Introduction

1In the invitation to the "(Re) Mapping the Centres: Membership and State" conference, the organisers asked participating linguists to "ponder the historical imposition of a national language as well as the issue today of regional languages". Another question was also asked, namely, "can the purity of a national language be retained by legislation from above or does language evolve through the 'democratic voice'"?

2In this article, I shall approach these questions from both a sociolinguistic and a socio-historical point of view as they apply to Wales. I shall also try to demonstrate how the concept of national as opposed to regional language is a highly subjective one, linguistic attitudes in Wales being shaped by an extremely complex mosaic of socio-economic, cultural, ethnic, historical and ultimately ideological factors.

1 - Language and Ethnicity: the "Welsh Ethnic Revival"

3When considering the question of 'Welshness' and 'Welsh identity' sociolinguists have often taken these concepts to be expressions of ethnicity within the context of the so-called "Welsh ethnic revival". Ethnicity, as defined by Fishman, is 'peopleness':

a very mystic, moving and powerful link with the past and an energizer of respect to the present and future. It is fraught with moral imperatives, with obligations to "one’s own kind"... As such, it is language-related to a very high and natural degree (both overtly and covertly). Indeed this is to such a degree that language and ethnic authenticity may come to be viewed as highly interdependent. (1992, 70-71)

4This emphasis on language, the past and ethnicity is articulated in a very personal way by Ned Thomas in his book The Welsh Extremist:

Languages are very delicate networks of historically accumulated associations, and a thought in Welsh has innumerable and untraceable connections with the thought of past centuries, with the environment, with the scenery even, with one’s mother and father, with their mothers and fathers, with the moral and emotional terms in which the community has discussed its differences. (1971, 30)

5This kind of statement seems to reinforce Fishman’s (1977, 19) contention that, in many societies, language is taken to be a kind of biological inheritance and the loss of one’s language can be equated with the extinction of one’s kinship group and all it represents.2 If hue, this would explain the intense emotion which underlies much of the debate on the subject of "language death".3

6In his article, "Ethnic awakening in the First World", Bud Khleif proposes that the resurgence of ethnicity in Wales is not only a relatively recent phenomenon but also part of a world-wide movement. Furthermore, the fact that the rise of ethnicity coincides with the process of decolonisation would seem to be more than mere coincidence.

[It] can be regarded as a search for roots, for identity... for coping with issues of alienation in a mass society. The resurgence of ethnicity cannot only be understood as a tool for social mobility but also as a widespread quest for community, a search for authenticity in the face of the overwhelming forces of modern life that are thought to be conducive to depersonalisation, bureaucratisation, and unresponsiveness. (1978, 103-104)

7He cites the cases of Corsica, Quebec and Wales as examples of the re-emergence of "old non-state nations". He affirms that the common denominator linking all these groups is that they have "developed new rising elites after 1945, an assertive new middle class, anti-colonial in character, aware of its past, bent on consolidating its future" (ibid.).

8He believes the emergence of this new middle class to be the consequence of a shift from a "goods producing society" to an "information, or knowledge society". The university, he argues "has become the arbiter of class position and has gained the quasi-monopoly in determining the future stratification of society". The new "knowledge class", he goes on, "is heavily ethnic: pro-Welsh, Welsh-speaking Welshmen in Wales; pro-Quebec, Joual-speaking Frenchmen in Canada, (and) a selfassertive, post-1945 university-trained professional class of non-WASPs (Catholics and Jews) in the USA" (ibid.).

9Edwards (1985, 11) states that this collective sharing of ethnic identity is only a short step away from 'nationalism' which he takes to be a form of politicised ethnicity:

although nationalism and ethnicity share much in common, most importantly the sense of "groupness" or "peoplehood", they are not identical. The obvious difference, however, is not one of principle but of scale. Nationalism is an extension of ethnicity in that it adds to the belief that the "only legitimate type of government is national self government".

10In this sense, ethnicity is the driving force behind "Welsh cultural nationalism" (Nairn, 1986), a form of nationalism which is defined primarily through language, history and literature. Although this "awareness of the past" plays a primordial role in defining Welsh identity, it should not be viewed as a simple, unilinear process, but rather as a highly elaborate multidimensional phenomenon.

2 - Welsh History and Language

11Perhaps the most important point to keep in mind is the extraordinary richness and the great antiquity of Welsh historical tradition, both in its Latin and Welsh-language sources. The mere fact that it focuses on Welsh, or more broadly speaking, Brythonic antagonism toward the English nation4 is of critical importance to our discussion. Brinley Roberts (1975), in his study of the Welsh sources of Geoffrey of Monmouth’s Historia Regum Britanniae, writes that the central theme uniting traditional British (i.e. Brythonic) history since the sixth century is that of the violation of the sovereignty of Britain by the Saeson, that is, the Saxon English (see also Roberts, 1971). Dumézil (1968) has produced convincing evidence showing that the sovereignty theme is itself a pre-Christian, Celtic belief with parallels not only in Ireland but in other Indo-European societies, most notably Iran.

