File Information
File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/intro/04/w04-0216_intro.xml
Size: 6,700 bytes
Last Modified: 2025-10-06 14:02:26
<?xml version="1.0" standalone="yes"?> <Paper uid="W04-0216"> <Title>Animacy Encoding in English: why and how</Title> <Section position="3" start_page="0" end_page="0" type="intro"> <SectionTitle> 2 The animacy dimension in natural </SectionTitle> <Paragraph position="0"> language The animacy hierarchy is one of the accessibility scales that are hypothesized to influence the grammatical prominence that is given to the realization of entities within a discourse. Three important scales (sometimes conflated into one, also called animacy hierarchy in Silverstein, 1976), are the definiteness, the person and the animacy hierarchy proper. We assume these are three different hierarchies that refer to different aspects of entity representation within language: the definiteness dimension is linked to the status of the entity at a particular point in the discourse, the person hierarchy depends on the participants within the discourse, and the animacy status is an inherent characteristic of the entities referred to. Each of these aspects, however, orders entities on a scale that makes them more or less salient or 'accessible' when humans use their language.</Paragraph> <Paragraph position="1"> The importance of accessibility scales is not widely recognized in computational treatments of natural language. This contrasts with the situation in linguistics where such scales have been recognized as playing an important role in the organization of sentence syntax and discourse.</Paragraph> <Paragraph position="2"> Even in natural language studies, however, their importance has been underestimated because the role of these scales is not always to distinguish between grammatical and ungrammatical utterances but often that of distinguishing mainly between felicitous and infelicitous ones, especially in languages such as English.</Paragraph> <Paragraph position="3"> Grammaticality and acceptability As long as one's attention is limited to the distinction between grammatical and ungrammatical sentences, the importance of the animacy hierarchy in particular is mainly relevant for languages with a richer morphology than English. In such languages animacy distinctions can influence grammaticality of e.g. case-marking and voice selection. To give just one example, in Navaho, a bi-form is used rather than an yi-form whenever the patient is animate and the agent is inanimate, whereas the yi-form is used when the agent is animate and the patient is inanimate as illustrated in (1) (from Comrie 1989 p. 193).</Paragraph> <Paragraph position="4"> (1) (a) At'eed nimasi yi-diilid girl potato burnt The girl burnt the potato.</Paragraph> <Paragraph position="5"> (b) At'eed nimasi bi-diilid girl potato burnt The potato burnt the girl.</Paragraph> <Paragraph position="6"> Other phenomena discussed in the literature are agreement limited to animate noun phrases (Comrie, 1989), overt case-marking of subject limited to inanimates or overt case-marking of objects limited to animates (Aissen, 2003, see also Bossong 1985 and 1991), object agreement in Bantu languages (see e.g. Woolford, 1999), choice between direct and inverse voice in Menominee (Bloomfield, 1962, see also, Trommer, n.d.).</Paragraph> <Paragraph position="7"> Recent linguistic studies have highlighted the importance of the category in languages such as English. For instance the choice between the Saxon genitive and the of-genitive (Rosenbach, 2002, 2003, O'Connor et al. 2004, Leech et al. 1994), between the double NP and the prepositional dative (Cueni et al, work in progress) and between active and passive (Rosenbach et al. 2002, Bock et al. 1992, McDonald et al. 1993) and between pronominal and full noun reference (Dahl and Fraurud, 1996, based on Swedish data) have all been shown to be sensitive to the difference between animate and inanimate referents for the arguments with variable realization. In these cases the difference between animate and inanimate does not lead to a difference between a grammatical or an ungrammatical sentence as in the cases cited in the previous paragraph but to a difference in acceptability.</Paragraph> <Paragraph position="8"> Interaction between animacy and other scales As mentioned above, the term 'animacy hierarchy' is used in two ways, one to refer to an ordering imposed on definiteness, animacy and person and the other where it refers to animacy proper. The reason for this lies in the interaction between the different factors that determine accessibility.</Paragraph> <Paragraph position="9"> It is conceptually desirable to distinguish between animacy and definiteness but in practice it is frequently the case that a linguistic phenomenon is conditioned by more than one of these conceptually independent factors (see e.g. Comrie, 1989 for discussion). The projects in the context of which the annotation tasks described here were performed (see description below, section 5) also encode some of these interacting factors. The LINK-project encodes information status (see Nissim et al, 2004) and the Boston project encodes definiteness and expression type (i.e., NP form) as proxies for information status.</Paragraph> <Paragraph position="10"> 3 Animacy as a factor in generation and translation As long as animacy was discussed as a relevant grammatical category in languages that had not been studied from a computational point of view, its importance for computational linguistics was perceived as rather limited. The fact that it permeates the choice between constructions in languages such as English changes this perception. The category is of obvious importance for high quality generation and translation.</Paragraph> <Paragraph position="11"> For instance, if one is faced with the task of generating a sentence from a three place predicate, P (a,b,c), and one has the choice of rendering either b or c as the direct object, knowing that c is human and b is abstract would lead one to choose c ceteris paribus. However, everything else is rarely equal and the challenge for generation will be to assign the exact relative weights to factors such as animacy, person distinction and recency.</Paragraph> <Paragraph position="12"> Moreover, the importance of these factors needs to be combined with that of heterogeneous considerations such as the length of the resulting expression.</Paragraph> <Paragraph position="13"> In the context of translation we need also to keep in mind the possibility that the details of the animacy ranking might be different from language to language and that the relative impact of animacy and other accessibility scale factors might be different from construction to construction.</Paragraph> </Section> class="xml-element"></Paper>