Transactions of the korea branch of the royal asiatic society volume XXXVIII

VIII The Hwarang Songs.
Our knowledge of the vernacular poetry of Silla is restricted to a meagre collection of fourteen pieces, usually known as hyangga (鄉歌), preserved in the Samguk Yusa. It is not usual to doubt that they really represent the songs of the period they claim to come from. The only other poems of the same genre are those of the monk Pyŏn Kyunyŏ (邊均如). There are eleven of these. He lived from 917 to 973, spending most of his life under the Koryo (高麗) dynasty after the disintegration of Silla. His eleven surviving poems are recorded in a biography compiled in 1070 (Kyunyŏ Chŏn 均如傳). They are buddhist, didactic and devotional. Thus all the texts we have of Silla poetry are very late.

The study of these poems is fraught with many difficulties, not least linguistic ones, owing to the fact that they are written in the complex system called idu (吏讀) using Chinese now ideographically, now phonet- [page 47] ically, in a primitive and clumsy attempt to record native Korean sentences. The initial work of deciphering these poems in modern times was done by Dr Ogura Shimpei (小倉進平)³⁾. But the labours of this great pioneer have been largely superseded by the work of Dr Yang Chudong⁴⁾.

Although there are debatable matters yet in the details of understanding the hyangga, the conclusions of Dr Yang are at least sufficiently established to give us a fair interpretation of the songs insofar as they affect our knowledge of the hwarang. But since the original texts are meaningless without extensive linguistic commentary, it seems that it will be enough here if I offer straightforward translations into English, prescinding from the discussion of either the philological or the literary aspects of the poems. The curious can refer to Dr Yang’s work and to the notes in such a work as Dr Kim Sayŏp’s history of Korean literature⁵⁾ of which I have also made considerable use.

However, I will prefix the translations with the note that at least three names are used to describe this genre. The names are hyangga (鄉歌), tonnorae or tosolga (兜率歌), and saenaennorae (詞腦歌 or 思內歌). The etymology of these words is a matter for discussion, but we shall not be far from the meaning of them if we regard it as being “Korean song”, except in the case of tosolga, which could also be a religious hymn, and must be mentioned again more specifically later.

The corpus contains six songs that were written by hwarang or about hwarang. Two them are by one man, so five authors are represented. One of them is apparently defective. Again it will be best to take them in chronological order as nearly as possible.
3) 鄉歌 及 吏讀研究 Seoul, 1929.

4) Op. cit., above page 7 note 2. Many writers treat the Toijang-ga (悼二 將歌) of the eleventh century as hyangga, bnt their form is notably different.

5) 金恩库: 改稿國文學史 Seoul 1956.
[page 48]

The first claims to be quite early, from the reign of Chinp’yŏng (579-632). It has already been referred to⁶⁾

The Song of the Comet (彗星歌)
See the fort by the Eastern Sea

Where Gandarva used to play.

The islanders have come,

There are the beacon flames.

But the moon, hearing

The Three Flower Boys are visiting the hills

Quickly shows her beams.

A star sweeps a road.

Some say: “See, a broom-star!”
Lo! The moon has gone down.

Now, what comet could this be?

According to the Yusa story⁷⁾ the author was the buddhist monk Yungch’ ŏn (融天師). The Gandharva (乾達婆) are the musicians of Indra, the sky god, associated with the moon and medicine, and also with ecstasy and eroticism. Yang Chudong suggests that the whole phrase means a mirage⁸⁾. The word for comet is etymologically “broom-star”.

The song is not by a hwarang, it adds nothing to what we have already learned from the accompanying story about the three youths and their connection with the buddhist monk. It is of interest to us in the hwarang connection, chiefly as indicating what their songs could be like. The same is true of most of the following songs too.

6) Above page 33.

7) Cf. above page 33 note 1.

