File Information
File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/metho/99/w99-0111_metho.xml
Size: 40,518 bytes
Last Modified: 2025-10-06 14:15:24
<?xml version="1.0" standalone="yes"?> <Paper uid="W99-0111"> <Title>Reference-based Discourse Structure for Reference Resolution</Title> <Section position="3" start_page="0" end_page="90" type="metho"> <SectionTitle> I Introduction, </SectionTitle> <Paragraph position="0"> The role of discourse structure in reference resolution, in particular pronoun and temporal reference resolution, is well recognised. However both Rhetorical Structure Theory (Mann and Thompson, 1988) and the intention-based approach to diacourse strutture associated with C~ntering Theory (GROSS and Sidner, 1986) involve high-level reamning. Such teasorting itself presupposes the availability of interpretations in which references have already been resolved. This rakes the question of the legitimacy of assuming such a discourse segmentation for the purposes of reference resolution. From a practical point of view a number of questions arise. Firstly, to what extent can We obtain the discourse structure we need for reference resolution without recourse to such higher-level reasoning? Secondly, having done our reference resolution, what further higher-level reasoning needs to be done if we are to adequately capture the meaning of the discourse, v Being able to label our discourse structure with discourse relations will only be useful insofar as these relations have implications for language understanding. Recognising an e.~hnafion can be seen to contribute to our understanding in concrete ways. It allows us to infer, for instance, the fact that one event caused another together with the fact that the first event preceded the second. Given that this information is implicit in the discourse, it is clear that our understanding would be impaired without it. With other discourse relations, however, the contribution to inference and so discourse understanding is le~ dear.</Paragraph> <Paragraph position="1"> We have implemented s unified approach to reference resolution and discourse structure in the systern of language understanding described in (Ramsay, 1999). Having used the existing discourse structure to remlve references in the utterance and anchor it in the discourse, we then use information about continuity of reference to attach the new discourse state generated by the utterance to the discourse structure, or discourse tree. Attachment is by means of a number of interchangeable attachment rules which we have used to explore how different referential cues contribute to discourse structure (Seville, 1999). The kinds of r~er~tial information we use include: ~eme I (as this term is used by (Hall/day, 1985)), since this is what the dismurse at any one point is about; reference ~/me ~, since sometimes discourse structure reflects the temporal structure of the events described; and pronommali~ation in general, since this recognisably contributes to the overall coherence of the discourse. There are further referential cues we have yet to explore, including adverbiais like &quot;Then&quot; and &quot;Before that&quot; and bridging descriptions like those in the following examples: (1) John I/~ Shirley.</Paragraph> <Paragraph position="2"> (2) H~ mo~er doesn't.</Paragraph> <Paragraph position="3"> (I) The house was grand.</Paragraph> <Paragraph position="4"> (2) The door was carved with gilded cherubs.</Paragraph> <Paragraph position="5"> 2 Tim is ~ailar in many ways to the center.</Paragraph> <Paragraph position="6"> 2Asain, this is ,;,,,;lax to the notion of tmpomi center (Kameyama, Pa~onean, and Poeslo, 1993), but with important differences; for instance, re\]crones time is an instant, while the tempond ~ is an interval.</Paragraph> <Paragraph position="7"> The way we evaluate the discourse trees generated by the different attachment rules is with respect to two main criteria. The first of these is pronominalisation. An adequate discourse structure should enable us to do long-distance pronominalisation, and, equally, prevent us from obtaining an antecedent in those cases where the referent would normally not be pronominalised; for example: - (1) Maryi loves Bill.</Paragraph> <Paragraph position="8"> (2) He loves John.</Paragraph> <Paragraph position="9"> (3) John loves Peter.</Paragraph> <Paragraph position="10"> ? (4) Peter loves her~. '&quot; Our second Criterion is discourse relations. While the discourse trees we generate for reference resolution are un\]abelled, we require that they get the attachments or d/sceurse 0wesd right, so that, in inferring how events are related, we consider only those events which are actually connected in the discourse. Below, we describe our approach to reference resolution and discourse structure building. In the case of reference resolution we are concerned with a wide range of referring expressions, including anaphors, pronouns, referential auxiliaries like &quot;had&quot;, names and definite descriptions (Seville and Ramsay, 1999). Of particular concern is the relationship between short, non-unique NPs like &quot;the man&quot; and both pronouns like &quot;he ~ and longer NPs like &quot;the man in the moon&quot;- Given that both pronouns and short NPs like &quot;the man&quot; are used to refer to familiar referents, in context, an obvious strategy is to exploit the discourse structure we have available to us for the resolution of the latter ks well as the former.