[Spellyans] An Abecedary Kernowek

Michael Everson everson at evertype.com
Sun Mar 29 16:17:02 BST 2009

On 29 Mar 2009, at 14:34, <ajtrim at msn.com> <ajtrim at msn.com> wrote:

> All vowels have their long sound as their name. That's OK, I think.
> For the long and short forms, I would just add ber or hir, rather  
> than trying to describe the diacritical mark used.

We need terms for the marks. Note too that circumflex is not only a  
mark of length; in some cases it marks quantity.

> For the alternative forms ë and ÿ, I would add eylyans.

I much prefer dewboynt.

> All consonants use the vowel, e (sometimes long; sometimes short),  
> except for q.
> Why not make q = qwê?

For one because q doesn't have the sound of [kw] unless the w is  
written with it.

> Most consonant names start with the consonant and have -ê suffixed.
> Why don't f, h, l, m, n, r, s, x conform?

Because of the history of the alphabet. Leaving aside "h" and "x", the  
alphabet which the Romans inherited from the Etruscans added -e to the  
letters except for those sonants which could be pronounced without a  
vowel [f:: l:: m:: n:: r::]. In letter-naming throughout Europe, these  
get an epenthetic e-.

> It's because these consonants can be pronounced long, so to  
> emphasise these, we start with a short e, we suffix the consonant,  
> and we pronounce it long.
> x is an exception to these "rules". It should be called xê.

What? [kse:]? I really do prefer [ɛks], and I think most would agree.

> v could be ev but it is called vê to distinguish it from ef. That's  
> OK.

And conforms with standard European practice. (I know the Welsh do  
that weird thing saying "ec" instead of "ce" for c, but that's  

> h could be eh. However, this is difficult to say distinctly, so I  
> recommend egh.

I think hâ does better.

> Teaching the alphabet will then reinforce the connection between h  
> and gh, and it would help to teach the sound of gh at the same time.
> The name, êch does not conform, and it isn't helpful.

Agreed; êch derives from French, and hâ seems sensible enough.


> You could use cû for the letter, and use qwê for the combination qw  
> in spelling. The word qwil then becomes qwê î ell.

That would be a bit unprecedented.
I tend to favour "cû wê î ell" or "kyû wê î ell".

> You may wish to consider m = ebm, and n = edn.

I think that would be a mistake. Every one including RLC speakers  
calls these [ɛm] and [ɛn]. In KS we write this easily, èm, èn.

> These would conform with the long consonant rule, and it would help  
> to teach the correct sound of bm and that of dn, where the b and d  
> should be incompletely exploded.

That should be learned in words which pre-occlude. There's no evidence  
that the letters m and n had pre-occluding names.

> So here is my suggestion:
> â, bê, cê, dê, ê, ef, gê, egh, î, jê, kê, ell, ebm, edn, ô,  
> pê, cû, èr, èss, tê, û, vê, wê, xê, yê, zê, (with qwê for  
> qw in spelling out words because you never use q alone.)
> Then, è would be è bèr, ê would be ê hir, ë would be ê  
> eylyans, etc.

"Alternation e"? I think "two-point e" is much better.

I had a chat offline with Eddie and so far he and I still think the  
best run is:
â, bê, cê, dê, ê, ef, gê, hâ, î, jê, kê, ell, èm, èn, ô,  
pê, cû, èr, èss, tê, û, vê, wê, ex, yê, zê

... though whether cû or kyû is better remains uncertain.

Michael Everson * http://www.evertype.com

More information about the Spellyans mailing list