Re: "Theory informs practice" - OK?

Hallo!

On Sun, 16 Nov 2008 10:00:37 -0500, Jim Henry wrote:

> In designing an engelang, I'm apt to look for ways to mark
> distinctions like the one you mentioned which seem logically
> or philosophically important but which few if any natlangs bother
> to grammaticalize.
Yes.  Engelangs allow for much experimentation, and it is
their unique distinctions and other features that make them
interesting.  They are not expected to work like human
natural languages.  Instead, a good engelang is based on a
well-defined design principle which is rigorously followed
through, without the layers of remnants of unproductive
mechanisms which are so characteristic of natlangs.

>       In designing a naturalistic human artlang,
> not so much.
A naturalistic human artlang should look like a human natural
language.  This *does* allow some room for interesting
distinctions (AFMOCL: degrees of volition in Old Albic), but
they should not do things one would never expect in a human
natlang.

>       For an exolang, probably somewhere in between
> -- looking to theory (not just in linguistics but in philosophy and
> math) for ideas about distinctions alien languages might make
> that human languages don't, for instance, but still assuming
> those alien languages will evolve ad-hoc into messy, irregular
> sets of trade-offs between conciseness and expressivity
> and not tend toward an extreme like an engelang.
Yes.  We know nothing about alien languages, but there are
a few things we can probably safely expect from them.
Their range of features is certainly wider than what can
be found in human natlangs, and many will be much "wilder"
than we can imagine.  Yet, there are some features a fully
functional language logically must have, and we can expect
alien natlangs to have *evolved* like human natlangs, which
means layers upon layers of relics of patterns no longer
productive and all that, so they won't look like engelangs.

... brought to you by the Weeping Elf