Implicit Causality, Pronominal Form and Anaphora Resolution in Spanish

• Author / Creator

  Cristerna Román, Dalia

• This thesis investigated the effects of implicit causality and pronominal form, i.e., null and
  overt pronouns, in Spanish anaphora resolution. A visual world eye-tracking experiment and
  a self-paced listening task were used with adult monolingual speakers of Mexican Spanish
  to assess two research questions: 1) whether implicit causality affects the resolution of
  pronominal subjects in Spanish, and 2) whether this effect was modulated by pronominal
  form.
  Implicit causality (IC) is the semantic bias that attributes the cause of an action to
  the subject (NP1) or object (NP2) of a sentence (Garvey & Caramazza, 1974). IC has been
  shown to affect pronoun resolution in various languages, including Spanish (Goikoetxea,
  Pascual & Acha, 2008).
  Pronominal form has also been shown to affect pronoun resolution in some pro-drop
  languages. Carminati (2002) proposed the Position of Antecedent Hypothesis (PAH) for
  Italian. PAH claims that null pronouns prefer to retrieve their reference from the most
  prominent antecedent (subject), while overt pronouns tend to be assigned to antecedents in
  a lower syntactic position (object). Alonso-Ovalle, Fernández-Solera, Frazier and Clifton
  (2002) tested the PAH in Spanish and they found similar results to Carminati’s. However,
  Chamorro, Sorace and Surt (2016) and Chamorro (2018) showed that speakers of Iberian
  Spanish seem to have clear preferences for overt pronouns retrieving objects, but not for null
  pronouns retrieving subjects.
  The effects of semantic and syntactic information on pronoun resolution in Spanish
  have been investigated separately; yet, to our knowledge, the combination and potential
  competition of the two strategies has not been examined.
  Accordingly, this thesis investigated the influence of these two sources of information
  in ambiguous and non-ambiguous sentences in Mexican Spanish. First, the visual world eye-

  iii

  tracking experiment showed that ambiguous null pronouns tend to be matched with first-
  mentioned antecedents, and this effect is stronger with subject-biased verbs. Ambiguous

  overt pronouns showed preferences in odds with the predictions of IC: with overt pronouns
  and NP1 verbs, participants’ looks pointed toward the second-mentioned antecedent, while
  with NP2 verbs they pointed to the first-mentioned. This IC unpredicted pattern could be
  explained by how sensitive overt pronouns are to less salient entities. Ariel (1990) claimed
  that the more specific an expression is, the better it is at retrieving a less salient antecedent.
  Relative to the effect of pronominal form, overt pronouns showed stronger effects than null
  pronouns. This might be because null subjects lack an auditory stimulus prone to be
  matched with a visual stimulus, whereas overt pronouns are a phonetic realization that can
  be more easily matched to a visual cue.
  Second, the self-paced listening task showed that non-ambiguous null pronouns are
  sensitive to IC information while overt pronouns not so. We found significant delays in
  participants’ listening times when the reference of the null subjects was non-congruent with
  the IC bias. Moreover, overt pronouns did not show significant delays in IC incongruent
  contexts. Relative to the effect of pronominal form, non-ambiguous overt pronouns did not
  show processing penalties when they retrieved their antecedent from the subject of the
  previous sentence. These results indicate that non-ambiguous overt pronouns in Spanish do
  not have a fixed preference for object antecedents.
  In addition, the role of the methodologies used was also discussed. We emphasized
  the importance of considering the differences in the outputs of online and offline methods;
  while visual world eye-tracking experiments give information about attentional preferences,
  self-paced listening tasks reveal actual language processing. Therefore, we treated the results
  of both experiments as complementary instead of assuming that both approaches revealed
  the same answers to our research questions.

  iv

  To recapitulate, this thesis shows that in Mexican Spanish when using a visual world
  eye-tracking experiment, null pronouns were linked with first-mentioned antecedents,
  especially with NP1 verbs, and overt pronouns showed preferences in contradiction to the
  predictions of IC. In the self-paced listening task, null subjects, but not the overt, showed
  sensitivity to IC. Overall, we showed that the use of null or overt pronouns and their specific
  features, such as their tendency to retrieve more or less salient antecedents interacts with
  implicit causality. Furthermore, we support the idea that different pronominal forms are
  sensitive to different kinds of information on varying degrees.

• Subjects / Keywords

  • Anaphora resolution
  • Implicit causality
  • Pronoun resolution
  • Spanish

• Graduation date

  Fall 2020

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/r3-ssag-4e36

• License

  Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.