The Future of Romani: Toward a Policy of Linguistic Pluralism

11 March 2005

Yaron Matras¹

Executive Summary

The past decade has seen the emergence of linguistic pluralism in the use of Romani in institutions such as media and education: language codification is primarily regional, targeting a regional audience. Regional initiatives are autonomous and implement their own solutions. At the same time, an international network of Romani language codification activities is emerging – through meetings, correspondence, exchange of publications, and via the internet. The question facing agencies engaged in language policy is how to pursue networking and international collaboration, while taking into account the de-centralised achievements of the past decade. The practical way forward is to adopt linguistic pluralism as a policy: to support regional initiative and creativity, while also strengthening international networking efforts and exchange. Users of written Romani should embrace the idea that different forms of the language may be used in different contexts, and that codification can be flexible and oriented toward practical communication, rather than rigid, serving as a symbol of loyalty to a particular Standard. Collective ownership of language will thus encompass a web of language varieties, and not just one single form of the language. Such a policy fits the specific Romani situation of a trans-national minority with dispersed, regional centres of cultural and public life. It is also suitable for the young generation of language users, who are accustomed to trial & error, individual creativity and flexibility in their use of written language in new communication technologies such as text-messages, internet chat-rooms and email.

Language Rights as Human Rights

In recent years, especially since 1990, the loose network of activists and initiatives known as the 'Romani movement' has managed to raise awareness among national governments and international institutions to the needs of the Roma as an ethnic minority. It has placed Romani human rights on the agenda of national and international politics. Countless resolutions by multilateral organisations such as the Council of Europe, the UN Human Rights Commission, the OSCE, and others, and various governmental initiatives testify to this achievement.

As in the case of other ethnic minorities, Romani human rights have two principal dimensions: First, the protection of the right of Roma to have equal access to the opportunities that society offers to all its other citizens. Second, the right to exercise control over a domain of cultural activities that are particular to the minority; in other words, the right of Roma to run their own cultural affairs.

Language is perhaps the most conspicuous of cultural assets which a minority may aspire to manage and develop on its own. This has long been acknowledged in the context of international debates on human rights: While discrimination on the basis of language is a violation of the Universal Declaration of Human Rights, the UN Convention on the Rights of the Child (1989) goes further and protects the right of the child belonging to a minority to use his or her own language. The right of minority communities to use their own languages in private and in public, and to develop their own languages in various spheres of public life, is anchored in the UN Declaration of the Rights of Persons Belonging to National or Ethnic, Religious and Linguistic Minorities (1993) and in greater detail in the European Charter for Regional or Minority Languages (1992). There is thus general awareness that support for minority languages is an integral part of the package of securing the human rights of a minority.

The need for measures in support of the Romani language has been mentioned explicitly in numerous international resolutions². One of the most frequently cited is Council of Europe Recommendation 1203 On Gypsies in Europe (1993), which calls for a European programme for the study of Romanes (Romani). In several countries, including Macedonia, Finland, Sweden, and Austria, there is at least some form of legislative or even constitutional recognition of the Romani language. Several other European countries have been pursuing policies of active encouragement and support of pilot projects engaged in teaching and broadcasting in Romani.

The Challenge of Language Planning

What do we envisage when speaking about the protection of minority languages? The European Charter for Regional or Minority Languages lists several concrete implementations of recognition and protection:

• General safeguarding, by promoting respect toward the language and encouraging its use in private and in public;

• Facilities to teach and to learn the language, for both speakers and non-speakers;

• Inclusion of the language in the education curriculum, especially at primary level;

• Provisions for pre-school education in the language for pupils whose families request it;

• Promotion of the study and research on the language, and teaching of the language at secondary and tertiary levels;

• Provision of translation facilities and interpreters for the language in public services;

• Provision of radio and television broadcasting in the language, and support of newspapers and the production and distribution of audio and audiovisual works in the language.

The task of governments in relation to these challenges are first of all to enable such activities (for instance by taking decisions on the shape of the school curriculum); next, to support them financially (by paying for the production of resources, or hiring and training personnel); and finally, to adopt a policy that encourages the use and development of the minority language. However, there is a further challenge in cases where, as in Romani, there is no tradition of using the language for the functions outlined above, and no pool of personnel – teachers, authors, broadcasters – or of resources – teaching materials, books, films – to enable to introduce the language into the relevant domains. Such situations call for language planning – a targeted effort to re-shape the language and to equip it with the necessary functions: a writing and spelling system, and a technical vocabulary to cover institutional domains of use.

Traditionally, language planning is viewed as involving two activity domains. The first is primarily social-political, and targets the status regulation of the language. The second is more technical-linguistic, and targets the body of linguistic material or corpus of the language. Here, decision-making processes traditionally aim at resolving the following questions: Which form of the language (e.g. which dialect) should be used for writing and in education, the media, and other public functions? Which writing system should be adopted or designed for the language (i.e. how should the language be codified)? How can the vocabulary of the language be expanded to cater to its new functions? And finally, how can the decisions taken in relation to choice of variety, writing system, and vocabulary be propagated among the potential users of the language?

