[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >


[Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: [jboske] grammar & pseudogrammar



#>>> fracture@hidden.email 04/29/03 03:23am >>>
#On Tue, Apr 29, 2003 at 02:56:00AM +0100, And Rosta wrote:
#> Jordan:
#> > On Tue, Apr 29, 2003 at 01:56:00AM +0100, And Rosta wrote:
#> > > Speaking as a grammarian, the current debate about the 'grammar'
#> > > of NAI strikes me as a waste of effort. A grammar of a language
#> > > defines a mapping between sound and meaning. The so-called
#> > > 'grammar' of Lojban does not do that; it is a pseudogrammar
#> > [...]
#> > 
#> > s/pseudo/formal/
#> > 
#> > A grammar of a language maps between symbols and the possible
#> > sentences in the grammar 
#> > 
#> > Perhaps you're just using a different (less formal) sense of the
#> > word "grammar" 
#> 
#> I'm using a different sense of the word grammar, the sense that
#> is applied to human language by the academic discipline devoted
#> to the study of human language. In linguistics, a 'formal grammar'
#> is just that, a formal grammar, and not a pseudogrammar like
#> Lojban's. As we have said before on this list, the thing that
#> Lojban (and computer languages, I gather) call a 'grammar', is
#> (or at least was) in linguistics called a 'grammaticality checker'.
#
#Since the terminology comes from Chomsky, and Chomsky is a linguist,
#I would have to respectfully suggest that at least two senses of
#the word "grammar" exist in linguistics.

I guess you're referring to _Syntactic structures_, rather than anything
from the subsequent 45 years. I will accept that the view of grammar
in _Syntactic structures_ is quite close to Lojban's, but in my view it
is something of a historical aberration.

To be sure, both "grammar" and "syntax" are often used in opposition
to "semantics" to describe the portion of the overall rule system (that is
also called 'grammar') primarily structural and has comparatively little to do 
with meaning -- but with the occasional aberration, the rule system is
taken, commonsensically, to be a mapping between sound and meaning.

#How long ago did you go to school?

1972.

#> > > It strikes me as silly to rule out a potentially meaningul string
#> > > just because the pseudogrammar prohibits it, given that the
#> > [...]
#> > 
#> > By definition any sentence which is not valid according to the
#> > formal grammar is simply not a sentence in the language 

My point is that it's pointless to accept the formal grammar as
defining the language -- it buys you nothing. It tells you that some
strings are kosher and some aren't, but contributes nothing to
the task of mapping sound to meaning.

#> > I'd love to see you making this kind of bullshit claim about a
#> > language like C++, btw.  
#> 
#> I wouldn't dream of making any sort of claim about a programming
#> language -- I restrict my claims to human languages. The former
#> is your area of expertise, the latter is mine. My contribution
#> to the debate is premised on the idea that Lojban is a human
#> language.
#[...]
#
#There's no difference.  Languages are languages (strings of symbols,
#presumably with meaning).

That sounds very much like Chomsky 1957. But, as I say, that was
an aberration. A language is a set of sentences consisting of a 
pairing of sound and meaning. Or it is a set of rules defining possible
sentences. 

Another thing I should make clear. Sometimes languages do have
apparently arbitrary restrictions ruling out sentences that one would
otherwise expect to be okay. So I'm not saying natlangs don't 
contain anything analogous to the Lojban grammar. But I do say
that such restrictions are functionally superfluous to the design of
a language. 

--And.