[Lexicog] semantic domains

[Wayne Leman]

--- In lexicographylist at yahoogroups.com, rrhodes at c... wrote:
Rudy is right that a simple ratio of nouns to verbs is not helpful.
Is a noun + light verb construction to be counted, following the
syntax, as a one noun and one verb? or just as a noun? Or do we
follow the semantics and count it as a verb.
	Rudy's question comes up in Iroquoian and Wakashan (but NOT,
as Ken Hill points out, in Hopi), i.e. do you count the verbally
inflected deverbal nominals as nouns or verbs? Similarly in
Algonquian where fully inflected verbs in a certain subordinate mode
of inflection can be used as a relative clause or, if there are no
other constituents in the relative clause, as a simple noun. The
choice again depends on both your morphosyntactic predilections and
your view of the syntax in particular tokens.
	To ask a question about languages preferring nouns or verbs
brings up a point that construction grammarians will like. The
preference is only meaningful if there is a difference in meaning
between being a noun and being a verb AND if that difference in
meaning is pre-linguistic, i.e., does it "mean" the same thing to be
a noun in English that it "means" to be a noun in Mohawk, or Makah,
or Ojibwe. This very rapidly becomes a Whorfian question and
therefore quite hard to answer (as per Fritz Goerling's comment on
Herbew and Greek).

	There are, however, more tractable questions in this
ballpark, that have lexicographic implications. For example, I'd like
to know what the difference in lexical pragmatics (in Horn's sense --
approx. usage) is between:

   	verb		light verb + noun
	nap		take a nap
	shower		take a shower
	look		take a look
	decide		make a decision/take a decision
	offer		make an offer
	work		do work
	walk		go for a walk

			etc., etc.

Several things can be said about these light verb constructions, e.g.
take as a light verb goes with things you can only do for yourself,
hence the apparently contradictory phrases:

	take a whiz, take a dump, (and their less euphemistic
counterparts)

in which you are leaving something, not taking it. But such insights
don't help in the usage department. There is a distinct preference in
English for adverbial notions to show up as modification on the noun
in the V + N form:

	nap for a short while	= take a short nap
	shower quickly		= take a quick shower
	?*decide firmly		= make a firm decision

But then that also works for genuine V + object as the famous
Wodehouse line goes:

	He smoked a lugubrious cigarette.

(I can't remember the exact reference to the thirty-or-so-year old LI
article in which this was the key example.)

	Then there are the collocational examples, in which the verb
isn't exactly light, but the choice of a quasi-light verb is governed
by the object either lexically or by semantic class.

   					quasi-light verb + object
	make an offer		=	submit an offer/tender an
offer
	??do a crime		=	commit a crime
	VERB time		=	spend time

			etc., etc.

These questions are the underpinning of the Apresian/Melchuk approach
to lexicography -- work not widely known in the English speaking
world.

	There is probably a coherence in lexical pragmatics with
these V + object constructions, but it's also a dissertation sized
question.

Rich Rhodes


>Peter is quite right about the issue, but it goes further than the
>event-thing distinction, since in many American Indian languages,
e.g.,
>the famous example of Hopi, "object" names like "house" are verbal
>predicates, so that the translation of "This is a house" would be
more
>like "This houses". This would indeed drastically affect the ratio of
>nouns vs verbs. I know of no study of this ratio, but it may be
because it
>is not a very significant point to study.
>
>Peter is also right about the history of Persian, in which older
single-
>morpheme verbs have been replaced by Noun + Light Verb
(like "do", "make")
>over the centuries as the only productive process, leaving few
original
>simplex verbs. This seems to be a common process in SOV languages,
as in
>Korean and Japan, where Chinese verbs were borrowed as Noun + Light
Verb
>("ha" in Korean, "su" in Japanese), and now English words can form
the
>basis of new verbs in this manner. The process in Persian seems to
have
>spread from the Dravidian languages in India, where it has gone on
so long
>that the light verbs have become suffixes rather than independent
bases,
>and are being phonetically eroded away, leaving the originally
attached
>morpheme as first a root and then as the full Verb itself, now
creating
>all new simplex verbs. This again would raise the question as to the
>significance of questioning the ratio of nouns to verbs, since this
might
>change over time in a single language.
>
>	Rudy Troike
>	University of Arizona
>
>
>
>
>
>Yahoo! Groups Links
>
>To visit your group on the web, go to:
>  http://groups.yahoo.com/group/lexicographylist/
>
>To unsubscribe from this group, send an email to:
>  lexicographylist-unsubscribe at yahoogroups.com
>
>Your use of Yahoo! Groups is subject to:
>  http://docs.yahoo.com/info/terms/


--
******************************************************************

  Richard A. Rhodes
  Department of Linguistics
  University of California
  Berkeley, CA 94720-2650
  Voice (510) 643-7325
  FAX (510) 643-5688

******************************************************************
--- End forwarded message ---