Re: a case-free language?

On Friday, October 1, 2004, at 01:41 , Henrik Theiling wrote:

> Hi!
>
> Rodlox <Rodlox@...> writes:
>>  just wondering...
>>
>>  are there any languages (natural or conlang) that lack (or very nearly
>> lack) cases?
>
> I had a long discussion about this some time ago.  And I was convinced
> (although probaby only by making the other people have flat foreheads
> from banging their head on the table because I am so stubborn) that
> 'case' is something purely morphological and many languages achieve
> the same effect (marking semantical roles and argument assignment)
> differently, e.g. by word order (e.g. Chinese) or adpositions
> (e.g. Japanese, Korean).  Thus these languages are said to not have
> case.
Yes, it does depend upon what you mean by 'case' and Rodlox has not made
it clear. Trask lists three meanings. The fact that he seems to think it
unusual for a language to lack case, makes me think he might not be
thinking in terms of morphology.

Trask's first definition is what Henrick and I (and many others)
understand by 'case', namely: "A distinctive, overtly marked form which
can be assumed by a NP [noun phrase] to indicate that that NP bears some
identifiable grammatical or semantic relation to the rest of the sentence.
"

As Henrick says, many languages lack cases entirely in this sense of the
word. One doesn't even have to go to the Orient; on my one island Welsh
has no cases. English retains overt case marking only with a few pronouns.
  (The possessive 's is, as has been argued very many times on this list,
an enclitic in modern English). Most of the African languages I've
encountered lack cases...and so on.

The other two meanings of 'case' given by Trask are:
2. "In GB [Government Binding Theory], a putatively universal abstract
property of noun phrases which is an extension of 'case' in sense 1. Every
overt NP, in this view, must be marked by the grammar as bearing exactly
one of a set of abstract 'Cases'....."
3. "See _deep case_"

I won't go into 'deep case' here, but under Case Grammar, Trask has:
"A theory of grammar which regards deep case as the grammatical primitives
in terms of which sentences are constructed."

Now clearly in the sense of 'Case' in GB and of deep case in the Case
Grammar theory, all languages must have cases.

I am not suggesting that Rodlox subscribes to either the Government
Binding or the Case Grammar systems, but his question does suggest he
might be thinking of case as some abstract concept similar to the meaning
of _Case_ or _deep case_ (yes, I know they are not indentical!).

The short answer to his question, I guess is:
- if you mean (as Henrick & I do) an overtly morphological marking that
identifies the relationship of the NP to the rest of the sentences, the
answer is lots of languages lack them;
- if you mean some abstract grammatical property or primitive, then
clearly no languages lack them.

Ray
===============================================
http://home.freeuk.com/ray.brown
ray.brown@freeuk.com
===============================================
Anything is possible in the fabulous Celtic twilight,
which is not so much a twilight of the gods
as of the reason."      [JRRT, "English and Welsh" ]

Reply