</Paragraph> <Paragraph position="11"> However, short NPs may be used to refer to entities for which a pronominal reference is no longer available: (1) Mary loves Bill.</Paragraph> <Paragraph position="12"> (2) He loves John.</Paragraph> <Paragraph position="13"> (3) John loves Peter.</Paragraph> <Paragraph position="14"> (4) Peter loves Mary/,he~.</Paragraph> <Paragraph position="15"> This suggests a functional division between pronouns and names/definite descriptions, with the formet, but not the latter, relying on the discourse structure for reference resolution. Full NPs can, of course, also be used to refer to entities which are pronominalisable, in the appropriate circumstances: (1) John~ loved Bill~. (1) John~ loved B'dl.</Paragraph> <Paragraph position="16"> (2) He~ adored hires. (2) Bill adored hin~.</Paragraph> <Paragraph position="17"> Given that we are not treating defmite descriptions as dependent on the discourse structure for their reference, we still have to consider how a short NP like &quot;the man&quot; may successfully refer when there are apparently several potential referents in the context: &quot;Alex~ was the man~ conducting the interviews. He~ arrived on time but couldn't find the entrance. A manj showed him/ the way and he~ settled into his office just in time for the first candidate. A mant walked into the room. Alex~ asked himt to sit down. TAe mant pulled out a gun.</Paragraph> <Paragraph position="18"> Alex~ recognised that her was the rnanj who had given hin~ directions.&quot;</Paragraph> </Section> <Section position="4" start_page="90" end_page="93" type="metho"> <SectionTitle> 2 Reference Resolution </SectionTitle> <Paragraph position="0"> For each utterance input to the system of language understanding, we obtain a logical representation; for example, the following is the representation obtained for &quot;John loved Mary&quot;:</Paragraph> <Paragraph position="2"> All referring expressions are represented in this as terms of the form ref(AzP=). Temporal expressions, or tense, also give rise to such terms. We use the inference engine described in (Ramsay, 1995; Cryan and Ramsay, 1997) to anchor the utterance in the common ground and update the common ground with its entailments. Before our representation is input to the theorem prover, it is converted to Skolemi-~ed normal form and then to sequent form.</Paragraph> <Paragraph position="4"> The order of the terms refO, B(named(B, John)))) and re.f(AB(named(B, Ma~ll)))) in these retains the ordering of &quot;John&quot; and =Mar~ in the utterance, and this is the order in which the terms will be dereferenced. Retaining the order of referring expressions is obvions\]y important for dealing with examples like &quot;John loves hlm~.', but alto for deter~inlng the relative salience of referents for the dereferencing of subsequent pronominals.</Paragraph> <Paragraph position="5"> In anchoring referring expressions, we treat each reference term as carrying with it an invitation to carry out an existence proof in the common ground.</Paragraph> <Paragraph position="6"> The nature of the proof is retie~ted in the properties of the reference term. Names and definite descriptions have no special properties, as they don't bring with them specific instructions as to where the referent is to be located in the discourse structure. We discuss how we deal with these below. Anaphors are characterised as salient and pronouns as centred, in the current discourse state (e.g., 4):.</Paragraph> <Paragraph position="7"> &quot;himself&quot; re/(~X(salient(Xo 4)&:re(X))) &quot;him&quot; re.f(AX(cerdred(X, 4)~m(X))) In order to dereference these terms, we need to consider the discourse structure. In the case of anaphors0 we consider the referents already mentioned in the new discourse state being constructed for the utterance currently being processed. In the case of pronouns, we consider all open nodes in the discourse tree we have constructed so far, i.e. those on its right-hand frontier:</Paragraph> <Paragraph position="9"> (I) John loved May.</Paragraph> <Paragraph position="10"> (2) He had hated her.</Paragraph> <Paragraph position="11"> (3) She had hated him.</Paragraph> <Paragraph position="12"> Each of these nodes represents a discourse state corresponding to an utterance and has assodated with it the list of referents mentioned in that utterance. These include, for example, in the case of &quot;She had hated him&quot;: 3, the speech t/me; #72, the reference time; #76, the event time; #75, the event itself; #47, Mary; and #46, John. As well as pronouns, temporal references are dereferenced ~ the open nodes in the discourse structure, as detailed below.