Predicative NPs and the annotation of reference chains
Kaja Borthen
Norwegian University of Science and Technology
7491 Trondheim,
Norway,
kaja.borthen@hf.ntnu.no
Abstract
In the development of machine learning sys-
tems for identification of reference chains,
hand-annotated corpora play a crucial role.
This paper concerns the question of how
predicative NPs should be annotated w.r.t.
coreference in corpora for such systems.
This question highlights the tension that
sometimes appears in the development of
corpora between linguistic considerations
and the aim for perfection on the one
hand and practical applications and the aim
for efficiency on the other. Many current
projects that seek to identify coreferential
links automatically, assume an annotation
strategy which instructs the annotator to
mark a predicative NP as coreferential with
its subject if it is part of a positive sentence.
This paper argues that such a representa-
tion is not linguistically plausible, and that
it will fail to generate an optimal result.
1 Introduction
Predicative NPs in positive sentences are
marked as coreferential with their subject in
many projects that develop corpora for coref-
erence chains. This is exemplified by the an-
notation scheme of MUC-7 (Hirschman and
Chinchor, 1997), on which several annotation
schemes are based. Also in more theoretical
works this has been a common mark-up, at least
for definite predicative NPs. For example, the
proper name David Beckham and the definite
phrase the second best player in the world in (1)
belowareregardedascoreferentialandanaphor-
ically related to each other by Mitkov (2002),
whose intention seems to be to present the stan-
dard view on this issue.
(1) David Beckham was voted the second
best player in the world behind Rivaldo.
There are several arguments in favor of treat-
ing a predicative NP and its subject as part of
the same reference chain if they appear in a pos-
itive sentence. A predicative NP and its subject
are often used to describe the same individual in
the world. As the goal for automatic identifica-
tion of reference chains is primarily to identify
equivalence sets, i.e. sets of nominals that refer
to the same entity, this is a strong argument for
assuming coreference between predicative NPs
and their subject.
A second argument in favor of this is that one
otherwiselosesimportantinformationaboutref-
erents in terms of descriptive content. Consider
the following example:
(2) Skogfoss is your parents’ farm, isn’t it?
I remember that farm.
In (2) above, we want to capture the fact that
Skogfoss and that farm refer to the same entity.
If one assumes an annotation strategy where
Skogfoss and the predicative NP your parents’
farm are marked as part of the same reference
chain, then the descriptive content contributed
by farm is added to the reference chain, which
in turn makes it highly likely that the later ex-
pression that farm belongs to the same reference
chain, as desired. In other words, the descrip-
tive content contributed by a predicative phrase
- if assumed to be coreferential with its subject -
can be useful for identifying desired subsequent
coreferences. Correspondingly, it can also be
useful for ruling out undesired ones.
The fact that the majority of annotated cor-
pora for coreference identification mark predica-
tive NPs in positive sentences as coreferential
with their subject, is in other words quite well
motivated. Still, we will claim that it is legit-
imate to question the plausibility of this prac-
tice, due to facts that we will look at in the
next section. These arguments give support to
van Deemter and Kibble (2000), who argue that
it is problematic that the annotation strategy
assumed in MUC-6 and MUC-7 goes beyond a
mark-up of coreference.
2 Why non-reflexive predicative NPs
are not anaphoric to their subject
Reference chains may be marked in terms of set
membership, but they can also be marked in
terms of anaphor-antecedent pairs. We there-
fore want to consider whether it is plausible to
assume that predicative NPs are anaphoric to
their subject.
An anaphor is often defined as a constituent
whose interpretation is dependent on some
meaning aspect of a prior constituent, called
the antecedent. The need for an antecedent
is assumed to be due to an impoverished se-
mantic content in the anaphor itself, and typ-
ically, the antecedent is a more complete se-
mantic description. There are different types
of anaphor-antecedent relations, but we will fo-
cusonanaphor-antecedentrelationsthatencode
identity of reference. In such relations, nominal
forms play an important role in identifying an
anaphor’santecedent,asdifferentnominalforms
signal different attention states for the associ-
ated referent (Gundel et al., 1993).
