Cult & Society/International Perspective

272

This chapter of the book deals with all the other languages of the world, languages that are neither the official language of a state, a region or a province nor important trading languages or intergroup lingua francas. This is an enormous topic because most of the world’s 7000 1 or so languages fall into this category. And most of these can be considered in danger.

Although, at first sight, the contexts of minority languages as dispa- rate as the three case studies in this chapter (the Albanian spoken in Sicily, the language of hunter-gatherers on the north-west frontier of China and the language of subsistence farmers in the Bolivian Altiplano) might appear to have little in common, on closer examination it becomes evident that some of the pressures on them are similar and the speakers’ experiences comparable. The processes of nation building and globalisation discussed in the earlier chapters of this book are at the root of these pressures. Indeed, as we have seen, nationalism is one of the main reasons why small groups come to be conceived as minorities, and nation building is the context that puts them under most pressure to become bilingual or to shift language. As education, modern state administration, modern forms of media and new technologies enter settings where the local language has no written form and only small numbers of speakers, it is easy to understand how difficult maintenance of the local language can become and how strong the pressure to shift entirely. Romaine reminds us of the four key reasons why so many languages are in an endangered position:

Fewer than 4 per cent of the world’s languages have any kind of offi- cial status in the countries where they are spoken ... most languages are unwritten, not recognised officially, restricted to local community

12 Endangered Languages

Endangered Languages 273

and home functions and spoken by very small groups of people. (2002: 1)

Given the external pressures on speakers and the likely choices that they will make, under these pressures, a number of these languages are likely to die out completely. Krauss (1992) suggests that the number to disappear might be in the region of 4000, or two-thirds of the languages currently spoken. Hale (1998) and Woodbury (1998) believe that only 600 out of the 6000–7000 languages are fully secure.

Crystal (2000) poses three questions that arise from this prediction: Should we care? Can anything be done? Should anything be done? To which one could add, if there is something to be done, are scholars and practitioners of LPLP the right people to do it? Can behaviour be influenced through policymaking? This chapter examines some of the reasons advanced for language maintenance and protection, and then reviews some of the issues in three different case studies where there has been policy and planning to promote and protect an endangered language.

12.1 Should we care?

There are several arguments advanced as reasons why humanity as a whole should not let languages disappear. We shall look at the strength of evidence for each one and then move to what can be done if we accept the arguments. The arguments can be categorised under a number of headings. The first is that diversity is good per se, in the way that biodiversity is good. Second, language maintenance allows members of a group to remain in touch with their own history and cultural heritage. Third, language is an essential element of identity that should be respected. Fourth, languages constitute an irreplaceable resource for humanity.

12.1.1 Language diversity is good per se

The hypothesis developed by the American linguists, Franz Boas, Edward Sapir and Benjamin Whorf, is a cornerstone of the argument that language diversity is good per se. Their fieldwork among Native Americans led to the development of the theory that groups develop their languages in reaction to and in relation with their different bio- niches and that group experiences mediated through language allow them to develop particular worldviews. The way that the lexis and

274 Language Policy and Language Planning

syntax of a language develops encourages its speakers to interact and portray experience in a certain way:

(T)he background linguistic system (in other words the grammar) of each language is not merely a reproducing instrument for voicing ideas, but rather is itself the shaper of ideas, the program and guide for the individual’s mental activity, for his analysis of impressions, for his synthesis of mental stock in trade. Formulation of ideas is not an independent process, strictly rational in the old sense, but is part of a particular grammar, and differs, from slightly to greatly, between different grammars. We dissect nature along different lines laid down by our native languages. The categories and types that we isolate from the world of phenomena we do not find there because they stare every observer in the face; on the contrary, the world is presented in a kaleidoscopic flux of impressions which has to be organized by our minds – and this means largely by the linguistic systems in our minds. (Whorf 1956: 212–13)

This led Whorf to a highly determinist position:

[N]o individual is free to describe nature with absolute impartiality but is constrained to certain modes of interpretation even while he thinks himself most free ... We are thus introduced to a new principle of relativity, which holds that all observers are not led by the same physical evidence to the same picture of the universe, unless their linguistic backgrounds are similar or can be calibrated. (1956: 214)

This means that Whorf believed that some concepts are ultimately untranslatable. Sapir was slightly more cautious. He is quite categorical that national characteristics (which he terms group temperament) and language have no intimate connection and that language and culture have no causal relationship. He did, however, advance the idea that there can be no thought without language, and that the human ability to conceptualise arises as the language faculty is refined (Sapir 1921: 15). Because of this, it is argued, humans are very much at the mercy of the particular language which has become the medium of expression for their society, and they interpret through the veil of their language or according to the framework it proposes. Sapir believed, however, that in terms of complexity, the language of the cultivated Frenchman and the South African Bushman are in essence perfectly comparable although different (Sapir 1921: 22). Moreover, he held that speakers of differing

Endangered Languages 275

languages are never imprisoned in their language, because all languages have the potential to be elaborated and concepts adopted.

The ecolinguistics school, which has developed in the last decades, begins from this premise, that there is a strong relation between a language and a worldview. A central claim is that certain languages are perfectly adapted to communicating certain values and attitudes. For example, the languages of pre-industrial societies are held to promote respect for the natural world (Mühlhäusler 2001; Harrison 2007). In contrast, Standard Average European languages (SAE) 2 deal with indus- trialisation and the pollution it causes with a semantic imprecision that fudges, euphemisms that mislead and an absence of terms that can direct attention to certain important processes. Mühlhäusler argues that the languages of the big polluting states allow them to systematically underestimate and play down the consequences of dangerous practices. Chawla (2001) compares lexis from Native American languages and English to illustrate how language frames what we see and how we act in our relations to the natural world. He explains how Cherokee and Pueblo extend kinship terms to nature and claims that:

[B]ecause of these beliefs, Indians have traditionally treated the inani- mate and animate world with awe and concern in ways that do not indiscriminately damage the natural environment. (Chawla 2001: 118)

He contrasts this with an analysis of English, where he concludes that its individualising tendencies contribute to English speakers’ failure to see the environment holistically and exacerbate their propensity to act for egotistical advantage, even when this may be against the greater good. Influenced by a scientific worldview, they interpret the Cherokees’ anthropomorphic attributions to the animate and inanimate world as superstition and dismiss them.

The ecolinguistic literature generally tends to assume a close and essential relation between language and knowledge. Accordingly, when intergenerational language transmission breaks down and a language disappears, there is likely to be a serious loss of inherited knowledge. The argument is that any reduction of language diversity diminishes the adaptational strength of our species because it lowers the pool of knowl- edge from which we can draw (Bernard 1992; Harrison 2007).

