Valerio Anselmo
A research on the name Kudara
(published in 1974)
The language of Paekche (2)

The language of Paekche (1)

Hypotheses on Paekche (1)


The language of Paekche (part two)

[page 9 of the original (cont.)]
Between the two languages must eventually have happened a kind of fusion, so that elements of one language would have passed into the other and viceversa, creating one single language, common to the two classes. To a similar conclusion, keeping in mind that the Chinese writing was adopted in Paekche around the half of the fourth century A.D., seems to lead also the study of one of the phonetic characteristics of the languages of Silla, Paekche and Koguryŏ, that is the palatalization of the t. From the place-names, in fact, we get that in the language of Silla the t  palatalizes in front of i  in all the positions of the syllable in the word, in the language of Koguryŏ the phenomenon happens in all the positions except in the initial one, while in the language of Paekche is midway between the one and the other 43. The studies on this subject have not yet been deepened enough and further explanations will surely appear in the near future.

Besides the Paekche’s place-names having connection with the Silla language, we remember here three names that, on the contrary, have clearly something to do with the language of Koguryŏ, that is Mich'uhol, previously already seen 44, Kimae 其買 and Kajinae 加知奈 (or Kaŭlnae 加乙乃), taken from the kwŏn 37 of the Samguk sagi, section «Koguryŏ» the first, and «Paekche» the other two.
[page 10 of the original]
The first name, Mich’uhol, which in the text concerning the origin of Paekche appears associated with that of Wiryesŏng, lets us understand that the last character hol is used phonetically to transcribe a word in the language of Paekche having the same meaning of the last character of the other name, sŏng (“fortress”), which is clearly the translation in Chinese of a local word. This tells us that also one of the centers of Paekche was called hol (fortress or encampment), as it happened in the language of Koguryŏ 45. For the second name, Kimae 46, from the analysis of the text (其買縣一云林川) we can argue a connection between the last character of Kimae and the meaning of “river” . As a matter of fact, in the above text, Kimae 其買 and Imch’ŏn 林川 are both mentioned as names of the same place, and the first one (Kimae) proves to be the phonetic transcription of the name, while the other one (Imch’ŏn) is its translation in Chinese. This character too is found in the place-names of Koguryŏ, where it transcribes a local word having the meanings of “water” and “river” 47. Finally, in the last case, the two names of Kajinae and Kaŭlnae, even if they are not the translation one of the other, let us understand that the word nae < MK 48 naj, which appears with the same pronounciation (even if with different characters) in both cases in final position, can be connected with a word of the language of Koguryŏ having a pronounciation similar to this one and the meaning of “land, territory” 49.
[page 11 of the original]
For these characters Karlgren reconstructs the archaic and ancient Chinese pronounciations: *χmwət / χuət  for , - / mai:  for , *nâd / nâi-  for and *nəg / nậi  for 50, from which we can obtain the pronounciations in Paekche's language *xol  for “fortress”, corresponding to the Mongolian qorga “fortress 51, fence”, to the ancient Turkish qurγan “fortress” and koriğ  “enclosure” 52 and to the MK 53 .ulh “enclosure, fence” 54, *mai  for “river”, perhaps to be connected to the MK mɨl  “water”, and with the Mongolian mari “great river” 55, *nai  for “land”, corresponding to the Jurchin náh, to the Manchu na, to the Goldi na and to the ancient Japanese *na, all with the meaning of “land”.

