WHICH CORNISH?
areth neb a vue rys gans an Dr Nicholas Williams yn Lostwydhyel mys Gwyngala 1996 yn dan gaderyans an Dr Philip Payton, Kevarwedhor Fundyans Lyen Kernow

[ Tre arta ]

I am very happy to be here in Cornwall after an absence of a mere 29 years. It is encouraging to hear so much Cornish spoken, even if it is distressing to know that the revival is very split - and completely unnecessarily as well.
You will, I am sure, want to know who I think I am, coming over here and telling you how you should revive your language. I can almost hear you asking yourselves 'Who exactly is this interloper?' I was born and brought up on the Essex side of London. While at school I taught myself Cornish from Caradar's Cornish Simplified (1959). I won a prize for Cornish verse in the Gorsedd competition of 1961 and still in my teens became a language bard in 1962 in Newquay, where I took the bardic name Golvan 'sparrow'. This wasn't because I saw myself as a Cockney sparrow; it was more because I thought of the sparrow like myself as small and chirpy.
I read Classics and English language in Oxford and then, because I had an interest in Cornish, I did a postgraduate course in Celtic there and learnt Irish. I went to live Ireland in 1967 and have been in Ireland virtually continuously ever since. I now make a comfortable living in University College, Dublin teaching Irish to the Irish through the medium of Irish. I have, of course, never lost my interest in Cornish, though I was a quiescent member of the revival for years. It was only the adoption of Kernewek Kemmyn that jolted me out of my semi-retirement from Cornish. I believe that Kernewek Kemmyn is not Cornish and am very unhappy that it was ever proposed or promoted.
Last year Kernewek dre Lyther published a book of mine, Cornish Today, in which I examine the three kinds of Cornish at present being used. I conclude in my book that Unified is by far the least unsatisfactory form of the language and Kernewek Kemmyn is so mistaken that it should be abandoned. Dr Charles Thomas of the Institute of Cornish Studies called it 'an unjustifiably wrong turn'. I believe he was absolutely right. I have not the slightest doubt that Kernewek Kemmyn will be abandoned. The only question is how soon.
I do not here wish to say much about so-called 'Modern Cornish'. It seems unwise to me to base the revival on the Late Cornish of the decline rather than on the rich and idiomatic language of the Middle Cornish period. The attempt to revive Cornish in the Anglicised spelling of the seventeenth century, I believe, has little to recommend it.
'Modern Cornish' and Kernewek Kemmyn alike make a fundamental error. They both assume that Middle Cornish and Late Cornish are very different and that we must therefore choose either the one or the other. This is not so. Middle Cornish and Late Cornish are effectively one and the same. This is not just my view, it was also the opinion of Henry Jenner, the first revivalist and the finest linguist Cornwall has ever had. The two forms of Cornish differ only in the matter of their spelling. Kernewek Kemmyn repudiates Late Cornish in favour of Middle Cornish. Yet Kernewek Kemmyn uses {k} in krev 'strong', kov 'remembrance', in hweg 'sweet', hwegh 'six' and {v} in ov 'I am', gwelav 'I see' where Unified uses {c}, {wh} and {f} respectively. These spellings in Kernewek Kemmyn are Late, not Middle Cornish.
As well as Cornish Today I have written a supplement to CT and have two further articles against Kernewek Kemmyn in the press. I have by the way some copies of the Supplement with me (a snip at �2 each and a ripping good read). If anyone wants a copy, he or she should speak to me afterwards. I intend to continue writing against Kernewek Kemmyn until the desired outcome is achieved, i.e. that Kernewek Kemmyn is abandoned. I am not doing so out of any personal vindictiveness. God forbid. It gives me no pleasure to find fault with other people's efforts. I criticise Kernewek Kemmyn because I believe it is my duty to do so. As a professional Celticist and someone who has known Cornish for more than 30 years, I know that Kernewek Kemmyn is not Cornish and I can explain why it isn't. While Kernewek Kemmyn continues to be taught and promoted, it is doing the revival enormous damage. I accept of course that Kernewek Kemmyn was devised and adopted in good faith by people who had the interests of Cornish at heart. One does not doubt their motives. Their judgment, however, is seriously open to question.
The supporters of Kernewek Kemmyn who have read my book and the Supplement fall into three discrete categories, A, B and C.
A. Are those who believe that my criticisms are valid and have already begun to distance themselves from Kernewek Kemmyn. I know of several such people. Since I live outside Cornwall, I don't know of them all, but I suspect that there are others.
B. The second group are those who believe my criticisms are valid but think that it is too late to change Kernewek Kemmyn now. The editor of the KDL Newsletter is one of these. Earlier this year (15 February) he wrote to a correspondent on the Continent and sent me a copy of the letter. He said:

