Resolving Anaphors in Embedded Sentences 
Saliha Azzam 
University of Sheffield 
Department of Computer Science 
Regent Court 
211 Portobello Street 
Sheffield S1 4DP U.K. 
S. Azzam~dcs. shef. ac. uk 
Abstract 
We propose an Mgorithm to resolve 
anaphors, tackling mainly the problem of 
intrasentential antecedents. We base our 
methodology on the fact that such an- 
tecedents are likely to occur in embedded 
sentences. Sidner's focusing mechanism is 
used as the basic algorithm in a more com- 
plete approach. The proposed algorithm 
has been tested and implemented as a part 
of a conceptual analyser, mainly to pro- 
cess pronouns. Details of an evaluation are 
given. 
1 Introduction 
Intrasentential antecedents, i.e., antecedents occur- 
ring in the same sentence as the anaphor, are a cru- 
cial issue for the anaphora resolution method. The 
main problem is to determine the constraints that 
intrasentential phrases must respect in anaphoric re- 
lations. These constraints are used to determine re- 
lations between a given anaphor and its antecedents. 
Until now, this kind of constraint has been tackled 
mainly in terms of syntactic aspects, see (Lappin and 
Leass, 1994) (Merlo, 1993) and (Hobbs, 1985). We 
propose to consider new kinds of criteria that com- 
bine semantic restrictions with sentence structure. 
One of these criteria is, for example, the way 
in which the verb meaning influences the sentence 
structure, then the way in which the sentence struc- 
ture influences the anaphoric relations between in- 
trasentential phrases. The structure we studied is 
the embedded sentence structure. Indeed, an im- 
portant assumption we have made is that embedded 
sentences favour the occurrence of intrasentential an- 
tecedents. We exploit the focusing mechanism pro- 
posed by Sidner (Sidner, 1979) (Sidner, 1981) (Sial- 
her, 1983) extending and refining her algorithms. 
The algorithm is designed for anaphors generally, 
even if we focus mainly on pronouns in this paper. 
Indeed, the distinction between different kinds of 
anaphors is made at the level of anaphor interpreta- 
tion rules. These resolution rule aspects will not be 
developed here; though they have been developed in 
the literature, e.g., see (Carter, 1987), and (Sidner, 
1981) (Sidner, 1983). We focus more on the mecha- 
nisms that handle these different kinds of rules. 
We first present how intrasentential antecedents 
occur in embedded sentences. We recall the main 
ideas of the focusing approach, then expand on 
the main hypotheses which led the design of the 
anaphora resolution algorithm. 
2 Intrasentential Antecedents 
2.1 Embedded sentences and elementary 
events 
An embedded sentence contains either more than 
one verb or a verb and derivations of other verbs 
(see sentence 1 with verbs said and forming). 
1) Three of the world's leading advertis- 
ing groups, Agence Havas S.A. of France, 
Young & Rubicam of the U.S. and Dentsu 
Inc. of Japan, said they are forming a 
global advertising joint venture. 
Broadly speaking embedded sentences concern more 
than one fact. In sentence 1 there is the fact of saying 
something and that of forming a joint venture. We 
call such a fact an elementary event (EE hereafter). 
Thus an embedded sentence will contain several EEs. 
Factors that influence embedded sentences are 
mainly semantic features of verbs. For example 
the verb to say, that takes a sentence complement 
favours an introduction of a new fact, i.e., "to say 
something", and the related fact. There are other 
classes of verbs such as want to, hope that, and so 
on. In the following, subordinate phrases, like rel- 
ative or causal sentences, will be also considered as 
embedded ones. 
2.2 Embedded sentences with 
intrasentential antecedents 
First of all, we will distinguish the Possessive, Recip- 
rocal and Reflexive pronouns (PRR hereafter) from 
the other pronouns (non-PRR hereafter). 
263 
On the basis of 120 articles, of 4 sentences on av- 
erage, containing 332 pronouns altogether, we made 
the following assumption (1): 
Assumption: non-PRR pronouns can have in- 
trasentential antecedents, only if these pro- 
nouns occur in an embedded sentence. 
The statistics below show that of 262 non-PRR 
pronouns, there are 244 having intrasentential an- 
tecedents, all of which occur in embedded sentences 
and none in a "simple" sentence. The remaining 18 
non-PRR pronouns have intersentential antecedents. 
