Re: How many verbs?

To summarize briefly (?):
- I think that the general concept of 'verb' has
little interest because different languages will give
a different idea of what is a verb
- and also because, even in a particular language, the
concept of 'verb' includes a whole lot of things
having as little to do together as cats and dogs

For ex:
- An oak 'is' a tree
- Two and three are five (French: 'font cinq')
- He 'is' sick (state)
- It is five o'clock (temporal situation)
- He 'became' a lawyer (change of state)
- The dog 'ate' the sausage (action)
- The sausage 'was eaten' by the dog (undergoing an
action)
- The rain 'is falling' (phenomenon)
- This road 'seems' to be closed (modal)
- He 'started' to run (inchoative)
- He 'went on' reading
- This 'leads' us to nowhere (?)
- Cela 'suffit' (English: This is enough = 'to be' +
adverb)
- J'ai 'failli' tomber (I nearly fell)
- etc.
And then, on the other side, you have examples like:
- La marche à pied est un bon exercice (Walking is a
good exercise; "marche" = noun)

This is just the same as the use of cases in different
languages. They rarely coincide.

Recently, there was a discussion here about the way of
treating the verb "to be" in a conlang. But "to be" is
an English verb, that means that, if you start from
this point of view, your idea will be biased from the
beginning.

AFAIK, in Chinese, there is no real distinction
between noun and verb. And to me, "development" might
well be a noun, syntactically speaking, this doesn't
mean anything, semantically speaking.

I opened my "Grammaire Basque" (Pierre Laffitte): "Le
verbe est le mot qui sert à indiquer l'état ou
l'action du sujet". True, first edition was 1944,
renewed 1962. But in my Larousse dictionary (1979):
"Verbe. Ling. Mot qui, dans une proposition, exprime
l'action ou l'état du sujet et porte des désinences de
temps et de mode". I could find more examples if I
took the time to look for them.

My 'Dictionnaire de Linguistique' (Larousse, 1991)
gives us more light upon the problem:
"En grammaire traditionnelle, le verbe est un mot qui
exprime le procès, c'est-à-dire l'action que le sujet
fait (comme dans "L'enfant écrit") ou subit (comme
dans "Cet homme sera battu", ou bien l'existence du
sujet (comme dans "Les méchants existent"), ou son
état (comme dans "Les feuilles jaunissent", ou encore
la relation entre l'attribut et le sujet (comme dans
"l'homme est mortel"). D'une manière purement
conventionnelle et sans que le sens le justifie
vraiment, on a admis que "faire l'action" s'étendait
dans ce cas à des phrases comme "La maison a reçu une
bombe" (où en réalité la maison subit l'action). On a
subdivisé les verbes en transitifs (...) et
intransitifs (...) Les transitifs ont été divisée
eux-mêmes en transitifs directs (...) et transitifs
indirects (...)

(This looks very much like ad hoc definitions. Every
time a case doesn't fit the general pattern, you
invent a new specific pattern to make it fit it.)

"En linguistique structurale, le verbe est un
constituant du syntagme verbal dont il est la tête; il
se définit par son environnement, c'est-à-dire par le
fait qu'il est, en français par exemple, précédé d'un
syntagme nominal sujet et suivi éventuellement d'un
syntagme nominal objet (+ marks for tense, person and
number).

(This is not even true. Ex: "Reste à savoir qui
décidera" (no "syntagme nominal" before "reste").
"Reprendre, c'est voler". "Laisse tomber !" etc.)

So semantically, one cannot define a verb except by an
accumulation of ad hoc clauses, and even
syntactically, it doesn't seem so easy. And anyway,
even if the definition were OK, it would be so only
for the very language you are considering.

What Trask says seems to lead to even more awful
conclusions: "Each verb typically requires, etc" : I
understand it as "every single verb will work the way
it decided to work". So you might as well say that
cats are dogs who decided not to bark, but to mew.

"You appear to dismiss everything except semantics".
Let's say that semantics (and cognition) is what is
really interesting me. The rest of it looks very much
like surface schemes to me, or apparences. Isn't it a
scientific way of thinking to try to go beyond
apparences, or to get to a higher level of abstraction
?

So if I had to think about a new conlang (but this is
only my personal point of view), I would first analyse
the different concepts we can find, for example when
examining English or French verbs (and if possible,
many other languages, and also other syntactical
categories) and then I would define classes of
conceptual relations, and it might well happen that in
the end, there would be nothing like "verbs" in that
language, because verbs don't exist on their own. They
are nothing but external shapes that the mind uses (or
not) because it has nothing better at hand in a
particular natlang.


=====
Philippe Caquant

"High thoughts must have high language." (Aristophanes, Frogs)


		
__________________________________
Do you Yahoo!?
Vote for the stars of Yahoo!'s next ad campaign!
http://advision.webevents.yahoo.com/yahoo/votelifeengine/

Reply