Sacred Texts  Shinto  Index  Previous  Next 
Buy this Book at

The Kojiki, translated by Basil Hall Chamberlain, [1919], at

p. 75


So thereupon His Swift-Impetuous-Male-Augustness sought in the land of Idzumo for a place where he might build a palace. Then he arrived at a place [called] Suga, 2 and said: "On coming to this place my august heart is pure," 1—and in that place he built a palace to dwell in. So that place is now called Suga. 2 When this Great Deity first built 3 the palace of Suga, clouds rose up thence. Then he made an august song. 4 That song said: 5

"Eight clouds arise. The eight-fold fence of Idzumo makes an eight-fold fence for the spouses to retire [within]. Oh! that eight-fold fence." 6


p. 76

[65] Then he called the Deity Foot-Stroking-Elder and said: "Thee do I appoint Headman 7 of my palace;" and moreover bestowed on him the name of Master-of-the-Temple-of-Inada-Eight-Eared-Deity-of-Suga. 8

p. 77


75:1 p. 76 I.e., "I feel refreshed." The Japanese term used is suga-sugashi, whence the origin ascribed to the name of the place Suga. But move probably the name gave rise to this detail of the legend.

75:2 The real derivation of Suga is unknown, all the native commentators accepting the statement in the text, and Motowori supposing that up to the time of the Deity's arrival it had borne the name of Inada. We may perhaps conjecture some connection between Suga and Susa-no-wo ("Impetuous Male," see Motowori's Commentary, Vol. IX, p. 49), and it may be mentioned that the "Eight-Eared Deity of Suga" is also mentioned as the "Eight-Eared Deity of Susa."

75:3 Or "began to build."

75:4 "Ode" is another rendering of the Japanese term uta, which has been used by the present writer and by others. Uta being however connected with utafu, "to sing," it seems more consistent to translate it by the English word "song."

75:5 Or perhaps rather "in that song he said."

75:6 This difficult song has been rather differently rendered by Mr. Aston in the Second Appendix to his "Grammar of the Japanese Written Language" (2nd Edition), and again by Mr. Satow in the note to his translation of the Ritual already quoted. Mr. Aston (premising that he follows Motowori's interpretation) translates it thus:

"Many clouds arise:
The clouds which come forth (are) a manifold fence:
For the husband and wife to retire within
They have formed a manifold fence:
Oh! that manifold fence!"

Mr. Satow's translation is as follows:

"Many clouds arise.
The manifold fence of the forth-issuing clouds
Makes a manifold fence
For the spouses to be within.
Oh! that manifold fence."

p. 77 In any case the meaning simply is that the multitudinous clouds rose up like a fence or screen behind which the newly-married deities might retire from public gaze, and Moribe suggests that the repetitions are an after-addition made to bring up to the usual number of thirty-one syllables what were originally but the three lines—

Tachi-idzuru kumo mo
       Tsuma-gome ni
Yo-he-gahi tsukuru yo


"The uprising clouds even, to shut up
       the spouses, make an eight-fold fence."

(See his discussion on this song in the "Idzu no Kotowaki Vol. I, pp. 1-3.)—The present writer has already stated in the Introduction (see p. lx) his reasons for always rendering the native word for "eight" (ya) by "eight" instead of by "many" or "numerous," as is done by the two eminent scholars above quoted. With regard to the word Idzumo which they, in deference to the opinions of the native commentators, render by "clouds which come forth" or "forth-issuing clouds" (the Chinese characters with which the word is written having that signification), the present writer cannot persuade himself that such a corruption as idzumo for ide-kumo either retained at the time of the composition of the song, or should now be credited with, the signification which this its supposed etymology assigns to it. The etymology moreover is far from being established, and in this, as in many other cases, the Chinese characters used to write the name of the province of Idzumo, may well have rested on nothing more than a vague similarity of sound, and probably no European scholar would endorse the opinion of the native commentators, to whom the "Records" are a sacred book, that the province of Idzumo received its name from this very poem. On the other hand, we need have no difficulty in conceding that the Pillow-Word ya-kumo-tatsu, by which Idzumo is preceded in poetical compositions, did probably here originate.—This song is in the "Chronicles" only quoted in a note, for which reason some authorities dispute its antiquity. In the note in question, we find the reading -gome (the "Records" have -gomi), the Transitive form instead of the Intransitive. If this were adopted, the translation would have to run thus: . . . . . "The eight-fold fence of Idzumo makes an eight-fold fence to shut up the spouse[s?] in;" and probably "spouse" should be understood in the Feminine to mean "wife."

76:7 p. 78 Obito, written with the Chinese character , while the Japanese word is probably derived from oho-bito, "great man." When used, as it often is, as a "gentile name," the translator renders it by "Grandee."

76:8 Inada-no-miya-nushi Suga-no ya-tsui-mimi-no-kami. It should be stated that Motowori, as usual, objects to the view that mimi signifies "ears" (its proper meaning) in this name. But he has no better explanation to offer, and the Chinese characters give us ya-tsu mimi, "eight ears." The author of the "Tokiha-gusa" ingeniously proposes to consider ya-tsu mimi as a corruption of yatsuko mi mi ( ) "servant august body," but this cannot be seriously entertained (Conf. Sect. XIII, Note 18).

Next: Section XX.—The August Ancestors of the Deity-Master-Of-The-Great Land