English in Wales

By Jeremy Martin

Copyright 2005


            The survival of the Welsh tongue, an insular Celtic language, in a largely uncorrupted form, despite its immediate proximity to densely populated urban centres of England, provides an English scholar with an increasingly rare opportunity to study the effects of a Celtic mother tongue on present-day English.  The Welsh tongue is in a unique position, being by far the oldest spoken language in the United Kingdom, and representative of the native language of Britain before Roman, Germanic, or Norman invasion, to demonstrate in its interactions with English the modifications and adaptations speakers of Britain’s original tongue make to phonology, grammar, and the English lexicon when confronted with the eclectic composition of our modern language (Davies 78). 

The Welsh language, rooted in the Celtic, and thereby the Proto-Indo-European family of languages, developed in relative isolation after the demarcation of Welsh territory by the construction of Offa’s dyke in 778 to 796 C.E. (no substantial Roman invasion reached further than Sedbury in the west).  The absence of widespread immigration and reliable infrastructure throughout Welsh history discouraged both the natural adoption and the legislated enforcement of the English tongue on most Welsh citizens until the mid-eighteenth century (Williams 21-32). 

As the English language gradually spread throughout Wales, an inevitable occurrence that began after the Acts of Union in 1536 and 1542 recognized Wales as a principality of England and Welsh citizens as British subjects, bilingualism was largely restricted to cities, border towns and coastal villages in which business was conducted with English gentry and merchants.  English was quickly regarded as the language of opportunity and upward mobility, gaining popularity in these areas as early as the mid-eighteenth century, as gentry were required to speak English at court and most secondary industrial enterprises in Wales were owned by English companies.  As immigrants to Wales tended to prefer these areas for settlement, for commercial and linguistic reasons; as infrastructure throughout the Welsh hills and valleys remained unreliable; and as the Welsh economy was largely export- rather than import-based due to the increasing demand for wool prior to and coal throughout the industrial revolution, smaller Welsh-speaking rural communities were often depleted of Welsh speakers.  These losses, however, for the same socio-economic reasons, were very rarely recouped with substantial numbers of Anglophone immigrants until the twentieth century (Williams 32-4).  Though the population of rural, Welsh-speaking communities atrophied, the language in these areas did not.

It was only at the turn of the twentieth century, with the fourfold convergence of the institution of mass British media, mandatory English education, highway and rail systems linking southern Welsh cities to major English urban centers, and a profound influx of postwar immigration that the Welsh language began to decline significantly in favour of the more commercially viable English language.  The linguistic and cultural impermeability of rural northern Wales, however, due to poor infrastructure continued well past the eighteenth century, to the point that in 2001, 76.21% of Gwynedd residents in the rural northwest reported some degree of bilingualism compared to the national average of 29.43% (ONS 7). The magnitude of the effects of this poorly-developed infrastructure in Welsh communities and the cultural distance resulting from such over the last five hundred years is difficult to overstate; the Nedd valley, the last Welsh community without electricity, was only provided with such in the month of this writing, despite being only about 36 miles from Cardiff, the Welsh capital (The Guardian, 2 Dec 2005). 

The ascendancy of the English language in Wales has thus been, by all accounts, a recent development spanning only a few generations, and one not evenly divided throughout the principality.  Historically, Welsh had been the language of the home and chapel, as compared to the commercial English, and the four primary forces of English-language assertion confronted the declining influence of the Welsh-speaking Nonconformist church and Sunday schools at the turn of the century, and a parental insistence that schoolchildren be taught the language of opportunity; it was common for concerned, bilingual parents to refuse to speak Welsh to their children (Thomas 101; Williams 36).  ‘Home and chapel’ ceased to be bastions of linguistic preservation. 

The Welsh language, however, though it has been on the decline in the last century, through its persistence and resistance to English throughout the last five hundred years, has fallen out of mainstream use in Wales recently enough that it remains the primary factor influencing the English language spoken there (McArthur 109).  The remnants of this very recently untouched Celtic tongue manifest themselves in “Welsh English”, the dialect of English spoken in Wales that is identifiably influenced by the three factors of British media, Welsh-border dialects of English, and the Welsh tongue itself.  It is this latter influence, studied specifically, and its effects on the dialect’s phonology, grammar, and lexicon that reflect the shaping of the English language by a Celtic tongue.