</Paragraph> <Section position="1" start_page="91" end_page="91" type="sub_section"> <SectionTitle> 2.1 .Anaphors </SectionTitle> <Paragraph position="0"> As stated above, we characterise the referents of anaphors as salient. This means that we must find their referents in the current discourse state. The other constraint we place on anaphors is that their referents play the role of arfumenf& This rules out cases li~e- .</Paragraph> <Paragraph position="1"> * &quot;Mary~ took John with/~rsei/i.&quot; While anapher resolution is not dependent on dis-course structure, we discuss it briefly here mainly for the purposes of comparison with pronoun resolution, which is heavily dependent on discourse strutture. Consider the following example: (1) John~ loves Bill i.</Paragraph> <Paragraph position="2"> (2) He~ adores him~.~.</Paragraph> <Paragraph position="3"> In dereferencing &quot;himeeli ~ in (2), the reference term to be considered is re/(AB(salient(B, 2)&re(B)))). In dereferencing this we consider, in the order in which they were mentioned and dealt with by the inference engine, the entities already added to the current discourse state: \[2,#ts,#4o\] These are speech time (2), the event itself (#78), and John (#46). Of these, we can prove that #46 satisfies the property m, which we use to represent male gender. It also meets the constraint of being an argument, so our reference term is dereferenced to it.</Paragraph> </Section> <Section position="2" start_page="91" end_page="92" type="sub_section"> <SectionTitle> 2.2 Pronouns </SectionTitle> <Paragraph position="0"> In the case of pronouns we consider two cases, that where the referent functions as an argument and that where it doesn'L In the following case, &quot;him&quot; plays the role of argument: (1) Jol~ loves Bill~.</Paragraph> <Paragraph position="1"> (2) He~ adores himj.</Paragraph> <Paragraph position="2"> What this means is that, in order to ensure that the constraint of dbjoint reference (i.e., between pronouns and the antecedents of anaphors) is met, we must start our search for the centred referent of,ef(AB(centred(B, 2)&re(B))) not in the current discourse state, but in the previous discourse state, represented by the lowermmt open node in the discourse tree. E}imlnating from consideration any of its referents which are also current entities, i.e., #46, this leaves us with: \[I,#80,#38\] These entities are, respectively, speech time, the event, and Bill. All of these are considered, in order, and we find that #38, i.e., Bill, matches the referring expression. &quot;:-In the following example, &quot;her TM does not refer to an argument, and we must begin the search for the referent of the corresponding term l'ey(AB(cenh'ed(B, 1)&f(B)))) not in the previous, but the current discourse state: (1) Maw~ took John with her~.</Paragraph> <Paragraph position="3"> Here, the t~ferents mentioned so far include: \[1, #82, #81, #47, #4el ofthese, we can prove that #47, i.e., Mary, satisfies the property jr, for female gender, and m our term is sueeeeefully dereferenced to this. We deal with the following cataphoric example in exactly the same way: %Vith heft Maryi took John.&quot; Because we treat &quot;With her&quot; as an instance of left extrapmition, we get e~Actly the same logical form for this utterance as for that above. It is this which determines the order in which referring expressions are dereferenced andso, in this case, as in that above, &quot;Mary&quot; is dereferenced before, and so provides the antecedent for, &quot;her &quot;s. Where we fail to find an antecedent for a pronoun in the previous discourse state, our search considers the next open node in the discourse tree. This is not necessarily the immediately previous discourse state, as illustrated by the following example4: (1) a man died in a park.</Paragraph> <Paragraph position="4"> (2) he had been sleeping there.</Paragraph> <Paragraph position="5"> (3) awoman loved him.</Paragraph> <Paragraph position="6"> (4) she had hated him.</Paragraph> <Paragraph position="7"> (5) he had hated himself.</Paragraph> <Paragraph position="8"> Here (3) is the next open node to be considered after (5), since (4) was closed by the attachment of the latter. It is in the list of referents mentioned in (3) that we are able to find the referent of &quot;her&quot; given the following ~ntinuation: (6) he had loved her.</Paragraph> </Section> <Section position="3" start_page="92" end_page="92" type="sub_section"> <SectionTitle> 2.3 Reference Times </SectionTitle> <Paragraph position="0"> For the purposes of reference resolution, we treat referential tenses similarly to pronouns. As an example, consider utterance (5) above. The logical form we obtain corresponding to th~ is ss follows:</Paragraph> <Paragraph position="2"> This =contains a reference to a centred time before speech time: ,'e l( AC( ,~,-~,,~( c) ) ) > B) ) ~The~z s~re fiuther cases of cataphors which we don't attempt to deal with here. These differ from the shove in that the~ is no \]eft extrapomtion of p~onoun~ont~ining constituests. Instead, the pronomin~dised items are marked as nero by being ~-eued, which is what distin. guisha them from unmarked, anaphoric uses (Hdliday and Huan, 1976); for ~*m_ple: &quot;THIS is wh~t worries me: I c~'t get ~y xeli,ble INFORMATION.