SUE WRIGHT

THE RIGHT TO SPEAK ONE’S OWN LANGUAGE:REFLECTIONS ON THEORY AND PRACTICE

(Received 9 October 2006; accepted in revised form 5 February 2007)

ABSTRACT. The right to speak one’s own language is a right that has onlyrecently been enshrined in law. At first this right was only a negative right (i.e. theright to use one’s language in the private sphere without persecution or prejudice),

not a positive right (i.e. the right to use one’s language in the public space, to beeducated in the language, to deal with the state in the language, etc.). As the Councilof Europe’s Charter for Regional and Minority Languages has been accepted andbecome law in a number of European states, we encounter a dilemma. In Europe,

institutions such as the courts, bureaucracy and education, have tended to use asingle standard. Now, minority and regional languages are rarely standardised. Whatshould be done? This paper considers how languages have been treated as defined

and static systems for the purposes of nation state language planning. Data fromItaly and France reveal that this has become unacceptable practice today. It is atodds with our greater acceptance of diversity and our recognition of the essentially

dialogic nature of language.

KEY WORDS: Regional or minority languages, language rights, languagemaintenance, theories of language, standardisation, heteroglossia, structuralist, post-

modernist, nation building

Introduction

The right to use one’s own language has only recently gained accep-tance as a fundamental human right. Until the end of the 20th cen-tury, whether or not a language community used its language in thepublic space depended on its political muscle or the tolerance of thedominant groups among which it lived. Those in power might makeit possible to use a language other than their own in the institutionsand forums of public life, but such use was not universally acceptedas a right. Moreover, where a ‘regional or minority’ language(RML) was employed, the purpose was often to maintain differenceand to exclude RML speakers from power. Even when internationalbodies finally enshrined in declarations the fundamental right ofindividuals to speak their language of choice, the intention was that

Language Policy (2007) 6:203–224 � Springer 2007DOI 10.1007/s10993-007-9050-y

speakers would be protected from persecution or unequal treatment.There was usually no explicit commitment, or even implicit inten-tion, to guarantee use of the RML in the public space. Positive, ra-ther than negative, rights did not develop until late in the 20thcentury. Only then did constituencies within the human rightsmovement begin to argue forcefully that the languages of minoritygroups should be promoted as well as tolerated.

According universal language rights has not proved unproblematic.The first, and perhaps the most potent reason for this, is that lan-guage use is a good barometer of power. To relinquish use of one’sown language to make space for the language of another group is al-most always indicative of a shift in power relations. Language renais-sance is rarely neutral but is rather a harbinger or reflection of powershifts. It is thus likely to be perceived as oppositional and there is of-ten resistance from those who could expect to lose by any change.

The second reason for the problematic nature of language rightsis that positive language rights are de facto group rights, even ifthey are presented de jure as individual rights. Where governmentsaccord access to government, participation in the legal process andeducational provision in the minority language, they tend to caterfor the group as a whole. When this happens, it is difficult for anindividual to opt out. Institutions function in a language and thechoice of the language to be used within them is often a zero sumprocess. Once one language is adopted, it ousts the others. Thelocal school will have a dominant language in the institution, evenif others are taught. There will be a dominant language in thecourt, even if there is provision for translation.

Thus, in effect, individual language rights may mean littlebecause to implement them they must become group rights. Evenwhere languages are made co-official in status, one language tendsto dominate and the others are relegated to a secondary rank inwhat necessarily becomes a hierarchical arrangement. Decisionmaking takes place in one language, leaving the speakers of the oth-ers at a disadvantage, dependent on interpretation and translation,with all the delay and distortion that may accompany the process.

Kymlicka (2001), Skutnabb-Kangas (2000) and May (2001) havedemonstrated clearly how minority groups faced with the dominantlanguage of the state are disadvantaged. What is less often consid-ered is that the problem may be replicated for the minorities thatminority groups themselves have among them. When a minoritylanguage is adopted for use in educational, legal, bureaucratic or

SUE WRIGHT204

governmental settings, such adoption may displace the disadvantageto another group. Speakers of varieties different from both the na-tional and the regional standard become, in effect, double minori-ties. As Bourhis (2001) has remarked, there is a Russian doll aspectof language: in each linguistic territory, there is always another lan-guage or dialect whose speakers wish to have acknowledged.

This brings us to the third and perhaps most problematic aspectof language rights, the nature of language itself. What is it that weactually mean when we say that a group speaks a certain language,which should be protected and promoted? In this paper I want to ex-plore this last problem at some length and examine how the way weconceive language has immense effect on our approach to languagerights. In the second part of the paper I hope to demonstrate throughcase studies how this plays out in legislation and policy and to ex-plain why promoters of language rights must always be concernedwith understanding the nature of what they are trying to conserve.

The Nature of Language

‘What is a language?’ In the research community there are twowidely divergent positions. The first derives from the scientific tra-dition that holds that there is a real world ‘out there’ that can beunderstood and described in language and which finds expressionin positivism in the 19th century and in some forms of structural-ism in the 20th. The other is the belief that the speaker/writer is anautonomous subject who, through free will, determines what will besaid and meant. Words mean what the subject intends them tomean. From 19th century romanticism to late 20th century post-modernism, some scholars have held that individuals create lan-guage from their own individual experiences and for their personalcommunication needs and that each set of language practicesframes reality for those who use it.

