Re: Never violate a universal unless it seems like a good idea atthe time

At 08:42 PM 9/4/03 -0400, David wrote:
><soapbox>Has anyone ventured the "universals are bunk" opinion yet?  Latin
>violates #4.  Several of my languages violate several.  And if these are
>to hold as universals, they should probably have an asterisk saying that
>they don't apply to creoles, or saying that creoles should have their own
>set, because many creoles violate many of these.  I'm also fairly certain
>I've seen a language that violates 38.  I'd like to see a natural language
>that doesn't violate at least one.  As for creation, I think it'd be
>unrealistic to create only languages which violate none of the universals
>in one way or another.  It'd be equally unrealistic to have it violate,
>say, more than five or six, but I don't think one should strive for
>violating none at all.  That said, these are good guidelines, but should
>be, IMHO, considered no more than guidelines.</soapbox>
I probably gave the impression in my post that I didn't want my languages
to violate any of the universals in any way whatsoever and that I didn't
want them to do anything too marked for fear of being implausible.  That's
not really what I meant to say.  My attitude is probably close to what you
outlined above: at least one of the less strict universals is probably
going to get violated, and violating more than 5 or 6 (or one of the
absolute ones, if they truly are absolute) would make it unrealistic.

For me, it is important to keep it realistic.  This probably matters less
for those who are creating languages simply for their own sake.  As I said
in a post earlier today, these languages that I am creating exist because
they have speakers, and the speakers exist because they are characters in a
story that I am writing.  Just as the characters in the story need to act
in a believable fashion and the plot has to be put together sensibly for
the story to be a good one, and the cultures in the story have to be well
thought-out to be believable, their languages also need to behave in a
believable fashion in order to complete the illusion.  (Not that I ever
plan to publish the story anyway or that most readers would know a violated
linguistic universal if it leaped up and bit them on the nose, but I'm
nitpicky with myself, and I'm a linguist, and *I'd* know.)

I would venture the opinion that Latin doesn't out-and-out *violate* #4
since that is a "with overwhelmingly greater than chance frequency," not an
absolute prohibition on the behavior.  Latin is just marked in that respect.

And I'll have to tell my husband all this about languages "violating"
universals.  He's been laughing lately when I've talked about "illegal
consonant clusters."  It put the image in his mind of the language police
coming out and arresting you or something for having illegal consonant
clusters :-)

Isidora

Replies