Obstacles Facing Language Planning in Romani

Romani is only one of many languages around the globe that does not have an established tradition of a single, Standard written variety. The absence of a Standard is sometimes confused with the supposed lack of 'a distinct Romani language' and the presence instead of numerous dialects. In fact, most European languages show dialectal variation, and the type of differences found between the various dialects of Romani is not at all unusual. The absence of a Standard language for use in cross-regional communication, in writing, or in institutions does however mean that there is no obvious choice of any single variety for the more public functions of an official or written language.

In some respects, Romani is indeed in a unique situation: It is dispersed among many different regions and countries across Europe and beyond. There is no single, accepted authority or agency that is, or could be, entrusted with taking language-planning decisions for Romani as a whole, much less so with implementing them in the various regions; the responsibility rests with individual governments, while codification activities are diverse and regionally based. Romani populations are all bilingual, and the respective state languages (and sometimes other minority languages) influence the individual dialects of Romani. This concerns both the internal shape of the language, especially the use of technical or institutional vocabulary, and its 'external' shape: The choice of writing system is often dependent, for reasons of convenience and accessibility, on the writing system of the respective state language.

The dilemma is therefore this: In order to protect and promote Romani language rights as human rights, there is a need to develop educational materials and media in the language, and to train teachers and writers. In the absence of an existing Standard written language, this cannot be done without language planning. However, there is no uniform concept on which to base language planning, and no obvious accredited or authorised body that could draft and implement such a concept.

Regional Codification in Romani: A Brief Overview

Nonetheless, Romani language planning is not a vacuum. Indeed, in the past decade or more there has been an upsurge of local and regional language planning activities across Europe. The reality on the ground is that of a dynamic, organic movement of language codification efforts that has yielded results in the form of regional networks of media, publications, and teaching resources, results that can no longer be ignored and must be taken into consideration in any global assessment of the state of the art of Romani language planning. I will present a brief survey of some of these activities³. It is a sample, with no claim to deliver a comprehensive description either of all countries, or of all relevant activities in those countries that are named.