Among linguists the dichotomy has played out as two distinctand oppositional traditions. Halliday has termed these philosophi-cal-logical and descriptive-ethnographic:

In the former, linguistics is part of philosophy, and grammar part of logic; in thelatter, linguistics is part of anthropology, and grammar is part of culture. The for-mer stresses analogy; is prescriptive, or normative, in orientation; and concerned

with meaning in relation to truth. The latter stresses anomaly; is descriptive inorientation; and concerned with meaning in relation to rhetorical function. Theformer sees language as thought; the latter sees language as action. The former

the right to speak one’s own language 205

represents language as rules.... The latter represents language as choices or as aresource (Halliday, 2003: 99–100)

An understanding of these contradictions is essential for allthose concerned with language rights. When we say that we willprotect a language, are we talking about protecting a freestandingstructure or a set of practices? Does a language exist as an idealsystem outside the speaker or only as individual or group behav-iour? Whether we conceive language as autonomous system or asindividual creativity, as rules or resource profoundly affects whatwe mean by language rights and how we respond to the demandsof RML speakers or act in their interests.

Language as Ideal System

It seems justifiable to begin the discussion of language as idealsystem with the works published in the name of de Saussure andwidely recognised as the origins of structuralism. De Saussuredirected attention to the discrepancies between language systems intheir formal description and what people actually say and write. Heconceived language as langue and parole, where the latter is theperformance of individual speakers with all the idiosyncracies oftheir idiolects and is an imperfect and incomplete reflection of theformer, which is the ideal system.1 De Saussure, himself, does notgo so far as to claim language is a system with a life of its own. Hetakes a more subtle view: langue is not completely present in anyspeaker, but exists perfectly only within a collectivity (de Saussure,1916: 14). Although language should not be reified and seen asexisting independently of speakers, it can be an imagined systemthat represents the totality of what all its speakers do. He isfamously reported to have said ‘Language is speech less speaking’(de Saussure, 1916: 77).

It is easy to understand why structuralism was so readilyaccepted in the wider world. The idea of an abstract, self containedconceptual system, a system of incontestable, normatively identicalforms was very attractive to those engaged in nationalist languageplanning. Nation builders needed the single standard language thatcould be employed, taught and acquired throughout the nationalspace and which would build the national community of communi-

1 Notebooks discovered in Geneva in 1996 (Bouquet & Engler, 2002) suggest thatde Saussure’s ideas were more complex than this. However, it is how de Saussure was

understood throughout the 20th century that is of interest here.

SUE WRIGHT206

cation they viewed as necessary for the creation of a homogenousnational group (May, 2001; Wright, 2000a). Nationalist languageplanning is normative and entails the imposition of a standard,sometimes from outside and always from above. Nationalist lan-guage planners thus embraced the concept of language as system,although de Saussure’s subtleties were often lost in the process!

An abstract ideal system rather than the nebulous and changingidiolects of individuals was also very attractive to all those whofollowed de Saussure in wanting to promote linguistics as ascience. Hjelmslev and the Danish school of Glossematics ‘tookthe Saussurean dictum that langue is form not substance to itslogical conclusion’ (Harris, 2001: 128) and worked in the sphere ofpossible and ideal systems rather than with naturally occurringexpression systems. Even those who based their research onempirical data concentrated on form over meaning and focussedon constancy rather than variety. They saw abstract levels ofanalysis as more ‘fundamental, more deep-seated – in a word morereal – than concrete ones’ (Joseph, 1995: 225).

The latest phase in the structuralist tradition is arguably theChomskyan school of generative linguistics. Chomsky set out to ‘ab-stract away from conditions of use of language and consider formalstructures and the formal operations that relate them’ (1968:111).Focusing on the mental and abstract nature of langue and the ‘gram-maticalness’ of human language acquisition, he was concerned withthe imagined ideal speaker rather than actual practice.

And even where scholars from the logico-philosophical or gram-matico-philological schools would reject the structuralist label assuch, we can still argue that their tradition is to divorce languagefrom its human source in the same way,

focusing instead on the internal logical or grammatical mechanisms by which

(language) putatively works as a situation-neutral device for encoding and decod-ing information about an objectively given extra-linguistic world (Joseph, Love,& Taylor, 2001: 92)

Outside linguistics, other disciplines embraced the structuralistperspective. It appealed to the strand of philosophy seeking tocounter the Romantic/Humanist view that language is an activityperformed by the autonomous human subject who articulates inter-nal thought processes. Heidegger argued that language users arejust that: they do not create language; they acquire a structure thatthey then use. ‘It is language that speaks’ (Heidegger, 1993: 441).In his view, we do not speak language, but language speaks us, and

the right to speak one’s own language 207

speakers merely inhabit pre-existing structures (langue) that enablethem to make any particular sentence (parole). Language use is notoriginal but merely the recombination of some of the elements inthe pre-existing system. Hence every new sentence we speak orwrite, is made up of the ‘already spoken’, every new text of the‘already written’.

Again such beliefs were coherent with nationalism, particularlyessentialist nationalism. We are the national group we are becausewe inhabit the national language, which makes us speak and thinkas we do.