I wonder whether you will have read the Nicholas Williams book yet. The small amount of research I have done myself suggests to me that most of what he says about Cornish phonology is right and that Ken George will have difficulty in countering it.

The editor of the KDL Newsletter nonetheless thinks that Kernewek Kemmyn cannot now be replaced. The same view was expressed by the author of a letter which appeared in An Gannas in March this year. The writer says:

Nyns eus edhomm dhyn ni a yeth yw pella diberthys yn hanow skolhygieth. Kernewek Kemmyn, martesen gans gwennogennow - yw gwrians da, hag yn sur [yth] yw gwell es chanj moy

[We do not need a language that is further divided in the name of scholarship. Kernewek Kemmyn, perhaps with warts, is a good creation and it is certainly better than a further change].

Since Unified Cornish was the only form of the revived language for 60 years and the Cornish Language Board did not think twice about replacing it with something inferior, this argument seems to me to have little force. Kernewek Kemmyn was introduced because it was wrongly believed to be more authentic than Unified. If it can be shown that Kernewek Kemmyn is far less authentic than Unified, as it can, then Kernewek Kemmyn loses any raison d'�tre it may have had.
C. The third group of supporters of Kernewek Kemmyn are those who believe (or at least, wish to believe) that my criticisms of Kernewek Kemmyn are insubstantial and of no importance. The editor of An Gannas appears to fall into this category of inveterate optimists.
If anybody really does believe that my criticisms are entirely invalid, I would ask him to remember the matter of {tj} and {dj}. Let me explain. In August 1987 the Cornish Language Board adopted Kernewek Kemmyn on the mistaken grounds that it was more authentic than Unified Cornish. At the same time I in Dublin acquired a copy of Dr George's book PSRC. I read it and was astonished. The sounds /t'/ and /d'/ were being recommended for revived Cornish and they were to be spelt {tj} and {dj}. I could understand writing {tj} and {dj} in Dutch, Norwegian or Serbo-Croat - but in Cornish? I wrote an article setting out why it seemed to me that /t'/ and /d'/ were mistaken. The article was finally published in 1990. Already by 1988 a typescript copy was being read here in Cornwall. In the periodical Carn, number 68 (Winter 1989/90), under the title Kernewek Kemmyn Up-Date we read the following by Dr Ken George, the deviser of Kernewek Kemmyn:

Nicholas Williams in a brilliantly argued paper to be published later this year, has put forward strong arguments to show that the phonemes /t'/ and /d'/ never existed. Close examination of his evidence confirms this....To everyone's relief, this will mean the disappearance of the corresponding graphemes {tj} and {dj} (Carn 68: 16).