Phonology and Pronunciation

Several phonological characteristics of Celtic languages present in Welsh reveal themselves in the pronunciation of Welsh English.  In recording the speech patterns of English speakers in Port Talbot, seven miles east of Swansea, Connolly codifies what may be taken to be the ‘average’ Welsh accent, if such a thing may be supposed to exist.  The Port Talbot area’s bilingual and monolingual-English populations are nearly the statistical median of all of Wales, and it is geographically located away from the English border, though proximal to a major highway (M4); though somewhat more representative of a South Wales accent, the population density of Wales is strongly weighted towards the southeast, making the case study of Port Talbot (southwest) a reasonable geographic mean: somewhat more southeastern than northwestern, just as is the principality’s population.  He identifies the following phonological system in Port Talbot English (PTE):


            Plosives: / p, b, t, d, k, g /

                        Affricates: / t∫, d3 /

                        Fricatives: / f, v, θ, , s, z, ∫, 3, h /

                        Approximants: / w, r, l, j /

                        Nasals: / m, n, ŋ /


                        Checked monophthongs: / I, ε, a, ɒ, ʊ, ǝ /

                        Free monophthongs: / i:, e:, a:, ɒ:, o:, u:, : /

                   Diphthongs: / e I,  ɒ I,  λ I,  I u, A u, ou / 

(Connolly 122)

Not present in Connolly’s study of the area is the rhotic ‘r’ of the insular Celtic tongues, which asserts itself in the accents of English speakers in the densely Welsh-speaking and bilingual north of Wales, particularly in word-final placement.  In metropolitan or heavily Anglicized areas such as Cardiff or Port Talbot, however, where infrastructural improvements have proved conducive to the promotion of other British accents, the ‘r’ is non-rhotic, though occasionally perceptibly lengthened (Collins and Mees 88).  Similarly, the degree of English contact in the exemplar has generated the sibilant / z / and the /d3/ affricate not present in the Welsh language, which are realized in more northern Welsh English as voiceless phonemes, resulting in homophonous pairs such as seal and zeal, sink and zinc, joke and choke, rich and ridge (Coupland, Introduction 9-10).  The voiceless realization of these sounds is evidence of Welsh influence supporting a distinctly Celtic adaptation of a sound not present in Britain’s original languages; in Scotland, speakers of Gaelic English – another insular Celtic tongue – perform the same substitution (MacArthur 110).

A quality of Port Talbot English that is derived from Welsh phonology and that is identifiable in several other case studies of Welsh English (as in Tench, Parry, and Coupland’s Standard) is the Celtic forwardness of phonological realizations, both in vowel sounds and the strong aspiration and dentalization of plosives, often manifested as weak affrication, especially in alveolar realizations of such (Connolly 123-4; Coupland, Introduction, 10).  Despite a very strong English influence in Welsh urban and border areas, even in Cardiff – coastal, well-connected, proximal to the English border and fully engaged by British media – general tongue advancement and vowel fronting is the rule, rather than the exception (Collins and Mees 88). 

            There exists as well in PTE and Welsh English as a whole, a strong preference for monophthongal vowel sounds, as vowel sounds in the Welsh language are pure, though they may follow one another with some rapidity (Vanderveen 1.4).  PTE exhibits no centring diphthongs whatsoever, as / ε: / is monophthongal and / a I / is very rarely exhibited (Connolly 122).  This preference is closely related to another peculiarity of Welsh English phonology: the Celtic syllable break.  As Welsh exhibits largely pure vowel sounds, with diphthongs clearly divided between two such sounds with no liaison existing even between semivowels and vowels, many Welsh English speakers create a syllable break between the vowel sounds provided them in the English language, pronouncing for example ‘beer’ ‘bee-uh’ (/ bi’- λ /), and ‘poor’ ‘poo-uh’(/ p ʊ- λ /) (Thomas 107-8; McArthur 100).  By virtue of the same linguistic dichotomy, most Welsh English speakers prefer a short, monophthongal vowel sound to a broad RP vowel, employing a short / ɑ / in ‘dance’ rather than the / ɑ: / of the RP accent, and a short ‘ʊ’ or schwa instead of the / λ / in words like ‘cut’ or ‘but’.  In addition, it is not at all unusual for these preferences for fronting and monophthongal vowels to combine; an / i / sound in y-terminal words such as ‘lovely’ replaces the often-diphthongized RP / I /.

            Perhaps the most famous of phonological qualities attributed to Welsh influence on the English spoken in Wales is the ‘singsong’ intonation of the language.  As intonation in Welsh and other Celtic languages is commonly distributed by stress and emotion rather than rigidly defined word inflections, nuclear pitch movement in Welsh English is often applied to the entire word, rather than just the tonic syllable of the sound unit (Connolly 126).  This aspect of the Welsh language also reveals itself in the Welsh English use of long vowels; the exception to the Welsh English preference for short vowels is almost universally localized in long vowels located in stressed syllables – if a long vowel exists in a stressed syllable, the ‘singsong’ Celtic influence will support its length (though the stress may continue almost indefinitely, based on the disposition of the speaker) (Thomas 121).