&quot; Dealing with such examples seems to require & theory of focus, and so ;- beyond the scope ofthi~ paper.</Paragraph> <Paragraph position="3"> ~This w~- built m/ng the cue of pronom;n~lisation. as described in the following section.</Paragraph> <Paragraph position="4"> When we are anchoring (5), the open nodes in the discourse tree are (4), (3), and (1). The entities mentioned in discourse state (4), the first open node to be considered, are as follows: \[4, #478, #480, #479, #476, #477\] These correspond to speech time, the reference time (i.e., the event time of (3)), the event time, the event itself, the woman, and the man. Our temporal ref* erence term is dereferenced to the first centred time which we can prove is before speech time s. This is .... not #480, the time of (4)'s event, but #478, the time of (3)'s event, which precedes it in the list of entities mentioned in (4). As will be apparent below; this choice of reference time has important implications for the discourse structure we build.</Paragraph> </Section> <Section position="4" start_page="92" end_page="93" type="sub_section"> <SectionTitle> 2.4 Names and Definite Descriptions </SectionTitle> <Paragraph position="0"> There are no reliable surface cues for distinguishing short, non-unique NPs used to refer to familiar referents and longer NPs which may denote simply by virtue of their meaning. We therefore need a unified approach to dereferencing the reference terms we obtain for both. As mentioned above, these are all of the form, re1(A-Pz), but, unlike in the case of anaphors and pronouns, incorporate no properties character/sing where in the discourse structure their referents are to be found: &quot;the man&quot; reJ~O~Y(man(X))) In finding our referent we have no recourse to dynamie properties of entities like salient and centre~ but only static properties like man. However, we still need to distinguish more and less prominent poten, tisl referents d a short NP like &quot;the man'. The way in which we do th~ is to co~der entities mentioned in the discourse in order of recency. Given a referring expression containing a property such as AX(man(X)), we prove that there is an entity sat~/ying the proper~y. In this way we obtain either an entity which has been explicitly mentioned or, in the case of bridging descriptions, one associated with such an entity but which has not itself been expl/citly mentioned. In the cue of a bridging description, we obtain the proof via a meaning postulate, such as the following:</Paragraph> <Paragraph position="2"> Due to the ordering of facts and meaning postulates in the database, we tend to obtain the most recently mentioned entity or, in the case of bddg/ng descriptions, that with the most recent antecedent, first.</Paragraph> <Paragraph position="3"> The following example serves to illustrate: awe ignore previous speech times from the point of view of subsequent de~ferencing.</Paragraph> <Paragraph position="4"> (I) Shirley arrived at the house~.</Paragraph> <Paragraph position="5"> (2) John opened the door/( 0.</Paragraph> <Paragraph position="6"> In resolving the reference to &quot;the door&quot;, we ohtain a proof that the entity #88(#123) satisfies this property. Although this entity has not previously been mentioned, the house s , #123, has been recently mentioned, and it is this, together with the meaning postulate above, which forms the basis of our proof. In order to recognise cases of ambiguity, rather than simply find the single most recent entity satisfying the referring expression, we simultaneously find all entities satisfying the property which are equally recent.</Paragraph> <Paragraph position="7"> (I) The cart arrived at the housej.</Paragraph> <Paragraph position="8"> (2) John opened the door/(~/j).</Paragraph> <Paragraph position="9"> Our default criterion of equal recency is having been mentioned, or having an antecedent, within the same discourse state. This means that we recognise the ambiguity in the example above. However, given this criterion, we do not detect any ambiguity in the following exampleT: (I) Shirley got in the cari.</Paragraph> <Paragraph position="10"> (1) She arrived at the housej.</Paragraph> <Paragraph position="11"> (2) John opened the door/0 ) .</Paragraph> <Paragraph position="12"> . Where no entity of the kind has been mentioned, our reference resolution procedure is exactly the same, with all potential referents being considered equally recent. We will obtain a unique referent in this case given a definite description, like ~the )Yr~ man on the moon&quot;, which denotes such a referent by virtue of its meaning.