Given this notion of anaphor, we will mention
four reasons not to assume that predicative NPs
are anaphorically related to their subject.
First, it is inherent in the notions anaphor
and antecedent that the antecedent is the source
of interpretation for the anaphor. Thus, if
an anaphor’s antecedent is identified and in-
terpreted, the anaphor is supposed to be inter-
pretable as well. If one assumes that predicative
NPs are anaphoric to their subject, this raises
the expectation that a predicative NP will al-
ways be interpretable as long as its subject is
interpreted. But this is not always the case:
(3) a. Fido is the dog I talked about.
b. A: Who is the witch in this play?
B: I am the witch.
In (3a), imagine a situation where Fido is
present in the immediate context, so that the
subject phrase Fido can easily be understood.
Assuming that predicative NPs are anaphoric to
theirsubject, thisshouldmeanthatthepredica-
tive NP the dog I talked about can be sufficiently
interpreted without any special contextual re-
quirements except from the sentence itself and
the presence of Fido. But this is not the case, as
the dog I talked about can only be fully under-
stood if the phrase has some other antecedent
than Fido. Thus, in this case, identifying the
subject as the antecedent of a predicative NP
is not sufficient to interpret the predicative NP.
The example in (3b) is another case where a
predicative NP does not seem to be anaphori-
cally related to its subject. The phrasethe witch
in B’s reply is an anaphor, as it cannot be in-
terpreted in isolation. As for its antecedent, it
makes more sense to say that it has the witch
in this play as its antecedent than the subject
phrase I. I is not anaphoric to the expression
the witch in this play, as the interpretation of I
is independent it. Thus, if one assumes that the
subject phrase I is the antecedent of the pred-
icative phrase the witch in (3b), then one does
not get access to the desired information that
the predicative phrase the witch refers to the
same witch as does the prior phrase the witch in
this play. In sum, the examples in (3) show that
assuming coreference between a predicative NP
and its subject in a positive predicative sentence
is not always enough to interpret the predicative
NP. This, in turn, suggests that a predicative
NP and its subject in a positive sentence do not
always stand in an anaphor-antecedent relation
to each other.
Something that supports this suggestion is
that it is hard to find differences between pos-
itive and negative predicative sentences w.r.t.
theinterpretationofthepredicativephrase. Ob-
viously, predicative NPs in negated sentences
arenotcoreferentialwiththeirsubjects, andnot
anaphorically related to them. If a predicative
NP and its subject were necessarily anaphor-
ically related in positive sentences, one would
expect that the interpretation of a predica-
tive NP in a positive sentence would be less
context-dependent than for a corresponding iso-
lated negative sentence, due to the presence of
a sentence-internal antecedent. But examples
that support this expectation are hard to find,
and have not yet been detected by this author.
Itseemsthatinbothpositiveandnegativepred-
icative sentences the predicative phrase can be
interpreted and sufficiently understood indepen-
dently of its subject.
A third argument for not assuming that pred-
icative NPs in positive sentences are anaphoric
to their subject is that they do not follow the
normal pattern for anaphoric expressions w.r.t.
NP-form. Whereas anaphora are often pronom-
inal and only very rarely indefinite, predicative
NPs are hardly ever pronominal and quite often
indefinite. The contrast w.r.t. nominal form is
illustrated by the following text fragments:
(4) a. Kari is in good shape. She is often
exercising. She/ a sporty girl bikes
to work every day.
b. Kari is in good shape. She is often
exercising. She is a sporty girl.
