Contribution to a Round Table discussion at 'Language, Literature and Frontier', Forum for Iberian Studies, The Queen's College, Oxford, 17 May 1997:

I must say straight away that the notion of drawing frontiers is one which filled me with gloom, for frontiers can imply confrontation and division rather than co-operation and unity. It is precisely the assumption of frontiers which has led the study of minority languages into what in some circles is called 'conflict linguistics'[1]. Yet the conflict, like the frontier, is not primarily linguistic, but political. Which 'linguistic modality' of the Iberian Peninsula (to use the sanitised terminology of the Spanish constitution, which dared not choose between lengua and dialecto for fear of arousing ignorant prejudice) is well-defined in terms of linguistic geography? Not even Catalan, in many ways the best-established besides Castilian and Portuguese, can claim such a distinction: the article on Catalan dialectology in the recent Lexikon der Romanistischen Linguistik bemoans that

Even linguists have been guilty of promulgating the notion of an artificial frontier, as studies entitled El dialecto leonés or El dialecto aragonés, neither of which objectively exist (or indeed ever existed) show.[3] And it comes as a surprise to many outsiders to learn that there lurk within the apparently unitary notion of 'Basque' a number of mutually unintelligible languages. The linguist's sin is on the whole no more than that facet of human nature which likes to pigeon-hole and generalise in order to make reality more manageable, which is why we speak of such artificial notions as 'Ibero-Romance', 'Rhaeto-Romance' and the like; though occasionally the boundary between ingenuous classification and subservience to political ends has been faint, as in the celebrated debate over the linguistic affiliation of Catalan, every argument concerning which is actually spurious.[4]

So: what is a linguistic frontier? We should identify at least the following:

1. Natural microfrontiers between very closely related lects, which exist between every community (indeed, at maximum resolution, between every speaker), an example of which would be the speech of La Coruña as opposed to the speech of El Ferrol.

2. Natural macrofrontiers between distantly related or unrelated languages (e.g. the frontier between Basque and Navarro-Aragonese).

3. Artificial macrofrontiers between imposed or agreed standards (e.g. the Spanish/Portuguese border, or the linguistic boundary between Flemish and French in Belgium).

It is clear that, insofar as we have been talking today about the boundaries between Galician and Castilian, Basque and Castilian or Catalan and Castilian, we have been talking today mostly about boundaries of type 3, though we must add the dimension of bilingualism (to be exact 'asymmetrical bilingualism'), which exists in the Galician, Basque and Catalan communities.

Does a linguist, then, have any place in a forum such as this, except as an interested observer of the games poiliticians play with language? Well, we might think in the first place that at the very least linguists could be expected to be on the side of the angels in pleading for the survival of as many languages as possible, since languages are our bread and butter. Alas, this leads us straight into a politically charged debate of the kind that is currently in full swing in Asturias and Aragón, namely, the debate between those for and against the establishment of an imposed common language (I will call the former 'interventionists' and the latter 'conservationists'). The interventionists have definitively won this debate in Catalonia and have tentatively won it in Galicia and the País Vasco, so any wisdom the linguist might have to offer there is too late. The question is this: the establishment of a common language is an obvious way of preserving it, whatever we mean by 'it' - 'it' is in fact a fiction, a figment of (usually) some academic's imagination. Unless there is a common language, the argument goes, there can be no agreement on official usage and no ennoblement of the language such as will make it suitable for use in a number of written registers ('status planning' or 'normalisation'), nor can there be uniform teaching in schools, so ensuring 'the language' for future generations ('corpus planning' or 'normativisation'). The current success of Catalan seems to speak eloquently in support of this approach, though we must never forget that at the heart of literary Catalan was a real spoken language which was a natural candidate for the exercise of hegemony, namely, that of middle-class Barcelona. So is this the camp the linguist should back? The well-known paradox is that if a common language succeeds, there must be casualties. Jesús Neira, for example, argued passionately against this policy in Asturias, and he had quite a weight of linguistic evidence on his side (not to mention Jovellanos).[5] When a common language begins to be imposed, speakers of the 'real' dialects resent it or feel inferior: it is in a way just as imperialist an act as the imposition of Castilian. In the País Vasco, the promotion of Guipuzcoan-based batua is reported as having alienated many vizcaíno speakers, to such an extent in fact that the learning of 'Basque' has visibly declined in Vizcaya.[6] The moral is clear: if you want to kill off the dialects (which as far as Romance languages are concerned, are the much-trumpeted 'true descendants of Latin'), then impose a common language. The linguist who values dialectal diversity must therefore think twice about movements towards a common language, yet in so doing he probably condemns himself to watching natural selection take its course. It is the classic conservationist conundrum: should you 'save' a natural state of affairs by intervening in it?

