[Spellyans] 'who?' in Cornish

[Daniel Prohaska]

> On 09.04.2021, at 13:29, Nicholas Williams <njawilliams at gmail.com> wrote:
> 
> My objection to the spellings <piw> ‘who’, <gwiw> ‘worthy’ and <liw> ‘colour’ in the Akademi dictionary is not just a quibble about orthography. It is a question of the implied pronunciation. If the rules were followed fully the diphthong in <piw> would be pronounced with a long nucleus. It is certain, however, that in ‘who’, ‘worthy’ and ‘colour’ in Middle Cornish the nucleus was short.

Nicholas, you say it is “certain” that the nucleus of a diphthong in open syllable was short. I do not know how you can be “certain” about this. I’m not sure even fits your overall theory. My two objections to this would be that Lhuyd and native writers show several spellings that would rather indicate a long nucleus in this environment. Also, since the Middle English diphthongs /iu/ and /eu/ fell in with each other at the end of the 13th century, ME and early ModE /iu/ (e.g. ‹newe› “new”) could be represented by ‹ew›. I cannot see how a short, or long for that matter, nucleus could be determined with “certainty” without direct observation, or at least commented observation at  the time Cornish was still traditionally and natively spoken.  And Lhuyd is the closest thing we have to such a comment. To me the evidence appears stronger that the nucleus was relatively long in this environment, allowing for a greater range of contrastive nuclei.   


> As a result the vowel written <y> alternated with /e/.
> We thus find, for example:
> 
> pyw OM 261 ~ pew TH 7
> gwyw PC 829 ~ guew BK 204
> lyw PC 3123 ~ lew CW 1051.
> 
> Indeed lyw ‘colour’ clearly rhymes for effect in the line du yw y lyw RD 2001 both with du and yw.  In the SWF if ‘colour’ and ‘is’ are liw and yw respectively, they do not rhyme. 

The natural rhythm of speech would emphasize ‹du› and ‹lyw›, - ‹yw› on the other hand would be unstressed in this context and its vowel/diphthong reduced/lowered. ‹Lyw› and ‹yw› need not rhyme perfectly to make this line work. What it does show is that /y/ was diphthongised early with the expected /iw/ diphthong: [ˈdiːw ɪw ɪ ˈliːʊ] (or if you don’t like the overlong [iːʊ], with a reduced length (< prosodic shift) [iˑʊ] - but I didn’t want to trigger you with representing a half-long vowel ;-) ). 
Phonologically the diphthong in ‹lyw› and ‹yw› are no different, i.e. /iw/, but ‹yw› is short by being unstressed, while ‹lyw› would have its fully stressed and long pronunciation. I say this with no degree of certainty, but evidence for long nuclei are there in the later texts. You have mentioned this problem, if memory serves, in “Cornish Today” and proposed a secondary lengthening in this environment, but I see no problem in assuming the nucleus was always (even after the prosodic shift) relatively long (in this environment) vis á vis a short vowel. Note that I’m not arguing specifically for spelling the digraph ‹iw›, just that there was a phonological distinction between /iw/ and /ew/.  



> 	The spellings piw, gwiw and liw occur in the SWF because such spellings were used for Kernewek Kemyn. The <iw> appeared in these three etyma in Kernewek Kemyn because the spelling system was constructed inter alia around the mistaken view that the nucleus in such words was long (or at least half-long). As a result the ensuing erroneous phonology imitated Breton which has piv ‘who’, liv ‘colour’ and gwiv ‘alive, lively’. In my view, and I have been writing about this since the 1990s, is that it is illegitimate to spell Cornish as though it were Breton. The Cornish scribes knew how to write their own language and we should as far as possible follow them. 

You are making my point for me… I personally agree, we should write Cornish as the scribes did, yet the phonological argument you make is far from certain in my opinion. Since you also alternate ‹i› and ‹y› in KS according to position, I don’t see what the huge problem is, with the grapheme ‹iw›. 


> 	The SWF (M) is the offspring of Kernewek Kemyn. Kernewek Kemyn was not chosen to be the official orthography and pronunciation for revived Cornish because it was seen to be far too alien. In 1995 the late Peter Pool tellingly wrote:
> 
> Kemyn 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, as indeed to those who prefer Kernuak, Kemyn has an alien and somewhat sinister appearance, as i f the language had somehow been taken over by robots and reduced to the status of a code.
> 
> The same, to a lesser degree, can be said of the SWF. Until the SWF is revised in a more authentic direction it will not gain universal acceptance. 

The SWF, for better or worse, is widely accepted, even if not particularly well liked, on all sides, for different reasons - as the lesser evil of not having a standard. But to suggest there was no contrast between an /iw/ diphthong and a /ew/ diphthong (in this environment) is untenable, not even to speak of certainty. I would agree with you, and I think, judging by the new edition of his Gerlyver Meur, that Ken has come around to this, too, that the threeway contrast /iw/ : /ɪw/ : /ɛw/ that KK makes/has made cannot be supported. 

The environment you describe essentially works like a long vowel, and long vowels (e.g. ‹da›) have not been shortened by the prosodic shift (by one mora you say, but that doesn’t make them short - and neither did it happen in monosyllabic words with a closed syllable, e.g. ‹tas›). 

On a side note, I discussed with Ken his transcriptions of diphthongs in open syllables and criticised that, systematically, his half-long vowels in open syllables of polysyllabic words, should also show half-length, which he never transcribed; e.g. ‹lowen› **[ˈlɔˑwɛn] would be a consistent pre-prosodic shift transcription, yet he gives [ˈlɔʊɛn] (with Ken’s weird turned upper case o-mega for ‹ʊ›). He transcribes KK ‹piw› as [piw] which also doesn’t line up with the rest of his phonology, which should show **[piːw] or **[piːʊ]. Phonologically his /piw/ would be correct in the context of KK. 

Dan




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://kernowek.net/pipermail/spellyans_kernowek.net/attachments/20210410/42a0b4e4/attachment.htm>