SOURCE: “Sources and Influences of Chikamatsu's Play,” and “Literary Analysis of ‘The Battles of Coxinga,’” in The Battles of Coxinga: Chikamatsu's Puppet Play, Its Background and Importance, Taylor's Foreign Press, 1951, pp. 76-99.

[In this excerpt from his study and translation of Chikamatsu's The Battles of Coxinga, Keene discusses both the sources of that play and the later texts that used Chikamatsu as a source. In his analysis of the play, Keene emphasizes the importance of the joruri (puppet play) genre to literary interpretation, arguing that Chikamatsu himself was highly aware of the limitations and capabilities of the puppet theater. This essay originally contained ideographic characters, which have been silently removed for this reprinting.]

SOURCES AND INFLUENCES OF CHIKAMATSU'S PLAY

(A) SOURCES

The slight relationship which Chikamatsu's play [The Battles of Coxinga] bears to historical fact may be inferred from the previous chapter. However, it is not a pure fantasy on the life of some person from a distant and exotic land, as was the tragedy of the little-known English dramatist, Aaron Hill, entitled The Fatal Vision: or, the Fall of Siam.¹ Hill, in his preface, described his reasons for having chosen an oriental subject.

That I might be at liberty to form the proper incidents with less confinement to my fancy, I resolved to choose no noted character of History. Our distance from, and dark ideas of, the Chinese nation, and her borders, tempted me to fix my scene in so remote a situation. The fable is fictitious; and the characters are all imaginary, even to the very names, excepting only Umcham, common to the Emperors of China, as at Rome, the Cæsars.²

The play by Chikamatsu's predecessor, Nishiki Bunryū,³ on the subject of Coxinga had no more connection with historical truth than had Hill's on Siam, but Chikamatsu preferred to elaborate on what he considered to be the facts.⁴ It is possible to ascertain some of the sources to which he had access concerning the life of the famous Ming loyalist.⁵ Although historical authenticity was not his intention, he desired to write a work that would bear some relationship to the truth.

Chikamatsu could have obtained information about Coxinga from one of several Chinese accounts written by his time,⁶ but his primary interest was in Coxinga's connections with Japan, and for this purpose a Chinese history would not be suited. Instead, the book he used for reference was the Minshin Tōki (Account of the Battles of the Ming and Ch'ing), a work written by Maezono Jinzaemon about 1665.⁷ Many of the distortions of historical fact and otherwise inexplicable embellishments of Chikamatsu's drama may be traced to incidents in Minshin Tōki. Such episodes in the play as the battle between the shrike and the clam, the game of go, and the prophecies of the spirits on the mountain were originally found in Maezono's work and were freely adapted by Chikamatsu for his dramatic purposes.

In any case, Chikamatsu's almost exclusive concern was to write a good play. For this reason, in writing a play on Coxinga's life he did not include all of the facts even as known to him from the inaccurate Minshin Tōki. Since a complete account of Coxinga from his boyhood to his death would not make for a dramatically unified work, he chose to treat only that part of Coxinga's life from the time when he decided to go to China until the battle of Nanking. Coxinga actually left for China as a boy in his seventh year,⁸ but some accounts that Chikamatsu may have read give other figures for his age. For example, it is stated in a book by Nishikawa Joken (1648-1724) that Coxinga was seventeen when he made the journey to China.⁹ However, as Chikamatsu wished Coxinga's connections with Japan to be as strong as possible, Coxinga is married to a Japanese wife. It was probably for this reason that Chikamatsu added a few years of age to Nishikawa's seventeen.

The play ends with the battle of Nanking; because of this fact, Chikamatsu changes the defeat into a great triumph for Coxinga. If Coxinga had died gallantly at Nanking the play might have been written as an inspiring tragedy, but a despondent Coxinga gazing at the ruins of his army and of his work of many years would not fit the Japanese pattern at all. The only possible way for the play to end if the earlier auspicious incidents were to be retained was with the complete triumph of Coxinga.

The beginning and end of the biography of his hero having been determined, Chikamatsu had to look for the usual accessories to a drama, a villain and love interest. Probably the most conspicuous evil-doer of the end of the Ming dynasty was Li Tzu-ch'eng, the famous bandit leader.¹⁰ Chikamatsu's attention may have been drawn to him by the Minshin Tōki in which he figures prominently. Rather than call him Ri Jisei, the normal Japanese rendering of Li Tzu-ch'eng, however, he named him Ri Tōten, possibly because the latter name was more suited to declamation.¹¹ There is very little to link Ri Tōten with the historical character on whom he is based. The fact that Li Tzu-ch'eng lost his left eye in fighting¹² may have been responsible for the scene in the first act in which Ri Tōten gouges out his left eye, as a warrant to the Tartars that he will join them in the attack on the Mings. That is the only direct point of resemblance in the play. The other wicked characters of the play are entirely fictitious; their villainy usually consists in the fact that they oppose Coxinga's forces, not in any inherent wickedness.