Language as Practice

The concept of language as system has been challenged by all thosewho see the human language facility as essentially creative, andtherefore divergent and heterogeneous. The Russian linguists,Medvedev, Voloshinov and Bakhtin,2 refuted any objective onto-logical status for language and pointed out that a view of languagethat stressed structure and system to the detriment of creativity andevaluation of meaning did not reflect how language actually works.Voloshinov framed the argument in the following way:

The basic task of understanding does not at all amount to recognizing the linguis-

tic form used by the speaker as the familiar, ‘that very same’, form, the way wedistinctly recognize for instance, a signal that we have not quite become used toor a form in a language that we do not know very well. No, the task of under-

standing does not basically amount to recognizing the form used, but rather tounderstanding it in a particular, concrete context, to understanding its meaning ina particular utterance, i.e. it amounts to understanding its novelty and not torecognizing its identity. (Voloshinov, 1994: 33)

Medvedev, Bakhtin and Voloshinov insist upon the social aspectof language, the need to consider its essentially dialogic nature. Allutterances are in accordance or in response to what has been saidor written before. They are dependent on the context in which theyare uttered. They are all evaluated and interpreted by their recipi-ents. Thus every utterance becomes ‘a responsive link in the contin-uous chain of other utterances which, in effect, constitute thecontinuity of human consciousness’ (Morris, 1994: 5)

Language can only exist in performance, and it survives bybeing taken up and reiterated in subsequent performance. The

2 There is some discussion over the authorship of individual works. I thus group

them and recognise the creative nature of their collaboration 1919–1929.

SUE WRIGHT208

continuous chain of utterances that results in effect constituteson-going human consciousness, with meaning being created andrecreated in every dialogic event.

In other words, the understander, belonging to the same language community,also is attuned to the linguistic form not as a fixed, self-identical signal, but as achangeable and adaptable sign (Voloshinov, 1994: 33).

In this interpretation, language becomes an on-going dialogue,learnt by being understood in context and subtly changed in reactionand response. Thus to speak a language means to be part of a commu-nity and language behaviour cannot easily be imposed from outside orabove, in the ways that language planning sometimes attempts.

This Russian work was not widely available to scholars outsidethe communist world until the early 1970s. However, in the Westtoo, a parallel and equally robust rejection of structuralism wasgaining ground. Wittgenstein (1953) reflecting on language con-cluded that there is no fixed meaning associated with linguisticforms and that language use is inextricably integrated with actions.The most that a linguist can do is situate the expression within itscontext. Austin (1962) agreed. He drew attention to the primacy ofthe written word in linguistic investigation and argued that such afocus encourages linguists to conceive words as having intrinsic andself-sufficient semiotic content, and to neglect that language isprimarily action in context.

These schools of thought hold that agency remains with the sub-ject and promote the idea that meaning is social and dependent toa large extent on the meanings constructed in interaction. In thistradition, discourse analysis theorists make the radical claim thatthe realities we take to define our social circumstances and our-selves within them are to a large extent socially constructed andthat the social world derives from discursive construction:

Important aspects of our social lives are constructed in and through languagewhether in the moment-to-moment social interchanges of everyday talk or in

the beliefs, understandings and principles that structure our lives (Coupland &Jaworski, 2001: 134).

Language as Ideology

Alongside this constructivist movement, there was a short livedstructuralist revival in the 1960s, led by the literary and philosophi-

the right to speak one’s own language 209

cal group gathered around Barthes, Foucault, Sollers and Kristeva.One of the key ideas was the existence of impersonal forces indiscursive texts (Barthes, 1968; Foucault, 1969). In the declarationthat the ‘author is dead’, they sought perhaps to demote the origi-nator of the text and redirect attention to how it was being inter-preted rather than to theorise on the existence of language outsideperformance. However, as theories were developed, they increas-ingly acknowledged that the text acts as a store of cultural meaningthat escapes its originator. Foucault spoke of ‘stepping into theflow of meaning’, Lacan of our entering, through language, intothe ‘Law of the Father’, the systems which govern conceptions inour culture.

This iteration of structuralism, however, soon gave way to post-structuralism and postmodernism as members began to reject thenotion of internal rules and mechanisms.3 Many thinkers underthese banners produced interesting analysis of the problems ofagency, subjectivity and constraint in relation to language andmeaning. Derrida contended that texts have meanings that we cansensitise and attune ourselves to but never fully determine or con-trol. He returned to de Saussure, propelling his ideas into newdirections and coining the term differance, to introduce a dynamicdimension. ‘An element functions or signifies...only by referring toanother past or future element’ (Derrida, 1981: 28–29). In this waythe ‘system’ can only be studied in use, as it is constantly renegoti-ated. In her work on semiotics, Kristeva argued that there was con-stant tension between the agreed symbolic and the changingsemiotic (Kristeva, 1986: 26). The ‘systematic constraints withineach signifying practice’ are profoundly different from the closedrules of structural linguistics. Whereas these can rule every sentenceas grammatical or ungrammatical, and as having some agreedmeaning, Kristeva’s ‘constraints’ derive from the social contexts inwhich the sign is produced and used. Meaning derives from thesign’s relation to the particular view of the world that comes fromacquiring and using it in particular dialogic situations and particu-lar settings, in other words, from the way it is ideologicallyanchored.

These ideas about agency and subjectivity have found expressionin the ‘language ideologies’ tradition. Within this group the

3 There is little to unite the protagonists of these movements except their rejection

of the full structuralist position.

SUE WRIGHT210

Foucauldian belief in the interconnectedness of power and knowl-edge and the way that discourses are produced and circulated byand in the interests of powerful groups has been refined and devel-oped. The core idea, that we only say what we are led to say bythe hegemonic forces at work which benefit particular groups,brings this school of thought close to the mainstream constructiv-ists, although always with the implication that agency is complexand manipulative.

