Re: Case

On Sat, 10 Jul 1999, From Http://Members.Aol.Com/Lassailly/Tunuframe.Html wrote:

> i'm not a linguist so i can't say much more than for "tan".
> maybe i'm completely wrong here but temptatively it makes me
> think of japanese nominalizing-topicalizing "no wa", "to iu no
> wa", "to iu koto", "to itte", etc.. so you can topicize any
> argument, verb, copula, clause, raccoons, etc... except that
> sometimes it seems also trigger a kind of dutch "er" or "daar"
> in the rheme and sometime it just points at the focus (?). also
> i guess that the fact that NEG is right after or before any
> negated item makes it possible to either topicize/focus a
> negation or negate the topic/focus (??). i must say that some
> of your examples i would have flatly misinterpreted.
>
Grinning... And I thought Denden was an easy language! Speaking on a
meta-level (as we used to say in the heady days of the Dutch Linguafiction
Society) I think these  constructions have a layered origin - the
adpositional particles appeared in the days when I thought that _French_
was an exotic language, and the particle _ga_ appeared when I was just
beginning to learn Classical Chinese, which was taught, appropriately
since Japanese was pushed out of the curriculum a few years before, with
lots of references to Japanese grammar.  I think that you are right in
discerning three functions, that can be subsumed under the epithetet
'nominalisation':

1. Genereral topicalisation (although, strictly speaking, the particle
_ka_ TOP is more suited in that role)
2. a sort of resumptive or reflexive article
3. a focus marker

The fourth function, as a copula might be convincingly derived from
these functions, as they all seem to point to the immediate relevance,
and thus the existence of the phrase that _ga_ points at.

>
> S = subject
> T/F = topic/focus
> iO = indirect object
> dO = direct object
> ?O = what kind of object ?
>
> your structure seems to me :
>
> S iO V dO
>
(EXP: experiential aspect, DEL: delimitative aspect)

But only when _ga_ NOM is used, since the general structure
for ditransitive sentences is S V iO dO:

.    Quandiyal jerat najan Lujis.
     Quandiyal give  kiss  Lujis
     Quandiyal gave Lujis a kiss

.    Yamei ka.nda    tau.tau.jeret teli    ryerp Gyezal.
     Yamei TOP.3sMGH NEG.NEG.give  nothing crumb Gyezal
     As for Yamei, she didn't even give no crumbs to Gyezal.

(_nda_ is a regional variant of regular 3sMGH _da_, which is bound to _ka_
TOP, also Denden is clearly a language with a lot of negatives!)

.    Sero  percha bachan.za.za knighdra.
     1sHGH cudgel beat.EXP.DEL ox
     I was just beating the ox a bit with a cudgel

But if the instrumental _hye_ INST is used, the order is S V dO HYE iO:

.    Sero  bachan.za.za khighdra hye  percha
     1sHGH beat.EXP.DEL ox       INST cudgel
     I was just beating the ox a bit with a cudgel

Which contrasts to the version with _ga_:

.    khighra ga  sero  bachan.za.za hye  percha nothaz (ga)
     ox      NOM 1sHGH beat.EXP.DEL INST cudgel dead    NOM
     The ox I beat with a cudgel has died/ is dead.

Which is a rather more complex sentence, as is (and here I transpose your
answer):

> in light cast by the following i guess ?O is actually iO:
>
> >       Keda   ga  naha   afran teshitlo gevir ram.
> >       father NOM mother go    market   give  money
> >       father gives money to mother who goes to the marked
> >
>
> S-T/F iO V dO
So I'd say that a phrase which is nominalized with _ga_ can be used as an
indirect object and then is placed after the topic/subject. (And in
Denden, topic-comment constructions are almost always of the form: topic
KA subject v o, as in _Yamei ka.nda tau.tau.jeret teli ryerp Gyezal_,
above.)

> stuff like "cu" and "at" reminds me of serial intransitive verbs
> or english up and down in put it up/cut it down or japanese
> copulas.
>
I think you're onto something here - I've never considered them as
verbs, since they are formally outside that class - no tense, aspect,
attitudinal qualifiers or even negation, but appear to 'finish' a
verbless sentence. I will think on this for a week, and then write
the relevant chapter in my 1999 grammar!

I earlier made the remark that _ga_ can be negated: _tauga_, but that that
wasn't the best or most accepted style. I've since then unearthed an
example of a clever negation of _ga_:

.    Manud yo   adan tau.adan ga! da    yumir'beru hye  qenan,
     Manud that man  NEG.man  NOM 1sMGH wife.first INST make_love

     xong.zi,  xong.zi   na da    yumir'verai hye  qenan
     tired.AUG tired.AUG so 1sMGH wife.second INST make_love

     tau.yindad. E.da       yumir.ir dilogh qenan!
     NEG.can     poss.1sMGH wife.DUP REFL   make_love

     That man Manud, he isn't a man! If he makes love with his first
     wife he's very tired, he's so tired he can't make love with his
     second wife. His wives must make love amongst themselves!

Where 'Manud isn't a man', is not expressed as _Manud adan tauga_, but as
_Manud tauadan ga_, 'Manud is a not-man'. It is likely that tauadan isn't
a lexical item, meaning for instance eunuch, but a productive grammatical
construction.

> >       Adanvough ga  barusha  goho qireze.
> >       old_man   NOM mountain high see
> >       The old man sees the mountain that is high.
> >       The old man sees that the mountain is high.
>
> S-T/F ?O V
>
a complicating fact is that the verb in the nominalized phrase
can have tense and aspect and so on, which means that it isn't
a simple noun phrase, I think:

.   Purdam ga  adan.dan noth.e.nai  cherdin.amoi.
    priest NOM man.p    die.RPT.PRF burn.RFUT
    The priest will burn the men who died.
    (or: the priest will burn the men after they have died.)

.   e.do        keda     ga  adim etand.alei mo   seras andal
    poss-1sMGH  ancestor NOM boy  be.PT3     then 3sHGH world

    can     hahan.alei.
    through tramp.PT3

    When my ancestor was a boy he travelled through the world.

I think that in the future I will have to distinguish between
two _ga_'s, one the nominalizing particle discussed above, and one
the etymologically related copula, which is used in sentences like:

.   Aya! Vlami e.serir    burgat tau.ga.
    EXCL Vlami poss.1pHGH god    NEG.NOM.
    Ha! Vlami is not our god.

However, sentences like:

.   koran hye  ka  yadir dox ga  lyan      ga
    veil  INST TOP woman all NOM beautiful NOM
    With a veil, every woman is beautiful

remain rather difficult to analyze:

    koran hye ka: Topic
    yadir dox: Subject
    ga: nominalizes lyan?
    lyan: stative verb? attributive verb? nominalized adjective?
    ga: copula

The exact meaning is perhaps something like: as for women with a veil,
all women are women who are beautiful, and a fuller form of the sentence
might be:

* yadir koran hye ka, yadir dox ga yadir lyan ga.

But I'm not very sure of this, myself.


Boudewijn Rempt  | http://www.xs4all.nl/~bsarempt