[Spellyans] th/dh

Daniel Prohaska daniel at ryan-prohaska.com
Fri Nov 19 09:28:41 GMT 2010

May I remind you that in CT 8.15 and 8.18 you write: 


   “8.15 As far as the continuants /ð/ and /z/ in JCH are concerned, it appears that there is a strong tendency to generalize the voiced form everywhere. Although Lhuyd writes noueth ‘new’ at §6 and §16 and diuath ‘end’ at §13, he writes duadh, diuadh, diuedh, diwadh ‘end’ at §§3, 5, 7, 9, 29, and 46. The final segment in medh ‘said’ is <dh> throughout and in vêdh ‘will be’ at §§15, 20, 24. 

   If one looks at the other spellings elsewhere in Lhuyd’s Cornish Grammar (AB vi) one finds the following: goruedh ‘lie’ (250a), dyhodzhedh ‘afternoon’ (249a), auêdh, enuêdh ‘also’ (249a), eledh ‘angels’ (249b), heruydh ‘according to’ (249c), hemladh ‘fight’ (249c), bedh ‘be!’ (245c), klêdh ‘ditch’ (242b), gurêdh ‘root’ (242b), blaidh ‘wolf’ (241b), menedh ‘mountain’ (230c), guragedh ‘wives’ (243a), brederedh ‘brothers’ (243a), abesteledh ‘apostles’ (243a).

   One might argue that Lhuyd was mistaken in hearing final /ð/ in bedh, dedh, etc. and that the analogy of Welsh led him astray. This is unlikely, I think. Lhuyd on a number of occasions makes it perfectly clear that what he heard on the lips of Cornish-speakers and what was written in Cornish manuscripts were at variance with one another. Lhuyd makes the following three statements:


1.	‘…for deyth [a day] must be read dedh’ (AB:227)
2.	‘For Aflauar, mute; yn few, alive, & Ty a fyth, Thou shalt be, must be read avlavar, in vêu, tî a vydh’ (AB:227c)
3.	‘Th supplied the use of Dh in the Cornish as in English. For Daveth, David; Forth, A way; Deth, A day; Fyth, Faith; Ethen, Birds; Goith, A goose, &c. should be read Davydh, Fordh, Dêdh, Fydh, Ydhyn, Gûydh’ (AB: 229b)


Moreover it is clear that he distinguished carfully between /ð/ and /θ/ in the Cornish he heard spoken for he says of ancient Cornish manuscripts:


‘For Diwet, an end, is pronounced Diuedh, and bitqueth, Ever, should be read Bithqueth…’ (AB:229a).


   8.16 It is apparent that in the Cornish of Lhuyd’s time the reflex of final /ð/ was /ð/ after stressed and unstressed syllables alike. This was the view taken by Williams in LCB. Whitley Stokes takes issue with Williams on this point for he says:


But I have invariably written th in auslaut. In this respect Mr. Williams has throughout his Lexicon been mislead by Welsh analogy, and failed to see that at the end of a word an old dh (ð) has always been sharpened into th (p). In the Passion, with the single exception of molloȝ, P.66, 3, (ȝ = dh) never occurs as a final, but only a th. Moreover, words which, in Old Cornish, may have ended in dh, rhyme in Middle Cornish poems with words which unquestionably end in th. See for example, the Passion, stanza 49, 4, where beth ‘erit’ (W. bydd) and feth (W. fydd from fides) rhyme with haneth (W. henoeth, Ir. innocht) ‘to-night’ and tergweth (W. teirgwaith) ‘thrice’. I conclude from this that in all word sending, according to the MSS, in th, that combination was pronounced sharp, as in Welsh, and should be so printed (CG: 138).


   I would suggest that both Stokes and Williams were right. Cornish originally devoiced final /ð/ in most positions and words in final etymological /ð/ could therefore be rhymed with words in final /θ/. As a result of the prosodic shift (see 8.19 below), however, final /ð/ was revoiced in those final positions where it had previously been devoiced. At first this revoicing may have applied only after stressed vowels. Later it was extended to unstressed vowels as well. The original devoicing, however, was allowed metri gratia in Middle Cornish verse long after it had ceased to reflect the spoken language.

    We can sum up the position best by giving some examples. Before the prosodic shift: cleth ‘ditch’ was /kleːθ/ before consonants and in pausa and /kleːð/ before vowels. Meneth ‘mountain’ was /ˈmeneθ/ in pausa and before consonants for a while, but eventually final /ð/ was generalized. Thus meneth became /ˈmeneð/ everywhere also. The only indication that the generalisation of final /ð/ was recent was the following: for the purposes of rhyme cleth, meneth could be treated as though ending in /θ/ rather than /ð/. 

