ABSTRACT
Three experimental studies are presented testing the choice of a left or a right dislocation in Catalan, depending on the bridging relation between the dislocate and its antecedent. We make the hypothesis that the stronger the anaphoric link between the dislocate and its antecedent, the more appropriate a right dislocation is, whereas the opposite is true for left dislocation. The results of two acceptability judgment experiments partially support our hypothesis: Ratings with right dislocation decrease proportionally to the strength of the anaphoric link; however, it is ratings of canonical sentences (without dislocation at all), rather than of left dislocation, that display the opposite direction. For left dislocation, the picture is more complex and reflects the pragmatic function of this construction as a topic shifter or marker of contrast. A third experiment tests the monotonicity hypothesis, according to which a left dislocation is chosen when the anaphoric relation with the antecedent is either downward monotonic or nonmonotonic, whereas a right dislocation is preferred with an upward monotonic relation. The results partially support such a hypothesis but also further support the importance of taking the pragmatic function of left dislocation into account to explain its choice in discourse by speakers.
Acknowledgments
We are grateful to the audience of the workshop “The Meaning of Functional Categories in the Nominal Domain” (March 21-22, 2019, Universitat Autònoma de Barcelona), where part of the material in this article was presented. Thanks are also due to two anonymous reviewers and the editor for useful comments, to Sílvia Planas-Morales for helping with the recordings for Experiments 2 and 3, and to Yair Haendler for his precious help with the statistical analysis.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1. The study of anaphoric reference in discourse has been a major issue in theoretical linguistics (Huang, Citation2000; Reinhart, Citation1983), discourse analysis (Schwarz-Friesel, Citation2007; Webber et al., Citation2003), natural language processing (Mitkov, Citation2012; Poesio, Citation2016), psycholinguistics (Haviland & Clark, Citation1974; Singer, Citation2012), and neurolinguistics (Burkhardt, Citation2006; Callahan, Citation2008).
2. A poset is a set defined by a relation that is either reflexive, antisymmetric, and transitive or irreflexive, asymmetric, and transitive. See Partee et al. (Citation1987, pp. 277–280), for technical details.
3. The choice of such a scale is motivated by the participants’ (mostly university students) familiarity with a 10-point scale, which is used in Catalonia for evaluation at school and at the university.
4. A Helmert contrast coding compares the mean scores of each condition with the mean scores of the totality of the preceding conditions. Given the order of the five bridging conditions in the model, the mean score of set-member was compared with the mean scores of epithet + hypernym, the mean score of necessary part was compared with the mean scores of epithet + hypernym + set-member, and so on.
5. Although no significant difference from epithet to hypernym is found.
6. We did not control for the relative informativeness of the object with respect to other material in the same sentence, which may vary from item to item. However, since we included items as a random factor in the statistical analysis, this potential variation was controlled by the model.
7. In our experiment, the rhetorical relation between the two utterances is always either an Elaboration or a Comment. On rhetorical relations, see Mann and Thompson (Citation1988) and Asher and Lascarides (Citation2003).