Re: deeply embedded VSO nightmare

In a message dated 10/19/01 2:38:24 PM, annis@BIOSTAT.WISC.EDU writes:

>Oy.
>
>So, Vaior is a VSO language.  This is fine and dandy most of the time,
>but while writing the primer -- I actually have a very enthusiastic
>friend learning the language, and he keeps pushing me on the primer --
>I regularly discover just how superficial my previous language
>creations were.
>
>Anyway, to my recent horror I have discovered that due to how Vaior
>word order works -- VSO, Noun - Modifier, Adj - Modifier -- that
>gigantic ambiguities develop when using participles and relative
>clauses.
>
>This is fine:
>
>    tuar-o   va tath-an   daip-oth-an        fid-íal  aldove-n.
>    see-PRES I  woman-ACC walk-PRES=PART-ACC fast-ADV home-ACC
>    I see a woman walking quickly home.
>
>But this:
>
>    ler-o   na  tath  daip-oth       fid-íal  aldove-n
>    go-PRES the woman walk-PRES=PART fast-ADV home-ACC
>
>could be either "the woman walking quickly is going home" or "the
>woman walking quickly home is going."  Granted the second
>interpretation sounds a bit odd, but it's an example of the
>ambiguities I've run into: which verb does 'aldoven' go with when the
>meaning doesn't make it entirely clear.
>
>I've not studied any strongly VSO languages enough to really grok how
>this sort of situation is disambiguated.  Some sort of word shuffle
>seems in order, but I thought I'd check to see if any real linguists
>:) had insight into how this sort of thing is usually handled.
>
>--
>William Annis  -  System Administrator -  Biomedical Computing Group
>"When men are inhuman, take care not to feel towards them as they do
>towards other humans."                       Marcus Aurelius  VII.65
I don't know anything about VSO languages, but there's one thing I can think
of. The two possiblities for your sentence "lero na tath daipoth fidíal
aldoven" are basically
V N (V Adv) N
V N (V Adv N)
where the parentheses represent what is basically a relative clause. In the
first, the last noun is not included in the relative clause, in the second it
is. To distinguish the two, you could, instead of indicating the relative
clause by a special participial form of the verb, or in addition to that,
stick in some kind of relative clause marker right where the ) is.
So you'd have:
Goes woman walking/walks quickly who home.
vs.
Goes woman walking/walks quickly home who.

I have a feeling that this is not the ideal solution for a perfectly VSO
language, because it would prefer a relative pronoun to the left of the the
clause rather than to the right (which would of course not disambiguate
anything) But I don't know what else to do other than accepting ambiguity, or
making special forms of words for nouns that aren't in relative clauses or
somesuch. There's something weird about VSO, or something I don't understand
anyway. They told us in LIN 101 that a verb and its object form a verb phrase
together, which makes sense in SVO, SOV, VOS, and OVS languages, where the
object and the verb are actually next to each other (and therefore you don't
get ambiguities like the one above). But in VSO and OSV they're not, so I
don't know how they can make a phrase together if there's a subject in
between. I asked my prof and he either wasn't too sure or didn't want to get
into it, but he hinted that maybe there was a transformation going on... how
screwy though, to need a transformation for every sentence! Maybe in these
languages it's the subject and verb that form a phrase together ... but that
doesn't seem too likely (could make a nice alien conlang though). So
anyway....

Good luck with your ambiguity nightmare,
Josh Roth
http://members.aol.com/fuscian/eloshtan.html

Replies