Pronouns 332 
non-PRR 262 
With intrasentential antecedents 244 
in an embedded sentence 
With intrasentential in a simple 0 
sentence 
With intersentential antecedents 18 
Our assumption means that, while the PRR pro- 
nouns may find their antecedents in an non embed- 
ded sentence (e.g., sentences 2 and 3) the non-PRR 
pronouns can not. 
2) Vulcan made i~s initial Investment in 
Telescan in May, 1992. 
3) The agencies HCM and DYR are ~hem- 
selves joint ventures. 
Without jumping to conclusions, we cannot avoid 
making a parallel with the topological relations de- 
fined in the binding theory (Chomsky, 1980), be- 
tween two coreferring phrases in the syntactic tree 
level. Assumption 1 redefines these relations in an 
informal and less rigorous way, at the semantic level, 
i.e., considering semantic parameters such as the 
type of verbs that introduce embedded sentences. 
2.3 Using Sidner's Focusing Approach 
To resolve anaphors one of the most suitable existing 
approaches when dealing with anaphor issues in a 
conceptual analysis process is the focusing approach 
proposed by Sidner. However, this mechanism is 
not suitable for intrasentential cases. We propose 
to exploit its main advantages in order to build our 
anaphora resolution mechanism extending it to deal 
also with intrasentential antecedents. 
We describe the main elements of the focusing ap- 
proach that are necessary to understand our method, 
without going into great detail, see (Sidner, 1979) 
(Sidner, 1981) (Sidner, 1983). Sidner proposed a 
methodology, modelling the way "focus" of attention 
and anaphor resolution influence one another. Us- 
ing pronouns reflects what the speaker has focused 
on in the previous sentence, so that the focus is that 
phrase which the pronouns refer to. The resolution 
is organised through the following processes: 
264 
• The expected focus algorithm that selects an 
initial focus called the "expected focus". This 
selection may be "confirmed" or "rejected" in 
subsequent sentences. The expected focus is 
generally chosen on the basis of the verb seman- 
tic categories. There is a preference in terms 
of thematic position: the "theme" (as used by 
Gruber and Anderson, 1976 for the notion of 
the object case of a verb) is the first, followed 
by the goal, the instrument and the location or- 
dered according to their occurrence in the sen- 
tence; the final item is the agent that is selected 
when no other role suits. 
• The anaphora interpreter uses the state of the 
focus and a set of algorithms associated with 
each anaphor type to determine which element 
of the data structures is the antecedent. Each 
algorithm is a filter containing several interpre- 
tation rules (IR). 
Each IR in the algorithm appropriate to an 
anaphor suggests one or several antecedents de- 
pending on the focus and on the anaphor type. 
• An evaluation of the proposed antecedents is 
performed using different kinds of criteria (syn- 
tax, semantics, inferential, etc.) 
• The focusing algorithm makes use of data struc- 
tures, i.e., the focus registers that represent the 
state of the focus: the current focus (CF) repre- 
sentation, alternate focus list (AFL) that con- 
tains the other phrases of the sentence and the 
focus stack (FS). A parallel structure to the CF 
is also set to deal with the agentive pronouns. 
The focusing algorithm updates the state of the 
focus after each sentence anaphor (except the 
first sentence). After the first sentence, it con- 
firms or rejects the predicted focus taking into 
account the results of anaphor interpretation. 
In the case of rejection, it determines which 
phrase is to move into focus. 
This is a brief example (Sidner 1983) : 
a Alfred and Zohar liked to play baseball. 
b They played it every day after school before din- 
ner. 
c After lheir game, Alfred and Zohar had ice 
cream cones. 
d They tasted really good. 
• In a) the expected focus is "baseball" (the 
theme) 
• In b) "it" refers to "baseball" (CF). "they" 
refers to Alfred and Zohar (AF) 
• The focusing algorithm confirms the CF. 
• In d) "they" refers to "ice cream cones" in AFL. 
• The focusing algorithm decides that since no 
anaphor refers to the CF, the CF is stacked and 
"ice cream cones" is the new CF (focus move- 
ment). 
We call a basic focusing cycle the cycle that in- 
cludes : 
• the focusing algorithm 
• followed by the interpretation of anaphors, 
• then by the evaluation of the proposed an- 
tecedents. 
2.4 What needs to be improved in the 
focusing approach? 
