/marnal,~
ELSEVIER
Journal o f P r a g m a t i c s 26 (1996) 417--436
Review
Language
comprehension
Rachel
Department
article
as structure
building
~
Giora*
o f P o e t i c s a n d C o m p a r a t i v e L i t e r a t u r e , T e l Avi~, U n i v e r s i t y , I L - 6 9 9 7 8 T e l - A v i v , I s r a e l
R e c e i v e d M a r c h 1995
1.
Introduction
The main thrust of current research into language processing
emphasizes
the modularity of knowledge
of language,
and focuses on those aspects of language
which
are specific to our linguistic ability. In contrast, Morton Ann Gerusbacher's
research
attests that many
aspects
of language
comprehension
involve
general
cognitive
processes
and mechanisms.
Gerusbacher
adduces a vast array of data which suggest
that these processes
and mechanisms
underlie both linguistic and nonlinguistic
phenomena.
Her findings show that language
comprehension
is strongly related to general cognitive
systems.
Gernsbacher
assumes
that the goal of comprehension
is to build a coherent
mental representation
of the information
being processed.
According
to this structurebuilding hypothesis,
in order to build a mental representation,
comprehenders
must
first lay foundations
for their mental
structures.
Next, they should develop
their
mental
structures
by mapping
information
o n t o it. W h e n
incoming
information
is
less coherent
with previous
information,
comprehenders
shift and build a new substructure.
The building blocks of mental structures are memory
cells. Memory
cells are activated by incoming
stimuli. Their initial activation forms the foundations
of the mental structures. When incoming
information
coheres with the previous
information,
it
is often mapped
onto developing
structures,
and does not necessitate
the activation
of new memory
cells. However,
when incoming
information
does not overlap with
the previous
information,
it is likely to activate different memory
cells. The activation of these new memory
cells forms the foundation
of a new substructure.
R e v i e w of: M o r t o n A n n G e r n s b a c h e r , L a n g u a g e C o m p r e h e n s i o n
N J: Erlbau~n, 1990.
* E - m a i l : g i o r a r @ c c s g . t a u . a c . i l ; F a x : + 9 7 2 3 6408980.
as Structure Building. Hillsdale,
0 3 7 8 - 2 1 6 6 / 9 6 / $ 1 5 . 0 0 C o p y r i g h t © 1996 E l s e v i e r S c i e n c e B.V. A l l rights r e s e r v e d
SSDI 0378-2166(95)00071-2
418
R. Giora
/ J o u r n c t l (?f" P r a g m a t & ' s
20 (1996) 417--436
Activated memory
cells transmit processing signals which either enhance or suppress other cells' activation. Enhancement
is a m e c h a n i s m
that highlights information necessary for further structure building. Suppression
serves to dampen informat i o n t h a t is n o l o n g e r f u n c t i o n a l i n s t r u c t u r e b u i l d i n g . T h e p r o c e s s e s o f l a y i n g t h e
foundation,
mapping,
and shifting, and the mechanisms
of enhancement
and suppression explain a great deal of linguistic and nonlinguistic phenomena.
2.
The
process
of laying
a foundation
Consider the following facts: Subjects spend more time viewing the first picture
of a story or episode than viewing
later occurring
pictures (Gemsbacher,
1983);
Subjects take longer to read the beginning
sentence of an episode than later occurring sentences within that episode (Haberlandt,
1980, ! 984; Haberlandt et al., 1980;
Mandler and Goodman,
1982); Initial words in a sentence take longer to read than
the same or other words occurring later in the sentence (Aaronson
and Scarborough,
1976; Aaronson
and Fcrres, 1983; Chang,
1980); Target words or phonemes
take
longer to identify when they appear in clause-initial position than when they occur
later (Cairns and Kamerman,
1975; Cutler and Foss,
1977; Foss,
1969, 1982;
Hakes, 1971 ; Marslen-Wilson
et al., 1978; Shields et al., 1974); First content words
of a sentence elicit larger N400 brain wave than words occurring later in the sentence, suggesting that words occurring initially in the sentence are more difficult to
u n d e r s t a n d t h a n l a t e r o c c u r r i n g w o r d s . N 4 0 0 is t h e n e g a t i v e c o m p o n e n t
of the eventrelated brain wave that occurs about 400 ms after the stimulus. N400 brain waves are
associated with difficulty of processing. They increase in response to low frequency
words and unexpected
words in a sentence (e.g., Kutas and Hillyard, 1982).
These facts suggest that comprehenders
spend more cognitive effort processing
information (both verbal and nonverbal) in discourse-initial
position than later in the
discourse. This, however, holds only for coherent discourses. Incoherent discourses
do not manifest such a pattern (Foss and Lynch, 1969; Greeno and Noreen,
1974;
Hakes and Foss, 1970; Van Petten and Kutas, 1987). Gernsbacher
concludes that in
a coherent discourse, comprehenders
use initial information
to lay a foundation
for
their representation
of the entire discourse.
Gernsbacher
views comprehension
in terms of structure building. According
to
the structure-building
hypothesis,
the goal of comprehension
is to build a coherent
mental representation
of the information
processed. To build this mental structure,
comprehenders
must first lay the foundation onto which subsequent information
will
be mapped.
However,
the extra cognitive effort comprehenders
spend at this stage
will be offset by the advantage of first mention. Serving as a foundation onto which
] In one of her notes G e m s b a c h e r explains wily she prefers q u o t i n g to paraphrasing: "'My Inotivation
is not laziness but the belief that paraphrasing is best reserved for investigating subjects' m e m o r y in laboratory e x p e r i m e n t s " (p. 242). W h e n trying to review G e r n s b a c h e r ' s work, I tried paraphrasing despite
m y a g r e e m e n t with her view, but failed. My paraphrasing of G e m s b a c h e r ' s work relies heavily on the
original. Her book is so well written, every paraphrase of it must be second best.
R. G i o r a / J o u r n a l o f P r a g m a t i c s 2 6 ( 1 9 9 6 ) 4 1 7 - 4 3 6
419
s u b s e q u e n t i n f o r m a t i o n is a d d e d , f i r s t m e n t i o n e d i n f o r m a t i o n m u s t b e m o r e a c c e s s i ble than later occurring information. Findings indeed show that first mentioned participants are more accessible than second mentioned participants, regardless of either
their semantic (agency) or syntactic (subjecthood) role, or whether they occupy sentence-initial position. Rather, the advantage of first mention must be due to the role
f i r s t m e n t i o n e d p a r t i c i p a n t s p l a y in f o r m i n g t h e f o u n d a t i o n s f o r t h e i r s e n t e n c e l e v e l
representation.
Regarding discourse comprehension,
the structure-building
hypothesis predicts
that the first clause of a discourse should enjoy first mention advantage, and be more
accessible than subsequent clauses, so that incoming information may be mapped
o n t o it. A t t h e s a m e t i m e , t h i s s a m e h y p o t h e s i s p r e d i c t s t h a t a n y c l a u s e c u r r e n t l y
b e i n g p r o c e s s e d ( i n its o w n s u b s t r u c t u r e ) s h o u l d b e f o r e m o s t o n t h e c o m p r e h e n d e r ' s
m i n d . T h i s s e e m i n g l y c o n t r a d i c t o r y a d v a n t a g e o f f i r s t m e n t i o n as o p p o s e d t o r e c e n c y
must be resolved when the comprehender has processed the second clause. By then,
the first clause (of a two-clause sentence) should be more accessible so that the seco n d c l a u s e c o u l d b e i n c o r p o r a t e d w i t h it. G e r n s b a c h e r e t al. ( 1 9 8 9 ) i n d e e d s h o w e d
that while comprehending
the second clause, subjects had greatest access to the
i n f o r m a t i o n p r e s e n t e d in the s u b s t r u c t u r e t h e y w e r e c u r r e n t l y b u i l d i n g . W h e n a c c e s sibility was measured 150 ms after the offset of the sentences" final words, the two
clauses were equally accessible. However, when measured about a second or two
later (after 1 4 0 0 m s a n d 2 0 0 0 m s ) , the first c l a u s e w a s s h o w n to e n j o y g r e a t e r a c c e s sibility. According to the structure-building hypothesis, these two seemingly contrad i c t o r y p h e n o m e n a ( o f first m e n t i o n as o p p o s e d to r e c e n c y a d v a n t a g e ) s h o u l d n o t be
m u t u a l l y e x c l u s i v e . W h i l e r e c e n t c l a u s e a d v a n t a g e is e x p e c t e d to b e s h o r t - l i v e d , first
mention advantage should be long-lived.
Research
into discourse structure provides support for the structure-building
h y p o t h e s i s . L i n g u i s t i c m a r k e r s o f h i g h a c c e s s i b i l i t y (e.g., p r o n o u n s ) are u s e d to r e f e r
e i t h e r t o t h e r e c e n t N P ( i n d i c a t i n g t h a t t h e p r e v i o u s / r e c e n t c l a u s e is h i g h l y a c c e s s i ble), or to the NP constituent of the discourse-topic proposition (indicating that the
f i r s t c l a u s e in t h e d i s c o u r s e is h i g h l y a c c e s s i b l e ) , a s s h o w n b y A r i e l ( 1 9 9 0 ) . I n f o r m a t i o n in d i s c o u r s e - i n i t i a l p o s i t i o n p r o v i d e s t h e b e s t c u e s f o r r e c a l l i n g t h e d i s c o u r s e
a s a w h o l e ( e . g . , f i r s t w o r d s o f a s e n t e n c e o r p i c t u r e s o f t h o s e w o r d s , as t e s t i f i e d b y
Bock and Irwin, 1980; Prentice, 1967; Turner and Rommetveit,
1968, or the beginn i n g o f s t o r y e p i s o d e s , as t e s t i f i e d b y M a n d l e r a n d G o o d m a n , 1982). W h e n a s k e d to
recall the m a i n i d e a o f a d i s c o u r s e , c o m p r e h e n d e r s t e n d to s e l e c t the initial s e n t e n c e
(Kieras, 1980). Indeed, discourse-initial position has been shown to be preferably
preserved for the discourse-topic proposition. Dooling and Lachman (197 I) and Garrod and Sanford (1977) showed that providing information about the discourse-topic
at t h e b e g i n n i n g o f a d i s c o u r s e a c t i v a t e s t h e r e l a t e d s c h e m a o r g e n e r a l k n o w l e d g e
which allow for incoming inforlnation to be integrated more easily. Bransford and
J o h n s o n ( 1 9 7 2 ) s h o w e d t h a t in c e r t a i n c a s e s , c o m p r e h e n s i o n
o f t e x t s is i m p o s s i b l e
w i t h o u t a d i s c o u r s e - t o p i c m e n t i o n in i n i t i a l p o s i t i o n ( e . g . , in t h e title). G i o r a ( 1 9 8 5 b )
s h o w e d t h a t d i s c o u r s e s w i t h a d i s c o u r s e - t o p i c p r o p o s i t i o n in i n i t i a l p o s i t i o n a r e r e a d
s i g n i f i c a n t l y f a s t e r t h a n i d e n t i c a l d i s c o u r s e s w i t h d i s c o u r s e - t o p i c m e n t i o n in f i n a l
p o s i t i o n . G e o r g e e t al. ( 1 9 9 4 ) d e m o n s t r a t e d t h a t d i s c o u r s e c o m p r e h e n s i o n
is f a c i l i -
420
R. G i o r a / J o u r n a l
¢~]'Pragmati,'s 2 0 ( 1 9 9 6 ) 4 1 7 ~ 4 3 6
tated when the discourse-topic
enjoys initial mention. Subjects recalled titled discourses better than untitled ones, and they searched for the discourse-topic
proposition in the beginning of the text. They further found that words in the untitled paragraphs elicited greater N400 amplitude
than words in the titled paragraphs.
