A Diachronic Analysis of Multiword Constructions in Indian English and Australian Aboriginal English

The use of the expressions ‘Indian English’ and Aboriginal English are both quite controversial among variationist linguists mainly because there is a number of Indian English and Aboriginal English varieties which cannot be neatly described in terms of features, even though the existence of a ‘pan-Indian’ and ‘pan-Aboriginal’ set of shared features can be generally recognised (Gupta 2001; Sailaja 2009; Leitner 2004; Malcolm 2013). Arguably in both cases the Indian and Aboriginal varieties have arisen as local linguae francae in extremely complex indigenous multilingual contexts and have later converged leading to the Indian and Aboriginal English varieties, which are recognizable across the two countries. In both varieties there are attested traces of innovative processes in the morphology and syntax continuum away from English, but also evidence of the restructuring continuum under the influence of codified varieties of English (Malcolm 2013). Recent research has shed new light on the interrelationship between geographically distant varieties giving birth to new challenging theories on contact varieties and language universals. In particular, the notion of “vernacular universals” (Chambers 2004) relies on the identification of universally shared features across varieties of English. Szmrecsanyi and Kortmann propose a similar hypothesis supporting the idea of “the universality of conspiracies of morphosyntactic features” (2004; 2012). In order to test these theories along with the possibility of deterministic internal developments occurring in geographically distant contact varieties, a study has been carried out analyzing and comparing two annotated corpora, the Diachronic Corpus of Indian English (835,000 tokens) and the Diachronic Corpus of Aboriginal English (430,000 tokens), specifically compiled to represent the different dimensions of linguistic variability over a period of 150 years (1833-2012). Spoken data such as speeches, reported legal cross-examinations and written data such as letters to the editor have been selected in order to be compared to similar samples from the BNC. The selection of such genres was motivated by the assumption that certain features at the lexico-syntactic interface may be traced to the oral level of language as ‘changes from below’ (Rissanen 1988, p.188) which are subjected to later stabilization with the consequent emergence of new language standards (Deumert & Vandenbussche 2003, p.456). The paper which is part of a larger project comparing pluricentric varieties of English examines the development and stabilization of multiword constructions which present “a number of intriguing analytical challenges which touch on the relation between morphology and syntax and more generally on the architecture of grammar” (Los et al., 2012). The two varieties may or may not present similar tendencies due to their different status, developmental stages, contact situations and cogno-cultural motivations. From this perspective, corpus-based evidence and linguistic diagnostics will be matched in order to establish the extent to which the frequency of such features across genres and time may contribute to the identification of language universals.