LINGUIST List 7.1812

Sat Dec 21 1996

Disc: Analogy

Editor for this issue: T. Daniel Seely <seelylinguistlist.org>


Directory

  1. G.J. VANDEN WYNGAERD, Re: 7.1794, Disc: Analogy
  2. Deborah Schmidt, analogy
  3. Patrick Farrell, Re: 7.1778, Disc: Analogy
  4. neulin!neulin.fmed.uba.ar!postmastccc.uba.ar>, Analogy

Message 1: Re: 7.1794, Disc: Analogy

Date: Fri, 20 Dec 1996 14:26:36 +0100 (MET)
From: G.J. VANDEN WYNGAERD <WYNGAERDrullet.LeidenUniv.nl>
Subject: Re: 7.1794, Disc: Analogy

Sentences of the type

(1)	I saw the barn red

have been discussed in Hoekstra (1987) "Small Clause Results", Lingua 74,
101-39. Hoekstra observes that resultative complements only occur with
activity verbs. 'see (sth)' being an accomplishment, the construction does
not permit a resultative object complement. If 'see' is given the semantics
of an activity verb, the resultative construction becomes acceptable, as G.
Nathan points out:

>	Consider, for example, a world similar to that proposed in The
>Lathe of Heaven in which some person has the ability to change the
>color of objects merely by choosing to see them that way. In such a
>case (1) becomes grammatical in just the same way as
>
>2)	I painted the barn red.
>
>	A change in the MEANING of `see' (or perhaps in the state of
>affairs in the world) renders the analogy fine.

While there is some truth to the claim formulated by the analogists that it
is something about the meaning of 'see' that makes (1) deviant, one would
still like make this general idea as precise as possible. The repeated
suggestion made on LINGUIST to the effect that certain analogies are
'unreasonable' or 'conceptually impossible' merely begs the question.
Observe that it is perfectly well possible to say

(3) The teacher watched his pupils into submission

with a verb ('watch') whose semantics is very close to 'see', but with the
difference that 'watch' is an activity verb. So it is not unreasonableness
or conceptual impossibility that is at stake here, but rather some more
subtle aspect of grammatical meaning that makes it impossible to draw
certain analogies.

The same holds for the easy/eager contrast: it is a truism to say that
ultimately some aspect of the meaning of both adjectives must entail that
they occur in different structural configurations; but saying that the
analogy is impossible because it is unreasonable or incompatible with the
meaning of 'eager' merely begs the question, which is, how do children
acquire the meaning difference between these words, and word meanings more
generally? While there are few insightful answers to give in this domain,
the classical poverty of the stimulus argument goes through to show that
large portions of this knowledge must be innate.

Guido vanden Wyngaerd
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 2: analogy

Date: Fri, 20 Dec 96 16:33:27 EST
From: Deborah Schmidt <DSCHMIDTuga.cc.uga.edu>
Subject: analogy

I must admit that much of the discussion on the question of analogy has been
too complex for me to follow. So perhaps I'm way off direction, but why has
there not been introduced into the discussion examples of ill-formed
sentences that are NOT semantically anomalous? I think that the same video
series that used the "see the barn red" example also compared the following:
 What did Bill see an elephant with? vs. *What did B see an elephant and?
These are old standbys for syntactic theorizing, along with sentences like:
 Who did John claim that Mary kissed? vs. *Who did John make the claim...
Would naive analogizing from well-formed wh-mvt data predict these, or not?
(I guess I should first ask whether "analogy" is at work when we produce
complicated novel sentences on the model of simpler sentences we've heard.)
Would some more sophisticated definition of the rules of analogy do? If so,

would the articulation of that more sophisticated definition have to include
the statement of universal syntactic principles? Would it be essentially
equivalent to the articulation of universal syntactic principles?

Blue is to purple as yellow is to orange, right?
How about: Blue is to green as yellow is to orange.
Depends on the rules of the game. Blue is to purple as sodium is to salt?
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 3: Re: 7.1778, Disc: Analogy

Date: Sat, 21 Dec 1996 02:10:01 -0700
From: Patrick Farrell <pmfarrellucdavis.edu>
Subject: Re: 7.1778, Disc: Analogy

Daniel Everett replied to my comments on analogy as follows:

>>If that is so, then I think one must
>> recognize that one kind of constraint on analogy is:
>>
>> 3. Avoid conceptual anomaly.
>
>Look, quite frankly I am happy that you and your students had a nice
>discussion about analogy. But there is a lot of work to answer before you
>propose ideas as serious counterproposals. I do not think this
>'avoidance' principle has been established at all.
>

I meant (3) to be a kind of brief expression of the general idea that
certain kinds of linguistic utterances are not produced because they are
nonsensical,
as suggested in one way or another by Straight and Israel. To my mind,
Nathan makes a good case for why this is the best approach to explaining
the failure of the seeing the barn red analogy. In suggesting
an explanation for why this particular analogy is not made, I was not aware
that I was making a counterproposal. I am still unclear. Did someone ever
show that a principle of universal grammar would explain the matter at hand?
Everett claims that

>Work on processing (see especially Ted Gibson's work in
>numerous articles and a forthcoming MIT Press book) shows fairly clearly
>that in the type of case raised here there are grammatical principles
>ruling out the relevant structures

It might be helpful to anyone interested in pursuing this matter to have
either a summary of the relevant principles and how they would apply to
the example, or a more specific reference to where and in what
particular work the relevant case is discussed.