Recall
that N400, the component
of event-related
brain potential, is associated with diffic u l t y o f p r o c e s s i n g . H e r e it w a s s h o w n t o b e s e n s i t i v e t o t h e a b s e n c e o f a d i s c o u r s e topic mention in initial position (e.g., a title).
Studies into aphasia further attest that mention of a discourse-topic
proposition
in the beginning
of a discourse
facilitates comprehension.
For example,
Hough
(1990)
showed
that right hemisphere
damaged
patients
have more difficulties
than other patients
and normal
subjects
comprehending
narratives
whose
discourse-topic
sentences are shifted to the end of the narrative. Schneiderman
et al.
(1992) showed
that the presence
of a discourse-topic
proposition
facilitates text
comprehension
for left hemisphere
damaged
and non hemisphere
damaged
individuals.
Right hemisphere
damaged
patients
do not benefit from its presence.
Schneiderman
and her colleagues
interpret this deficit as stemming
from a more
general
impairment
in formulating
macrostructure.
There
is ample
evidence,
then, that information
in the beginning
of a discourse
tends to be functional
in
laying the foundation
for building the mental representation
of the discourse
as a
whole.
3. T h e p r o c e s s e s
3. I.
of mapping
and shifting
Mapping
According
to the structure-building
hypothesis, coherence
should facilitate mapping. Findings concerning four types of 'coherence'
(referential, temporal, locational
and causal) are cited in support of the claim. Below I will review the data as presented by Gemsbacher,
criticize the notions of 'coherence'
employed,
and propose
an alternative view:
(i) Referential
'coherence'
facilitates mapping:
Sentences that maintain a referent i a l l i n k ( e . g . , 1) a r e m o r e l i k e l y t h a n s e n t e n c e s w h i c h d o n o t r e f e r t o p r e v i o u s l y
mentioned entities (e.g., 2) to be mapped onto developing
structures:
(1)
(2)
We got some beet- out of the trunk. The beer was warm.
Andrew was especially fond of beer. The beer was warm.
However,
Gernsbacher
(1983: 379), suggests
(3) My
notes that counterexample
(3), taken from Johnson-Laird
t h a t a r e f e r e n t i a l l i n k is i n s u f f i c i e n t f o r c r e a t i n g c o h e r e n c e :
daughter
works in a library in London.
London
of natural history. The museum
is organized
theory ...
museum
is the home of a good
on the basis of cladistic
R. G i o r a / J o u r n a l
The problem
to one entity
(3.1)
is n o t a l l e v i a t e d
only (e.g., Ida's
even
o f P r a g m a t i c s 26 (1996) 4 1 7 - 4 3 6
421
when the discourse manifests referential
as shown in Giora, 1985b):
linking
husband,
This first time she was married her husband came from Montana. He
kind that when he was not alone he would look thoughtful. He was
that knew that in Montana
there are mountains
and mountains
have
them. He had not lived in Montana. He would leave Montana. He had
Ida and he was thoughtful.
(Taken from Ida by Gertrude Stein)
In fact, as I show below, referential "coherence'
is n o t e v e n
i n ( 4 ) e x h i b i t s n o r e f e r e n t i a l " c o h e r e n c e " y e t it i s c o h e r e n t :
necessary.
was the
the kind
snow on
to marry
The
passage
(4) Just how far have women risen in the film community?
According
to PM, who
w a s a t W o m a n in F i l m l u n c h e o n r e c e n t l y , it h a s a c t u a l l y b e e n a v e r y g o o d y e a r
for women. "Demi Moore got $5 million for her part in Indecent Proposal. Urea
Thurman
earned $4 million in Mad Dog and Glory. Just three years ago, in
P r e t t y W o m a n , J u l i a R o b e r t s g o t -- w h a t w a s i t ? $ 3 , 0 0 0 ? " ' W o m e n h a v e h a d r e a l
progress"
concluded
PM.
(ii) T e m p o r a l
"coherence'
facilitates mapping:
Sentences that manifest temporal
'coherence'
in that the events they report occur in the same time frame (e.g., 6 in the
context of 5) are more likely than sentences which are not temporally coherent (e.g.,
7 in the context of 5) to be mapped onto developing
structures. Data indeed support
t h e c l a i m : ( 6 ) is r e a d f a s t e r t h a n ( 7 ) :
(5)
(6)
(7)
a. I a r r i v e d a t t h e s t a r t l i n e a t 7 : 4 5 a . m .
b. I was the only female marathon
runner.
c. T h e m a r a t h o n w a s s c h e d u l e d to b e g i n at 8 : 0 0 .
d. As 1 nervously awaited the start, I talked with other runners.
e. I a l s o t r i e d to s t r e t c h a n d r e l a x .
f. A t e i g h t o ' c l o c k s h a r p t h e s t a r t e r f i r e d h i s p i s t o l . 2
H a l f a n h o u r l a t e r , it b e g a n t o r a i n .
T h r e e d a y s l a t e r , it b e g a n t o r a i n .
However,
temporal
"coherence',
I c l a i m , is n e i t h e r s u f f i c i e n t n o r n e c e s s a r y f o r d i s c o u r s e c o h e r e n c e . T h e s e n t e n c e i n ( 8 ) , w h i c h is t e m p o r a l l y i n c o h e r e n t , c o h e r e s w i t h
(5), and must be as easy to map onto (5) as (6), or at least easier to map than (7). (9)
on the other hand, which is temporally
c o h e r e n t , is i n c o h e r e n t , a n d m u s t b e at l e a s t
as difficult as (7) to map onto (5):
(8) A year later, two other women
2
Proposition
(b) has been added
showed
to the original_
up at the marathon
line. I felt much
better.
422
R. G i o r a / J o u r n a l
(9) Half
an hour
later, ! regretted
~f Pra,~matics 20 (1996) 417--436
not going
to the movies
the previous
night.
(iii) Locationai
'coherence'
facilitates mapping:
Sentences
which report events
occurring
at the same place (e.g., 11) are more likely to be mapped
than sentences
which are not locationally
coherent
(e.g., 12). Data support this prediction.
(11) is
read faster than (12) in the context of (10):
(10)
(11)
(12)
a.
b.
In
In
Mike and I were standing in the hallway near my office.
We were enthusiastically
discussing
some new data.
a nearby office, people had difficulty concentrating.
a nearby town, people had difficulty concentrating.
However,
locational
'coherence',
I claim, is neither
sufficient
discourse
coherence.
Consider
(13) which, though locationally
cohere
with (10), and (14), which
coheres
with (10) despite
'coherence'
:
(13)
(14)
In a nearby
In a nearby
doors.
office, a man was holding a receiver.
university/city,
people wouldn't
discuss
data
nor necessary
for
coherent,
does not
lack of locational
even
behind
closed
(iv) Causal
'coherence'
facilitates mapping:
Sentences
which are causally related
to previous events (e.g., 15e in the context of each of 15a--c) should be read faster
than sentences which are unlikely logical consequences
thereof (e.g., 15e in the context of 15d). Data support the structure-building
hypothesis
(cf. Keenan et al., 1984):
(15)
Context
sentence
a.
b.
c.
d.
Joey's big brother punched
him again and again.
Racing down the hill, Joey fell off his bike.
Joey's crazy mother became furiously angry with him.
Joey went to a neighbor's
house to play.
e.
The
Consequence
next
sentence
day, Joey
was
covered
But causality,
I argue, is not necessary
causally connected,
yet incoherent:
(16)
in bruises.
(cf. 4 above).
Nor
is it s u f f i c i e n t .
(16)
is
Dana hit Dan. He therefore cried. His shirt got wet as a result. So he went home
t o w a s h it. H i s m o t h e r w a s h o m e , s o h e h a d t o g o o n e r r a n d s . H e h a t e d h i s
mother. He was unhappy.
He cried a lot. His mother hated that. He left her. He
met Dina as a result. He fell in love with her. They got married
years later.
Dina had a baby. As a result, there was too much laundry. Now she was very
tired. She dreamt all day. Therefore
she wanted to have a career.
The conclusion
one is to draw from these counterexamples
is that referential,
temporal, locationai
and causal 'coherence'
do not satisfy the conditions
for discourse
R. G i o r a / J o u r n a l
423
o f Pragmati('s 26 (1996) 417--430
coherence.
As a result, they cannot
really support
the claim
that coherence
facilitates
mapping.
To be able to maintain
the claim
that coherence
facilitates
mapping,
a different, more
global
notion
of discourse
coherence
is required.
3.2.
Discourse
coherence
In Giora
(1985a,b,c,
1988, partly following
Grice,
1975) I proposed
to view
discourse
coherence
in terms
of Relevance
to a discourse-topic
and Graded
Informativeness.
Specifically,
an informative
discourse
is coherent
if and only if it
(a)
conforms
to the Relevance
Requirement
of as related/similar
to a discourse-topic
generalization,
preferably
made
explicit,
course.
It functions
as a reference
point
tions are assessed
and stored,
in that all its propositions
are conceived
proposition.
The
discourse-topic
is a
and placed
in the beginning
of the disrelative
to which
all incoming
proposi-
and
(b)
conforms
to the Granded
Informativeness
Condition
which
requires
that each
proposition
be more
(or at least not less) informative
than the one that precedes
it in relation
to the discourse-topic.
Along
the lines suggested
by e.g., Shannon
(1 9 5 1) a n d A t t n e a v e
(1959),
a message
is considered
informative
to the extent
that it has properties
unshared
by the previous
proposition,
which,
in tum, allow
it to reduce
possibilites
by half,
and
(c)
marks
marker
It has
easiest
any deviation
e.g., 'by the
from
way',
Relevance
and Graded
Informativeness
'after all" (cf. Ariel,
1985,
1988). 3
been
shown
that discourses
to process
(Giora,
1985b;
which
conform
to
Giora
et al., 1995). 4
the
above
by
conditions
an explicit
are
the
1 proposed (Giora, 1985b) that the conditions of discourse well-formedness are derived from more
g e n e r a l c o n s t r a i n t s o n s t o r a g e o f g e n e r a l k n o w l e d g e in m e m o r y . A c c o r d i n g l y , a w e l l - f o r m e d i n f o r m a t i v e
d i s c o u r s e is o r g a n i z e d l i k e a p r o t o t y p e - o r i e n t e d c a t e g o r y (e.g., t h e c l a s s o f b i r d s , cf. R o s c h , 1973). T h e
p r i n c i p l e o f o r g a n i z a t i o n is s i m i l a r i t y ( t o a p r o t o t y p i c a l m e m b e r ) . C o n s e q u e n t l y , t h e i n t e m a l s t r u c t u r e o f
t h e c a t e g o r y is g r a d e d . It is s t o r e d r e l a t i v e to t h e l e a s t i n f o r m a t i v e m e m b e r in t h e set. a n d is o r d e r e d
informatively from the least to the most informative member. Likewise. a well-formed informative disc o u r s e e v o l v e s g r a d u a l l y f r o m t h e l e a s t to t h e m o s t i n f o r m a t i v e m e s s a g e . It b e g i n s w i t h a g e n e r a l i z a t i o n
a n d b e c o m e s m o r e i n f o r m a t i v e as t h e d i s c o u r s e p r o c e e d s .
