Welsh Phonetics:
a preface

J.C. Wells

I am not a native speaker of Welsh. But it is a language in which I have for many years had a great interest.

When I was an undergraduate reading classics at Cambridge, one of the subjects I chose to study was comparative philology. This involved discussing numerous examples of sound correspondences between the various Indo-European languages. I remember then feeling envious of one of my fellow-students who spoke Welsh, since he could use his native-speaker knowledge to come up with Celtic examples that the rest of us had to learn by heart. I duly learnt ten or twelve Welsh words to quote in the examination -- but I didn't take it any further.

When I came to UCL to do a master's in phonetics, I remember hearing two of my fellow postgraduates speaking together in Welsh. I felt a mixture of emotions: annoyance that I couldn't understand what they were saying or take part in their conversation, but also a lively phonetic interest in the exotic sounds and rhythms I was hearing. I wished I could speak Welsh -- but I didn't take it any further.

When I started teaching, becoming a lecturer at the UCL Department of Phonetics and Linguistics, every now and again there would be a student who was a native speaker of Welsh. After all, Welsh is the second most widely spoken indigenous language of the United Kingdom in which I live. And I began to feel ashamed. Here was I, holding forth about the phonetics and grammar of a wide range of different languages -- yet I couldn't even speak this language that was on our doorstep, a few hours' drive from London.

So I decided to learn Welsh. I studied books, I went to evening classes. I took a Welsh O level, I took a Welsh A level. (There's a lot to be said for doing advanced-level language exams in one's thirties, rather than as a teenager: you can make your own mature judgments about the literature you study, rather than having to reproduce the teacher's views.) The peak of my achievement came when I managed to hold my own in a BBC radio interview, in Wales, in Welsh.

This means that I was delighted to hear of plans to publish Welsh Phonetics. This book will enable a wide audience to become familiar with the phonetic work that has been done on the language, work that has not until now been widely known.

Non-linguists looking at Welsh notice mainly the spelling, which makes it look unpronounceable, with its abundant w's and y's. Linguists know there's more to a language than its orthography.

• Syntacticians can savour the VSO word order of Welsh and the interesting consequences it entrains.
• Morphologists can marvel at the conjugating prepositions and the singulative suffix of nouns.
• Phonologists are fascinated by the consonantal mutations. They ought to investigate, too, the strange behaviour of schwa, which happily occurs in stressed syllables while being barred from certain unstressed ones (namely, final syllables).
• Classicists and etymologists can explore the legacy of Latin left in Welsh: the obvious loans such as pont 'bridge' and ffenestr 'window', and the less obvious ones such as gwenwyn 'poison', cadwyn 'chain', and saeth 'arrow' derived by regular sound changes from Latin venenum, catena, sagitta. Even the history of English can be seen in Welsh, with sad retaining its old meaning of 'firm, steady' and the place-name Betws taking us back to the obsolete English bedehouse 'house of prayer, almshouse'.

What can phoneticians find? Look inside.

• Many will know that Welsh has a voiceless alveolar lateral fricative, as in llaw [ɬaʊ] 'hand'.
• Some will have heard of the remarkable high central vowel of North Welsh, the [ɨ] of dyn [dɨːn] 'man'.
• But most have still to make the acquaintance of the remarkable set of diphthongs all suggestive of English [aɪ]: the tei [təɪ] 'tie' and tai [taɪ] 'houses' distinguished in all types of Welsh, and the further diphthongs heard, in north Wales only, in cau [kaɨ] 'to shut' and cae [kaːɨ] 'field'.
• Few will know, until they read it here, that -- unlike in most languages -- stress in Welsh seems to be associated with the presence of level pitch, and lack of stress with a change of pitch. In consequence, non-Welsh-speakers tend to fail to perceive Welsh stress correctly, which reminds us that auditory perception may be language-dependent.

Yn wir, trysordy o iaith yw'r Gymraeg.