Theiling Online    Sitemap    Conlang Mailing List HQ   

More thoughts about S11 grammar

From:Henrik Theiling <theiling@...>
Date:Tuesday, March 22, 2005, 13:32

To break the creative (relative) silence on the list with some
on-topic stuff, here're some thoughts about S11.

Memory Refresh

A few weeks ago I presented my first thoughts about S11, my new
project that has no transitive verbs, only intransitive ones.  All the
rest is done by SVC (serial verb construction).  Short feature list:

   - Word order is noun-verb,
   - Verbs are enclitics: they attach to the preceding word,
   - First noun is topic,
   - Evidence/mood enclitic attaches to topic,
   - States and entities are nouns,
   - Events and actions are verbs.

   For 'read', the lang has two verbs: 'read' and ''.

    'John reads a book (so they say).'

The REL behind a clause makes relative clauses, with the original
topic being the referent:

    John-HEAYSAY-read REL ...
    'John, who reads a book, ...'

New Stuff

Now, after some investigation on how far this can be pushed, I arrived
at a few point where it was unelegant to stick to the rules strictly:

   - noun-noun modification:
        John's book
        the house's inside  (needed in: 'John is in the house.')

   - sentences expressing states:
        John is a butcher.
        John has a book.
        John is in the house.

Since states are nouns, it seems these two items can be unified.
Indeed, 'John has a book' and 'John's book' seem to be very similar.

I think these sentences show that some relations cannot easily be
split into two verbs: they feel inheritently binary (=transitive).
E.g. 'have' or 'be': it feels very awkward to me to split these to
mark a) 'the one who has/is' and 'the one who is had/been'.

Therefore, there is probably going to be closed class of special
verbs.  I don't dare to call them transitive yet, since they feel
special: put differently -- they are verb-like clitics that convert a
noun into a predicate.  The predicate's argument is left unmarked.
These constructions are often used as relative clauses:

    - to be:          John-HEARSAY butcher-be.
                      'John is a butcher.'

                      John-HEARSAY butcher-be REL
                      'John the butcher' (apposition)

    - to be located:  John-HEARSAY house-be.located.
                      'John is in/at/with the house.'

    - to have:        John-HEARSAY book-have.
                      'John has a book.'

                      book-HEARSAY-have John.
                      'A book is what John has.'

                      book-HEARSAY-have John REL
                      'John's book'

      This may be used without an argument (like in many langs):

                      'There's a book.'

Some feel like inheritent noun-noun modifications, and maybe these
do need yet more special treatment:

    - part of whole:  inside-HEARSAY-part.of.whole house

                      inside-HEARSAY-part.of.whole house REL
                      'the inside of the house'

In SVC, this works as follows:

    book-HEARSAY-have inside-HEARSAY-part.of.whole house REL-be.located.
    'There's a book inside the house.'

In contrast to the other special verbs above, 'to have' seems more
naturally splittable into two verbs 'have' and 'own', however, and
could in fact express several shades of 'to have':

    John-HEARSAY-own book-have.
    'John has a book.'
    'There is a book and John owns it.'

    John-HEARSAY-be.located book-have.
    'John has a book.'
    'There is a book and it is with John.'

But then, it seems like 'be.located' can be split, too. *sigh*

I don't know what exactly will be needed.  Hopefully you have some
thoughts about what will most probably be special verbs?

It is strange that I wanted to ultimately eliminate the need for
distinguishing arguments and adjuncts and now arrived at a similarly
arbitrary borderline between special verbs and normal verbs.  Yet the
normal intransitive verbs just don't feel elegant in some cases.

Comments, suggestions, criticism, please! :-)


PS: Phonology is almost done, but I have no native words yet.
    Stay tuned.


H. S. Teoh <hsteoh@...>
Paul Sherrill <psher@...>