Re: HELP: Is this sensible?

Hi!

Waah: my internet connection was broken for more than three hours!  I
hardly survived this...

Carsten schrieb:
> Thanks, but I've got another question on what you wrote:
>
> [...] It depends on how you assign agent and patient. If the
> assignment is highly lexicalised, you could do anything. [...]
>
> What do you mean with "lexicalized" here?
By this I meant if the lexicon entry for most verbs gives you a fixed
case-assignment phrase around it, then the assignment of cases would
be highly lexicalised.  You could look up the word 'to be strange' and
the lexicon would say that the one how is strange in a sentence would
be assigned case X.

In contrast to that, your language could be highly based on rules that
apply to every verb.  In such a language, the assignment of case for
the same verb could be different from sentence to sentence, and it
could also be regularly inferred.  E.g. in the case of 'to be
strange', my Tyl Sjok allows two assignments:

  Paul.AGT strange.
  = Paul is (~behaves) strange.  (he is in control)

  Strange Paul.PAT
  = Paul is strange. (inherently, he is not in control)

It's simply a matter of definition in your grammar.  If the assignment
is lexicalised, you can make it as chaotic as you like, of course. :-)

>...
> who does something is the agent and the one who is affected is the
> patient.
>...
If your language is supposed to regularly assignment agent and
patient, not by definition in the lexicon, defining that the agent is
'doing' something would probably not be enough for clarification of
case assignment.  Have a look at the distinction in the following
sentences:

  Peter dries.
  Peter falls.
  Peter sneezes.
  Peter speaks.
  Peter runs.

In English, you could argue that for each of these sentences, the
answer to the question 'What does Peter *do*?' is the verb.  But this
is simply because Peter is the subject of each of these sentences and
by asking for the doer, you ask for the subject.  It is a recursive
definition.  In this case, your case assignment should be equal to
that of English, which, I'm sure, you do not want, since adjectives
can select different cases in your language.

If you decide to use assignment by control, the first two examples
select patientive case, while the latter two would select agentive
case.  The one in the middle may select P or A, depending on language,
speaker or situation. :-)

You could use other things than control for a definition.  For
Tyl-Sjok, I use a system very close to Central Pomo's, but with
simplified rules (IIRC, Central Pomo assigns by control, but not if
the controller is personally affected, I think).

Other reasons for assignment of agentive case could be state
vs. event, or volition or performance/effect/instigation (PEI).  You
might want to check out the archives for discussions of this.  Usually
Daniel Andreasson and Marcus Smith write nice descriptions about
active languages.  E.g.:

http://listserv.brown.edu/archives/cgi-bin/wa?A2=ind0005A&L=conlang&P=R17644
http://listserv.brown.edu/archives/cgi-bin/wa?A2=ind0102B&L=conlang&P=R1214

>...
> you will see adjectives are marked depending on whether their noun
> is (agent, ) patient or something else ("oblique").
>...
Hmm, assigning oblique case seems strange to be.  Having looked at
your web-page, I understand that your language is supposed to be
similar to Tagalog's trigger system.

The one book about Tagalog that I have suggests that it has three
cases which are called nominative, genitive and dative/locative there.
It also dependents on the trigger (called voice in that book, which
seems odd).  It seems Tagalog has active, benefactive,
dative/locative, instrumental and objective triggers, the active one
being the default (without ending).  Depending on the trigger (on the
verb), the nominative fills the thematic role the trigger is named by
and the other supplements seem to be reassigned by some rules,
too. :-) It gives interesting examples illustrating the trigger
system:

   a) B-um-ili    ang=lalake ng=isda  sa=tindahan.
      PERF.AT-buy NOM=man    GEN=fish DAT=store
      The MAN bought fish at the store.

   b) B-in-ili    ng=lalake ang=isda sa=tindahan.
      PERF-buy-OT GEN=man   NOM=fish DAT=store
      The man bought FISH at the store.

   c) B-in-ilh-an ng=lalake ng=isda  ang=tindahan.
      PERF-buy-DT GEN=man   GEN=fish NOM=store
      The man bought fish at the STORE.

   d) Ip-in-am-bili ng=lalake ng=isda  ang=pera.
      IT-PERF-buy   GEN=man   GEN=fish NOM=money
      The man bough fish WITH the money.

   e) I-b-in-ili   ng=lalake ng=isda  ang=bata.
      BT-PERF-buy  GEN=man   GEN=fish NOM=child
      The man bough fish FOR the child.

AT=active trigger
OT=objective trigger
DT=dative/locative trigger
IT=instrumental trigger
BT=benefactive trigger

The book does not say how the active/objective triggers are used for
intransitive verbs, but from the name 'nominative' I conclude that
Tagalog is not split-S, but nominative/accusative, the accusative
function being taken of what is called genitive in my book.  This is a
wild guess, though.

Your language seems to be different in structure, in that it has as
case what Tagalog has as triggers.

Hmm.  This is long again.  Sorry.

Bye,
  Henrik