On the other hand, linguists should not perhaps be so coy about language planning, for I believe they have a great deal to learn from what is going on at the moment in the Spanish autonomías. If we are worried about the artificiality of common Galician, for instance, we might do well to reflect that Castilian, and indeed all the Romance standards, owe much to such language planning, albeit of a less sophisticated and self-conscious kind. The idea that 'the Spanish language' (is such a concept viable?) derives naturally from Latin is of course a nonsense - it too has undergone its eclectic interventionist normalisation and normativisation from Alfonso X onwards.

Incidentally, we must realise that the linguistic ecological balance-sheet is not all negative. It is rather sad in my view that so little attention has been paid to the new varieties of Castilian that are created by diglossic contact with other languages, and to the fascinating phenomenon of code-switching and hybridisation, well-documented for US Spanish and the fronteiriço of the Uruguay/Brazilian border, but much less well-studied in those areas reported as using languages which go by the name of chapurrao or amestao and which often, alas, are sneeringly dismissed in the 'frontier' debate (almost, to continue the rhetoric of confrontation, as if they were 'double agents').[7]

I return to frontiers. It is of course the common language that creates the macrofrontier. Indeed, the desire for a frontier is a motivating factor in the elaboration of a common language in the first place, since the chief political point in Spain is usually that of insisting on difference from Castilian and the unity of the region in question - ironically, the antipathy towards Castilian may even lead to a reintegrationist stance with another, non-Castilian, standard: Galician with Portuguese, Valencian with Catalan.

Another confrontational feature of common languages (and it must now be appropriate that I include Castilian in this category as well) is the pursuit of a single, rather than a polynomic, norm. Very occasionally we find reluctant admittance of variants (usually in extremely unimportant areas, like the first person singular of the verb roer in Castilian) and often for the wrong reasons (the Real Academia Española has long admired loísmo since it appears to reflect an isomorphism between case form and case-function of the kind often - though not as often as is sometimes thought - encountered in Latin). But the idea that a common language can admit many variants goes against the grain, possibly because normativisation costs effort and the in-crowd then don't want to see that effort wasted.

In the end, all a linguist can do is clarify. I would plead that in our discussion we are at least clear about the nature of languages and frontiers and that we separate in our minds at least (for it's probably impossible anywhere else) the linguistic from the political.

[1] See especially Rebecca Posner (1993), 'Language conflict or language symbiosis? Contact of other Romance varieties with Castilian', in David Mackenzie and Ian Michael (eds), Hispanic Linguistic Studies in Honour of F.W. Hodcroft (Llangrannog: Dolphin), pp.89-106 and John N. Green (1994), 'Language status and political aspirations: the case of northern Spain', in M.M. Parry, R.A.M. Temple & W.V. Davies (eds), The Changing Voices of Europe: Social and Political Changes and their Linguistic Repercussions, Past, Present and Future. Papers in Honour of Professor Glanville Price (Cardiff: University of Wales Press/MHRA), pp.155-72.

[2] Joan Veny, 'Katalanisch: Areallinguistik', in Günter Holtus, Michael Metzeltin and Christian Schmitt (eds) (1991), Lexikon der Romanistischen Linguistik, V (Tübingen: Niemeyer), pp.243-61, at p.243.

[3] Ramón Menéndez Pidal (1962), El dialecto leonés (Oviedo: Diputación de Oviedo, Instituto de Estudios Asturianos); Manuel Alvar (1973), Estudio sobre el dialecto aragonés (Zaragoza: Institución 'Fernando el Católico').

[4] See, for example, Gerhard Rohlfs, trs. Manuel Alvar (1979), Estudios sobre el léxico románico (Madrid: Gredos), p.259, who concluded that 'el catalán es en lo esencial una "dépendance del provenzal' (on the basis of 33 arbitrary lexical items).

[6] See Jesús Neira (1982), El bable: estructura e historia (Salinas: Ayalga).

[7] Catrin Hughes (1992), 'Evaluating linguistic competence in a Basque-Castilian speech community', Bulletin of Hispanic Studies, 69, 105-26.

[8] Berta Piñán (1991), Notes de sociollingüística asturiana (Xixón, Llibros del Pexe), pp.49ff., calls attention (objectively) to the phenomenon of hybridisation in Asturias.

© Christopher J. Pountain 1997