2.4.1 Intrasentential antecedents 
The focusing approach always prefers the previ- 
ous sentences' entities as antecedents to the current 
sentences. In fact only previous sentence entities 
are present in the focus registers. Thus phrases of 
the current sentence can not be proposed as an- 
tecedents. This problem has already been under- 
lined, see (Carter, 1987) in particular who pro- 
posed augmenting the focus registers with the en- 
tities of the current sentence. For example in sen- 
tence 4 while the focus algorithm would propose 
only "John" as an antecedent for "him", in Carter's 
method "Bill" will also be proposed. 
4) John walked into the room. He told Bill 
someone wanted to see him. 
2.4.2 Initial Anaphors 
The focusing mechanism fails in the expected fo- 
cus algorithm when encountering anaphors occur- 
ring in the first sentence of a text, which we call 
initial anaphors, such as They in sentence (1). The 
problem with initial anaphors is that the focus reg- 
isters cannot be initialised or may be wrongly filled 
if there are anaphors inside the first sentence of the 
text. It is clear that taking the sentence in its classi- 
cal meaning as the unit of processing in the focusing 
approach, is not suitable when sentences are embed- 
ded. 
We will focus on the mechanisms and algorithmic 
aspects of the resolution (how to fill the registers, 
how to structure algorithms, etc.) and not on the 
rule aspects, like how IRs decide to choose Bill and 
not John (sentence 4). 
3 Our Solution 
As stated above, embedded sentences include sev- 
eral elementary events (EEs). EEs are represented 
as conceptual entities in our work. We consider 
that such successive EEs involve the same context 
that is introduced by several successive short sen- 
tences. Moreover, our assumption states that when 
non-PRR anaphors have intrasentential antecedents, 
they occur in embedded sentences. Starting with 
these considerations, the algorithm is governed by 
the hypotheses expanded below. 
3.1 Main hypotheses 
First hypothesis : EE is the unit of processing in 
the basic focusing cycle. 
An EE is the unit of processing in our resolution al- 
gorithm instead of the sentence. The basic focusing 
cycle is applied on each EE in turn and not sentence 
by sentence. Notice that a simple sentence coincides 
with its EE. 
Second hypothesis : The "initial" EE of a well 
formed first sentence does not contain non-PRR 
pronouns just as an initial simple sentence can- 
not. 
For example, in the splitting of sentence 1 into two 
EEs (see below), EEl does not contain non-PRR 
pronouns because it is the initial EE of the whole 
discourse. 
EEl) '¢rhree of the world's leading adver- 
tising groups, Agence I-Iavas S.A. of France, 
Young & Rubicam of the U.S. and Dentsu 
Inc. of Japan, said" 
EE2) "they are forming a global advertis- 
ing joint venture." 
Third hypothesis : PRR pronouns require special 
treatment. 
PRR could refer to intrasentential antecedents in 
simple sentences (such as in those of sentences 3 and 
4). An initial EE could then contain an anaphor of 
the PRR type. Our approach is to add a special 
phase that resolves first the PRRs occurring in the 
initial EE before applying the expected focusing al- 
gorithm on the same initial EE. In all other cases, 
PRRs are treated equally to other pronouns. 
This early resolution relies on the fact that the 
PRR pronouns may refer to the agent, as in sentence 
3, as well as to the complement phrases. However 
the ambiguity will not be huge at this first level of 
the treatment. Syntactic and semantic features can 
easily be used to resolve these anaphors. This relies 
also on the fact that the subject of the initial EE 
cannot be a pronoun (second hypothesis). 
Having mentioned this particular case of PRR in 
initial EE, we now expand on the whole algorithm 
of resolution. 
3.2 The Algorithm 
In the following, remember that what we called the 
basic focusing cycle is the following successive steps 
• applying the resolution rules, 
• applying the focusing algorithm, i.e., updating 
the focus registers 
• the evaluation of the proposed antecedents for 
each anaphor. 
265 
The algorithm is based on the decomposition of 
the sentence into EEs and the application of the ba- 
sic focusing cycle on each EE in turn and not sen- 
tence by sentence. 
The complete steps are given below (see also figure 1): 
Step 1 Split the sentence, i.e., its semantic repre- 
sentation, into EEs. 
Step 2 Apply the expected focus algorithm to the 
first EE. 
Step 3 Perform the basic focusing cycle for every 
anaphor of all the EEs of the current sentence. 
