Single-word incorporations in Ukrainian-English bilingual discourse: Little things mean a lot.
|Title:||Single-word incorporations in Ukrainian-English bilingual discourse: Little things mean a lot.|
|Abstract:||This research develops theory-independent diagnostics for identifying the status of single-word incorporations of one language occurring in the discourse of another. Our investigation is based on the premises of Variation theory requiring systematic evaluation of a spoken-language corpus, and exhaustive quantitative analysis of inherent variability not only in mixed or bilingual language, but more innovatively in the unmixed or monolingual vernaculars, which are in contact. We hypothesize that (1) if the contentious lone items are code-switched, i.e. produced by the donor language, their patterns of behavior should be similar to those of their unmixed donor-language counterparts, while at the same time differing from those of their unmixed recipient-language counterparts; (2) if such items pattern similarly to their unmixed recipient-language counterparts, while differing from their unmixed donor-language counterparts, we infer that they were produced by the recipient language, and hence are borrowed; and (3) if their patterns follow neither donor- nor recipient-language counterparts, they could be produced by some other language contact phenomena. We test this hypothesis by analyzing natural performance bilingual data involving two typologically different languages, Ukrainian (a Slavic language) and English. Numerous morpho-syntactic differences between these two languages provide a wide range of 'conflict sites', which allow us to observe the differences in their patterns, and hence, to reveal unambiguously the language membership of the items in question. The results showed that lone English-origin items occurring in Ukrainian discourse were not produced exclusively by one process. The majority of them were borrowed, whereas some were code-switched. None, however, were produced by any other phenomena. Borrowed items, either nonce or established, were shown to acquire all morpho-syntactic features required for a particular part of speech, and applied those features with the same variability as their Ukrainian counterparts. Code-switched English-origin utterances clearly failed to fulfill Ukrainian requirements (with respect to gender assignment, different types of agreement, syntactic position, etc.), and exhibited behavior similar to their unmixed English counterparts. We have also found that a switch site constitutes a 'barrier' for language-specific features, but not for the universal ones. Having succeeded in applying this diagnostics to code-switching/borrowing identification, this study reveals a detailed view of what the process of integration is at the moment of borrowing, i.e. on the synchronic level. It consists of applying the patterns of the language in which they are incorporated, either standard or non-standard, to other-language items in exactly the same way, as to the recipient-language counterparts. Thus, this research provides clear indication that the source grammar of inherently ambiguous forms in bilingual discourse can be empirically determined by using a comparative method applied to an entire corpus of natural performance data, using appropriate diagnostics, focusing on language-specific features, and making use of, rather than avoiding, inherent variability.|
|Collection||Thèses, 1910 - 2010 // Theses, 1910 - 2010|