12Welsh historical tradition is dominated by the fact that what is today England was once Brythonic-speaking territory. The setting for the earliest Welsh bardic poetry, for instance, is more often southern Scotland or the north of England than Wales. Indeed, Cumbria (also Cumberland) is an older cognate form of the modern Welsh name for Wales itself – Cymru (from Primitive Welsh *Cum-brogi, meaning "land of the fellow countrymen"). The recovery of this lost territory from the English fills the pages of early Welsh (and Breton) vaticinatory literature from Nennius’s ninth-century Historia Brittonum (Roberts, 1975)5 and the tenth-century "Armes Prydein" (Williams, 1955) to Geoffrey of Monmouth’s twelfth-century "Vita Merlini" (Thorpe, 1966). Bardic prophesies that a Celtic king would once again reclaim the throne of England were commonplace in Wales until the fifteenth century and were fully exploited for political purposes by nume rous Kings of England, most notably by Henry Tudor (Parry, 1955).

13This tradition has conserved the names and deeds of scores of Brythonic heroes from regions long since Anglicised, the most famous of these, of course, being King Arthur himself. Many Welshmen today still carry names such as Caradog, Aneirin, Owain or Llewelyn, names which bear witness to this distant past. In modern Welsh, even towns such as York, London or Monmouth are still known under their ancient Brythonic names: Efrawg, Llundein, Monwy. Indeed, the Welsh name for England, Lloegr, was in already in use during the sixth century. The Anglicisation of Wales today, then, is viewed by an increasing number of scholars as the end of a process of language shift which began in England 1,600 years ago with the arrival of the first Germanic-speaking colonists, the Anglo-Saxons.6

14On the Welsh sense of history, Raymond Garlick (1982, 33), poet and former editor of the Anglo-Welsh Review, writes:

Welsh myth and legend recede into Celtic prehistory.... So long and unbroken a sweep of history not unnaturally gives rise to a tendency to look back (intensified for some, no doubt, by an unsatisfactory national present, and an uncertain future). Its continuity allows a telescoping effect whereby such figures as Magnus Maximus, Arthur, Glyndwr, such events as Camlan, Catraeth, Cilmeri, can be referred to quite naturally in contemporary contexts ranging from political discourse to pop songs.

15The past is an omni-present factor in defining Welsh nationhood where the link between territorial sovereignty, ethnicity and language is paramount. Gwynfor Evans (1976, 25), the former head of the Welsh nationalist party, Plaid Cymru, expresses this view in his book, Land of Our Fathers:

Of very few nations can it be said (as it can be said of the Welsh), that their territory has remained the same for most of the last three millennia. The probable result of this unusual Welsh situation is that the majority of the Welsh of this generation are descended from the people who inhabited the country when Ostorius and his legions came here in 47AD.7

16In this scheme, the place of the Welsh language is clearly central. Evans (ibid., 16) continues, "Down the ages the arts which received the most attention were those based on language. To a greater degree than in most other countries, the history of Wales is the history of its language".

17Paradoxically, it is the progressive Anglicisation of the Welsh gentry following the crowning of Henry Tudor which provoked the gradual disappearance of the class of bards who had been the guarantors of Welsh literary tradition. The inevitable consequence was that the Welsh language and culture increasingly came to be associated with the majority of those who continued to speak it, namely the gwerin, the Welsh peasantry. Moreover, the memory of the Welsh language as belonging to a broader Brythonic or "British"-Celtic family began to fade.

18The introduction of Protestantism into Wales, however, and the subsequent translation of the Bible into Welsh by order of Queen Elizabeth I, probably saved the Welsh language from extinction and provided it with a modern literary model. Indeed, one of the consequences of this remarkable event was that by the end of the eighteenth century three-quarters of the population was literate in Welsh thanks to the efforts of men such as Gruffydd Jones who organized travelling Bible schools throughout Wales (Clement, 1960).

19In spite of this impressive accomplishment, Welsh identity ultimately came to be identified with the working class, poverty and ignorance, especially in the eyes of the English. A British government study conducted in 1847, called in Welsh "Brad y Llyfrau Gleision" (the Treachery of the Blue Books)8, even went as far as to blame the Welsh language for the perceived backwardness and, particularly hurtful, the supposed immorality of the Welsh people. The consequence of this report was that English was to become the sole language of education at a time when most were monolingual Welsh speakers (Davies, 1993).