The Consequences for Language Rights and Language Policy

So how have these different scholarly conceptions of languageinfluenced activists wishing to claim language rights and languagepolicy makers faced with such demands? The interesting thing isactually how little these debates have impinged on the wider publicand governmental agencies. Although most linguists would nowrefute the idea that language is straightforward mapping of objec-tive reality, the complexities of the discipline’s ontological debateare largely ignored. Policymakers, in particular, seem to disregardlinguists’ claims that ‘it is actually impossible to define languages asdistinguishable objects’. (Lamb, 2004: 394). This is understandable.If authorities were to conceive the linguistic world as a constantlyshifting landscape of idiolects, it would be impossible to implementany kind of language policy, including the one that interests ushere, the right to speak one’s own language. If this level of diver-sity were recognised, the state would only grant negative not posi-tive language rights. It is difficult to manage anything but a verylimited number of language varieties in schools, courts and statebureaucracy.

The history of Modernity confirms this; the introduction ofpositive rights (the language as record in administration, of plead-ing in justice or as medium in education) has historically always ledto standardisation of at least the written form of the language, withpressure on the periphery to accept the variety of the centre. Thuswhen a group fights for positive rather than negative rights, it mustexpect that the pressures will be normative. Where language rightsare granted, it is likely they will be the right to employ an ideal,agreed system rather than the right to communicate with the insti-tutions of the state in one’s language of primary socialisation. Thisleads to the situation where there is pressure on those speakers who

the right to speak one’s own language 211

do not use the standard employed to converge, to shift to thatnorm.

RML rights legislation can thus reproduce the centre-peripheryconflict that accompanied the spread of national languages andRML rights activists need to consider whether all the individualsconcerned by the linguistic changes that standardisation willprovoke will want to converge. If standardisation means a sub-stantial and unwelcome shift, this defeats the original aim of secur-ing the right to use one’s own language.

There is no easy solution to this. Even where language activistsand policymakers fully understand the theoretical dichotomy, thiswill not necessarily affect the way they address the rights question.In theory, policies and actions grounded in the recognition thatlanguage is practice and in the awareness of the force of languageideologies should be different from those deriving from the beliefthat language can be treated as a discrete system. In practice, theremay be little difference because of the difficulties of translating rec-ognition of diversity into policy. So, even if we conceive languageas contextually bound performance and ideologically groundedpractice, we tend to implement policy as if language were an idealsystem. The fundamental dilemma of the practical implementationof linguistic rights lies here: how to reconcile political exigency,which tends to be centripetal, with actual language practice, whichtends to be centrifugal?

Rights for Regional or Minority Languages in Europe

We can see this dilemma at work in the flurry of legislation thatfollowed the Council of Europe’s Charter for Regional or MinorityLanguages, made available for signature November 1992.4 Statescommitting to the Charter agree to promote use of historic RMLson their territory in a number of public arena, including education,the media, the courts and state bureaucracy. Much of the debateabout the Charter has been about the extent and location of theserights, since the Charter is not a bill of rights set in stone but anegotiable package. Central government and RML groups have toagree on where the language will be protected and promoted in apick and mix deal. There has been little discussion about the prob-lem we are currently considering: the form of the language to be

4 At the start of 2007 33 states had signed the charter and 22 had ratified it.

SUE WRIGHT212

promoted. This has turned out, however, to be one of the centralproblems of the initiative. The next sections reveal differentoutcomes in states with contrasting approaches to RML rights.

Minority Language Legislation and Implementation in Italy

The Italian state has a history of accepting the rights of certainregional and minority groups to use their own language. The borderregions of Valle d’Aosta, Trentino-Alto Adige, Friuli-VeneziaGiulia have enjoyed limited legislative and administrative autonomywithin the state for several decades, including the right to use theregional language(s) in certain public settings. For other, smaller,linguistic groups there was less autonomy, but various Regionalinitiatives since 1945 introduced minority language rights legislationpiecemeal throughout the state (Eurydice, 1998). In 1999, Law No482 Regulations regarding the protection of historic language minori-ties gave a national framework to regional minority language pres-ervation and renaissance. Thus, although the Italian governmenthas signed but not ratified the Charter,5 the Italian state has instru-ments for some recognition of minority language rights and aculture of doing so.

In practice, the first two problems mentioned in the introductionabove, the reluctance of majorities to countenance minoritylanguage rights and the need for the minority to accept grouprights have not proved major difficulties in the Italian case. How-ever, the third problem, the nature of the language to be protectedhas been less easily resolved. The experiences of the Arbresh andLadin speakers, recounted below, reveal how difficult it is to resisthomogenisation.

Arbresh

There are eleven Arbresh6 settlements in Sicily, of which threeremain Arbresh speaking (Derhemi, 2006). The factors that haveled to stable Italian/Arbresh diglossia in towns such as Piana degli

5 At the time of writing (2006), the Charter was still to be ratified in Italy. Thegovernment signed the Charter 27/06/2000. The Berlusconi government halted the

process. It is expected that ratification will now happen, since parliament gave ap-proval in 2001, but progress has been very slow.6 There are also groups on the mainland, e.g. in Calabria, Abruzzo and Molise.

the right to speak one’s own language 213

Albanesi are classic. Here the community has maintained thelanguage since it came to the island from Albania five centuriesago. Orthodox Christians among a Catholic population, theArbresh conserved their religion and practised endogamy, whichkept them a separate and cohesive group. They constituted a rela-tively affluent urban society and were reportedly rather contemp-tuous of the rural Sicilian majority. Dissociated from thesurrounding population and prosperous and confident, thesegroups maintained Albanian to a greater degree than the Albanianspeakers of the Italian mainland (C. Perta, 2004, Language mainte-nance and loss in three Albanian villages in Molise. UnpublishedPhD thesis. University of Edinburgh).

