@phdthesis{Horsch2022, author = {Horsch, Jakob}, title = {The comparative correlative construction in World Englishes}, doi = {10.17904/ku.opus-833}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:824-opus4-8335}, school = {Katholische Universit{\"a}t Eichst{\"a}tt-Ingolstadt}, pages = {XX, 346 Seiten : Diagramme}, year = {2022}, abstract = {Employing a two-pronged approach to analyzing language, this study uses evidence obtained from a large-scale corpus study and psycholinguistic experiments to investigate the Comparative Correlative (CC) construction (e.g. The more we get together, the happier we'll be) across twenty varieties of English. More than 10,000 tokens from three of the largest corpora of English (COCA, BNC, GloWbE) and experimental data from almost 600 participants from four locations (Texas, South Africa, Singapore, and Kenya) are analyzed using advanced statistical methods, providing striking evidence for a high degree of parataxis between the subclauses of the construction and massive redundancy regarding cross-clausal associations which, it is argued, are best explained with a Usage-based Construction Grammar (CxG) approach that posits the existence of an elaborate network of interconnected mesoconstructions. Addressing more general questions about the nature of human language, it is argued that this 'bottom-up' approach to language can account for one of the most fundamental riddles that the discipline of linguistics is striving to solve: Accounting for the potential of infinite creativity of language based on a finite set of rules. Furthermore, it is claimed that as an inclusive approach to language, CxG is better at accounting for construction-specific features that Mainstream Generative Grammar (MGG)-based approaches have struggled with. As it turns out, the evidence suggests that the English CC construction is anything but 'peripheral'. Additionally, this study examines the CC construction from a World Englishes perspective, addressing a significant gap in research: Schneider, for example, has noted that "[o]nly rarely have New Englishes been employed as test cases for general questions of language theory" (2003: 237), and Hundt writes of the challenge of "bring[ing] intricate, quantitative modelling of usage data to bear on theoretical modelling" (2021: 298). Accordingly, this study provides evidence for the predictive power of Schneider's Dynamic Model (DM) (2003, 2007), which posits that there is a correlation between evolutionary stages of World Englishes and linguistic features. In this context, it is also shown how principles of cognitive approaches to language can explain both generalizations across varieties as well as variety- and DM stage-specific effects. Furthermore, this study provides strong evidence to support Kortmann's claim of a continuum ranging from Pidgins and Creoles as "deleters" to L1 varieties as "preservers" (2014: 9) of grammatical formatives, contra Mesthrie and Bhatt's claim of a geographical dichotomy between African and Asian varieties.}, subject = {Englisch}, language = {en} }