</Paragraph> <Paragraph position="13"> Because the referents of NPs needn't have been mentioned and don't have to satisfy the property'!: * centred, we are able to obtain referents for names and definite descriptions in cases where a pronominal reference would have failed. However, the reverse situation is also poaible~ In the following example, &quot;it&quot; dearly picks out the red car, whereas &quot;the car' doesn't seem an appropriate way of referring to it. (1) Mary drive, a red carl.</Paragraph> <Paragraph position="14"> eThis was also a fizst mention. In this case. the referent .was accommodated.</Paragraph> <Paragraph position="15"> TThis scc--m approps~te in th/s cue: (I) Shirley Sot in the cart.</Paragraph> <Paragraph position="16"> (1) She arzived at the housej.</Paragraph> <Paragraph position="17"> (2a) John opened the doorj(#) to let herin.</Paragraph> <Paragraph position="18"> ? (2b) John opened the door;( 0 to let her out.. However, in other cases, it does seem that we need to consider entities other than those w~thin the same discourse state to be equally, recent. Discourse structure may be a factor, and this is something we may be able to capture. However, other factors which are harder to track, like discourse topic, also seem to be involved. (2) She hates iti/the car~.</Paragraph> <Paragraph position="19"> Our explanation of this is that, in the same way that we must use an anaphor rather than s pronoun when an anaphor is available (i.e., for arguments within the same clause), there is a convention whereby we normally use a pronoun rather than a definite description when this option is open to us a . It is possible to use a full definite description in such circumstances, but there should be a reason for such a choice of referring expression, Of which discourse structure is the obvious candidate, since this may well be affected in such a case, as we will show below; We define normal use as follows. A pronoun should be used to realise any entity mentioned in the predous discourse state, except in those cases where this entity is being promoted over a more salient referent with the same pronominal properties. According to this definition, &quot;Bill&quot; in (2) below is a normal use of a full NP, since &quot;He&quot; would have been dereferenced to John, over whom Bill is being promoted.</Paragraph> <Paragraph position="20"> (1) John loves Bill. (I) John loves Mary.</Paragraph> <Paragraph position="21"> (2) Bill loves him. ? (2) Mary loves him.</Paragraph> <Paragraph position="22"> However, the use of &quot;Mary ~ is regarded as exceptional. As such, it is likely to be interpreted as conveying some implicature (Orice, 1975). Given some associated discourse move (i.e., a further convention associated with the exception), the use of &quot;Mary&quot; can be regarded as felic'ito~ (Austin, 1962). However, in the absence of any reason for the exceptional choice of referring expr_~__m_'on, it may be regarded as iafelicitous. While we have not associated any discourse moves with the use of definite descriptions, we do find such moves eme~hg in the work described below. This happens because we treat pronouns, but not names and definite descriptions, as anchored to some particular discourse state, and this is one kind of information used in building the subsequent discourse structure.</Paragraph> </Section> </Section> <Section position="5" start_page="93" end_page="456" type="metho"> <SectionTitle> 3 Building the Discourse Structure </SectionTitle> <Paragraph position="0"> Once we have anchored an utterance and created a new discourse state corresponding to it, we-use information about continuity of reference to attach it to the existing discourse tree. Since there are a variety of referential cues and ways of using (and combining) them to be considered, this attachment information is represented in a variety of interchangeable rules.</Paragraph> <Paragraph position="1"> The referential cues explored so far include: 8This is not merely an orb/tm~ convention. The use of a pronoun assures us that we are continuing to talk about the same referent in circumstances where a name or definite description might signal a change of referent. Here, we briefly consider how useful the different ref~entiel cues seem to be, by comparing the results of some of the more succesdul attachment rules.</Paragraph> <Section position="1" start_page="94" end_page="94" type="sub_section"> <SectionTitle> 3.1 Theme </SectionTitle> <Paragraph position="0"> The tree below was constructed using a very simple attachment rule whereby a discourse state which continued the existing theme was attached to its predeeeuor as a sister, while one which changed the theme was instead attached as a daughter s.</Paragraph> <Paragraph position="2"> (1) a man died in a park.</Paragraph> <Paragraph position="3"> (2) he had been sleeping there.</Paragraph> <Paragraph position="4"> (3) a woman loved him.