A Regional Law (26/98), as well as the National Law 482/99,gave the Arbresh legal instruments to strengthen their position andto promote Arbresh in new settings. In particular, the legislationmade it possible to use Arbresh in education, in public offices, inlocal government, in the judicial system, in the mass media, andallowed for the reinstatement of place and personal names. Thestatus and prestige of the language thus appeared to be strength-ened. Having witnessed decline and shift in other Arbresh commu-nities, the Piana leaders expressed their determination to work tostem attrition. The principle course of action in the language main-tenance and revitalisation programme was to be courses in andthrough Arbresh in the schools. This had the support of the major-ity of parents according to a survey conducted in 2001 (Derhemi,2002). The plan was to use Arbresh in the earliest years of primaryeducation as the language of school socialisation. This would buildupon a practice that already existed, as teachers habituallyprovided a bilingual bridge for Arbresh speaking children. Olderstudents were to have classes to allow them to acquire literacy inArbresh.

However, since Arbresh had never been a language of institu-tions nor of governance, it had not been developed as a writtenlanguage and there was no written standard. Until this was agreedupon, all educational initiatives faltered. Agreement provedimmensely difficult. There were two camps: those who believed thata standard should be developed for the 7000 Arbresh speakers inSicily and those who argued that it made more sense to adopt thewritten standard of Albania, which already existed and which gaveaccess to wider sources and contacts, and to an extensive literature(Derhemi, 2002; Fieldwork Italy, 2002).

SUE WRIGHT214

However, the ready made standard is exterior and slightly alien.Opponents to its adoption reasoned that there is little point instruggling to conserve a language, if what is conserved is not thegroup’s language but another, which, although close in some ways,is alien in others. They argued that importing standard Albanianwould introduce another idiom and that adopting it would create asituation of double diglossia for Arbresh speakers.

The alternative solution for codifying and standardising Ar-bresh also proved problematic. The difficulty stemmed from theway the project was eventually assigned to a small group ofscholars and poets and their subsequent failure to consult widely(Fieldwork Italy, 2002). The result is a highly literary standardthat was introduced top down to speakers. The few school textsproduced in it so far have been ill received. Teachers report thattheir pupils found the language difficult to understand and didnot relate it to what they actually spoke (Derhemi, 2002; Field-work Italy, 2002). The controversy, the lack of commitment tothe form and the subsequent slow rate of progress in producingtexts in it is one key reason for the slow implementation of theeducational initiative.

In Sicily, the Arbresh encountered the fundamental problemdiscussed above. Planners need to deal with language as systemin order to introduce Arbresh into institutions and promote itsacquisition in education. Thus, it was language as system notlanguage as practice and on-going dialogue that policymakers setout to protect and promote in these initiatives. This was not,however, what Arbresh speakers had hoped for. They wantedlanguage renaissance so that their practice could be enshrined asa language of power. The imposition of a norm, even thoughonly slightly different from practice, was experienced as alien-ation. Recent research shows that the home continues to be theprincipal site where Arbresh is used and that, despite continuingpositive attitudes towards the language, it has not become oneof the languages of the Piana public space (Derhemi, 2006), andmay even be in further decline despite having gained the legalframework for support.

Ladin

Ladin speakers live principally in five valleys of the Dolomite Alpsin the north of Italy. There are approximately 30,000 members of

the right to speak one’s own language 215

the Ladin speaking community7 who speak six (in some opinions,seven) distinct varieties of the language (Schmid, 1998). The areahas only acquired good communication systems in the recent pastand the relative isolation of the communities in separate valleysallowed the different varieties to develop and survive.

Since 1989, Ladin has been an official language in the provinceof Bolzano/Bosen, alongside Italian and German. In 2001, newrights for Ladin speakers were implemented in Trento province fol-lowing the 1999 legislation.8 Pressure for a single standard hascome from the provincial authorities who, in a multilingual situa-tion, have seen no good reason to accept variety in Ladin, whenboth Italian and German speakers of the Trentino-Alto Adigeregion, who also have their own distinct regional varieties, havetraditionally accepted the standard languages in their contacts withlocal government (Schmid, 1998).

Schmid recounts the debates that have taken place in the pursuitof a standard acceptable to all. As the differences among the dia-lects are quite significant, extensive accommodation is necessary ifthere is to be interdialectal comprehension. He recognises that sinceLadin competes in the local public space with two major Europeanstandard languages, it would be unlikely that Ladin could survivein any official role without some agreement on a norm. He suggeststhat the solution could be a Dachsprache, a variety which couldexist alongside the different dialects as a koine. He argues that theDachsprache cannot be one of the varieties, even that of the mostnumerous groups, since it would be rejected by the others. It can-not be a Mischsprache (a mixture of terms from all the varieties)since this would cause dissension at the micro level of choice anddecision rather than at the macro. Furthermore, he notes that theLadin speaking area is extremely difficult to delimit, since, as in alldialect continua the various particularities of the dialects arenot regular and congruent and many ‘Ladin’ features appear inneighbouring dialects.