A n y d e v i a t i o n r e q u i r e s e x t r a p r o c e s s i n g a n d is j u d g e d b y s p e a k e r s as l e s s n a t u r a l ( ( S i o r a . 1988).
Along the lines suggested by Grice (1975), overt violations of the requirements are intended to be reco g n i z e d a n d t r i g g e r t h e g e n e r a t i o n o f i m p l i e a t u r e s . T h e y are a i m e d at a c h i e v i n g s p e c i a l e f f e c t s o r p r o d u c t s . a n d a r e m o r e d i f f i c u l t t o p r o c e s s ( G i o r a , 1 9 9 0 , 1 9 9 3 ) . In c o n t r a s t , u n i n t e n d e d v i o l a t i o n s c o n s t i t u t e
a n o m a l ies.
424
R. G i o r a / J o u r n a l
~/'Pragmati~'s 26 (1996) 417-436
The theory of discourse
coherence
I propose explains why the four types of discourse connectivity
(i i v ) c o n s i d e r e d
by Gernsbacher,
do not guarantee
coherence.
First, they do not block violation of the Relevance
Requirement,
since they allow for
two sentences to cohere via only one sentence constituent.
Given that the discourset o p i c is a p r o p o s i t i o n
(cf. Giora, 1985a, and (a) above), for a discourse to be coherent, both the argument
and the predicate constituents
of the discourse-topic
proposition should control the various propositions
of the discourse segment. Any reference
to just one constituent
of the discourse-topic
proposition,
b e it a n N P o r a P P , w i l l
n o t s u f f i c e ( c f . 3, 3 . 1 a b o v e ) .
The second problem
with the set of "coherence"
conditions
cited by Gernsbacher
has to do with their scope. Being constraints on adjacent sentences
only, referential,
temporal,
locational
and causal
"coherence"
fail to guarantee
relevance
to a discourse-topic.
At best, they constitute
local connectivity.
In Giora
(1985a,b,c)
I
showed
that coherence
is n o t a l i n e a r p r o p e r t y
of the discourse.
Rather, coherence
relations obtain between
the various propositions
of the discourse
and a superordihate proposition
(cf. 3 vs. 4 above). The counterexamples
I propose
above, which
achieve coherence
despite absence of cohesive
(referential, temporal,
locational and
causal) links (e.g., 4), are all subsumable
under a discourse-topic
proposition,
as
required by the Relevance
Condition
((a) above).
The structure-building
framework
has been verified only as far as adjacent
sentences are concerned.
However,
regarding mapping,
it s e e m s s a f e t o h y p o t h e s i z e
that
w h a t is t r u e o f t h e s e c o n d c l a u s e m u s t b e t r u e o f a l l s u b s e q u e n t
clauses:
The third
clause, for instance, must be most accessible
while being processed
in its substructure. A second or so later, though, the first clause must regain its superior accessibility. By then, the third clause, which
has been fully comprehended,
should be
mapped
onto the first clause, and so forth.
Given that comprehenders
build a mental structure of a discourse in relation to the
first clause, I propose
that the theory of discourse
coherence
I delineate
supports
Gernsbacher's
claim that coherence
facilitates mapping.
The processes of laying the
foundation
and of mapping
incoming
information
onto developing
structurcs constitute precisely
(part of) the processing
model predicted
by the theory of discourse
coherence
in question.
Put differently,
the findings
which
support
the structure-
The discourse in (4) a b o v e is an e x a m p l e o f a w e l l - f o r m e d text. It o b e y s the R e l e v a n c e R e q u i r e m e n t
in that it begins with a gelaeralization ( " i t has ... been a very g o o d year for w o m e n [in the film c o m m u n i t y ] " ) wh i ch presents the set of properties ( g o o d year, for w o m e n , in the film c o m m u n i t y ) shared by all
the propositions in the text. T h e various propositions exhibit instances o f w o m e n in the film c o m m u n i t y
w h o had a g o o d y e a r financially. T h e r e f o r e they can be included in the c a t e g o r y "'good year for w o m e n
in the film c o m m u n i t y " . In addition, the text c o n f o r m s with the G r a d e d i n f o r m a t i v e n e s s R e q u i r e m e n t .
E a c h proposition (apart form the last one) adds s o m e n ew information, e.g., m e n t i o n o f specific
actresses, m e n t i o n o f specific instances o f financial success, c o m p a r i s o n with past success. Th e last
pr opo s i t i o n repeats in f o r m a ti o u presented in the d i s c o u r s e - t o p i c proposition. R e p e a t e d m e n t i o n of the
d i s c o u r s e - t o p i c functions as a s e g m e n t a t i o n m a r k e r ( L o n g a c r e , 1979; Giora, 1983, 1986, f o r t h c o m i n g ) .
"'The Structure B u i l d i n g F r a m e w o r k explicitly p r o p o s e s that the c o g n i t i v e process o f laying a tbundation occurs w h e n e v e r c o m p r e h e n d e r s are building mental structures, regardless o f w h e t h e r the units
are clauses, sentences, story e p i s o d e s and regardless o f modality'" (p. 232).
R. Giora / Journal of Pragmati¢'s 20 (1990) 417~130
building
bacher's
3.3.
hypothesis
examples
also support the theory of discourse
all obey the Relevance
Requirement.
coherence
425
I propose:
Gerns-
Shifting
According
to the structure-building
framework,
when incoming
information
does
not cohere
with previous
information,
and cannot
be mapped
onto a developing
structure, comprehenders
shift and develop
a new substructure.
The structure-building framework
is supported
by theories of discourse
coherence
in this respect too.
Various
researches
looking
into how texts progress
from sentence
to sentence
suggest that the introduction
of new information
triggers text segmentation
(e.g., Dane~,
19 7 4 ; R e i n h a r t ,
1980).
Discourse
segmentation
at the level of discourse units larger than a sentence (e.g.,
paragraphs)
also falls out of the structure-building
hypothesis.
The conditions
of discourse coherence
I propose (cf. b above), require that discourse segments
(e.g., paragraphs)
should end at peaks of informativity.
This means that the boundaries
of a
discourse
segment
are determined
by the Relevance
Requirement.
An informative
message
that no longer bears relevance/similarity
to the topic of a given discourse
segment
signals segment boundary,
and starts a new segment. As required by condition (c) of the discourse
coherence
theory
I propose
above,
a new substructure
should
be explicitly
marked
by a digressive
marker,
e.g., paragraphing,
semantic
connectors,
adverbial phrases, etc. 6 Such markers signal digression
from Relevance. 7
However,
since the structure-building
framework
deals primarily
with relations
between
adjacent
sentences,
it d o e s n o t p r e d i c t d i s c o u r s e
segmentation
after the
introduction
of new information.
I have shown
(Giora,
1983a,b,
1986, 1988) that
segmenting
the text after introducing
the next discourse-topic
is a w i d e s p r e a d
phenomenon.
For example,
the future discourse-topic
of the second paragraph
o f (1 7 )
below (the chance discovery of penicillin) appears
in the final position of the first
segment.
It is s t i l l h i g h l y r e l e v a n t t o t h e d i s c o u r s e - t o p i c
under discussion
presented
in the first sentence of that paragraph
(The chance discovery of important scietltific
discoveries). However,
a highly informative
description
o f it, w h i c h i s i n t r o d u c e d
in
the beginning
of the next paragraph,
is beyond
the scope of the given discourse-
6 N o t e that a n ew d i s c o u r s e - t o p i c m a y drastically d i v e r g e from the d i s c o u r s e - t o p i c u n d er discussion, or
just m i l d l y so. Vario u s discoursal d e v i c e s indicate the e x t e n t to w h i c h a n e w l y introduced d i sco u r se
topic is new. U n l i k e the notion o f accessibility, w h o s e graded nature has been w i d e l y a c k n o w l e d g e d , the
terrain o f i n acces s i b i li ty hardly k n o w s any degree. T h e literature a b o u n d s in graded notions such as
A w a r e n e s s (Chafe, 1974, 1976), G i v e n n e s s (Prince, 1981b), A c c e s s i b i l i t y (Ariel, 1990--1991; G u n d e l et
al., 1993). It s eem s that a g r a d e d notion o f n e w n e s s w h i c h is a natural e x t e n s i o n o f the e x i s t i n g notions,
has still to await research_
7 Note, h o w e v e r , that these c o h e s i v e d e v i c e s do not function e x c l u s i v e l y as d i g r e s s i o n markers. Apart
from i n t r o d u c i n g n ew d is c o u r s e topics to the d i s c o u r s e , they have o t h e r functions. H e n c e the possibility
<~f m a n i p u l a t i n g the c o h e s i v e d e v i c e s ( t e r m e d by G e r n s b a c h e r t e m p o r a l , location, causal c o h e r e n c e ) for
the pu rp o s e o f a f f e c t i n g both c o h e r e n c e and i n c o h e r e n c e .
C o n s i d e r also the possibility o f e n d i n g a d i s c o u r se s e g m e n t by a c o d a ( " A n d t h a t ' s it". cf. Labov.
1972) or by r e p e a t e d m e n t i o n o f the d i s c o u r s e - t o p i c p r o p o s i t i o n w h i c h f u n ct i o n s as a coda too.
426
R. G i o r a / J o u r n a l e~f P r a g m a t i c s 2 6 ( 1 9 9 0 ) 4 1 7 ~ 1 3 0
topic. The theory of discourse
coherence
which I propose predicts that when a discourse-topic
shifts, comprehenders
should shift and develop a new structure. Howe v e r , it i s n o t c l e a r w h y , a c c o r d i n g
to the structure-building
framework,
comprehenders should
shift when
segmentation
occurs
alter the introduction
of the next
discourse-topic.
Under such circumstances,
the last clause of the given segment
is
coherent with the given discourse-topic
and with previous information,
and the first
clause of the next segment
is coherent with previous
information,
placed in the end
of the previous
segment.
Examples
such as (17) suggest that comprehenders
shift
and build a new substructure
when information
no longer coheres with the given discourse-topic
rather than with 'previous
information':
(17)
It has often occurred
in the history of science that an important
discovery
was
come upon by chance. A scientist looking into one matter, unexpectedly
came
upon another which was far more important
than the one he was looking into.
Penicillin
is a result
Penicillin
was
~['su¢'h
accidentally
a discovery.
discovered
by Fleming
in t928
...