• Czech Republic and Slovakia: Codification efforts began already in the late 1960s, and were revived after the political transition in 1990⁴ . Most written material is in the East Slovak dialect, but some in other varieties spoken in the country, including Lovari. The writing system is based on those of Czech and Slovak and features the diacritic symbols diacritic {š č ž} as well as {ď ľ}. Aspirates are expressed by {ph th kh čh} and {h} and {ch} are distinguished. Publications in Romani include original literature and poetry, educaation materials, several bilingual periodicals, some with their own websites, and at least one bilingual academic journal. Romani language instruction is carried out in various parts of the both countries at primary and secondary levels, and in the Czech Republic also at univeristy level.
• Hungary: Written Romani is usually based on the Lovari dialect. The writing system makes use of the symbols {ny ly gy} for palatals, as in Hungarian, but differs from Hungarian in the use of {sh ch zh} and the avoidance of vowel length distinctions. Aspirates are {ph th kh}. Publications in Romani include grammatical descriptions and language teaching materials, literary translations, and periodicals, some of which appear on the web. Romani is taught sporadically at primary and secondary level, and at least at one university. There are also regular television broadcasts in Romani.
• Romania: The Romanian ministry of education adopted Romani into the official curriculum in the early 1990s, and it is now taught at all levels in schools and at university. The ministry has produced and adopted a series of Romani language textbooks and other teaching materials for this purpose. The variety used is based to some extent on the Kelderash dialect, but attempts are made to form a ‘common language’ by incorporating elements from other dialects as well. The official curriculum adopts a writing system designed by Marcel Cortiade (Courthiade). It is a rather unusual alphabet, which features alongside the diacritics {symbols: {ations k/g, s/ts and t/d respectively in a series of grammatical endings – specifically: case endings of the noun and pronoun – thus: man-ge ‘for me’ is spelled {manqe} whereas tu-ke ‘for you’ is {tuqe}. There are also independent text productions in Romani, whose authors tend to use writing systems based on Romanian, with some adoption of more international conventions.
• Macedonia: A flexible form of literary Romani was proposed already in 1980 by Kepeski and Jusuf ⁵ , based on either the Džambazi (Gurbet) or Arli dialects, and using the South Slavic writing system in both its Roman and Cyrillic forms, the Roman variant featuring the diacritics {š č ž} and aspirates {ph th kh čh}. Similar principles were adopted by a standardisation conference in the early 1990s, and implemented in a series of publications, including the national census documents and periodicals⁶. Alongside printed periodicals there are several Romani websites based in Macedonia, some of them preferring the symbols {sh ch zh} to the counterpart diacritics. There are also regular television broadcasts in Romani.
• Serbia: Most publications appear to be in the Gurbet/Džambazi dialect, and follow the writing system of Serbian, either in its Cyrillic or, more frequently, Roman variant, featuring {š ž č} and {ph th kh čh}. Other publications include political journalism and biographical collections. A series of teaching materials have been prepared at the local level, drawing on the private initiative and often resources of individual teachers, and there are pilot attempts to introduce them into primary school curriculum. Courses on Romani are now being introduced at least at one university. There are regular radio broadcasts in Romani, and several Romani websites are based in Serbia.
• Bulgaria: Authors tend to use their own individual dialects in writing, and there are Romani language publications in such dialects as the Erli of Sofia, the Kalburdžu varieties of the Varna district, and the so-called ‘Drindari’ dialect of Sliven. There is now an overwhelming tendency to use the Roman script for Romani, sometimes adopting the diacritics {š nthologies and political journalism, and the bilingual periodical ‘Andral’. Sporadically there has been inclusion of Romani in the curriculum at primary school level, and the ministry of education has appointed a coordinator for Romani language instruction, serving schools across the country. Recently, a Romani language course has been launched at university level.
• Austria: In the early 1990s a project was launched with the aim of codifying the endangered Romani variety of the Burgenland district⁷. A survey was carried out by a team of linguists and community representatives, and the speaker population was polled in respect of their preferred writing system. As a result of the survey, a written form of the Burgenland Romani dialect was introduced, based on the German writing system. The project has received government support and has produced anthologies of traditional tales and biographical texts in books and audio CDs, language education materials including interactive computer games in Romani, comics, and two regular periodicals, one of them directed at children. Romani is now a regular part of the regional curriculum at primary and secondary school levels, and is offered at university level as well. Projects are under way to codify and develop similar materials for other Romani varieties of Austria, including Kelderash, Lovari, and Arli. Other publications include biographical anthologies in Lovari, a bilingual periodical in Kelderash Romani, and occasional local broadcasting in Romani.
• Russia: There has not been very intensive exchange between the Russian Romani community and other parts of Europe in recent years, but there is a Russian tradition of codification of Romani, based on the Russka Roma dialect and the Russian writing system in Cyrillic script, going back to the 1930s. Many of the older texts have been destroyed, but several hundred books and periodicals in Romani have recently been re-published on the web by the ‘rombiblio’ project.
• Finland: Codification efforts began here in the mid-1990s. Using the local Finnish Romani variety, a group of linguists and community representatives designed a writing system which makes use of the diacritic symbols {š } and the aspirates {ph th kh čh}, and introduces a new diacritic {□} to represent a unique sound adopted into the dialect from Swedish. As in Finnish, doubling of letters represents a long articulation of sounds. This writing system has served for the production of several educational textbooks and public information material. There is occasional teaching of Romani at primary and secondary school levels, and research on the language is being carried out at a state research institute. There are also teacher training activities and weekly radio broadcasts in Romani.
• Sweden: Here, the production of educational material has been ongoing since the mid-1980s. The groups catered for are the local Kelderash/Lovari speaking community, and more recent immigrants speaking Arli varieties. Some texts adopt the former dialects, others are written in a mixture of two or more varieties. The writing system is generally that adopted in central Europe, featuring the diacritics {š č ž ć} and aspirates {ph th kh čh}.
• Germany: Although Germany ratified the European Charter of Minority or Regional Languages and recognised Romani as one of its minority languages, recognition is restricted to the dialects of Romani that are spoken by German citizens. Paradoxically, the most established Romani association in the country, the Central Council of German Sinti and Roma, has opposed both the development of written materials in Romani and the use of the language in public life. Nevertheless, there are various local initiatives to develop both printed and audio-visual materials in Romani for inclusion in the primary school curriculum. The author supervised the translation of a series of primary and pre-school level booklets into three different dialects (Gurbet, Lovari/Kelderash, and Polska Roma) in 1996, using the central European type or ‘international’ writing system with {š č ž ć} and {ph th kh čh}

Though far from complete, this survey demonstrates firstly that Romani is gradually occupying a position in the public life of Romani communities, including periodicals, broadcasting, educational material and the school curriculum. It also shows that initiative is regional, and often local. Authors tend to write for an audience consisting of their immediate community, and language planners adopt solutions that can be implemented within the framework of their own region, or sometimes within the boundaries of the state. The resources that support these activities are also largely local: Sometimes the authors’ private resources are invested in codification activities, sometimes support is received from local, regional, or state authorities, and quite often it is distributed via NGOs to individuals from grants provided by international foundations and multilateral organisations.

The most distinctive feature of current Romani codification attempts is therefore their polycentric character. Calls such as those made by Romani activists I. Hancock⁸ or V. Kochanowski ⁹ to adopt a single dialect as a Standard are not being followed. Nor has M. Cortiade’s proposal for a uniform alphabet and an artificially constructed Standard, a proposal backed by a resolution of the International Romani Union from 1990¹⁰, gained any wide popularity apart from its use within the Romanian education system, as well as by a small number of individuals mainly in Serbia, Macedonia and Albania. Instead, the overwhelming trend is toward a ‘polycentric’ model¹¹, where various codification models coexist side by side, as legitimate and coherent concepts in their own respective contexts.