Step 4 Perform a collective evaluation (i.e., evalu- 
ation that involves all the anaphors of the sen- 
tence), when all the anaphors of the current sen- 
tence are processed. 
Step 5 Process the next sentence until all the sen- 
tences are processed: 
• split the sentence into EEs 
• apply Step 3 then Step 4. 
F~rst Sentence 
1 
Sentence Splitting Algorithm \[ 
Next EE 
I Sentence Splitting 
Algorith~r 
Expected Focus Algorithm \[ I 
\[ Interpretation of each Anaphor \] l 
Evaluation of the proposed antecedents I 
1 
| 
Next EE \[ No more EEs 
Nex~t Collective Evaluation of the Antecedents \[ ! 
Sentence | No more sentences 
Basic 
Focusing 
Cycle 
Figure 1: The Algorithm 
Main Results : 
1. Intrasentential antecedents are taken into ac- 
count when applying the focusing algorithm. 
For example, in sentence 1, the intrasentential 
antecedent Bill will be taken into account, be- 
cause EEl would be processed beforehand by 
the expected focusing algorithm. 
2. The problem of initial anaphors is then re- 
solved. The expected focusing algorithm is ap- 
plied only on the initial EE which must not con- 
tain anaphors. 
3.3 Examples and results 
To illustrate the algorithm, let's consider the follow- 
ing sentence : 
Lafarge Coppee said it would buy 10 per- 
cent in National Gypsum, the number two 
plasterboard company in the US, a pur- 
chase which allows it to be present on the 
world's biggest plasterboard market. 
At the conceptual level, there are 3 EEs. They are 
involved respectively by the said, buy, and allows 
verbs. They correspond respectively to the following 
surface sentences: 
EEl "Lafarge Coppee said" 
EE2 "it would buy 10 percent in National Gypsum, 
the number two plasterboard company in the 
US" 
EE3 "a purchase which allows it to be present on the 
world's biggest plasterboard market." 
Consider the algorithm : 
• the expected focusing algorithm is applied to 
the first EE, EEl, which contains non-PRR 
anaphors. 
• the other phases of the algorithm, i.e., the basic 
focusing cycle, are applied to the subsequent 
EEs : 
- EE2 contains only one pronoun it, which is 
resolved by the basic focusing cycle 
- it in EE3 will be resolved in the same way. 
The anaphora resolution has been implemented as 
a part of a conceptual analyser (Azzam, 1995a). It 
dealt particularly with pronouns. It has been tested 
on a set of 120 news reports. We made two kinds 
of algorithm evaluations: the evaluation of the im- 
plemented procedure and an evaluation by hand. 
For the implementation the success rate of resolu- 
tion was 70%. The main cases of failure are related 
to the non implemented aspects like the treatment 
of coordination ambiguities and the appositions, or 
other anaphoric phenomena, like ellipsis. 
For the second evaluation which concerns the real 
evaluation of the approach,i.e., without going into 
the practical issues concerning implementation, the 
success rate was 95%. The main cases of failure were 
due to the cases that were not considered by the 
algorithm, like for example the pronouns occurring 
before their antecedents , i.e., cataphors. Such cases 
occur for example in sentences 5 and 6 pointed out 
266 
by Hobbs (IIobbs, 1985) to discuss the cases that are 
not handled easily in the literature. 
5) Mary sacked out in his apartment before 
Sam could kick her out. 
6) Girls who he has dated say that Sam is 
charming. 
Our algorithm fails in resolving his in 5, because 
the algorithm searches only for the entities that pre- 
cede the anaphor in the text. The same applies for 
he in 6. However improving our algorithm to process 
classical cases of cataphors, such as that in sentence 
6, should not require major modifications, only a 
change in the order in which the EEs are searched. 
For example, to process pronouns of the sentence 
6 split into two EES (see below), the algorithm must 
consider EE2 before EEl. This means applying the 
step 2 of the algorithm to EE2, then step 3 to EEl. 
The sentence 5 should require specific treatment, 
though. 
EEl) "that Sam is charming" 
EE2) "Girls who he has dated say" 
IIobbs also pointed out the cases of "picture noun" 
examples, as in sentences 7 and 8: 
7) John saw a picture of him. 
8) John's father's portrait of him. 
