Robert D. Borsley

University of Essex


1. Introduction

It is sometimes said that there is a new confidence in Wales. The country has had its own assembly for two years. Its rock bands, actors, and film makers are increasingly visible outside Wales, and there is talk of ‘cool Cymru’. Naturally, the Welsh language has an important place in this newly confident Wales, and the prospects for the language are an important subject of debate.

In this situation one might suppose that Welsh would be increasingly prominent as an object of academic study in Wales. Sadly, this is not the case. A great deal of energy is devoted to teaching Welsh, and to promoting its use, and also to studying how it is and has been used, but there is little research on the language itself, its sound system and its grammatical structure, how they are acquired by children, how they have changed over time, and how they vary from place to place. Increasingly, such research is carried out outside Wales and by non-Welsh linguists. Consider the area of syntax. Probably the most important recent work is French linguist Alain Rouveret’s Syntaxe du gallois: principes généraux et typologie. Other important work has been carried out by linguists based in England, notably Maggie Tallerman of Durham, David Willis of Cambridge, and Louisa Sadler of Essex (see e.g. Tallerman 1998, Willis 1998 and Sadler 1988). The situation is broadly similar in phonetics and phonology. Probably the most important work here is that of Martin Ball in Ulster and Briony Williams in Edinburgh. Other important work on Welsh has come from scholars in the United States and Germany, notably James Fife and Erich Poppe (see Fife and Poppe 1991).

There is clearly nothing wrong with linguists who are neither Welsh nor based in Wales working on Welsh, but one would expect Wales to produce linguists of her own. In fact, there are very few Welsh linguists. There are a few whose main interests are in dialectology, notably Robert Owen Jones and Glyn Jones in Cardiff and David Thorne in Lampeter. There are two others who have done important work in a number of areas: Gwen Awbery in Cardiff and Bob Morris Jones in Aberystwyth (see Awbery 1976 and Jones and Thomas 1977). All these linguists began to publish in the 1970’s, and no comparable figures have emerged since then.

One might argue that it is too soon for the new Welsh confidence to have any impact in the slow moving world of academia. This is probably right. It is important, however, to appreciate how bad the situation is and how much it has deteriorated over the years


2. Some details

One might expect to find some research on the Welsh language in the Welsh departments of the various colleges that make up the University of Wales. However like most English, French, German, etc. departments, these are largely literature departments. They teach Welsh as a second language, but they do little to promote linguistic research on Welsh. It was not always like this. Sir John Morris Jones, was Professor of Welsh in the University of Wales Bangor in the second and third decades of the 20th century. His A Welsh Grammar (1913) and his posthumous Welsh Syntax: An Unfinished Draft (1931) are still useful sources of information about aspects of the language. Interestingly, the former was published in the same year as The Welsh Vocabulary of the Bangor District (Fynes-Clinton 1913), by O.H. Fynes-Clinton, Professor of French in Bangor. This is an important source of colloquial Welsh data. Arguably 1913 represents the high point of research on Welsh in Wales. There have been others in Welsh departments whose main interests were linguistic. One example was Melville Richards, who taught Welsh in Swansea and Bangor (and Liverpool) (see Richards 1938). Another was T.J. Morgan, Registrar of the University Of Wales in the 1950’s and Professor of Welsh in Swansea in the 1960’s (and father of Rhodri Morgan), whose Y Treigladau a’u Cystrawen (‘The Mutations and their Syntax’) (Morgan 1952) was a major study of the Welsh mutation systems. A further example was Arwyn Watkins, a lecturer in Aberystwyth, who in 1961 published Ieithyddiaeth (‘Linguistics’) (Watkins 1961), a Welsh medium introduction to the subject. Interestingly, he eventually had to move to University College Dublin to get a chair. At one time there was also considerable work on Welsh dialects in the Cardiff department. See especially Ceinwen Thomas’s three volume study of the dialect of Nantgarw near Cardiff (Thomas 1993). However, Ceinwen Thomas retired some time ago, and such research is largely a thing of the past. Now, it appears that research on Welsh in Welsh departments is limited to philological studies and the writing of traditional grammars (see Thorne 1993, Thomas 1996). The latter are useful resources but no substitute for the detailed and precise descriptive work that is characteristic of modern linguistics.

One might also expect to find research on Welsh in a linguistics department in Wales. There is in fact just one, in Bangor. The department has produced a significant amount of work on Welsh over the years. In the 1970’s and 1980’s Alan Thomas did important work in Welsh dialectology (see e.g. Thomas et al. 2000, published some time after Alan Thomas’s retirement). In the early and mid 1990’s Welsh syntax was a central concern for Ian Roberts and myself (see Borsley and Roberts 1996). Among other things this led to the establishment of an annual Welsh Syntax Seminar, which continues to this day.2 Now, however, with Roberts in Cambridge after a spell in Stuttgart and myself in Essex, the department has no one working on Welsh.3

Thus, there is very little research on Welsh in Wales and those departments that one might expect to be promoting such research are doing very little.


