Chapter 6. Naturalization of the Japanese loanword sushi in English

Abstract

Japanese ranks very high among the non-Indo-European languages that provide loanwords to English (Durkin 2014: 25). The naturalization process of Japanese loanwords in English is explained by Doi (2014) in a three-stage scale, with “totally foreign” at one end and “fully incorporated” at the other. While Doi’s scale offers a refined generalization of the process of naturalization, it is based on examples from the Oxford English Dictionary (OED). Therefore, it is by no means certain that the scale can be extended to account for the patterns of loanwords outside the OED examples. To answer this question, this paper focuses on a Japanese loanword sushi. It examines its characteristics using iWeb corpus (Davies 2018) and a Google web search. Doi (2014) identifies three stages distinguished by morphological and semantic characteristics. In Stage 1, the loanword accompanies a paraphrase as in “sushi or rice sandwiches” (an 1893 citation in OED). In Stage 2, the loanword occupies the attributive slot of a compound (e.g. sushi bar). In Stage 3, the loanword involves a semantic change (e.g. sushi ‘vinegar-flavored rice with fish/vegetables’ > ‘raw fish’ for some speakers) or occurs with an affix (e.g. the plural marker (sushis)), evaluated as evidence for “productivity”. These criteria, however, leave out two types of web data. One is the case where sushi is used in a phrase or sentence as-is (e.g. flavor of the sushi; you can order sushi ...), and these appearances constitute approximately 80% of the iWeb corpus data. The web search also reveals that sushi can be part of a unique construction: e.g. appearing in a snowclone, X is the new Y, as in Sushi is the new pizza., or in an Internet meme (cf. Dancygier and Vandelanotte 2017), One Does Not Simply ..., as in One Does Not Simply Make Sushi (https://makeameme.org/). The other is the case where sushi develops a nomenclatural network in a ‘restaurant frame’ (cf. Coulson 2006): sushi has a subset category, sushi rolls, which continues to expand membership by adding new dish names (Volcano roll, Samurai roll, etc.); or sushi creates new food names via blending (e.g. Fauconnier and Turner 1998) (sushi donuts or doshi [donut-shaped sushi] etc.). To consider data beyond OED, a more flexible and encompassing model of the naturalization process is called for. This paper proposes a preliminary model, which considers different factors that can interact with each other (e.g. morphological, word-network development). It also assumes a layered model, which permits co-existence of “old layers” and “new layers” (Hopper 1987: 159). Examination of other loanwords would offer insight into how the process of naturalization can be more appropriately modeled for Japanese loanwords in English and loanwords in general.