In 7 our algorithm is successful, i.e., it will not iden- 
tify him with John because of our previous assump- 
tion (section 2.2). However our algorithm would fail 
in 8 because the non-PRR pronoun him could refer 
to John which occurs in the same EE. 
Notice that Hobbs' (I-Iobbs, 1985) remark that 
"the more deeply the pronoun is embedded and the 
more elaborate the construction it occurs in, the 
more acceptable the non reflexive" is consistent with 
our assumption. 
For example in the embedded sentence 9 where ei- 
ther the reflexive (himself) or non reflexive pronouns 
(him) may be used, it is more natural to make use 
of him. 
9) John claimed that the picture of him 
hanging in the post office was a fraud. 
4 The Conceptual Level 
We comment here on the main aspects of the con- 
ceptual analysis that are related to the anaphora 
resolution process. They concern mainly the way of 
splitting embedded sentences and the problems of 
determining the theme and of managing the other 
ambiguities and the several readings. 
The conceptual analyser's strategy consists of a 
continuous step-by-step translation of the original 
natural language sentences into conceptual struc- 
tures (CS hereafter). This translation uses the re- 
sults of the syntactic analysis (syntactic tree). It is 
a progressive substitution of the NL terms located 
in the syntactic tree with concepts and templates of 
the conceptual representation language. Triggering 
rules are evoked by words of the sentence and allow 
the activation of well-formed CS templates when the 
syntactico-semantic filter is unified with the syntac- 
tic tree. The values caught by the filter variables 
are the arguments of the CS roles, i.e., they fill the 
CS roles. If they are anaphors, they are considered 
to be unbound variables and result in unfilled roles 
in the CS. The anaphora resolution aims therefore 
at filling the unfilled roles with the corresponding 
antecedents. 
4.1 Splitting into EEs 
The splitting of a sentence in EE is done on the 
corresponding CS. A minimal CS is a template com- 
prising a predicate that identifies the basic type of 
the represented event and a set of roles or predicate 
cases. 
For example, the sentence "to say that they agree 
to form a joint venture" is represented, in a simpli- 
fied way, with three templates, corresponding to the 
predicates: 
• move information (from "to say"), 
• produce an agreement (from "to agree"), 
• produce a joint venture (from "to form"). 
Given that one template at the semantic level repre- 
sents an elementary event, the splitting is implicitly 
already done when these templates are created in the 
triggering phase. Indeed, the syntactico-semantic 
filter of the triggering rules takes into account the 
semantic features of words (mainly verbs) for recog- 
nising in the surface sentence those that are able to 
trigger an elementary event. 
4.2 Determining the theme 
Gruber and Anderson characterise the theme as fol- 
lows: if a verb describes a change to some entity, 
whether of position, activity, class or possession, 
then the theme is the changed entity, (Gruber, 1976) 
and (Anderson, 1977). As Carter (Carter, 1987) 
demonstrated, this definition of Gruber and Ander- 
son is sufficient to apply the focusing mechanism. 
This assumption is particularly apt when we dispose 
of a conceptual representation. Indeed, to deter- 
mine the thematic roles, we established a set of the- 
matic rules that affect for a given predicative occur- 
rence, its thematic functions according to the predi- 
cate type, the role type and the argument's semantic 
class. 
4.3 Managing other ambiguities 
An important aspect appears when one designs a 
concrete system, namely how to make other dis- 
ambiguation processes cohabit. In the concep- 
tual analyser, the general disambiguation module 
(GDM) deals with other ambiguities, like preposi- 
tional phrase attachment. It coordinates the treat- 
267 
ment of different kinds of ambiguities. This is nec- 
essary because the conceptual structures (CS) on 
which the rules are performed could be incomplete 
because of other types of ambiguities not being re- 
solved. For example, if the CF of the sentence is 
a PP object that is not attached yet in the CS the 
thematic rules fail to fill the CF. The GDM ensures 
that every disambiguation module intervenes only 
if previous ambiguities have already been resolved. 
The process of co-ordinating ambiguity processing is 
fully expanded in (Azzam, 1995b). 
4.4 Multiple readings 
When dealing with ambiguities, another important 
aspect is managing multiple readings. At a certain 
point when the GDM calls the anaphora module to 
deal with a given anaphor, the status of the concep- 
tual analysis could be charaeterised by the following 
parameters : 
• The set of conceptual structures for the current 
reading Ri on which the resolution is performed, 
given that several readings could arise from pre- 
vious ambiguity processing. 
