Irina Rempt <[log in to unmask]> wrote:
> I'm curious how other people come up with words - I know some people
> who can sit down and *make* them, but if I try that I usually have to
> throw away ninety-five percent of the result.

I usually use this method. As a result, of course, I may take
two or three minutes to create a single word. But the "correct"
sounds tend to pop up easily.

Also, JOEL MATTHEW PEARSON <[log in to unmask]> wrote:

> My general strategy for making up words is to decide on a meaning
> that needs to be lexicalised, and then go from there.  Sometimes I make
> do with a compound based on existing roots.  Other times - especially
> if the meaning is a 'basic' one which should by rights have its own
> root - I just let the proper combination of sounds 'come to me'.  As
> a result, a lot of sound symbolism and onomatopoeia has crept into
> Tokana.

All my conlangs so far are quite symbolic and onomatopoeic, due to the
method I use to create words. In Drasele'q, I at first tried to derive
as much as possible from roots, but I didn't have enough of them and
some derivations were pretty twisted. Then I decided I would try to
create new words for all simple concepts. If something can be derived,
I might create both, the derived word and the new one. I like to have
synonims and words that apply to different registers (formal, informal,
archaic, etc.).

Some examples of s/o in Drasele'q (you may see it or not):
_qoft_ "hard, rigid, strong"
_kra`t_ "toad"
_kri`k_ "frog"
_i`r_ "smile"
_mara'menden_ "to free-fall"
_fkif_ "shaft, beam"
_stonksen_ "to explode"
_kruvden_ "to break, tear apart"

--Pablo Flores

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
 Stewart's Law of Retroaction:
  It is easier to get forgiveness than permission.