Although the various models are independent of one another, some global tendencies may be identified in the choice of writing system. First, most if not all codification models seek a kind of compromise between the writing system of the respective state alphabet, and the ‘international’ transliteration conventions adopted by linguists for the purpose of descriptive analysis of Romani dialects, which feature the use of {š tem – for instance in Sweden, Finland, Hungary, or Bulgaria – the greater the compromises that authors are willing to make in favour of the ‘international’ system. Parallel to the three diacritics with wedge accents {š č ž} and {ph th kh čh}. The stronger the clash between the state language and this international transliteration system - for instance in Sweden, Finland, hungary, or Bulgaria - the greater the compromises that authors are willing to make in favour of the ‘international’ system. Parallel to the three diacritics with wedge accents {š č ž} we find a second option, an ‘anglicised’ one, featuring {sh ch zh}. The latter has advantages especially in email and web communication, where diacritics are not always transmitted across different software platforms.

The fact that authors adopt compromises and show tendencies to accommodate to international solutions indicates that international networking does play a role even in regional and polycentric codification activities. This is becoming even more apparent in recent years, where many Romani periodicals and other publications have websites which are easily accessible and are read by an audience of readers outside their base country. Some Romani-language web-publications even incorporate contributions from other countries, written in different dialects and in a different writing system. Indeed, variations in writing conventions and choice of dialect are often found even within a single printed publication. Apart from the technical aspects of orthography, exchange and mutual accommodation is recognisable also in the spread of new vocabulary. Terms such as aement.

Email correspondence is probably the most powerful vehicle of written communication in Romani. Orthographic variation in emails is levelled due to the absence, by and large, of diacritics in most systems. On the other hand, email brings together writers of different dialectal backgrounds. By its very nature – as a loose, spontaneous, rapid, yet effective means of communication, both private and public – email supports a trial-and-error approach to writing: Writers use their own dialects, but respond to individual usages coined by their interlocutors. They experiment with terms and writing conventions without the fear of either embarrassment or sanctions of any kind, creating their own individual blends of what they might consider their ‘own, genuine’ variety and the ‘other, distinct’ structures to which they choose to accommodate. Emailing in Romani is thus a pool of linguistic diversity, and at the same time a powerful force of convergence.

The Meaning of ‘Pluralism’

One of the remarkable features of the polycentric language planning landscape in Romani is the absence of any overt competition. On the whole, those engaged in codification activities appear tolerant of the diversity of codification models, and although discussions and consultations are commonplace, there are few if any visible attempts to interfere with solutions and strategies adopted by others. The first conclusion to be drawn from this observation is that linguistic uniformity and the symbolism attached to it do not, for most Romani cultural activists, constitute an agenda item of high priority. Indeed, if we examine the historical circumstances in which Standard languages emerge, we find that they generally satisfy a quest for power – by imposing one single variety of the language on all users in the public spheres such as education, public services, and broadcasting; a quest for control – by rewarding those who adhere to the Standard, and imposing sanctions on those who don’t, usually via tests within the education system and the qualifications that it awards; and finally a need for a national symbol of unity, with which users of the language can identify and call their ‘own’. For the bulk of users of written Romani, none of these demands can be identified beyond the local or regional domain. Although political unity (in the sense of pursuing a common cause) is on the agenda of most associations and initiatives, most do not regard linguistic diversity as an obstacle to unity.

It may be useful at this point to return to the question of dialect differences in Romani, whose role as potential obstacles to mutual comprehensibility is often emphasised. In fact, Romani dialects form a continuum across Europe, with neighbouring dialects tending on the whole to be quite similar to one another. In addition there are of course those dialects that are spoken by communities whose ancestors emigrated into their present locations after the 18th-19th century, whose dialects are in some sense ‘insular’. On the whole, the dialects of Romani are quite homogeneous, having descended from the same ancestor language (which we call ‘Early Romani’) only about 600 years ago and having still been exposed to mutual influences since their dispersion. Basic vocabulary and grammar do not generally offer any barriers to mutual intelligibility. A greater obstacle are loanwords from the surrounding languages, which differ of course among the dialects. However, in oral conversation among people from different backgrounds speakers tend to avoid incorporation of a large number of foreign words, switching off code-switching, as it were, and paraphrasing many terms instead.

From the point of view of rhythm and phonology (so-called ‘accent’), which do impose difficulties on mutual comprehensibility, Romani dialects might be divided into three main groups: The dialects of south-eastern and central-eastern Europe (from Greece and Turkey to Hungary and Slovenia, including Moldavia) are all mutually intelligible in face-to-face communication, with little effort. The same can be said for the dialects of the Baltic areas and eastern Europe (Poland, Slovakia, Ukraine, Russia). Finally, the dialects of western Europe (Germany and neighbouring countries) form a coherent group, with Finnish Romani and some of the Romani dialects of southern Italy in a somewhat more isolated position. Face-to-face communication across any of these groups is more difficult and requires somewhat more adjustment and experience, but it is far from impossible.

Bearing this in mind, one can appreciate that written exchanges between writers of different dialects can certainly allow efficient communication, as long as three main conditions are met:

• The participants are willing to accept other forms that are not their own, and to insert mild adjustments into their own writing; in short, participants must be prepared to accommodate to the requirements of the situation.
• The use of loanwords must be kept to a minimum and instead paraphrases of terms, or well-known internationalisms must be chosen.
• The writing systems used by the participants must be similar enough for them to be able to decode the graphic representation of sounds and words.

