Re: of cakes and men -- doraya syntax

On Wed, 10 Nov 1999, Grandsire, C.A. wrote:

>         As I see it, Doraya would be a VSO language where the position in front
> of the verb is available for topicalisation (like my Chasma"o"cho) or
> old information, in which case the fronted noun or nouns must be
> recalled by a pronoun after the verb. On the other hand, a pronoun
> always refer to old information, so if its antecedent is not in the
> sentence itself, the pronoun must be put in front of the verb, even when
> it's a contracted form of two pronouns where only one refers to old
> information not in the sentence. Do I understand it right?
>
        Sounds like you've got it.  Here are my rules so far: 1)
fronting of a subject pronoun that occurs alone (i.e., without an
antecedent in the same clause) is mandatory, thus giving:

        tai tyepylen rian
        3ps perf.eat cake
        He ate a cake. (he = given info)

instead of the expected *_tyepylen tai rian_.  2) Fronting of pronoun
contractions is mandatory.  To illustrate, the "underlying" structure of
this sentence:

        kos tasa tyepylen
        person 3ps.3pn perf.eat
        The person ate it. (person and it = given info)

would be something like *_kos tyepylen tai sae_.  3) Fronted
non-subject NPs are marked with the clitic _ui_, as in:

        rianui tasa tyepylen
        cake.TOP 3ps.3pn perf.eat
        He ate the cake (As for the cake, it was eaten by him)

or, to demonstrate _ui_'s cliticness further:

        rian mayaui tasa tyepylen
        cake good.TOP 3ps.3pn perf.eat
        He ate the good cake (As for the good cake, it was eaten by him)

>         Well, I don't feel it is unnaturalistic, even if I cannot see any
> examples in natural languages. I find it a very natural way to do it (a
> little like the object being after the verb in French, except in pronoun
> form or topicalised). I think some examples of spoken French that
> Mathias has already given fit well this behaviour, so maybe there is
> indeed a precedent in a natlang. So go for it, it appears very natural
> (at least for me).
>
        My (admittedly inadequate) knowledge of French grammar has
always been an inspiration for Doraya in many regards, particularly
the pronoun fronting feature (a feature of French that I absolutely
adore; nothing beats a good "je lui l'ai donne'" or "il s'en souvient").
But as I mentioned, Castilian Spanish seems to have a structure similar
to Doraya's, except in reverse (Mr. Brandt-Young will no doubt recognize
this example from the Ling 5 midterm):

        Que' hizo Consuelo?
        'What did Consuelo do?'

        Consuelo comio' tu bocadillo.
        'Consuelo ate your sandwich.'  ("Consuelo" = given; "comio' tu
bocadillo" = new)

but

        Quie'n comio' mi bocadillo?
        'Who ate my sandwich?'

        Tu bocadillo lo comio' Consuelo.
        'Your sandwich, Consuelo ate it.'  ("Tu bocadillo lo comio'" =
given; "Consuelo" = new)

>         Just one last question: could the 10th sentence be put like this:
>
>         kos rianui tyepylen tasa.
>         person cake.TOP eat.perf 3ps.3pn
>         The person ate the cake.
>
> or is it ungrammatical? In my understanding of this pattern, this last
> sentence is perfectly possible, unless you have an additional rule that
> says that compounded pronouns must be put in front of the verb anyway.
> Go for what you prefer!
>
        See rule #2 above.  It probably wouldn't be *completely*
ungrammatical; you'd get across the point that there's a cake and
there's a guy and there's some eating going on, but it'd sound a little
odd.
        But anyway, I guess the part of this that I'm most concerned
about is the clitic _ui_, which (again) marks non-subject NPs that are
moved out of the main phrase.  I like the way it looks, but I wonder if
it's necessary and/or naturalistic.  It doesn't seem to be necessary in
either French or Spanish -- but both of those languages have separate
pronoun forms for different cases, a feature that Doraya's pronouns
lack.

Thanks for your feedback!
Adam
----------------------------.
             myth@inquo.net |
http://www.inquo.net/~myth/ |
----------------------------'