I dispute two aspects of that passage. In the first place, my article was in no sense brilliantly argued. I said nothing that was, in my view, not obvious to any scholar who had examined the question in detail. In the second place I take issue with the expression 'his evidence'. I lived and indeed live in Dublin. I had no evidence that was not available to Dr George. All I had were the texts of MC and LC: the Passion Poem, the Ordinalia, Beunans Meriasek, John Tregear, the Creation, the writing of the Bosons, Wella Kerew, Edward Lhuyd, Pryce and also Padel's book on place-names. None of this is my evidence; it was readily available to Dr George when he suggested his {tj} and {dj}.
At all events I was instrumental in relieving Kernewek Kemmyn of {tj} and {dj}. Had it had not been for my article, Kernewek Kemmyn would still write Y'n termyn eus passys yth edja trygys yn Synt Leven den ha benyn yn tyller kriys Tji an Hordh (I am quoting from Dr George's book The Pronunciation and Spelling of Revived Cornish). But for me, Kernewek Kemmyn speakers would still be saying wodja 'after', adjwon 'know', kerentja 'love', gantjo 'with him', boghodjek 'poor', gallodjek 'powerful', pedjwar 'four', krydji 'believe', pydji 'pray', nyndj udji 'is not', etc. If I was correct about {tj} and {dj} (as Dr George has admitted), is there not a prima facie probability that there is at least some substance to my further criticisms of Kernewek Kemmyn? Kernewek Kemmyn shorn of and is the still the same old Kernewek Kemmyn. It is all the same vintage. Those who tell you that my criticisms are as nothing, deceive only themselves.
In my article about {tj}, {dj} I praised other aspects of Kernewek Kemmyn. This I bitterly regret, since I was not telling the truth. I lauded Dr George's efforts because I wished to encourage this new apparently more scientific approach to Cornish. I hoped that by being polite, I would be able to get rid of the absurd {tj} and {dj}. I was aware that Kernewek Kemmyn was mistaken, though I did not then realise just how mistaken it was. At the same time as I was praising Dr George, I was writing to the publisher Len Truran advising him not to have anything to do with the new system. Foolishly I thought that Kernewek Kemmyn would not be accepted without considerable revision. How naive I was! My greatest regret is not that I praised Kernewek Kemmyn insincerely, but rather that I did not write my book until 1995. Had Cornish Today been published in 1988, Kernewek Kemmyn would never have got as far as regrettably it has.
Dr Ken George was the inventor of Kernewek Kemmyn. I say inventor advisedly, since most of the sound system and the spelling of Kernewek Kemmyn is fiction rather than fact. Now that I have written in detail and at length against Kernewek Kemmyn, Dr George finds himself in something of a quandary. And he certainly has my sympathy. At one moment he claims that my criticisms are worthless. The next moment he appears to admit that perhaps I may have a point here and there. In An Gannas December 1995 he described my critique of Kernewek Kemmyn as a chi kartennow or 'house of cards' and asserted that the more he looked at my arguments, the weaker they became.
In his interim response to my criticisms in the magazine Kernow, however, he appears to be admitting that Kernewek Kemmyn is not as perfect as he had once hoped. Let me give you some passages to compare and contrast. In his book PSRC of 1986 Dr George wrote:

I can therefore state with confidence that Revived Cornish, as exemplified by the phonological base described in this chapter [i.e. Kernewek Kemmyn], is closer to the Cornish of 1500 than were either OldC or LateC. What is more, it is closer to the Cornish of 1500 than is, say, the "Geordie" dialect to standard English (PSRC: 91).

Note the self-assured tone when describing Kernewek Kemmyn: 'I can therefore state with confidence...'. That was in 1986. Since the appearance of my book in 1995, however, Dr George is apparently not quite as convinced as he was of the excellence of Kernewek Kemmyn. In Kernow (January 1996) he wrote:

There is a measure of uncertainty ("experimental error") associated with the reconstruction of Cornish. Some is due to the variation within the traditional language.....Once a reconstruction approaches traditional Cornish so closely as to be within the zone of uncertainty, it is possible for two reconstructions to be within the zone of uncertainty and yet to differ (Kernow 14: 9-10).

A tone of certainty has been replaced by a zone of uncertainty. Yet what Dr George says is quite simply untrue. There is room for some uncertainty in revived Cornish and indeed I can tell you, if you wish, exactly where the uncertainties are. There is no room for the huge differences in spelling and sounds between Kernewek Kemmyn and UC. It is no use invoking experimental error here, since there is no experiment. We are talking about historical linguistics, not natural science. We can't put Kernewek Kemmyn and Unified Cornish in test-tubes and leave them overnight, nor can we examine them under the microscope. We can do nothing but read the remains of traditional Cornish as closely as possible and draw our conclusions. It is not a question of experimental error but of error pure and simple. Kernewek Kemmyn is mistaken.
Dr George wrongly believes that MC had two long /o/ vowels. Since he believes that there were two such vowels in MC, one open and one closed, Dr George is compelled to spell one of them : koes 'wood', goes 'blood', hwoer 'sister' etc. Indeed he criticises Nance for not making the distinction. In his book PSRC Dr George says of Nance:

There are so many such pairs [i.e. words distinguished according to Dr George by a closed versus an open long o], that is is difficult to believe that Nance was unaware of this difference; and one suspects that he deliberately over-simplified the language to make it easier to learn (PSRC: 126).