This would only be the case if the strong Whorfian claim that translation is ultimately impossible were true. However, translation is possible. The very fact that Mühlhäusler and Chawla can point out that Westerners ignore important aspects of the environment, and that

276 Language Policy and Language Planning

new ways of expressing our experience might change how we behave towards it, exemplifies Sapir’s claim that elaboration and adaptability are always possible. Speakers of SAE could gain insights from peoples whose languages make the need to respect the environment more explicit. If a certain language does not have the lexis for a particular subject, it can be borrowed or invented. If it does not have a particular structure, it can be developed. If there is a particular way of reasoning or conceiving a topic in a language, it can be copied. If precise ways of talking about the ecosystem make speakers more aware of diversity and of damage and encourage them to be active stewards of biological and ecological diver- sity, then these can be copied and learnt.

The argument that protecting languages where relationships are encap- sulated in particular ways might make us more respectful of nature is a weak one for the preservation of the language. We would have to believe that there was an essential relationship between the language and the concepts expressed for the argument to hold. However, many people do take this line. Skutnabb-Kangas argues that the language/concept link is never fully translatable:

Traditional knowledge may indeed linger even after a native language is lost, but the richness and diversity of that knowledge cannot survive even one generation of language loss. (2000: 259)

But, if we take a non-essentialist position, we cannot use this argu- ment for the preservation of the language. There is no reason why the insightful framing of other languages could not be imported into SAE languages if their speakers decided to develop the competence to reflect different ecological positions. The question is not whether it can be done but whether it will be done. Are SAE speakers being alerted to the way that their languages obfuscate and mislead in environmental affairs by the ecology lobby? Yes, they are. Articles from the ecolinguistic group are doing so. Whether such awareness will result in more careful linguistic formulation in other languages and more ecologically aware behaviour are further developments and not at all inevitable. Certainly there are precedents for believing that language change may be associ- ated with change in other behaviour and attitudes. The feminist move- ment argued that aspects of practice in the SAE languages contributed to patriarchal attitudes. A reworking of terms and structures to change this was advocated as a way of changing attitudes, influencing behaviour in relationships and promoting greater equality in society. In another example, the anti-vivisection lobby points out that the impersonal

Endangered Languages 277

constructions of English scientific discourse may well contribute to the desensitising necessary for researchers to function in the laboratory (Kahn 2001). Campaigners argue that if those engaged in vivisection were forced to articulate what they were actually doing, then they might feel less comfortable doing it.

There is an important point here for small language survival, because, if the argument is valid, it could just as easily be used against minority languages. Anyone adopting a strong Whorfian position, whereby languages are not ‘interchangeable’ (Mithun 1998), would have also to accept that a group could never transcend its position, could not learn from others more advanced in some area, through translation. According to this logic, to profit fully from advances in cognition, new concepts in technology, novel art forms and so on, members of the small group would need to acquire the language in which they had been developed, with the consequent risk of language shift.

There is also a wider claim from the ecolinguists. Parallels have been drawn between linguistic diversity and biodiversity. Extrapolating from the ecological mainstream, they argue that diversity is good per se, and that the multiplicity of languages contributes to the health of an envi- ronment. This seems to be a metaphor that cannot be taken too far and one that is difficult to support with evidence. Adaptation to an environ- ment is positive, and can be shown to be so in the case of language. Mühlhäusler points out that

Languages over time become fine-tuned to particular environmental conditions. It is language that allows people to become efficient users of the environment. But it takes time to get to know a place. (Mühlhäusler in a university newspaper interview 1998, quoted by Skutnabb-Kangas 2001: 259)

But language diversity can also be seen as contrary to human well-being. Although Crystal (2000) and Skutnabb-Kangas (2000) cite the Mexican myth that holds that the goddess Iatiku created different language groups so that they could not quarrel, it is difficult to find examples from history where incomprehension safeguarded peaceful coexistence. There is more data to confirm the view expressed by George Steiner, already mentioned in the Introduction to this book:

Time and again linguistic differences and the profoundly exasperating inability of human beings to understand each other have bred hatred and reciprocal contempt. In short: languages have been throughout

278 Language Policy and Language Planning

human history zones of silence to other men and razor edges of divi- sion. (Steiner 1998: 58) 3

Most importantly perhaps, too much reliance on parallels with the natural world raises the spectre of ‘survival of the fittest’. This aspect of the biological metaphor is not compelling if we want to argue for the preservation of languages under threat.

Thus, the ecolinguistic case seems to rest on analogies that do not fully hold, on a conviction that language and thought are so inextri- cably linked that worldviews cannot be understood outside the language group where they originate, and that languages are ultimately untrans- latable. The claim that a language may provide insights and make rela- tionships evident is probably true, but it does not furnish a sufficient reason to maintain a language, where a group is shifting language for its own project. The argument does not stand on its own unless one has accepted the strong Whorfian position.

12.1.2 Language, history and group continuity

Another reason advanced for preserving languages is to ensure that the past remains contactable:

With language loss the link with the recorded and transmitted cultural legacy is broken. (Bourhis 2001: 6)

Where languages have a stable standard, the written form allows access to the group’s past, beyond living ancestors. The texts of former genera- tions are available for their descendants, because the language does not change substantially. In some groups too rigid adherence to historical norms sometimes causes problems. Where a written standard has been based on the literary forms of an earlier period, there will be some distance between the written and the modern spoken language, with all the difficulties this entails for universal literacy. This was what occurred in Italy and Greece where early standardisation drew from the canon of a past golden era. The diglossia and social exclusion it can cause may be seen as a high price to pay to maintain historical continuity. In other cases the planners of the written standard may have consciously broken with the past in order to simplify or to align the language with a new ideology. This was the case with the alphabet reform in Turkey, and the change has made old texts inaccessible for modern Turkish readers. 4

If the language has no written form, the texts of the group are passed on in the oral tradition. The data on how oral transmission is affected by

Endangered Languages 279

language shift is not extensive and, as yet, is inconclusive. Eades (1988: 97) finds that Australian Aboriginal people maintain ‘significant Aboriginal and social aspects’ and ‘distinctively Aboriginal interpretations and mean- ings’ even when their language of interaction is English. Woodbury (1998), commenting on Eades, agrees that content and style can be transported from language to language, but wonders what is left out. Jocks (1998) gives an account of Native American religious ceremonies in which he tries to evaluate how far those who participate in translated ceremonies can expe- rience the event in the same way as those who speak Kanien’keha, the language used in the rituals he studied. He concludes that, although trans- lation is possible, something is inevitably lost in translation.

This may be true but Jocks’ research on these translated/interpreted ceremonies itself provides just one example to show that group history, memory and knowledge can be preserved even when there is language shift. Whether or not they will be preserved will depend on group attitudes to past organisation of the society and its values and atti- tudes. Memory may be jettisoned along with the language if there is a strong modernising current in the group. It will also depend on the group composition during and after shift. If there is dispersal, penetra- tion and mixing, then opportunities and occasions for the transfer of group history, memory and knowledge may be fewer. It is not inevitable, however, and the core of ideas and values can survive a change in the language that carries them as they can survive a change in the speaker that expresses them.