The result of such exchanges in written Romani across different dialects can be compared with the reading ability of Scandinavians (speakers of Norwegian, Danish and Swedish), who can easily understand each other’s written languages, even if they have difficulties understanding some of the Scandinavian languages in oral face-to-face communication. Linguistic pluralism in Romani can thus be taken to mean three basic principles:

• Regional pluralism: Different forms of the written language can be used in different regions with no substantial obstacles to mutual comprehensibility, and so without constituting a hindrance to trans-national communication among Roma.
• Contextual pluralism: Individual users of written Romani may choose to use different forms and even different codification (writing) systems in different contexts. For example, certain writing conventions may be followed in the education system in a given country, while a different system may prevail in periodicals appearing in the same country; imported literature might follow a third set of conventions, and a favourite web site might have a fourth, while informal internet communication will allow individuals to choose their own preferences. Users can learn to switch among different systems, as required by the context.
• Functional pluralism: The idea should prevail that writing, especially in the present age, is there primarily in order to facilitate communication and the transfer of information, and as such it is in the first instance of functional use to individuals. Users of a written language should be allowed the flexibility to work creatively with language. This entails a free-enterprise approach to the use of language by individuals and groups, free from the control of power centres. Efficiency of communication ought to be the only sanction or reward that is associated with the choice of variants in either phonological shape, lexicon, or spelling.

One might contest that a model of linguistic pluralism along the lines suggested here is perhaps idealistic, but not feasible. I would argue that there are several factors that favour pluralism in written language in the contemporary situation – both globally, owing to the role of trans-national communication, post-modern attitudes, and new technologies; and with specific reference to the Romani experience:

Pluralism already represents the overwhelming trend on the ground, with written Romani showing regional codification with some international orientation. No unification effort will succeed in bringing dozens or even hundreds of authors and thousands of other users of written Romani under the control of one, single authority. And, conversely, no language policy that ignores or tries to bypass these pioneers of written Romani will have a chance to succeed.

A new generation of Romani intellectuals is exposed to various forms of the language, both oral and written, through encounters with other Roma at international conferences, through internships and training seminars in the NGO domain, and through regular email communication and text messaging. It is especially via the latter two media that writing, including trans-national correspondence, has acquired a new position in the daily communication patterns of individuals. This generation can accept, comprehend and make creative use of different forms and varieties of the Romani language.

These young Romani intellectuals, the future of any Romani literacy movement, belong to a global generation of creative and flexible users of written media, who are at ease in experimenting with different variants of the written word via internet chat rooms, emails, and text messages. To them, linguistic pluralism is not just a concept, but a day-to-day reality.

Arguably, linguistic pluralism is gradually having a global impact, making it the trend rather than an exceptional handicap. Even in languages where there is a firm and rigid tradition of a uniform Standard, a young generation of users is now taking the liberty to embrace more flexibility and functionality. Witness the mixture of UK and US spellings even in academic publications in English, not to mention the disappearance or otherwise random use of apostrophes in most informal writing in English. In Britain, a new method of teaching literacy – ‘Jolly Phonics’ – is gradually being adopted, which encourages pupils to experiment flexibly with writing conventions for several years in primary school in order to encourage confidence and creativity of expression and written communication; here, communicative function is placed above adherence to the formal norm. Or witness the freedom with which anglicisms are being incorporated into media-language in German, or the sudden appearance of apostrophes in informal written German (including advertising), or the confusion caused by the introduction of a spelling reform in German, immediately followed by its retraction from various public domains and media.

These and similar developments suggest that a Standard must not necessarily be interpreted in the narrower modern sense – as a symbol of the acceptance of the power and control of a central authority. Rather, ‘standard’ usage in its evolving, contemporary context can be taken to mean a network of options from which users in a particular context can pick and choose in order to sustain efficient communication. Arguably, in the absence of a centralised Romani political authority, and in view of the geographical dispersion and cultural blends that make up the diverse communities of Roma across Europe and beyond, nothing but ownership of a diverse set of norms and options would meet both the moral and practical expectations of the Romani population.

In the age of new communication and information technologies, where texts can be transferred instantly from one format into another, and search engines can deliver both precise matches and approximations, where applications can correct both spelling and style, and machines can provide crude but instant translations, there is arguably less need to impose regulation on the individual who engages in written communication, and even less of a need to insist on homogeneity of formats, styles and shapes. Moreover, to the extent that regional norms remain in place and cooperation is sought at the international level, networks can be formed to produce solutions for teaching materials or media, which can then be transferred easily into the respective regional formats for ground-level distribution.

Implications for Resource Development

The latter point means firstly, that it is certainly possible and desirable to pursue international networking for the production of texts and teaching materials in Romani, even if we accept the fact that operational centres of text production are regional and local; there is no contradiction between regional pluralism and international networking. Moreover, as our survey above suggests, the two go hand in hand. There is every reason to draw on a wider pool of talent, experience and expertise and pursue the development of language resources for Romani in an international context.