20The remarkable tenacity with which the Welsh resisted English linguistic expansion from the Reformation to the present has often been explained by the intimate association that has existed between the Welsh language and Protestant nonconformity. Even if the reality is more complex, this hypothesis fits neatly into the scheme presented in Ferguson’s classic article (1959) entitled "Diglossia" in which he demonstrates how weaker, less prestigious languages must be socially and functionally compartmentalized in order to survive. According to this view, most Welsh-speakers continued to speak their language, above all else, for religious reasons, the Welsh language being perceived to be the "language of heaven", the language which was imperatively to be used when communicating with the omniscient. English, on the other hand, was considered to be the language of the "fallen world" (Williams, 1989; Tudor Jones, 1973; Gruffydd, 1960). Although this view is on the wane, it can still be heard in some circles to this day.

3 - Anglicisation of Wales

21In his thesis Language, Identity and Education in a Liberal State: The Anglicisation of Pontypridd, Mid-Glamorgan, 1818-1920, Tim Williams (1989) stresses a rather unpopular view, in nationalist circles at least, claiming that the Welsh language "shrivelled in the air of freedom". In other words, he argues that it is the Welsh people themselves who voluntarily chose to abandon the Welsh language, not the British State which suppressed it.

22Be that at as it may, Williams’s position does highlight an ambiguous side of Welsh identity, particularly in Anglicised areas of urban South and East Wales. "Welshness" would thus have a plural nature: Wales may indeed be a "nation" but it was and is also a principality subordinate to the Crown. In other words, it is an integral part of the British State. The "profound Britishness of the Welsh", as Williams expresses it, was perhaps motivated by a desire to play an active part in the British Empire.9 Furthermore, the profound attachment of the Welsh to the royal family, particularly in the person of the Prince of Wales, is still widespread. As one elderly lady from Penclawdd, Dyfed, expressed it: "The Prince of Wales was ours and no one else’s!" (German, 1998, personal interview).

23Two conflicting conceptions of "Welshness" emerge as a result. Older Welshmen tend to consider Wales to be a nation, not necessarily in the political sense, but as a part of the United Kingdom, while increasing numbers of younger, well-educated10 Welshmen seek to redefine Wales as a "nation" fundamentally Celtic in language and culture.

4 - A Sociolinguistic model

24At this point, I should like to address the question of the role of language in this ideological framework. A sociolinguistic model proposed by Le Dû and Le Berre (1994) is helpful in structuring the discussion of national as opposed to regional languages, or for our purposes: State versus non-state national languages. Originally conceived to describe the linguistic situation in Brittany, it could provide some interesting insights into the situation in Wales and the United Kingdom as a whole.

4.1 - The Norm

25Simply put, Le Dû and Le Berre argue that, in sociolinguistic terms, there are three basic levels of language, the first being the "Norm", the linguistic model sanctioned by the State. It is equated with what they call "la langue de disparité", a form of language associated with the governing political and intellectual elite.

26One of the unfortunate consequences of the Norm’s high status is that it is frequently associated with prescriptive notions of linguistic "purity" and "correctness". This approach towards language is highly destructive, often insulting and, linguistically speaking, untenable, in that it pre-supposes that minority languages or dialects are "impure" and that they have somehow "degenerated" from an original model embodied by the literary model. The idea is, nevertheless, commonplace... both among those who master the Norm and particularly by those who do not. Inevitably, this situation results in linguistic insecurity on the one hand and linguistic prejudice on the other, prejudice which often leads to the establishment of very real social and economic hurdles which are particularly difficult to overcome.11 Attitudes towards language are governed above all by extralinguistic judgements on the people who use given forms of speech.

27Translated into British terms, the Norm of the British State and, very significantly, that of the English nation, is RP/standard English. It is a striking fact, however, that no more than 3% to 5% of the English population use the RP accent naturally (Tmdgill and Hughes, 1979; Trudgill, 1994) leaving us to ponder whether a prescriptive definition of correctness is truly warranted nowadays. Furthermore, in a study of the speech of thirteen Welsh politicians, Nicolas Coupland (1990) demonstrates that the RP accent is still intimately associated with Tory party members and, by extension, the English social elite, whereas politicians having the strongest Welsh accents are inevitably Labour politicians or members of Plaid Cymru, the Welsh Nationalist Party.

4.2 - The local "Standard"

28The second category defined by Le Dû and Le Berre is the local "Standard", a written model normally designed by ecclesiastical authorities for preaching to a given linguistic community. In the case of Wales, the local Standard is represented by the form of literary Welsh originally codified in the sixteenth century by the translators of the Welsh Bible and modernized during this century. A spoken form of the Welsh literary language is clearly gaining in strength at the expense of local forms of Welsh dialect. If one accepts this scheme, it explains why Welsh, although clearly a national language, has, until recently, suffered from its inferior status with regard to the English Norm – the official language of the State.

