Re: Tagalog & trigger idea: I'd like comments. :)

On Tue, Nov 16, 2004 at 10:44:48PM +0100, Henrik Theiling wrote:
[...]
> Hmm, I would not say this is triggerish, but I'd say the language uses
> word order to mark the focus, just like German.  The example sentence
> you gave works exactly like that in German:
>
>    give=       geben     (with 3.sg.pres.ind.act. = gibt)
>    the man=    der Mann
>    the woman=  die Frau
>    the flower= die Blume
>
> Funny, even the genders are the same. :-) Case markers are in front in
> German and also contain definedness information -- they are articles,
> of course.  And there is no verb complement. :-)))
> The case mapping is:
>
>    giver     -> nominative:  der Mann
>    recipient -> dative:      der Frau
>    patient   -> accusative:  die Blume
Interesting, I didn't think of this aspect of it. But, Tatari Faran
does not have passives. The fronted NP also takes a "nominative" role
in a sense. This doesn't quite show in these examples because a
relatively simple sentence is involved; but it becomes clearer if you
tack on several clauses together:

diru nei kira firasa sei kiran ka esan, henan dei niba' tara' ka tsum.
"The girl was given flowers by the young man, and he kissed [her]."

firasa sei kira kiran ka diru nei esan, henan dei niba' tara' ka tsum.
"The flowers were given to the girl by the young man, and he kissed
[the flowers]."

The second clause takes the fronted NP of the first clause as its
"subject", independent of case:

henan dei niba' tara' ka       tsum.
CONJ  RCP kiss  3sp   ORG-MASC COMPL
"And [was] kissed by him."

The elided subject here refers to the fronted NP in the previous
clause; hence, changing word order in the 1st clause alters the
meaning of the 2nd.

The point here is that there is no passive form of the verb; the
selected "subject" in the 1st clause becomes the "subject" of the
elided 2nd clause independently of its noun case.


[...]
> (Just as a remark: for more emphasis plus demotion of 'der Mann', you
> may also choose to use passive voice in German, in this case, indirect
> passive voice (with 'bekommen' or 'kriegen') to get nominative case
> marking for 'Frau':
>
>    Die Frau bekommt die Blume von dem Mann gegeben.
>    NOM              ACC       OBL
> )
[...]
> (Again, with passive voice, this time direct passive with 'werden':
>
>    Die Blume wird der Frau von dem Mann gegeben.
>    NOM            DAT      OBL
> )
This is where a fundamental difference occurs: the case markings of
the NP's in Tatari Faran cannot change without also changing the
factual content of the sentence. There is no such thing as passive
verbs in TF. If you change the marking of _diru_ to originative, the
girl becomes the giver rather than the recipient.


[...]
> > How is this related to triggers? Tatari Faran, unlike Tagalog, does
> > not move the case marking of the trigger NP onto the verb. It also has
> > no special trigger marker.
> >...
>
> I think that's the point.  The system I see is focus by word order,
> not triggers.  Triggers as I understand them involve changing the verb
> form to indicate which role is the trigger, and then requiring the
> trigger case for that role.  (The other roles may keep their case
> marking or be changed as well, depending on language).
Hmm OK, so what Tatari Faran has isn't triggers after all. :-)


> This is similar to voices, but also different, since usually triggers
> mark *roles* instead of purely syntactically changing case assignment,
> and they usually do not demote anything to obliqueness.  Marking roles
> is the reason why trigger languages often have more choices for
> triggers that for cases (Tagalog, if I remember correctly, has five
> (or only four?) choices for triggers on the verb, but only three
> cases).
And a significant difference between Tatari Faran and German is that
Tatari Faran has no concept of obliqueness.


> At least, this is my understanding.  Triggers have created so much
> confusion here that I will not claim that I understood them correctly.
> I just describe what *I* understood. :-)))
OK, I don't think I adequately understand triggers myself. :-)


[...]
> Hmm, I don't think so, since the system is really identical to that of
> German, which is clearly accusative and never even has anyone I met
> considered that it had triggers. :-)
[...]

OK, so Tatari Faran doesn't have triggers after all.

But I don't think you can say Tatari Faran is accusative. The case
assignments are purely semantically-based, and you can't change the
case of an NP without also changing its semantic role. There's no such
thing as oblique cases nor passive verbs, and hence no way nor need to
"promote/demote" NP's to/from obliqueness. Any 3 core cases can
adequately serve as a "subject"; there is no nominative case which
must always be present (explicitly or implicitly).


T

--
Once bitten, twice cry...

Replies