</Paragraph> <Paragraph position="5"> (4) she had hated him.</Paragraph> <Paragraph position="6"> (5) he had hated hin~Jf.</Paragraph> <Paragraph position="7"> (6) he had loved her.</Paragraph> <Paragraph position="8"> Of the rules we considered which ut'dised theme, this one was relatively suer~seful, supporti~ the long-distance reference to &quot;he~' in (6). It can be seen that this is because the open nodes in the tree serve to keep track of any previous themes. However, theme, taken by itself, does not seem to provide a good guide to discourse structure. The discourse structure We obtain h~re is, in elect, linear. Alternative theme-based attachment rules did result in ~We distinguish between dize~ly attaching a2 as a sister of al and attaching it as a daughter of*ts's mother. In both cases nl would no longer be open either for dereferencing pronouns or as an attachment site for future nodes. The difference becomes apparent when we start labellinlg the tree with discourse relations, a matter which iJ not discussed in the current paper.</Paragraph> <Paragraph position="9"> more interesting structures. For instance, (5) could be treated as resuming the theme of (2), and so attached there; however, in this case the reference to &quot;her&quot; in (6) would fail due to the closure of nodes (3) and (4). A further alternative involves treating smooth transitions from mention in the rheme to mention as theme as cases of thematic continuity (Hahn and Strube, 1997). We found this approach to result in some trees which were useful for pronoun resolution but which, again, didn't seem to reflect the discourse structure.</Paragraph> </Section> <Section position="2" start_page="94" end_page="94" type="sub_section"> <SectionTitle> 3.2 Reference Time </SectionTitle> <Paragraph position="0"> We illustrate the cue of reference time using the same example as above. In this case, attachment of a discourse state with a referential tense was to the discourse state which tint introduced the time referred to.</Paragraph> <Paragraph position="1"> Again, as in the case above, we are able to manage the reference to &quot;her&quot; in (6). There is a difl~erence in this case, however, in that we anchor the ref~.rring expression to (3) in this case but to (4) above. Atguably this tree, which has fewer open nodes, does a better job of retaining just that information which is needed for pronoun resolution. What it clearly does better, however, is represent aspects of the discourse structure like the dependency of (2)on (1) and the parallelism of (4), (5), and (6). It can be seen to do this insofar as the temporal structure reflects the dismune structure. Where it is unsuccessful is in capturing that (3) is a continuation of (1), but then this is unsurprising since the simple past tense of (3) is existential rather than referential Ideg.</Paragraph> </Section> <Section position="3" start_page="94" end_page="456" type="sub_section"> <SectionTitle> 3.3 Pronomlnalbmtion </SectionTitle> <Paragraph position="0"> The mndts we obtain using the cue of ~ference time alone can be enhanced either by considering theme together with reference time or by. considering all pronomie~t;~-d items, which we take to include reference time. The problematic attachment of (3) above is not helped by considering its theme, since this is, like its tense, .existential rather than referential. However, if we use a pronominal/sation-based rule, whereby we attach a new discourse state as a daughter of the highest open node at which all its sdegIt refers only to the current speech time.</Paragraph> <Paragraph position="1"> pronouns can be dereferenced, then we obtain the following tree.</Paragraph> <Paragraph position="2"> (1) a man died in a park.</Paragraph> <Paragraph position="3"> (2) he had been sleeping there.</Paragraph> <Paragraph position="4"> (3) a woman loved him.</Paragraph> <Paragraph position="5"> (4) she had hated him.</Paragraph> <Paragraph position="6"> (5) he had hated himself.</Paragraph> <Paragraph position="7"> (6) he had loved her.</Paragraph> <Paragraph position="8"> By considering all pronouns, we realise that (3) should not be attached higher than (1), where the referent of &quot;him&quot; was introduced. It is arguable that the attachment of (3) to (I) should have been as a sister rather than a daughter, thereby closing (I) to future pronoun resolution and attachment, but this is a distinction which is hard to capture.</Paragraph> </Section> </Section> <Section position="6" start_page="456" end_page="456" type="metho"> <SectionTitle> 4 Discussion </SectionTitle> <Paragraph position="0"> By experimenting with a variety of interchangeable attachment rules we have found that, at least in those cases where it is available, temporal reference provides a reliable constrmnt on discourse structure. Where it is not available, tree-building can be guided by other cues such as continuity of theme and pronominal/sation. However, nnllke reference time, these are heuristics rather than constraints and in some crees can be unreliable. In the first example below, the theme, aShe', of (3) seems to cue attachmdnt to (I), where this referring expr ~,~on is anchored, but (3) is, rather, an elaboration of (2).