Next, there is a need for resources that will transmit a message of linguistic pluralism and help users of the language acquire proficiency and confidence in accessing different variants and making choices among them. The acquisition of literacy itself is best carried out in the language variety that the learner – child or adult – can call their own. But subsequent language teaching can and should incorporate strategies to acquaint learners with different forms of written Romani. Multidialectal teaching materials in Romani have already been produced and tested in Germany and the Czech Republic¹², and have acted as catalysts for pupils to develop respect and curiosity toward other dialectal variants. A central, electronic pool of teaching resources would allow teachers to have access to a range of materials, and to choose and adapt those that may be of use to them.

Finally, there is a need to base new language resources on new technologies, and to make the maximum of what technology can offer. Even simple programming at the level of word-processor macros can enable users to convert texts from one writing system to another. By incorporating professional programmers as well as linguists into the consultation process, procedures can be developed to facilitate the adaptation of texts to different regional and local dialects and spelling conventions, thus enlarging the pool of materials. The potential for wider distribution of texts, through format and style conversion and local printing and publishing-on-demand, is likely to create further incentives for writers to produce quality material in Romani.

Needless to say, this requires proficiency and consistency in electronic production of texts. It is vital to invest all the available resources to allow those who assume managerial and authorship roles in the production of teaching materials and other texts to undergo appropriate training in basic information technology skills, and to benefit from a pool of expert technicians and programmers. These expert technicians might be hired at one or several locations and be available for consultation by email and at occasional workshops, entrusted with the task of consulting a network of co-opted writers, authors, and publishers of Romani material.

Electronic dictionaries and other learning tools can offer users similar advantages and support flexibility and pluralism in writing conventions. This has been demonstrated already by Romlex, an international collaborative project based at the universities of Graz, Manchester, and Aarhus, with joint funding from the Open Society Institute and the Austrian Chancellary¹³. Romlex is an online multidialectal dictionary, covering 25 varieties of Romani and up to 15 target languages. As a resource that is committed to pluralism, it is both symbolic and practical in allowing the user to choose among numerous different dialectal variants when keying in a search word in order to obtain a dictionary definition. The resource also enables the user to access separate entries for new vocabulary, and to choose his or her own preferred spelling conventions in the key-in window, while the application searches the database for approximations. Unlike conventional dictionaries, Romlex is thus a ‘bottom-up’ resource, one that is defined by user needs, rather than by a wish to impose a uniform norm on the user. Romlex also offers a pool of dialect-to-dialect as well as Romani-to-target language dictionaries which can be printed and distributed on demand.
Even in the short term, taking into account poor access to electronic resources in many regions where Romani is already part of the primary school curriculum, the incorporation of new technologies into a close-knit international network devoted to the production of teaching and learning materials in Romani will have great advantages: Electronically produced manuscripts can be edited at a designated location for the benefit of collaboration partners across the continent, and converted into the appropriate regional variants upon demand. This would require

• an information network through which author participants can inform other (client) participants of the existence of a manuscript or text, and share this text in electronic form; and through which client participants can file a request to adapt that text to their own local format;
• a technical support team with the task of serving the collaboration network by archiving the material that is submitted, and carrying out style and format conversions upon request;
• a network of publishers, able to produce print-on-demand copies and distribute them in the relevant regions.

Such a resource pool seems desirable in the first instance for the production of much-needed teaching and learning materials, but could in principle be extended to other kinds of texts – translations, anthologies, or periodicals.

More linguistic research is needed in order to be able to identify the most relevant dialectal and orthographic variants. The Romani Morpho-Syntax (RMS) Database at the University of Manchester already contains detailed information on over 100 different varieties of Romani, a product of international collaboration among researchers, language students and local language enthusiasts. It allows swift linguistic comparisons among dialects and offer new prospects in dialect classification, for both academic and practical purposes. The information it contains can also be used to develop a programme that will convert texts from dialect into another. Close cooperation with both programmers and authors of teaching materials will enable to develop a benchmark for basic text compatibility. Texts that follow that benchmark – i.e., are coherent and consistent in dialectal form and choice of orthography and composed in a recognisable format, might then be converted at the push of a button.

Implications for Curriculum Design

Many issues pertaining to the Romani school curriculum are still at a very initial stage of planning and discussion. Among them are issues of principle, such as the usefulness of autonomous Romani schools, versus integration into mainstream schools. Some regard the former as an expression of cultural autonomy, while others view it as a form of segregation, and integration as a measure of success. A number of conditions apply however, regardless of the preferred education strategy:

• Romani education will continue to be bilingual, as Romani communities will continue to cherish the bilingual skills that have been an asset to them for many centuries, and as any form of education will seek to equip pupils with the skills to operate as comfortably outside of their own Romani environment as within it.
• Whatever the overall framework, teaching Romani requires a pool of trained teachers who are insiders to the Romani community, and have not only a general teaching qualification, but specific training in teaching Romani as a subject; teacher training in Romani language is therefore a top priority, and a curriculum for teacher training – at the national and possibly also international level – is an urgent necessity.
• Romani language teaching is not a uniform procedure, and so it cannot involve an unvarying curriculum. Rather, the teaching of Romani must be embedded into the relevant context of pupil profile, level, and overall teaching context and goals. Curriculum is therefore a plurality of possible activities; the challenge is to match the appropriate activities to the situation and goals.
  The latter point suggests that teaching materials must be carefully designed and selected to cater for particular types of school contexts. Some of the basic questions that must be addressed when designing a Romani language curriculum are these:
• Are the pupils native speakers of Romani, who use Romani in their family and perhaps also with their peers, or do they lack any Romani language skills and are expected to acquire them entirely via the school curriculum?
• Is Romani the medium through which literacy is first acquired, that is, do the pupils speak Romani and have their first encounter with literacy in their native language, Romani? Or are the pupils already familiar with the concept of literacy when they start taking up Romani as a subject, having already learned the basic principles of reading and writing in the state language?
• Is the aim of the programme to teach the Romani language as a subject – be it a foreign language, or a native language –, or is it intended also to teach general subjects (such as Romani history or culture) in Romani?
• Are the students children or adults?