Actually, the idea that certain kinds of utterances are not produced because
they are nonsensical has been, as far as I can tell, generally assumed both
within
and outside generative grammar. Consider the following analogy
failure:

The child bothered me : The withdrawal of the toops bothered me ::
The child loved me : The withdrawal of the troops loved me

Presumably "The withdrawal of the troops loved me" is precluded as nonsense
because of what we know about the meaning of the words and the construction
in which they occur and our general belief system about the world. The
withdrawal
of the troops is not the kind of thing that we would consider to be capable
of loving.
In generative grammar these kinds of cases have been typically treated as
violations
of "selection restrictions" (see for example Chomsky 1957, 42-43,
1965, Radford 1988, 369-371), which Jackendoff (1990, 50-53),
for example, characterizes, in essence, as constraints on the expression of
a verb's arguments
that follow from its meaning or conceptual structure.

Along similar lines, consider the following analogy that we do not draw:

She brought a letter to her mother : She brought her mother a letter ::
She carried a letter to the mailbox : She carried the mailbox a letter

Pinker (1989) justifies at length an account of such cases that goes, in
essence as follows. "...The double object form expresses causation of a
possession
change" (98). Since carrying a letter is not an action that can terminate in the
mailbox possessing anything, "She carried the mailbox a letter" is basically
nonsense. I know of no alternative approach to explaining
this kind of case (short of simply marking "carry" with a diacritic
that precludeds dative movement or something
analogous), although there are of course various slightly different
analyses that take the same general approach.

In short, I meant (3) to be simply a brief expression of what I take to be
a basic assumption underlying this approach and all accounts of
selection-restriction-type
phenomena.

Is there some alternative explanation of these cases that utitlizes principles
of universal grammar (an empty category must be properly governed,
move around a tree only at logical from if you can, etc.) that anyone would be
willing to explain and defend as fundamentally distinct from and
preferable to the admittedly simple-minded "avoid nonsense" approach?

Now, what I tried to express earlier was:
I can't help but wonder why one might implicate universal grammar (an empty
category must be properly governed, move around a tree only at logical form if
you can, etc.) as a source of explanation for what appear to be failed
analogies due to semantic/conceptual constraints of the selection-restriction
variety. I think that Straight was wondering more or less the same thing.
I am still wondering ... which is OK by me.

Stirling Newberry writes:

>> 3. Avoid conceptual anomaly.
>
>This doesn't work at all - there are plenty of analogies that *work* in
>language which are conceptually anamolous - and in fact were chosen because
>they allowed the dealing with concepts which are irreducible directly. A
>simple example is anthropomorphization - many things do not seem to follow
>any sort of pattern which we can reduce to words. The analogization of
>ships and storms as feminine is an example of this - by creating the
>lingusitic analogy it reinforced the behavioral analogy in other parts of
>the brain. In language having classes of nouns be "masculine" or "feminine"
>or meter schemes or rhyme schemes. These are all - if examined -
>conceptually anomalies.

Well, yes of course "conceptual anomaly", like "nonsense", is not at all
precise. If one were to say that everything that we say that is not
"literally" true
(whatever that means) is conceptually anomalous, then the bulk of what we
say is probably conceptually anomalous ("I don't hold that belief", "My
time is up", etc.).
Obviously, one would want to say that nonsense is stuff that is anomalous GIVEN
our knowledge of word meanings, our overall conceptual system, our beliefs
about how the world is, our knowledge of the grammar of our language, etc.
Our overall conceptual system might itself be
said to be anomalous from the perspective of a computer or a Vulcan or
something,
but that doesn't seem interesting. For human beings, anthropomorphization, for
example, is not anomalous. It is normal. We go the the back of the bus or
the foot of
a mountain, we think verb phrases have heads, we drive on the shoulder of
the road, and
so on (see Lakoff & Johnson 1980 and much subsequent work on the principled
metaphorical nature of our conceptual system and how this manifests itself
in routine
language).

Benji Wald writes:

>So far, all I'm saying is that "she saw the barn red" and "to please her is
>eager" seem to be weird in quite different ways, not surprisingly, since
>they are based on quite different "analogies" (one with properties of
>"easy" and one with those of "paint"). Quite a while ago I even claimed
>that "she saw the barn red" was "grammatical" as far as I was concerned (I
>wouldn't exclude it from the grammar of English, and can even interpret it,
>as could quite a few other discussants at that time, including in ways I
>grasped but hadn't thought of beforehand.) I wouldn't say the same for "to
>please her is eager". It is not interpretable to me by any stretch of the
>grammar -- which is not to say that I insist the violation is "syntactic"
> than "semantic".