As for a love interest, Chikamatsu appears to have felt that Coxinga should remain faithful to his Japanese wife. Thus it was that the michiyuki, which almost always is a journey of lovers, had to be modified in this instance; faute de mieux, we have Komutsu dressed as a man accompanying the Princess Sendan. In place of the usual love interest Chikamatsu described Coxinga's relationship with three women, Komutsu, Sendan and Kinshōjo. Komutsu seems to have been entirely a creation of Chikamatsu's invention.¹³ Sendan was apparently derived from the Princess Sendara in Nishiki's play.¹⁴ Kinshōjo, the most appealing of the three women, was also original with Chikamatsu.

The other men in the play of importance, Go Sankei and Kanki, are both historical. Go Sankei is derived from Wu San-kuei (1612-1678), one of the chief figures of the early Manchu dynasty. When Li Tzu-ch'eng was pressing on Peking in 1644, the Emperor ordered Wu to return to the capital. He delayed so long in doing so that before he reached Peking he learned of its fall to Li Tzu-ch'eng. His father was held hostage by the bandit and Wu might have yielded to Li in order to save his father's life, but Li so offended him by taking his favourite concubine that Wu preferred to negotiate with the Manchus. When the latter refused to bargain with him over the spoils of the Ming empire, he surrendered to them. He was granted positions of high responsibility by the Manchus and remained a loyal soldier to them until 1673 when he revolted. He later proclaimed himself emperor of a new dynasty, but his revolt was eventually put down.¹⁵

That Chikamatsu could make of this man the model of loyalty that Go Sankei becomes in the play should serve as adequate testimonial of the strange uses to which history was sometimes put by him. However, Chikamatsu's interpretation of the man may have been influenced by such a description of him as one finds in Nishikawa Joken's Night Tales of Nagasaki.¹⁶

The father of Coxinga was called Tei Shiryū, and had the appellation of Ikkanrō. He was from Fukien. When the Ming Dynasty reached a time of troubles, he … did not submit to Tartary, but communicated with Go Sankei …

[Later, in describing Coxinga's grandson, who surrendered Formosa to the Manchus.] However, he was far from resembling his father and grandfather in his abilities. He was at a loss what to do; he had no supporters in the fifteen provinces. Go Sankei was dead and his survivors scattered. He therefore surrendered to the Manchus.¹⁷

If Chikamatsu read some work of similar content, he would have been given the impression that Wu San-kuei (like Coxinga's father) was loyal to the Ming. This was enough knowledge of the man for Chikamatsu to possess. He needed a loyal minister to counterbalance the wicked Ri Tōten, and such an account of Wu San-kuei could have supplied him with the man. Once Chikamatsu had his loyal minister, a stock character of Chinese history, he used him throughout the play whenever needed.¹⁸

Very little else in the play is even remotely related to history, although Chikamatsu may actually have regarded details derived from the Minshin Tōki as documentation for his work. For example, the lecture on the significance of ideographs delivered by Go Sankei to the Emperor in the first act comes directly from the Minshin Tōki. In the first volume of the work, under the heading of the “Emperor's Dream”, we are told of an incident said to have occurred on the ninth day of the ninth moon of 1635. The interpretations offered by a soothsayer to the emperor concerning his dream about an ideograph correspond exactly to Go Sankei's explanations in the first act.¹⁹

Similarly, the fight between the shrike and the clam in the second act is also found in the Minshin Tōki. If the scene is not very convincing in Chikamatsu's play, it is partially because the moral of the incident does not apply. The moral would seem to be that when two parties are fighting bitterly they will not have time to look out for an enemy with designs on both of them, and thus will fall easy prey to a clever outsider. But it is not Watōnai's ambition to conquer both the Manchus and the Ming adherents. He intends instead to help the latter against the former. In the Minshin Tōki the moral actually applies. The story serves to describe how the Manchus were able to take Peking without effort while the Mings and the bandit forces of Li Tzu-ch'eng were fighting one another.²⁰ Chikamatsu adopted the image for his play without considering its inappropriateness.

A better piece of adaptation was done by Chikamatsu in the case of the scene of the Mountain of the Nine Immortals. This story is originally scattered in different parts of the Minshin Tōki. In the latter work it is Kanki who visits the mountain. There he bows before a temple of the immortals and prays that he be granted a divine message telling of his future. He falls asleep and in a dream an old man comes to deliver a prophecy. Chikamatsu substituted Go Sankei for Kanki, and the first Ming emperor and Liu Po-wen for the nameless prophetic old man of the original. These two men figure in the first book of the Minshin Tōki in connection with other prophecies. Another legend which is treated in the scene on the mountain is that of Wang Chih, the man who became so absorbed in a game of go that he lost all sense of time. Go Sankei also is substituted for Wang Chih. From these scattered pieces in the Minshin Tōki and from various bits from other works he built the famous mountain scene.²¹

