly in the western world, rather than specific nations, are colonizing developing nations without ruling them politically. While European settlement and exploitation colonizations since the 15th century had expanded the Indo-European dispersal and spread major European languages around the world, the latest form of  world-wide globalization has spread English beyond the political boundaries set up by the earlier colonial regimes. It is this latter form of economic globalization that has shaken the imperial position of French and has made it obvious that Spanish, Portuguese, and Russian as major languages are not real world languages. The situation is in fact a challenge for linguists to articulate more clearly the distinction between “major” and “world” languages. Although they may have more speakers world- wide, major languages such as Mandarin, Cantonese, and Hindi have traditionally been spoken in one geographical area, are associated with single ethnic populations, and continue to function as vernaculars even in their diasporas. On the other hand, world languages are spoken at various places around the world, by ethnically diverse populations, a large proportion of whom use them only as lingua francas. As a matter of fact the majority of English and French users are non-native speakers, who are incidentally the major agents of the spread of these languages, a phenomenon that is reminiscent of the way Latin was spread in the European Romance countries (France, Portugal, Spain, part of Switzerland, and Romania), originally by its legionaries and during the post-Roman period by the local elite (Mufwene 2004).

The recent rapid spread of English around the world is thus comparable to that of major European languages during the exploitation colonization of Africa, Asia, and the Pacific since the 19th century. It is spreading among the same kind of intellectual and/or political-economic elite that had not only appropriated and indigenized European languages as their lingua francas but also maintained their indigenous languages as their vernaculars. Thus, while English has shaken the position of French and other colonial European languages in these former exploitation colonies, it is far from endangering the indigenous ones. English is not really the “killer language” that non-global approaches to language endangerment have painted it to be, certainly not in relation to the indigenous languages of former exploitation colonies. Moreover, it does not have the kind of agency that would have made it a killer, because it is (would-be) speakers who kill their languages by opting to speak another language, which amounts to what is known as language shift.

In the case of sub-Saharan Africa, for instance, it is important to take account of the expansion of the indigenous lingua francas that also function as urban vernaculars and are spreading at the expense of traditional ethnic languages and apparently also of the colonial European languages which continue to function as official languages.24 Non- negligible is also the still-ongoing effect of indigenous population contacts that preceded European colonization, such as those of the Bantu with the Pygmy and Khoisan populations, at the expense of the languages of the latter groups. History also reminds us of other settlement contacts between Bantu and other non-Bantu populations, such as in the northern part of DRC and in Rwanda and Burundi. The latter case raises, for instance, the interesting question of why and how the Tutsi in both countries have “Bantuized” linguistically whereas the Maasai (also Nilotic) in Kenya have not. Recent history marked by genocidal wars and refugee resettlements have also produced new intimate contacts between diverse ethnolinguistic populations. Time will tell what the outcomes of these new population contacts will be.

There are also many places where small Bantu populations are found surrounded by larger Bantu populations speaking different languages. Such enclave populations have nonetheless maintained their languages as markers of their separate ethnic identities. Linguistics ought to articulate the specific conditions under which this has remained possible, especially in a history marred by periodic interethnic wars, even though some of these may not have been local. All the above considerations suggest that a global approach capable of discriminating a wide range of variation in how language competition and selection have proceeded around the world is what linguistics needs to understand language endangerment. What is true of Europe and its former settlement colonies is not necessarily true of the rest of the world. While it is certainly regrettable that some languages are vanishing, it is undeniably evident that the socio-economic ecologies in which speakers have evolved have changed to points of no return. We cannot address the question of language loss without concurrently addressing that of costs and benefits to would-be-speakers of the lost and endangered languages. A problem as “wicked” and “vexed” as this one cannot be addressed unidimensionally, nor unglobally.



1 Authors such as Dollfus (2001) and Carroué (2202) actually make it clear that French scholars should have used globalisation, the correct translation of globalization, instead of mondialisation as the unmarked term. The reasons are the same as those I give below regarding English.

2 The French term langue minorée reflects a process in which vernaculars have been marginalized even if they are not spoken by minority populations, as in the case of Provençal in the southern part of France.

3 The difference can be explained as follows: In the traditional colonial system, one particular nation had the monopoly of control over a particular territory and of access to its raw materials for the development of the metropolitan industry. In the globalized form of colonization, one company holds privileged relations with another company or many others, and it derives preferential profits from the connection(s). The common colonial feature is that in both cases the exclusionary relation is asymmetrical, favoring the exploitation of the weaker by the stronger. For a succinct and accessible history of how colonialism and regional globalization have evolved in various civilizations since the dawn of agriculture, see Cowen (2001).

