Re: OT: What makes a good conlang? (was Re: Super OT: Re: CHAT: JRRT)

On Mon, Mar 08, 2004 at 06:55:48PM -0500, Jeffrey Henning wrote:
> On Mon, 8 Mar 2004 17:10:43 -0500, H. S. Teoh <hsteoh@...> wrote:
>
> >> other hand demonstrate.  I'd always prefer a brief sketch which shows
> >> masterful treatment of certain details over a complete conlang which
> >> consists of a humdrum, obviously unreflected SAE grammar and a
> >> randomly generated vocabulary.
> >
> >I fully agree. This is one of the main reasons I stayed away from
> >automatically generating Ebisédian vocabulary in any way, even though that
> >meant that I have to be content with its incomplete lexicon for many more
> >years.
>
> I think you are unnecessarily conflating what can be done as two steps:
> * generate random forms that fulfill the phonotactics of the language
> * evaluate the forms to intuit what meanings they correspond to
[snip]
=========================
This is in fact exactly the way I work: generate a whole bunch of forms, and then
browse the list as I am trying to fill a certain semantic need. I found that
that helped me avoid making all of the roots in Orēlynna sound the same. In my
new project, eminkahken, I am taking the same approach. If it ain't Baroque
(ahem... "broke"...)

>>>sandat hsteoh@QUICKFUR.ATH.CX 3/9/2004 8:27:23 AM >>>
Hmm, that's a very good point. :-) But I guess I have a problem there 'cos
Ebisédian's phonology wasn't very well thought up (I had basically no
experience with phonologies when I did it, so it was a bit too
unrealistic). I might consider doing something like this for Tamahí,
although there may not be a need for it since Tamahí would've inherited a
whole bunch of words from Ebisédian.
=========================
I'm finding that creating the phonology for emindahken is more difficult than
Orēlynna's, perhaps because I didn't know what I was doing the first time
around. I still don't, but at least I know that I don't... (see the emindahken
phonology post of a couple of days ago).

James W.