One other major borrowing from the Minshin Tōki was the group of stratagems suggested by Go Sankei and Kanki in the final act. The concealed hornets in bamboo tubes advocated by Go had been used in the theatre before, notably in Nishiki's play on Coxinga.²² It may have been to discredit the device his predecessor used that Chikamatsu rejected the plan. It is more likely, however, that Chikamatsu found it in the sixth volume of the Minshin Tōki. In the seventh volume of the same book is found Kanki's plan of poisoned baskets of fruit, there described as an invention of Coxinga's.²³

To summarise, it may be said that The Battles of Coxinga owes less to history than it does to the Minshin Tōki, and less to the Minshin Tōki than it does to Chikamatsu's imagination. The plot was nevertheless regarded as historical by many people of Chikamatsu's time and for years afterwards. Of this fact one soon becomes aware in examining the works written under the influence of the Battles of Coxinga.

(B) INFLUENCES OF THE PLAY

It is interesting to learn of the sources of the Battles of Coxinga, but the influences of Chikamatsu's play are of far greater significance in Japanese literary history. In addition to the various versions of the Battles of Coxinga itself that have been made to suit the demands of theatres of later times, there exist numerous works which are based entirely or in part on the play. Some of these works have been lost or are now difficult to obtain, but a number have become famous in their own right and are included in collections of Japanese literature.

A typical work derived from The Battles of Coxinga is the play by Ki no Kaion, Chikamatsu's rival dramatist. It is entitled Keisei Kokusenya.²⁴Keisei, meaning “courtesan”, is a word found at the head of the titles of many plays of the Tokugawa period, serving to show that the protagonists will be creatures of the demi-monde. Ki no Kaion's play was a demi-monde version of The Battles of Coxinga, written about a year after Chikamatsu's play in order to profit by its popularity.

Although the change of milieu necessitated many revisions in the structure of the original, there still remains a number of striking resemblances to Chikamatsu's work. For example, in the second act there is a scene in which the hero is pursued while carrying the heroine, Kaguyama, on his back. By showing the pursuers a chart of Kaguyama's pedigree, he makes them his followers, the keisei equivalent of the scene in the bamboo forest.

Again, in the third act, we have a close adaptation of the original. The hero and heroine, together with a friend, appear before the house of a wealthy man, Iwabuchi Sazen, the husband of Kaguyama's elder sister. Kaguyama, who was separated from her sister when a child, comes to seek her help in raising funds for redeeming her from the brothel in which Kaguyama works. She raps on the garden gate and asks to speak with her sister. The sister then appears at a window in the tower. Kaguyama shows her an image of Buddha with a broken left arm, a souvenir of their childhood, and calls her sister by name. The husband, Sazen, is absent, but the sister lets Kaguyama in. She is about to hand over the necessary sum of money when Sazen returns. He declares that he intends to kill the courtesan. He puts down his sword for a moment and Kaguyama snatches it up. Her sister rushes to stop her. At that moment, Kaguyama's companions, who have been waiting impatiently outside the house, burst into the room. But all are too late; Kaguyama opens her robe to disclose that she has already stabbed herself.

Ki no Kaion's play was highly regarded by certain critics, but it never enjoyed much success. Some kabuki versions of The Battles of Coxinga achieved more popularity, usually because of the actor who played the role of Coxinga. Kokusenya Takenuki Gorō (1727), for instance, was successful because the adaptor, Ichikawa Danjūrō, was the most famous actor of his day. In his version the characters were all given Japanese names (Coxinga became Soga Gorō) and the “rough business” of the original was emphasised to such an extent that the play was popularly called “Oshimodoshi” or “Pushing Back”, referring to the hero's energetic behaviour in the last act.²⁶

One unusual adaptation of Chikamatsu's play was the Nō drama Watōnai, written by Hara Kanchiku in 1756. Many jōruri plays by Chikamatsu and other dramatists were derived from the Nō, but this is probably the only case of a Nō drama having been inspired by a jōruri. Watōnai is based on the beginning of the second act of The Battles of Coxinga.²⁷

Other imitations of the play were produced through the eighteenth and well into the nineteenth centuries.²⁸ The last of the line, The Magic Lantern Picture of Coxinga (1872), is one of the most interesting. This work by the dramatist Kawatake Mokuami (1804-1893) was designed as a farewell piece for the retiring actor, Sawamura Tanosuke. Frankly intended to excite the tears of the audience, it incorporated one of the most touching scenes of Chikamatsu's play in new and even more exotic surroundings. Kokin, a geisha, is shipwrecked off Naruto in a storm, and is rescued by a foreign ship. In the course of time she becomes the mistress of one Kankisu, an Englishman, and lives with him for seven years in London. Her husband, Hikosō, journeys all the way from Japan to see her again. They speak to one another, she up in a tower and he down on the ground. Kokin declares that she is unable to return to Japan with Hikosō because of the illness of her benefactor, Kankisu. Hikosō has just been informed that his father is critically ill and cannot tarry any longer in England. The two part, accompanied in all likelihood by the sighs and tears of the audience which was moved equally by the sorrow of the spectacle of husband and wife parting forever and by the knowledge that it was seeing its favourite actor play for the last time. The play was never repeated.²⁹