The suggestion
that "previous information"
in the context of processing
a whole
discourse must mean "the given discourse-topic"
seems to follow from the structurebuilding framework.
The process of mapping
incoming
information
onto that clause
which constitutes
the foundation
involves
mapping
information
in the most recent
clause onto the first clause.
Why do speakers and writers segment a text after the introduction
of the next discourse-topic?
Given the structure-building
hypothesis,
I propose that they do so to
facilitate mapping:
By placing the next discourse-topic
at the end of a given paragraph, writers allow for the next paragraph
to be mapped
onto the previously
developed substructure.
Such concatenation
of new discourse-topics
must facilitate processing: It does not require the activation of new memory
cells for the foundation
of
a new mental structure.
How do readers identify new discourse-topics
when they are introduced
ill a p r e vious substructure?
In Giora
(1983a,b),
I suggested
that segment-final
position
enhances
r e c a l l , s i n c e it e n j o y s r e c e n c y e f f e c t . I n f o r m a t i o n
in discourse-final
posit i o n is f o r e g r o u n d e d
so as to be (as) highly accessible
(as first mentioned
information) for further discussion.
But according
to the structure-building
hypothesis,
this
explanation
cannot hold. Once this information
is p r o c e s s e d ,
it loses its superior
accessibility
(due to recency), and first mentioned
information
regains greater accessibility (of. The process of laying the foundation).
However,
information
in segmentfinal position must at least be more accessible than previously
presented information
(apart from first mentioned
information).
In Giora (1988) I offered an alternative
explanation
which agrees with the structure-building
hypothesis.
Along the lines suggested
by Gemsbacher,
I proposed
that
highly informative
messages
enjoy a privileged cognitive status due to informativity.
Informativity
affects conlprehenders's
ability to access information
(by the mechanism of enhancement
and suppression
to be discussed
later on). Since information
in
R. G i o r a / J o u r n a l o f P r a g m a t i c s 2 6 ( 1 9 9 6 ) 4 1 7 - - 4 3 6
427
the final position of a discourse segment must be highly informative, if not the most
i n f o r m a t i v e in t h a t s e g m e n t ( c f . b a b o v e ) , it m u s t a l s o b e e a s i e r t o a c c e s s , r e l a t i v e t o
previously mentioned information (first mentioned information not included). This
e x p l a i n s t h e s e g m e n t a t i o n in ( 1 7 ) a b o v e . T h e c h a n c e d i s c o v e r y o f p e n i c i l l i n , a s p e c i f i c i n s t a n c e o f scientOqc c h a n c e d i s c o v e r y o f s o m e i m p o r t a n c e , is i n t r o d u c e d b e f o r e
p a r a g r a p h i n g o c c u r s . Its i n f o r m a t i v i t y ( w h i c h s u p p r e s s e s the a c c e s s i b i l i t y o f p r e v i o u s l y m e n t i o n e d i n f o r m a t i o n , as w i l l b e d i s c u s s e d b e l o w ) a l l o w s it t o b e e a s y to
access and selected for further discussion.
S h i f t i n g , t h e n , o c c u r s w h e n n e w i n f o r m a t i o n , e . g . , a n e w d i s c o u r s e - t o p i c , is i n t r o d u c e d to the d i s c o u r s e . G i v e n the s t r u c t u r e - b u i l d i n g f r a m e w o r k , I p r o p o s e that s e g m e n t a t i o n s t r a t e g i e s v a r y in a c c o r d a n c e w i t h t h e d e g r e e o f ' n e w n e s s ' / ( i n ) a c c e s s i b i i ity o f i n c o m i n g i n f o r m a t i o n . W h e n i n c o m i n g i n f o r m a t i o n c o h e r e s w i t h the p r e v i o u s
d i s c o u r s e - t o p i c a n d r e q u i r e s little a c t i v a t i o n o f n e w m e m o r y ceils, text p r o g r e s s i o n
may proceed smoothly, with no explicit discourse connectors (cf. example
17
a b o v e ) . W h e n a n e w d i s c o u r s e - t o p i c is l e s s c o h e r e n t w i t h t h e p r e v i o u s d i s c o u r s e topic and requires activation of new memory cells, speakers/writers will cue comp r e h e n d e r s b y m e a n s o f e x p l i c i t l i n g u i s t i c c o n n e c t o r s (as r e q u i r e d b y (c) a b o v e a n d
in a c c o r d a n c e w i t h t h e s t r u c t u r e - b u i l d i n g h y p o t h e s i s ) .
4.
The
mechanisms
of suppression
and enhancement
M e n t a l r e p r e s e n t a t i o n s o f d i s c o u r s e c o n s t i t u e n t s in m e m o r y a r e e q u a l b u t h i e r a r c h i c a l . S o m e i n f o r m a t i o n e n j o y s e n h a n c e d r e c a l l , a n d s o m e is l e s s a c c e s s i b l e . F o r
instance, the time course of comprehension reveals that the advantage of first ment i o n is l o n g - l i v e d , w h e r e a s t h e a d v a n t a g e o f c l a u s e r e c e n c y is s h o r t - l i v e d . T o b u i l d
coherent mental structures, comprehenders
s h o u l d k e e p a c t i v a t e d i n f o r m a t i o n t h a t is
f u n c t i o n a l in d e v e l o p i n g t h e i r m e n t a l s t r u c t u r e s . A t t h e s a m e t i m e , t h e y s h o u l d s u p p r e s s i n f o r m a t i o n t h a t is n o l o n g e r f u n c t i o n a l in d e v e l o p i n g t h e s e s t r u c t u r e s . S i m i larly, coherent mental structures require that relevant meanings be activated and
r e t a i n e d , w h i l e i r r e l e v a n t i n f o r m a t i o n be s u p p r e s s e d a n d d a m p e n e d . A c c o r d i n g to the
s t r u c t u r e - b u i l d i n g f r a m e w o r k , it is t h e m e c h a n i s m s o f e n h a n c e m e n t a n d s u p p r e s s i o n
that are responsible for the activation and dampening of information.
How do the mechanisms of enhancement and suppression operate? Mental struct u r e s a r e r e p r e s e n t e d in m e m o r y c e l l s . M e m o r y c e l l s r e p r e s e n t p r e v i o u s l y s t o r e d
m e m o r y traces, a n d are a u t o m a t i c a l l y a c t i v a t e d b y i n c o m i n g stimuli. A c c o r d i n g to
the structure-building framework, activated memory cells transmit processing signals which either enhance or suppress the activation of other memory cells.
Gernsbacher capitalizes on the role the mechanisms of enhancement and suppress i o n p l a y in u n d e r s t a n d i n g a m b i g u o u s a n d u n a m b i g u o u s
w o r d s , in i m p r o v i n g t h e
a c c e s s i b i l i t y o f c o n c e p t s r e f e r r e d to b y a n a p h o r i c a n d c a t a p h o r i c d e v i c e s , a n d in
i m p r o v i n g m e m o r y f o r t h e m a t i c as o p p o s e d t o s u r f a c e l e v e l i n f o r m a t i o n .
According to many models of word understanding (e.g., Becker, 1976; Kintsch,
1988; Marslen-Wilson
and Welsh, 1978; Norris, 1986; McClelland and Rumeihart,
1981), immediately after comprehenders
hear or read an ambiguous word, multiple
428
R. Giora / Journal of'Pragmatit's 26 (1996) 417--436
m e a n i n g s a r e a c t i v a t e d . F o r e x a m p l e , r e s e a r c h b y S w i n n e y ( 1 9 7 9 ) s h o w e d t h a t in
understanding ambiguous words like bug (insect/listening device), both meanings
a r e a c t i v a t e d i n i t i a l l y . T h i s h o l d s e v e n if t h e c o n t e x t is b i a s e d in f a v o r o f o n e m e a n i n g , as in ( I 8), w h i c h f a v o r s t h e ' i n s e c t ' i n t e r p r e t a t i o n :
(18) The man was not surprised when he found several spiders, roaches, and other
b u g s ...
As sentence comprehension
progresses, the appropriate meaning remains activated, while the inappropriate meaning becomes less activated. According to some
theories (McClelland and Kawamoto,
1986; Waltz and Pollack, 1985), the inapprop r i a t e m e a n i n g s a r e i n h i b i t e d b y t h e a p p r o p r i a t e o n e s . T h e a s s u m p t i o n is t h a t c o n c e p t s c o m p e t e f o r a f i x e d a m o u n t o f a c t i v a t i o n . I n a p p r o p r i a t e m e a n i n g s d e c r e a s e in
activation because appropriate meanings have increased. Gemsbacher's
findings do
not support this assumption:
T h e a p p r o p r i a t e m e a n i n g s d o n o t i n c r e a s e in t h e i r
accessibility when the inappropriate meanings decrease.
A c c o r d i n g to o t h e r t h e o r i e s , i n a p p r o p r i a t e m e a n i n g s s i m p l y d e c a y , b e c a u s e t h e y
are not stimulated by the context (e.g., Anderson, 1983). These theories predict that
in a n e u t r a l c o n t e x t ( e . g . , 21 b e l o w ) , b o t h m e a n i n g s s h o u l d b e l e s s a c t i v a t e d w h e n
a c t i v a t i o n is m e a s u r e d a f t e r a d e l a y . H o w e v e r , if m u l t i p l e m e a n i n g s o f n e u t r a l s e n t e n c e s are j u s t as a c t i v a t e d w h e n m e a s u r e d i m m e d i a t e l y a n d after a d e l a y , t h e s e
results support the suppression hypothesis. According to the suppression hypothesis,
m u l t i p l e m e a n i n g s o f a n e u t r a l s e n t e n c e s h o u l d b e j u s t a s a c t i v a t e d a f t e r t h e d e l a y as
they are immediately, because there are no bases from which suppression signals can
b e t r a n s m i t t e d . T h i s is n o t t r u e o f i n a p p r o p r i a t e m e a n i n g s . A c c o r d i n g t o t h e s u p p r e s sion hypothesis, inappropriate meanings should become less activated via the mechanism of suppression. Memory cells representing the context transmit signals which
dampen the inappropriate meanings.
Gernsbacher
and Faust (1990, 1991) tested these hypotheses. They presented
a m b i g u o u s w o r d s l i k e q u a c k in d i f f e r e n t s e n t e n c e c o n t e x t s : in a c o n t e x t t h a t b i a s e d
o n e m e a n i n g ( d o c t o r , a s in 19), in a c o n t e x t t h a t b i a s e d a n o t h e r m e a n i n g ( d u c k , as in
2 0 ) , a n d in a n e u t r a l c o n t e x t w h i c h is e q u a l l y r e l a t e d t o e i t h e r m e a n i n g ( a s in 2 1 ) :
(19)
(20)
(21)
P a m w a s d i a g n o s e d b y a q u a c k ...
P a m h e a r d a s o u n d l i k e a q u a c k ...
P a m w a s a n n o y e d b y a q u a c k ...
