Re: How many verbs?

On Saturday, July 17, 2004, at 08:29 , Philippe Caquant wrote:

[snip]
>> But then, what did Larry Trask know about
>> linguistics?
>>
> I was referring to the permanent confusion between the
> syntactical notion of "verb", and the semantics of
> that so-called "verb".
This is not what you said. Let me remind you what you did write on 16th
July:
<< What is a verb ? Nothing. A verb in English, or in Russian, or in
Georgian, might be something, but such a thing as "a verb" (or: "an adverb"
) simply does not exist.) >>

As far as I understand this, you are saying that, while in the analysis of
an individual language like English, Russian or Georgian, there might be,
but not necessarily is, a class of words called 'verbs', nevertheless the
abstract concept 'verb' has no existence. I asked you what your evidence
is that "such a thing as 'a verb' simply does not exist"? It seems a
reasonable question.

> It is nearly impossible to
> define a verb otherwise than syntactically, and yet,
> in every grammar, you will find explanations like
> "verb is used to express action, or state, or changing
> of state", mixed up with syntactical notions,
Strange - I've just consulted three grammars at random - one on Welsh,
another on Xhosa and the third on Czech - and none of them give any such
explanations. I guess you mean "In every grammar I have read". Maybe you
need to read some more enlightened grammars.

> and this
> simply doesn't work. As other list addicts already
> pointed out many times, a verb in language A doesn't
> have to be a verb in language B.
Yes, the set of actual verbs in one language will not coincide with the
set of verbs in another. One important case is of descriptive adjectives
which, in some languages, are stative verbs. But it is singularly odd that
we can recognize them as verbs if the concept 'verb' has no existence.

> To me, the notion of
> "verb" is uninteresting except for purely syntactic
> purposes.
So in terms of syntax 'verb' does have a meaning? So what am I to
understand by: "but such a thing as
'a verb' ..... simply does not exist."?

But this is, surely, a very narrow outlook. More often than not, verbs
have interesting morphologies; and some us - *shudder* - even find them
interesting from semantic point of view.

> What is interesting is the semantic relation
> defined between two concepts.
I refer again to Trask's definition:
"Grammatically speaking, verbs are most obviously distinguished by the
fact that each verb typically requires the presence in its sentence of a
specific set of NP [noun phrase] arguments, each of which typically
represents some particular semantic role and each of which may be required
to appear in some particular grammatical form (particular case marking,
particular preposition etc.)."

Consider carefully:
"..each verb typically _requires_ the presence in its sentence of a
specific set of noun phrase arguments, each of which typically represents
some particular *semantic* role...."

Isn't this to do with semantic relations between concepts? Our troublesome
verbs often involve more than just two concepts, but that's the problem
with real life.

What did Larry Trask's "Dictionary of Grammatical terms in Linguistics"
say about semantic?
"*semantics* /sI'm&mtIks/ n. The branch of linguistics dealing with the
meanings of words and sentences. The relation between syntax and semantics
and the location of the dividing line between them have long been matters
of controversy. Several extreme positions have been maintained. One,
associated with Generative Semantics, holds that syntax and semantics are
not distinct at all, but constitute a unified area of investigation.
Another, associated with the Standard Theory of TG [Transformational
Grammar], holds that the two are sharply distinct, with syntactic
structures being primary and the semantics serving only to interpret
syntactic structures. A third, arguably represented by some versions of
functional grammar and of cognitive grammar, is that the two are distinct
but that the syntax serves only to realize underlying semantic structures.
  Most linguists today would probably advocate some kind of intermediate
position in which the two areas are distinct and make independent
requirements but in which the dividing line is somewhat blurred and each
area can have effects upon the other."

I fall into the category of "most linguists today.....". But your position
seems even more extreme that Trask's "A third, arguably represented.....";
not only do you clearly draw a hard & fast distinction between semantics
and syntax and also quite clearly put primary importance on semantics, but
you appear almost to regard syntax as of little or no relevance.

I, on the other hand, while admitting that semantics and syntax are
different areas, find the dividing line between them somewhat blurred and
consider that the one affects the other.

> Oh, I nearly forgot to add: IMO.
I assume they're your opinions. What I ask is on what evidence you base
your opinions.

You hold, presumably, that it is possible to discuss the 'semantic
relation defined by two concepts' without any reference to syntax. I do
not understand how this is possible. I am seeking enlightenment.

You maintain that 'verb' has no meaning as an abstract concept. Could you
explain to me why you consider Trask's definition, which I quoted above,
is invalid?

Language, surely, exists to mediate between externalizable events (sounds,
  among other things) and non-externalizable events (thoughts). Language
does this by providing a loosely hierarchical collection of signs, each of
which, generally speaking, mediates between sound (the sign's form) and
thought (the sign's meaning). Part of this hierarchy is provided by syntax.

You can, of course, have sounds without thoughts (animals & small children
are good at this!) and thoughts without sounds, but you cannot have
meaning without form nor form without meaning. Unless your thoughts are
given external form (typically, in sound or writing) they have no meaning.
  Language provides a hierachy of levels  ultimately represent just such a
loosely hierarchical collection of signs. Until the sound bits are
associated with the thought bits, it's not language. Semantics, syntax,
morphology, phonology are all, IMO, important and inter-related. You
appear to dismiss everything except semantics.

Ray
===============================================
http://home.freeuk.com/ray.brown
ray.brown@freeuk.com    (home)
raymond.brown@kingston-college.ac.uk   (work)
===============================================
"A mind which thinks at its own expense will always
interfere with language."         J.G. Hamann, 1760

Reply