Schmid, a German-speaking linguist, well known for his work inpromoting Romansh, a language very close to Ladin on the dialectcontinuum, suggests that the experience of the German-speakingworld permits optimism. The German standard (Hochdeutsch) isaccepted across the German dialect continuum as the acceptable

7 Only the province of Bosen/Bolzano takes a language census. These figures are

from estimates in Schmid (1998), Beninca (1998).8 Only in the province of Belluno are Ladin speakers unable to use their language

extensively in public domains.

SUE WRIGHT216

written form of spoken forms that differ widely. Schmid himselfhas produced a Dachsprache with as much ‘elemental’ Ladin in itas possible so that all speakers should be able to see how it relatesto their own version. To do this he has returned to the Latin rootsof the language and made his choice based on historical linguisticsas well as present usage. This is, of course, the ultimate perfect sys-tem. An outsider, a professional linguist, constructs a language thatno-one actually speaks by adjudicating among dialectal choices andwith reference to Latin roots.

Whether Ladin standard will be widely used or not remains tobe seen. It is logical and it is ‘fair’ as far as that is possible, but itis not part of ‘the continuous chain of other utterances which, ineffect, constitute the continuity of human consciousness’ and whichis what speakers wish to protect.

The Italian State, Language Rights and Standards

The history of language rights in Italy exemplifies how languagerights are linked to standardisation. A list of the languages that thestate has been prepared to recognise reveals this. Seven out of the12 languages recognised by Law 482/99 (Albanian, Catalan,Croatian, French, German, Greek, and Slovene) could haverecourse to written standards developed as part of their status asnational/official languages in other states. The language varietiesthat the state is not yet ready to recognise are those which theItalian centre considers dialects of Italian and which it argues canbe served by the Dachsprache of standard Italian (Coluzzi, 2004).

France: the Jacobin Tradition

France differs from Italy in that it has a history of muscularcentralisation and the institutions of the state have made rigorousattempts to mould citizens into a linguistically and culturallyhomogenous group. However, in 1999, the socialist governmentdeparted from traditional French Jacobinism and signed theCharter for Regional or Minority Languages. In swift reaction, theConstitutional Council ruled that this was unconstitutional, statingthat any recognition of the other languages of France could not be

the right to speak one’s own language 217

reconciled with the status of French as national language.9 TheCharter has not been ratified and there are currently no plans to doso. Public reaction expressed in the press at the time (andconfirmed the following year in the debate about state funding forBreton-medium Diwan schools) showed that the majority in Francecontinue to subscribe to the French republican tradition of ‘oneterritory, one people, one language’.

The result of this muscular centralism is that the situation ofminority language speakers in France differs radically from theposition of minorities in France’s neighbours. State support is mini-mal or non-existent and there is a marked lack of interest in theminority languages among even those whose families were speakersas recently as two or three generations ago.

Occitan

This lack of interest is particularly interesting in the case ofOccitan, because the language possesses an extensive literature andan imposing heritage. If Occitan speakers wanted to celebrate theirlinguistic and cultural specificity there is an enormous legacy fromwhich to draw. If Occitan elites wanted to mobilise this large popu-lation it would be possible to present ‘Occitania’ as victim, a siteof French invasion and occupation, and subsequent economicneglect. However, only a tiny minority of the southern French haswanted any kind of autonomy, however mild, and only a slightlylarger proportion has shown any enduring interest in regionalparticularity and heritage.

The appellation, Occitan’ would not be recognised by many inthe Occitan speaking groups that have survived. They describethemselves as speakers of Languedocien, Gascon, Provencal, Lim-ousin, Vivaro-alpin or Auvergnat. Occitania is not and never hasbeen a linguistic space with a recognised centre. Occitan is a set ofrelated dialects, and, unsurprisingly, has no single agreed standard.Occitan is a scholarly term, employed to describe the part of theRomance dialect continuum bounded by Catalan and Aranese inthe west and Piedmontese and Ligurian in the east.

There is, however, a major difference between this RML caseand many others, in that Occitan has existed as a written language

9 It may well have been responding to pressure from the President on this matter

(Wright, 2000b).

SUE WRIGHT218

and a language for literary production for one thousand years.Poets in the various dialects of Occitan from the 11th century pro-vide some of the earliest examples of European vernacular litera-ture.10 Occitan continued as a pan-European language of literatureto the 13th century with an audience well outside its homeland.After the incorporation of the various kingdoms of the south intothe French state, however, the language disappeared from publicdomains and there was a move to French in the towns and amongthe elite. Occitan medium literature waned.

In the 19th century, at the height of interest in ethnic originsand national particularity, there was a revivalist movement. TheFelibrige, led by Frederic Mistral, differed, however, from mostnationalist movements in having little or no political agenda. Thegroup was only interested in the cultural revival of the written lan-guage. It promoted an orthographical system based on that ofFrench, adapted to the distinctive Provencal of Arles and Avignon(Jourdanne, 1897). Unsurprisingly, there was a considerable gapbetween the scholarly and literary Provencal in which the Felibrigewrote and the various spoken practices of the whole Occitan area.This was not an issue; the Felibrige had no programme for the pro-motion of standard Provencal among the population as a whole.

In reaction to the elitist and regionalist position of the Felibrige,groups of activists in Limousin and western Languedoc advocatedthe adoption of a koine, usable throughout the area. Interestingly,although their motivation was more inclusive, their decision toreturn to the medieval language for their standard actually meantthat the written language they promoted was in some ways evenless accessible to the mass of speakers than Mistralien Provencal.The division between the two movements continued into the 20thcentury and it was in the very recent past that a rapprochementcame about.

