Use this URL to cite or link to this record in EThOS:
Title: The inference-driven model of quantifier focus
Author: Dawydiak, Eugene Jurij
ISNI:       0000 0001 3418 8054
Awarding Body: University of Glasgow
Current Institution: University of Glasgow
Date of Award: 2001
Availability of Full Text:
Access from EThOS:
Access from Institution:
Previous research (e.g. Moxey & Sanford, 1987) has demonstrated that positively and negatively quantified statements place attentional focus upon different subsets of a logical superset and are, thus, associated with different patterns of subsequent anaphoric reference. Typically, positive quantifiers (e.g. many (x)) are associated with anaphoric reference to the subset of entities for whom the sentence predicate is true (the reference set or refset) while negative quantifiers (e.g. not many (x)) are associated with reference to the subset for whom the predicate is false (the complement set or compset). The primary objective of this thesis is to provide an empirical evaluation of a model (the Inference Model) which has been advanced to explain the occurrence of complement set reference. The experiments reported in this thesis address three central issues associated with the Inference Model: the relationship between denial and complement set reference; the impact of connective on complement set reference; and, the time course over which complement set focus is established. The empirical chapters contained in Section B present the results from four off-line sentence completion studies. Experiments 1 and 2 investigated the relative effects of denial and downwards monotonicity on complement set reference in the context of both free continuation (Experiment 1) and forced reference (Experiment 2) tasks. Consistent with the Inference Model, the results of both studies indicated that denial was the major determinant of complement set reference. Experiment 3 represented a slight digression from the main direction of the thesis and was conducted in order to test an explanation for an unexpected observation made in the first two experiments. Experiment 4 investigated the sensitivity of set reference pattern to the inferential constraints imposed by different connectives. The results indicated that the pattern of set reference was largely insensitive to connective and were interpreted as being consistent with focus having been established prior to the connective being encountered. Section C reports the results from three on-line studies which attempt to identify the time course over which complement set focus develops. This question is important as the Inference Model predicts that compset focus should be established less immediately than refset focus. Experiments 5 and 6 measured reading-times for an anaphoric reference sentence which appeared immediately after a quantified statement. The results demonstrated no systematic asymmetry in the time taken to resolve compset and refset anaphors and were interpreted as being consistent with focus for both subsets having been established during comprehension of the quantified sentence. Experiment 7 attempted to index the time course of set focus effects using the more sensitive on-line measure of eye-movements. The results were generally consistent with the Inference Model in that they suggested that compset focus was established later than refset focus. The overall pattern of results presented here can be interpreted as being broadly consistent with both the Inference Model and a less complex mechanism based upon the general account of negative processing suggested by Clark (1976). The results are not, however, consistent with an alternative account of complement set reference proposed by Kibble (1997).
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Natural language; Positive; Negative; Anaphoric