Apart from such direct imitations as the ones above cited, there are numerous instances of indirect borrowing from The Battles of Coxinga. In the celebrated Chūshingura by Takeda Izumo, for example, we find a number of such passages, although the plays have no similarity of plot. Most striking is the scene in which Oboshi Yuranosuke reads a long scroll. At the same time Okaru is reading it by its reflection in her mirror, and the villain Kudayu, who is under the verandah, reads it by the light of the moon as it unrolls. The passage is derived from the third act scene in Chikamatsu's play in which Kinshōjo on the wall sees the moonlit reflection of her father in her mirror.³⁰

The influence of The Battles of Coxinga was not restricted to plays. As early as 1716 there was the novel Military Tales of Coxinga, a work in five books that are slavishly modelled on the acts of the play.³¹Keiseiya Gundan (1717) by Andō Jishō (c. 1660-1745) goes to the opposite extreme in that it preserves of the original only a few similarities in the names of the characters.³²

Probably the best work which was inspired chiefly by Chikamatsu's play was the novel Kokusenya Minchō Taiheiki (1717) by Ejima Kiseki (1667-1736).³³ The author retained the main outlines of Chikamatsu's plot, but expanded it into a novel of complexity and humour. A comparison of the extracts below with their originals in the first act of The Battles of Coxinga may best show Kiseki's technique.

Women's long tongues are the source of all evil. Misfortunes do not originate in Heaven but with women. All disturbances in the nation, all derelictions in the administration, have their sources in women: of this there are many examples both ancient and modern.

Now, he who was called the Emperor Shisōretsu, the seventeenth sovereign of the Ming, was the second son of the Emperor Kōsō. When he first ascended the throne the world was at peace, and thus is was that he devoted himself to his revels and did not stint himself in his luxuries. He gave himself to sensual pleasure and drinking parties. On spring days before the flowers had come out, rare incense was buried in the ground to scent the shoes of the passers-by. On summer nights when there was no moon over the palace, fireflies were massed together to serve as lanterns. Yet, though one day of sensual pleasure succeeded another without pause, his corrupt ministers still flattered the emperor, and did not admonish him. The emperor in his besotted state, seemed to have forgotten everything.

Now, of the many consorts who were favoured by him, the Lady Kasei and the Lady Rishi surpassed all the others of the three thousand court-ladies, being gentle and elegant in their figures and faces. They were his two hands, and he bestowed on them his favours, alike in the season of spring flowers and in that of autumn leaves; thus it was his years were spent in deep delight.

The Lady Kasei had conceived in the autumn of the previous year, and the imperial birth was to occur in the following month. Her ladies-in-waiting, and all the others of the court down to the lowly attendants, prayed that a crown prince would be born as heir to the throne. There had been a sudden change in the language used toward the Lady Kasei, and she was already addressed with reverence as “Her Majesty, the Empress-Mother”.

Now, there was an eldest son of an old minister named Ri Chin, whose name was Ri Tōten. When he had grown to manhood, his behaviour was so cruel that his father Ri Chin had turned his back on him, disinherited him, and had him driven from his doors. Ri Tōten was not permitted to remain in Nanking, and, having nowhere to turn, had gone to Tartary. For over ten years he had lived in friendship with the people of that land. However, as his father Ri Chin was growing old and weak, his mind dwelt more and more on Ri Tōten, and he longed for the son he had once cast forth. He pardoned Ri Tōten his past misdeeds and called him back. Later, he presented him with a house, and made him the successor to the family name. He also found for him an official position in the palace.

Ri Tōten was by nature of surpassing cunning. A wily person, with a talent for fabrication as his birthright, he soon won over people's affections. Frequenting both the mighty and those of low degree, he ingratiated himself with all. He was lionised by those close to the throne, and even the emperor wondered, “What would we do without him?” Thus it was that he gradually rose to eminence, even surpassing his father in the official emoluments he was granted. His power grew with each day. Those who did not meet with his favour, even if they were loyal ministers, were slandered and driven from the court. Everything was done as he desired.

Now, when he had been in Tartary he had promised that if he returned to China he would destroy the Ming dynasty, and would divide the capital city of Nanking in two, sharing rule with the Tartars. It was his plan to wait for something exceptional to occur, and then to take advantage of it to throw the country into agitation. Then he would raise a rebellion in the midst of the resulting confusion. To this end he watched with unremitting attention. When he heard that the Lady Kasei was with child, he thought that this would prove a splendid opening for the disasters he was plotting, and his heart was suffused with smiles.