Although this state of affairs seems inauspicious for the preser-vation of the dialects on the Occitan continuum, there is actuallyone positive consequence. As there is no general support for theadoption of Occitan in institutions, in governance and as a mediumof education, there is no need for a standard written language foruse in such forums. The situation has allowed all those who dowish to maintain their language to maintain the language they

10 Guilhem de Poitier, Jaufre Rudel, Marcabrun, Bernard de Ventadour, Ber-trand de Born, Arnaud Daniel, Guiraut de Borneil, Peire Vidal, Raimbaud d’Or-

ange, Arnaud de Mareuil etc (Dupuy, 1972).

the right to speak one’s own language 219

actually speak and the different forms of Occitan are tolerated.Language variety is explicitly catered for in the few settings whereOccitan is employed in public space: the daily newspaper, LaMarseillaise, prints a weekly page in Provencal and mentions infootnotes where practice is divergent for Languedocien and Gasconspeakers; the weekly news magazine, La Setmana, prints articles inany and all of the varieties; where students opt to study Occitan asa subject in secondary schools, the programme can accommodateto difference and some textbooks (e.g. Oc-ben!) are adapted to thelanguage practice of the region. The association, Leis Amics deMesclum, organise an annual literary competition Escriure enLenga d’Oc. They explicitly invite submissions in any and everyvariety of the language: ‘Toutes les graphies sont admises’.

Most interestingly, some email and website communicationseems to be written in hybrid varieties, i.e. texts exhibit features ofmore than one dialect (Wright, 2005, 2006). There is evidence thatOccitan is sometimes written in email, chat-room communicationand even websites according to French orthographic rules, presum-ably by those who have not acquired literacy in formal settings(Wright, 2005, 2006). There is a conscious rejection of purism. Onecan find numerous examples of peer pressure to maintain standardFrench on French medium websites (e.g. attack on form as well ascontent on the discussion forum of the Conseil superieur del’audiovisuel http://www.csa.fr/outils/forum), but this seems to betotally absent from the Occitan medium sites. Rather the invitationis inclusive and sites urge participants to communicate as they wantand can (e.g. http://www.cyberbougnat.net; http://www.occitan.skyblog.com).

The trade-off seems clear. Where a language becomes a languageof power in any way (the language used in interaction in demo-cratic institutions and in bureaucracies and the language spreadthrough a state run education system) the cost is acceptance of lan-guage as system, a codified, stable written standard, that may notentirely reflect the practice of those designated as its speakers.However, where the speakers of a language have minimal linguisticrights (i.e. the right to maintenance in private and as expression ofcultural heritage) differing practices are tolerated. This is, of course,of little worth if language shift occurs anyway, because the lan-guage does not have cultural capital, in Bourdieu’s sense ofeconomic value and political advantage.

SUE WRIGHT220

Conclusion

It seems reasonable to argue that language rights are always goingto be problematic: a constant tension will exist between the accep-tance of the heterogeneity of practice and the necessity of fixing aset of forms that will remain invariant across all domains. In thepast, this tension was resolved in the context of nation buildingand its homogenising processes. Where a group accepted the assim-ilatory imperative within nationalist ideology, they accepted lan-guage as system and converged linguistically. In the presentcontexts of RMLs and the desire of their speakers to have theirvariety recognised, this ideological framework is mostly missing.

The case studies clearly illustrate different aspects of the prob-lem. First, in a fairly small group such as the Arbresh, it might seememinently sensible to use the ready made standard of Albanian,which will save money and give speakers and readers access to alarger community of communication. However, if the desire of theArbresh is to use their language to underpin the homogeneity oftheir own small group and be a vehicle for its distinctive history andculture, then trying to impose an external standard is pointless. Alanguage similar to one’s own but with distinctions that make itappear alien may not be any more acceptable than a language thatis radically different. It is clear that if a group is concerned todefend its own language, it cannot do this except from within.

Second, in a situation where there is great intra-group diversity,the state may require some convergence among dialects to achievea widely accepted written standard to be employed in state bureau-cracy, in the courts and in education. This will mean that somegroups judged to be on the periphery will be pressed to accept lin-guistic assimilation to a putative centre. Now, if one is going toassimilate, one may prefer to assimilate to the national languagerather than to the standard devised by a neighbour.

Third, where the solution to diversity has been a standard writ-ten language devised by scholars and intellectuals from outside thecore community of traditional speakers, there have also been prob-lems. Even when the intellectuals in question come from communi-ties where the language is part of the heritage, but where use isdisappearing or has disappeared, the effects for the core group ofspeakers may be alienating. The latter are (stereo)typically rural,poor, old and illiterate in their mother tongue. The aspirationalspeakers, who acquire the language as a second language are

the right to speak one’s own language 221

(stereo)typically young, educated and politically literate. If theintellectual activists, the educated urban speakers undertakethe work of standardisation, this often leaves the rural speakers onthe periphery of language practice once again, stigmatised by theassociation of their version of the language with poverty and pow-erlessness. Hoffmann (1996) and Green (1994) have described thisprocess as one that puts minority speakers in a situation of doublejeopardy.

The rights movement derived from the desire of groups to usetheir language on a larger stage. Given the difficulty in deciding onwhat is to be protected even in small, demographically concen-trated groups, it remains to be seen how the right to maintain one’sown language, now framed in international law, will develop. Dolanguage rights depend on the acceptance of standardisation andthe concept of language as ideal system in the mould of nationalistlinguistics or will the political, technical and intellectual inventionsand developments of the 21st century provide new frameworks andstrategies which will make it possible to respect and conserve theheterogeneity of language practices in which local identities arerooted.