The already cited 56 Chou-shu passage mentions for the king of Paekche the names Wolaγa 於羅瑕 and Kjʌnkitsi 鞬吉支 and for the queen the name Woljuk 於陸 57. Let's reconstruct, always with the help of Karlgren, the northern Chinese pronounciations of the 6th-7th century of these names.
[page 12 of the original]
We will obtain: ·uo-lâ-γa  for 於羅瑕, ki̯ɒn-ki̯ět-tśie  for 鞬吉支 and ·uo-li̯uk  for 於陸 58. The cited text says that Kjʌnkitsi 59 was the name given to the king by the people and, since it seems to be phonetically different from the other two, will be analysed later. Woljuk “queen” connects directly with the name for the “wife” (夫人) of the king of Koguryŏ, orikuku 60, and with the names for “queen” quoted by the Nihon shoki as reading of the characters 大后 and 王后 when it is the name of the wife of the king of Paekche: komuoruku, komuworuku 61, konioruku and koniworuku 62. In Mongolian uruγ means “relative-in-law”, while xan-u uruγ means “the king's family; dynasty” 63, An analysis of the forms komuoruku and komuworuku demonstrates the presence, in the second syllable, of a suffix -u of the same type of that used in classic Mongolian to form the genitive of a name terminating with the -n consonant 64. The -i of the forms konioruku and koniworuku would probably be the same suffix transcribed in a different way 65. The coincidence with the cited Mongolian expression lets us presume that komuworuku (王后) and the other forms of this name originally meant “the wife of the king” or “the wife of the chief”, hence we could get, for the word “king” or “chief”, the term kom or kon. As a logical consequence we get that the name Woljuk, which in this way connects with the Mongolian language, should be a “northern” word.
[page 13 of the original]
The pronounciation ·uo-li̯uk does indeed surprise us because of its perfect correspondence with the above mentioned term woruku 66.

Wolɑγɑ , the name or title of the king 67, although of a much more difficult solution, could perhaps be connected with the Mongolian ulaγan  “red” 68, which in turn can be connected to the MK pɑlk- “to be clear, bright, luminous”. The adjective “luminous” seems to have been an attribute of the king: this is attested, indeed, by the expression pɑlgɨn nimgɨm  (“luminous king”) of the Chŭngbosamnyakchikhae  增補三略直解 69, while also the name of the Mongolian town of Ulan Khoto (Ulaγan Xota), previously called Wang-yeh-miao 王爺廟, seems to lead us to a similar conclusion.
[page 14 of the original]
Wolɑγɑ  would then mean “the luminous” and be a title of the king, used also with other Mongolian or Tungus populations. Both these names, Wolɑγɑ  and Woljuk , seem to be linkable to similar forms in Mongolian, as we have seen. The first name, moreover, is to be connected to the Mongolian in a way different from pɑlk- , perhaps in a more direct way than what we can do with this latter form, which still keeps the initial voiceless bilabial occlusive from which later developed the corresponding fricative attested by the Middle Mongolian hulɑ’ɑn 70. The Mongolian term ulaγan is in connection with ula (Monguor fula < *pula; Manchu hula< *pula) with the meaning of “tinder” (Feuerschwamm) 71, a name that lets us go back to a root *pul- or *pula- with the meaning “fire”, which can be directly connected to the MK pɨl  “fire”. In Mongolian, too, ulaγan, being connected to “fire”, probably had the meaning "luminous" besides the meaning "red"; the name of the town of Ulan Khoto, that in Chinese is written with the character wang  “king”, lets us suppose that the title “luminous” (ulaγan) was attributed to the king also by the proto-Mongolian populations. Hence we have a parallel with the name of the king of Paekche, a name which is clearly of northern origin.

The name of the king of Paekche mentioned by the Chou-shu as a title given to the king by the people, Kjʌnkitsi, is equivalent to that quoted by the Nihon shoki, Konikishi 72. Since this title has already been analyzed in another occasion 73, here it will be enough to briefly remember that the name is made up of two parts, the first one (kjʌn = koni ) should probably correspond to “king, prince (with the meaning of the Latin princeps )” and the second (kitsi = kishi ) to “sir”. This name, on the other hand, seems to reveal a norhern origin, but the fact is not so sure.

From what has been expounded until now it appears clearly that the language of Paekche, perharps started originally from a state of bilingualism, became probably a language of a northern type, even admitting that there could have been some contributions from the language of the conquered people. This, by now, is what, at a rough estimate, we could find out.

