Re: How to start to make a language?

En réponse à "H. S. Teoh" <hsteoh@...>:

>
> I'd say, don't worry about Latin letters right from the start; get
> your
> phonology going first, and *then* consider what's the most
> straightforward
> way to write them in Roman letters.
>
Indeed, that's the best advice possible. I'd add this one: for a first
language, don't try to make too original a phonological inventory. You would
end up without being able to pronounce it, and in creating the morphemes you
would forget most of the exotic sounds anyway, making them so rare that they
wouldn't need to be here anyway...

>
> Oh, Maggelity will drive one insane unless he already is. ;-) (j/k,
> Christophe.)
>
Actually, I quite agree with that remark ;))) .

>
> Yeah. My approach is still to invent the sounds first, and perhaps even
> a
> native script, and *then* consider how a group of explorer-linguists
> might
> transcribe this language in Latin letters. Obviously, they will choose
> the
> most straightforward, clearest system they can think of.
>
That's exactly the approach I had for Itakian. So the native script, though
phonemic, doesn't mark tone, but the latin transliteration does, since it's
linguists which made it. Moreover, I decided that the first person to ever
propose a transliteration of Itakian was French, which explains why it uses |c|
for /k/ instead of |k|, and |i| and |u| for /j/ and /w/, instead of |y| and |w|
like an Anglo-Saxon would have imagined.

The nice part of this approach is that you can make some twists in your
transliteration like the ones I explained above, just by giving "historical"
reasons. It's practical because this way you needed use a perfectly precise
transliteration, and the language looks a bit less bland on paper :))) .
However, as was said before, refrain from using unnecessary diacritics. Look at
natural languages and their orthographies. In your own native language, accents
are always there for a reason. It's the same in French, despite its looks. No
accent in French is present when it could be avoided.

> Unless, as mentioned above, your goal is to create an Indo-European
> language with spelling conventions dating several from mythological
> times,
> shoe-horned into a modern language whose pronunciations have *nothing*
> to
> do with the spelling at all. Like Maggel. :-P (OK, I'm kidding. Maggel
> at
> least is regularly written. But its regularity is the kind that will
> drive
> your brain all irregular. :-P)
>
Maggel, regularly written? Hahahahahahihihihihihihohohohohohoho!!!!! Although
you're right about Maggel's regularity, and I've been emphasizing it on the
list to show how bad it already begins, I've also been repeating that a "rule"
in Maggel is just something which explains a plurality of cases, not a majority
of them ;))) . A word like |imuohf| [mbu:]: cow is not regularly written at
all! (only |uo| representing [u:] is regular. The rest is not :)) ).

But since one of the points of Maggel is to make the distinction between
regularity and irregularity obsolete, in a strange, Maggellish way, your point
stands ;))))) .

Christophe.

http://rainbow.conlang.free.fr

Take your life as a movie: do not let anybody else play the leading role.