With thanks to Dr Robert Penhallurick of Swansea University
‘Let me begin by mentioning two cases in Wales itself, where Welsh was simply driven out, and where we have a West-Saxon dialect, certainly much worn out under the influence of education, but still purely English without any Welsh influence. These are the peninsula of Gowerland, in the south of Glamorganshire, west of Swansea, and the south-west corner of Pembrokeshire, about Tenby, Pembroke, and Haverfordwest. They are merely English settlements of the twelfth century.’
These are the words of one of the most eminent English language scholars of the nineteenth century, Alexander John Ellis, from a talk that he gave in London in 1882.
Ellis was the author of the first major survey of the regional dialects of English in Britain, published in 1889. He was among the first to attempt systematically to distinguish the chief dialect regions of English, a language which had arrived in Britain in the fifth century AD as a group of related dialects spoken by Germanic tribes known as the Angles, Saxons, Jutes and Frisians. These tribes took part in the unrest and conflict among the Celtic peoples of Britain, who themselves had arrived a good while earlier, perhaps around 1000 BC. From around AD 450, then, there started in Britain an intimate contact between two quite different language groups, Germanic and Celtic, contact which continues to the present day. By the eighth century AD, the group of Germanic dialects had acquired the name English, and by the sixth century one of the Celtic dialects had evolved into Welsh.
Within this tale of contact, the linguistic history of the Gower Peninsula has a rather overlooked but nevertheless significant role. It also plays a small but notable part in the history of the English language generally.
In the late nineteenth century, as an important aspect of his research, A. J. Ellis wanted to discover in which parts of Britain did the locals habitually speak English as an everyday language and in which parts did a Celtic language still predominate. In Wales, where Welsh prevailed over most of the territory, the enclaves of South Pembrokeshire and most of the Gower Peninsula grabbed his attention because they were English-speaking to the extent that, as he concluded in 1882, they ‘must be regarded as part of England’. This is a bold statement. On what evidence is it based?
Naturally, Ellis’s evidence was linguistic: the words, grammar and pronunciations of the dialect of English that was spoken in nineteenth-century Gower.
At that time in Gower there were words such as caffle or caffled meaning ‘to entangle’ or ‘entangled’, clit ‘to stick together’, cloam ‘earthenware’, drashel or dreshel ‘a flail for threshing’, evil ‘a three-pronged agricultural fork’, fleet ‘bleak, exposed, unsheltered’, flott ‘aftergrass, the grass that grows after mowing’, foust ‘to tumble’, frithing ‘a wattled fence’, nesseltrip ‘the smallest piglet in a litter’, ovice ‘the eaves of a house’, planche ‘to make a wooden-boarded floor’, purty ‘to sulk’, quat ‘to press down or flatten’, showy ‘to clear (speaking of weather)’, slade ‘ground sloping towards the sea’, soul ‘anything eaten (such as cheese, butter) with bread’, suant ‘regular, working smoothly’, sul or zul ‘a plough’, toit ‘a small straw seat’ and ‘frisky or wanton’, want ‘a mole’, and wimble ‘to winnow’. These are all regional dialect words of English.
Some of them (for example, caffle and slade) have also been recorded in regions distant from Gower, such as Scotland and the north of England. Some of them are rarely recorded elsewhere, such as wimble in the sense ‘to winnow’. Some (like evil ‘a fork’ and sul) are descended from words that existed in Old English, the earliest historical period of the language, which lasted from the fifth century to about 1100. Most have interesting origins, like drashel, which is formed on the verb thrash and goes back to Old English, and want ‘a mole’, which goes back even before the birth of English, probably to a Germanic form which meant ‘to turn’, possibly referring to the winding passages made by the animal.
As well as recording that they were in use in nineteenth-century Gower, Ellis noted in his 1889 book another detail that all of these words had in common: they had also been found in the traditional dialects of the south-west and west of England, many of them heard especially in Somerset, Devon, Cornwall and Dorset. This in itself suggested a historical social connection between the peninsula and the land lying the other side of the Bristol Channel.
