[Spellyans] Late Cornish adaptations

Owen Cook owen.e.cook at gmail.com
Fri Aug 1 13:56:21 BST 2008

2008/8/1 Michael Everson <everson at evertype.com> rug scrifa:
> I would write and recommend <hy>, <cry>, <chy> for these words. <ei>
> is a Lhuydian innovation; I can see a use for this graph, if one were
> writing dialogue where one character had a strong Penwith accent or
> something... but I agree with Neil, there's no *need* to write <ei>,

Yes, I agree completely. The SWF also generously "permits" apostrophes
in words like for' and por'h, where there is absolutely no need for
them. Late Cornish readers will know that <dh> is usually not
pronounced in any word ending in <rdh>; the apostrophe only divides
Cornish users and benefits no one (save RMC users ignorant of RLC
pronunciation). In the same way, ffordd is normally /fOr/ in modern
spoken Welsh, but nobody writes this as ffor'.

> Having said that, it's unambiguous in terms of interpretation when
> encountered and so no problem for KS. In principle one might write
> <hy>, <crÿ>, <chÿ>.

Huh? There is no alternation with <ë> here...

> Nor I, and again I think KS should discourage this: <brâs>, <gwlân>,
> <clâv> is a better solution.

Since we already use <â> with a different significance (i.e. irregular
length), would it not be a good idea to use <å> instead? Length here
is not irregular, but the vowel quality is. The ring over <å> also
evokes the o in variants like broas or braos. I feel as though I have
also seen <å> in RLC for Lhuyd's turned a, but perhaps this was my

Unlike Craig, I do think words like these have to be specially marked.
Michael has found us a minimal pair, so we do have to do with a
phonemic distinction.

> I object here, Eddie. If you say ['beUn at ns], write <bewnans>. If you
> say ['boUn at ns], write <bownans>. This is a dialect feature. I suppose
> we could write <bëwnans> and <böwnans> (with diaeresis to show
> alternation) but you're adding more diacritics then, and you'd end up
> with <Kernëwek>~<Kernöwek> and I don't know if that's a credible
> solution.

The word Kernowek is not in the ew~ow set. It is in the Jowan,
lowarth, lowen set. We see some early evidence for /ow/ and a lot of
late evidence for /u:/. However entrenched it may now be among the UC
and KK user communities, *Kernewek is wrong.

As for the ew~ow set as in bewnans, I'm inclined to think that we
already have an umbrella graph: <ew>. For any <ew> words could be
found that remained stable in Late Cornish as /ew/, we could press
<êw> into service. I can only find a handful of such words in
Gendall's dictionary:

RG bêu, KS bew 'alive'
RG bêu 'owns'
RG bleau 'hair'
RG deau 'two', KS dew 'two'
RG Deu, KS Duw 'God'
RG knêu 'fleece'
RG kynêu 'young animal'
RG lêu, KS lew 'rudder', 'lion'
RG plêu, KS plu 'parish'
RG reau, KS rew 'frost'
RG têu, KS tew 'fat'

Oll an gwelha,

More information about the Spellyans mailing list