He hurried with all speed to the Lady Rishi and whispered to her, “Do you not feel aggrieved? Of all the many court ladies, you and Kasei are the ones who have enjoyed the warmest favours of his Majesty. Neither one is inferior or superior to the other; the emperor considers you his two hands. The court ladies and all the others who frequent the palace treat you two consorts with equal and profound respect. But now, if the Lady Kasei is delivered safely, and an imperial prince is born, no words will suffice to describe the awe in which she, as Her Majesty, the Empress-Mother, will be held by subjects and ministers alike. Your own prestige will sink day by day, and finally no one will know if you are here or gone, and you will be relegated to some distant quarter of the palace. You will feel as do those court ladies who have been forgotten. Until your hair has turned white, yours will be the life of a widow, and you will learn to lament the length of autumn nights. If this should come to pass your state will be most pitiable”.³⁴

Ri Tōten goes on to suggest means whereby Lady Rishi can prevent this melancholy situation from arriving; his plans finally succeed in wrecking the dynasty. The amplifications of Kiseki are often very good. For example, there are several amusing episodes arising from the troubles of the Japanese Komutsu and the Chinese Sendan in understanding one another.

(One day) Komutsu's father, who claimed to be an authority on language matters, caught the family tortoiseshell cat and put it on the chopping board. As he stood there, butcher's cleaver in hand, about to cut the cat up into cutlets, Komutsu cried in astonishment, “Hey! Why are you trying to kill our dear little kitty?” And she snatched the cat away. Her father, pretending to understand Chinese, said, “Just now when the princess was chattering, she was saying, ‘Let's eat the cat fried in sesame oil’. You wouldn't know about such things. Chinese are meat-eaters, and they put anything at all into their dishes, even cats and rats.”³⁵

In the field of “history”, The Life of Coxinga (1855) by Kanagaki Robun (1829-1894) may be mentioned.³⁶ The author attempted to combine historical facts with Chikamatsu's invention. This may be seen in the names used for the characters of his history. We are told of “Watōnai, the only son of Tei Shiryū Ikkan”, following Chikamatsu's appellation, but the Ri Tōten of the play is here called Ri Jisei (Li Tzu-ch'eng), showing the author's superior knowledge of Chinese history.³⁶The Life of Coxinga cannot be taken seriously by a Westerner as an historical work, but the distinction between fact and legend is much less clear in Japanese minds, and most of the book's readers must have believed it to be factual.

The popularity of Coxinga in Japan continued to grow as a result of the continued successes of the play when revived, and of the imitations and adaptations, some of which have been described. It could not have come as a shock to Japanese, then, when in 1898, at the time of the cession of Formosa to Japan, the first Governor-General of the island, Admiral Count Kabayama, at once proceeded to Coxinga's temple in the city of Tainan, and recommended that it be known henceforth as a jinja (or Shinto shrine) instead of as a miao (or Chinese temple).³⁷ The request was granted and Coxinga took his place among the other Japanese immortals. The move was undoubtedly politically inspired, serving to stress the half-Japanese, half-Chinese parentage of the hero, but even if it is admitted that Coxinga, as a loyal defender of the Ming, was entitled to veneration by the Chinese, it is difficult to see why Japanese should have wanted to worship at his shrine. Yet already in the late eighteenth century there was a shrine to him in Kyoto,³⁸ and none expressed surprise at his deification a hundred years later. By what means was it that Coxinga, the son of a prostitute by some accounts, a man who left Japan in his seventh year and never returned, was enabled to scale the Japanese heaven? Is it not likely that Chikamatsu's Battles of Coxinga was responsible for elevating a rather dubious half-breed, the “Attila of the East” for some, to the status of a Japanese hero, and even a Japanese god?

LITERARY ANALYSIS OF THE BATTLES OF COXINGA

The first question a European is likely to ask about a play with which he is unfamiliar, is whether it is a comedy or a tragedy. This is a traditional and useful way of dividing works for the theatre, but at times it has been found that a given play does not fit easily into either category. The history plays of Shakespeare, for example, are usually not classified under either heading, but form a group of their own. Even in the case of the history plays, a prevailing tone generally leads the reader or spectator to consider the play as comedy or tragedy, in spite of contrary incidents that may occur in the work. That is to say, neither the death of Falstaff nor the massacre of the boys in the English camp, tragic though they are, disturbs the essentially comical and high-spirited tone of the play Henry V, nor, for that matter, do Faulconbridge's banterings make King John seem the less a tragedy. In the case of The Battles of Coxinga, however, we have a play in which the tone itself is constantly changing. We may decide on the basis of the happy ending that the play is a comedy, but its most memorable incidents are pathetic if not tragic. This strange ambivalence might appear a fault to Westerners, but it is precisely the quality that Japanese critics have most often praised in the play, and was certainly an important reason for its success.

