[Spellyans] -ita

Michael Everson everson at evertype.com
Tue Jul 1 08:16:03 IST 2008

At 01:19 +0100 2008-07-01, A. J. Trim wrote:

>>(1) Initially <i> is [I]; initial [i:] would be 
>>written <î> (modulo <ÿs/ës>). This rule 
>>increases the visibility of the letter <i> 
>>considerably -- a concern many RLC users have 
>>expressed to us.
>Rule (1) is crazy for Middle Cornish folk.

I beg to differ. Initial <i> is common in the 
texts. So is initial <y>. You may have a 
preference for <y> in general if you learnt 
Unified Cornish, because Nance didn't use <i> at 

>Best to use <y-> for initial [I] and <i-> for initial [i:].

In the first place there aren't all that many 
words with either, though some of them are fairly 
high-frequency words.

>RLC folk could use <i> instead of <y>, throughout the system, if they want to.

Do you suggest that all of these words could have 
two different spellings? There's no advantage to 

>They would then need to mark long <i> with a diacritical mark, perhaps a
>circumflex, thus <î>.

Frankly the RLC folk have a lot to adjust to in 
the SWF and one of the reasons has to do with the 
distribution of <i> and <y>. For whatever reason, 
a number of them feel that "y is medieval" and "i 
is modern". That's not a very sophisticated view 
(and neither is your view that "initial i is 
crazy for Middle Cornish folk) but it *is* 
something we can address, and something we 
*should* address.

Initial [i:} is not common at all, so it would be 
marked <î> if it occurs. Initial [I] is rather 
common, so initial <i> wouldn't be ambiguous. And 
some of the function words which use what we 
propose to keep initial <y> for have [@] rather 
than [I]. Moreover, distinguishing the 
preposition <in> which does not mutate the 
following word from the particle <yn> which does 
mutate the following word makes the orthography 
much more user-friendly for learners and teachers 

>>(2) In monosyllables and their derivatives <i> 
>>is [i:] and <y> is [I] (modulo <bÿs>/<bës>. 
>>When unstressed, these <i> letters are 
>>shortened to [I].
>Rule (2) is OK (but see (3) below.) Not sure 
>about bÿs/bës yet. See Note below.

<bÿs>/<bës> solves a fundamental problem that is 
otherwise insoluble given the option <bys>/<bes> 
in the SWF.

>>(3) In polysyllables not derived from 
>>monosyllables, the sound is almost invariably 
>>[I]. In final unstressed position, <y> will be 
>>[I]~[@] and <i> will indiate that the vowel is 
>>[i:] and that the syllable is stressed 
>>(crodo'dil, pl. croco'dilys; con'strin, ppl 
>>con'strinys). Elsewhere in medial position <y> 
>>will be used since the vowel is almost always 
>>[I]; in those rarer instances where the vowel 
>>is [i:], it can be marked with the circumflex 
>><î> (bîbel (if not beybel) sîra, personal 
>Rule (3) is OK for Middle Cornish folk.
>However, it would not be necessary to mark the <i> in words like Bibel and
>sira, as the <i> is already regularly long [i:].

Yes, it would, I think, because <i> is not 
regularly long [i:] in polysyllables because of 
Rule (2).

>Marking is only thought to be necessary because 
>you want to say that the <i> in words like 
>tiryow (from tir) and gwiryoneth (from gwir) are 

That's correct.

>I would rather write these tyryow and gwyryoneth but, either way, I believe
>that any confusion would be minimal.

If you learned UC or UCR, you're already writing 
<tyr> and <tyryow> and you've learnt to pronounce 
one long and one short. So changing this to <tir> 
and <tiryow> doesn't affect that: the rule, 
"shorten [i:] to [I] in polysyllables" holds 
either way.

>RLC folk would need to mark these, e.g. Bîbel and sîra if they were using
><i> in place of <y> [I].

Well, we're not going to make <i> and <y> 
orthographic options in initial or medial 
position. That would solve nothing.

>Note: Many people have problems, or mental blocks/bloody-mindedness, with
>diacritical marks.

Too bad. That's just plain too bad.

The phonology of Revived Cornish (with its two 
dialects complicating things) simply can't be 
represented in a coherent yet authentic 
orthography that doesn't use diacritical marks. 
If people don't want to write a mature and robust 
orthography, they can write UC, or UCR, or RLC, 
or SWF. (I can't bring myself to recommend KK.)

The Irish, Welsh, Scottish Gaels, and Bretons all 
manage with diacritical marks (though the Welsh 
are disadvantaged by character set issues). Never 
mind the French, the Germans, the Danes, the 
Estonians, the Luxemburgers and the Maltese.

People are smart, and children will never have 
problems with KS's few diacritics.

>Words like bÿs/bës would inevitably be written 
>bys/bes much of the time (as they are in the 
>current SWF).

Would they? I'm going to publish books using 
them. I expect some of these might be popular. 
We're going to use them in our teaching grammars, 
and we will teach that the marks are important and

>However, the <y> is long so they should be 
>bïs/bës. That way, when the mark is left off we 
>get bis/bes which is better.

No, it isn't. It just shifts ambiguity from the 
<res>-'necessary'-words to the 

>For RLC folk, bis would be a definite error as 
>for them the <i> would be short (assuming rule 
>(1) above.)  I don't know but I would expect RLC 
>folk to use only the <ë> form.

In KS, in monosyllables, <i> is [i:], <y> is [I], 
<ÿ> is [i:] or [e:], and <ë> is [i:] or [e:]. We 
grasped this nettle: <ÿ> is a different letter 
from <y>. The diacritic can't be "left off" if 
the orthography is to be coherent.

Sorry if I seem impatient. I don't mean to.
Michael Everson * http://www.evertype.com

More information about the Spellyans mailing list