</Paragraph> <Paragraph position="1"> In the second example, 0) is a continuation of (I) even though its theme, aHe n, triers to John who was not even mentioned anti\] discourse state (2).</Paragraph> <Paragraph position="2"> (1) Sam arrived at Hmmal~'s house.</Paragraph> <Paragraph position="3"> (2) He had lost the key.</Paragraph> <Paragraph position="4"> (3) She~ would be annoyed with him.</Paragraph> <Paragraph position="5"> (I) Mary ran.</Paragraph> <Paragraph position="6"> * (2) 3ob~ had seen her.</Paragraph> <Paragraph position="7"> (3) H~ ~ her.</Paragraph> <Paragraph position="8"> While reference time is of i/mited appficabi\]ity, we do not regard this u a problem since we can recoguise precisely those cases in which it is applicable. There is one other kind of cue which we have found to be useful, and a~ain this only applies to a particular set of cases. This is the use ors full NP to refer to an already familiar referent, particularly the current theme, when, for the purposes of reference resolution, a pronominal reference would be perfectly adequate. While this is not a cue we explicitly set out to investigate, interestingly different discourse structures emerged from certain of the attachment rules specified, depending upon whether or not references were made pronominally. Compare the following discourse structures, both constructed using the cue of pronominalization, as outlined above. The first, with its pronominal references, can be regarded as an elaboratzon, but, if continuity is regarded ~ a pre-requisite of this relation, then the second cannot Ix.</Paragraph> <Paragraph position="10"> (1) Mary, loves John. (1) Mary loves John.</Paragraph> <Paragraph position="11"> (2) She~ adores him. (2) Mary adores John.</Paragraph> <Paragraph position="12"> This is not s result we set out to achieve. It emerged as a result of our different approaches to reference resolution for pronouns and full NPs. Recall that the former but not the latter are regarded as anchored within the discourse structure, thus providing us with potential attachment cues. In constructing the first of the trees above, the anchoring of the pronouns in (2) to discourse state (I) provides the basis for the attachment made. There are no alternative rules for full NPs in the second example, however; rather, attachment simply fails to be made to (1) becanse there is no evidence of a discourse connection. null The di~erent discourse structures we obtain given pronomlnal and non-pronominal zeferences to the * same thematic entity suggest the importance of the choice of referring expression to discourse structure. One of our goals is to use referring expressions to recognise the discourse moves associated with them.</Paragraph> <Paragraph position="13"> The following examples serve to illustrate: (1) Sam/arrived at the house.</Paragraph> <Paragraph position="14"> (2) He~ had lost the key.</Paragraph> <Paragraph position="15"> (3) He~/Sann rang the bell.</Paragraph> <Paragraph position="16"> (1) San~ arrived at the house.</Paragraph> <Paragraph position="17"> (2) He~ had lost the key.</Paragraph> <Paragraph position="18"> (3) He~/?Sam? dropped it.</Paragraph> <Paragraph position="19"> In the first of these, &quot;Sam&quot; in (3) seems to be s felicitous choice of referri~ expression. This makes sense, since, in this case, (3) is not & continuation SSlf the~ were a chanse of referent, then the relation would seem to be z parallel or. contrastive one, as in: (1) Mary Io,= ~o~.</Paragraph> <Paragraph position="20"> (2) Jue adores hi,., The discourse structure we set when the (ull NP &quot;Mary&quot; is used is one assoc~ted with such s disco~, relation. However, these rel~tiou are inappropriate in this case. Rather than attempt to label the tree resulting from the use of &quot;Mary', we would prefer to resard this choice of referring expression as infelicitous.</Paragraph> <Paragraph position="21"> of (2). In the second example, however, &quot;Sam&quot; does not seem to be available as a choice of referring expression. Again this makes sense insofar as (3) is here a continuation of (2). The cue of pronominalization seems to indicate continuity. Where it could have been used but isn't, the deliberate choice of an alternative referring expression serves to traplicate (Grice, 1975) the absence of continuity. However, our rules as currently implemented don't really capture this explanation. Given a non-pronominal reference to the current theme, there may well be a lack of evidence of continuity, so we avoid the attachment of (3) to (2), but we may also fail to attach (3) to (I)! One way of dea/ing with such cases would be to introduce a new kind of attachment rule. All the rules explored so far rely on choceing a preferred node for attachment. These cases might be better dealt with by negative constraints on attachment. Given a default attachment of (3) to (2) z~, this attachment will be prevented in these cases where (3)'s theme is not pronominalised (but could have been), thereby resulting in the alternative attachment of (3) to (I). Having identified two kinds of referential cues which seem to provide useful guides to discourse structure in particular eases, it is worth considering what problems remain with these. Firstly, we consider temporal reference as a cue to d/scourse structure. We have argued that this provides a relatively reliable constraint. However, this assumes that we resolve our temporal references appropriately in the first place.