4 In this case too, one can point out the kinds of asymmetries that have become all too common in today’s global world system. However, I argue below that not all such power asymmetries lead to the endangerment of indigenous languages by western European world languages.

5 I use the term language evolution here as in Mufwene (2001) to cover not only changes in the structures of a language but also those cases where a language dies or splits into different varieties, such as the speciation of Latin into the Romance languages. The case of speciation should be extended to creoles, although linguists have typically treated them as unusual developments.

6 This is basically also the position developed by Cowen (2001), who interprets colonization as a quest for more resources for subsistence and sees in the history of the world “patterns of exchange and exploitation, patterns of conquest and military rule, and patterns of military conviction and endeavor” (9). All this is also correlated with “tension between movement and settlement,” from a perspective that shows globalization as a process or condition that has always been inherent in the history of mankind but has consistently increased in complexity, speed of expansion, and the size of the territories involved.

7 To be sure, these languages, which now fall in the category known as expanded pidgins should be associated with the exploitation colonization of the relevant territories. The rudimentary pidgins from which they have evolved are no longer spoken. Nor is the Portuguese pidgin that, according to Huber (1999), used to function as the language of the Euro-African trade on the coast of Africa before, or up to, the early 19th century.

8 Just like in the European metropoles, citizenship in the settlement colonies is typically also associated with some expected ability to speak the official or dominant language. This factor accounts for why not only some European immigrants but also the indigenous populations have felt the pressure to shift to the official or dominant languages of the European-styled new nations in the Americas, Australia, and New Zealand.

9 There are several reasons for both the endangerment and the loss of these languages, including, in the early stages of colonization, the relocation of indigenous groups as they escaped enslavement and the illnesses that the Europeans brought from the Old World. Since the 19th century, the gradual integration of the Natives in the new, mainstream socio-economic world order, which requires mastery of the relevant European language of the new polity, has probably been the most important factor. Urbanization and schooling, associated with monolingualism in the European language, are additional factors. Note that the boarding schools, in which Jesuit priests subjected their indigenous students to humiliating punishments if they were caught speaking their native languages, had no significant effect, not any more than they did in Africa or Ireland, because the students could not speak the European language with their uneducated relatives once they came back home.

10 The history of plantation settlement colonies makes implausible the myth that the slaves gave up their languages because their masters forbade their usage. The masters could not have controlled what the slaves spoke and taught their children in private. Another myth not to entertain is that the slaves’ languages died as soon as their creoles developed. Up to the time of Emancipation, there always were among the bozal slaves those who were fortunate enough to meet somebody who spoke their language (native or not), especially at times when some planters preferred slaves from the same region. The main question if whether their children were interested in speaking their parents’ languages, even in addition to Creole. The outcomes suggest an evolution similar to what can be observed in African cities, in which children typically select as their vernacular the urban vernacular or lingua franca over their parents’ ethnic languages. It is the cumulation of such selections by individual speakers that has driven indigenous ethnic languages out of African cities, at least among the young. It is apparently the same practice that led to the demise of African and East Indian languages in former plantation settlement colonies, even when the post-Emancipation contract laborers came in groups that were relatively homogeneous ethnolinguistically and lived in their own segregated camps. An important partial exception is Hawaii, which does not have the same peopling and population contact history as the American-Caribbean region and the Indian Ocean.

11 Since about the 18th century, Europe has generally subscribed to a nation-state philosophy which promotes unity around the same national language that must be spoken by all as a vernacular. Few of its minority languages have survived. Regarding the impact of this policy on language diversity see passing comments by Hamel (2003:136).


12 I will ignore here cases such as in Zimbabwe, where a settler class emerged and segregated itself, with a host of socio-economic privileges, from the indigenous populations. The case of South Africa must also be treated as a case of dual colonization style, that of the Afrikaners, on the settlement model, and that of the British rule, on the exploitation model. Needless to say that the two regimes clashed, which accounts for many of the social woes of South Africa since the institutionalization of the Apartheid regime in 1948.

13 For those who assume that “global languages” necessarily endanger the indigenous ones in the territories where they have been adopted, it is worth noting that Greek did not endanger Latin even in the Eastern Roman Empire, where the Roman elite spoke it as a lingua franca. Latin died in the Western Empire through its own geographical expansion and transformation, whereas in the East, as in England and North Africa, it simply did not have a chance to evolve into a vernacular, “global” though it must have been as a language.