It has previously been mentioned that the five acts of the jōruri play are believed to have been imitated from the five Nō plays customarily presented at one theatrical performance. It was also noted that in between the individual Nō plays there were short comic interludes called kyōgen. These interlude-plays were often parodies of the pieces they followed; a Nō play about a devoted priest might be followed by a kyōgen about a roguish one. It might seem that this alternation of mood from the tragic tone of the Nō to the broad farce of the kyōgen and back again would be too great a wrench for the sensibilities of the audience, especially when presented in roughly equal amounts. On the contrary, the kyōgen were considered a highly important part of a day's entertainment, and appear to have responded to some need felt at a certain stage of development of the theatre in every country. We find a curious parallel, for example, in the theatre of Italy during the eighteenth century.

Metastasio wrote not only the tragic play itself but wrote also two little comic interludes, according to that illogical jumbling fashion which prevails whenever an art is adolescent. After the curtain had fallen upon the intensely tragic Dido upbraiding the stately and statuesque Aeneas, it rose upon Signorina Santa Marchesini, as a prima donna, quarrelling with the stage tailors about the length of her train, and interrupted by the arrival of a famous buffo as Nibbio, a ridiculous manager from the Canary Islands … These interludes are droll enough—caricatures of the very tragedy they relieve, caricatures of the wrath and faintings of just such a queen as Dido, caricatures of just such airs, with tremendous nautical, botanical, meteorological similes … [It was the] turning of everything into ridicule fearlessly, from the certainty that as soon as the tragedy was resumed people would weep as much at the originals of the caricature as they had laughed at the caricature itself.³⁹

It was just about the time of The Battles of Coxinga that the Japanese puppet theatre began to emerge from its “adolescence”; or, rather, it would be more accurate to say that it was at this time that Chikamatsu was lifting it to a level of maturity. His domestic tragedies, for example, have a uniform tone, and there is no catering to the public craving for low comedy. But one may find in any country a certain reluctance to give up old forms, and Japan, of all countries, is the one which preserves the things of the past the longest. It was thus that, even after the new stage represented by the three-act domestic tragedies was attained, a demand continued to exist for the older style of plays, plays which still preserved the heterogeneous elements of a programme of Nō tragedies and kyōgen. In The Battles of Coxinga Chikamatsu offered an excellent example of this mixture. To appreciate the play as a Japanese can, we must be able to admit the possibility of sandwiching regularly in one theatrical piece scenes of solemn splendour, low comedy, pathos, poetic description divorced from plot, fantasy and rough action. This is not the same as the inclusion of short comic scenes in Western tragedies; the play is in a composite form, of which we should have more examples in the West had the Italian entertainment described above been fused into one drama, instead of remaining three acts of a play and two comic interludes.⁴⁰

The similarities that may be noted between the structure of The Battles of Coxinga and the order of a programme of Nō and kyōgen are not exact, but they are striking enough to afford us insight into the form of Chikamatsu's play. The five Nō plays that were offered at one performance were in a traditional order, which was a fixed one except for some accepted alternatives. The normal order was (1) the gods (2) men (3) women (4) mad people and (5) demons, but in place of (4) mad people, it was possible to have a dramatic realistic piece, and in place of (5) demons, a congratulatory piece. Much leeway was permitted in the interpretation of these categories, and certain Nō plays have been assigned at different times to one or another of them. The third, or “woman” play, was the high point of the programme, and most of the best dramas belong to this group. In between the Nō plays were presented kyōgen. Sometimes there would also be short interpolated kyōgen episodes even within the Nō play itself; these, however, were generally pertinent to the plot, unlike the interlude-plays, and have some correspondence with the “comic relief” found in Western plays.

The first act of The Battles of Coxinga has nothing to do with the gods, but it does treat of the court of the Chinese emperor, an exalted subject. The effect of this act on the audience must have been to inspire awe by the loftiness of its diction and the splendour of its décor and costumes.⁴¹ This, presumably, should have been the effect of the first of a Nō series. It is the only act of the play which has this effect, although the mood is not maintained to its conclusion, but breaks up into scenes of battles between the Chinese and the Tartars.

The second act opens with a very different setting. Instead of the Chinese emperor and his glorious court, we have two simple Japanese fisherfolk wandering along the beach. This first part of the second act seems to me to correspond in effect to a kyōgen interlude. It is not farcical, but in the abruptness and violence of the change of mood, we have a parallel to the effect of a kyōgen after a Nō.