</Paragraph> <Paragraph position="22"> In this fragnent from the discourse used above, we attach (5) to (3) only becausewe first dereferenced its reference time to the event time of (3). Now consider the alternative continuation: (5) he had beaten her for years.</Paragraph> <Paragraph position="23"> In this case, we would also attach (5) to (3) because again its r~erence time would be preferentially dereferenced to the event time of (3)aS, rather than that of (4), but in this case we get it wrong. The mistake is not necessarily one which is irrecoverable. If there were no sensible discourse relation to be inferred between (5,) and (3), as in this case, then the mistake could be recognised and both the temporal reference and the attachment revised. However, this Z2By this we mean only that con~deration of (2) as a potential attachment node precede, con~deration of (1). &quot; 13A similar preference would be made by other approaches such as Tempora/Centering (Kameyama, Pussons,u, and Poesio0 1993).</Paragraph> <Paragraph position="24"> example does serve to illustrate the problems with in effect using temporal structure em a cue to discourse structure, when perhaps the two are better treated as mutually constraznlng.</Paragraph> <Paragraph position="25"> In exploiting the distinction between pronominal and non-pronominal references to the current theme we find a related, albeit less serious, problem to that above. In order to exploit the choice of referring expression in building discourse structure, we need to know that it is felicitous. However, whether it is felicitous itself depends on the discourse structure: Note that, in the examples discussed above, discourse states (I) and (2) taken together provide identical contexts for the dereferencing of the referring expressions &quot;He/Sam&quot; in (3). The only differonce is the discourse thread, which joins (3) to (I) in the first case, but to (2) in the second case. Since this is not the kind of information we have available to us when we are doing our reference resolution, we simply haveto ~s*,me that the choice of referring expression is felicitous. However, this assumption could be retracted were we to fail to label the discourse structure constructed. In this example, &quot;Sam arrived at the house ...Sam dropped it \[the key\].&quot; can be rejected ass narrative sequence, because Sam can't drop the key he has lost x4.</Paragraph> </Section> <Section position="7" start_page="456" end_page="456" type="metho"> <SectionTitle> 5 Further Work * </SectionTitle> <Paragraph position="0"> R~erential continuity is a complex phenomenon, and in considering the impact of reference on discourse structure there are many more examples of reference to be considered than those diso,~d here.</Paragraph> <Paragraph position="1"> These include referential adverbials which, as an additional source of temporal reference to tense, can be expected to further constrain the discourse structure. Some examples of these are discmmed in (Seville, 199g). Another important category of referential expressions are bridging descriptions. While pronominali~tion is a key indicator of referential continuity where it is available, where it is not, brid~Ong descriptions seem to provide a similar degree of referential continuity: (I) The hous~ was grand.</Paragraph> <Paragraph position="2"> (2a) Ik was in the baroque style.</Paragraph> <Paragraph position="3"> (2b) The door/( 0 was carved with ~lded cherubs.</Paragraph> <Paragraph position="4"> s4Similazly, &quot;he had beaten her for years&quot; would be rejected as a potential ezplanat/on of &quot;a woman loved him&quot; in the example above, and, to return to a previous example: * (1) Mary loves John.</Paragraph> <Paragraph position="5"> (2) Mary adore, b;m~ could not be labelled as a coast, u. no element of contrast can be identified.</Paragraph> <Paragraph position="6"> Integrating bridging descriptions into our treatment of reference resolution was straightforward, because of the way in which the terms they refer to are dervived, via meaning postulates, from terms representing existing referents. For the same reason, recognising and exploiting this kind of referential continuity to build discourse structure should be relatively straightforward.</Paragraph> </Section> class="xml-element"></Paper>