Many of the teaching materials produced so far lack a clearly-defined target audience and learning setting. Yet defining those is crucial for the design of the programme and its success. Thus, pupils who can speak Romani and who use it actively in their families and communities will require training in reading and writing the language, but not in memorising basic vocabulary or grammatical inflections. Pupils who do not speak the language (whether they are of Romani origin, or Gaje joining the Romani class) on the other hand will need extensive practising of grammar and vocabulary.

Bilingual pupils who have already learned to read and write in the state language will require an introduction to the specific sounds and spelling conventions of Romani, but not to the principle of graphemic representation and syllable building as such. On the other hand, if pupils are monolingual or virtually monolingual in Romani at the age at which they enter school, alphabetisation (the acquisition of the basic principles of literacy) should be carried out in Romani – one cannot learn to read and write in a language that one cannot speak and understand! Forcing children to acquire literacy in an official state language of which they are merely semi-speakers is likely to be a cause for poor performance, alienation, and often subsequent segregation in separate custodial classes, creating a vicious circle of learning deprivation and social exclusion.

Teaching the Romani language is not the same as teaching various subjects in Romani, and although thematic learning widens the scope of language training, when contemplating priorities for the curriculum one must carefully select the aims of the programme, and design the materials accordingly. In this connection, it is vital that the pupils’ existing spheres of knowledge be taken into appropriate consideration. One of the great advantages of native language instruction, and the reason it is considered a ‘human right’, is that only initial instruction in the native language allows the child who is entering school to be able to draw a positive link between his or her cognitive skills which have been acquired before entering school, and the content of the school curriculum, and use the former as a bridge to tackle the latter. The imposition of a language in which the child is not at home right at the start of a child’s school career sets a demarcation line between prior knowledge and curriculum knowledge. In this respect, teaching materials aimed at children and adults must also differ in content and the skills on which they draw.

How can the concept of linguistic pluralism be integrated into the Romani language curriculum? Following the principle of native language alphabetisation, we must ensure that children have access to literacy classes in their native language – not just any form of Romani, but the variety of Romani that is closest to that spoken in the child’s native environment. This of course may create practical problems, in particular in areas where several different dialects of Romani are spoken. Such areas are not, however, innumerable, and a linguistic survey of ‘problem’ communities, if carried out by local teachers in collaboration with linguists, might be able to identify the principal varieties, and ensure that literacy materials are available for each.

Even in those situations where initial alphabetisation might take place in Romani, a transition to the state language will have to follow quite rapidly in order to ensure that children stay in touch with the general national curriculum. But Romani language instruction can and should continue. Given the paucity of reading materials available in Romani, some texts are likely to be selected from outside the immediate region, or even from outside the country. One advantage of an international network running a pool of recognised teaching resources is that materials could be authorised in advance by the respective education ministries, allowing teachers to select materials according to their immediate needs. Thus, gradually, the stage following the acquisition and training of literacy itself could include text samples written in other formats, in other spelling conventions, or in other dialects, representing perhaps a picture of the culture, folklore and lifestyles of Roma from other regions and even from other countries. Pupils will become accustomed to reading texts in other forms of Romani, and will become to some extent familiar with images of other Romani populations. Pluralism can, in this way, be a vehicle toward strengthening mutual interest, respect and solidarity.

The Role of Experts and External Support

In the programmatic outline that was presented in the previous sections, the key role belongs to those who are at the heart of the Romani language movement: the writers and teachers who cultivate the language and propagate its use. But outside this core of language pioneers there are also others who can offer their expertise and support. Let us first review a list of urgent items on the implementation agenda. We have identified the following needs:

• Official recognition of Romani language rights, and government support for curriculum design, media, translation and research;
• Training facilities and training curriculum for Romani language teachers.
• Development of curriculum concepts designed to match different types of teaching situations, and development of teaching materials in individual regions.
• Surveys of the bilingual and dialect situations in relevant regions, to support the choice of curriculum and the form of teaching materials.
• An international pool of teaching resources, ideally authorised or partly authorised by the relevant ministries of education, to be made accessible to teachers and curriculum officers.
• Training of teachers and those responsible for curriculum design in basic information technology skills, and networking activities to foster links and exchanges among them.
• Sharing of electronic resources and texts, and the development of new learning and teaching resources, with adaptations for individual regions.
• In pursuit of the latter, development of a benchmark for text compatibility, and the creation of a scientific and technical support team to facilitate exchange and adaptation of resources.