8) Op. cit. page 574.
[page 49]

The next song is the one written by Siro about the senior hwarang, Taemara. In spite of the doubt referred to above about the Yusa’s dating of this story, it seems reasonable to attribute the song to the reign of Hyoso (692-702), or earlier.⁹⁾

Song of Yearning for the Flower Boy Taemara (慕竹旨郞歌)
The whole world weeps sadly

For departing Spring.

Wrinkles lance

Your once handsome face.

For the space of a glance

May we meet again.

Fair lord, what hope for my burning heart?

How can I sleep in my alley hovel?

Again there is little here to detain us, beyond noting the intensity of the expression of devotion.

About fifty years later come the next two songs, from the reign of Kyŏngdŏk (742-765), both by the same singer, Wŏlmyŏng.¹⁾ Wŏlmyŏng was no longer a hwarang when the songs were written, but he states clearly enough that the first of the two was written in the hwarang style. In the Yusa it is described as a tosolga. In other connections this word is taken to refer to the refrain or genre of Korean songs of Silla,²⁾ but in this case it is undoubtedly an ordinary buddhist term and refers to the Tusita Heaven (兜率天) within which is the pure land of Maitreya. In fact this is a devout song in honour of Maitreya. It is unusual in that the Yusa itself provides a Chinese paraphrase of it.

9) See above page 40. Also, for an earlier date and another interpretation of the song, Sin Susik (辛秀植:) in 국어국문학 No. 23, Seoul, 1961. Pp. 13ff.

1) See above page 42.

2) Cf. Yang Chudong op. cit. page 525.
[page50]

兜 率 歌

龍 樓 此 日 散 花 歌

挑 送 靑 雲 一 片花

殷 重 直 心 之 聊 使

遠 邀 兜 率 大 僊 家

Or, in translation from the Korean:
The Tusita Hymn
O flowers scattered here today,

As we sing the scattered petals song,

Heed the orders of this upright heart

And haste to serve Maitreya’s throne.

It is notable that the character used to indicate Maitreya in the Chinese translation (僊) is a variant of the character (仙) used so often to indicate the hwarang themselves. According to the story Maitreya answered the prayer in person.

The second song is a prayer for a departed soul.

Song for a Dead Sister (爲亡妹營齋歌)

Here is the road of life and of death.

Were you afraid

When you left, not even saying:

“I am going now”?
Like fallen leaves borne here and there

On the early autumn wind,

Though born from the same branch

We know not whither we go.

How I strive to perfect my road, waiting

To meet you in the abode of Amitabha.

This is strictly a buddhist song, relevant to hwarang solely because of the history of the author.

The next song dates from the same reign, and is attributed to the monk Ch’ungdam (忠談師). Its con-[page 51] nection with hwarang is merely that it praises one of them. The Yusa¹⁾ gives no further information about the circumstances of its composition.

Song in Praise of the Flower Boy Kilbo (讚耆婆郞歌)
Moon

Appearing fitfully

Trailing the white clouds,

Whither do you go?

The face of the Flower Boy Kilbo

Was reflected in the pale green water.

Here among the pebbles of the stream

I seek the bounds of the heart he bore.

Ah, ah! Flower Boy hero,

Noble pine that fears no frost!

There is little here to comment on for the sake of learning more about the hwarang. At this distance of time it is hard even to perceive the exact nuance of the emotion. But we do know that the author was in the habit of making a libation of tea every year to Maitreya at the Samhwaryŏng (三花嶺), which may have been a hwarang memorial.

The last of the series is the latest song recorded by the Yusa, and possibly the most famous of all. It is said to date from the time of Hŏn’gang (875-886). It is the shaman song that passed into the repertoire of the Koryŏ dynasty in an extended form and was believed potent against diseases.²⁾ Neverthless on reading its earliest version one is subject to many imaginative possibilities as to what its origin may have been. It belongs to the story of Ch’ ŏyong.³⁾

1) See above page 43.

2) Cf Kim Sa-yop, op. cit., page 291.

3) See above page 44.
[page52]

Chŏyongs Song (處容歌)

Playing in the moonlight of the capital

Till the morning comes,

I return home

To see four legs in my bed.

Two belong to me.

Whose are the other two?

But what was my own

Has been taken from me. What now?

Again the text of the song adds nothing to what we can learn from the story itself as already recounted.

In fact the songs do not materially help in our reconstruction of the hwarang picture at all. They are interesting as being the nearest we are ever likely to get to the actual words and thoughts of the hwarang. But even so they are full of obscurities and difficulties It is easy but dangerous to treat modern translations of such old texts as though the full implications of the English were present in the mind of the original writer. We cannot accurately fathom the emotions of their structure and meaning. But they do give us a glimpse of the degree of sophistication to which poetry in Old Korea had arrived, and infuse a little life into our discussion of p’ungnyu.⁴⁾

4) After completing the above section on the hwarang songs I first had opportunity to examine Peter Lee’s Studies in the Saenaennorae — Old Korean Poetry (Serie Orientale Roma XXII), Rome 1959. This was the first discussion of the poems of Silla to be published in the English language, and is of great interest to readers of English who are not familiar with Korean. It is in the main a digest of Dr Yang’s work.

Mr. Lee translates hwarang as ‘knight.’ I believe I shall have here adequately demonstrated why I believe this to be a misleading translation. His versions of the songs differ in several respects from my translations given above.

I notice in addition that Mr Lee seems not to have been acquainted with Mishina’s book on the hwarang, and so he accepts the fact that Taemara was still a hwarang though in old age, and thus commits himself to the corresponding interpretation of the song and story in question. (Op. cit. page 106).
[page53]

VIII The Hwarang after Silla
Fascinating though the problem of the hwarang of Silla is. it is scarcely more fascinating than the problem of how they faded from the scene. That the institutions of any dynasty should become effete with the decline of the ruling power is not in itself remarkable, and this seems to what happened to the hwarang. I think that Professor Reischauer has begged altogether too many questions in his statement that “The Hwarang bands lost their fighting prowess and degenerated into groups of effeminate dilettantes”⁵⁾ partly because he relies on the common modern idea that the hwarang were primarily warriors, but also because the word effeminate seems stronger than the evidence will support.

Indeed the factor that makes the disappearance of the hwarang so interesting is the suggestion that their development was not so much a matter of degeneration as of specialisation in some aspects of their original activities in the changed conditions of later society. I said early in this paper that the lexicography of our subject shows that the word hwarang has come to be connected with shamans and travelling singers. Since music and religion are significant elements in our earliest descriptions of hwarang, it seems altogether too much to ask us to believe that some vital sea-change overtook the hwarang of late Silla and they changed their character completely, although that is the view that some modern writers would urge us to take.

There is however the danger of the contrary error, which would ride equally rough shod over the paucity of the evidence and accept as proven what is in my view only the best tenable hypothesis, that the hwarang were a religious cult that was the direct ancestor of the
5) E. O. Reischauer and J. K. Fairbank, East Asia the Great Tradition, London,I960, page 415.
[page 54]

later dancing boys. This is the view represented by Ayukai and Mishina.

The writers of the Koryŏ period (918-1392) have a group of references to the subject that read more like reminiscences than historical accounts, explaining Koryŏ facts in terms of Silla history. The material has been fully collected by the two Japanese scholars, and a full discussion can be found in Mishina’s work.⁶⁾

The key point in the matter seems to be the palgwanhoe (八關會) a festival of the Koryŏ dynasty which is recorded as having been founded in the 33rd year of the reign of Chinhŭng, the Silla king who organized the hwarang. At this festival the dancing of a group called the sŏllang (仙郞) was a distinctive feature. I have already quoted the article by An Kyehyŏn⁷⁾ in which he argues for an earlier date for the founding of the p’algwanhoe and also for the fact that the hwarang performed at it from its inception.

This festival continued to be observed throughout the Koryŏ dynasty, only finally disappearing with the advent of the Yi and their strictly confucian policies. It was a blend of buddhist with earlier Korean religious elements. In the later days the sŏllang who took part in it were understood by the men of the time to be the direct descendants of the hwarang. The tradition of touring the country to sing and dance in the high places was remembered. The four known as the Sasŏn (四仙), An Sang (安詳), Yŏngnang(永郞), Sullang (述郞), and Nam Sŏkhaeng (南石行), whose name was later associated with a religious dance (四仙舜), were especially recalled in the Tongchŏn area. They are clearly the men mentioned already.⁸⁾

In the Koryŏ dynasty the name was given to youths

6) SKK pp 273ff.

7) Above page 19 note 9.

[page 55]

trained in buddhist monasteries, and under the reign of Ch’ungnyŏl (忠烈 1275-1308) the title of sŏllang was in use for such lads at the palace.⁹⁾ And then Ayukai discovered that there was an apparently revived military significance for the word at this period. But the references suggest that by this time there was an antiquarian element in ideas about the institution.

There is thus ample evidence that the idea of kuksŏn neither died nor came into disrepute during the Koryŏ dynasty, although the classic institution and the word itself seem to have been no longer in use. The next problem is how did the name get transferred to the wandering players, and why did the word hwarang survive with them when it was a Silla word, and the later words were sŏllang and kuksŏn?

So far as I can discover we have only one literary link between the hwarang and the players⁸⁾. This tells us that the masked dance plays and the Ch’ŏyong dance were performed at the p’algwanhoe of the Koryŏ court. When, during the Yi dynasty, the masked dance play became the property of the lower classes and its religious origins got more heavily overlaid with peasant satire and slapstick at the expense of the clergy and gentry, it might be natural for the players to be called by the name they had had in the days when it had been a court performance.

A possible explanation of the vocabulary difficulty may be suggested by the fact that the Koryŏ court was centred on the central and north western part of the country, whereas the term hwarang has survived better in the South and East where the original hwarang had lived.
8) Cf. Tongguk Yŏji Sŭngnam, article on T’ongch’ ŏn (通川). Also above page 41.

9) Cf. SKK pp 287ff. Cf also note on Min Chong yu above page.

8) The details of references to the Koryŏsa and others are given by Yang Chaeyŏn (梁在淵) in his paper on the Sandaegŭk (山臺戱) in the 30th Anniversary Commemoration Theses of Chungang University, Seoul 1955.
[page 56]

This area also incidentally represents some of the most strongly buddhist parts of the country during the Yi dynasty, and the connection between the wandering players and the buddhists is an important one. Even long after the players had become much busied with satire against the monks, they were normally sheltered in the temples during the off seasons, and they were at times associated with the bands of begging monks and others working for the temples or other rural communities and which were called kŏlliptae (乞粒險) and performed dances and acrobatic tricks very like those done by the troupes of male entertainers called sadangp’ae (寺堂牌 or 男寺黨牌). Either of these groups might be associated also with shamanistic activities (kut 굿) and various sacrifices. In any case the troupes would almost certainly contain boys called hwadong (花童) or hwarang (花郞)⁹⁾ (For an account of their sexual behaviour see the following section).

These were a low class of society, generally outcast or even feared by the common people. Racially many of them derived from the so far little studied mujari (무자리, 揚水尺)¹⁾, the so-called “Korean gipsies”, who appear in the Three Kingdoms period as nomads keeping cattle and selling basketwork. In later days they were noted for producing exorcists and as the forbears of the strolling players, prostitutes, butchers and other low caste trades. They were of very different stock from the first hwarang, it would seem, but in the end their descendants inherited the hwarang name.

The transference of the hwarang title would seem to have been a natural one. If the latter day hwarang lacked the qualities that gained universal reverence for their forerunners, at least the popularity which they did enjoy was based in what were essentially the same activities transferred to another social plane.