In (4a), the two occurrences of the pronoun
she are interpreted as referring to Kari. The in-
definite phrase a sporty girl, on the other hand,
is not likely to refer to Kari in (4a); rather, the
phraseisinterpretedgenerically. Thisillustrates
the fact that indefinite NPs are bad candidates
for being anaphora. This tendency does not
hold for predicative NPs, though, assuming that
they are anaphora. This is shown in (4b), where
weseethatanindefinitepredicativeNPcanvery
wellbeprecededbyapronounthatdescribesthe
same individual as the predicative NP. In other
words, if one assumes that predicative NPs are
anaphoric to their subject, one has to explain
why predicative NPs do not have the expected
formsofnominalanaphora.1PredicativeNPsare
typically either indefinite or initiated by the def-
inite article, which are forms that do not require
previous familiaritywith the referent (Gundel et
al., 1993).
A fourth argument for not treating predica-
tive NPs as anaphoric to their subject has to do
with binding principles (Chomsky, 1981). The
interpretation of a reflexive predicative NP is
dependent on the interpretation of its subject,
so we take it as uncontroversial that a reflexive
predicative NP like the one in (5a) is anaphoric
to its subject, just as the reflexive in (5b).
(5) a. She has always been herself.
b. She blamed herself.
Usually, in positions that license reflexives,
full lexical NPs cannot be inserted without af-
fecting the interpretation. This is illustrated by
thecontrastbetween(5b)aboveand(6b)below.
(6) a. She has always been my best friend.
b. She blamed my best friend.
Thereisaverystrongpreferenceforinterpret-
ing my best friend and she in (6b) as referring
1Mitkov (2002) suggests that whereas definite pred-
icative NPs should be regarded as anaphoric to their
subject, indefinite ones should not be regarded as such.
This solves the problem that predicative NPs are often
indefinite, whereas most anaphora are not. However,
this distinction between indefinite and definite predica-
tive NPs does not solve the more general problem that
definite as well as indefinite predicative NPs do not seem
to get their interpretation through their subject.
to different individuals, and it is impossible to
use (6b) to enforce coreference between the ob-
ject and the subject, as in (5b). This illustrates
themotivationforabindingprinciplewhichpre-
dicts, amongotherthings, thatalexicalNPcan-
not be anaphorically bound by a co-argument.
If we say that the lexical predicative NP in
(6a) is anaphoric to, and coreferential with, its
subject, then we have to assume that there are
different binding principles for the arguments of
the verb be than for other verbs. The difference
is then that with predicative NPs, unlike non-
predicative ones, both reflexive and nonreflexive
NPs can appear in the same position with the
same anaphoric interpretation. If we, on the
other hand, do not assume coreference and an
anaphor-antecedent relation in (6a), we can do
with one general binding principle.
3 Discussion
In section 2 we have seen arguments which sug-
gest that non-reflexive predicative NPs are not
anaphoric to their subject. First of all, as-
suming that a predicative NP is coreferential
with its subject is not always enough to as-
sign it an interpretation. Secondly, contrary
to what one would expect, there is no differ-
ence between positive and negative predicative
sentences w.r.t. the context-dependency of the
predicative phrase. Thirdly, non-reflexive pred-
icative NPs do not have the forms that are
expected for nominal anaphora. And finally,
non-reflexive predicative NPs behave differently
w.r.t. binding than what is expected if they are
assumed to be anaphora.
These facts suggest that the reason why many
predicative NPs are intuitively seen as describ-
ing the same individual as their subject is a dif-
ferent reason from that which identifies the an-
tecedent of an anaphor. We propose that non-
reflexive predicative NPs are not anaphoric to
their subject, but rather part of a predication
which holds of the subject referent.
But marking coreference is not necessarily the
same as marking anaphor-antecedent pairs, as
reference chains can also be thought of in terms
of equivalence sets. So does the conclusion that
non-reflexive predicative NPs are not anaphoric
to their subject mean that one should never
annotate predicative NPs as coreferential with
their subject when building corpora with refer-
ence chains? As far as we can see, that depends
on the intended application for the system, on
the machine learning system’s abilities, and on
what additional information is available or re-
trievable from the corpus and/or the system.