Linguists have a role to play in a wide range of activities directly related to this agenda. First, more research is needed into differences between the dialects of Romani, in order to be able to advise on the extent to which materials can be expected to be compatible or comprehensible to particular audiences. Such research would involve both the documentation of the structures of individual dialects (sounds, grammar, lexicon), and observation about the dialect and language repertoires used in individual communities. More observations and evaluations of the codification process in individual regions and locations are needed in order to assess the impact of media, teaching, and new technologies, and in order to be able to keep language policy in touch with developments on the ground. Of particular interest and need are first-hand observations of cross-dialectal communication, of which there are only very few studies; these can help assess the prospects and the obstacles to mutual comprehensibility. Training in basic concepts of Romani linguistics should form a part of any serious teacher-training curriculum, and linguists must contribute their part to the design of such curricula. Finally, the development of language support tools such as dictionaries and grammars, the bread-and-butter of applied and descriptive linguists, are vital to the enrichment of Romani language teaching and learning resources.

Alongside the professional expertise of linguists, backing by multilateral organisations and foundations is a crucial component in any Romani language network. Only those can offer the middle- to long-term financial and logistic support that is necessary in order to ensure continuity in the process of exchange of ideas and creative collaboration among teachers, education officials, academic and technical specialists, and only they can provide a supporting environment that is relatively free of the various pressures of local or national administrations, yet equally committed to improving resources, opportunities and participation at the local and regional levels.

Outlook

Romani is a trans-national language. For the benefit of its users, those engaged in drafting and implementing Romani-language policy must embark on a course of trans-national cooperation. They must form an organic network that will inspire and support, but not direct nor control. Their most immediate task is of a practical nature: to nourish the development and expansion of Romani literacy. In order to work in harmony with ongoing efforts and with the immense pool of talent and energy that are already involved in promoting Romani literacy, language policy-makers face a special challenge: They must disentangle language from ideological allegiances. Literacy must not be regarded as a mere rallying expression of loyalty toward a central authority, nor of the acceptance of a pre-fabricated, imposed set of norms. Rather, it must be viewed as a space that is open to negotiation between the participants in the communicative interaction; a space which the participants are able to shape according to their own needs and wishes. The goal of language education is to give users of language the skills they need in order to claim ownership over language as a means of expression and communication. Users of Romani have made a choice in favour of linguistic pluralism. It is the experts’ duty to support them in pursuit of their choice.

Endnotes:

1. Yaron Matras is Professor of Linguistics at the University of Manchester, author of Romani: A linguistic introduction (Cambridge University Press, 2002), and Editor of the journal Romani Studies. Correspondence address: School of Languages, Linguistics, and Cultures, The University of Manchester, Manchester M13 9PL, United Kingdom. e-mail: yaron.matras@manchester.ac.uk.
2. See P. Bakker & M. Rooker. 2001. The political status of the Romani language in Europe. Mercator Working Papers 3, Barcelona: CIEMEN (http://www.ciemen.org/mercator/pdf/wp3-def-ang.PDF).
3. For other surveys of codification practices in Romani see: Matras, Yaron. 1999. Writing Romani: The pragmatics of codification in a stateless language. Applied Linguistics 20, 481-502; Halwachs, Dieter & Ztreanu, Mihaela. 2004. Romani in Europe. Romani in Europe.
4. See Hübschmannová, Milena. 1995. Trial and error in written Romani on the pages of Romani periodicals. In: Matras, Yaron. ed. Romani in contact. Amsterdam: Benjamins. 189-205.
5. Jusuf, Saip, and Kepeski, Krume. 1980. Romani gramatika. Skopje: Nasa Kniga.
6. Friedman, Victor A. 1999. The Romani language in the Republic of Macedonia: status, usage, and sociolinguistic perspectives. Acta Linguistica Hungarica 46: 317-339
7. Halwachs, Dieter W. 1996. Verschriftlichung des Romani. (Arbeitsbericht 2 des Projekts Kodifizierung und Didaktisierung des Romani). Oberwart: Verein Roma.
8. Hancock, Ian F. 1993. The emergence of a union dialect of North America Vlax Romani, and its implications for an international standard. International Journal of the Sociology of Language 99: 91-104.
9. Kochanowski, Vania de Gila. 1995. Romani language standardization. Journal of the Gypsy Lore Society, fifth series 5: 97-107.
10. Courthiade, Marcel. 1990. Les voies de l'iimergence du romani commun. Études Tsiganes 36: 26-51.
11. See Hübschmannová, Milena, and Neustupný, Jií V. 1996. The Slovak-and-Czech dialect of Romani and its standardization. International Journal of the Sociology of Language 120: 85-109.
12. Jekh du trin " romanes, published in Hamburg by Verlag für pädagogische Medien (1996), is a series of twelve booklets in the Lovari/Kelderash, Gurbet, and Polska Roma dialects; Co uz umim, published in Prague by Fortuna publishers (1999), is a primer in East Slovak Romani, Lovari Romani, and Czech.
13. Romlex, and soon also accessible as http://www.romlex.org.