At 7:39 pm -0400 29/9/00, Nik Taylor wrote:
>Thomas Nordengen wrote:
>> Does anyone know a way to distinguish conlangs from natural ones?
>I don't think it would be possible to distinguish a well-designed
>conlang from a natlang.

I think you're right.  If, e.g. Tolkien had not been the honest fellow he
was, but had tried to pass Quenya and Sindarin off in some other medium
than fiction as actual examples of real language, I think he'd have got
away with it or at least have provided something that future generations
would argue over - Are they conlangs or not?

>Some conlangs would probably stand out as not
>being natural, simply because they violate so many tendencies in

Yes - if a language is deliberatedly contructed to be different in some way
from human language (e.g. lojban based on predicate logic; Fith, which is a
'stack manipulated' language; AllNound; Srikanth's 'Lin', the language of
being who communicate by telepathy etc, etc), then it's fairly easy to tell
the difference.

The artificial auxiliary languages almost always betray their articiality.
The majority of a_priori languages do so in the way their vicabularies are
constructed on strict classificatory principles in a way found in no
natural languages; nearly all the a_posteriori ones that I know do so in
that one meets mixes of natural languages elements which would not occur in
any of their source languages.  Interlingua avoids these strange mixes but
it gives itself away in that it 'ought' to be a Romancelang, but it lacks
certain features common to all the Romancelangs, i.e. no grammatical
gender, practically none of the Romance verbal conjugation apparatus etc.

There's, I think, a continuum of conlangs from those well-crafted products
that probably can't be distinguished from natlangs, through to those which
are deliberately designed to be different.


A mind which thinks at its own expense
will always interfere with language.
                   [J.G. Hamann 1760]