In another branch of linguistics there is a long tradition of research into one aspect of linguistic continuity. Historical linguists track lexis in language shift where groups are conquered and assimilated, tracing a word through a set of languages to show patterns of contact and how words passed from one language to another as goods or ideas were exchanged. 5 There is usually a trace of substratum groups in the language and the gene pool of a people (Cavalli-Sforza 2000). It is, of course, of little comfort to a conquered and decimated group to know that elements of their language, perhaps along with some of their genes, survived, incorporated into the new society. This is a very weak form of language maintenance! However, the fact that there are more languages that have died out than are now spoken means that it is a very widespread phenomenon. These previous languages survive in the vocabulary of the languages that come after. Naturally, topography is particularly likely to bear witness to former speakers on the territory.

For there to be no legacy from a language that is eclipsed to the language that supersedes it suggests conscious repression of the transfer process.

280 Language Policy and Language Planning

Where there is no trace of a language in situations of contact, the reasons are sometimes to be found in paranoid attitudes among the dominant group and fear among the subordinate. Humphrey (1989) discusses how speakers of Buryat, in south-east Siberia, avoided importing Buryat terms into Russian in order not to offend the Soviet authorities by using Mongolisms. This was a widespread reaction among local political elites who were aware of Stalin’s fear of pan-Mongol nationalism. They would not risk misrep- resenting themselves as unorthodox in a dangerous political period and so rejected codeswitching and tag-switching when speaking Russian. This was one of the factors that halted the process of borrowing and mixing and maintained strict compartmentalisation of the languages.

12.1.3 Language and identity

The third reason advanced for maintaining languages is that a language is a component of identity and that the disappearance of the language of a group has immense repercussions for healthy self-regard (Joseph 2004; Edwards 2009). Language is a robust marker of group member- ship and one that is not easily changed. It is one of the stronger markers of identity because there are cognitive as well as psychological barriers to be overcome when individuals shift language. An outsider cannot decide to join a linguistically distinct group simply by willing it. The apprenticeship before being able to make the change will be long and possibly hard. Particularly in the case of an adult learner, it may never be complete. The strong poststructuralist argument that language is contingent falls apart when we witness the difficulties that individuals have when they are forced to shift language.

This is not to reify or essentialise language. The relative difficulty of changing language group does not mean that it cannot be done, only that it is done with difficulty. There is nothing intrinsic about language and identity: they can both shift; they can both be hybrid; they can both be layered. That being said, if a group views membership of their language group as essential to their identity and believes their society would be fundamentally altered in ways that they do not want by language shift, then it seems their perfect right to couple identity and language. As May (2001) argues, even if language is a contingent factor of identity, that is not to say that it is not experienced as significant for or constitutive of identity. The strongest argument for taking the neces- sary institutional decisions to support maintenance of a language has to be because its individual speakers wish to do so.

However, the problem then arises that language maintenance cannot be an individual decision. An individual does not want to be the only

Endangered Languages 281

person maintaining a language. S/he wants the group to maintain the language so that s/he can function in that language within it. In Chapter 10 we considered whether language rights can be seen as indi- vidual rights or whether they are de facto a kind of group right. The point is fundamental to the identity issue. Using the language of a group is a kind of social glue. Where speakers are part of a small language group and know only that language, their repertoire constrains them to social commerce within that group, often limits their choice of marriage partner and usually dictates where they can work. If the group’s concern to maintain solidarity and continuity placed a limitation on the other languages that members may learn, this would be an infringement of individual rights. However, discouraging the learning of other languages may seem to group leaders the only way to preserve group language in the face of economic and political pressures on individuals to shift. 6

There are several dilemmas here. On the one hand, it would be invid- ious if the individual’s right to acquire a language that gives access to higher levels of education and allows social and geographical mobility were curtailed in any way to encourage language maintenance among the group. 7 On the other hand, the decision of individuals to acquire a language that is instrumentally more useful affects not just them but the group as a whole, because each individual decision to acquire a dominant language (L2) has an incremental effect and makes mainte- nance of the dominated languages (L1) more difficult. In the best case scenario, learning the L2 leads to personal bilingualism and societal diglossia. The L1 may not develop the lexis necessary for many domains but remains secure in the private spheres. In the worst case scenario for language maintenance, the L2 penetrates the private sphere as well and there is personal and family shift. Where there is education in a standard language and where there is scope for social and geographical mobility, there may be exit from the language group, and then numbers of the group may dwindle and those remaining become isolated. This has always happened. It is the scale of the phenomenon, which is the new development:

If members of a subordinate population have the opportunity to learn the language of the dominant group, some or all of them will usually do so. It seems that it is not so much the tendency to learn a dominant group language which has increased a great deal in modern times, but rather the opportunity to do so, and concomitantly and more importantly for linguistic diversity, the tendency to abandon one’s ancestral language in the process. (Dorian 1998: 5)

282 Language Policy and Language Planning

The second problem is concerned with freedom of choice. It is difficult for individuals to choose the language(s) in which they are socialised and educated since this happens in childhood. The adults responsible for the socialisation and education of the younger generation determine the medium. The literature abounds with numerous examples of how the decision of parents or educators to socialise and educate in the majority language has been experienced as deprivation, exclusion and loss by those for whom the choice was made (Dorian 1998; Mithun 1998; Crystal 2000). Hale (1998) believes that not being taught the language of one’s group can even give rise to grief. On the other side, a major theme in all literature is the individual who rails against the constraints of the primary group and struggles to transcend them. Literature may gloss over the linguistic dimension of this but transcendence mostly involves language acquisition. The fundamental problem is that the choice of language(s) of socialisation and education is not one that individuals can make for themselves.

Before leaving the subject, there is perhaps one further point to be made about the contingency of the link between identity and language. Does group identity survive when the group no longer speaks its ances- tral language? For example, can the southern, mainly non-Welsh- speaking Welsh claim equivalent Welshness to those in the north-west where the Welsh language is more widely spoken? Can immigrants who assimilate linguistically in the host country maintain their iden- tity? Clearly in the first case there will be different contributing factors to what makes these two groups see themselves as Welsh, but they may be no less powerful. The powerful factor in the cultural vitality of both groups may be the fact that both are cohesive groups that continue to be a majority in their settings. This is not the case in the second situ- ation. Identification with the land of origin may continue to be very strong and maintained through religious practices or cultural main- tenance in diet and dress, even where, given the need to survive and the hope of prospering in the new environment, there is linguistic accommodation. However, where there is complete loss of the original language of the group in a situation of migration this must affect iden- tity, because new language behaviour will affect patterns of contact. Migrant groups in the United States have tended to shift language within three generations (Fishman 1977) and this has happened in some cases of migration into Europe. In such cases contact between grandparents and grandchildren is cut, and, at the very least, a new, hyphenated identity becomes likely for the younger group.

It seems unproblematic to suggest that where there is language change there will be changing identity. This is not to agree with the Whorfian

Endangered Languages 283

notion that a language carries other aspects of the culture in a particular way and makes likely certain ways of being and that a change of language necessarily affects that. It is to recognise that patterns of association are central to identity formation. Where individuals acquire the medium that opens the door to other groups, diverse contacts and different influ- ences, it is difficult not to accept that identity will be altered. Very few who become members of new networks and who are exposed to new ideas will not change their perception of their identity in some way. Bourdieu’s concept of habitus describes the process: the embodied mean- ings of groups do not determine how individuals act but they provide an influential frame of reference. To acquire more embodied meanings is to widen and alter frames of reference.