This connection was evident too in Gower grammar and pronunciation, in forms such as beant for ‘isn’t’, know ’n for ‘know him’, and zay, zee, zide and zo for ‘say’, ‘see’, ‘side’ and ‘so’. The use of beant and baint as negative forms of the verb to be, in constructions like bain’t I and bain’t they, has been frequently recorded for the traditional dialects of the south-west of England, as has the masculine pronoun ’n or ’en, which is descended from Old English hine, which was pronounced like the word inner with an h added at the beginning. The use of a z-sound instead of an s-sound in the initial positions of words, as well as the use of an initial v-sound in words such as vather and vind (‘father’ and ‘find’), also are particularly associated historically with south-west England. A. J. Ellis went so far as to classify Gower English as a sub-section of the contemporary Southern English dialect region of Britain, seeing it as descended ultimately from the West Saxon dialect of Old English, centred in south-west England.
Ellis was not the first scholar to take an interest in the English dialect of Gower. His remarks were informed by the mid-to-late-nineteenth-century observations of two Gower clergymen, the Reverend J. Collins and the Reverend J. D. Davies. Two centuries before them, in 1697, Isaac Hamon of Bishopston had compiled a short list of Gower English words and pronunciations (he mentions the use of v instead of f and z for s), some of which he said could be traced back a further 200 years. This Gower English was indeed ‘ancient and not acquired in modern times’, as A. J. Ellis put it in 1882. The linguistic evidence does suggest that the Gower dialect had its roots in the Old English of Somerset and Devon in particular, and that speakers from over the channel must have started settling in numbers in Gower some time before the late fifteenth century.
In fact, meaningful English-speaking settlement of Gower began further back than that, taking place under Norman protection from the early twelfth century onwards. The Normans had conquered Gower by that time and they established a Lordship of Gower which extended beyond the peninsula into mainland South Wales and which was divided into two administrative areas: Gower Anglicana or ‘English Gower’ in the south and west of the peninsula, and Gower Wallicana or ‘Welsh Gower’ to the north. ‘English Gower’ contained land that was more fertile and easily farmed than ‘Welsh Gower’, and this fundamentally geological divide lies beneath the cultural and linguistic partition that emerged in Gower after the Normans, traces of which survive up to the present. For ‘English Gower’ was that part of the peninsula which over time became mostly English-speaking, and, moreover, English-speaking in a dialect that was carried over the sea from the west country, initially brought by settlers encouraged by the Normans. This dialect took hold in south-west Gower. It is an overstatement to say (as A. J. Ellis did) that ‘Welsh was simply driven out’ of Gower, because the north-east corner of the peninsula (in the Wallicana) stayed Welsh-speaking and there were speakers of Welsh in the Englishry down to the early decades of the twentieth century. The occasional Welsh loanword can even be found in the accounts of the old English dialect of south-west Gower, such as pentan, meaning ‘the hob of the fireplace’, and that common exclamation of disgust, ach-y-fi. But linguistically and culturally south Gower developed an Anglo identity. Along with south Pembrokeshire, which was similarly colonized in the twelfth century, south Gower is among the very first staging posts for English on its journey outwards from England, a journey which eventually spread the language across the world.
For how long did the old Gower dialect survive?
Can it still be heard today? In 1882, A. J. Ellis said that the West Saxon features had been ‘certainly much worn out under the influence of education’, suggesting a dialect ebbing away in the face of increasing educational provision and the growing influence of Standard English. But surprisingly, in 1697, Isaac Hamon had already talked of the ‘Old English of West Gower, which is now out of use’, stating that its forms were ‘strange to the people now that are under 50 years of age’.
In light of this, it is yet stranger that research carried out in the 1960s by David Parry and Clive Upton of Swansea University found evidence of the survival of west England forms in the vocabulary, grammar and pronunciation of the 60-plus age-group of Gower natives, including most of the features listed above. How do we explain all this?
