Re: theory

Marcus Smith wrote:

> Returning this topic to conlanging:
>
> I try to make my languages by-and-large fit current syntactic and phonological
> theory, but I am not tyrannical about it.  In fact, I have intentionally
> introduced features into Telek that many theories predict shouldn't occur.
> Some of them I give a historical explanation to, but others I don't.  For
> example, my active language has a passive construction, but that is because
> there was originally an indefinite subject that has been as a passive
> morpheme.
The reason that Nur-ellen doesn't have a passive construction is not
that
"active languages never have a passive construction", but because I
found out
that it didn't feel "right".  When I started designing Nur-ellen, I
hadn't heard
anything of active languages.  I had heard about ergativity back then,
and
when I was thinking about it, the thing linguists call an active
language
came to my mind.

When I was trying to find out how Nur-ellen forms passive, I noticed
that,
in accordance with the "spirit" of the active-type agreement, the formal
subject in the passive construction would have to be put in the
objective case,
which would have meant that there would be merely a shift of the
semantic subject from agentive to some oblique case.  Hence, I decided
that the whole thing was completely redundant and unnecessary, and as I
thought that Elves wouldn't use it anyway, I decided to dump it.  All
without knowing that "active languages never have a passive
construction".

What this episode shows is that you should neither be hang up with
getting your conlang conformant with theory at any cost, nor bash theory
whereever you can.
Just do what feels right to you; chances are that your language turns
out to fit quite well.

Joerg.