Appendix:Middle Korean h-final nouns

Middle Korean had a number of nouns and numerals that ended in /-h/, which is believed to arise from Old Korean */-k/.

When following a sonorant (a vowel or nasal or liquid consonant), /-h/ did not appear in isolation but surfaced when followed by a vowel- or plosive-initial suffix. When following an obstruent, /-h/ fused with the preceding consonant to form an aspirate consonant. However, because /h/-initial consonant clusters also surfaced as an aspirate, it is impossible to distinguish between /-hC/ clusters and /-Ch/ clusters. Accordingly, aspirate-final nouns are not included in this list.

The /h/-final nouns have been the focus of particular attention by linguists because not only are they a distinctive feature of Middle Korean lost in the modern dialects, /h/ appears to be clustered around the single semantic field of nouns for places. This raises the possibility that it stems from a fused place suffix. Both Alexander Francis-Ratte and Martine Robbeets, who believe in a genetic relationship between Korean and Japanese, suggest a connection to Japanese 処(こ) (ko, “place”). Both Korean and non-Korean scholars have also suggested a potential link to Proto-Turkic *-ag as in *ōtag (“fireplace”).

The list below combines those of Heo 1975, Yi 2010, and Park 2018, excluding compound words. The three lists differ slightly in what words are included. As final /-h/ was being lost even in the mid-fifteenth century, during the time of the earliest Hangul sources, many forms below are also attested without the coda. Due to the conditioned surfacing of the /h/-coda, it is impossible to determine whether /-h/ existed or not for many nouns attested only a few times in the surviving corpus.

Middle Korean had a diminutive suffix 악 (Yale: -ak) / 억 (Yale: -ek) which was not extremely productive by the fifteenth century but may have been more productive earlier, and which is seen as a source for some of these forms.

The reason for the pitch realization differences is unclear. It is often suggested that these reflect a borrowed layer from a non-standard variety of Middle Chinese, rather than being nativised from Sino-Korean (based primarily on the late Tang prestige dialect). For many of these words, it is disputed whether they are truly of Chinese origin or are native words that coincidentally resemble Chinese. The wordlist is from Yi 2010.

Whether the /-h/to /-k/ correspondences here reflect an indigenous Korean weakening of the latter, or whether the Chinese source variety was a late form of Middle Chinese that had already weakened coda stops to /ʔ/ (Modern Mandarin lacking coda stops altogether), is disputed.

• 허웅 [heoung] (1975) 우리옛말본—15세기 국어 형태론 [uriyenmalbon-15segi gugeo hyeongtaeron, Historical Korean Grammar: The Morphology of the Fifteenth-century Language], Saemmunhwa
• 이준환 [ijunhwan] (2010) “중세국어 ㅎ말음 체언의 음운론 [jungsegugeo nmareum cheeonui eumullon, The phonology of Middle Korean /h/-coda substantives]”, in Gugyeol yeon'gu, volume 24, pages 163—199
• 박용찬 [bagyongchan] (2018) “ㅎ 말음 체언의 역사적 변화 [h mareum cheeonui yeoksajeok byeonhwa, The diachronic change of Middle Korean /h/-coda substantives]”, in Urimalgeul, volume 78, pages 51—77