Nance was absolutely meticulous in all things. If he had believed there was a difference, he certainly would not have suppressed it. The problem here is not about Middle Cornish so much as Late Cornish, since in Late Cornish some words with Middle Cornish long /o:/ have long /u:/. Dr George does not seem to realise that Nance was perfectly well aware of the problem and attempts to solve it in the introduction to his 1938 dictionary.1 Note, however, in the short passage that I have just quoted Dr George says: 'one suspects [Nance] deliberately over-simplified the language to make it easier to learn.' That was in 1986.
In 1996 after the appearance of my book Dr George wrote as follows:

The orthography of Kernewek Kemmyn has got its priorities right: it balances the need for historical authenticity with the need for a phonemic system, giving priority to the latter (Kernow 14: 10).

By 'phonemic' Dr George means that the spelling should be regular and involve one symbol or set of symbols per sound. Kernewek Kemmyn replaced Unified Cornish for one reason only: that it was believed to be more authentic. Now apparently, as a result of my criticism of Kernewek Kemmyn, Dr George is saying that authenticity isn't all it's cracked up to be. Authenticity is all very well but a regular spelling takes precedence. Dr George believes that it is more important for revived Cornish to be regular than to be authentic. If I may be allowed to paraphrase, Dr George seems to be saying that Kernewek Kemmyn simplifies Cornish in order to make it easier to learn. Yet in 1986, before my book had appeared, Dr George actually rebuked Nance for doing just that. He wrote: 'one suspects that [Nance] deliberately over-simplified the language to make it easier to learn.' Ironically by insisting on two long o vowels, when MC had only one, Dr George is making Cornish harder to learn as well as less genuine.
Let me digress here for a moment to say something about the difficulty or otherwise of revived Cornish. It is the job of those reconstructing Cornish to render the reconstruction as authentic as possible. Authenticity is the overriding criterion - indeed it ought to be the sole criterion. When learners come to the reconstruction, however, they will learn it at their own pace and to their own degree of competence. They may favour 'easy' constructions and forms over 'hard' ones. That is at the level of the learner only. The underlying reconstruction must be as authentic as possible, irrespective of complexity or simplicity.
To those who espouse Kernewek Kemmyn I would say this: we are reviving a language as we believe it to have been. If we know our reconstruction is mistaken, then we must change it, whatever the cost. Otherwise we are not reviving Cornish. Kernewek Kemmyn was devised among other things to give revived Cornish academic respectability. You will not be astonished to hear that Celticists outside Cornwall are very sceptical indeed of Kernewek Kemmyn's claims to be authentic. Now that Kernewek Kemmyn itself is no longer claiming to be authentic, who shall blame them?
Why and how is Kernewek Kemmyn mistaken? In the first place, one must question the wisdom of the Cornish Language Board in adopting Kernewek Kemmyn so quickly. From 1928-87 there was only one kind of revived Cornish in existence, Unified. Moreover UC contained within itself a considerable amount of variation and scope for revision. Correction of a few points is probably necessary. But the Cornish Language Board did not set up a committee to examine UC and recommend improvements. That would have been sensible. Instead they adopted Kernewek Kemmyn, a totally new and unhistorical form of Cornish - a radical rewriting of the language that does not resemble in sounds or spelling the traditional Celtic speech of Cornwall at any period in her history. Kernewek Kemmyn is the work of one man, not a professional linguist, let alone a professional Celticist. Before adopting Kernewek Kemmyn in its entirety, the CLB did not submit the new system to a panel of Celtic scholars outside Cornwall. Nor did they canvass users of Cornish to ask them what they thought of the new form of the language. The CLB adopted and began vigorously to promote Kernewek Kemmyn without any independent assessment and with a minimum of consultation. To say that the Board acted imprudently is an understatement.
Now that I have shown, and continue to show, that Kernewek Kemmyn is very mistaken indeed, various members of the Cornish Language Board are expressing anger towards me. A classic case of shooting the messenger.
The errors in Kernewek Kemmyn are threefold: theoretical, orthographical and phonological. I'll deal very briefly with these in turn.