   It should be noticed, however, that Lhuyd spells the word for ‘truth’ with a final <th>: gwyroneth (AB: 240c) although the Welsh cognate, gwirionedd has final /ð/. It seems that while gwyroneth was still pronounced with final /θ/ in pausa, the word was assimilated to other abstracts in -eth /eθ/ < *-akta. Such abstract nouns in -eth (Welsh -aeth) include skiantoleth ‘prudence’, gowegneth ‘fraud’ (AB: 240c).”   


If I were to be sarcastic I could ask you why you were bending over backwards to defend KK spellings, but I accept and realise that you weren’t. Please accept this of me also and take note that I have arrived at the same conclusions you had in the mid 90s independently. Yes, I had read CT intensively, but it’s been years and I’d forgotten about this particular passage. Our little tiff here made me rummage through CT again and I was quite surprised to find this.  


In the third edition of CT you added a footnote that you had changed this position: 


“7 A thorough examination of the evidence shows that Lhuyd did not really hear unstressed -edh as a plural termination. He says “The Fourth Plural Terminated formerly (as still in the Welsh) in edh; as Brederedh, Brothers; Eledh, Angels; Abesteledh, Apostles. Which pronunciation was more anciently expressed by t; as Guraget, Wives for Geragedh. It’s at present changed into es according to their writing; but into ez according to their Pronunciation” AB: 243a. I think it likely that the variation stressed vowel + b, g ~ unstressed vowel + p, k (for example mab, gwreg ~ morrep, carrek) also operated with ð and θ. I now, therefore, write fedh ‘faith’, gwedh ‘trees’, and scodh ‘shoulder’ but eleth ‘angels’, bugeleth ‘shepherds’, deweth ‘end’ and gwyryoneth, gwyroneth ‘truth’ in the vocabulary of UCR at the end of this book.”


And this is also the position you hold and defend today.



From: nicholas williams
Sent: Friday, November 19, 2010 9:13 AM


“I have an explanation for the th/dh which is the same as that for v/f. And for b/p and g/k. You appear to agree with v/f, b/p and g/k. You seem to allow that the variation is a question of the preceding vowel's being stressed or unstressed. But in the case of th/dh you adduce a rule about position and pausa, for which there is not a shred of evidence—“


Well, you seemed to think so in the mid 90s. And why do you think there is no explanation for the different treatment of b/p and g/k? There was no voiced ~ voiceless contrast here and the alternation started as an allophonic realisation, whereas dh : th always were contrastive. So the system of voiced : voiceless opposition was asymmetrical when comparing the stops and the fricatives. It is difficult to say where v/f are concerned, because of the MC scribal conventions and the frequent loss of final /v/ in unstressed syllables. Lhuyd at least gives us a look into the possible retention of a final contrast dh : th.


”and for which there are no parallels in the system. In the one case it's historical. In the other it's position. It looks as though you are "bending over backwards" to defend the KK based spelling of the SWF.”


Again, were you bending over backwards to defend KK based spellings when you wrote CT? No, I didn’t think so. Please give me the benefit of the doubt when I claim the same. Thanks.


“And not very convincingly.”


Well, I see Lhuyd’s spellings and I stop and think and say, hey looks like Cornish had /ð/ in word final position after all. So, your extending the voiced/voiceless alternation by analogy with g/k and b/p is not very convincing to me either.


“I agree that -eth (< ed) is -edh before vowels, e.g. dewedha 'to finish'. That is already in UC. What I will not agree is that the simplex is dywedh rather than dyweth.”


It there! Attested several times!!! Not a shred of evidence, I have explained that Lhuyd’s diuadh and similarly spelt forms with final dh occur in set phrases in JCH, but the only occurrence of this word with final th, a pausa is imaginable. 


“I regret that we are never going to agree and I am certainly not going to write the SWF.



Fine. At least this is an honest statement. So you, like Ken George do not want to use the SWF. You both think it is inferior to your respectively preferred orthographies. It’s really not such a huge problem and I explained several times how a compromise between these two positions can be achieved. Write <dh> in this position and devoice by rule. Easy. It’s done elsewhere, why not here. 



On 2010 Du 19, at 01:09, Daniel Prohaska wrote:

No, Nicholas, it doesn’t. You’re interpreting with your phonological model in the back of your mind. You’re not willing to let go of that – you needn’t, not if it were even shown in a Revived Cornish orthography, as this ca be done by a simple rule. As I said, I’m not even arguing against the possible unvoicing in this position – in fact I show it in my dictionary, but I also accept that the voiced sound can re-appear before vowels and in derivatives. What’s so difficult to understand about that?



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://kernowek.net/pipermail/spellyans_kernowek.net/attachments/20101119/931ba888/attachment-0001.html>

More information about the Spellyans mailing list