Re: polysynthetic languages

>  > Which leads to my next question...Are there any natural
>  > polysynthetic languages that do mark the nouns for case?  It seems
>  > to me that it would certainly be *possible* for polysynthesizm and
>  > a case system to be found together because there are a number of
>  > languages out there that are fond of redundancy and wouldn't mind
>  > marking everything twice.  Alternatively, are there any natural
>  > languages that fall below the morpheme-to-word ratio needed to be
>  > considered polysynthetic, rather than merely synthetic, that both
>  > mark the verb for both subject and object agreement and also mark
>  > nouns for case?
>
>Possibly Georgian.  The person and number of the subject, direct
>object and indirect object can be indexed on the verb, and nouns are
>declined in seven cases.  I don't know how free the word order is, and
>I don't know whether Georgian is considered polysynthetic (I don't
>recall ever seeing the term applied to it).  There are people here
>who've actually studied the language - hopefully they can provide more
>information.  I the meantime,
>http://www.armazi.demon.co.uk/georgian/grammar.html
>is a nice introductory grammar (and the source of most of what I know
>about the language).

I'll take a look at it when I get back online.  Right now I am working on
my laptop at the skating rink while my daughter practices.  (I knew there
was a reason we bought a laptop.)  I have a power outlet, which is very
good, considering that the Thinkpad's internal battery somehow failed to
charge properly despite the fact that it has been plugged in for the last
couple weeks straight.  I'll have to hope that I can find another power
outlet when I take her to violin class after this.

Mainly, I was looking for confirmation that there were natlangs that were
so redundant that they put both subject and object markers on the verbs and
also marked nouns for case.  Cristophe's comments about Basque and yours
about Georgian confirm that this sort of redundancy is found in more than
one language family.

It's also good to know that Georgian marks verbs for indirect object, since
I had the idea that I wanted Cwendaso to be able to do this for a very
important grammatical reason, since it may not be possible to put case
markings on the vast majority of proper nouns in the language.

What I now have in mind for Cwendaso (and this is going to involve a good
deal more work on my part because it is involves developing the language's
syntactic structure over a 1500-2000 year period) is a language that
started out as heavily agglutinative with proper names being generally
either simple words or compounds.  Then the compounded proper names became
more and more complex until they became actual polysynthetic names.  (This
happened first with personal names and only later to place names.)  So at a
certain point in the language's history, a very substantial (and
increasing) proportion of the proper names were polysynthetic, but the
language itself was not polysynthetic yet but merely highly
agglutinative.  The polysynthetic names created some grammatical issues
(which I won't go into here) about how to properly incorporate them into a
sentence since they were both full sentences and nouns at one and the same
time.  The grammatical markers that arose to deal with the polysynthetic
names later ended up having their use extended so that they were also used
to join a main clause to its subordinate clause, largely displacing the
original methods of subordinating clauses in everyday language.  Meanwhile,
the proportion of polysnthetic names was steadily increasing and was
putting pressure on the language as a whole to use polysynthetic
constructions outside of proper names.  This trend continued until Cwendaso
became a polysynthetic language, and nearly all of the proper names (both
place names and personal names) were of a polysynthetic form.

The Cwendaso (or Tovláugad, as they call themselves) have migrated twice,
and in the first migration they left behind the Emitovláugad, so there now
exists a sister language (or perhaps more than one) to Cwendaso which is
not polysynthetic nor gives polysynthetic names.  The protolanguage was
highly agglutinating, marked subject and direct object on the verb (and
possibly indirect object as well, though that may have arisen later in the
Cwendaso language alone), and had a fully developed case system consisting
of at least nominative, accusative, genitive, dative, instrumental, and
locative (and probably others).  There may have been a vocative case in the
protolanuage, but, if so, it no longer exists in Cwendaso; an entirely
different grammatical structure is used in place of vocative.  The word
order in the protolanguage was either SOV or SVO.  I am going to guess that
the protolanguage already had four grammatical numbers and that Cwendaso
did not develop two of them later.

The Cwendaso are an entirely oral culture, (but won't be for much longer)
and have a very large oral poetic, historical, and ceremonial corpus, the
very earliest portions of which were composed in the last period of the
protolanguage just before the first migration that separated them from the
other speakers of the protolanguage.  Since the Cwendaso are avid
historians, it is possible to pin down the date of composition of very many
of the pieces in the oral corpus to within half a century with fair
reliability.  Phonological changes cannot be tracked through the poetic
corpus, since the pronounciation is continually updated.  (I would be
possible to track phonological changes by attempting to reconstruct the
protolanguage through reference to the sister language(s) and by analyzing
loan words taken into the Trehelish or Nidirino languages at various
times.)  The layers of syntactic change that the language underwent can, to
some extent, be traced through the oral corpus.  The Cwendaso themselves
recognize that the oldest of their songs have a distinctly different style
to them than the more recent ones.  In addition to the modern speech, a
very conservative dialect of the language, based on the oral corpus, is
used for ceremonial purposes, giving speeches, etc., and is both understood
and spoken with varying degrees of eloquence by all speakers of the
language except for young children.

Isidora

Replies