Yes they are quite different. As Nathan points out "She saw the barn red" seems
fine in the context of an imaginary world in which beings have the ability
to change the color of objects by how they choose to see them. This is a kind of
world that is readily imaginable. One might even make a movie incorporating this
idea. Take the case of "The withdrawal of the troops loved
me". What would be the imaginary world that would make this OK? Well suppose
the world is such that events can become beings that we can talk to. Then
it would be OK to say things like "Has she spoken with the destruction of
the city yet?" and "The withdrawal of the troops sure seems to be in love with
me" and so forth. The basic problem here is that nobody would want to see
this movie.
And who would want to see a movie in which people's wishes were constantly
turning
into sentient beings ("to win the lottery" pops out of the star's head and
starts
talking to her, "to finish my dissertation" pops out of someone else's head
and starts walking around, and so forth). This would be the imaginary world
in which "To please her is eager" would be OK. Ridiculous, right?

On a final note, I wish to express my regret for having ventured out into this
discussion, whose rules I'm not sure I understand (like, I'm probably
breaking them now). I think I now understand
why my first posting elicited several private responses. It seems that people
are willing to talk to each other in this forum in ways that are considerably
less civil than what they would employ in face-to-face interactions, private
e-mail exchanges, or in academic publications. I know I feel rather wounded by
a few of the things that have been said. I hereby apologize for anything
that I may have
said that had a similar effect on anyone else. I only intended to contribute
my thoughts about theoretical issues that interest me. I will assume that
the same is true for all the participants in this discussion, whom I hold
in the highest respect.


References
Chomsky, Noam. 1957. Syntactic Structures. The Hague: Mouton.
Chomsky, Noam. 1965. Aspects of the Theory of Syntax. Cambridge, Mass.: MIT
Press.
Jackendoff, Ray S. 1990. Semantic Structures. Cambridge, Mass.: MIT Press.
Lakoff, George, and Mark Johnson. 1980. Metaphors We Live By. Chicago:
University of Chicago Press.
Pinker, Stephen. 1989. Learnability and Cognition: The Acquisition of
Argument Structure. Cambridge, Mass.: MIT Press.
Radford, Andrew. 1988. Transformational Grammar: A First Course. Cambridge
and New York: Cambridge University Press.

Happy Holidays

Patrick Farrell
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 4: Analogy

Date: Fri, 20 Dec 96 16:26 GMT-3:00
From: neulin!neulin.fmed.uba.ar!postmastccc.uba.ar> <neulin!neulin.fmed.uba.ar!postmastccc.uba.ar>
Subject: Analogy

Robert Petterson correctly points out the plausibility of the fourth sentence
in a certain context. 

> > The boy paints the red barn : The boy paints the barn red ::
> > The boy sees the red barn: The boy sees the barn red

> Sorry, I just had to do this...
> 
> The boy puts red-tinted glasses on, I put on green. There is a white barn
> outside. When we go out, **the boy sees the barn red**, while I see it
> green!
 
That example may not be a happy one, but that can be easily overcome
by replacing "sees" with "discovered" or lots of other verbs. It would be
possible to explain why we don't make that analogy with reference to the
semantics of the verbs involved. But I think the discussion has somewhat
missed the point with respect to what Chomsky claims: the 
acquisition of sintax cannot be analogy-based. That doesn't mean we (both
adults and children) don't make analogies at all or that analogy is not
a valid way of reasoning. It only (and it is quite a lot) means that if
we were to acquire sintax by making analogies from sentences we hear nothing
could prevent us from producing:

(1) To please John is eager 

out of 

>> John is easy to please : To please John is easy ::
>> John is eager to please.

And (1) is not produced by adults nor by children. This is the kind of
evidence that Chomsky and others analize to support their claim for a 
specific language faculty. Language acquisition (and pathology of
language) cannot be explain on terms of general cognitive structure
and processing, at least with our present knowledge about how things are.
 
With respect to Petterson's question

> What about this? Does it count if you use commas?
> 
> A: John seems so eager to please and help out.
> B: Well yes... **to please, John is eager**, it is true, but to help out
> ... well, he's more of a hindrance than a help.

The answer is no, it doesn't count because the idea of the
example is to raise the phrase "to please" and make it the subject of 
"is easy" together with "John". With the use of commas you are just
displacing or moving an element but the "structure" of the sentence is
the same as "John is eager to please"; while the idea was to have another
sintactic structure with the same meaning. That is possible for the first
case (easy) but not for the second (eager). This is of course much better 
and more interestingly explained in the original reference (cf. Chomsky), 
which I highly recommend for those interested in the topic.
 
Maria Cristina Cuervo
Universidad de Buenos Aires
Argentina 

 Maria Cristina Cuervo 
 postmastneulin.fmed.uba.ar
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue