Author's comment: Japanese names referred to in the body of this essay and in the notes are given in the Japanese way–last name first.

1. Although in this essay I impose my own perspective upon the question of the relation of movement to song I want at the outset to give a brief version of how a Japanese scholar, whose ideas have been influential on my own in this area, Professor Honda Yasuji, regards the link between movement and word in dance to song in Japan. (A presentation, related to the following though from a slightly different angle, may be found in Nihon no matsuri to geinō, Honda Yasuji, 1974, p. 42Google Scholar).

Professor Honda thinks there are three kinds of relationship in Japanese dance. First, song may be used as a musical accompaniment for dance. The human voice as it sings is an instrument which provides rhythm, for example, as a drum might, or a melodic line, as a flute might. Second, movement may be an attempt to “express” something, a feeling perhaps, in the sense of the song. This relationship—movement expressive of the meanings of a song—Honda finds in dance both in India and in Japan. Third, movement may be made up of specific gestures used as “translations” of meaning in song. For example, in dance which is characteristic of the main stream of Kabuki, words like “write,” “brush-pen,” “letter” etc. may be rendered by a dancer's gesture of manipulating his folded fan as if writing.

2. Dances where a strong element of possession remains alive today, such as the miko (priestess) dance at the Haushiwake shrine to be discussed later in this section, are the exception. Another exception, though strictly speaking this is considered an example of Izumo kagura by Honda rather than miko kagura, is Ōmoto Kagura of Shimane Prefecture. The dividing line between a true shamanistic event and choreographed dance is clearly one which needs to be reassessed. Comparative evidence from other parts of the world may give us more rigorous criteria in the future. The term miko mai (dance of a priestess) is the term used by Honda in his classification system of dance in the countryside to designate dance of the type mentioned here as derived from the movements of a performer in true shamanism. It may be helpful to have a translation of Honda's definition of this type of dance; it is hypothetical in part, showing his own sense of the limits of the evidence in Japan.

Pillars and poles etc. can temporarily be inhabited by a god (kami no yoriyasui mono.) Sometimes people themselves become the residence of a god (shinza) … What is called possession dance (kamigakari no mai) begins first with a dance in which the dancer purifies (kiyomeru) himself. The dancer holds grass or a bell (suzu) in his hand. In this way shaking the grass or sounding the bell, he continues just turning around and around in one place until he enters a state where he is no longer himself (muga.) What is performed under the name ”miko kagura” is evidently what resulted when the dance (mai) element of this process was abstracted as a single entity. (Kagura, , Nihon no minzoku geinō #1, Mokujisha, Tokyo, 1966, p. 29Google Scholar).

A discussion of miko kagura among other types is found in “Kagura: the Search”, King, Eleanor, Dance Research Monograph One, 1971–1972, New York: CORDGoogle Scholar.

3. We may have to reconsider the adequacy of a strict polarity, just as we may need to rethink the question of “spontaneous” vs. choreographed dance in the shamanistic tradition. The relation of an art form intended for appreciation to its ritual underpinnings is described in “Yamabushi kagura and bangaku: Performance in the Japanese Middle Ages and Contemporary Folk Performance.” Honda, Yasuji, translated by Hoff, Frank, Educational Theater Journal, May, 1974Google Scholar. See especially the section “From ritual stamping to choreography for dance.”

4. See Hoff, Frank, “The Evocation of okina and his katari. A Performance Version of Ritual Shamanism.” to be published in Alcheringa. Ethnopoetics. New Series, Volume Three, Number 1, Boston UniversityGoogle Scholar.

5. Shimotsuki Kagura no kenkyū. Honda, Yasuji. Meizendō, Tokyo, 1954, p. 667 ffGoogle Scholar. Songs translated include numbers 1, 4, 3, 6, 10.

6. Kagura, p. 171.

7. Nō no kenkyū, Kanai, Kiyomitsu. Ōfusha. Tokyo, 1969, pp. 139–140Google Scholar.

8. A study such as we find in Kodai kayō to girei no kenkyū, Tsuchihashi Yutaka (p. 161 ff)Google Scholar of the primitive magical efficacy of holding some green and natural object, bough of leaves for example, in the hand would throw light on an archetypal experience of renewal which has made it inevitable even in later Japanese dance types for something held in the hand, if only a fan, to be a necessary concomitant of its paraphanalia and choreography.

9. Honda, Yasuji, Ennen, Mokujisha, Tokyo, 1969Google Scholar.

Honda himself danced Hana-ori a few years ago at a private performance to mark the anniversary of the death of dance scholar Kodera Yūkichi.

In annotating this dance, Honda simply records his own observations in words, but he does provide examples of an interesting older manner of documenting dance movement to illustrate the high points of two other dances performed at the Ennen of Mōtsuji. As given in Ennen, these are Honda's own transcription of drawings, one of which dates to 1822, which exist in records kept at the temple. Along the lines of the drawings are written the words of the song which accompanies the movement indicated. The writing at the top of several identifies the main image of the temple, thus indicating the principal frontal axis of movement. At the side of one diagram (#177) is written a fuller verbal description of movement taken from another record.

It is clear, then, that older records of movement in dance can be found in the countryside. These can be studied by themselves as innovative attempts at notational systems and can also be used to compare how dancers today render what was recorded in this way.

Editor's note: An addendum to this article is under consideration for the forthcoming issue of the Dance Research Journal in which will appear examples of this older notation for two dances performed at the Ennen of Mōtsuji. There will be translations of the songs which accompany the movement of the dances.

10. Texts of the Izumi version of komai are given in Kyōgen mai-utai-shū, Nonomura, , Yōkyoku-kai, TokyoGoogle Scholar. Sumiyoshi is on page 10. Nanatsugo is on page 11 and the text of the latter song is annotated with a brief account of movement which accompanies it.

11. See Hoff, Frank, “Furyū Odori. Joint Treasure of Apollo and jthe Violet-haired Muses.” Studies on Japanese Culture, Japan PEN Club, 1973, Tokyo, Vol. 1, p. 623Google Scholar.

12. Nihon koyō-shū, Honda, Yasuji. Miraisha. 1962. p. 209Google Scholar. Stanzas 79, 80, 81, 83.

13. Kyōgen mai-utai-shuū. p. 1.

14. See Hoff, Frank, “Nihon to Girisha ni okeru katarimono no butaika. Hikaku-bungaku- engeki kenkyū” (“The Staging of the Spoken Tale in Greece and Japan. A Study in Comparative Literature and Theatre.”) in Geinō, ronsan (A volume in honor of Professor Honda Yasuji) Kinshosha, Tokyo. 1976Google Scholar.