[Spellyans] Front unrounded vowels, was: The quantity system

Michael Everson everson at evertype.com
Tue Jun 24 20:43:05 BST 2008

At 21:32 +0300 2008-06-24, Owen Cook wrote:
>2008/6/24 Michael Everson <everson at evertype.com> rug scrifa:
>>  Actually, I think I've caught you here. ;-)
>>  Consider German and its dialects. <ö> is (in
>>  practice if not in theory) an umbrella graph for
>>  /ø/~/e/ (schön [Sø:n]~[Se:n]) and <ü> is an
>>  umbrella graph for /y/~/i/ (grün
>>  /gry:n]~[gri:n]). So what we propose here is that
>>  <ÿ> and <ë> are both umbrella graphs for
>>  [i:]~[e:].
>It's a clever trick, but I'm not taken in! ;-)

It's not a trick. It's true.

>The real reason for the diaeresis is simply that 
>we don't want to discrimate against Mac users.

Like it or not, fonts are tools. If an 
orthography users cannot type or fonts do not 
support, then you're sunk.

>While a noble goal in itself, this is not worth 
>the vexation of using the wrong diacritic when a 
>better alternative exists.

There is nothing "wrong" with the diaeresis.

>The words we're talking about all have <ë> or 
><ÿ> in a stressed syllable anyway -- no problem 
>there, were we to use the acute.

The acute is not generally available on the 
letter y in fonts, and even when found in a font 
on a Macintosh, the British Mac Roman keyboard 
(default at install) doesn't support it in input. 
The answer has got to be NO. This one is a 
non-starter. I cannot agree to hobbling the 
Cornish language with y-acute, y-grave, or 
y-circumflex. Choosing ANY of those is GUARANTEED 
to lead to failure for us. It is going to be 
hellish enough for us when they see accents used 
in our publications. Having them be UNAVAILABLE 
accents, world of Unicode or not, cannot do us 
any good.

And they will tell us how great Kernowek Kebmyn 
is without any diacritics at all.

>Furthermore, anybody with exposure to Irish will 
>know the acute as a length marker already.

More Cornish learn Spanish or French I suspect. 
Both of which use diaeresis (Spanish on ü, French 
on ÿ (!). Welsh uses ÿ as well, interestingly.

><é> is familiar to most of us from French as a 
>higher variety of <e> (potentially convenient 
>for those who want to continue the fiction of 
>'aspirational' /I:/), and was used this way in 
>Breton (in the Orthographie universitaire). 
>Long, stressed, alternating with a higher 
>variant -- our dédh and prés words are all of 
>these things. On hiatus, umlaut, centralized -- 
>dëdh and prës are none of these.

No, they aren't any of those. But the accent we 
need to use isn't *marking* length or stress or a 
higher variant (because it's meant to be used on 
the y as well).

This use of diaeresis is that of an umbrella 
graph for *dialect preference*. I have already 
demonstrated that this usage is found in German 
as ö and ü, in practice, when you take dialect 
into account. Exactly as in Cornish!

Both vowels have to be marked -- with the same 
diacritic (there is no other choice in terms of 
sane orthography design) -- and only one 
diacritic is really available. Whether that 
availability is "fair" or not is irrelevant. 
That's what the world gives us. And I don't think 
there's any argument that a diaeresis "must" mean 
one thing or another. When I read Spanish, I know 
that the güe means /gwe/ not /ge/. When I read 
Quenya, I know that it is pretty much decorative, 
meaning "I am not silent!". When I read Albanian, 
I know that it's a schwa. When I read Russian, I 
know that ë means [jo].

Lhuyd uses a single dot over y (for a different purpose)

>And if a diaeresis is arguably unobtrusive, an acute is certainly more so by
>virtue of being more simply formed.

Owen, lots and lots of people really disliked 
<ei> when it was first mooted. We got used to it, 
and came to like it. (The KK users never gave it 
a chance -- likely because they wanted the SWF to 
be as similar to KK as possible.) We can't have 
it. We've got to have something else, and that 
has to be *practical*.

I ask you earnestly to open your heart to the 
dewboynt. Else we are surely going to be stuck.
Michael Everson * http://www.evertype.com

More information about the Spellyans mailing list