The Integration of Lone English Nouns into Bilingual Sonoran Spanish

Abstract

Using data from Arizona, United States, the present study seeks to further our understanding of lone other language items (LOLIs) in bilingual discourse and their status as either borrowings or codeswitches by measuring the degree of incorporation that can indicate a LOLI's status as a borrowing or codeswitching. To accomplish this aim, nouns from 40 sociolinguistic interviews from 8 Spanish monolingual speakers from Sonora, Mexico, 8 English monolingual speakers from Arizona, and 24 Spanish-English bilinguals from Arizona (from Sonoran families) are compared. Codeswitching can be defined as the "juxtaposition of sentences or sentence fragments, each of which is internally consistent with the morphological and syntactic (and optionally, phonological) rules of the language of its provenance" (Poplack, 1993, p. 255). Borrowing involves the incorporation of LOLIs from a donor language incorporated into a recipient language and need to be morphologically and syntactically adapted into the recipient language (Poplack, Sankoff, and Miller, 1988; Sankoff, Poplack, and Vanniarajan, 1990). Accordingly, the key difference between codeswitching and borrowing is that borrowings are morphosyntactically incorporated into the recipient language while codeswitches are not incorporated. It is important to note that in terms of LOLIs' status, phonological integration has been discarded for being too variable and therefore not a reliable factor in discerning one-item codeswitches from borrowings (Poplack and Sankoff, 1984; Poplack, Sankoff, and Miller, 1988). In order to measure the degree of incorporation that can in turn indicate the LOLI's status as a borrowing or a codeswitch, the present study applies a sociolinguistic comparative method to loanwords, following Poplack and Meechan (1995, 1998) by comparing nouns from Spanish (recipient language), nouns from English (donor language), and LOLIs from English in Spanish discourse. Since phonology has not been applied to the method of analysis, this study also seeks to explore if phonological integration is correlated to morphosyntactic integration of determiner realization of LOLIs. The results show, in accordance to previous studies, that the LOLIs overall act morphosyntactically like patrimonial Spanish words in terms of the variables that condition determiner usage. In terms of how phonological integration interacts with morphosyntactic integration, it does seem that the two correlate. LOLIs with Spanish morphology are more morphosyntactically similar to Spanish patrimonial nouns and LOLIs with English phonology are more morphosyntactically similar to English patrimonial nouns in both overall frequencies and the factors that condition determiner usage, leading to the hypothesis that LOLIs that are integrated phonologically are established borrowings and LOLIs that are not integrated phonologically are either codeswitches or nonce borrowings. We provide further evidence for this hypothesis by examining the pauses and false starts that are present before LOLIs with Spanish versus English phonology. The results indicate that LOLIs with English phonology are more often preceded with pauses and false starts than LOLIs with Spanish phonology. The findings of this study suggest that phonological integration is a factor that should be brought back to the discussion on discerning LOLIs' status as a borrowing or a codeswitch.