            The syntactical structure of the Welsh language reveals itself frequently in the grammatical constructs of speakers of Welsh English.  The subject pronoun, required in English but commonly zeroed as unnecessary in the Welsh language, is similarly left out of common Welsh English, as in saw ’im goinand did it ‘isself.  The Celtic syntactical tradition of foregrounding for emphasis is also commonly employed throughout Wales, as in money he’s not short of and hurt, are you? (McArthur 111).  The Welsh language does not make a distinction between clefting and pseudo-clefting as English does, so the practice of fronting for emphasis can most easily be explained by a Welsh influence conventionally unconcerned with dividing semantic importance with syntactical placement (Thomas 137).  This foregrounding may also lend itself to what is considered the grammatical shibboleth of Welsh English, the imperative clause look you or mind you, as in look you, the sheep are loose, which is also reflective of Welsh word order, in which verbs always precede their subjects (McArthur 111; Thomas 108). 

            Perhaps the most striking remnant of Celtic influence on Welsh English is the syntactical use of the verb bod, having been transposed into the English language.  In the Welsh language,

bod (be) is followed by a subject nominal, a predicator yn (semantically related to the English preposition in), and an uninflected verb-noun, as in:

                        Mae ef yn mynd i’r sinema bob wythnos

                        (he goes to the cinema every week)

                        (lit. ‘is he in go to the cinema every week)”

(Coupland 6)

The insertion of the be verb in Welsh English, analogous to the bod formation in Welsh, represents a repetitive or habitual action.  Thomas illustrates this phenomenon with a variation on Coupland’s example: He do go to the cinema every week; McArthur offers he did go regular like and he do go to the rugby all the time (Thomas 135; McArthur 111).  This be confusion is also present in the general Celtic use of will, followed by zero, as an English modal auxiliary, as in is he ready yet? – No, but he will in a minute (Thomas 139).  Just as Welsh and other Celtic languages do not require the be verb to supplement the existing one, so does Welsh English omit the extraneous word. 

Finally, the expletive there as in there’s tall you are! or there’s strange it was! can be correlated directly to the Welsh dyna, an adverb equivalent to the English there, as in dyna dal wyt ti (lit. ‘There’s tall are you’) and dyna od oedd ef (lit. ‘There’s strange was it’).  Here the Welsh-equivalent there simply replaces the expletive how in a straightforward case of preserved grammatical structure.



The polar effect of the Welsh language having enjoyed relative isolation from English influence until the mid-eighteenth century is the English language having enjoyed similar isolation from the Welsh tongue.  The environmental code-switching between home and business, chapel and government, as well as the geographical divisions between Welsh communities and bilingual ones throughout much of Welsh history have so restricted Welsh influence on English that only very few words of Welsh origin survive in the English language, and even these are largely descriptive nouns and turns of phrase without parallel in English.  Pengwin, the most famous Welsh adoption into English, is very nearly the only exception to the rule of the isolation of Welsh terminology.  Even the most vehemently patriotic of Welsh publications in English – dictionaries of Welsh terms and phrases published to generate a sense of national identity – provide only English words taken into a dialectical register or Welsh words applied to structures and items never seen outside Wales (and, likely, infrequently even therein). 

Terms of endearment like bach and del remain popular in Welsh English, used sufficiently into the English register to be considered adoptions, and nouns without direct English corollaries such as hiraeth (a profound, consuming longing for a place or person), hwyl (a sense of enthusiasm and rambunctiousness) and clennig (a gift of money) also persist in the English register.  The vast majority of words promoted by scholars as being drawn from Welsh, however, simply are Welsh words used to describe peculiarly Welsh customs or entities with no exterior currency, such as eistedfodd (a cultural festival unique to Wales), cymanfa (a chapel singing festival unique to Wales), and penillion (a form of choral singing peculiar to Wales), that cannot be properly thought to be nativized terms to a monolingual Welsh speaker of English (Thomas 142; McArthur 112).  While the cognitive elements of the Welsh language certainly influence a Welsh speaker’s ability to use and shape English, the concrete elements – words – have largely remained the province of each disparate language by extensive code-switching and pervasive bilingualism; in speaking the prestigious English language in the context of attempted advancement and the pursuit of political and economic prosperity, borrowing from the rural, lower-class tongue may easily have been construed as counterproductive in the extreme.  Concurrently, with Wales’ significantly bilingual population, the adoption of Welsh words was largely unnecessary; if the correct word did not exist in English, a Welshman simply spoke Welsh.