14 Both authors see languages of the world as organized in galaxy-like constellations that permit communication on at least three levels: 1) the vernacular level, at which indigenous ethnic languages function in former exploitation colonies; 2) the regional or national level, which may or may not include the official language, at which a lingua franca makes it possible for speakers of the various ethnic vernaculars to communicate with each other; 3) national or international level, at which a former colonial/imperial language functions as a supra-lingua franca. Perhaps in the original spirit of language wars, different macro-constellations, similar to la Francophonie, would work as groups to face the competition of other constellations. The idea would make sense if the vitality of the indigenous ethnic vernaculars depended on that of the regional or national lingua francas. As it turns out, the latter may change without affecting the former. Although Kikongo-Kituba is a threat to Kiyansi in the urban environment in the Bandundu province of the Democratic Republic of Congo, (DRC) things would not change much if Lingala became the exclusive vernacular of the city of Bandundu and completely displaced Kikongo-Kituba. Likewise, the communicative importance of Lingala and other regional lingua francas of DRC would not change if English replaced French as the official language. As I will state again below, the dynamics of language practice are very local and nothing in the way of membership in a particular language constellation or “galaxy” crosses the instantaneous decisions that speakers make during their various communicative events. It is the cumulative effect of such decisions that drive language evolution, including language death (Mufwene 2001).

15 There is a misguided tradition that confuses the notions of “world language” and “major language”. For instance, Chinese varieties and Hindi are major languages confined primarily to where they originated, despite the fact that they are also spoken by Chinese and Hindi populations in the relevant diasporas. However, they are not world languages, because they are hardly targeted as lingua francas by non-Chinese and non-Hindi populations outside China and India. In this respect, Spanish is not as much of a world language as French is, although it has more native speakers than the latter does. The vast majority of Spanish speakers are those native ones, from Spain and its former settlement colonies in Latin America. The situation is just the opposite with English and French, most of whose speakers are non-native. See also Phillipson (2003) for a similar perspective.

16 Phillipson (2003) must refer to this same management level when he highlights asymmetries in the ways American and European firms function abroad, with the former being less accommodating to the local official language (especially in the USA) than the latter, owing largely to the monolingualism of American CEOs. Accurate as his observation certainly is, this is not the level of population contact that affects the vitality of a language in a polity. The behavior of American senior managers is comparable to what Phillipson reports of senior administrators in the European Union,

in contrast with the individual multilingualism of mid-level bureaucrats. I am sure lower-ranked employees can function monolingually in the official language of the city where the office is located.

17 The Agence Intergouvernementale de la Francophonie is fully aware of this. Several meetings were held, at the expert council and at the head of state levels to debate ways in which the spread of English must be faced in Francophone countries. Calvet (2002) is largely a response to some of the absurd strategies of La Francophonie that we need not discuss here. Many of them stem from not knowing what socio-economic ecologies are (dis)favorable to particular languages and not willing to distinguish the vernacular from the lingua franca functions of an imperial language.

18 Language maintenance has to do with the activities of a population that help them keep their language alive. It is related to language revitalization, which literally means “bring life back” to a language that may die without the relevant intervention. Language preservation amounts to no more than documenting a language through writing texts and/or grammars and through compiling dictionaries, interventions that do not necessarily get its would-be speakers to use it. The closest analog I can think of is suggested by the term preservation itself which is reminiscent of preserving lifeless organisms, plants, or body parts in a jar – i.e., the preservation of form without life, which is good for museums.

19 It may be informative to clarify here that linguists have approached the issues without as much scholarship as environmentalists. The latter are generally biologists well trained in macroecology, in the dynamics of the coexistence of species, and in the ways that biota balance themselves out in ecosystems. They are usually well informed in biotic disequilibria that follow from mismanagements of the physical ecologies, such as through urban and industrial developments. Zealous though concerns with non-human species sometimes appear to be, the future of humans typically emerges at the center of controversies. The question can often be recast in terms of (ultimate) costs and benefits to mankind, if not to the whole ecosystem, in developing a project one way rather than another. To date, linguistics has had no specialization area comparable to macroecology. We have a vague notion of what constitutes the ecology of a language, how it affects language competition and selection – in spite of vague invocations of globalization and/or colonization – and what are the conditions that are most (dis)favorable to language.

