<?xml version="1.0" standalone="yes"?>
<Paper uid="P85-1021">
  <Title>amp;quot;Grammatical Relations and Montague Grammar&amp;quot;,</Title>
  <Section position="4" start_page="167" end_page="169" type="metho">
    <SectionTitle>
2 Parsing
</SectionTitle>
    <Paragraph position="0"> As in the earlier GPSG system, the primary job of the parser in the HPSG system is to produce a semantics for the input sentence. This is done compositionally as the phrase structure is built, and uses only locally available information. Thus every constituent which is built syntactically has a corresponding semantics built for it at the same time, using only information available in the phrasal subtree which it immediately dominates. This locality constraint in computing the semantics for constituents is an essential characteristic of HPSG. For a more complete description of the semantic treatment used in the HPSG system see Creary and Pollard \[2\].</Paragraph>
    <Paragraph position="1"> Head-driven Active Chart Parser A crucial dilference between the HPSG system and its predecessor GPSG is the importance placed on the head constituent in HPSG. \[n HPSG it is the head constituent of a rule which carries the subcategorization information needed to build the other constituents of  the rule. Thus parsing proceeds head first through the phrase structure of a sentence, rather than left to right through the sentence string.</Paragraph>
    <Paragraph position="2"> The parser itself is a variation of an active chart parser \[4,9,8,13\], modified to permit the construction of constituents head first, instead of in left-to-right order. In order to successfully parse &amp;quot;head first&amp;quot;, an edge* must be augmented to include information about its span (i.e. its position in the string). This is necessary because heaA can appear as a middle constituent of a rule with other constituents (e.g. complements or adjuncts) on either side. Thus it is not possible to record all the requisite boundary information simply by moving a dot through the rule (as in Earley), or by keeping track of just those constituents which remain to be built (as in Winograd). An example should make this clear.</Paragraph>
    <Paragraph position="3"> Suppose as before we are confronted with the task of parsing the sentence The manager works, and again we have available the grammar rule R1. Since we are parsing in a ~head first&amp;quot; manner we must match the H constituent against some substring of the sentence.</Paragraph>
    <Paragraph position="4"> But which substring? In more conventional chart parsing algorithms which proceed left to right this is not a serious problem, since we are always guaranteed to have an anchor to the left. We simply try building the \[eftmost constituent of the rule starting at the \[eftmost position of the string, and if this succeeds we try to build the next \[eftmost constituent starting at one position to the right of wherever the previous constituent ended. However in our case we cannot ausume any such anchoring to the left, since as the example illustrates. the H is not always leftmost.</Paragraph>
    <Paragraph position="5"> The solution we have adopted in the HPSG system is to annotate each edge with information about the span of substring which it covers. In the example below the inactive edge E1 is matched against the head of rule R1, and since they unify the new active edge E2 is created with its head constituent instantiated with the feature specifications which resulted from the unification. This new edge E2 is annotated with the span of the inactive edge El. Some time later the inactive edge I,:3 is matched against the &amp;quot;np&amp;quot; constituent of our active edge E2, resulting in the new active edge E.I. The span of E4 is obtained by combining the starting position of E3 {i.e. t) with the finishing postion of E2 (i.e. 3). The point is that edges ~Lre constructed from the head out, so that at any given tame in Lhe life cycle of an edge the spanning informatiun on the edge records the span of contiguous substring which it covers.</Paragraph>
    <Paragraph position="6"> Note that in the transition from rule ill to edge 1~2 we have relabeled the constituent markers z, cl, ~nd h with the symbols ~, np, ~utd VP respectively.</Paragraph>
    <Paragraph position="7"> This is done merely a.s ~t mnemouic device to reflect the fact that once the head of the edge is found, the subcategorization information on that head (i.e. the values of the &amp;quot;SUHCAT&amp;quot; feature of the verb work.s) is An edi\[e is, Iooe~y spea&amp;ing, ,-tn inlCantiation of a nile witll ~nnle of tile \[e~urml on conlltituentll m~de ntore spm:iflC/.</Paragraph>
    <Paragraph position="8"> propagated to the other elements of the edge, thereby restricting the types of constituents with which they can be satisfied. Writing a constituent marker in upper case indicates that an inactive edge has been found to instantiate it, while a lower case (not yet found) constituent in bold face indicates that this is the next constituent which will try to be instantiated.</Paragraph>
    <Section position="1" start_page="168" end_page="169" type="sub_section">
      <SectionTitle>
Using Semantics Restrictions
</SectionTitle>
      <Paragraph position="0"> Parsing ~head first&amp;quot; offers both practical and theoretical advantages. As mentioned above, the categories of the grammatical relations subcategorized for by a particular head are encoded as the SUBCAT value of the head. Now GR's are of two distinct types: those which are ~saturated&amp;quot; (i.e. do not subcategorize for anything themselves), such as subject and objects, and those which subcategorize for a subject (i.e. controlled complements). One of the language-universal grammatical principles (the Control Agreement Principle) requires that the semantic controller of a controlled complement always be the next grammatical relation (in the order specified by the value of the SUBCAT feature of the head) after the controlled complement to combine with the head. But since the HPSG parser always finds the head of a clause first, the grammatical order of its complements, as well as their semantic roles, are always specified before the complements are found. As a consequence, semantic processing ~f constituents can be done on the fly as the constituents are found, rather than waiting until an edge has been completed. Thus semantic processing can be do.e extremely locally (constituent-to-constituent in the edge, rather than merely node-to-node in the parse tree as in Montague semantics), and therefore a parse path ,an be abandoned on semantic grounds (e.g. sortal iltconsistency) in the rniddle of constructing an edge. la this way semantics, as well as syntax, can be used to control the parsing process.</Paragraph>