12.1.4 Diversity as a research resource

Another argument from those who are striving to save endangered languages is that they are an irreplaceable source of information on the way that the human brain processes thought. If every linguistic system finds a slightly different way for individuals to conceptualise, reason, communi- cate, be expressive and so on, then the disappearance of a language takes away a building block that might allow us to understand the process more clearly. Thus preserving the thousands of human languages that currently exist may help scholars to understand how language works. Hale (1998) says categorically that the loss of linguistic diversity will be a loss to schol- arship. Like Mühlhäusler (1996), Fill (1993), and a selection of contribu- tors to Dascal (1992, 1995), he contends that we must have as much data from as many languages as possible to reveal how language works. Every language that is lost is one less example of the phenomenon, and with it may go a particular feature that exists nowhere else. At the same time, linguists argue that the more languages available for study, the more likely the search for the universals of language will be successful:

While a major goal of linguistic science is to define the grammars of all natural languages, attainment of that goal is severely hampered, some would say impossible, in the absence of linguistic diversity. If English were the only language on the face of the earth, we could not know literally hundreds of things which are permitted, even predicted, by universal grammar and accidentally missing in English, or any other single language. (Hale 1998: 192)

Further research data comes from the fact that the languages of people in apparently simple societies are as complex in linguistic terms as the

284 Language Policy and Language Planning

languages of societies, which are much more technologically and socially elaborate. Hale (1998) claims that so-called simple societies should be seen as rich in mental wealth with respect to language formulation and ritual interaction. Difference among groups is very revealing about deep and surface levels of the evolution of the human brain. However, it would be naive to think that the needs of scholarly research alone could ever be a compelling reason for a group in the process of shift to reassess their behaviour and change tack.

12.2 What can we do?

12.2.1 Describing the problem

The first step is to recognise and describe the problem. The sheer scale and speed of language shift and loss led to heightened interest in the fate of the smaller languages of the world in the last decades of the twen- tieth century. Linguists started making it the focus of major interna- tional meetings as, for example, in the conferences on maintenance and loss of minority languages, held in Noordwijkerhout, the Netherlands, in 1988 and 1992, the conference on language endangerment held at Dartmouth College in the United States in 1995, and the conference on language rights in Hong Kong in 1996. In 1993, UNESCO set up the ‘Endangered Languages Project’, a programme to record the world’s disappearing languages before their last speakers died ( www.endan- geredlanguages.com ), published an Atlas of the World’s Languages in Danger and maintains a website to list funding opportunities. Individual programmes to protect, maintain or revitalise endangered language groups have proliferated.

A growing number of linguists have made the position of endan- gered languages and policy reactions to endangerment their specialism. Representative scholarship can be found in Dorian (1989), Hale (1992); Fase, Jaspaert and Kroon (1992), Grenoble and Whaley (1998 and 2006), Nettle and Romaine (2000) Harrison (2007), King and Schilling- Estes (2008); Austin and Sallabank (2011); Kasten and De Graaf (2013); Thomason (2015).

Numerous typologies have been developed to describe the different situations along the continuum of endangerment and to identify the variables that lead to the disappearance of a language. Conflating them 8 gives a five-point scale indicating the declining probability of survival.

Language groups that are viable and unproblematic. These languages 1. will survive.

Endangered Languages 285

Language groups that are viable although small. On demographic 2. grounds these languages could be considered to have too few speakers to sustain them, but are likely to survive because of the presence of other factors. The reasons for expecting survival could derive from the isolation of the community, the traditional and conservative nature of the society, the high status of the community relative to neighbouring groups, the value placed on language as a marker of identity, and so on. Language groups whose future is uncertain in the long term. In 3. these cases, past experience suggests that the languages will die out if nothing changes. Speakers do not appear to be committed to the language, perhaps because they feel that they are socially and economically disadvantaged by maintaining it. Peer pressure to use the language is weak. Typically, child speakers are fewer than adults in number as families pursue social mobility through shift. Language groups that are contracting now. The languages are in 4. imminent danger of disappearing. The younger generation have been educated through another medium and have shifted. The youngest monolingual speakers are aged 50+. Groups that have shifted. The languages are either nearly extinct 5. or completely extinct. In the former, the tiny number of surviving monolingual speakers are very old. In the latter all mother tongue speakers have died. The language may survive in written or recorded form but it is now a dead language.

12.2.2 Responses to language loss

Languages can cease to be spoken for two reasons: the group that speaks the language ceases to exist or the group that speaks the language shifts to another language. The first case implies the destruction of the speaker or the speakers’ habitat. This was what happened in the genocide that took place in some European colonisation and what happened in Ireland in the famines of the nineteenth century or in the Sahel in the famines of the late twentieth century. Language death was a side effect of exploi- tation, war and genocide or catastrophe; as the safety of a people was threatened so was its language. Such language death must have occurred frequently in history as small weak groups were conquered and displaced by larger and stronger peoples or as natural disasters struck. Whether groups were physically wiped out or became too small to exist as a separate society, their languages disappeared. Pagel (1995) suggests that perhaps 100,000 languages may have once been spoken by humans and have since disappeared, probably in a violent way.

286 Language Policy and Language Planning

The second case is mutation not extinction. Here one group (A) in contact with a second (B) disappears as a discrete group because it makes the shift to B’s language. This can also result from oppression, but in other kinds of contact there can be an element of choice. The reasons are usually to be found in the imbalance of economic and political power between the two, but the pressure can also derive from more positive reasons: marriage, educational opportunity, professional advancement, economic advantage. The shift can happen with or without overt pres- sure from the more powerful group.

The violence of obliteration in the first case and the large degree of external coercion and pressure found in a good deal of the second has coloured much recent scholarship in the area. It is perhaps in conse- quence that some of the language rights literature treats those who become bilingual or who shift from a smaller to a bigger language as pawns in globalisation or capitalist economics or dupes of nationalism. There is sometimes a lack of recognition that it may be a considered and conscious decision to accept language shift as the price to pay for a desired move from one’s original group to the wider world. Depending on the individual situation, shift can be associated with transcendence rather than tragic loss. However, to date, understanding aspirations and transcendence does not seem to have attracted as much research interest as understanding coercive pressures and loss. More investigation of pull factors would be a very useful contribution to this debate.

More emphasis on this focus would also remind scholars and plan- ners that the objects of LPLP are also agents, and their choices are ultimately decisive in determining how language behaviour develops. Speakers themselves are the ultimate arbiters of language revitalisa- tion, and the other players need to be sensitive if they aspire to play a role. The academic community can put the case and identify the vari- ables but ultimately language maintenance is not their choice and they can only help if they are asked to do so. The activists of the minority group may have to compromise on goals if the rest of the group is not committed to language revitalisation, and accepts shift and mutation of identity to further other goals. A central government, that supports its minorities, needs to understand the process it implements, if it is not to have unlooked for results. These points are illustrated in the case studies discussed below.

