This paper attempts to reconstruct the correct form and meaning of the word kocha (古次) that corresponds to ku (口) in the Goguryeoic toponym Hyeolgu-gun (穴口郡) in Samguksagi (三國史記). As to kocha (古次) and the analogous holcha (忽次), a couple of hypothesis have been put forth: Yang (1942, 1965) and Shin (1958) suggested that the words referred to a cape (串) or a promontory (岬), even though they acknowledged that the words were comparable to the Old Japanese *kuti ‘mouth’. Lee (1968) saw that the word form kocha (古次) and holcha (忽次)’ showed dialectal differences in the Gogureoic toponym. On the other hand, Kim(1968) said that the two forms originated from different sources: one originated from the word referring to ‘mouth (口)’, the other from the word indicating ‘promontory (岬)’.
However, it is dubious that kocha (古次) and holcha (忽次) or Chinese letter ku (口) should always be interpreted as referring to a cape (串) or a promontory (岬) in Gogureoic toponyms. This paper emphasizes that the shape of the mouth of the beasts like tigers, lions, dogs, foxes, cats, etc. are very similar to a cape (串) or promontory (岬) in that the mouth of a beast is lowering from the back jaws to the front, just like a cape (串) or a promontory (岬) heading for the sea. This means that the mouth sense and the cape sense of kocha (古次) or holcha (忽次) form a polysemy, not a homonym in Gogureoic toponym. And this paper suggested that the Goguryeoic word form for kocha (古次) and holcha (忽次) be /*kutsi/, considering the vowel shift and Old Japanese form *kuti, and other altaiic forms indicating mouth.
The Altaic lexical entries relative to a semantic category ‘mouth’ inStarostin et al. (2003) are the forms like <SUP>*</SUP>ki??? ’? u, <SUP>*</SUP>aga and <SUP>*</SUP>?m? o. Starostin et al. (2003) treats the Old Japanese /<SUP>*</SUP>kuti/ at the entry <SUP>*</SUP>ki???’u. However, the Altaic lexical items listed in the entry <SUP>*</SUP>ki????’u in Starostin et al. (2003) cannot be seen to be related with the Old Japanese /<SUP>*</SUP>kuti/. The Altaic lexical entries comparable to the Korean ‘mouth’ might be said the forms listed in the entries like <SUP>*</SUP>aga and <SUP>*</SUP>?m? o. It is claimed that the proper item treated in the entry <SUP>*</SUP>aga should be the Korean word /akari/ or /akuri/ indicating the animal’s mouth or the entrance of objects. This paper claims that the Altaic varieties could be captured by postulating the Abstract Reconstructed Form (ARF) like <SUP>*</SUP>AKVTW, in which *A stands for the initial vowel, <SUP>*</SUP>K the first consonant, <SUP>*</SUP>V the second vowel, and <SUP>*</SUP>T the second consonant, and <SUP>*</SUP>W the final vowel.