They showed that the appropriate and neutral meanings remained equally activated
i m m e d i a t e l y a f t e r r e a d i n g e a c h o f t h e a b o v e s e n t e n c e s , a n d at a d e l a y e d i n t e r v a l . H o w ever, the contextually inappropriate meanings remained equally activated only immediately. At a delayed interval (of about 1 second), the inappropriate meaning was less
activated. These results support the suppression hypothesis, according to which only
memory cells representing the inappropriate meaning should be suppressed.
The suppression mechanism may contribute to testing the difference between vari o u s f i g u r e s o f s p e e c h . F o r i n s t a n c e , in G i o r a ( 1 9 9 1 , 1 9 9 5 ) 1 s u g g e s t t h a t , a m o n g
429
R. G i o r a / , l o u r n a l o f P r a g m a t i c s 2 6 ( 1 9 9 6 ) 4 1 7 - - 4 3 6
o t h e r t h i n g s , j o k e s d i f f e r f r o m o t h e r f i g u r e s o f s p e e c h ( e . g . , i r o n y , m e t a p h o r ) in t h a t
their understanding involves suppressing the most probable (the so-called appropriate) m e a n i n g . Its s u p p r e s s i o n a l l o w s o n l y f o r the a t y p i c a l m e a n i n g o f the a m b i g u o u s
utterance to be retained. Indeed, Gernsbacher
and Robertson (1995) suggest that
understanding puns and jokes relies on comprehenders'
ability to quickly suppress
the prototypical meaning of such utterances. They show that lessskilled readers took
l o n g e r t o u n d e r s t a n d p u n s t h a n m o r e s k i l l e d r e a d e r s , b e c a u s e t h e y w e r e s l o w e r at
s u p p r e s s i n g t h e t y p i c a l m e a n i n g ( e . g . t h e t y p i c a l m e a n i n g o f ' b a r " in T w o m e n w a l k
into a bat', and
a third
man
ducks).
The suppression mechanism
is a l s o f u n c t i o n a l in d a m p e n i n g i r r e l e v a n t a s s o c i a tions. Nonambiguous
words may have multiple associations, some of which might
be irrelevant. For instance, the association between the word apple and tree may be
m o r e r e l e v a n t in t h e c o n t e x t o f ( 2 3 ) t h a n in t h e c o n t e x t o f ( 2 2 ) , in w h i c h p i e w i l l b e
more relevant :
(22)
(23)
Jim baked the apples.
Jim picked the apples.
Gernsbacher and Faust (1990) show that whereas multiple associations are activated immediately upon comprehending
the sentence, only the relevant associations
remain activated after a delay. The less relevant associations are suppressed after a
brief period.
T h e c l a i m t h a t m e n t a l r e p r e s e n t a t i o n s o f d i s c o u r s e c o n s t i t u e n t s in m e m o r y a r e
hierarchical holds for representations of referents as well. According to the structurebuilding framework, for comprehenders
t o b e a b l e t o a c c e s s p r o n o u n s ' r e f e r e n t s , at
some point following the mention of the pronouns, antecedents and nonantecedents
m u s t b e a c t i v a t e d a t d i f f e r e n t l e v e l s . F o r i n s t a n c e , in ( 2 4 ) , i n w h i c h P a r e is t h e
antecedent of the anaphor she, Pam should be more activated than the nonantecedent
Ann, after comprehenders
finish reading the sentence. Findings indeed support this
prediction (Corbett and Chang, 1983):
(24)
Ann predicted
easily.
that Pam
would
lose the track race, but she came
in f i r s t v e r y
A c c o r d i n g t o t h e s t r u c t u r e b u i l d i n g f r a m e w o r k , t h e d i f f e r e n c e in l e v e l s o f a c t i v a tions between antecedents and nonantecedents
s h o u l d i n c r e a s e r e l a t i v e to the i n f o r m a t i v e n e s s o r e x p l i c i t n e s s o f t h e a n a p h o r , i.e., t h e r e f e r r i n g e x p r e s s i o n . T h e e x p l i c i t n e s s o f t h e a n a p h o r s h o u l d e n h a n c e t h e a c c e s s i b i l i t y o f its o w n r e f e r e n t w h i l e
s u p p r e s s i n g o t h e r s . F i n d i n g s i n d e e d s h o w t h a t e x p l i c i t a n a p h o r s l i k e P a r e in ( 2 5 )
i n c r e a s e the level o f a c t i v a t i o n o f their r e f e r e n t s as o p p o s e d to n o n r e f e r e n t s , c o m p a r e d to less e x p l i c i t a n a p h o r s like she in (24) ( C o r b e t t a n d C h a n g , 1 9 8 3 ; G e r n s bacher, 1989):
(25)
Ann predicted
easily.
that Pam
would
lose the track race, but Pam
came
in f i r s t v e r y
430
R . G i o r a / , l o u r n a l ~)f P r a g m a t i c s 26 ( 1 9 9 6 ) 4 1 7 - - 4 3 6
Furthermore,
findings show that new concepts improve their referential accessibility by triggering suppression
of previously
mentioned
concepts (Gernsbacher,
1989; Dell et al., 1983). Findings from the various studies support the Explicitness
Principle :
+'The more explicit the concept, the more likely it is to trigger the suppression of other concepts; when
used anaphorically, the more likely it is to enhance its referent.'" (p. 133)
Explicit anaphors are used to retrieve referents which are relatively hard to access
( s e e a l s o G i v 6 n , 1 9 8 3 ; A r i e l , 1 9 9 0 , 1 9 9 1 ) . G e r n s b a c h e r m e n t i o n s t h r e e c o n d i t i o n s in
which referents are hard for comprehenders
to retrieve. First, she discusses referent i a l d i s t a n c e . A t l o n g r e f e r e n t i a l d i s t a n c e s , r e f e r e n t s a r e h a r d e r t o a c c e s s . T h e r e is
a m p l e e v i d e n c e in t h e l i t e r a t u r e a s s o c i a t i n g l o w a c c e s s i b i l i t y a n d d i s t a n c e . H o w e v e r ,
Gemsbacher
s u g g e s t s t h a t it is n o t d i s t a n c e , b u t r a t h e r t h e i n t e r v e n t i o n o f t h e o t h e r
concepts which trigger suppression, and affect distant referents" accessibility. When
t h e d i s t a n c e is n o t f i l l e d b y i n t e r v e n i n g c o n c e p t s , d i s t a n t r e f e r e n t s ' a c c e s s i b i l i t y is
not affected (e.g., Carroll and Slowiaczek,
1987).
Second, she addresses the question of topicality. Topical concepts are easy to
access, because they are enhanced by both the advantage of first mention (cf. The
process of laying the foundation) and by frequent mention, which make them resistant to suppression. Less topical concepts are more difficult to access (e.g., Anders o n e t a l . , 1 9 8 3 ) , a n d r e q u i r e e x p l i c i t a n a p h o r s f o r t h e i r r e t r i e v a l (e+g., C h a f e , 1 9 7 4 ,
1976; Marslen-Wilson
et al., 1982; Giv6n 1983; Ariel, 1985, 1990, 1991).
T h e t h i r d c o n d i t i o n u n d e r w h i c h e x p l i c i t a n a p h o r s a r e u s e d is a t e p i s o d e b o u n d aries (see Marslen-Wilson
e l al+, 1 9 8 2 ; F o x , 1 9 8 6 ; T o m l i n , 1 9 8 7 ; H o f m a n n ,
1989;
Ariel, 1990, 1991). Beginnings of episodes and paragraphs are where new discoursetopics are introduced. Introduction of a new discourse-topic
involves shifting (el.
Shifting) and triggers suppression
of previous information. Previous information,
t h e r e f o r e , is l e s s a c c e s s i b l e in s e g m e n t - i n i t i a l p o s i t i o n , w h i c h e x p l a i n s w h y a n a p h o r s
are expected to be more explicit at the beginning of a paragraph or a new episode.
H o w e v e r , it h a s b e e n o b s e r v e d ( G i o r a , 1 9 8 3 a , b , 1 9 8 6 ; G i o r a a n d L e e , 1 9 9 6 ) t h a t
a n a p h o r s a l s o t e n d t o b e m o r e e x p l i c i t in s e g m e n t - f i n a l
position. One explanation
f o l l o w s f r o m t h e t h e o r y o f d i s c o u r s e c o h e r e n c e I p r o p o s e . S e g m e n t - f i n a l p o s i t i o n is
the preferred position for the most informative message in the discourse. As mentioned above, this information may function cataphorically to introduce future discourse-topics. Hence their explicitness: Their enhanced cognitive status makes them
highly activated and easier for comprehenders
to access while they shift and build
t h e n e x t s u b s t r u c t u r e . I n a d d i t i o n , it h a s b e e n p r o p o s e d ( e . g . , L o n g a c r e , 1 9 7 9 ; G i o r a
and Lee, 1996) that explicit anaphors signal segmentation.
Their explicitness and
enhancement
do not necessarily follow from the relative accessibility of their referents. Nor are they a function of their future role in the discourse. Rather, in segmentfinal position, explicit anaphors serve as a segmentation cue.
The ~nechanisms of enhancement
a n d s u p p r e s s i o n p l a y a r o l e in i m p r o v i n g t h e
accessibility of concepts marked by cataphoric devices. Cataphoric devices enable
comprehenders
to access concepts to be subsequently mentioned. Gernsbacher
and
R. G i o r a / J o u r n a l o f Pragrncttics 26 ( 1 9 9 0 ) 4 1 7 - 4 3 6
431
Shroyer (1989) show that the unstressed indefinite this article, which introduces new
concepts to the discourse (Prince, 1981a), makes them more highly activated than
c o n c e p t s i n t r o d u c e d v i a " a ' o r ' a n ' . T h e y a r e b e t t e r at s u p p r e s s i n g t h e a c t i v a t i o n o f
other concepts, and they better resist suppression by other concepts (Gernsbacher
and Jescheniak, 1995).
The mechanisms
of enhancement
and suppression
p l a y a r o l e in i m p r o v i n g
memory for thematic versus surface information. As the text progresses, surface
i n f o r m a t i o n ( e . g . , s y n t a c t i c f o r m ) is a l w a y s c h a n g i n g . A s a r e s u l t , n e w s u r f a c e i n f o r m a t i o n s u p p r e s s e s o l d s u r f a c e i n f o r m a t i o n . I n c o n t r a s t , t h e m a t i c i n f o r m a t i o n is c o n s t a n t l y b e i n g r e i n t r o d u c e d , a n d is t h e r e f o r e r e p e a t e d l y e n h a n c e d . T h e r e s u l t is t h a t
t h e m a t i c i n f o r m a t i o n is m o r e a c t i v a t e d t h a n s u r f a c e i n f o r m a t i o n , a n d is t h e r e f o r e
more accessible.
5.
Individual
differences
in structure
building
The general cognitive processes and mechanisms
underlying the structure-building framework also help explain individual differences. For instance, less skilled
comprehenders
are less able t h a n m o r e s k i l l e d c o m p r e h e n d e r s to r e m e m b e r r e c e n t l y
comprehended
information. According to the structure-building
framework,
they
shift too often.