The position that a language group finds itself on the vitality continuum, and an understanding of its aspirations will determine what action can be taken to maintain or revitalise the language. There are three possible options. 9

Endangered Languages 287

The first is to do nothing; accept changes in language use as normal. Such a philosophy would perhaps reflect Edwards’ (1985: 86) assertion that it is natural for language use to change, and ‘more reasonable to consider group and individual identity altering ... than it is to see the abandonment of original or static positions as decay or loss’. Language will adapt as speakers are in contact with others or in new situations. Those holding this position would say that the dynamic and accommo- dating nature of language and its users will always counter any attempt to preserve a language as if it were an artefact in a museum, frozen in its present state for all time. 10 The argument is also that one cannot pres- surise people to give up life chances to revitalise or preserve a language and that there has to be a reason beyond a simple desire to conserve.

The second option is to turn to scholarship. Linguists can preserve the language, by documenting and recording as much data as possible. Then, if the endangered language disappears as practice, it can still survive as knowledge and system. People will know about it, even if no one uses it as a medium. The arguments for such activities are, as discussed in 12.1.4, that they safeguard linguistic diversity and contribute to a knowl- edge base for language universals. The Western idea that knowledge in and of itself is valuable is at work here. Some 11 would see an implicit ethical problem in the fact that such scholarly activity is likely to be of little immediate benefit to the remaining speakers, who may feel their ownership of the language under threat, as well as the language itself.

The third option is to attempt some sort of language salvage, revi- talisation, or maintenance programme, depending on the position of the language on the continuum of vitality. This would include language development strategies, such as producing a written form of the language; encouraging acquisition of the language and literacy in it; using it as a medium of education; funding and encouraging literary production; translating government and administrative documents into it; funding media in it; requiring business and commerce to make their services available in it; and other such tactics.

A number of the strategies for revitalisation will need the approval of the state and may require changes in state law. Their implementation will need extensive financial support from general taxation. Revitalisation of a minority language is language policymaking and language planning at a local level and the activities encompassed by it are the status planning, corpus planning and acquisition planning usually undertaken by the state: that is designating the language as a medium in certain institutions (status planning); making the language fit for that purpose, by codifying and standardising it (corpus planning); educating speakers to use it in

288 Language Policy and Language Planning

both written and spoken forms (acquisition planning). Where an organ of civil society undertakes the role of promoting the language, then it will need to do so under the aegis of the state. Little can happen here if the governing elite or the dominant group is opposed to extended use of the minority language or subscribes to the ideology of the ‘neutral’ state. 12 The activity may be judged balkanisation rather than a beneficial promotion of diversity.

This has led a number of sociolinguists to argue that official support for a language should be a key component of language maintenance (Bourhis 2001; Skutnabb-Kangas 2000). Other scholars concede that this is true, but warn against too great a reliance on official intervention. Fishman stressed that top-down policies cannot compensate for lack of bottom-up support and that use in the family is ultimately more impor- tant than use in institutions:

Endangered languages become such because of the lack of informal intergenerational transmission and informal daily life support, not because they are not being taught in schools. (Fishman 1997: 190)

Nettle and Romaine (2000: 39–40) agree and believe that state support for the language and culture of a group relatively lacking in power does not necessarily ensure the reproduction of a language unless other measures are in place to ensure intergenerational transmission at home. Romaine wrote elsewhere that:

looking to schools and declarations of official status to assist endan- gered languages is much like looking for one’s lost keys under the lamp post because that is where the most light appears to shine, rather than because that is where they have been lost. Just as it is easier to see under the lamp post it is far easier to establish schools and declare a language official than to get families to speak a threat- ened language to their children. Yet only the latter will guarantee transmission. (2002: 04)

If use in the home is the key to language maintenance, community activ- ists who want to maintain a language have to address the problems of status linked to minority language use and the advantage bound up with majority language use. Grillo (1989) believes that the attitudes of major- ities towards minority groups can often be categorised as an ‘ideology of contempt’. Basing his analysis on languages in Europe, he concluded that any language that is not a language of state or power is despised by

Endangered Languages 289

non-speakers and that this attitude is sometimes transferred to its speakers. In consequence, those who are not part of the dominant majority may want to shift to acquire prestige (Joseph 1987). Nettle and Romaine (2000) suggest that conferring power and thus prestige on a minority language group is one of the surest ways of reversing language decline.

However, conferring power and prestige through designating a language an official language may still not be sufficient to reverse decline. The relative failure of Irish revitalisation in Ireland is one illustration of Fishman’s point that official language promotion needs the backing of families committed to intergenerational transmission and private use of the language (O’Laoire 1996).

12.2.3 The variables in maintenance and loss

Edwards (1992) suggests, in a predictive typology, the sites where pres- tige is located. The wide variety of variables that he identifies as playing a role and the difficulty of altering circumstances in some of them is a clear indication of the limitations of LPLP.

Edwards distinguishes 33 elements that favour language maintenance. The best case scenarios fall broadly into the following areas:

1. The group is demographically secure, concentrated in an area, with a tradition of endogamy and little in or out migration;

2. The geographical situation protects the integrity of the group and inhibits contact;

3. The socio-economic status of speakers is high and a range of employ- ment opportunities is available within the group, in the group language;

4. The group has achieved political recognition, rights and a degree of autonomy;

5. Some political life takes place in the language. Dominant and other groups have a positive attitude towards the minority group;

6. The group is cohesive, practising a single religion and maintaining cultural practices in the group language;

7. All generations of the group have high levels of competence in the group language, which is the habitual language of the home and transmitted intergenerationally;

8. The language has a standard written form and speakers are literate in it. It is used in literary production, education and the media.

There are a number of variables in this list that no amount of state support or group effort could affect. Geographical settings that isolate

290 Language Policy and Language Planning

the group and belief systems that discourage any exit from the group, in particular exogamy, will always help maintenance, but cannot be planned for. A large, healthy group that provides a choice of marriage partners and an attractive bio-niche that supplies all the material needs of the group are basic human desiderata. They will aid language main- tenance if they promote stability and lessen the reasons to pursue pros- perity and general human happiness outside the group.

Other variables depend heavily on the will of central government. If the dominant group agrees to a measure of devolution and autonomy, this will make a significant contribution to minority language main- tenance. Where central government relaxes its hold and the group achieves a measure of political control over their institutions, use of the minority language can become the medium for civil administration, the judiciary, education, the media and democratic forums with significant effects on language revitalisation. This provides instrumental motiva- tion for maintaining the language and reinforces the affective reasons for maintenance. Where there is no devolution of power and where the minoritised language does not become the language of that power, there is usually attrition. Bourhis (2001) is one among many who has demon- strated that even pluralism leads to language shift in the long term, because social mobility in a pluralist society still requires that minorities learn the dominant language.

