[Spellyans] th/dh

Daniel Prohaska daniel at ryan-prohaska.com
Sat Nov 20 17:02:08 GMT 2010

-----Original Message-----
From: Michael Everson
Sent: Saturday, November 20, 2010 12:32 PM

"On 20 Nov 2010, at 11:01, Daniel Prohaska wrote:

> Sorry Nicholas, I didn’t mean to hostile against you – it was probably just a little frustration creeping in because I wasn’t getting through to you.


Nicholas and I both understand you."


Good, I feel we’re just having a good old heated academic argument over our pet theories ;-) so no personal slur was intended.


“> What I meant was, that if your solution is implemented orthographically, Cornish will write final <dh> in stressed syllables and final <th> (for etymological /ð/) in unstressed syllables.


That's final <dh> (for etymological /ð/) in stressed syllables and final <th> (for etymological /ð/) in unstressed syllables. Because the evidence (of /ð/ as well as /f/ and the stops) suggests that in unstressed syllables the consonant devoices.”


But there is also evidence that it didn’t, or didn’t necessarily so, per Lhuyd, who is our only direct evidence for /ð/ as a whole.  


“> Thus your theory is manifested orthographically, basically forcing the Cornish learner to accept it for real because s/he will not know which final <th> in unstressed syllables is etymological /θ/ and which is /ð/.


People speak synchronically.”


Yes, and orthographies are often historicising. Any more general statements that come to mind? 


“> This is fine, of course, for speakers who devoice in this position and for people who believe the phonemic contrast to have been lost in this position.

> For those who don’t, i.e. the speakers who devoice only in pause or don’t devoice etymological /ð/ at all, the orthographic representation of your theory is useless.


Who are these people, Dan? You speak as though there is a community of native speakers out there, passing the language on from generation to generation. If George hadn't revised the phonology according to his preference (etymologize in the direction of Welsh and Breton) this wouldn't be an issue. But George's phonology wasn't taken up, apart from errors like [ɪ] for what we write <ÿ>.”


[ɪ] is not an error. It existed in Cornish. And no I don’t speak as if there’s a community of native speakers. There are people though that use final voiced [ð] in unstressed syllables. There are people that don’t. If it were for UC along no one would be saying [deːð], but [deθ], which I also consider to be incorrect.  


Going by Lhuyd, George had some evidence to go by to conclude that [ð] was possible in that environment, as did Nicholas when he wrote the first edition of CT, though he later changed his mind about the issue. 


“> While the former would be able to read their [ˈdɪʊəθ] as <dyweth> (or <diweth>), they would wonder why the word is thus spelt when saying [pæn ˈɛɾə ˈdɪʊəð ən ˈvlɛðən].


Where do you get this Breton-style sandhi across word-boundaries from? I don't believe there is evidence for it in the texts.”


Edward Lhuyd’s transcription of JCH. 


“> The latter group would be puzzled anyway because they wouldn’t understand why their final [ð] in syllables would be written <dh> in stressed syllables, but <th> in unstressed syllables.


> My solution to write <dh> but devoice by rule is much more inclusive, will give speakers a choice, help the learner make a connection between simplex and derivative.


But it abandons traditional orthographic forms (as Ray has noted)”


But we have already abandoned traditional orthographic forms by using <dh> in the first place. The texts have <th> for both /ð/ and /θ/, yes and the few instances of <ȝ> can represent /ð/ but also /θ/.


“and in fact, although it is all very nice for you to say "they can learn a rule that says 'devoice in final unstressed syllables' the reality of it is that such a spelling will inevitably lead to people voicing the segment, which we don't in fact recommend.”


You don’t recommend because of the opinion (or theory) you hold that /ð/ in this position was [θ], which is why I say that you are pushing your theory on the Revival; if you spell <dh> there’s a cop out because you can devoice by rule, if you write <th> you are codifying Nicholas’ opinion, forcing those who don’t want to devoice to devoice. 


“Because the evidence of the texts suggests that it was devoiced.” 


The evidence from Lhuyd, who was the only one to write <dh> in the first place suggest it wasn’t. The texts didn’t distinguish between <th> and <dh> at all, which is why it is impossible for them to distinguish them in final position. I can repeat this sentence as often as you can yours.


“Your rule is not orthogonal to the way other consonants are treated in the orthographic system. Kernowek, Kernoweger. Nowyth, nowodhow.”


These sounds are not analogous. The g/k pair is a stop. The native vocabulary of Cornish didn’t distinguish the two in final position, apart from the absolute initial position they were allophones. /θ/ and /ð/ were distinct phonemes. Maybe this accounts for the differing treatment.  


“> People who generally devoice in final unstressed position can easily do so by rule,


They'll fail and start saying /ð/ just as the KKers fail to say dÿdh with a long vowel.”


 That was in the crystal ball you keep on your couch table? ;-)


“> speakers who devoice only in pausa and before voiceless consonants can do so by rule and context, people who generally don’t devoice use a reading pronunciation of the orthography.


Again you talk as though there were a community of native speakers.”


No, I don’t. But the fact remains, that there are Cornish speakers who have [ð] and [θ] in this position. The SWF is supposed to be inclusive. Sorry, the evidence to me isn’t strong enough to suggest that final [ð] did not occur in traditional Cornish. That’s why I cannot extrapolate a rule from g/k or b/p (which isn’t even as consistently applied as the g/k pair) and apply it to <th>. Again, absence of evidence isn’t evidence of absence.  


“> My solution is more inclusive, it is less “dogmatic” (because it doesn’t force one theory upon the speaker), and it is workable. Since <dh> wasn’t used in the MC MSS anyway it is not even less authentic or traditional.


I don't believe that Lhuyd's examples prove your case. As I have said, the simpler explanation for his anomalous use of both <dh> and <th> in the same words is influence of Welsh on the former.”


I don’t know if Lhuyd’s evidence proves my case, I’m not 100 % sure (how could I be), but doubt remains, and I cannot in good conscience apply a rule that orthographically enshrines devoiced final [θ] for etymological /ð/ when there is evidence to the contrary. 




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://kernowek.net/pipermail/spellyans_kernowek.net/attachments/20101120/2f4900e0/attachment-0001.html>

More information about the Spellyans mailing list