3. Some arguments 

In response to the above, it might be argued that the situation is not so bad or that it as good as one could expect in the circumstances. Someone might argue that those university departments that might be involved in research on the language need to focus on other matters because their students’ interests lie elsewhere. Members of Welsh departments might argue that their students are interested in Welsh literature and not in the Welsh language. Similarly members of the Bangor linguistics department might argue that that their students are predominantly English and are only interested in aspects of English. The obvious response to this argument is that there is no reason why academics’ research activities should be circumscribed by the interests of their students. This is illustrated inter alia by Bob Morris Jones. He has made an invaluable contribution to the study of Welsh, but he has never taught courses focusing on Welsh. It seems, then, that the problem is not that students are not interested in the language but that academics are not interested (or lack the expertise to pursue their interest).

Someone might also argue that the priority for those who are interested in Welsh must be to teach it or to promote its use in various ways and that studying the language itself is a luxury the country can’t at present afford. The trouble with this argument is that promoting a language and studying it are not independent matters. In the long run the possibilities of promoting the language will be significantly reduced if the language is not an object of study. Consider a teacher of Welsh with an advanced student who asks for a detailed description of some aspect of Welsh syntax, something more detailed than a general grammar provides. In all probability, there will be nothing that the teacher can point to. Similarly, consider a computer scientist looking for precise descriptions of aspects of the language to incorporate into some natural language processing system. Again there is likely to be nothing suitable. Finally, consider a speech therapist concerned with developmental disorders in Welsh speaking children. He or she needs a clear picture of the normal pattern of development in the language. There has been some research on the acquisition of Welsh as a first language, but our knowledge of the normal pattern of development is quite limited. In general, then, promotion of the language requires better descriptions of the language than are presently available. Thus, the idea that study of the language is a luxury the country cannot afford is not a tenable one.

Even if satisfactory descriptions of the language were available for all practical purposes, there would still be a good reason for studying Welsh. A good reason for studying any language is to contribute to the development of a general picture of what human language is like. This is the central aim of theoretical linguistics. Welsh deserves to be as influential in this enterprize as English, French, Japanese or any other language, but it will only have an influence if there is a significant body of rigorous analytic work on Welsh. Producing such a body of work should not be left to linguists outside Wales.


4. Some comparisons

There are a number of comparisons that are relevant here. Wales is not the only small country with an interesting language. Another is Iceland. In 1998 Iceland had a population of under 300,000, while Wales had a population of nearly 3 million. More importantly, according to the 1991 census, the number of Welsh speakers in Wales was 500,000. Thus, there are considerably more Welsh speakers than Icelandic speakers. In the circumstances, one might think that there would be at least as much work on Welsh in Wales as there is on Icelandic in Iceland. In fact, there is far more work on Icelandic in Iceland. Over the last 20 years Icelandic linguists have produced a large body of sophisticated work on their language and as a result Icelandic has had a major impact on linguistic theory.

One might argue that Icelandic is in a rather different position because it is not under pressure from any other language. Consider, then, the case of Basque. Like Welsh, Basque is under pressure from a much stronger language (two languages in fact given that some Basque speakers are in France). It has more speakers than Welsh (660,000 according to the 1991 census) but the difference is not that great. No doubt, then, Basques who care about their language have every incentive to seek to promote its use in whatever way they can. Nevertheless there has been considerable research on Basque in the Basque country, and Basque like Icelandic has had a considerable impact within theoretical linguistics.

There is a much more specific comparison that is worth making. Wales is not the only country with a single linguistics department. Another is the Irish Republic, where the only department is that of University College Dublin. This is in fact smaller than the Bangor department, but two members of the department, Máire Chiosáin and Cathal Doherty, are best known for their work on Irish, Chiosáin for work on phonology and Cathal Doherty for work on syntax.

Thus, a variety of comparisons suggest that the situation could be much better than it is.


5. The future

What of the future? It is fairly clear that the prospects for the study of Welsh in Wales are not good. If there were good young scholars working on Welsh one might think that there is some hope for the future. There were in fact two Ph.D. students at Welsh Syntax Seminar in 2000. Both, however, were Germans, one based at University College London and the other at the University of Bonn. Similarly, David Willis has recently received AHRB funding for a project to develop a large computerized corpus of Middle Welsh, and he has another German linguist as his research assistant.

As far as I can see, the only bright spot is the Bangor Psychology department, which has two American psycholinguists, Marylin Vihman and Ginny Gathercole, conducting ESRC financed projects on aspects of the acquisition of Welsh. Both have Welsh Ph.D. students working with them. Here, then, we have some good news. However, psycholinguistics presupposes linguistics, and that remains in very short supply.

