Newsgroups: sci.lang
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!news.ultranet.com!news.sprintlink.net!pipex!uknet!festival!edcogsci!iad
From: iad@cogsci.ed.ac.uk (Ivan A Derzhanski)
Subject: Esperanto and UG (Re: One point against Esperanto)
Message-ID: <D69y70.FDH@cogsci.ed.ac.uk>
Organization: Centre for Cognitive Science, Edinburgh, UK
References: <D5ICH0.Ho1@indirect.com> <D61nCy.LGL@cogsci.ed.ac.uk> <3lcacl$2rj@lyra.csx.cam.ac.uk>
Date: Thu, 30 Mar 1995 22:23:22 GMT
Lines: 96

In article <3lcacl$2rj@lyra.csx.cam.ac.uk> etg10@cl.cam.ac.uk (Edmund Grimley-Evans) writes:
[replying to me]
>> Note that in Esperanto prepositions govern the nominative, and that
>> doesn't seem to be the case in any language which has oblique cases.  [...]
>> I suspect that Esperanto is typologically unique in this respect, [...]
>
>Of course you could make Esperanto fit your proposed universal simply
>by defining it to contain a third case - the "prepositional case", or
>whatever you would like to call it - which happens to be identical to
>the "nominative".

Yes, of course.  A more interesting question is whether our mind,
which supposedly applies laws modelled by the ones we formulate,
can introduce such a distinction (in case it needs it at all).
If it does not, then an isolated community in which E-o is learnt
as a first language and never taught prescriptively is likely to lose
the _-n_ marking.  We can't set up such an experiment, but perhaps
there is something to be learnt from the errors learners make.

>That's the problem with linguistic universals; they tend to be
>universals of descriptions rather than of languages themselves ...

True.  Though that doesn't make linguistic universals any different
from universal observations made in any other field of research:
after all, all we ever get to deal with are our representations
of various aspects of reality rather than reality itself.

But you have a good point.  I've sometimes wondered what happens when
a Greenbergian universal of the form `If condition C holds, distinction
D must be made' comes to be violated in some language in the process of
regular evolution.  Will the language strive to undo condition C?

For example, a universal states that no language has a trial unless it
also has a dual.  Suppose a language has 4 numbers (Sg, Dl, Tr and Pl),
but a sound shift causes the Dl and the Pl inflexion to fall together.
Will every generation of speakers reconstruct the underlying distinction
between the superficially indistinguishable Dl and Pl?  Will the language
necessarily evolve towards eliminating Tr also (undoing the condition of
the universal)?

Opinions welcome.

>It's interesting that Esperantists hardly ever talk about the
>"nominativo", only about presence or lack of the "akuzativo",

Except for the way Zamenhof formulated the rule about prepositional
case marking.

>which makes one suspect that we are not really dealing with case at all
>but rather with some kind of postposition. If there are difficulties
>with using the accusative ending - which some people claim - it might be
>because it is the only postposition in an otherwise prepositional language.

That happens more frequently than one might expect.  The Russian _radi_
and the German _wegen_, both meaning `for the sake of', are postpositions
in otherwise prepositional languages (though they are also prepositions).
The Persian <rA>, also one of a very few postpositions in a prepositional
language, marks direct objects, though it differs from _-n_ in that it
requires its objects to be specific.

>Anyway, what I really wanted to mention was that the "accusative" in
>Esperanto is also used with adverbs:
>
>    tie    there
>    tien   thither
[...]
>Does this not also suggest that we are not dealing here merely with a
>regularised Indo-European noun inflection?

It does.  Whatever adverbs may do in IE languages, they don't decline.

>Another linguistically relevant fact about the accusative in Esperanto
>is that there is an apparent tendency for speakers to use the incorrect
>form [2] while there is no tendency to use the incorrect form [3]; [1]
>is the correct form:
>
> [1] Li ricevis multe da mono.
> [2] *Li ricevis multe da monon.
> [3] *Li ricevis multen da mono.

Does [1] assume that _multe da mono_ is not a direct object, but
rather an adjunct of some sort?  That's how it looks, and it is natural
that speakers should revolt against that, since _ricev-_ is transitive.

In [2] it is (re)interpreted as a direct object, anyway, and it seems
that _-n_ is assigned to the adverbial phrase and somehow percolates
to the noun, overriding the prepositional case assigned by _da_.
I don't think this is linguistically implausible.  (Compare the
Slavic numerals from 2 up, which assign an oblique case unless
another oblique case is being assigned from higher up.)

-- 
`"Na, na ... ah mean, *no wey*, wi aw due respect, ma lady," stammers Joe.'
Ivan A Derzhanski (iad@cogsci.ed.ac.uk)    (J Stuart, _Auld Testament Tales_)
* Centre for Cognitive Science,  2 Buccleuch Place,   Edinburgh EH8 9LW,  UK
* Cowan House E113, Pollock Halls, 18 Holyrood Pk Rd, Edinburgh EH16 5BD, UK