The third set of variables in the typology derives from the economic value attached to knowing and using the language. The central role of economics in language maintenance is twofold. Where the group sees itself as a group and defends its interests as a group, there are positive effects for solidarity and group language maintenance. Where interests are defended successfully, it remains possible to live economic life in the minority language. Where the minority language is used in business and industry, the language is an economic asset and becomes a language which non-speakers want to acquire. The revitalisation of Welsh illus- trates how this can happen. Devolution for Wales (1997), building on the Welsh Language Act (1993), led to a legal framework for language revitalisation. Government departments were required to make services available in Welsh as well as English. Business and commerce followed the lead of local government and took the voluntary decision to provide bilingual services, seeing this as advantageous in a situation of national self-confidence and enthusiasm. With more and more employment hinging on bilingual competence, bilinguals increasingly found them- selves in a position of economic advantage. English monolingual families became more willing to choose bilingual education for their children.

Endangered Languages 291

The results of the 2001 census record that there was growth in those claiming knowledge and use of Welsh with 21 per cent stating that they were able to speak it. In the 2011 census the percentage claiming to speak Welsh fell to 18.5 per cent but the percentage declaring that they had some knowledge of Welsh was 74 per cent. The economic advan- tage of knowing Welsh, allied to an increasingly positive image of the nation, seems to have affected patterns of identity with many identi- fying themselves as Welsh rather than British, and exhibiting this orien- tation by choosing to claim some knowledge of Welsh, even if they do not consider themselves competent speakers.

One factor in these patterns of identification may well have been the creation of the Welsh Assembly, which took power after elections in 2007. Responsibility for education, health, local government, trans- port, planning, economic development, social care, culture, environ- ment, agriculture and rural affairs was devolved to the Assembly and government with central government in London retaining control of police, prisons and the justice system, defence, national security and foreign affairs. With so much power devolved the political conditions strengthen the economic conditions propitious for the maintenance of the language of the territory.

Other variables have also contributed to the vitality of the Welsh language. The informal networks of civil society, religion, cultural production and sport have been influential for revitalisation. There is also a tradition of pride in history and culture within the group, and a custom of using the language for literature, in particular song and poetry. However, these elements were present in the years when there was a decline in the use of Welsh and a language shift to English that seemed irreversible. They do not appear to have been adequate on their own to ensure language survival. Economic and political factors seem to have been necessary to halt the shift to English monolingualism.

12.3 The problems of small groups and language maintenance

These are dispiriting findings for all those committed to maintaining the language of their group, but who have no prospect of achieving the political independence and socio-economic buoyancy that seem to guarantee success. The evidence seems to be building incontrovert- ibly that once again the tactics suggested for language maintenance and language revitalisation are not materially different from those under- taken in nation building, if perhaps on a smaller scale.

292 Language Policy and Language Planning

Now, out of the 4500 or so groups whose languages are threatened, few are engaged in struggles for self-determination. Few are large enough to have the political clout to gain autonomy or to sustain it. Many have no aspirations to do so. Many are penetrated by other speakers and to achieve autonomy and then employ the strategies of nation building would simply produce new inequalities. The kind of solutions devel- oped in nationalism are not appropriate where groups are nested one inside the other like the proverbial Russian doll. There will need to be new solutions other than endless fracturing. We shall return to this in the conclusion. Before doing so, let us look at three case studies that illustrate how difficult it would be to implement the kind of strategies for reversing language shift and revitalising language that those working on endangered language revitalisation often suggest.

12.3.1 Development and the end of isolation

Many small languages have survived because of the isolation of their speakers. Whaley (2002) documents how development and the end of isolation have caused language shift among the Oroqen, a group living on the north-west frontier of China. The Oroqen were originally a nomadic hunter-gatherer people, organised in a number of clans, that arrived in the frontier area of the Hinggan mountains sometime in the seventeenth century. Whaley cites sources that estimate their number at around 4000 before the Second World War.

Under the communist regime, their situation altered radically. In the first decades of the PRC, there were positive attitudes to minorities and minority rights were enshrined in legislation. These rights were based on the territorial principle and resulted in the creation of autonomous regions, prefectures and counties for minority groups. The Oroqen were allotted the Oroqen Autonomous Banner, and thus became a group identified with a clearly defined territory for the first time. The Chinese government instituted a number of measures to raise the standard of living of minorities, including the provision of health care and educa- tion. One aspect of this policy was the settlement of nomads so that children could be schooled. The Oroqen ceased being nomadic in the late 1950s and began to settle in villages (Whaley 2002).

Other aspects of modernity altered traditional ways. In order to exploit natural resources, the Beijing government sent large numbers of workers into the area, mostly Han Chinese. To transport resources and products out of the area, Beijing built roads and rail track. The new infra- structure and industrial activity had a number of effects. In particular, the increased population density and the network of roads affected the

Endangered Languages 293

Oroqen hunting grounds. It became more difficult to live by the old methods and increasingly easy to live by the new. In a few decades, the Oroqen went from living through hunting to participating in the PRC economic system, from being an isolated group, numerically dominant in the area, to a tiny minority (1 per cent) in most of the villages and cities that they inhabit (Whaley 2002).

By its own criteria, the state treated minorities well in a number of ways. First, secondary education was free for them and there were quotas to guarantee minimum minority numbers at university. A rela- tively high proportion of Oroqen thus went through the Chinese educa- tion system. Second, the Oroqen, like other minorities, were exempt from the national one child per family policy. Third, to encourage them to settle, many Oroqen were housed in government built houses, which were of good standard for the region. These various advantages meant that the Oroqen became relatively pro-government and never felt the need to mobilise as a group to defend their interests. 13 They also gave the group relative prosperity and standing which have been incentives to non-Oroqen to marry into the group. The Oroqen found themselves in a classic melting pot situation and there has been a massive language shift from Oroqen to Chinese in a few decades.

As the Oroqen became conscious that their language was not being transmitted to the next generation, there were initiatives for basic Oroqen language classes in school and on children’s television to give the children of the community some familiarity with, if not competence in, Oroqen. These were not an unqualified success, mainly because Oroqen is a collection of dialects rather than a unitary language with a standard form. The acquisition programmes each based their course on one of the dialects and this alienated other Oroqen groups who did not recognise their own idiom in the form taught. 14 This is the problem experienced by the Galicians and the Basques discussed in Chapter 11, and indeed wherever speakers of a set of dialects set about devising a standard language that can serve them all. Some of the group will be speakers of a variety far from the norm adopted and, unless there is a compelling reason for accepting it, may not do so. Standardisation may be rejected by those who do not recognise it as their speech.

The case study illustrates how even where the political elite of a state has shown some support for minorities, 15 there can still be language shift. Whaley believes that it may even have been the government’s treatment of them as a minority group that actually weakened their identity patterns and made assimilation more likely. As highly inde- pendent and fluid hunter groups, speaking related dialects of the Oroqen

294 Language Policy and Language Planning

language, they did not originally see themselves as a unified group. The various policies of the PRC, however, treated them as if they were. Being conceived as an entity was as much a new development for the Oroqen, as settlement, work in a cash economy, education, intermarriage and all the other innovations that they have experienced. Now language revitalisation also requires them to see themselves as unified. It may not happen if the very recognition of group membership itself constitutes a break with the past.

