The “Gradient Structure” of Korean Words

Hailey Hyekyeong Ceong


A probabilistic approach, in which morphological structure is non-discrete - gradient word structure - has been proposed for Indo-European languages (Hay and Baayen, 2005). This paper demonstrates a “gradient structure” of Korean words with a particular focus on Hannate words (so-called “Sino-Korean” words). Hannate words are usually considered loanwords, but most of them acquire their lexical category by combining with native suffixes,  as with the adjective namca-tapta ‘manly’. The word namsengmi ‘masculine beauty’ can be a complex or compound word, depending on the treatment of mi. This problem of determinability is similar to the ambiguity encountered in English neoclassical compounds. For example, it is unclear whether socio- is a prefix or a compound element in a complex word like sociolinguistic, given that the socio- element is treated as a stem in word like sociology (Bauer 1998) . The adoption of Hannate roots into Korean eliminates the ideographic and tonal information that fixes their meanings in Chinese. This paper helps shed light on the understanding of Hannate words in Korean and ways that language contact and the borrowing of words have consequences for the expanded lexicon of the borrowing language, which includes native items, borrowed items, and the products of reanalysis and analogy by speakers over time.


gradient; word formation; Korean; Sino-Korean

Full Text:


Working Papers of the Linguistics Circle

EISSN: 1920-440X
ISSN: 1200-3344

University of Victoria