It is hard to see where exactly the blame lies for this situation. It seems, however, that the Welsh Academic community has failed rather badly here. Iceland and the Basque country have shown what small countries can do in the field of linguistic research. There is no reason why Wales should not do as well, but it will take some big changes to give the academic study of Welsh the position one would expect it to enjoy in Wales.





In a reply to the published version of this article, Patrick Sims-Williams, Professor of Welsh at the University of Wales, Aberystwyth, took exception to my remark that Welsh departments ‘do little to promote linguistic research on Welsh’ and provided a cataloque of recent research on Welsh in Welsh departments. However, his list simply confirmed my observation that that ‘research on Welsh in Welsh departments is limited to philological studies and the writing of traditional grammars’. There is nothing wrong with philological work or with traditional grammars, but Welsh, like every other language, deserves more. Philology is concerned with the past, and has nothing to say about contemporary languages. Traditional grammars seek to describe a language in a single volume utilising a descriptive apparatus uninfluenced by the insights of modern linguistics. They can be useful resources, but it is not possible to do justice to any language in a single book, and it is not possible to do this without drawing on the insights of modern linguistics. Like other languages, Welsh deserves detailed and precise work on its grammar and phonology drawing on linguistic theories (and contributing to the evaluation of these theories). As I indicated in the article, such work is taking place but overwhelmingly outside Wales. As a philologist, Sims-Williams would no doubt think it very odd if philological research on Welsh was largely conducted outside Wales and barely existed within the country. It is every bit as odd that linguistic research on Welsh is largely conducted outside Wales and barely exists within the country. I also noted in the article that the situation in Wales contrasts sharply with the situation in Iceland and the Basque country, where Icelandic and Basque scholars have done extensive linguistic research on their own languages. Anyone who says the situation in Wales is fine is saying that Welsh does not deserve the kind of attention in Wales that Icelandic and Basque are getting in Iceland and the Basque country. I hope no Professor of Welsh would say this.




1. I am grateful to Gwen Awbery, Bob Morris Jones, Ian Roberts, Janig Stephens, Maggie Tallerman and David Willis for various helpful comments on this note. It should not be assumed, however, that any of them agree with the views expressed here. A shortened version has appeared in Planet.

2. See

3. There are other linguists in Wales, notably in the School of English, Communication and Philosophy in Cardiff. Not surprisingly, Welsh is not a major focus of interest for these linguists although some have done some work on the use of Welsh.



Awbery, G. (1976), The Syntax of Welsh: A Transformational Study of the Passive, Cambridge, University Press, Cambridge.

Borsley, R.D. and I. Roberts (1996), The Syntax of Celtic Languages, Cambridge, University Press, Cambridge.

Fife, J. and E. Poppe (1991), Studies in Brythonic Word Order, John Benjamins, Amsterdam.

Fynes-Clinton, O.H. (1913), The Welsh Vocabulary of the Bangor District, Oxford University Press, Oxford .

Jones, B.M. and A.R. Thomas (1977), The Welsh Language: Studies in its Syntax and Semantics. Cardiff: University of Wales Press.

Morgan, T.J. (1952), Y Treigladau a’u Cystrawen, Caerdydd: Gwasg Prifysgol Cymraeg.

Morris Jones, J. (1913), A Welsh Grammar, The Clarendon Press, Oxford.

Morris Jones, J. (1931), Welsh Syntax: An Unfinished Draft, University of Wales Press, Cardiff.

Richards, M. (1938), Cystrawen y Frawddeg Gymraeg, Gwasg Prifysgol Cymru, Caerdydd.

Rouveret, A. (1994), Syntaxe du gallois: principes généraux et typologie, Editions du CNRS, Paris.

Sadler, L. (1988), Welsh Syntax: A Government-Binding Approach, Croom Helm, London.

Tallerman, M.O. (1998), ‘The uniform case-licensing of subjects in Welsh’, The Linguistic Review 15, 69-133.

Thomas, A.R., G.E. Jones, R.O. Jones and D.A. Thorne, C. Ó Dochartaigh (eds) (2000), Welsh Dialect Survey, University of Wales Press, Cardiff.

Thomas, C.H. (1993), Tafodiaith Nantgarw: Astudiaeth o Gymraeg Llafar Nantgarw yng Nghwm Taf, Morgannwg, Gwasg Prifysgol Cymru, Caerdydd.

Thomas, P.W. (1996), Gramadeg y Gymraeg, Gwasg Prifysgol Cymru, Caerdydd.

Thorne, D.A. (1993), A Comprehensive Welsh Grammar. Gramadeg Cymraeg Cynhwysfawr, Basil Blackwell, Oxford.

Watkins, T.A. (1961), Ieithyddiaeth: Agweddau ar Astudio Iaith, Gwasg Prifysgol Cymru, Caerdydd.

Willis, D.W.E. (1998), Syntactic Change in Welsh: A Study of the Loss of Verb-Second, Oxford: Oxford University Press.