The future of Welsh English is a matter of considerable debate amongst linguists; Thomas and McArthur believe Welsh English will persist as an accent, but not a dialect, as the ancient tongue continues to fall out of use.  Coupland and Connolly venture no predictions in any of their articles, and confine their socio-political commentary to the past and present exclusively.  Jenkins and Williams predict that a concentrated effort to preserve the Welsh language may continue to provide a Celtic influence upon the English language.  In any event, as the four primary forces of English dissemination throughout Wales continues and even accelerates, and the traditional cultural strongholds of the Welsh language grow in irrelevancy, the only hope of preserving a true dialect informed by a genuinely Celtic influence – actual speakers of maternal Welsh, seems to lie in a concentrated effort for preservation.  Considering, however, that even the most optimistic proponents of linguistic preservation provide Welsh-language education of English-speaking students as the most appealing option, it seems very likely that after thousands of years of perpetuation, this will be the last generation in which one will be able to study the English of a natural speaker of an ancient Celtic tongue.



What follows is a conversation between Sylvia Bailey and Edwin Richards, native and fully bilingual residents of Swansea, Wales.  They exhibit an accent essentially identical to the Port Talbot accent that forms the framework of the Phonology and Pronunciation section of this paper.  Specific and unique attributes of Welsh English are isolated and identified below:

Full Conversation

            - .jpg with Transcript

Shortened Monophthong

Unvoiced, Lengthened Fricative (1)

Unvoiced, Lengthened Fricative (2)

Equal Stress (1)

Equal Stress (2)

Works Cited

Connolly, John H.  “Port Talbot English.”  In English in Wales: Diversity, Conflict, and Change, Nikolas Coupland, ed.  Avon: Multilingual Matters, 1990.

Coupland, Nikolas.  “’Standard Welsh English’: A Variable Semiotic.”  In English in Wales: Diversity, Conflict, and Change, Nikolas Coupland, ed.  Avon: Multilingual Matters, 1990.

Coupland, Nikolas and Thomas, Alan R.  “Introduction: Social and Linguistic Perspectives on English in Wales.”  In English in Wales: Diversity, Conflict, and Change, Nikolas Coupland, ed.  Avon: Multilingual Matters, 1990.

Davies, Janet.  “Welsh.” In Languages in Britain and Ireland, Glanville Price, ed.  Oxford: Blackwell, 2000.  

Thomas, Alan R.  “English in Wales.”  In The Cambridge History of the English Language 5, Robert Burchfield, ed.  Cambridge UP, 1994. 

McArthur, Tom.  The Oxford Guide to World English.  Oxford UP, 2002. 

            Office of National Statistics.  Census 2001: Report on the Welsh Language.  2004.  Accessed 08 Dec 2005. http://www.statistics.gov.uk/downloads/census2001/Report_on_the_Welsh_language.pdf

Parry, David.  “The Conservative English Dialects of North Carmarthenshire.”  In English in Wales: Diversity, Conflict, and Change, Nikolas Coupland, ed.  Avon: Multilingual Matters, 1990.

Parry, David.  “The Conservative English Dialects of South Pembrokeshire.”  In English in Wales: Diversity, Conflict, and Change, Nikolas Coupland, ed.  Avon: Multilingual Matters, 1990.

Tench, Paul.  “The Pronunciation of English in Abercrave.”  In Parry, David.  “The Conservative English Dialects of North Carmarthenshire.”  In English in Wales: Diversity, Conflict, and Change, Nikolas Coupland, ed.  Avon: Multilingual Matters, 1990.

The Guardian Unlimited,  02 Dec 2005.  Accessed 02 Dec 2005.  http://www.guardian.co.uk/uk_news/story/0,3604,1656089,00.html

Williams, Colin H.  “The Anglicization of Wales.”  In English in Wales: Diversity, Conflict, and Change, Nikolas Coupland, ed.  Avon: Multilingual Matters, 1990.


Works Consulted

Aitchison, John W. and Carter, Harold.  “The Welsh Language 1921-1991: A Geolinguistic Perspective.”  In ‘Let’s Do Our Best for the Ancient Tongue’: The Welsh Language in the Twentieth Century.  University of Wales, 2000.

Hanson, J. Ivor.  We Also Speak English: A Study of the Speech Mannerisms of the Welsh.  Bridgend: Bridgend Publishing Co., 1978.

Jenkins, Geraint H.  “Wales, the Welsh, and the Welsh Language.”  In The Welsh Language and its Social Domains 1801-1911.  University of Wales, 2000.

Rollinson, William.  The Cumbrian Dictionary of Dialect, Tradition and Folklore.  Otley: Smith Settle, 1997.

Vanderveen, Roger.  “A Welsh Course,” Mark Nodine, ed.  Cardiff University.  Accessed 14 Dec 2005.  http://www.cs.cf.ac.uk/fun/welsh/Lesson01.html