Adjective Placement in Korean Numeral–Classifier Constructions: A Corpus Study of Sortal Classifiers and Measure Expressions

This study examines the placement of adjectives in Korean numeral–classifier constructions, using Mandarin Chinese as a baseline for comparison. Drawing on the morphologically annotated NIKL POS-tagged corpus as the primary dataset and koTenTen concordances as a qualitative follow-up source, we examine where adjectives occur relative to numerals and classifiers and whether corpus evidence allows adjectives to intervene between the numeral and the classifier. The Korean data show that postnominal numeral–classifier constructions are substantially more frequent than prenominal constructions, and that adjective-bearing numeral–classifier strings are rare overall. In contrast, data from the Mandarin CCL corpus show that adjectives can occur between the numeral and the classifier, particularly when the classifier has a more noun-like or measurement interpretation. In the Korean corpora examined, no cases were found in which an adjective appears between a numeral and a highly grammaticalized sortal classifier; the only apparent exceptions involve a single noun-like measure expression, swul ‘spoon’. These findings suggest that Korean strongly preserves numeral–classifier adjacency, and that apparent exceptions arise primarily from elements that retain nominal properties rather than from canonical sortal classifiers.