There is another quick shift of mood in the second part of the act. The audience, which has been listening with amusement to Komutsu's shrewish comments and to Sendan's nonsensical Chinese, is told that the bizarre stranger in bedraggled clothes is the princess last seen leaving China in her little boat. A link is then established with the serious business of the first act. The next of the Nō series now begins, the “man” play. The “man” in question is, of course, Watōnai, the future Coxinga, and his mighty deeds are the subject of the act. Watōnai realises that he is a man of destiny, and that he is to be the agent for the restoration of the Ming. He crosses over to China with his mother, overcomes a tiger with divine help, and then, with the aid of the now friendly tiger, defeats a hostile Chinese force, recruiting its members as his own followers. The second act ends in farce, with Watōnai and his mother shaving the heads of the Chinese soldiers in the Japanese manner. The Chinese soldiers, bereft of their locks, catch cold and sneeze, while their “noses ran and eyes watered like the showers of summer”. The followers are required to assume comical names denoting their places of origin, and the whole happy troupe rides off.

The third act, like the third of the Nō play series, is easily the most important, and contains the chief elements that were retained in later performances of the plays as well as in imitations and adaptations. The third is the “woman” play, and the woman is Kinshōjo, Ikkan's daughter. She figures in all of the famous scenes of the act—the recognition between father and daughter, Kanki's various changes of decision, the “flowing rouge” and her suicide. The third is the one act without any sharp changes of mood. There is comic relief in the conversation of Kinshōjo's maids, but it is slight and not sustained. There is also a little “rough business” when Watōnai breaks into the castle, but it is over in a sentence, and is not the swinging, crashing, slaughtering of a real “rough business” scene.

The third act develops to its lyrical conclusion with no break, but the fourth act is divided into two parts. The first of these is composed of two rather light-hearted episodes, Komutsu's fencing practice and the michiyuki of Princess Sendan. The second part of the act is in an entirely different mood, describing the adventures that befell Go Sankei on the Mountain of the Nine Immortals and afterwards. He is shown in a vision the various triumphs won by Coxinga. As he watches time passes rapidly, and five years have already elapsed by the time the vision comes to an end. He then meets his old friend Ikkan, who is with a party of the Ming adherents. The Tartars spot them, and they are able to escape only by means of a miraculous cloud bridge supplied by the god Sumiyoshi. This bridge is the source of the undoing of the Tartar troops and the act ends with a Ming victory.

The fourth act of the Nō series was usually the “mad person” play. One Japanese critic felt that the effect of the michiyuki and of the scene on the Mountain of the Nine Immortals was that of fantasy and illogicality, which represented an equivalent of the madness of the Nō series.⁴² It may also be that the scenes of battle are realistic enough for this fourth act to be likened to the dramatic realistic piece that was a permitted substitute for the “mad person” play, but in either case the parallel is not as close as elsewhere. The first part of the act may be considered a kyōgen-like episode, as may be the high-spirited closing moments.

As we have seen, a Nō programme was generally concluded with a “devil” piece, but a congratulatory play could take its place. The last act of The Battles of Coxinga is of the latter sort, ending as it does with the complete triumph of Coxinga's forces, and with prayers for the continued happiness of China under Japanese auspices.

When we examine the structure of the individual acts of the play, we are struck by their self-sufficiency. That is to say, each act virtually stands on its own; this would be explained, of course, by the fact that the play's structure was influenced by the tradition of the performance of a series of self-contained Nō plays. An act usually ends in such a manner as to give us an effect of completion, and not of leading us on to the following act.⁴³ The first act ends with an encomium of Lady Ryūka, the second with a description of Watōnai's triumphal parade of his new followers, the third with praise of Watōnai (now Coxinga), the fourth with the victorious entry into Foochow, and the last with the final successes of Coxinga and the praise of Japan. Only at the end of the fourth act is there any suggestion of action to come. The following is a typical act ending of The Battles of Coxinga.

That he was the most wonderful of men from Japan, where such heroes are born, where the land is as land should be, and the sovereign is a true sovereign, he showed by illuminating a foreign country with the brilliance of his martial talents.

(Act III)

Contrast with this a typical act ending from Shakespeare.

                              Two things are to be done,—
My wife must move for Cassio to her mistress;
I'll set her on;
And bring him jump when he may Cassio find
Soliciting his wife:—ay, that's the way;
Dull not device by coldness and delay.

(Othello, Act II, Scene 3.)

In Othello we are led on to the next act by Iago's revelation of his plans. He tells us that the play is not ended, and even specifies what action is to come. In Chikamatsu's play the incompleteness of the action is implicit and not explicit; we expect that the hero will restore the Ming dynasty, but Chikamatsu does not promise us this, much less give us any detailed plans. If the play had ended after the third-act lines quoted above, one would not have been quite satisfied perhaps, but the eventual success of the hero Coxinga could be inferred. Certainly had the story of Coxinga been as well known to a Japanese audience as that of the Soga brothers, for example, Chikamatsu would not have hesitated to end his play at that point, knowing that the audience would supply the eventual resolution. In the case of the comparatively unknown Coxinga, Chikamatsu had to present a more complete picture. Although the audience could have been quite certain that the plea of Coxinga's mother for vengeance on the Tartars would not go unanswered, it wanted to know how the triumph was effected. When, in later years, the story of Coxinga became thoroughly familiar to Japanese playgoers, it became possible to stage only one act of The Battles of Coxinga without fear that the audience would be unable to understand what was going on.⁴⁴