9) cf Ayukai op. cit pp 78ff.

1) cf. Ch’oe Sangsu, A Study of the Korean Puppet Play, Seoul 1961.

[page57]

IX The Question of Homosexuality
The question of homosexuality in connection with hwarang needs to be considered if only because Ayukai²⁾ has brought it up. But there is also the fact that the latter-day hwarang, the players, are notoriously a homosexually liable group.

During the Yi dynasty homosexual practices were regarded with disgust by the confucian gentry. The early apologists of the dynasty held paederasty as one of the crimes of the Koryŏ kings. The most famous case is the scholar-painter-calligrapher King Kongmin (恭愍王 reigned 1352-1374) who towards the end of his life appointed royal catamites called chajewi (子弟衛) five of whose names are recorded: Hong Yun (洪倫), Han An (韓安), Kwŏn Chin (權瑨), Hong Kwan (洪寬) and No Sŏn(盧瑄)³⁾.

The word used to describe homosexual activity in the Koryŏ Sa in this connection is the Chinese literary expression lung-yang-chih-ch’ung (龍陽之寵, Korean pronunciation yongyang-chi-ch’ong), which does not have the expected reference to the two male symbols, the dragon and the sun, but refers to a favourite of the feudal lord of Wei (魏) who was known as Lung-yang⁴⁾. This is a very intellectual expression and has little bearing on the subject in Korea. Like such modern expressions as namsaek (男色) and the more figurative kyegan (鷄姦) it is literary.

But the Korean language has a native vocabulary on this subject which is not obviously created by a figurative use of words. In Chinese, as in English, the relevant vocabulary is entirely either figurative or learnedly concocted. But in Korean we find piyŏk (비역) as a verb and a verbal noun, while both myŏn (면) and

2) Op. cit. page 36.

3) 高麗史 列傳嬖辛二 金興度傳

4) See 戰國策 卷第二十五 魏四
[page 58]

t’otchangi (톳장이, possibly a figurative expression) are used for a catamite. For native words of no clear etymological significance to have survived suggests that the practice has at some time had an appreciable place in the culture.

It is certain that homosexuality was well known in rural society during the Yi dynasty. I have heard of it from older men in the villages of South Kyŏnggido, and Bishop Cooper has spoken of its occurrence in the same area at the beginning of this century. There is a vaguely unsavoury reputation sometimes connected with the chibang yangban (地方兩班) or provincial gentleman, but I heard in the villages more of the practices of the lower classes, among whom, for instance, paederasty seems to have formed a recognized outlet for a young widower, and caused very little stigma to be attached to his favourite who on growing older could turn to normal sexuality in marriage. I was told that the presents, especially of clothing, given to the boy would make his status public knowledge in the village.

Something of this is also reflected in the salacious chatter of village youths and in some versions of the Kkoktu Kaksi (꼭두각시 ) traditional puppet play of Korea. In these the wastrel hero, Pak Ch’ ŏmji (朴僉知), who is a satire on the provincial yangban, is made to have spent some of his money on a pretty boy, midongaji (美童아지)⁵⁾, and a number of coarse homosexual puns are introduced. Collections of coarse anecdotes such as the Myŏngyŏp Chihae (蓂葉志諧) of Hong Manjong (洪萬宗, flourished, during the reign of Sukchong 1675-1721) contain tales, though not many, of like ‘import. It is also noticeable that the essentially innocuous expression used in the Kkoktu Kaksi text referred to, midong (美童), meaning a goodlooking boy, usually carries overtones of paederasty. It is used thus in Korean translations of the bible to translate the Hebrew qadesh or male prostitute⁶⁾.
5) Ch’oe Sangsu (崔常壽), A Study of the Korean Puppet Play, Seoul 1961.

6) I Kings xv. 12, xxij. 46, et al.