Theoretical errors of Kernewek Kemmyn

Kernewek Kemmyn is constructed on the assumption that Middle Cornish is very close to Breton and Welsh. If the Welsh and Breton evidence and the evidence of the MC texts are at variance, Kernewek Kemmyn gives more heed to Welsh and Breton than to the texts. Yet it is quite clear that Middle Cornish is very different from Welsh and Breton. The Welsh for 'father' is tad and for 'after' is wedi. The Breton for 'father' is tad and the word for 'after' is goude. In Middle Cornish, however, 'father' is tas with an s and 'after' is wose with an s or woge with a g. This change of d to s or g occurred in Cornish at some time between Old Cornish and our earliest Middle Cornish texts, ca 1150-1250. The assibilitation of d is unique to Cornish and does not occur in either of its sister languages. Such an important change indicates that something very far-reaching has occurred between Old and Middle Cornish. If Cornish is unlike Welsh and Breton in this important respect, the experienced historical linguist would expect it to differ from them significantly in a host of other ways. Yet it apparently never once occurred to Dr George that Middle Cornish might be very different from both Welsh and Breton.
The real difference between Cornish on the one hand and Welsh and Breton on the other is the influence of English. Until the early nineteenth century there was little English in Wales and there was, of course, no English in Brittany. In Cornwall, to judge by the Bodmin Manumissions, Old English (Anglo-Saxon) had made inroads already in the Old Cornish period. It seems that between Old and Middle Cornish the language was strongly affected by Middle English and had become in some ways more like English and less like Welsh and Breton. Nowhere was this so important as in the vowels. This whole development I call the Prosodic Shift and I can show exactly how it has influenced Middle Cornish phonology and why therefore Kernewek Kemmyn is seriously at odds with the evidence.
Cornish has been heavily influenced by an early form of English, but that does not make Cornish any less Cornish or any less Celtic. Cornish became extinct in the late 18th century and revived Cornish is the language of an earlier period. As a result there are ways in which revived Cornish is more Celtic than, for example, modern Welsh.

Orthographical errors in Kernewek Kemmyn

Even if MC were not radically different from OC, Kernewek Kemmyn can be faulted in another way. Unified Cornish seeks to regularise the spelling of the medieval texts. It widens the use of {j} to places where the scribes write {g} for example yn y jy for MC yn y gy 'in his house'. Moreover UC borrows {dh} from Lhuyd to represent the sound of voiced th. Some people criticise on the grounds that it is never used in traditional Cornish. The Passion Poem, however, uses the symbol yogh {3} to represent the sound of voiced th. I like to understand {dh} as just a convenient and modern way of representing the symbol {3} on contemporary keyboards. After all Edward Lhuyd used for the yogh of the Passion Poem as early as 1707 (see item 14 on your hand-out).
Generally speaking therefore there is little or nothing in UC that is not in the MC texts. Kernewek Kemmyn on the other hand rewrites the spelling of Middle Cornish entirely and produces a bizarre and exotic spelling. Kammbronn is only the most notorious example. The late Peter Pool rightly observed:

Kernewek Kemmyn is something quite different, an entirely artificial creation which does not resemble Cornish as used by Cornish people at any time in history. To those accustomed to Unified...Kemyn has an alien and somehow sinister appearance, as if the language had somehow been taken over by robots and reduced to the status of a code.

Except that Kernewek Kemmyn is not a code, since a code is consistent.
Had the sound system of Kernewek Kemmyn been even partially correct, this distortion of the orthography might (and I say might) have been justified. Since the phonology of Kernewek Kemmyn is very mistaken, the distortion is as inauthentic as it is unnecessary. Kernewek Kemmyn looks wrong because it is wrong.

Phonological errors in Kernewek Kemmyn

The spelling of Kernewek Kemmyn with its novel use of {i} and {y}, its {oe} and its curious doubling of letters is based on certain mistaken assumptions about the the sound-system of MC. These assumptions can be summarised as follows:

1. Kernewek Kemmyn is unaware of the alternation of vowels in monosyllables and disyllables: byw but bewnans, bedheugh but bydh, etc. In fact Kernewek Kemmyn is inconsistent since it writes blydhen 'year' not bledhen, gwydhenn 'tree' not gwedhenn, hwyja 'vomit' not hweja but it does write Kernewek rather than *Kernywek and kewsel rather than *kywsel. I have discussed this whole question elsewhere and do not intend to say anything more about it here except this: Kernewek Kemmyn is badly mistaken on this point because the deviser does not seem to know about the vocalic alternation in Welsh which also occurs in Cornish. The alternation has been largely lost in Breton and this may in part explain why the deviser of Kernewek Kemmyn is ignorant of this essential feature of MC.