T h e d i f f e r e n c e s in a p p l y i n g t h e m e c h a n i s m s o f s u p p r e s s i o n c a n e x p l a i n w h y l e s s
skilled comprehenders
are less able to reject contextually inappropriate meanings,
incorrect forms of homophones,
typical but absent members of nonverbal scenes,
and why they ignore words written on pictures, and pictures surrounding words.
However, less skilled comprehenders
are not less able to appreciate contextually
appropriate meanings
of ambiguous
words, and typical members
of nonverbal
scenes. These findings suggest that less-skilled comprehenders
have deficient suppression rather than enhancement mechanisms.
The extent to which comprehenders
u s e g e n e r a l c o g n i t i v e p r o c e s s e s in b u i l d i n g
mental representations while comprehending
a discourse may further distinguish
between skilled and less-skilled comprehenders.
Research into general cognitive
p r o c e s s e s a n d t e x t - s p e c i f i c s t r a t e g i e s in c o m p r e h e n s i o n
( G i o r a e t al., 1 9 9 6 ) r e v e a l s
that less-skilled readers are more likely than more skilled readers to use general cognitive processes. Skilled comprehenders,
on the other hand, rely more heavily on
text-specific strategies. The text-specific strategy we studied relies on coherence
structure. When trying to comprehend a text, the skilled reader processes deeply only
t h e f i r s t p r o p o s i t i o n o f t h e p a s s a g e , a n d t h e n s k i m s t h r o u g h t h e r e s t o f it f o r d i s c o n f i r m a t i o n s . T h i s i m p l i e s little or n o e f f e c t o f the c o n t e n t s o f the r e m a i n d e r o f the p a s sage on her comprehension. The less-skilled reader, on the other hand, employs the
general comprehension
strategy. She forms a representation of the whole text based
o n e a c h a n d e v e r y p r o p o s i t i o n . T h e f i r s t p r o p o s i t i o n is still t h e m o s t i m p o r t a n t o n e ,
s i n c e it l a y s t h e f o u n d a t i o n s f o r t h e r e p r e s e n t a t i o n (cf. T h e p r o c e s s o f l a y i n g t h e
foundation above). However, unlike the skilled reader, the less-skilled reader does
not skim through the remaining propositions but reads them carefully.
432
R_ Gior¢l / .loto-tlcH ¢~] P r a g r t l a t i ¢ ' s 2 6 ( 1 9 9 6 ) 4 1 7 ~ 4 3 6
These strategies explain the difference we found between skilled and less-skilled
c o m p r e h e n d e r s . W h e n t h e t e x t w a s c o h e r e n t , i.e., m a n i f e s t i n g d i s c o u r s e - t o p i c m e n t i o n in i n i t i a l p o s i t i o n , s k i l l e d r e a d e r s w e r e m o r e l i k e l y t h a n l e s s - s k i l l e d r e a d e r s t o
identify the discourse-topic proposition. However, under the same condition, lessskilled rather than skilled readers benefitted from analogy. Less-skilled readers, who
employ general-comprehension
strategies, may benefit from each and every proposition they integrate into their mental
representations,
analogical
comparisons
included. However, skilled readers who employ text-specific strategies and process
d e e p l y o n l y the first p r o p o s i t i o n , c a n n o t b e n e f i t f r o m a n a l o g i c a l c o m p a r i s o n s o c c u r r i n g l a t e r in t h e t e x t .
E s p e c i a l l y s t r i k i n g is t h e f i n d i n g t h a t s p e l l s o u t t h e c o s t o f a s k i l l : t h e m o s t s k i l l e d
and academically advanced readers performed worse even than the least skilled and
academically advanced readers, when their strategy was no longer functional. When
w e m a n i p u l a t e d text c o h e r e n c e b y d i s p l a c i n g the first p r o p o s i t i o n , l e s s - s k i l l e d r e a d e r s o u t p e r f o r m e d s k i l l e d r e a d e r s b y f a r . T h i s m a n i p u l a t i o n is c o n c e p t u a l l y s i m i l a r t o
t h a t u s e d b y G e r n s b a c h e r e t al. ( 1 9 9 0 ) , w h o c o m p a r e d s u b j e c t s " m e m o r y f o r s c r a m bled/incoherent and unscrambled/coherent
materials. Gemsbacher
e t al. f o u n d t h a t
skilled comprehenders
outperformed
less-skilled comprehenders
on unscrambled
texts. However, the two groups performed similarly on the scrambled texts. According to the structure-building framework, comprehenders
use the first proposition to
lay the foundations
for the representation
structure. Incoming
propositions are
m a p p e d o n t o t h a t s t r u c t u r e . H o w e v e r , w h e n m a p p i n g is i m p o s s i b l e , s u b j e c t s s h i f t t o
a new structure. Shifts involve poorer access to recently comprehended
information,
w h i c h r e s u l t s in p o o r c o m p r e h e n s i o n . A c c o r d i n g t o G e r n s b a c h e r a n d h e r a s s o c i a t e s ,
s k i l l e d a n d l e s s - s k l i l l e d c o m p r e h e n d e r s d i f f e r in t h e i r s h i f t i n g l i k e l i h o o d s . H o w e v e r ,
when these likelihoods are similar, differences among subjects disappear. Discourse
scrambling increases the likelihood of shifting and therefore eliminates the skilled
comprehenders" superiority.
However,
Gernsbacher's
structure-building
framework
does not predict lessskilled comprehenders'
superiority under incoherence condition. In contrast, the use
of structure-coherence
strategy predicts such superiority. Furthermore, the use of
s t r u c t u r e - c o h e r e n c e s t r a t e g y p r e d i c t s skill b u t n o t g e n e r a l c o m p r e h e n s i o n a b i l i t y diff e r e n c e s , as f o u n d . T h e s e a n d o t h e r a s p e c t s o f t h e d a t a , s u c h a s a n a l o g y e f f e c t s , s u g gest that skilled readers do not identify discourse-topics by using general structurebuilding processes.
In sum, the structure-building framework can account for a great number of linguistic and nonlinguistic phenolnena. The process of laying the foundation explains
w h y c o m p r e h e n d e r s take l o n g to p r o c e s s the first s e g m e n t o f a ( v e r b a l or n o n v e r b a l )
d i s c o u r s e s e q u e n c e ; it e x p l a i n s t h e a d v a n t a g e o f f i r s t m e n t i o n w h i c h e n h a n c e s t h e
accessibility of this segment. The process of mapping accounts for discourse coherence, and the process of shifting explains why and how comprehenders
shift and
start a new substructure. The process of shifting accounts for comprehenders" poor
menlory for the exact form of recently comprehended
information. The mechanisms
o f s u p p r e s s i o n a n d e n h a n c e m e n t e x p l a i n h o w i r r e l e v a n t i n f o r m a t i o n t h a t is n o t f u n c t i o n a l in s t r u c t u r e b u i l d i n g is d a m p e n e d , w h i l e i n f o r m a t i o n i n v o l v e d in s t r u c t u r e
R. G i o r a / J o u r n a l o f P r a g m a t i c s 2 6 ( 1 9 9 6 ) 4 1 7 - - 4 3 6
433
b u i l d i n g is e n h a n c e d . T h e y f u r t h e r e x p l a i n h o w s k i l l e d a n d l e s s s k i l l e d c o m p r e h e n d e r s d i f f e r in t h e i r a b i l i t i e s . G e r n s b a c h e r ' s
Language
comprehension
as structure
b u i l d i n g is a b o o k l i n g u i s t s a n d d i s c o u r s e a n a l y s t s h a v e l o n g b e e n w a i t i n g f o r .
References
Aaronson, Doris and Steven Ferres, 1983. Lexical categories and reading tasks. Journal of Experimental
Psychology. Human Perception and Performance
9: 675~99.
Aaronson,
D o r i s a n d H o l l i s S. S c a r b o r o u g h ,
1976. Performance
theories for sentence coding: Some
qualitative evidence. Journal of Experimental
Psychology.
Human Perception and Performance
2:
56--70.
A n d e r s o n , A n t h o n y , S i m o n C . G a r r o d a n d A n t h o n y J. S a n f o r d , 1 9 8 3 . T h e a c c e s s i b i l i t y o f p r o n o m i n a l
a n t e c e d e n t s a s a f u n c t i o n o f e p i s o d e s h i f t s in n a r r a t i v e t e x t . Q u a r t e r l y J o u r n a l o f E x p e r i m e n t a l
Psychology 35A: 427--440.
A n d e r s o n , J. R o b e r t , 1 9 8 3 . T h e a r c h i t e c t u r e o f c o g n i t i o n . C a m b r i d g e ,
MA: Harvard University Press,
Ariel, Mira, 1985. The discourse function of given information. Theoretical IAnguistics 12(2/3): 99--113.
Ariel, Mira, 1990. Accessing NP antecedents. London: Routledge.
Ariel, Mira, 1991. The function of accessibility in a theory of grammar.
Journal of Pragmatics
16:
443--463.
Attneave, Fred, 1959. Applications of information theory of psychology. New York: Holt, Rinehart and
Winston.
Becker, Curtis A., 1976. Semantic context and word frequency effects in visual word recognition. Journal of Experinlental Psychology.
Human Perception and Performance
2: 556--566.
B o c k , J. K a t h r y n a n d D a v i d E . I r w i n , 1 9 8 0 . S y n t a c t i c e f f e c t s o f i n f o r m a t i o n a v a i l a b i l i t y in s e n t e n c e
production. Journal of Verbal Learning and Verbal Behavior 19: 467~,84.
Bransford, John D. and Marcia K. Johnson,
1972. Contextual
prerequisites for understanding:
Some
investigations
of comprehension
and recall. Journal of Verbal Learning and Verbal Behavior
1 I:
7 ! 7--726.
Cairns, Helen S. and John Kameraa,
1975. Lexical information processing during sentence comprehension. Journal of Verbal Learning and Verbal Behavior 14: 170--179.
C a r r o l l , P a t r i c k J. a n d M a r i a L . S l o w i a c z e k ,
1987. Modes and modules:
Multiple pathways
to the
language
processor.
In: J.L. Garfield, ed., Modularity
in k n o w l e d g e
representation
and natural
language-understanding,
221-248. Cambridge, MA: MIT Press.
Chafe, Wallace L., 1974. Language and consciousness.
Language 50:11--t33.
Chafe, Wallace L., 1976. Givenness, contrastiveness,
definiteness, subjects, topics, and point of view.
In: Charles N. Li, ed., Subject and topic, 25--56. New York: Academic Press.
Chang, Frederick R., 1980. Active memory processes in visual sentence comprehension:
Clause effects
a n d p r o n o m i n a l r e f e r e n c e . M e m o r y & C o g n i t i o n 8: 5 8 - - 6 4 .
Corbett,
Albert T. and Frederick
R. Chang,
1983. Pronoun
disambiguation:
Accessing
potential
antecedents. Memory & Cognition I 1 : 283--294.
C u t l e r , A n n a n d D o n a l d J. F o s s , 1 9 7 7 . O n t h e r o l e o f s e n t e n c e s t r e s s in s e n t e n c e p r o c e s s i n g . L a n g u a g e
and Speech 21:1
10.
