Isolating, Inflected, Word Building, ETC.

These are some really fascinating topics that make me
wish I had majored in linguistics instead of
engineering. ;-)

But all this makes me suspect that the best way to
arrive at a really naturalistic conlang is not to
build it from scratch to completion in one step, but
to evolve it, step by step, from some starting point,
either an existing natlang, or a very rudimentary ad
hoc proto-language. If one started, for example, with
the few dozen words from the language of the movie
"Caveman", and applied, a few hundred years at a time,
ten thousand years worth of mutations, it would seem
that the result would be a very plausible imitation of
a non-existent natlang. Assuming, of course, that the
mutations applied at each step are plausible.

--gary

Reply