The language of Paekche (1)

Hypotheses on Paekche (1)


  1. See Pak Pyŏng-ch’ae, Kodae..., cit., p. 122-124. On the palatalization of the language of Koguryŏ, see also Kim Wan-jin 金完鎷, Koguryŏŏ-e issŏsŏ-ui t kugaeŭmhwa hyŏnsang-e taehayŏ 高句麗語에 있어서의 t 口蓋音化現象에 對하여, in Yi Sung-nyŏng paksa songsu kinyŏm nonch’ong 李崇寧博士頌壽紀念論叢, Seoul 1968, p. 135-141. For another phenomenon, the drop of the intervocalic -k-, Paekche seems to precede the other two states (see Pak Pyŏng-ch’ae, Kodae..., cit., p. 124).
  2. Quoted in the section concerning Koguryŏ, because the Paekche’s territory where it was, was in 475 A.D. occupied by Koguryŏ. In this work the old names written in Chinese characters are given in the modern Korean pronounciation, if a more accurate analysis is deemed not essential.
  3. Mich’uhol is definitely a name of Paekche, because it is mentioned in the tale of the foundation of the state, even if, because of the fact that the area where it was located became afterwards territory of Koguryŏ, is listed in the Samguk sagi among the place-names of Koguryŏ. Because of this fact it remains a doubt, small indeed, that the last character hol could have been added by Koguryŏ to the first two characters, substituting another previous character with the same meaning. But the short period when this territory remained under the control of Koguryŏ (from 475 to 551-552) seems to suggest the exclusion of such an hypothesis.
  4. See Samguk sagi, kwŏn 37, p. 579 of the cit. ed.: 其買縣一云林川. The correspondence of the character to the character lets us think that “wood” in the language of Paekche was ki (B. Karlgren, Grammata serica recensa, Stockholm 1957, 952a: *ki̯əg / kji). This ki can be connected to the form kɨnɨl “tree” of the Koguryŏ language and with the ancient Japanese (see Yi Ki-mun, A Genetic View..., cit., p. 102) and perhaps with the MK .kɨ.nɨl “shadow”. On the contrary, S. E. Martin advances the hypothesis that the Japanese ki “tree” could be connected with the Korean khi (MK chi ) “rudder” (see S. E. Martin, Lexical Evidence Relating Korean to Japanese, in «Language», 42, 2, 1966, p. 227).
  5. See Yi Ki-mun, Kugŏsa..., cit., p. 68.
  6. About the division of the Korean language in “ancient” (AK), “middle” (MK) and “modern”, see Yi Ki-mun, Kugŏsa..., cit., p. 22-23, and Lee Ki-moon (Yi Ki-mun), Linguistics, in Korean Studies Today, Seoul 1970, p. 82-83.
  7. See Samguk sagi, kwŏn 37, p. 579 cit. ed.: 加知奈縣一云加乙乃. On nae (na) with the meaning of “land”, see Yi Ki-mun, Kugŏsa..., cit., p. 69.
  8. See Karlgren, Grammata..., cit., 5031, 1240c, 318b, 945a. In these quotations, the numbers correspond to those of the individual characters in the graphical form that interests us. In quoting the Chinese pronounciations we neglected (when not expressely specified) the modern Chinese pronounciation, and limited ourselves to transcribe the archaic (first Chou) and the ancient (Ch'ang-an language, around the 600 A.D.) forms.
  9. See J. E. Kowalewski, Dictionnaire Mongol-Russe-Français, Kasan 1846, repr. 1964, vol. II, p. 968.
  10. See N. Poppe, Vergleichende Grammatik der altaischen Sprachen, 1, Vergleichende Lautlehre, Wiesbaden 1960, p. 88, and G. Clauson, An Etymological Dictionary of Pre-Thirteen-Century Turkish, Oxford 1972, p. 652.
