Re: Verb-second ... verb-penultimate languages?

On Fri, 21 Apr 2006, Jim Henry wrote:
[snip]
> Does anyone know of any verb-penultimate languages?
> I suppose they would be VS in intransitive sentences,
> SVO in transitive, and SOVA where there is a sentential adverb;
> maybe SOVR (source/recipient last) in ditransitive sentences?
> Part of me suspects this is unnatural, and yet it doesn't seem any
> odder w.r.t. V2 languages than OVS or OSV languages, which
> do exist in small numbers.
> ------------------------------
How would this work?  To get an idea of it, I need to
create an example or two.

Let's try to modify a (nonsense) English text:
"You exist.  I know this, for you tell me that you see
ghosts.  I am happy, that I know this truth."

to have verb-penultimate clauses:
"Existest thou.  This know I, for me tellst thou that
ghosts seest thou.  Happy am I, that this truth know I."

In this example, I've simply insisted that each clause is
subject-final.  Assuming that each clause of the required
language *must* have a subject (a reasonable restriction?),
it's but a short step from an isolating, verb-penultimate,
subject-final language to an inflecting, verb-final language.
We now have:
"Existest-thou.  This know-I, for me tellst-thou that
ghosts seest-thou.  Happy am-I, that this truth know-I."

And I think that this would be a natural progression from
one kind of language to another.

An example that is not subject-final is a little harder for me
to envisage.  Suppose we had a language all of whose
clauses has the form: Subject Predicate.  For the Verb to
be penultimate in all clauses, we'd need to ensure that we
could always split the Predicate into Predicate-Part-1 Verb
Predicate-Part-2, where Predicate-Part-1 is optional and
Predicate-Part-2 is a single word.  But how big can that
word be?  If your language were highly inflecting, it could
dispense with (or absorb) most prepositions, expressing
their senses ("from", "to", "with", "near", "under", "after" ...)
by different cases, inflections of Verb.  Then what common
function remains for Predicate-Part-2 to perform?  It can't
take, for example, the rôle of Direct Object in an intransitive
clause.  The only obvious candidate for its function is to
mark the end of the clause.  I've seen this happen in three
actual languages; two are natlang dialects and the other a
conlang.  The first case is in Melayu Pasar (the Malay of
the market) as spoken by those whose L1 is not Malay.
They form the habit of suffixing the emphatic particle "-lah"
to *every* utterance, which somewhat diminishes its force
when actually needed ;-).  The second case is in the
Queensland dialect of Australian English, where the informal
question marker ", eh?" has gradually transmuted into a
formal end-of-utterance marker ", eh".  The other case is in
Elomi and its descendant Ilomi, in which the word "a" (used
to mark a pause) has evolved to mark a syntactic bracket.

I've looked at two general clause structures with the verb
second-last:
1.  OTHER  V  S
and
2.  S PRED-1 V PRED-2

The first collapses to:
1a.  OTHER  V-S
and the second to:
2a.  S PRED V FIN

Codes:
S = Subject
V = Verb
OTHER = all non-V, non-S clause components
PRED = Predicate
PRED-# = Predicate part #
FIN = Clause-end marker

Other forms may exist, but they're not obvious to me.

So I conclude that for a verb-penultimate language to remain
with the verb stably in that position, you'd want it to have an
explicit clause-end marker, that speakers felt to be indispen-
sable.  But I think that time would erode its felt necessity.


And Patrick Littell replied to Jim:
[snip]
> A number of languages -- not just Germanic ones -- have V2-like
> constructions, but I can't name a single one with "V(-2)"-like
> constructions.   If I'm looking through a grammar and find "X must be
> the second constituent in a sentence" I'm no longer surprised; I'd be
> surprised (and of course very interested) if I ran across "X must be
> the second-to-last constituent in a sentence."  (This is assuming some
> variety of constituent order -- if every sentence ended in, say, a
> verb followed by a evidential particle, this "V(-2)" phenomenon would
> not be much to write home about.)
>
> So far as I've gathered, it's thought to be impossible.  I'd greatly
> like to run across one, but I never have.
I don't think it impossible, just likely to be unstable.


Regards,
Yahya

--
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.385 / Virus Database: 268.4.4/319 - Release Date: 19/4/06

Reply