12.3.2 The disappearance of a language and its speakers

The language in the next case study is as endangered as the Oroqen dialects but the reasons for this are quite different. The Uru of Iru-Itu in the Bolivian Altiplano did not lose their language because of economic development but through lack of it. The community where it was spoken almost disappeared after a period of natural disaster and hardship.

The Uru live on the banks of the river that flows between Lake Titicaca and Lake Poopo, surrounded by Aymara-speaking peoples. There are three Uru languages/dialects, Uchumataqu, Murato and Chipaya, and they appear to be distinct from surrounding languages (Muysken 2002).

The Uchumataqu community was decimated after a three-year drought between 1939 and 1942, which caused death and dispersal. ‘In 1942 only six men and a few women, all elderly remained’ (Muysken 2002) quoting Vellard (1954). The tiny size of the group and the fact that the women were past childbearing age meant that the men who had remained sought wives outside the group. These women were Aymara- speaking and this became the main language used in the new families. Muysken (2002), again drawing from Vellard (1954), suggests that in the area the prestige of Aymara was then higher than that of Uchumataqu, the dialect of Uru spoken by the community. The mothers did not raise their children to speak ‘a despised dialect’ (Vellard 1954: 93).

In 2001, Pieter Muysken worked with the villagers to record what elements of the language the community could remember. 16 He reports that there are no speakers for whom Uchumataqu is a first language and the reconstruction depended on the memories of those who have passive or partial knowledge of the language. Muysken records that there was little consensus and much imprecision among his informants. There were unbridgeable gaps where memories failed and some lexis could not be recalled. Those who remembered some of the language had experienced the fundamental problem of maintaining competence in a situation of shift. De Bot (2001) suggests that language loss is a spiral. If there are few opportunities to use the language because of dwindling

Endangered Languages 295

numbers of interlocutors and decreasing domains, there will be loss of language ability deriving from the shrinking occasions when the speaker employs the language. It will take longer to retrieve words for objects in the language that is undergoing attrition, which insecurity in itself discourages use of the language. There is a demographic point at which this process begins. The process is then immensely difficult to reverse. The critical mass factor not only affects competence in the language but motivation to improve competence. If there are few interlocutors and few domains, why maintain the language?

In this situation it is interesting to speculate why the Uru are taking steps to revitalise Uchumataqu. Muysken suggests a possible motivation. The Uru have been part of the tremendous labour migration from the Altiplano. This migration is seasonal rather than definitive and those involved return regularly. The contact with the wider world has given both the desire and finance for modernisation. Iru-Itu is no longer a poor community, but amongst the most prosperous and innovative in the region, ‘looked upon with some jealousy and respect by their Aymara neighbours’ (Muysken 2002). There is now reason to feel proud of group membership. Moreover, there has been significant ethnic mobilisation in Bolivian politics since the mid-1990s in which the neighbouring Aymara have been major players. In their turn, the Uru are becoming more self- aware, either because of rivalry with or emulation of the Aymara. Desire to remember the language and record it may reflect a new pride in group membership. Muysken recognises the limits of what can be done:

It would not be a purely automatic and unconscious reversal of a process of language shift, of course. That shift took place much too long ago for that, and the language is too far gone. It would be a conscious effort to give the language its place alongside, not in place of, Aymara and Spanish. It would involve the activities of a small group of cultural brokers, community leaders, and be linked to processes such as folklorisation and musealisation of Uru culture. It would also need to be a modern development, relying on literacy and possibly even on modern media. For some, this makes the possible revitalisation of Uru unreal, artificial or suspect. However, it may be the way in which many such revitalisation processes take place in different parts of the world. Situations such as that of the Uru and the Uchumataqu language cast doubt on traditional notions of authenticity and spontaneity, and show that even rural communities are capable of ‘language planning’. (Muysken 2002: 313–14)

296 Language Policy and Language Planning

Muysken does not mention another possible factor that may be contrib- uting to this desire for revitalisation among the Uru. Those who work outside the village are in contact not with Aymara speakers but with Spanish, possibly English, speakers. There may be a possibility that where the language that one needs for work is far removed from the cultural base, the desire to cultivate a language of identity becomes stronger. This seems to be an effect associated with globalisation and is an issue to which I will return in the conclusion.

12.3.3 Status and literacy

The third example of an endangered language is taken from a study carried out by Eda Derhemi in Europe. Piana Arbresh is a dialect of the variety of Albanian currently spoken in five communities in Sicily. 17 The Arbresh-speaking population in Piana degli Albanesi numbers about 7000 (Derhemi 2002). Arbresh has been the language of the community for the past 500 years, since the Albanians came to settle in Sicily. Derhemi suggests that the reasons for its survival derive in part from the fact that the Albanians were Orthodox Christians among a Catholic population. They maintained their religion and practised endogamy, which kept them a separate and cohesive group. Other factors encouraging maintenance included the relative prosperity of the group; the issue of status worked for the minority rather than against it, since the Albanians were rather contemptuous of the rural Sicilian majority and dissociated themselves from the surrounding population. The prosperity and confidence of the Piana degli Albanesi group contributed to the survival of Albanian in the town and may be the significant variable missing in the other centres where Albanians settled, but which have not conserved the language into the twenty- first century. 18

Arbresh speakers appear to be in a much stronger position of language vitality than either the Oroqen or the Uru. Despite their centuries long contact with Italian/Sicilian speakers, they conserved their language and maintained stable diglossia. In addition, Law 482/1999 19 passed by central government gave them legal instruments to strengthen their position and put them in a position to promote Arbresh in new settings. In particular it was now possible to use Arbresh in school. The status and prestige of the language appeared undiminished.

At the turn of the century Derhemi reported, however, a general perception in Piana degli Albanesi that the use of Arbresh was declining, both in the number of domains in which it was used and among younger speakers. Having witnessed ‘the semispeakers or non-proficient Arbresh

Endangered Languages 297

speakers’ in other communities, community leaders determined to work to stem attrition in Piana itself (Derhemi 2002).

The principal thrust in the language maintenance programme was the introduction of courses in and through Arbresh in the schools. This had the support of the majority of parents according to a survey conducted in 2001 (Derhemi 2002). Arbresh would be used in the earliest years of primary education as the language of school socialisa- tion, which would formalise a practice that already existed as teachers provided a bilingual bridge for Arbresh-speaking children. In addition, older students would be able to acquire literacy in Arbresh in more advanced classes.

Despite these various indicators of the vitality of Arbresh, there were two obstacles which needed to be overcome if the language was to survive in the long term. The first problem was the absence of an agreed standard written language, which meant that the educa- tional initiatives were delayed while the community negotiated an agreement on what this should be. There were two camps: those who believed that a standard should be developed for the 7000 Arbresh speakers in Sicily and those who argued that it made more sense to adopt the written standard of Albania, which existed already and which gave access to wider sources and contacts, and a more exten- sive literature.