2. The second assumption wrongly made by Kernewek Kemmyn is even more serious. Kernewek Kemmyn assumes that Cornish, like Welsh and Breton had three lengths of vowel, short, half-long and long. Kernewek Kemmyn also assumes that this system of three lengths disapppeared in the early 17th century (ca 1600) since it is clear that Late Cornish had only two lengths, short and long as in English. Kernewek Kemmyn is badly astray on this point.

The reshaping of vowel length could not possibly have happened ca 1600 for several reasons. The loss of half-length was a major change in the language. If it had occurred ca 1600, one would expect to find vast differences in the sound system of John Tregear (ca 1555) and The Creation of the World (1612). Such differences are wholly absent - indeed The Creation of the World is more archaic in ways than Beunans Meriasek (1504). Secondly we know from Lhuyd that vowels in monosyllables had lengthened before ca 1690. They can hardly have shortened ca 1600 only to lengthen again within 80 or so years. The change from three lengths to two lengths (the Prosodic Shift) is part of the same movement seen in the assibilation of tad to tas. It occurred at the end of the Old Cornish period under the impact of Middle English. It is overwhelmingly apparent from the MC texts themselves that there was no half-length in MC.
This question is absolutely crucial for understanding why Kernewek Kemmyn is mistaken but it is rather technical. Welsh and Breton have three possible lengths of vowel: long, half-long and short. This means that vowels can either be of a single length (short), of a double length (half-long) or of a triple length (long). Take the Welsh word tad 'father'. That has a long vowel. You can almost here the three a's in it: taaad. The plural tadau 'fathers' has a half-long vowel: taadau. It is of only two lengths. The word mam 'mother' and the plural mamau 'mothers' have a short vowel of one length only. I have taken my examples from Welsh but Breton is similar.
Notice two things. In the first instance native speakers feel that half-long and long are really aspects of the same thing since neither is short. Vowels to the native-speaker are either short or not-short. Modern Breton grammars actually go one further and treat half-long as the same as long. The second thing to notice is that a unit of length is known as a mora, the Latin word for delay. Tad therefore has a trimoric vowel, whereas tadau is bimoric and mam is unimoric.
Old Cornish had this threefold system. Under the impact of Middle English it appears that Middle Cornish lost the threefold distinction and acquired a simple twofold one, long and short. In fact what happened was this: stressed syllables where possible were all shortened but were correspondingly more vigorously pronounced. This meant that three morae became two morae and two morae became one mora. This process I refer to as the Prosodic Shift. Take a word like myres, myras 'to look' and the monosyllabic myr! 'look!' or ef a vyr 'he looks'. Before the shift the monosyllable myr! look! had three morae: miiir. The dissyllable myres, myras had two morae: miires. After the shift miiir with three morae reduced to two: miir, whereas miires reduced to one mires. Miir is still not-short. It is still longish, because it still has two morae, the length of the original half-long.
The stressed vowel in mires {myres} on the other hand is now short. The vowel was half-long and is now short. A short vowel is relatively more lax and less tense than its long counterpart. This means that short i is open, i.e. the tongue and organs of speech are slacker than for long i. Compare English sheep with English ship. The open slack vowel in mires is so much lower in the mouth than the vowel in miir that it is near to e. Short y and short e are not always kept separate in Middle Cornish.
If what I am saying is correct, we should expect in the MC texts to find myr! 'look!' and ef a vyr 'he looks' always spelt with {y}, but the dissyllabic forms meras, veras to be spelt on occasion with {e}. This is precisely what we do find. Notice the following examples of the disyllable with e: meras PA 215d, TH 49a, 50a; veras PA 168b, OM 2325, BM 4074, TH 2, 3, 7, 9; veres BM 4351; verys PC 1257 and the following examples of the monosyllable with y: myr BM 935, 1805, 3194, 3229, 3270, 3656; cf. also the monosyllabic form in my a vyr 'I will look' OM 1251. There are instances of myras with {y}, but none of *mer with {e}. This indicates that e and y in such items are not in free variation. Meras clearly has a short vowel. If so, Kernewek Kemmyn is mistaken in believing that it has a half-long one and therefore in spelling it with the same vowel /i/ as the monosyllable mir. Kernewek Kemmyn mires is therefore wrong.
But there are very many other words which Kernewek Kemmyn spells with {i} when it is quite clear that the vowel is not /i/ but /I/ {y} or /e/ {e}. Examples include whelas 'to seek', y gela 'the other', screfa 'to write' but scryf! 'write!'. The word trega 'dwell' is interesting. Dr George is perplexed by the way in which this word frequently has e/y rather than /i/ and he admits he doesn't understand why. At all events he spells it with {y} rather than {i}. In fact of course trega has e in disyllables only. In monosyllables the vowel is always y. Note the following from John Tregear: may hallan ny bewa ha trega in charite, ha nena ny a dryg in du ha gans du, ha du a dryg innan ny 'that we may dwell in charity and then we will dwell in God and with God and God will dwell in us' (TH 30). The monosyllable dryg < tryg has two morae: triig, whereas the dissyllable has one mora in the stressed vowel: trega < tryga. That one sentence is enough to show that in TH there was no half-length. Similar examples could be taken from BM and the rest of the texts. If Middle Cornish had no half-length and it didn't, then the arbitrary rewriting of the spelling system in Kernewek Kemmyn is not only unnecessary it is mistaken.
There are many further errors in Kernewek Kemmyn that are a direct result of the same fundamental error, namely the assumption that Middle Cornish had half-length. I have set them out in detail in Cornish Today and elsewhere. One mistake in particular I have already mentioned. That is the belief that Middle Cornish had two different long o sounds spelt {o} and {oe} in Kernewek Kemmyn respectively. Perhaps in the discussion after this talk there will be an opportunity to explain how Kernewek Kemmyn has misread the evidence and why therefore such spellings as koes 'wood', loes 'grey', hwoer 'sister, ev a woer ' he knows', oen 'lamb', etc. are inauthentic. The editor of the KDL Newsletter by the way agrees with me (and Nance and Jenner and Caradar) on this point. In January of this year he wrote to Dr George and sent me a copy of the letter. He wrote:

Mes gans hemma oll kales yw krysi nag yw an gwir gans N[icholas] W[illiams]... Mars yw gwir nag eus marnas unn son o, ni a yll defendya yn mes an lytherennans "oe" yw nebes hager dell yw res avowa ha ny vydh res perthi kov mars yw res skrifa po ha leverel koedh po kodh. Ytho nowodhow da vydh

[All the same it is difficult to believe that Nicholas Williams is not right...If it is true that there is only one sound o, we can remove the spelling "oe" which, it must be admitted, is rather unsightly and it will not be necessary to remember whether to write {o} or {oe} and to say "koedh" or "kodh". It will therefore be a change for the better].

Unified Cornish is greatly superior to Kernewek Kemmyn and should never have been abandoned by the CLB. I have recommended a number of minor emendations to the spelling of Unified Cornish. These have been accepted by Agan Tavas. I also suggest simplifying the construction of sentences in Unified Cornish in line with the language of John Tregear and the other Tudor texts. I am currently, with the approval of Agan Tavas, working on a handbook of revised Unified Cornish or Kernowek Ewnys. I hope the book will be finished before the end of this year and will be available early in 1997. Emending Unified Cornish is sensible. Rewriting Cornish is not.

Before I finish let me make a plea to you all. If you use Kernewek Kemmyn, please give it up as soon as you can. If you are thinking of learning it, don't. If you hesitate between Kernewek Kemmyn and UC, stick with Unified. The revival cannot afford three systems, particularly when the most widely used form of the language is so mistaken. Cornwall needs her own language but that language is not Kernewek Kemmyn. The time has come to begin the retreat from Kernewek Kemmyn. The revival will achieve neither authenticity, academic respectability nor unity until Kernewek Kemmyn is rejected. Only when Kernewek Kemmyn is abandoned, will Cornish speakers find peace of mind.

N.B. Y'n text a-uhon yma grafemow dysquedhys ynter camvahow crullys: { ha }.

[ Tre arta ]