  11. The first documents written in Korean alphabet date back to the 1445. The creation of an alphabet which accurately established the sound of the vowels and of the consonants related to the sounds of the Chinese characters used at that time in Korea (and already pronounced in a way that was different from China) marks a fundamental date for the following studies on the phonetics of the MK (see Yi Sang-baek 李相佰, Han’gŭr-ŭi kiwŏn 한글의 起源, Seoul 1957). Therefore, being the forms of the 15th century the most ancient that were written in a very accurate and exact phonetical alphabet, in this work we will refer very often to them when we shall have to reconstruct pronounciations not reconstructable in other ways.
  12. We preferred to connect xol with the mongolian xorγa, instead of the manchu holo “valley” and the MK kol “valley, cave” (see Yi Ki-mun, Kugŏsa..., cit., p. 79). Also in a place-name of the section devoted to Koguryŏ there is the character , apparently with the meaning of “valley” (see Samguk sagi, kwŏn 37, p. 578 of the cit. ed.: 大谷君一云多知忽). Anyhow, the word “valley” in the Koguryŏ language is generally rendered with the characters 呑, 旦, corresponding to a pronounciation tan, that can be linked to the japanese tani “valley” (see Yi Ki-mun, Kugŏsa..., cited, p. 70). The MK .ulh can perhaps be also connected with the Turkish ula:ğ «something joined on» (see Clauson, An Etymological Dictionary..., cit., p. 136a).
  13. See Kim Sŏn-gi 김선기, Han-il-mong tanŏ pigyo 한·일·뫇단어비교 in «Han’gŭl» 한글, 142, Seoul 1968, p. 27.
  14. See note 28.
  15. The names are given in the Korean pronounciation reconstructed keeping in mind both the reconstruction of the Karlgren's Chinese pronounciations and the pronounciations of the Nihon shoki concerning these names.
  16. See Karlgren, Grammata..., cit., 61e, 6a, 33h, 249c, 393a, 864a, 61e,1032f. Since the pronounciation of the Chou-shu characters is involved, we quote here just the ancient form given by Karlgren (see note 50) and not the archaic form.
  17. On the pronounciation of the affricates in middle Korean, see Yi Sung-nyŏng 李崇寧, Chungse kugŏ munpŏp 中世國語文法, Seoul 1961, p. 35. We are inclined to believe that the pronounciation of the affricates [ʧ] and [ʧh] of modern Korean goes back to [ʦ] and [ʦh].
  18. See NS, Iwa, K. 19, Kimmei tennō 欽明天皇, 7th year, p. 95 of the 2nd vol.: 夫人 (オリクク). The edition Kokushi taikei 國史大系 of the Nihon shoki (Tōkyō 1952), on the contrary, gives the readings orikeno and woriworo (I-2 p. 73). Maruyama Rimpei 丸山林平, Jōdaigo jiten 上代語辭典, Tōkyō 1967, p. 405, gives the reading oriku.
  19. See Nihon shoki, ed. Kokushi taikei, K. 14, Yūryaku tennō, 20th year, I-1, p. 388: 大后 (コムオルク · コムヲルク).
  20. See NS, Iwa, K. 26, Saimei tennō 齊明天皇, 4th year, p. 336 of the 2nd vol.: 王后 (オニオルク). The koniworuku form is quoted by Yang Chu-dong 梁柱東, Koga yŏn’gu 古歌硏究, Seoul 1969, p. 605.
  21. See F. D. Lessing, Mongolian-English Dictionary, Berkeley 1960, p. 885 and 927; see also Kowalewski, cit., p. 718 of the 2nd vol., and Poppe, Vergleichende..., cit., p. 142. In ancient Turkish too the term uruğ could mean “progeny” or “clan” (see Clauson, An Etymological Dictionary..., cit., p. 214b).
  22. See N. Poppe, Grammar of Written Mongolian, Wiesbaden 1964, p. 74, § 282.
  23. The genitive suffixes generally used in MK were -ăi / -ŭi, -i and -s.
  24. Perhaps, could this word also be in relation with the MK ə.rɨ - “to join sexually, to temptate, to exite sexually” and the MK ərun “an adult, a married person”? The MK ə.rɨ - goes back to a reconstructed ancient form *erə - (which is strangely similar to the greek ἔρως “love”), but this form differs very much from the pronounciation woruku / ·uo-li̯uk.