[page 59]

But the word midong and the reputation for homosexuality were particularly attached to the wandering players and musicians. It was almost normally assumed that the all male teams used the boys as catamites. They were dressed attractively, often in girl’s clothes, though not always, and on occasion it seems that they were also prostituted. Professor Ch’oe Sangsu tells me that regular berdache marriages were sometimes entered into within the bands, and he has met and interviewed such cases.

Song Sokha in the work already referred to⁷⁾ speaks of the namsadang as troupes of peformers whose chief purpose was to earn money as boy prostitutes, and says that they were formed on the analogy of the strange husband-and-wife teams for travelling prostitution which were a feature of rural Korea from the middle of the Yi dynasty onwards. He claims that the male teams were not set in circulation until the end of the dynasty, but adduces no evidence for this statement. He points out that they were often associated with buddhist temples and with young monks collecting alms for their establishments. It is in such a connection that one may come across the word namch’ang (男娼) or boy entertainer.

The namsadang, however, were also associated with shamanistic practices. Very occasionally one can still meet them performing kosa (告祀) and other religious ceremonies in the Korean countryside, were the boys have a specific role in the dancing. The connection of shamanism with homosexuality is a little obscure. Transvestitism is a common and well attested practice for shamans of both sexes, although in the recent periods transvestitism among Korean adult male shamans does not seem to have been normal. Homosexuality among the shamans of Siberia is attested by various writers⁸⁾,

7) Op. cit. page 102.

8) Cf Ch’oe Namsŏn (崔南善) Salman Kyodapki (薩滿敎剳記) in Kyemyŏng (啓明) No. 19, Seoul, 1927.

[page60]

but modern Korean shamanism has diverged in many respects from the forms found in the more primitive cultures of north-east Asia, and it would be rash, without further evidence, to suggest too close a relation between shamanism and homosexuality in this country.

The namsadang itself is not a very clearly defined institution. It shades off into the allied groups of begging monks on the one hand and into the village farmers’ bands on the other. I do not know of any really adequate study of this aspect of Yi dynasty society. I know of one case where the village band maintained a midong chosen for his good looks, who was not expected to work, but to dress prettily and entertain the labourers. He had reached the age of twenty and still held this position, which was beginning to be thought of as undesirable by other people. There was no clear imputation of paederasty, but a strong sense of inversion setting in.

This then is the Korean background on the subject. It is strikingly different from the luxurious literary and theatrical homosexual tradition in China, and even more so from the glamorized and pseudo-chivalrous homosexual code of late mediaeval Japan, with its manuals and novelettes. It is a matter belonging to a lower stratum of Korean life, with possible primitive religious connections, though they are now exceedingly dim ones. And against that background we must consider Ayukai’s⁹⁾ suggestion that the hwarang of Silla may have been a homosexual cult.

The chief bases for the contention are the constantly stressed prettiness of the boys, their gay clothes and cosmetics, the extreme of affection displayed in such a story as that of Sadaham¹⁾ and the foundation myth about the women, especially if the latter is taken in conjunction with the theory that the hwarang cult was inspired by an ancient shamanistic cult.
9) Op. cit. page 36.

1) See above page 32. Cf. also Siro, above page 39 and An Sang, page 41.

[page61]

Mishina has shown that none of the arguments are conclusive. Even the story of the Miruk hwarang,2) though it has romantic tones in it, is not necessarily any more homosexual than the poems of St John of the Cross. On the other hand in any organization of young men inversion is bound to appear to some degree. To assert that the hwarang never practised it would temerarious in the extreme. Yet it cannot be proved that it ever occurred at all.

The evidence that after Silla times the people who carried on the name of hwarang of ten indulged in homosexuality is another matter. I have said enough to show that among them if anywhere could such activity be found in Korea a couple or more generations ago. It does not by any means mean that the habit was handed down with the name. There were many other ways also in which the hwarangi of the Yi dynasty differed from the hwarang of Silla.