Hybrid loans: a study of English loanwords transmitted to Korean via Japanese

Abstract

In this paper we present the results of a study of Japanese-influenced English loanwords in Korean. We identify some 10 phonological indexes that identify the loan's Japanese provenance and examine their relative rate of retention in 287 loan forms that show a mixture of Japanese-style and direct-English-style phonological characteristics. Our chief finding is that certain traits of Japanese-style loanwords are more resistant to change to the direct English style of adaptation that is applied to contemporary loans. We consider two possible explanations for such a hierarchy. First, the resistance hierarchy reflects the relative perceptual saliency of the relevant features (cf. Steriade, 2001). Second, the resistance hierarchy reflects the degree of confidence Korean speakers have about the ''correct'' direct-English borrowing pattern.

The Adaptation of English Liquids in Contemporary Korean: a Diachronic Study

“…So, one would like to see more evidence of duration of input liquid as a relevant factor in singleton adaptation. Heo and Lee (2004), Kang et al (2008) and A. Lee (2009), on the other hand, propose that the singleton adaptation is a remnant of Japanese-mediated loans.…”

“…This is a factor I did not consider going into the study but added to the study after examining the data. Also, I examined the effect of Japanese influence on the rate of singleton realization by examining if the singleton realization occurs more frequently in variants that contain other known characteristics of Japanese influence (Kang et al 2008) The coefficients show the direction and the size of each effect. The factors are contrast-coded so that the condition that favors geminate adaptation (i.e., Japanese influence, non-initial position, cluster context, and singleton spelling) is coded as a reference category (=0) for each factor and the other category is coded as 1.…”

“…As for the Japanese influence, it is too time-consuming to code the entire data for the presence of Japanese influence and an approximation is made by examining the adaptation of word-initial voiceless stops only. 15 In direct English to Korean adaptation, English voiceless stops are consistently adapted as aspirated stops but in Japanese-mediated loans, word-initial voiceless stops are adapted as lenis stops of Korean because English voiceless stops are adapted as voiceless stops of Japanese which are in turn adapted as lenis stops in Korean as in Columbus → EPK <koLomposɨ> vs. <kʰoLLəmpəs'ɨ> (Ito et al 2006, Kang et al 2008. 16 Therefore, words that do not contain a word-initial voiceless stop are not coded for the Japanese influence factor.…”

“…16. According to Kang et al (2008), voiceless stop adaptation is the most identifiable trait of Japanesemediated loans. In hybrid loans, i.e., in loan forms where a subset of Japanese traits is replaced by the traits of direct English-to-Korean loans, the voiceless stop adaptation pattern is one of the first to be eliminated.…”

“…In Korean, the pervasive influence of Japanese-mediated loanwords, i.e. Western loanwords borrowed into Korean through Japanese (Tranter 1997, Kang et al 2008 complicates the matter further. Also, the interpretation of written data from time periods when no spoken data are available requires some amount of detective work that takes into account the orthographic conventions of the time.…”