20 A by product of linguists’ stand on the ethnographic status of creoles is that, in Jamaica, for instance, tourism and other government agencies have been sending conflicting messages to outsiders, claiming in some publications that everybody in Jamaica speaks English and in others that Jamaican Creole or Patois is a separate language (Irvine 2004). Most Patois speakers assume they speak an English dialect and are shocked when a visitor that claims to speak English does not understand them. As Irvine also shows, the distinction made by linguists between creole and non-creole features also reflect a-prioristic decisions and linguistic norms external to the relevant polity.

21 Monolingual countries that challenge the generalization include Rwanda and Burundi, which have maintained ethnic divisions and are not more developed than their multilingual neighboring countries. Multilingual countries that also challenge the rule include Belgium and Switzerland.

22 French sociolinguistics has another adequate term for this ethnographic condition, viz., langue minorée. When applied to the language of an economically powerless majority population, it captures the ironic fact that the majority is treated as a minority and the economically powerful but small elite of the overall population acts as the dominant group.

23   A combination of these reasons may account for the situation in Japan.

24 In Kinshasa, for instance, more and more members of the government who have migrated from non- Lingalaphone parts of the DRC have found it imperative to develop fluent competence in the local vernacular in order not to be treated as outsiders. This is a good incentive for letting their own children develop native competence in Lingala (very often at the expense of French), which entitles them to claim cultural roots in the capital.



Alesina, Alberto/Arnaud Devleeschauwer/William Easterly/Sergio Kurlat/Romain Wacziarg. (2003). “Fractionalization”, in: Journal of Economic Growth 8: 155-194.

Batibo, Herman M. (1998). “The fate of the Khoisan languages of Botswana”, in: M. Brenzinger (ed.). Endangered languages in Africa. Cologne: Rüdiger Köppe Verlag, pp. 267-84.

Brutt-Griffler, Janina. (2002). World English: A Study of its Development. Clevedon: Multilingual Matters.

—. (2004). “The sound of retreat: The linguistic imperialist camp in disarray”, in:

Journal of Language, Identity, and Education 3: 134-139.

Calvet, Louis-Jean. (1998). Language Wars and Linguistic Politics. Oxford: Oxford University Press.

—. (2002). Le marché aux langues et les effets linguistiques de la mondialisation. Paris: Plon.

Carroué, Laurent. (2002). Géographie de la mondialisation. Paris: Armand Colin. Cowen, Noel. (2001). Global History: A Short Overview. Cambridge, UK: Polity Press.

Crosby, Alfred W. (1986). Ecological Imperialism: The Biological Expansion of Europe, 900-1900. Cambridge: Cambridge University Press.

—. (1992). “Ills”, in: Alan L. Karras/J.R. McNeill (eds.). Atlantic American Societies: From Columbus through Abolition 1492-1888. London: Routledge, pp. 19-39.

Crystal, David. (1997). English as a Global Language. Cambridge: Cambridge University Press.

—. (2000). Language death. Cambridge: Cambridge University Press.

—. (2004). The Language Revolution. Cambridge, UK: Polity Press.

Dalby, Andrew. (2002). Language in Danger. New York: Columbia University Press. Dollfus, Olivier. (2001). La mondialisation. Paris. Presses des Sciences Po.

Fenton, Steve. (2003). Ethnicity. Cambridge, UK: Polity Press.

Friedman, Victor A. (2001). “The Vlah minority in Macedonia: language, identity, and standardization”, in: Juhani Nuoluoto/Marti Leiwo/Jussi Halla-aho (eds.). Selected Papers in Slavic, Balkan, and Balkan Studies (Slavica Helsingiensa 21). Helsinki: University of Helsinki, pp. 26-50.

Goodman, Martin. (1991). “Babatha’s Story”, in: Journal of Roman Studies 81: 69-175.

Guglielmo, Jennifer/Salvatore Salerno (eds.). (2003). Are Italians White: How Race is Made in America. New York: Routledge.

Hagège, Claude. (2000). Halte à la mort des langues. Paris: Editions Odile Jacob. Hale, Ken. (1992). “On endangered languages and the safeguarding of diversity”, in:

Language 68: 1-3.

Hamel, Rainer Enrique. (2003). “Regional blocs as a barrier against English ? The language policy of Mercusor in South America”, in: Jacques Maurais/Michael A. Morris (eds.). Languages in a Globalising World. Cambridge: Cambridge University Press, pp. 111-142.

Huber, Magnus. (1999). “Atlantic creoles and the Lower Guinea Coast: A case against Afrogenesis”, in: Huber/Parkvall (eds.), pp. 81-110.