Firstly, dialects do not disappear too easily in response to the promotion and spread of Standard English via education or the media, though this can contribute to changes in local speech. It requires something more powerful than this alone to kill off a traditional dialect. Secondly, anxiety about linguistic change is a common symptom of sensitivity to wider societal and technological change, and can lead to some alarmism. In the case of the old Gower English, predictions about its demise have been made ever since the first accounts of the dialect were written. Nevertheless, it must be said, it is now no longer a living dialect. The older generation of 1960s Gower natives was the last to use the old dialect habitually. In the early twenty-first century there are those who can remember it as a way of speaking from their youth, but the West Saxon words and sounds are no longer a part of everyday Gower life. What happened?
The main cause of its passing was the end of its isolation. For centuries, Gower English had existed as a linguistic island, connected to its mother language by sea route. That isolation helped preserve the social and linguistic character of south Gower. During the course of the nineteenth century, however, the population of Wales became increasingly anglicized, with English-speaking escalating, especially in the south-east. Locally, the peninsula experienced marked rural depopulation between 1850 and 1930, but in contrast the population of Swansea grew spectacularly, as did its English-speaker numbers. Extensive housing development ensured that this population expanded into nearby Gower. It was this new wave of anglicization and its much greater speaker numbers that eventually overwhelmed the old Gower English during the course of the twentieth century.
Yet, in a sense, it is still there, surfacing in the often-expressed sentiment among the southern Welsh that Gower people are a bit different and not quite as Welsh in their make-up as their neighbours. These days high property prices on Gower and a perceived social-class difference from the mainland fuel this feeling, but it is based on something deeply embedded in folk history.
And it seems that linguistically the south Gower community continues to be different. The English spoken there now is not the old inherited West Saxon dialect, but is a version of the Welsh English of the mainland, which is a considerably younger variety than old Gower English, and which gets much of its flavour, particularly in pronunciation, from features of the Welsh language which transferred into English as it was acquired by the southern Welsh. When I did some research in the early 1980s comparing the pronunciation of English in one village located in the old Welshry (Penclawdd) and one in the old Englishry (Reynoldston), I found the southern Welsh English accent present in both, except that Reynoldston was less Welsh-influenced than Penclawdd. It was a more English Welsh English, so to speak. There was, for example, hardly any sign in Reynoldston of that Welsh rolled pronunciation of r that can often by heard in the English of South Wales, including the English of Penclawdd. One can interpret this as a kind of ingrained resistance to the ‘Welshness’ of Welsh English and as a continuing testimony to 900 years of English-speaking in south Gower.
Collins, J. (1849) ‘A List of Words from the Gower Dialect of Glamorganshire’, Proceedings of the Philological Society, Volume IV, Number 87 (June, 1849), 222-3.
Davies, J. D. (1877-94) A History of West Gower, Glamorganshire Parts I-IV (Swansea: The Cambrian).
Ellis, A. J. (1882) ‘On the Delimitation of the English and Welsh Languages’, Transactions of the Philological Society, Volume 19 (1882–4, 1885), 5–40. Read before the Honourable Society of Cymmrodorion, 24th May 1882, and in abstract before the Philological Society, 2nd June 1882.
Ellis, A. J. (1889) On Early English Pronunciation, with especial reference to Shakspere and Chaucer, Part V: Existing Dialectal As Compared With West Saxon Pronunciation (London: Trübner & Co.).
Hamon, I. (1697) ‘The Old English of West Gower, which is now out of use’. MS Carte 108 (Collections for Wales) at the Bodleian Library, folios 21-30. Reproduced in ‘Edward Lhuyd and Some of his Glamorgan Correspondents: A View of Gower in the 1690s’, by F. V. Emery, The Transactions of the Honour¬able Society of Cymmrodorion, Session 1965, Part I, 59-114.
Parry, D. R. (ed.) (1977) The Survey of Anglo-Welsh Dialects, Volume 1: The South-East (Swansea: privately published by the editor).
Penhallurick, R. (1982) ‘Two Gower Accents: A Phonological Comparison of Penclawdd and Reynoldston’, Transactions of the Yorkshire Dialect Society, Part LXXXII, Volume XV (1982), 29-41.
Upton, C. (1970) Studies in the Linguis¬tic Geography of Pembrokeshire and the Gower Peninsula (University College of Swansea, unpublished MA dissertation).