ACKNOWLEDGEMENTS

This article is a reworking of a paper originally given at the 2004Conference on Endangered Languages in Barcelona. My thanks toall those who have helped me revise and refine the ideas, in partic-ular the Language Across Borders research group at the Universityof Portsmouth. The mistakes remain, of course, my own.

REFERENCES

Austin, J. L. (1962). How to do Things with Words. Oxford University Press.

Barthes, R. (1968). The Elements of Semiology. New York: Hill and Wang.Beninca, P. (1998). Dolomitic Ladin. In G. Price et al. (Eds), The Encyclopedia of the

Languages of Europe. Oxford: Blackwell.

Bourhis, R. (2001). Acculturation, language maintenance and language shift. InJ. Klatter-Holmer & P. Avermaetvan (Eds), Theories on Maintenance andLoss of Minority Languages. Munster: Waxmann.

Bouquet S. & Engler R. (2002). Ferdinand de Saussure: Ecrits de Linguistiquegenerale. Paris: Gallimard.

Chomsky, N. (1968). Language and Mind. New York: Harcourt Brace.

SUE WRIGHT222

Coluzzi, P. Regional and minority languages in Italy. Mercator Working paper 24.Available http://www.ciemen.org/mercator/.

Coupland, N. & Jaworski A. (2001). Discourse. In P. Cobley (Ed.), Semiotics andLinguistics (pp. 134–148). Routledge.

Derhemi, E. (2002). The endangered Arbresh language and the importance of

standardised writing for its survival. The case of Piana degli Albanesi, Sicily.MOST Journal on Multicultural Societies 4(2).

Derhemi, E. (2006). Features of dysfunctional attrition in the Arbresh of Piana degli

Albanesi. International Journal of the Sociology of Language 178.Derrida, J. (1981). Positions. (trans. A. Bass). London: Continuum.Dupuy, A. (1972). Petite encyclopedie occitane. Montpellier: Saber.Eurydice. (1998) The Italian Education System. Biblioteca di documentazione

pedagogica. Unita italiana di Eurydice.Foucault, M. (1969). L’archeologie du savoir. Paris: Gallimard.Green, J. (1994). Language status and political aspirations: the case of Northern

Spain. In M. Parry et al. (Eds), The Changing Voices of Europe. Cardiff:University of Wales Press.

Halliday, M. (1977/2003). Ideas about Language. In J. Webster (Ed.), Collected

Works of M. A. K. Halliday, Vol. 3, (pp. 92–115). London: Continuum.Harris, R. (2001) Linguistics after Saussure. In P. Cobley (Ed.), Semiotics and

Linguistics (pp. 118–133). Routledge.Heidegger, M. (1993). The way to language. In D. Krell et al. (Eds), Basic Writings.

London: Routledge.Hoffmann, C. (1996). Twenty years of language planning in contemporary Spain. In

S. Wright et al. (Eds), Monolingualism and Bilingualism: Lessons from Canada and

Spain (pp. 59–91). Clevedon: Multilingual Matters.Joseph, J. (1995). Saussurean tradition in linguistics. In E. Koerner & R. Asher

(Eds), Concise History of the Language Sciences (pp. 233–239). Oxford: Elsevier

and Pergamon.Joseph, J., Love, N. & Taylor, T. (2001). Landmarks in Linguistic thought II.

London: Routledge.

Jourdanne, G. (1897/2000). Histoire du Felibrige. Nimes: Lacour/Redivida.Kristeva, J. (1986). The Kristeva reader. In T. Moi (Ed.), Columbia University Press.Kymlicka, W. (2001). Politics in the Vernacular. Oxford University Press.Lamb, S. (2004). Language and reality. In J. Webster (Ed.). London: Continuum.

May, S. (2001). Language and Minority Rights: Ethnicity, Nationalism and thePolitics of Language. Harlow: Longman.

Morris, P. (Ed.) (1994). The Bakhtin Reader. London: Edward Arnold.

Saussure, F. (1916). de Cours de linguistique generale. Paris: Payot.Schmid, H. (1998). Wegleitung fur den Aufbau einer gemeinsamen Schriftsprache der

Dolomitenladiner. Istitut Cultural Ladin.

Skutnabb-Kangas, T. (2000) Linguistic Genocide or Worldwide Diversity and HumanRights. Lawrence Erlbaum.

Voloshinov, V. (1994). Marxism and the philosophy of language. In P. Morris et al.(Eds), The Bakhtin Reader. London: Edward Arnold.

Wittgenstein, L. (1953). Philosophical Investigations. Oxford: Blackwell.Wright, S. (2000a). Community and Communication: The Role of Language in Nation

State Building and European Integration. Multilingual Matters: Clevedon.

the right to speak one’s own language 223

Wright, S. (2000b). Jacobins, Regionalists and the Council of Europe’s Charter forRegional and Minority Languages. Journal of Multilingual and Multicultural

Development, 21(5), 414–425.Wright, S. (2005). ‘L’occitan et la Toile’. Paper given at Centre de Recherche et

Documentation. Nice, October.

Wright, S. (2006). Languages and the Internet: Publishing in Lesser-used Languages.Language and Politics 5.2.

SUE WRIGHT

School of Languages and Area StudiesUniversity of PortsmouthPark Building, King Henry lst StreetPortsmouth, PO1 2DZUKE-mail: sue.wright@port.ac.uk

SUE WRIGHT224