  25. The NS, Iwa (2nd vol., p. 554, note 10) reads it orake and puts it in relation with the reading orikoke of the character , when it concerns the king of Koguryŏ: (orikoke); 狛王 (koku-orikoke) (NS, Iwa, K. 19, Kimmei tennō, 7th year, p. 95 of the 2nd vol.). See also NS, Iwa, 2nd vol., p. 95, notes 28 and 29.
  26. Pelliot, speaking of the Mongolian titles of the Koryŏ-sa 高麗史, proposes the link of the (woman's) name Hou-la-tch'e 胡剌赤 with the Mongolian hula'an > ulān “red” (see P. Pelliot, Les mots mongols dans le Korye Să, in «JA», 217-2,1930, p. 261). The initial voiceless bilabial fricative which here still appears and that certainly was present in the proto-Mongolian at the time when the Chou-shu was compiled (which gives us the passage quoted in the note 24), leaves nevertheless open the doubt on the validity of the correspondence Wolaγa-Ulaγan (< *pulagan) to which we pointed. If this correspondence is valid, then this title shows that in the Paekche's language the transformation of the initial voiceless bilabial occlusive (*p) of the proto-altaic form took place rather early compared to the other altaic languages. The pronounciation ·uo given by Karlgren for the character suggests the passage *p > *b > > w. The sonorization of the proto-altaic *p- is suggested for the Turkish and for the Mongolian by Sinor (cf. D. Sinor, Ouralo-Altaïque - Indo-Européen, in «T'oung Pao» 37, 1943, chart at p. 235), but N. Poppe (Introduction to Altaic Linguistics, Wiesbaden 1965, p. 197) reconstructs for the proto-Altaic two phonemes *p and *b, distinguished by the sonance. The validity of the correspondence Wolaγa-Ulaγan is even more questioned by the presence of the same character in the first syllable of the name Woljuk. The fact that the same character was used lets us think that there were no difference in the pronounciation of the first syllable of both names, but, if for Wolaγa we can assume that the w- comes from an initial *p- of the proto-Altaic, will it be possibile to do the same for the w- of Woljuk when we put this term in relation with the Mongolian uruγ ? On the other hand the initial w- of Woljuk is proved also by a reading of the Nihon shoki, and because of the etymologies quoted in the text it is correct to think that the Mongolian uruγ and the forms ·uo-li̯uk and woruku be in close kinship. Another hypothesis could be to say that the character was not used to transcribe the pronounciation uo, but u, and this would solve the Woljuk-Wolaγa problem, but the readings of the Nihon shoki go against this solution. The derivations proposed in the text, which on the other hand appear to be evident enough, leave open many problems, that by now we are not able to solve.
  27. Quoted by Yu Ch'ang-don 劉昌惇 in Ijoŏ sajon 李朝語辭典, Seoul 1964, p. 369.
  28. See Poppe, Vergleichende..., cit., p. 74.
  29. See Poppe, Vergleichende..., cit., p. 12. Ki-Moon Lee (Yi Ki-mun) in A Comparative Study of Manchu and Korean, in «UAJ», 30, 1-2, 1958, p. 110, links the term Mo. ulagan (< hulaγan < *pulaγan ) with the Ma. fulgiyan “red”, Ma. fulahūn “light red”, Nü. fû-lâ-kiāng id., fu-liang id., and the MK pi̮rk- “to be red”.
  30. See NS, Iwa, K. 14, Yūryaku tennō, 9th year, p. 483 of the 1st vol.: 百濟王 (クダラノコニキシ).
  31. See V. Anselmo, Armonia consonantica nel coreano antico, in «AION», 33, 1973, 1, p. 61-76.

To the top of the page
© Valerio Anselmo