Irvine, Alison. (2004). “A good command of the English language: Phonological variation in the Jamaican acrolect”, in: Journal of Pidgin and Creole Languages 19: 41- 76.

Krauss, Michael. (1992). “The world’s languages in crisis”, in: Language 68: 4-10.

Maffi, Luisa (ed.) (2001). On Biocultural Diversity: Linking Language, Knowledge, and the Environment. Washington, DC: Smithsonian Institution Press.

Mayor, Federico/Jérôme Bindé. (2001). The World Ahead: Our Future in the Making. Paris: UNESCO Publishing; London and New York: Zed Books.

Mazrui, Ali A./Alamin M. Mazrui. (1998). The Power of Babel: Language and Governance in the African Experience. Oxford: James Currey; Nairobi: E.A.E.P.; Kampala: Fountain Publishers; Cape Town: David Philip Publishers; Chicago: University of Chicago Press.

Mufwene, Salikoko S. (1994). “On decreolization: The case of Gullah”, in: Marcyliena Morgan (ed.). Language and the Social Construction of Identity in Creole Situations. Los Angeles: UCLA Center for African-American Studies, pp. 63-99.

—. (1997). “The ecology of Gullah’s survival”, in: American Speech 72: 69-83.

—. (2001). The Ecology of Language Evolution. Cambridge: Cambridge University Press.

—. (2002). “Colonization, globalization, and the plight of “weak” languages: A rejoinder to Nettle & Romaine’s Vanishing voices”, in: Journal of Linguistics 38: 375-395.

—.(2003a). “Colonization, globalization, and the future of languages in the twenty- first century”, in: MOST: Journal of Multicultural Societies 4.



—. (2003b). “Language endangerment: What have pride and prestige got to do with it?” in: Joseph D. Bian/Johanna Destafano/Neil G. Jacobs/Ilse Lehiste (eds.). When Languages Collide: Perspectives on Language Conflict, Language Competition, and Language Coexistence. Columbus: Ohio State University Press, pp. 324-345.

—. (2004). “Language birth and death”, in: Annual Review of Anthropology 33: 201-222.

Nettle, Daniel/Suzanne Romaine. (2000). Vanishing Voices: The Extinction of the World’s Languages. Oxford: Oxford University Press.

Odlin, Terence. (2003). “Language ecology and the Columbian exchange”, in: Joseph D. Bian/Johanna Destafano/Neil G. Jacobs/Ilse Lehiste (eds.). When Languages Collide: Perspectives on Language Conflict, Language Competition, and Language coexistence. Columbus: Ohio State University Press, pp. 71-94.

Phillipson, Robert. (2003). English-only Europe? Challenging Language Policy. London: Routledge.

Pieterse, Jan Nederveen. (2003). Globalization and Culture: Global Mélange. Lanham: Rowan & Littlefield Publishers, Inc.

Polomé, Edgar C. (1983). “The linguistic situation in western provinces of the Roman Empire”, in: Principat 29: 509-553.

Price, Glanville. (1984). The Language of Britain. London: Edward Arnold.

Sapir, Edward. (1921). Language: An Introduction to the Study of Speech. New York: Harcourt, Brace, & World.

Skutnab-Kangas, Tove. (2000). Linguistic Genocide in Education – or World-wide Diversity and Human Rights? Mahwah, NJ: Lawrence Erlbaum.

Steger, Manfred B. (2003). Globalization: A Very Short Introduction. Oxford: Oxford University Press.

Stiglitz, Joseph E. (2002). Globalization and its Discontents. New York: W.W. Norton

& Co.

Swaan, Abram de. (2001). Words of the World: The Global Language System. Cambridge, UK: Polity Press.

Tomlinson, John. (1999). Globalization and Culture. Chicago: University of Chicago Press.

Tosco, Mauro. (1998). “‘People who are not the language they speak’: On language shift without language decay in East Africa”, in: M. Brenzinger (ed.). Endangered Languages in Africa. Cologne: Rüdiger Köppe Verlag, pp. 119-142.

Vansina, Jan. (1990). Paths in the Rainforest: Toward a History of Political Tradition in Equatorial Africa. Madison: University of Wisconsin Press.

Warner-Lewis, Maureen. (1996). Trinidad Yoruba: From Mother Tongue to memory. Tuscaloosa: University of Alabama Press.


No comments yet
Leaving the first to comment on this article.
You need or account to post comment.