Daneg, Franti.~ek, 1974. Functions of sentence perspective and the organization of the text. In: Frantigek
Dane~, ed., Papers on functional sentence perspective, 106-128. The Hague: Mouton.
Dell, Gary S., Gail McKoon and Roger Ratcliff, 1983. The activation of antecedent information during
the processing of anaphoric reference in reading. Journal of Verbal Learning and Verbal Behavior 22:
121--132.
Dooling, D. James and Roy Lachman,
1971. Effects of comprehension
on retention of prose. Journal of
Experimental
Psychology
88:21 6-222.
F o s s , D o n a l d J., 1 9 6 9 . D e c i s i o n p r o c e s s i n g d u r i n g s e n t e n c e c o m p r e h e n s i o n :
Effects of lexical item difficulty and position upon decision times. Joumal of Verbal Learning and Verbal Behavior 8 : 457-462.
F o s s , D o n a l d J., 1 9 8 2 . A d i s c o u r s e o n s e m a n t i c p r i m i n g . C o g n i t i v e P s y c h o l o g y
14: 5 9 0 - - 6 0 7 .
434
R. G i o r a / . l o u r n a l o f P r o g m a t i c s
2 0 ( 1 9 9 0 ) 41 7 ~ 4 3 6
F o s s , D o n a l d J. a n d R i c h a r d H . Jr. L y n c h , 1 9 6 9 . D e c i s i o n p r o c e s s i n g d u r i n g s e n t e n c e c o m p r e h e n s i o n :
E f f e c t s o f s u r f a c e s t r u c t u r e o n d e c i s i o n t i m e s . P e r c e p t i o n a n d P s y c h o p h y s i c s 5: 1 4 5 - - 1 4 8 .
F o x , B a r b a r a A . , 1 9 8 6 . L o c a l p a t t e r n s a n d g e n e r a l p r i n c i p l e s in c o g n i t i v e p r o c e s s e s : A n a p h o r a in
w r i t t e n a n d c o n v e r s a t i o n a l E n g l i s h . T e x t 6: 2 5 - 5 1 .
Garrod, Simon and Anthony Sanford. 1977. Interp re ting a na phoric re la tions : T he inte gra tion of s e m a n
tic i n f o r l n a t i o n w h i l e r e a d i n g . J o u r n a l o f V e r b a l L e a r n i n g a n d V e r b a l B e h a v i o r 1 6 : 7 7
90.
G e o r g e , M a r i e S t . , S u z a n n e M a n n e s a n d J a m e s E. H o f f m a n , 1 9 9 4 . G l o b a l s e m a n t i c e x p e c t a n c y a n d l a n g u a g e c o m p r e h e n s i o n . J o u r n a l o f C o g n i t i v e N e u r o s c i e n c e 6( 1 ) : 7 0 - - 8 3 .
G e r n s b a c h e r , M o r t o n A n n , 1 9 8 3 . M e m o r y f o r t h e o r i e n t a t i o n o f p i c t u r e s in n o n v e r b a l s t o r i e s : P a r a l l e l s
and insights into language processing. Unpublished doctoral dissertation, Austin, TX: University of
T e x a s at A u s t i n .
G e r n s b a c h e r , M o r t o n A n n , I 9 8 5 . S u r f a c e i n t - o r m a t i o n l o s s m c o m p r e h e n s i o n . C o g n i t i v e P s y c h o l o g y 17:
324-363.
G e m s b a c h e r , M o r t o n A n n , 1989. M e c h a n i s m s that i m p r o v e r e f e r e n t i a l a c c e s s . C o g n i t i o n 3 2 : 9 9
156.
G e m s b a c h e r , M o r t o n A n n a n d M a r k E. F a u s t . 199(t. T h e r o l e o f s u p p r e s s i o n in s e n t e n c e c o m p r e h e n s i o n .
I n : G . B . S i m p s o n , e d . , U n d e r s t a n d i n g w o r d a n d s e n t e n c e , 9 7 128. A m s t e r d a m : N o r t h - H o l l a n d .
G e r n s b a c h e r , M o r t o n A n n a n d M a r k E. F a u s t , 1 9 9 1 . T h e m e c h a n i s m o f s u p p r e s s i o n : A c o m p o n e n t o f
g e n e r a l c o m p r e h e n s i o n skill. J o u r n a l o f E x p e r i m e n t a l P s y c h o l o g y : L e a r n i n g M e m o r y a n d C o g n i t i o n
17(21: 2 4 5 - - 2 6 2 .
G e r n s b a c h e r , M o r t o n A n n a n d D a v i d J. H a r g r e a v e s , 1 9 8 8 . A c c e s s i n g s e n t e n c e p a r t i c i p a n t s : T h e a d v a n tage o f first m e n t ion. Journal of Memory and L a ngua ge 27:699
717.
G e r n s b a c h e r , M o r t o n A n n , D a v i d J. H a r g r e a v e s a n d M a r k B e e m a n , 1 9 8 9 . B u i l d i n g a n d a c c e s s i n g c l a u s a l
representations: The advantage of first mention versus the advantage of clause recency. Journal of
Memory and Language 28: 735--755.
G e r n s b a c h e r , M o r t o n A n n a n d J o r g - D i e t e r J e s c h e n i a k , 1 9 9 5 . C a t a p h o r i c d e v i c e s in s p o k e n d i s c o u r s e .
C o g n i t i v e P s y c h o l o g y 29: 24--58.
Gemsbacher,
M o r t o n A n n a n d S u z a n n e S h r o y e r , 1989. T h e c a t a p h o r i c u s e o f t h e i n d e f i n i t e t h i s in
s p o k e n n a r r a t i v e s . M e m o r y a n d C o g n i t i o n 17: 5 3 6 - - 5 4 0 .
G e r n s b a c h e r , M o r t o n A n n , K a t h l e e n R. V a r n e r a n d M a r k E. F a u s t , 1 9 9 0 . I n v e s t i g a t i n g d i f f e r e n c e s in
g e n e r a l c o m p r e h e n s i o n skill. J o u r n a l o f E x p e r i m e n t a l P s y c h o l o g y : L e a r n i n g , M e m o r y a n d C o g n i t i o n
16: 4 3 9 ~ ¢ 4 5 .
Gemsbacher,
M o r t o n A n n a n d R a c h e l W . R o b e r t s o n , 1995. R e a d i n g s k i l l a n d s u p p r e s s i o n r e v i s i t e d .
P s y c h o l o g i c a l S c i e n c e 6: 1 6 5 - - 1 6 9 .
G i o r a , R a c h e l , 1983a. S e g m e n t a t i o n a n d s e g l n e n t c o h e s i o n : O n the t h e m a t i c o r g a n i z a t i o n o f the text.
Text 3(21:155
182.
G i o r a , R a c h e l , 1 9 8 3 b . F u n c t i o n a l p a r a g r a p h p e r s p e c t i v e . I n : J a n o s S. P e t 0 f i a n d E m e l S ( i z e r , e d s . , M i c r o
and macro connexity of texts, 153--182. Hamburg: Buske.
Giora, Rachel, 1985a. Towards a theory of coherence. Poetics Today 6(4): 699--716.
G i o r a , R a c h e l , 1985b. A text b a s e d a n a l y s i s o f n o n - n a r r a t i v e text. T h e o r e t i c a l L i n g u i s t i c s 12(2/31
115--135.
O i o r a , R a c h e l , 1 9 8 5 c . W h a t ' s a c o h e r e n t text'? I n : E m e i S~3zer, e d . , T e x t c o n n e x i t y , t e x t c o h e r e n c e :
A s p e c t s , m e t h o d s , r e s u l t s , 16--35. H a m b u r g : B u s k e _
G i o r a , R a c h e l , 1986. P r i n c i p l e s o f text s e g m e n t a t i o n : T h e c a s e o f the f o r m a l l y u n s e g m e n t e d text. J o u r n a l o f L i t e r a r y S t u d i e s 2 ( 2 ) : 15--28. I R c p r i n t e d in H e b r e w L i n g u i s t i c s 1 9 9 0 : 2 8 - - 3 0 . S p e c i a l i s s u e o n
Stylistics and the literary text.[
G i o r a , R a c h e l , 1988. O n the i n f o r m a t i v e n e s s r e q u i r e m e n t . J o u r n a l o f P r a g m a t i c s 12(5/61: 547 565.
[ R e p r i n t e d in A s a K a s h e r , e d . , 1 9 8 9 . C o g n i t i v e a s p e c t s o f l a n g u a g e u s e , 6 3 - - 9 6 . A m s t e r d a m :
NorthHolland. 1
G i o r a . R a c h e l . I 9 9 0 . O n t h e s o - c a l l e d e v a l u a t i v e m a t e r i a l in i n f o r m a t i v e t e x t s . T e x t 10(41: 2 9 9 - - 3 2 0 .
G i o r a , R a c h e l . 1 9 9 1 . O n t h e c o g n i t i v e a s p e c t s o f t h e j o k e . J o u r n a l o f P r a g n a a t i c s 16(51: 4 6 5 - - 4 8 6 .
G i o r a , R a c h e l , 1 9 9 3 . O n t h e f u n c t i o n o f a n a l o g i e s in i n t - o r m a t i v e t e x t s . D i s c o u r s e P r o c e s s e s
16:
59t--611.
G i o r a , R a c h e l , 1 9 9 5 . O n i r o n y a n d n e g a t i o n . D i s c o u r s e P r o c e s s e s 19: 2 3 9 - - 2 6 4 .
R. G i o r a / Jour*~al o f P r a g m a t i c s 2 6 ( 1 9 9 6 ) 4 1 7 - - 4 3 6
435
Giora, Rachel, Nachshon
M e i r a n a n d P a z O r e f , 1996. I d e n t i f i c a t i o n o f w r i t t e n d i s c o u r s e - t o p i c s b y
structure coherence and analogy strategies: General aspects and individual differences. Journal of
Pragmatics 26(4).
Giora, Rachel and Cher leng Lee, 1996. Written dis c ours e s e gm e nta tion: T he func tion of uns tre s s e d pron o u n s in M a n d a r i n C h i n e s e . I n : J e a n n e t t e G u n d e l a n d T h o r s t e i n F r e t h e i m , e d s . . R e f e r e n c e a n d r e f e r ence accessibility. Amsterdam:
Benjamins.
G i v 6 n , T a l m y , 1 9 8 3 . T o p i c c o n t i n u i t y in d i s c o u r s e : A q u a n t i t a t i v e c r o s s l a n g u a g e s t u d y . A m s t e r d a m :
Benjamins.
G r e e n o , J a m e s G . a n d D a v i d L. N o r e e n , 1 9 7 4 . T i m e t o r e a d s e m a n t i c a l l y r e l a t e d s e n t e n c e s . M e m o r y &
C o g n i t i o n 2: ! 1 7 - - 1 2 0 .
G r i c e , H. P a u l , 1975. L o g i c a n d c o n v e r s a t i o n . In: P e t e r C o l e a n d J e r r y M o r g a n , eds., S y n t a x a n d s e n r a n t i c s , V o l . 3: S p e e c h A c t s , 4 1 - - 5 8 . N e w Y o r k : A c a d e m i c P r e s s .
