<?xml version="1.0" standalone="yes"?>
<Paper uid="W05-1005">
  <Title>Automatically Distinguishing Literal and Figurative Usages of Highly Polysemous Verbs</Title>
  <Section position="2" start_page="0" end_page="38" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Due to a cognitive priority for concrete, easily visualizable entities, abstract notions are often expressed in terms of more familiar and concrete things and situations (Newman, 1996; Nunberg et al., 1994).</Paragraph>
    <Paragraph position="1"> This gives rise to a widespread use of metaphor in language. In particular, certain verbs easily undergo a process of metaphorization and meaning extension (e.g., Pauwels, 2000; Newman and Rice, 2004). Many such verbs refer to states or acts that are central to human experience (e.g., sit, put, give); hence, they are often both highly polysemous and highly frequent. An important class of verbs prone to metaphorization are light verbs, on which we focus in this paper.</Paragraph>
    <Paragraph position="2"> A light verb, such as give, take, or make, combines with a wide range of complements from different syntactic categories (including nouns, adjectives, and prepositions) to form a new predicate called a  light verb construction (LVC). Examples of LVCs include: 1. (a) Azin took a walk along the river.</Paragraph>
    <Paragraph position="3"> (b) Sam gave a speech to a few students.</Paragraph>
    <Paragraph position="4"> (c) Joan takes care of him when I am away.</Paragraph>
    <Paragraph position="5"> (d) They made good on their promise to win.</Paragraph>
    <Paragraph position="6"> (e) You should always take this into account.</Paragraph>
    <Paragraph position="7">  The light verb component of an LVC is &amp;quot;semantically bleached&amp;quot; to some degree; consequently, the semantic content of an LVC is assumed to be determined primarily by the complement (Butt, 2003). Nevertheless, light verbs exhibit meaning variations when combined with different complements. For example, give in give (someone) a present has a literal meaning, i.e., &amp;quot;transfer of possession&amp;quot; of a THING to a RECIPIENT. In give a speech, give has a figurative meaning: an abstract entity (a speech) is &amp;quot;transferred&amp;quot; to the audience, but no &amp;quot;possession&amp;quot; is involved. In give a groan, the notion of transfer is even further diminished.</Paragraph>
    <Paragraph position="8"> Verbs exhibiting such meaning variations are widespread in many languages. Hence, successful NLP applications--especially those requiring some degree of semantic interpretation--need to identify and treat them appropriately. While figurative uses of a light verb are indistinguishable on the surface from a literal use, this distinction is essential to a machine translation system, as Table 1 illustrates. It is therefore important to determine automatic mechanisms for distinguishing literal and figurative uses of light verbs.</Paragraph>
    <Paragraph position="9"> Moreover, in their figurative usages, light verbs tend to have similar patterns of cooccurrence with semantically similar complements (e.g., Newman, 1996). Each similar group of complement nouns can even be viewed as a possible meaning extension for a light verb. For example, in give advice, give orders, give a speech, etc., give contributes a notion of  Sentence in English Intermediate semantics Translation in French Azin gave Sam a book. (e1/give Azin a donn'e un livre `a Sam. :agent (a1/&amp;quot;Azin&amp;quot;) Azin gave a book to Sam.</Paragraph>
    <Paragraph position="11"> Azin gave the lasagna a try. (e2/give-a-try a0 try Azin a essay'e le lasagne.</Paragraph>
    <Paragraph position="12"> :agent (a1/&amp;quot;Azin&amp;quot;) Azin tried the lasagna.</Paragraph>
    <Paragraph position="13">  &amp;quot;abstract transfer&amp;quot;, while in give a groan, give a cry, give a moan, etc., give contributes a notion of &amp;quot;emission&amp;quot;. There is much debate on whether light verbs have one highly abstract (underspecified) meaning, further determined by the context, or a number of identifiable (related) subsenses (Pustejovsky, 1995; Newman, 1996). Under either view, it is important to elucidate the relation between possible interpretations of a light verb and the sets of complements it can occur with.</Paragraph>
    <Paragraph position="14"> This study is an initial investigation of techniques for the automatic discovery of meaning extensions of light verbs in English. As alluded to above, we focus on two issues: (i) the distinction of literal versus figurative usages, and (ii) the role of semantically similar classes of complements in refining the figurative meanings.</Paragraph>
    <Paragraph position="15"> In addressing the first task, we note the connection between the literal/figurative distinction and the degree to which a light verb contributes compositionally to the semantics of an expression. In Section 2, we elaborate on the syntactic properties that relate to the compositionality of light verbs, and propose a statistical measure incorporating these properties, which places light verb usages on a continuum of meaning from literal to figurative. Figure 1(a) depicts such a continuum in the semantic space of give, with the literal usages represented as the core.</Paragraph>
    <Paragraph position="16"> The second issue above relates to our long-term goal of dividing the space of figurative uses of a light verb into semantically coherent segments, as shown in Figure 1(b). Section 3 describes our hypothesis on the class-based nature of the ability of potential complements to combine with a light verb.</Paragraph>
    <Paragraph position="17"> At this point we cannot spell out the different figurative meanings of the light verb associated with such classes. We take a preliminary step in proposing a statistical measure of the acceptability of a combination of a light verb and a class of complements, and explore the extent to which this measure can reveal class-based behaviour.</Paragraph>
    <Paragraph position="18"> Subsequent sections of the paper present the corpus extraction methods for estimating our compositionality and acceptability measures, the collection of human judgments to which the measures will be compared, experimental results, and discussion.</Paragraph>
  </Section>
class="xml-element"></Paper>