Re: Avoiding near-collisions in vocabulary coinage

Alex Fink <000024@...> schrieb am 05.08.2008 3:05 +0200:

> Anyone else had that happen?
>
Yes, sometimes. It's also interesting that sometimes I want to assign
semantically related words to the existing word by coincidence. On the
other hand, Ayeri's phonetic inventory (15 consonants, 5 vowels if you
don't count long vowels¹) don't allow for much difference for words if
you want to keep its sound consistent -- which I try -- and my approach
on wordmaking is not root-based although I have acquired a habit to
search the dictionary for existing words that I might want to derive a
meaning from. To avoid some bad cases of too similar words I do however
search the dictionary, usually for the first/main syllable of the word
and see what the database spits out. Thus I often need to slightly tweak
words that seem nice to me at first sight. Still there are way too many
similar words I feel. But then, these cases exist in English as well or
otherwise there wouldn't have minimal pairs. Anyway lack of redundancy
is nothing I planned for in Ayeri. It has just grown out of some initial
ideas about morphology and overall sound. On a side-note, Roger, are
having rather many minimal pairs an issue of possible confusion in
Indonesian as well?

Regards
Carsten

-----
¹) /p t k b d g m n ŋ v s h r l j a a: e e: i i: o o: u u:/, although
/o:/ is very uncommon and /u:/ in fact doesn't appear.