On the one hand: As long as the machine
learning method applied is capable of learning
two different systems at once, it may be desir-
able to include predicative NPs in the same ref-
erence chain as their subject, if part of posi-
tive sentences. As argued in section 1, predica-
tive NPs in positive sentences do describe the
same individual as their subject does, and the
semantic information in predicative NPs can be
useful for identifying desired later coreferences
and ruling out undesired ones. For practical
applications it is a goal to collect equivalence
classes, and the richer these are, the better.
The approach is furthermore not very effort-
demanding. The drawback of the approach is
that it seems to mix two distinct phenomena,
i.e. reference tracking and predication. The an-
notation strategy is therefore theoretically ques-
tionable and is likely to lead to unsatisfactory
results in some cases, as illustrated by the ex-
amples in (3).
On the other hand: Imagine a system where
referentsarenotseenasindividualsintheworld,
but abstract entities where e.g. (1) states a be-
relation between the discourse referent associ-
ated with David Beckham and the intensional
object corresponding to the second best player
in the world. Imagine further that one dis-
tinguishes between (discourse) referent tracking
and collection of predications. Predicative NPs
willonlybeannotatedascoreferentialwiththeir
subject if they are reflexive, but any predicative
NP will be kept track of as part of a predication
that holds of its subject referent. Thus, there
are two sources of information about referents;
the descriptions used to refer to them, and pred-
ications made about them. The latter source
will be of importance for reference tracking also
independently of predicative NPs. Consider (7).
(7) Fred drives a taxi, whereas Joe studies
math. Who would you prefer to meet,
the math student or the taxi driver?
The information retrieved from the predica-
tions about Fred and Joe is crucial to assign
the correct interpretation to the expressions the
math student andthe taxi driver. If predications
about referents are kept track of just as well as
reference chains, then we are in principle able to
achieve the correct interpretation for (7).
Thus, with the present approach, which
keeps reference tracking and predication track-
ing apart, it is not necessary to assume that
predicative NPs are coreferential with their sub-
ject in order to exploit the important semantic
information they may contribute about the sub-
ject referent. The present approach is also more
theoretically plausible in that it singles out two
arguablydistinctphenomena, anditisnotlikely
to go wrong in cases like those in (3). The main
drawback of this approach is that it is effort-
demanding, as it requires a semantic component
that keeps track of predications as well as ref-
erents. Such a component is not part of most
current automatic systems for reference track-
ing, and without it, the approach produces less
information about referents than its competitor
described above. For instance, without the com-
ponenttokeeptrackofpredications, onedoesn’t
capture the fact that a predicative NP in a pos-
itive sentence is a property of the subject ref-
erent, whereas a predicative NP in a negative
sentence is not.
4 Conclusion
We claim that the common practise in anno-
tation of reference chains conflates two dis-
tinct phenomena when it comes to predica-
tive NPs; i.e. reference tracking on the one
hand and tracking of predications on the other.
We acknowledge that this conflation may be
the most efficient and successful approach at
the time being, but we believe that treat-
ing these two phenomena as distinct will turn
out to be the most successful approach when
richer systems are being developed. The Nor-
wegian coreference task project BREDT (see
http://ling.uib.no/BREDT/) follows this latter
strategy.

References

Chomsky, N. 1981. Lectures on government and
binding. Foris, Dodrecht.

Gundel, J., N. Hedberg and R. Zacharski. 1993.
Cognitive status and the form of referring ex-
pressions in discourse. Language, 69:274-307.

Hirschman, L. and N. Chinchor. 1997. MUC-7
coreference task definition. MUC-7 Proceed-
ings. Science Applications International Cor-
poration. See www.muc.saic.com.

Mitkov, R. 2002. Anaphora Resolution. Pear-
son Education.

van Deemter, K. and R. Kibble. 2000. On Core-
ferring: Coreference in MUC and Related
AnnotationSchemes. Computational Linguis-
tics, 26(4).