      <Paragraph position="1"> Anaphora ill HPSG Another example of how parsing ~head first&amp;quot; pays oil is illustrated by the elegant technique this strategy makes possible for the binding of intr~entential a~taphors. This method allows us to assimilate cases of bound anaphora to the same general binding method used iu the HPSG system to handle other non-lexicallygoverned dependencies ~uch a.s ~ap.~, ~,ttt~ro~,t.ive pronouns, and relative pronouns. Roughly, the unbound dependencies of each type on every constituent are en- * coded as values of a,n appropriate stack-valued feature  (&amp;quot;binding feature&amp;quot;). In particular, unbound anaphors axe kept track of by two binding features, REFL (for reflexive pronouns) and BPRO \['or personal pronouns available to serve as bound anaphors. According to the Binding Inheritance Principle, all categories on binding-feature stacks which do not get bound under a particular node are inherited onto that node. Just how binding is effected depends on the type of dependency.</Paragraph>
      <Paragraph position="2"> In the case of bound anaphora, this is accomplished by merging the relevant agreement information (stored in the REFL or BPRO stack of the constituent containing the anaphor) with one of the later GR's subcategorized for by the head which governs that constituent. This has the effect of forcing the node that ultimately unifies with that GR (if any) to be the sought-after antecedent. The difference between reflexives and personal pronouns is this. The binding feature REFL is not allowed to inherit onto nodes of certain types (those with CONTROL value \[N'rRANS}, thus forcing the reflexive pronoun to become locally bound. In the case of non-reflexive pronouns, the class of possible antecedents is determined by n,,difying the subcategorization information on the hel,,l governing the pronoun so that all the subcategorized-fl~r GR's later in grammatical order than the pronoun are &amp;quot;contra-indexed&amp;quot; with the pronoun (and thereby prohibited front being its antecedent). Binding then takes place precisely as with reflexives, but somewhere higher in the tree.</Paragraph>
      <Paragraph position="3"> We illustrate this d~ttttction v, ti, kh I.~O examples.</Paragraph>
      <Paragraph position="4"> \[n sentence S I below told subcategorizes for three constituents: the subject NP Pullum, the direct object Gazdar, and the oblique object PP about himself.' Thus either PuUum or f;uzdur are po~ible antecedents of himself, but not Wasow.</Paragraph>
      <Paragraph position="5"> SI. Wasow was convinced that Pullum told Gazdar about himself.</Paragraph>
      <Paragraph position="6"> $2. Wasow persuaded Pullum to shave him.</Paragraph>
      <Paragraph position="7"> \[n sentence 52 shave subcategorizes for tile direct object NP him and an NP subject eventue.tly tilled by the constituent Pullum via control. Since the subject position is contra-indexed with tile pronoun, PuUum is blocked from serving a~ the a,tecedent. The pro,mun is eventually bound by the NP WanouJ higher up in the tree.</Paragraph>
      <Paragraph position="8"> Heuristics to Optiudze .&amp;quot;Joareh '\['he liPS(; system, based as it is upon a carefully developed hngui~tic theory, has broad expressive power. In practice, how-ver, much of this power is often not necessary. To exploit this fact the IiPSC, system u.~cs heuristics to help r,,duve the search space implicitly defined by the grammar. These heuristics allow the parser to produce an optimally ordered agenda of edges to try ba.sed on words used in tile sentence, and on constituents it has found so far.</Paragraph>
      <Paragraph position="9"> * The pNpOlltiOll tl treltt~ |eel~lttt~tllF :is a c:~se tam'king. One type of heuristic involves additional syntactic information which can be attached to rules to determine their likelihood. Such a heuristic is based on the currently intended use for the rule to which it is attached, and on the edges already available in the chart. An example of this type of heuristic is sketched below.</Paragraph>
      <Paragraph position="10"> RI. x -&gt; cl h a* Heuristic-l: Are the features of cl +QUE?  Heuristic-I encodes the fact that rule RI, when used in its incarnation as the S -- NP VP rule, is primarily intended to handle declarative sentences rather than questions. Thus if the answer to Heuristic-1 is &amp;quot;no&amp;quot; then this edge is given a higher ranking than if the answer is &amp;quot;yes&amp;quot;. This heuristic, taken together with others, determines the rank of the edge instantiated from this rule, which in turn determines the order in which edges will be tried. The result in this case is that for a sentence such as 53 below, the system will prefer the reading for which an appropriate answer is &amp;quot;.a character in a play by Shakespeare&amp;quot;, over the reading which has as a felicitous answer &amp;quot;Richard Burton&amp;quot;. S3. Who is Hamlet? It should be empha~sized, however, that heuristics are not an essential part of the system, as are the feature passing principles, but rather are used only for reasons of efficiency. In theory all possible constituents permitted by the grammar will be found eventually with or without heuristics. The heuristics siu,ply help a linguist tell the parser which readings are most likely, and which parsing strategies are usually most fruitful, thereby allowing the parser to construct the most likely reading first. We believe that this clearly diifereutiares \[IPSG from &amp;quot;ad hoc&amp;quot; systems which do not make sharp the distinction between theoretical principle and heuristic guideline, and that this distinction is an iznportant one if the natural language understanding programs of today are to be of any use to the natural language programs and theories of the future.</Paragraph>
    </Section>
  </Section>
  <Section position="5" start_page="169" end_page="170" type="metho">
    <SectionTitle>
ACKO WLED(4EME1NTS
</SectionTitle>
    <Paragraph position="0"> We would like to acknowledge the valuable assitance of Thomas Wasow ~md Ivan Sag ht tile writing of this paper. We would also like to thank Martin Kay and Stuart Shi.e~:r lot tlke~r tte\[p\[u\[ cuttutteut, aly on an earlier draft.</Paragraph>
  </Section>
class="xml-element"></Paper>