X-Apparently-To: [log in to unmask] via; Sun, 30 Oct 2005 15:19:25 -0800
X-Originating-IP: []
Authentication-Results:; domainkeys=pass (ok)
Received: from  (EHLO (
  by with SMTP; Sun, 30 Oct 2005 15:19:25 -0800
Received: by with SMTP id i28so968982nzi
        for <[log in to unmask]>; Sun, 30 Oct 2005 15:19:24 -0800 (PST)
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws;
Received: by with SMTP id s65mr2746525nzi;
        Sun, 30 Oct 2005 15:19:24 -0800 (PST)
Received: from ? ( [])
        by with ESMTP id 5sm399524nzk.2005.;
        Sun, 30 Oct 2005 15:19:24 -0800 (PST)
Mime-Version: 1.0 (Apple Message framework v734)
In-Reply-To: <[log in to unmask]>
References: <[log in to unmask]>
Content-Type: text/plain; charset=US-ASCII; format=flowed
Content-Transfer-Encoding: 7bit
From: "David J. Peterson" <[log in to unmask]>
Subject: Re: OFFLIST: Applicatives (was: Re: Requesting some challenging sentences)
Date: Sun, 30 Oct 2005 15:19:20 -0800
To: tomhchappell <[log in to unmask]>
X-Mailer: Apple Mail (2.734)
Content-Length: 1616

Howdy Tom,

You wrote:
No, that would be a "demotion"; I meant what I wrote.

Ohh...  Now I see what I confused.  In what you wrote, you wrote
"third position" to "second position" twice, but called them two
different things.  Thus, I thought one of them had been mixed up.
The difference, though, was in the *type* of language: indirect/direct
vs. secundative/primative.  Incidentally, I think Hawaiian's the

It was my impression that the question "Which roles can be
applicativized?" was language-specific.
Is that what you are saying here?
I wanted to say that, and forgot to.

Yes.  I could've said it more clearly, though, like you did here.  :)

That means that, in those
languages, by a process of two transformations, any participant, no
matter how peripheral, can become the Subject.

Yes.  This is probably because passivization is more restrictive--
that's my guess, at least.  One thing more applicatives does not
predict, though, is the lack of passive morphology, or dearth thereof.
Tukang Besi makes abundant use of applicativization, but actually
has five different types of passives (sensitive to all different kinds
of things).  They're just rarely used.  They're only used, in fact,
when an applicativized oblique needs to be put into the subject

In my yet-to-be-incarnated conlang, I am thinking of having
information-salience ("topic" and "comment", and "focus"
and "background", and "given" and "new") morphologically marked;
separately from semantic or thematic or theta- roles.


You know, I always had this idea where if you had a language
that marked just about everything, you could speak in proforms
pretty much all the time.  That is, you'd only use an overt nominal
(or overt verb) when it needed to be introduced for the first time,
or was not obvious from context.  In essence, you'd be speaking in
pure morphology.

Hey, since we're on the subject, maybe you can help me out.  For
one of my languages, I came up with a series of applicatives.  In
this language you can only relativize subjects; nothing else.  Thus,
applicatives and passives are used to expand the number of things
that can be relativized.  I came up with a number of them (source,
goal, non-motive location, instrument, possessor) that are seen
commonly.  Then I have this one called "human".  I invented this
awhile ago, and I had the whole system worked out, somewhere,
but I've misplaced that information.  So now I have this system,
and I have *NO* idea what I meant by having a "human" applicative
prefix!  Absolutely none!  I can't even *fathom* how it would work!
Do you have any guesses?  'Cause I know the system was neat,
and it'd be nice to know what the heck I was thinking.  You can see
the forms here (the page isn't done yet):

"sunly eleSkarez ygralleryf ydZZixelje je ox2mejze."
"No eternal reward will forgive us now for wasting the dawn."

-Jim Morrison