LINGUIST List 3.223

Fri 06 Mar 1992

Disc: Parsing Problems

Editor for this issue: <>

Directory

1. Vicki Fromkin, Re: 3.212 Parsing Problems
2. Georgia Green, Re: 3.212 Parsing Problems
3. Herb Stahlke, Re: 3.212 Parsing Problems
4. , Ambiguous parse

Message 1: Re: 3.212 Parsing Problems

Thanks Carl Alphonce -- for information re the ambiguity of
THE PLAYER KICKED THE BALL KICKED HIM. Interesting that my 'only'
interpretation was your second one -- The player kicked - the ball kicked
him. rather than - The player -- kicked the ball -- kicked him. Wonder
what factors dominate re one's interpretation of ambiguous sentences.
I know work has been done on this. Glad it is not a linguistic problem
or at least not adccording to my view. VAF

Message 2: Re: 3.212 Parsing Problems

I trust everyone realizes now that "The player kicked the ball kicked him"
is not only grammatical, but at least 5-ways ambiguous:

The player who kicked the ball kicked him. [=someone else]
The player who was kicked the ball kicked him. [=someone else]
The player kicked the ball that was kicked to him.
The player kicked the ball that kicked him. [takes more imagination]
The player complained that the ball kicked him.

Message 3: Re: 3.212 Parsing Problems

Has the term "garden path" undergone a semantic shift? I recall that
in graduate school, shortly after Haj Ross' dissertation hit the world
at the Illinois LSA Summer Institute, the term referred to sentences
with island constraint violations like

	Who do you know the woman that married?

and other sentences that were imparsible for because of complex
grammaticality problems. Grammatical sentences with insufficient
redundancy to be easily parsed were of less interest since, at that
time, we could pass them off as performance problems. Not.

Herb Stahlke
Ball State University

Carl Alphonce <alphoncecs.ubc.ca> wrote:
>Subject: Re: 3.194 - Another Parse
>I just realized that there is another possible parse
>for the sentence
> "The player kicked the ball kicked him"
>In addition to
> [ the player [ kicked the ball ] ] kicked him
>there is also the structure
> the player kicked [ the ball [ kicked him ] ]
A few years ago, I developed a grammar of English for a
parsing program written by Phil Harrison at Boeing
Computer Services. After reading the above, I thought
it would be fun to see if "my" grammar parsed the
sentence ambigously. Phil tried it out, and behold,
there is also a third parse. It is distinct from the
second parse above only by the labels on the bracketing,
or putting it differently, the second parse above is
ambiguous, depending on the labels. The least unnatural
(at least to me) labeling of the second parse is:
 [ the player [ kicked [ the ball [ kicked him]]]
 S VP NP S
--in which the first "S" is an independent clause, while
the second is a reduced relative. The other labeling of
the second parse (i.e. the third parse) is:
 [ the player [ kicked [ the ball [ kicked him]]]
 NP S NP S
which can be paraphrased as "the player who was kicked
the ball that was kicked to him." (Both Ss are reduced
relatives.)

Parsing "kicked him" as a reduced relative, as in the
second and third parse structures, requires being able
to passivize the second (direct) object of a
double-object verb, which for some people is
questionable. ("John was given the book" is OK for
everyone, while "The book was given John" is in that
gray area.) I should mention that Phil had to add a
subcategorization for this passive to the lexical entry
for "kick" before the sentence would parse.