By having a sufficiently small lexicon that a simple visual scan of
the root list suffices to check for redundancies?  Admittedly, it's
not a solution that scales well, but since my largest lexicon to date
is still under 100 roots, it works so far...



On 8/4/08, Jim Henry <[log in to unmask]> wrote:
> On Mon, Aug 4, 2008 at 4:34 PM, Michael Poxon <[log in to unmask]> wrote:
>> Sometimes this is simply going to be unavoidable, especially if one is
>> working with a definite phonological area. Indeed, sometimes it will
>> probably be intentional (witness Tolkien lifting both the word and the
>> meaning from Finnish, with quenya tie "road" and tul- "to come" for
>> example). In my case, where there is inherently some causal connection
>> being
>
> I should clarify; I meant, not avoiding new words in one's conlang
> that sound too similar to existing natlang words, but avoding words
> that are too similar to words already in the same conlang.
>
>> .............. Omina's not
>> derived from either a conlang or a natlang (some overt Basque influences
>> though) but what I'm after is an impression of inner consistency. I
>> wouldn't
>> dream of using any computer resources for generating words. I have to feel
>> them.
>
> Indeed, I don't like to use script-generated words for any conlang
> where beauty is a high-priority design goal.  Even when I use
> script-generated
> words for an engelang, I tend to generate the list of words in one
> step, and then match them up with appropriate meanings in
> another step, not entirely automatically.
>
> For gzb all the words were made up by hand, either adapted
> to gzb phonology from a natlang source or made up a priori
> -- and yet, before definitely adopting a word form I've thought
> of that sounds intuitively good, I think it's usually a good idea to
> use this findsimilar.pl script to make sure I don't already have
> another word that sounds too similar and is likely to occur in
> the same contexts.   It seems that this would be a good
> design principle for any engelang, or engelangesque artlang,
> where unambiguity or ease of learning are design goals
> at higher priority than euphony.   (Euphony is a design goal
> of gzb, but it's not one of my highest priority goals.)
>
> --
> Jim Henry
> http://www.pobox.com/~jimhenry/conlang/fluency-survey.html
> Conlang fluency survey -- there's still time to participate before
> I analyze the results and write the article
>

-- 
Sent from Gmail for mobile | mobile.google.com

Mark J. Reed <[log in to unmask]>