Gender in Insular Celtic : a functionalist account of variation and change in Irish and Welsh
Citation:Alessio S. Frenda, 'Gender in Insular Celtic : a functionalist account of variation and change in Irish and Welsh', [thesis], Trinity College (Dublin, Ireland). Centre for Language and Communication Studies, 2011, pp 300
Frenda_PhD_2011.pdf (PDF) 1.293Mb
In this study, I investigate linguistic variation and change in a situation of language contact and obsolescence; speciﬁcally, I look at grammatical gender in Insular Celtic. The analysis of this linguistic phenomenon is conducted using data sets pertaining to spoken Irish and Welsh, taken as the two main representatives of this language group. The data set for each language includes two main components, relative to older and contemporary varieties, respectively. These data were in part drawn from existing corpora and in part collected in the course of this study. The sociolinguistic status of these languages is taken into account; background information is provided and the concepts of language contact, obsolescence and convergence are presented and discussed. The two main goals of this study are (i) to characterize gender agreement in Insular Celtic, using the theoretical framework provided by Functional Discourse Grammar (FDG) and (ii) to provide a quantitative analysis of any observed pattern of variation in order to achieve a better understanding of the system and contribute to the ongoing debate on grammatical gender. The main ﬁnding of my analysis is that whereas conservative varieties exhibit a fairly consistent agreement system across the board, i.e. across agreement targets and grammatical genders, contemporary varieties do not; in the latter, grammatical agreement is still fairly consistent between the article and the noun but signiﬁcantly less frequent with other agreement targets such as the attributive adjective and the anaphoric pronoun. These ﬁndings are problematic on two different levels: typologically, distinctive agreement patterns are well known in relation to anaphoric elements but not to targets lying within the noun phrase; from a theory-internal point of view, FDG does not recognize, within the noun phrase, any smaller constituent which contains only the determiner and the noun, so the observed patterns run contrary to what we expect given a model of gender agreement based exclusively on the notion of feature copying across syntactic boundaries. To overcome these difficulties, I propose a model of noun-phrase agreement for Insular Celtic which is consistent with FDG and is capable of accounting for the language-internal aspects of the observed linguistic change; in this model, feature copying operates alongside a different mechanism which is responsible for the peculiar way in which agreement between the article and the noun is marked in Insular Celtic. Then, looking at anaphoric agreement in particular, I argue for the necessity that the current FDG theory be expanded in order to provide a better account of how different types of information capable of inﬂuencing pronominal agreement forms are stored and accessed. Finally, I argue that the observed variation in gender agreement is to be interpreted as the result of a process of resemantization and address two related questions: what might have started this process, and why it has resulted in the overgeneralization of masculine agreement forms. Considering comparable cases, I argue that the resemantization of pronominal agreement may be explained as a result of the gradual demise of gender agreement marking within the noun phrase. As regards the latter question, the switch leading to the overgeneralization of masculine forms, rather than a logically equally possible switch in the opposite direction, is explained as a combination of two factors, language-internal (markedness) and contact-induced (convergence).
Author: Frenda, Alessio S.
Advisor:Saeed, John I.
Publisher:Trinity College (Dublin, Ireland). Centre for Language and Communication Studies
Note:TARA (Trinity's Access to Research Archive) has a robust takedown policy. Please contact us if you have any concerns: firstname.lastname@example.org
Type of material:thesis
Availability:Full text available