On Mon, 15 Oct 2012 09:26:34 +0200, Christophe Grandsire-Koevoets <[log in to unmask]> wrote:

>On 15 October 2012 05:17, Alex Fink <[log in to unmask]> wrote:
>> On Sun, 14 Oct 2012 18:14:51 -0400, neo gu <[log in to unmask]> wrote:
>> >AFMCL, this is a transliteration of a not-yet-developed native
>> orthography, using 18 letters.
>> This is delightful!  Essentially no-one designs native alphabetic
>> orthographies to have any non-transparentness (even Maggel was written
>> in the local Roman alphabet!)
>Actually, Maggel is written in its own alphabet, which happens to be a
>descendant of the Uncial script that also resulted in the Gaelic type, and
>thus is similar to it, and yet different enough that it can be considered a
>separate alphabet rather than a form of the Roman alphabet (in any case,
>that's what I advise you to say in the presence of Maggel speakers, at
>least if you value your fingers! ;) ). Some of the main differences is that
>the Maggel alphabet has no capital forms, special ligatures, and features
>letters that mandatorily attach to the next one while others forbid it (a
>bit like the Arabic script, and unlike any cursive form of the Roman
>alphabet). However, it also has less letters than the Roman alphabet, so a
>one-to-one transliteration in the Roman alphabet (based on the historical
>origin of each Maggel letter) isn't that difficult, even though it means
>that some letters get strange values (for instance the letter
>transliterated as <t> never has the value [t]!).

*puts on my steel gauntlets*  Alright, fine, Maggel uses a daughter script of Roman!  

But on further thought I shouldn't've excluded it from the contention, as whatever its script is, Maggel does use it in a way which wholly refuses to draw on European traditions.  And that's what I really was getting at.  There are people out there, mostly European althisters, who will specify that one of their consonant phonemes is spelled <ch> or whatnot because the Roman script is what they've got and the neighbouring langs use such strategies too.  But beyond that, people don't do much.  There's Maggel; there's BPJ's scripts with some nice instances of underspecification and whatnot; there's, umm, ...

>*sigh* I wish Maggel orthography was as easy to describe as this!

No you don't!  Where would the fun be in that?  :-p

Far from contending for your crown, I was rather hoping that showing another non-bizarre system without a one-to-one letter-sound correspondence would make the notion of doing so seem more accessible.  Making something like Maggel is intimidating...

For those who read my description: embarrassingly, I entirely forgot the phoneme <kx> /kx/!  And I forgot to note the behaviour of some of the other velars before <i>.  In inflection one can get <ŋi> /Ji/, but the other velars are respelled, so e.g. <kx> + <i> = <tši>~<kti> /tSi/.  (And I transcribe the esh letter as <š>, not <S> per my last message.)