Alyawarr children's variable present temporal reference expression in two, closely-related languages of Central Australia

Abstract

In a small, remote central Australian community, young children go about their daily lives mainly speaking Alyawarr English (AlyE), a new Central Australian contact language. At school they are acquiring Standard Australian English (SAE) as a second language. Much of what they encounter in SAE will seem familiar and much will be obviously different. In between, there will be many subtle differences that are possibly harder to detect, parse and maintain. This study investigates this remarkable bi-varietal language use, considering whether separate ‘codes’ are indeed evidenced, how they can be quantitatively modelled, and what they tell us about the impact of formal similarity between languages and emerging bilingualism. The use in childhood of multiple closely-related languages is somewhat of a final frontier for several strands of research: child language development, second language acquisition, and creole studies. However, the methodologies common in these fields don’t easily import to the present scenario: for while the assumption of existing, separate codes may be operationally practical (even if theoretically contestable), the presence of overlapping structures (i.e. morphemes that are used in both AlyE and SAE) in the present data set makes code separation on structural grounds problematic. The solution tested here takes a corpus of 50+ hours of naturalistic video recordings of six focus children, and first creates two maximally contrastive, contextually defined data sets: HOME (home, Alyawarr interlocutor) and SCHOOL (school, non-Alyawarr interlocutor). Each data set is then analysed using the Comparative Variationist method. Three variables of present temporal references clauses were selected: aspect morphology with variants V (e.g. ‘look’), Ving (e.g. ‘looking’), and Vbat (e.g. ‘lookbat’), 1sg subject pronouns (‘I’ and ‘AM’) and verb transitivity marking (-im and unmarked). In both HOME and SCHOOL data, variants, their distributions, respective envelopes of variation and variable grammar (modelled using logistic multiple regression conducted in Goldvarb X) are examined. The results show that code-separation is evidenced for all three variables, but that the locus of change is not the same in each case. For transitive marking and 1sg subject pronouns, the same variable grammar is deployed in both HOME and SCHOOL data, while non-SAE-like variants are increasingly avoided in the SCHOOL (i.e. -im and ‘AM’) and thus the envelope of variation for the SAE-compatible variant (i.e. -Ø and ‘I’) expands its range of use on the way to becoming the categorical variant in the children’s SAE. For aspectual morphology, the locus of change is located within the variable grammar where complex patterns of reorganisation are evidenced (e.g. HOME V is strongly associated with ‘stative’ clauses; SCHOOL V with ‘stative’ and ‘habitual’). Additionally, fundamental changes in the envelope of variation (e.g. Ving is not used on transitive verbs in the HOME but is in the SCHOOL) and the range of variants (e.g. Vbat is not used in the SCHOOL) indicate that these remain central considerations in the use of all three variables. This study therefore breaks new ground in both methodological terms, with the application of variationist modelling to child bi-varietal language use, and in advancing our understanding of the vectors of code-separation in the complex ecologies of the region.