Re: basic vocab

dirk elzinga wrote:

> When mapping meanings to roots, I found myself unconsciously selecting
> certain types of roots over others (eg; roots ending in -u are not as
> common in Tepa as roots ending in other vowels, though there is not
> phonotactic prohibition against it). This process has given Tepa a
> perhaps more natural feel. In fact, I'm wondering now if it wouldn't
> have been better (if more tedious) to create words on the fly rather
> than selecting from a automatically generated list. It certainly would
> have revealed some interesting things about my phonotactic
> predilections ...
I've always created the words for my conlangs 'on the fly' rather than working
from computer-generated lists.  With my first conlang, my phonetic predilictions
at the time shone through quite clearly:  /l/, /r/, /m/, /n/, /h/, and /w/ were
each about three to four times as common as the next most common consonants,
giving the language a rather nasal, whiny feel (or so I think now).  With
subsequent projects I learned not to favour certain sounds quite so much, while
still avoiding an artificial situation where each sound was equally represented.

In my current conlang, Tokana, the inventory of phonemes is so small that I have
no choice but to use my favourite sounds over and over, a happy result.  Even so,
I've tried to keep the distribution of sounds somewhat asymmetric--e.g., favouring
stops over fricatives, etc..  One asymmetry, namely the comparative rarity of the
/f/ sound, is a by-product of the fact that this sound was introduced rather late
in my conception of the sound system, after I had already created quite a few
words.  (Within the fictional context for Tokana, I explain the rarity of the /f/
sound in historical terms:  /f/ was once a more common sound in the language, but
is now  disappearing, undergoing gradual merger with /h/ in most dialects.)

Matt.