Antecedent (grammar)

About: Antecedent (grammar) is a research topic. Over the lifetime, 1392 publications have been published within this topic receiving 41824 citations.

An organizational communication‐based model of individual customer orientation of nonmarketing members of a firm

Abstract: A model of individual customer orientation is proposed and tested Building on past research from the organizational communication and marketing literatures, the model (which incorporates facets of the organizational environment, interfunctional dynamics, and communication practices) is tested to examine the impact of these antecedent variables on customer‐oriented outcomes of nonmarketing members of the firm Results indicate that highly valued interfunctional communications by marketing results in nonmarketing members being more customer‐oriented and more willing to engage in behaviors designed to satisfy internal customers Managerial implications are discussed along with directions for further research

Pragmatic principles in anaphora resolution at the syntax-discourse interface

Abstract: Previous experimental research has shown that the syntax-discourse interface can be a locus of deficits for learners, even at very advanced levels. In this paper an L2 Spanish corpus is used to investigate Anaphora Resolution (AR) at the syntax-discourse interface (i.e., how null/overt pronouns and NP subjects refer to their antecedents in discourse). A fine-grained tagset was designed to annotate formal, pragmatic and information-status AR factors in a sample of very advanced L1 English – L2 Spanish learners vs. a Spanish native control subcorpus from the CEDEL2 corpus. The corpus analysis results reveal that, though very advanced learners can attain similar patterns to Spanish natives with AR, they show certain deficits: they are pragmatically more redundant than ambiguous. This is explained in terms of a new proposal: the Pragmatic Principles Violation Hypothesis. Results also show that the syntax of topic shift is more complex than previously assumed, with higher rates of NPs than overt pronouns. This can be accounted for by the nature of the antecedent (number of potential antecedents and their gender differences).

What does he mean

Abstract: What does he mean? Maria Wolters David Beaver Rhetorical Systems Ltd., Department of Linguistics, Margaret Jacks Hall 4 Buccleuch Place, Edinburgh EH8 9LX, Scotland Stanford University 94305 Stanford, CA, USA maria@rhetoricalsystems.com dib@stanford.edu Abstract Empirical results on the meaning of accented pronouns often conflict. This is a problem for formal semantic models. In this paper, we intend to broaden the empiri- cal basis of two of these models. First, in a corpus study, we checked whether properties of the antecedent influ- ence whether a pronoun is be accented. We found that pronouns with NP antecedents are more likely to be ac- cented than those with pronominal antecedents. In a pro- duction experiment, we investigate whether accented pro- nouns signal topic shifts. Although this effect is present in our data, it is very weak. We conclude that a comprehen- sive model of pronoun accentuation will need to account for the fact that in most cases, that accent appears to be optional. In fact, most instances of accented pronouns in our data could be explained as signs of a rhetorical con- trast. Introduction In order to interpret a personal pronoun, a listener needs to determine which discourse entity or entities the pro- noun specifies. Since pronouns themselves carry very lit- tle semantic information, the listener needs to tap into a variety of information sources in order to find out which discourse entity a given pronoun specify: lists of salient discourse entities, grammatical conventions, discourse structure, assumptions about the discourse model of the speaker, and so on. Formal work on pronoun interpretation has focussed on written language. In this paper, we explore how the insights gained so far can be translated to speech, where phrasing and accentuation might provide important cues to resolution algorithms. Most semanticists take accent on pronouns to signal somewhat “unusual” resolution strategies. In particular, accented pronouns are assumed to signal topic shifts. But do the theories developed so far describe successfully how accented pronouns actually used in speech? This is the question we ask here. Our paper is organised as follows: First, we sketch the theoretical basis of our analyses, the work of [Nakatani, 1997] and [Kameyama, 1997]. Then, we ex- plore to what extent the patterns postulated by the theo- ries can indeed be found in corpus data and experimental data. Finally, we discuss the consequences of our results for semantic and cognitive approaches to resolving ac- cented pronouns. Theoretical Claims It is not clear what exactly speakers mean when they accent a third person personal pronoun. Some peo- ple apparently accent a pronoun if the distance in clauses between pronoun and antecedent is larger than usual [Giv´on, 1983]. A number of sample discourses that have been discussed extensively in the literature [Kameyama, 1997, Beaver, 2000] show that when sub- stituted for their unaccented counterpart in a given dis- course, accented pronouns are frequently resolved to a different discourse entity. In the following example, most listeners would resolve the unaccented pronoun in (2) to Ian, whereas the accented pronoun in (3) would be resolved to James. (1) Ian i often meets James j for dinner. (2) He i prefers Italian restaurants. (3) HE j prefers Italian restaurants. We will now examine two theoretical analyses of ac- cented pronouns. Nakatani: Shifts in Attentional State Christine Nakatani develops a model of stressed pro- nouns in terms of attentional state [Nakatani, 1997]. Her framework is Centering Theory [Grosz et al., 1995]. The goal of Centering is to develop a theory of local discourse structure, i.e. to describe what makes a discourse seg- ment coherent. From a psycholinguistic point of view, Centering models how the attentional state of speaker and hearer change during a discourse. Each attentional state contains a set of discourse entities which are the current “centers” of attention – hence the name “Center- ing”. In the Centering model, each sentence is associated with a list of the discourse entities which have been re- alised in that sentence. These discourse entities consti- tute the list of forward-looking centers. The forward- looking centers are ranked according to their salience. The most salient center on the list is called the preferred forward-looking center C p . The most salient entity of the previous utterance U n−1 that is realised in the current utterance U n is the backward-looking center C b . Each sentence has at most one C b . Transitions between sen- tences can be classified according to two criteria: 1. whether the backward-looking center of the current sentence is the same as that of the previous sentence

Pictures and anaphora: Evidence for independent processes

Abstract: Pictures enhance our comprehension of written texts, but the perceptual and cognitive processes that underlie this effect have not been identified. Because integrating the information contained in a text places demands on working memory, the effect of a picture may be to expand the functional capacity of working memory and thereby to facilitate comprehension. Reasoning thus, we predicted that the availability of a diagram would interact with the difficulty of resolving anaphoric references in texts. The resolution of an anaphor distant from its antecedent (which should stress working memory) should benefit greatly from the presentation of a picture, whereas the resolution of an anaphor near to its antecedent should benefit less from the presentation of a picture. Picture availability and distance separating the anaphor from its antecedent were manipulated in experiments involving both cumulative and moving window presentations of texts. Although picture presence and ease of anaphor resolution significantly improved comprehension of the material, no evidence was found for an interaction of these factors. The results are interpreted as consistent both with dual code theory and with aspects of working memory management that do not involve anaphor resolution.