4.3 - The Badumes

29The third category is called "Badume" or "Basilect"12, strictly oral, highly variable, heavily stigmatised forms of speech which were widespread in the tiny rural communities of western Europe until the nineteenth and early-twentieth centuries. In Welsh terms, the situation is complicated by bilingualism (and even bidialectalism). Two classes of Badume-speakers can thus be distinguished: a) speakers of Welsh Badumes and b) speakers of Welsh-English13 Badumes.

a - The Welsh Badumes

30The first are speakers of local Welsh dialects. Although extreme dialectal variation within Welsh is not as commonplace as it once was (Beth and Peter Thomas, 1989), Welsh dialects tend to be neglected for prescriptive reasons. Yet, the linguistic insecurity this breeds is one of the reasons why educated Welsh speakers (and this includes those who have learned Welsh) can sometimes find themselves excluded from conversations involving local colloquial speakers (Bellin, 1994). This can be explained by the fact that some standard Welsh speakers tend to favor a linguistically unified form of speech characterized by neologisms and a refusal to use English borrowings. As Tim Williams (1989, 15) bluntly puts it, native speakers identify this form of Welsh as "artificial, non-dialectal, and alien, setting up barriers to acceptance and use".

b - The Anglo-Welsh Badumes

31The second class of Badume is made up of the local Anglo-Welsh varieties of English spoken throughout Wales. These have been well documented by David Parry (1977, 1979) and Robert Penhallurick (1991) of the University of Wales, Swansea. Although many varieties are devoid of overt Welsh influence, particularly along the English border, South Pembrokeshire and the Gower Peninsula, most, particularly in the west where Welsh is still widely spoken, are c haracterized by a considerable Welsh substratum.

Conclusion

32In the sociolinguistic model described above, the relation between language as a symbolic marker of identity14 and the social status of the speaker stands out clearly. As Frankenberg (1971, 91 quoted in Khleif, 1978) expresses it,

The industrial and agricultural experience of the past confronted Welsh-speaking, non-conformist, liberal wage-earners and tenant farmers with Anglicised, Anglican, Tory landlords and employers.... English and Welsh were almost synonymous with landlord and tenant or capital and labour.

33As has been repeatedly demonstrated (Giles et al., 1970, 1972, 1976; German, 1996), the acceptability of a given accent or dialect has relatively little to do with linguistic criteria per se and much to do with judgements about the cultural and social model associated with the speakers who use it.

34Assuming one accepts Le Dû and Le Berre’s sociolinguistic model, it would appear to apply not only to Wales but, with nuances, to many other regions of Britain and the Western world more generally. Interestingly, in countries such as France and the United Kingdom, the old Badumes, under the pressure of the State education systems, have gradually given way to new forms of "langues de parité", informal, colloquial forms of speech which may retain traces of the local Badumes, particularly in terms of phonology, but which are increasingly characterised by their grammatical proximity to the Norm. In assuming the affective function formerly retained by the Badumes, these new informal styles of speech pose the greatest threat to local national languages such as Welsh on one hand, and the Badumes on the other.

35Taking the argument one step further, these more casual modes of speech can also be seen to challenge the RP/standard English "Norm" itself, as one can note by the strident complaints of prescriptivists to the BBC about supposed declining linguistic standards. Indeed, the recent sacking of Boris Johnson, former Radio Four announcer, for having too posh an accent would seem to prove that the pendulum is swinging the other way.15 This, it could be concluded, is an example of evolution through the "democratic voice".

36If one accepts this hypothesis, one could surmise that Welsh English will survive, not in the form of heavily Welsh-influenced, non-standard forms of local English, but rather as grammatically homogenous varieties (relatively speaking) with the preservation of attenuated Welsh phonological and syntactical traits of the kind described by Coupland (1988, 1990).

37As for the Welsh language, the intimate association it once had with Protestant Christianity is increasingly a thing of the past. This association is being steadily replaced by a new vision of Welshness which stresses the link between the Welsh language and its social, cultural, historical and literary heritage. The knowledge of Welsh history, as we have seen, would seem to be one of the defining features of modern Welsh identity (see German, 1996).

38Politically empowered by the recent creation of the new Welsh Assembly, the Welsh language, or at least a literary-based form of it, is potentially on its way to being upgraded from a local "Standard" to something resembling the status of a State "Norm", on a par with the RP/standard English Norm. Whether this reinforces the position of the Welsh language over the long term, however, will largely depend on the desire of Welsh-English speakers to learn the mother tongue. Furthermore, if a situation of diglossia within the Welsh language itself is to be avoided, a situation which could precipitate its demise, teachers of the new Welsh Norm must also be willing to valorise and promote the use of local Welsh dialects.16 Likewise, if this is to be achieved, older speakers of local Welsh must also make an effort to use the language with young learners, thereby familiarizing them with "authentic" forms of the language. Such an approach could do much to encourage linguistic and social cohesion within Welsh-speaking communities in the future.