This dilemma has similarities to the case of the Galicians, described in Chapter 11. The ready-made standard is exterior and slightly alien. Opponents to its adoption reason that there is little point in strug- gling to conserve a language if what is conserved is not the group’s language but another, which, although close in many ways, is alien in others. They argue that the differences between standard Albanian and Arbresh cannot be treated as negligible and that adopting it would create a situation of double diglossia for Arbresh speakers. The other solution, to codify and standardise Arbresh has also attracted strong criticism. The problem is that the project was assigned to a small group of scholars and poets who did not consult widely (Fieldwork Italy 2002). The result is a highly literary standard that would need to be introduced top-down to speakers. The first school texts produced in it were ill received. Teachers reported that their pupils found the language difficult to understand and did not relate it to what they actually spoke (Derhemi 2002). The controversy and the subsequent slow rate of progress in producing texts is a key reason for the slow implementation of the educational initiative. Moreover, informants (Fieldwork Italy 2002) suggested that the highly public arguments

298 Language Policy and Language Planning

were likely to drive those families where use of Arbresh is fading even further away from the language.

This brings us to the second problem for language maintenance in the group. The point has been made several times in this text, that percep- tions of status and prestige are key to intergenerational transmission and language loyalty. Among Arbresh speakers a very keen pride in group membership seems to have been a major reason for the survival of the language. In much of the twentieth century, contact with Albanians from Albania was minimal and in the communist period links were cut entirely. Since the opening up of Albania, Italians’ attitudes towards their neighbours across the Adriatic have fluctuated. In 1997 there was widespread support for those opposing the regime and a warm reception for the first refugees who crossed the Adriatic in small boats to leave the crumbling communist regime. In the years since then, many Albanians, particularly young men, have entered Italy, mostly in search of work and mostly without the papers to allow them to do so. Their difficult position in the margins of economic activity has produced an impres- sion among the Italian majority that they are employed illegally in the black economy and may even be involved in criminal activities. Italian prejudice against Albanians has grown, with the result that, however justified or unjustified this attitude may be, association with Albania, Albanian and Albanians may have become less attractive at the present time. 20 Semi-speakers of Arbresh may be less motivated to take steps to improve their competence and to work to maintain the Arbresh- speaking community in Piana degli Albanesi, if they feel the language associates them with a community that is the object of majority suspi- cion or racism. In 2014 Derhemi placed Arbresh speakers at 3 on the five point probability of survival scale, in other words intergenerational appeared to have broken down.

12.4 The difficulties of implementing universal linguistic rights

Bratt Paulston (1997: 82) suggests that claims for universal language rights may ‘merely serve to weaken potential rights since many claimed rights patently cannot be enforced in some situations’. The three studies in this chapter demonstrate that many of the strategies suggested for language revitalisation (Fishman 2001) are not an option for the Uru, the Oroqen, or even possibly for the Sicilian Arbresh speakers. These groups cannot copy language groups that have fought their way out of subjugation and which are commonly held up as models. The Welsh,

Endangered Languages 299

the Catalans, the Basques that I have used as examples of successful revitalisation of ‘minority’ languages are not in the same situation as the Uru and the Oroqen. The languages of the latter are among the 4500 under threat; the former are not. Even Arbresh speakers, with some legal recognition in Italy, larger and more concentrated numbers of speakers, and closely related to a language which is national and official in another state, are not in a comparable position. There is a continuum along which the likelihood of revitalising or maintaining a language grows, closely related to whether the group takes or is accorded some measure of political independence. The probability of maintaining the language diminishes for many practical reasons of doability as well as conscious repression.

Without the political muscle to gain separate development, minori- ties depend entirely on dominant majorities being ready and willing to confer on the language of the weaker group the necessary prestige (Grillo 1989) and utility (May 2002) for long-term stable bilingualism. If majorities refuse to accept arrangements that allow languages to coexist within the same space, minorities with little power can do little to fight their case. Of the four ideological approaches to dealing with a multicul- tural/multilingual state, 21 even pluralism which is founded on equality and even-handedness and which is the most favourable to minorities cannot guarantee bilingualism and minority language survival.

Pluralist regimes ensure that minorities enjoy freedom from interfer- ence and provide positive support, for instance, by funding the cultural and linguistic activities of the minority in the same way that it does for the majority. The pluralism ideology works with the concept of ‘reasonable accommodation’ (Bourhis 2001: 11) and it is assumed that the cultures of the dominant majority and that of linguistic minorities will be more or less transformed by the sustained contact between these communities:

However, it is acknowledged that by virtue of their weaker vitality position, linguistic minorities are more likely to be transformed by such contact in the long run, than is likely to be the case for the dominant majority. (Bourhis 2001: 11)

Thus, in practice, even pluralism may not be robust enough to change perceptions of prestige within and towards the weaker group and main- tain distinction.

A number of scholars have pointed out the symbolic nature of much policymaking in pluralism. Schiffman (1996) suggests that central

300 Language Policy and Language Planning

authorities of the state are often cynical in that they grant status to languages that are fast disappearing. Where a policy appears to be accommodating linguistic pluralism, it is often symbolic and little actual planning follows from the declaration. May (2001) criticises pluralist policymaking for often being grandiose but delivering little. It rarely changes the power relationships. Romaine (2002) cites the Native American Languages Act of 1990 as a classic example of a policy where the rhetoric is not reflected in measures to implement the law. She also notes that where official support comes too late, after the community has ceased to transmit the language, there is no threat to the status quo. This is probably why it is allowed to happen and may account for an apparent increase in support for pluralism. So we should be aware that state support for minority language maintenance may be meant to be token.

The other state ideologies found in multilingual states either reject language rights or, where they appear more supportive may actually derive from rather sinister reasons. Civic and assimilationist ideolo- gies both promote the language of the dominant group, along with its culture, attitudes and core values, although the rhetoric of the former sounds more egalitarian. Their inclusivity leads to an intolerance of difference. Ethno-nationalist ideology may support language rights for minorities living among the majority, but the price the latter pay for this has often be exclusion. 22 The minority has been marginalised, or in the worst cases, held in enclaves (apartheid, reserves), expelled (ethnic cleansing) or even physically eliminated (genocide) (Bourhis 2001). It is the separate development required in a system of apartheid or reserves that allows the language of the minority group to be taught within the group. Indeed it serves the marginalisation and differentiation desired. Language provision for allochthonous minorities can help to prepare an immigrant group for return to the country of origin when they are no longer required as a workforce.

There is thus a fundamental dilemma; the more desirable current state ideologies all encourage linguistic shift in the minority group. The alternative, marginalisation within the state, is a high price to pay for language maintenance. In the setting of the strong state, the only way to acquire the prestige necessary for maintenance without marginalisa- tion, seems to be to go it alone. The coexistence of languages without hierarchy is claimed for certain societies where language is exclusively oral and interaction face-to-face (Fairhead 1994), but it does not seem an option where there is a strong state or a literate culture. Even where stable bilingualism can be maintained there is still diglossia. The hierarchical

Endangered Languages 301

arrangement of languages seems to be the price that societies pay for universal literacy and education, social mobility and free movement of citizens, and an inclusive political system and democratic participation. The interesting question to ask is whether there will be any change in this, if the state weakens as structures become more global.