The learnability of semantic distinctions: the case of evidentiality
Date
2022
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
It is often assumed that cross-linguistically more prevalent distinctions are easier to learn potentially due to their conceptual naturalness. Prior work supports this hypothesis in phonology, morphology and syntax but has not addressed semantics. This work aims to unravel the potential factors that contribute to the learnability and the cross-linguistic prevalence of certain semantic distinctions over others. Within the semantic domain, evidentiality offers an excellent test case for learnability patterns due to the cross-linguistic frequency of certain evidential systems along with the fact that such distinctions are not grammaticalized in English and can be taught to English-speaking adults within an Artificial Language Learning paradigm without native language interference. ☐ Our first aim is to test to test whether conceptual naturalness can account for ease of learning and cross-linguistic frequency patterns of evidential systems. When exposed to these miniature evidential morphological systems, adult speakers of English learned the typologically most prevalent system (marking indirect, reportative information) better compared to less-attested systems (Experiments 1-2). Similar patterns were observed when non-linguistic symbols were used to encode evidential distinctions (Experiment 3). ☐ Our previous experiments have shown that learners preferentially encode reportatives compared to other sources but left open the nature of the asymmetry because of design-specific issues. Our second aim is to adjudicate between a pragmatic bias and category partition issues as an explanation for the learnability of evidential distinctions. In Experiments 4-6, we directly compared two simple evidential systems, each marking only a single source (visual vs. reportative) and leaving the other source unmarked. Similarly to our previous findings, participants learned more easily to mark reportative information sources over visual/direct sources of information. Our results provide support for a pragmatic bias that shapes both the cross-linguistic frequency and the learnability of evidential semantic distinctions. ☐ Our last aim is to test whether learnability patterns change when the same information source meanings are mapped on different linguistic and non-linguistic forms. Developmentally, evidential meanings seem to follow a protracted trajectory. According to a prominent hypothesis, this difficulty stems from the complexity of the underlying concepts. On an alternative proposal, the difficulty often lies in the mapping between linguistic expressions and concepts, even if the concepts themselves are available. In Experiments 7-8, we offer a novel argument for the role of mapping factors in acquiring a evidential meanings, specifically pointing to the importance of the syntactic and pragmatic factors that lead to the form-meaning association.
Description
Keywords
Artificial language learning, Evidentiality, Learnability, Pragmatics, Semantics