(Sundel. J e a n n e t t e K., N a n c y H e d b e r g a n d R o n Z a c h a r s k i , 1993. C o g n i t i v e status and the l o r m o f referr i n g e x p r e s s i o n s in d i s c o u r s e . L a n g u a g e 6 9 ( 2 ) : 2 7 4 ~ , 0 7 .
H a b e r l a n d t , K a r l , 1980. S t o r y g r a m m a r a n d r e a d i ng tim e of s tory c ons titue nt. P oe tic s 9:99--118.
Haberlandt, Karl, 1984. Components
o f s e n t e n c e a n d w o r d r e a d i n g t i m e s . I n : D a v i d E. K i e r a s a n d
M a r c e l A . J u s t , e d s . , N e w m e t h o d s in r e a d i n g c o m p r e h e n s i o n r e s e a r c h , 2 1 9 - - 2 5 2 . H i l l s d a l e , N J : E r l baum.
H a b e r l a n d t . K a r l , C l a i r e B e r i a n a n d J e n n i f e r S a n d s o n . 1 9 8 0 . T h e e p i s o d e s c h e m a in s t o r y p r o c e s s i n g .
J o u r n a l o f V e r b a l L e a r n i n g a n d V e r b a l B e h a v i o r 19: 6 3 5 - - 6 5 0 .
H a k e s , D a v i d T., 1971. D o e s v e r b s t r u c t u r e a f f e c t s e n t e n c e c o m p r e h e n s i o n .
Perception and Psy
c h o p h y s i c s 10: 2 2 9 - - 2 3 2 _
H a k e s , D a v i d T . a n d D o n a l d J. F o s s , 1 9 7 0 . D e c i s i o n p r o c e s s e s d u r i n g s e n t e n c e c o m p r e h e n s i o n :
Effects
o f s u r f a c e s t r u c t u r e r e c o n s i d e r e d . P e r c e p t i o n a n d P y s c h o p h y s i c s 8 : 4 1 3 4 16.
H o f m a n n , T h o m a s R., 1 9 8 9 . P a r a g r a p h s & a n a p h o r a . J o u r n a l o f P r a g m a t i c s 13: 2 3 9 - 2 5 0 .
H o u g h , M o n i c a S., 1990. N a r r a t i v e c o m p r e h e n s i o n
in a d u l t s w i t h r i g h t a n d left h e m i s p h e r e - d a m a g e :
T h e m e o r g a n i z a t i o n . B r a i n a n d L a n g u a g e 38. 2 5 3 - - 2 7 7 .
J o h n s o n - L a i r d , P h i l i p N i c h o l a s , 1983_ M e n t a l m o d e l s . C a m b r i d g e . M A : H a r v a r d U n i v e r s i t y P r e s s _
K e e n a n , J a n i c e M . , S u s a n D. B a i l l e t a n d P o l l y B r o w n , ! 9 8 4 . T h e e f f e c t s o f c a u s a l c o h e s i o n o n c o m p r e
hension and memory. Journal of Verbal Learning and Verbal Behavior 23: 115--126.
K i e r a s , E. D a v i d , 1 9 8 0 . I n i t i a l m e n t i o n a s a c u e to t h e m a i n i d e a a n d t h e m a i n i t e m o f a t e c h n i c a l p a s sage. Memory and Cognition 8:345
353.
K i n t s c h , W a l t e r , 1 9 8 8 . T h e r o l e o f k n o w l e d g e in d i s c o u r s e c o m p r e h e n s i o n .
Psychological Review 95:
163-182.
Kutas, martaand
S . A . H i l l y a r d , 1982. T h e l a t e r a l d i s t r i b u t i o n o f e v e n t r e l a t e d p o t e n t i a l s d u r i n g s e n t e n c e
processing. Neuropsychologia
20(5): 579--590_
L a b o r , W i l l i a m , 1 9 7 2 . T h e t r a n s f o r m a t i o n o f e x p e r i e n c e in n a r r a t i v e s y n t a x , L a n g u a g e in t h e i n n e r c i t y ,
354~405. Philadelphia, PA: University of Pennsylvania Press.
g o n g a c r e , R o b e r t , E_, 1 9 7 9 . T h e p a r a g r a p h a s a g r a m m a t i c a l u n i t . I n : T a l n l y G i v 6 n . e d . , S y n t a x a n d
s e m a n t i c s , V o l . 12: D i s c o u r s e a n d s y n t a x , 6 9 - - 1 3 5 . N e w Y o r k : A c a d e m i c P r e s s .
m a n d l e r , J e a n M . a n d M a r s h a S. G o o d m a n , 1 9 8 2 . O n t h e p s y c h o l o g i c a l v a l i d i t y o f s t o r y s t r u c t u r e . J o u r n a l o f V e r b a l L e a r n i n g a n d V e r b a l B e h a v i o r 21 : 5 0 7 - - 5 2 3 .
marslen-Wilson,
W i l l i a m , E l e n a L e v y a n d L o r r a i n T y l e r , 1982. P r o d u c i n g i n t e r p r e t a b l e d i s c o u r s e : T h e
e s t a b l i s h m e n t a n d m a i n t e n a n c e o f r e f e r e n c e . I n : R o b e r t J. J a r v e l l a a n d W o l f g a n g K l e i n , e d s . . S p e e c h ,
place and action, 328 339. New York: Wiley.
Marslen-Wilson,
William, gorrain Taller and Mark Seidenberg,
1978. S e n t e n c e p r o c e s s i n g and the
c l a u s e b o u n d a r y . I n : W i i l e m J . M . g e v e l t a n d G i o v a n n i B. F l o r e s d ' a r c a i s , e d s . . S t u d i e s in t h e p e r ception of language, 219-246. New York: Wiley.
Marslen-Wilson,
W i l l i a m a n d A . W e l s h , 1978. P r o c e s s i n g i n t e r a c t i o n a n d l e × i c a l a c c e s s d u r i n g w o r d
r e c o g n i t i o n in c o n t i n u o u s s p e e c h . C o g n i t i o n I(1: 2 9 - - 6 3 .
M c C l e l l a n d , J a m e s L. a n d A l a n H. K a w a m o t o ,
1986. M e c h a n i s m s o f s e n t e n c e p r o c e s s i n g : A s s i g n i n g
r o l e s to c o n s t i t u e n ts of sentences. In: James g. m c Cle lla nd a nd Da vid g. Rum e lha rt, eds., Parallel distributed processing: Explorations of the microstructure of cognition. Vol. I : Foundations, 272 325.
Cambridge, MA: MIT Press.
436
R. G i o r a / J o u r n a l o f P r a g m a t i c s
20 (1996) 417~36
M c C l e i l a n d , J a m e s L . a n d D a v i d E. R u m e l h a r l , 198 I. A n i n t e r a c t i v e a c t i v a t i o n m o d e l o f c o n t e x t e f f e c t s
in l e t t e r p e r c e p t i o n . P a r t 1 : A n a c c o u n t o f b a s i c f i n d i n g s . P s y c h o l o g i c a l R e v i e w 8 8 : 3 7 5 - 4 0 7 .
Norris, Dennis, 1986. Word recognition: Context effects without priming. Cognition 22: 93--136.
P r e n t i c e , J o h n L . , 1 9 6 7 . E f f e c t s o f c u e i n g a c t o r v s . c u e i n g o b j e c t o n w o r d o r d e r in s e n t e n c e p r o d u c t i o n .
Psychonomic
S c i e n c e 8: 1 6 3 - - 1 6 4 .
P r i n c e , E l l e n F., 1 9 8 1 a . O n i n f e r e n c i n g o f t h e i n d e f i n i t e - t h i s N P s . I n : A r a v i n d J o s h i , B o n n i e W e b b e r a n d
Ivan Sag, eds., Elements of discourse understanding. Cambridge:
Cambridge University Press.
P r i n c e , E l l e n F.. 1 9 8 1 b . T o w a r d a t a x o n o m y o f g i v e n - - n e w i n f o r m a t i o n . I n : P e t e r C o l e , e d . , R a d i c a l
pragmatics, 223 255. New York: Academic Press.
R e i n h a r t , T a n y a , 1980_ C o n d i t i o n s f o r t e x t c o h e r e n c e . P o e t i c s T o d a y 1 ( 4 ) : 1 6 1 - 1 8 0 .
Rosch, Elinor, 1973. On the internal structure of perceptual and semantic categories. In: T.E. Moore, ed.,
Cognitive development and the acquisition of language, 111--144. New York: Academic Press.
Schneidennan,
E t a I., K u n i k o G . M u r a s u g i a n d D o u g l a s J_ S a d d y , 1 9 9 2 . S t o r y a r r a n g e m e n t a b i l i t y in
right hemisphere-damaged
p a t i e n t s . B r a i n a n d L a n g u a g e 4 3 ( 1 ): 107 120.
Shannon, C.E., 1951. Prediction and entropy of printed English. Bell System Technical Journal 30:
50--64.
S h i e l d s . J o y c e L . . A s t r i d M c H u g h a n d J a m e s J. M a r t i n . 1 9 7 4 . R e a c t i o n t i m e t o p h o n e m e t a r g e t s a s a
f u n c t i o n o f r h y t h m i c c u e s in c o n t i n u o u s s p e e c h . J o u r n a l o f E x p e r i i n e n t a l P s y c h o l o g y
102: 2 5 0 - 2 5 5 .
Swinney, David A., 1979. Lexieal access during sentence comprehension:
(Re)consideration
of context
e f f e c t s . J o u r n a l o f V e r b a l L e a r n i n g a n d V e r b a l B e h a v i o r 18: 6 , ~ 5 ~ 5 5 9 .
T o m l i n , R u s s e l S . , 1 9 8 3 . O n t h e i n t e r a c t i o n o f s y n t a c t i c s u b j e c t , t h e m a t i c i n f o r m a t i o n a n d a g e n t in E n g lish. Journal of Pragmatics 7:411~432.
T u r n e r , E l i z a b e t h A. and Ragnar Rommetveit,
1968_ F o c u s o f a t t e n t i o n in r e c a l l o f a c t i v e a n d p a s s i v e
sentences. Journal of Verbal Learning and Verbal Behavior 7 : 543--548.
V a n D i j k , "Feun A , . 1 9 7 5 , R e c a l l i n g a n d s u m m a r i z i n g
complex discourse. Ms., University of Amsterdana.
Van Dijk, Teun A., 1977. Text and context. London: Longman.
Van Petten. Cyma and Maria Kutas, 1987. Ambiguous
w o r d s in c o n t e x t : A n e v e n t r e l a t e d p o t e n t i a l
analysis of the time course of meaning activation. Journal of Memory and Language 26: 188--208.
W a l t z , D a v i d L. a n d J o r d a n B. P o l l a c k , 1 9 8 5 . M a s s i v e l y p a r a l l e l p a r s i n g : A s t r o n g l y i n t e r a c t i v e m o d e l
of natural language interpretation. Cognitive Science 9: 51-74.