• The set of conceptual structures of the current 
sentence Si where the anaphor occurs; 
• The set of conceptual structures of the current 
elementary event EEi where the anaphor occurs 
after the Si splitting. 
• The state of the focus (content of the registers), 
SFi 
The main assumption is that the anaphora resolu- 
tion algorithm always applies to a single state, (Ri, 
Si , EEi, SFi) when resolving a given anaphor (Step 
3): 
a If several antecedents are still possible after the 
individual evaluation of the anaphor, Ri is then 
duplicated, in Rij, as many times as there are 
possibilities. 
b When performing the collective evaluation of 
all Si anaphors, every inconsistent Rij is sup- 
pressed. 
c The result is a set of readings (Rij, Sj , EEj, 
SFi). 
5 Conclusion 
We have proposed a methodology to resolve 
anaphors occurring in embedded sentences. The 
main idea of the methodology is the use of other 
kinds of restrictions between the anaphor and its an- 
tecedents than the syntactic ones. We demonstrated 
that anaphors with intrasentential antecedents are 
closely related to embedded sentences and we 
showed how to exploit this data to design the 
anaphora resolution methodology. Mainly, we ex- 
ploited Sidner's focusing mechanism, refining the 
classical unit of processing, that is the sentence, to 
that of the elementary event. The algorithm has 
been implemented (in Common Lisp, Sun Spare) to 
deal with pronouns as a part of a deep analyser. The 
main advantages of the proposed algorithm is that 
it is independent from the knowledge representation 
language used and the deep understanding approach 
in which it is integrated. Thus, it could be set up in 
any conceptual analyser, as long as a semantic rep- 
resentation of the text is available. Moreover Sid- 
ner's approach does not impose its own formalisms 
(syntactic or semantic) for its application. The im- 
provement of the proposed algorithm requires deal- 
ing with special cases of anaphors such as cataphors 
and also with specific cases which are not easily han- 
dled in the literature. For example, we saw that a 
solution to processing cataphors could be to recon- 
sider the order in which the conceptual structures 
(elementary events beforehand) are searched. 
6 Acknowledgements 
This work has been supported by the Euro- 
pean Community Grant LE1-2238 (AVENTINUS 
project). 

References 
Anderson, S.R. 1977. Formal syntax. In Wasow 
and Akmajian, editors, Comment on the paper by 
Wasow in Culicover. Academic Press, pages 361- 
376. 
Azzam, Saliha. 1995a. Computation of Ambiguities 
(Anaphors and PPs) in NL texts. CLAM : The 
prototype. Ph.D. thesis, Paris Sorbonne Univer- 
sity. 
Azzam, Saliha. 1995b. Anaphors, pps and disam- 
biguation process for conceptual analysis. In 14th 
International Joint Conference on Artificial In- 
telligence (IJCAI'95). San Mateo (CA): Morgan 
Kaufmann. 
Carter, David. 1987. Interpreting Anaphors in nat- 
ural language texts. Chichester : Ellis Horwood. 
Gruber, J.S. 1976. Lezical structures in syntax and 
semantics. North-Holland. 
Hobbs, Jerry. 1985. Resolving pronoun references. 
In B. Grosz K. Sparck-Jones B. Webber, editor, 
Readings in Natural Language, volume 44. Morgan 
Kaufmann Publishers Los Altos California, pages 
311-338. 
Lappin, S. and H.J. Leass. 1994. An algorithm for 
pronominal anaphora resolution. Computational 
Linguistics, 20:535-561. 
Merlo, P. 1993. For an incremental computation of 
intrasentential coreference. In the 13th Interna- 
tional Joint Conference on Artificial Intelligence 
(IJCAI'93), pages 1216-1221. San Mateo (CA): 
Morgan Kaufmann. 
Sidner, C. 1979. Toward a computation of in- 
trasentential coreference. Technical Report TR- 
537, MIT. Artificial Intelligence Laboratory. 
Sidner, C. 1981. Focusing for interpretation of pro- 
nouns. American Journal of Computational Lin- 
guistics, 7:217-231. 
Sidner, C. 1983. Focusing in the comprehension 
of definite anaphora. In Brady. M and Berwick 
R.C, editors, Computational Models of Discourse. 
Cambridge (MA) : The MIT Press. 