The acts of the play also show an individual development corresponding to the similar development in a Nō drama. Japanese critics, using the terminology first applied to the Nō, distinguish three phases of action in each of the acts of a jōruri. These are called jo, ha and kyū⁴⁵ in Japanese, corresponding roughly in speed, if not in mood, to an andante, allegro and presto. Where the line of demarcation between the sections is to be drawn rests largely with the particular critic, but a shift from one tempo to another is readily observable in each act. A Western play, on the other hand, is usually treated as a unity, and the acts may or may not embody any particular mood. In the case of the fourth act of Macbeth, for example, we have an interlude of the relative calm of England, to contrast with the terrible doings in Scotland that have preceded and are to culminate in the following act. More often, however, there will be no divisions corresponding to the formal structure of the play; the graph of the action will show a continued rise to the climax and then a falling off to the denouement. Seldom will each act show a rise and fall of its own. In many cases, indeed, the end of an act will be purely formal; modern revivals of Shakespeare which present an entire play in two acts do not suffer on that account.

The acts of The Battles of Coxinga are joined only in a rather loose federal union, but the union exists. A graph of the action of the entire play might also be made, and divisions of jo, ha and kyū sections established. This second graph is more difficult to draw than one for an individual act, partly because the fourth act, which was the high point of the original jōruri performances,⁴⁶ is probably less interesting for a reader than the third act, thus throwing off balance for him the development Chikamatsu may have intended.

At this point we touch on a most difficult and delicate matter: to what degree may a jōruri be judged on the basis of a reading knowledge of it, divorced as it then is from the accompanying music and spectacle? Chikamatsu was considered to have been the first Japanese dramatist whose writings were of sufficient literary quality to make them worth reading,⁴⁷ but the neglect from which he suffered until the search was begun for a Japanese Shakespeare⁴⁸ was probably due to the feeling that his works could be appreciated only in performance.

The text of a jōruri has been compared to that of a motion picture script.⁴⁹ Very seldom is it possible for one to read a script with pleasure because it is usually not much more than the outline of the unspoken action. It is in the nature of the motion picture medium to exploit the possibilities of chases across the house-tops, charges of soldiers, and the excitement of inarticulate crowds. It is difficult to realise such scenes on the stage, and they are likely to embarrass us when attempted; words alone must bear the action. A jōruri is somewhere in between the two. The narrator describes the action as it occurs, but however skilful the words of the dramatist, hearing or reading the description will not be the equal of seeing the action. For example, there is the sentence recited by the narrator, “When suddenly they lift together their thousands of paper lanterns on poles, it is just like seeing the thousand suns and moons of a thousands worlds at the same instant, and the soldiers of the castle, dumbfounded, stand in utter bewilderment”. This is Chikamatsu at his best, but the words still do not have the startling power that the sudden brilliant flash of light must have had on the audiences. The wild battles, which are the chief feature of the kyū sections of four of the acts, are relatively uninteresting to read, but must have been fascinating to view. Chikamatsu was, of course, counting on such effects.

Of the two elements, music and spectacle, the reader probably loses most by the absence of the latter. Japanese music, while often capable of expressing great pathos, is not able to afford a very wide range of emotions. The music of The Battles of Coxinga was not composed afresh for the play, but had been used years before by Takemoto Gidayū, who composed it.⁵⁰ The relationship between Chikamatsu's text and the music was one of form, rather than of emotion, as would be true of a Western opera. That is to say, if the music called for a series of low notes, Chikamasu's text could not demand at that point sounds that were necessarily, or at least more comfortably, sung in a high voice. Nor, of course, could he write more sounds than the music would accommodate. On the other hand, there was no correspondence between the words and the music that accompanied them; the same music might be used in different plays for dissimilar emotions. There was also nothing in the music to distinguish any of the characters. It is even impossible to tell from the music whether a man or a woman is supposed to be singing.⁵¹ The music was necessary to a jōruri if only because it helped to avoid monotony in the recitation of the narrator. It occupies a place of about the importance of music in a motion picture, rather than in an opera or a stage play. In an opera the music can make us accept a ridiculous plot told in clumsy words. In a stage play, on the other hand, no music, however beautiful, must ever be permitted to interfere with the intelligibility of a single word. In the motion picture or the jōruri it can heighten the effect of the words and sometimes even take their place, but the music is always subservient to the spectacle.⁵²

A Westerner who has never seen a jōruri performance, but only photographs of the expressionless-faced operators in their elaborate dress standing beside the puppets, which are about two-thirds as tall as the operators, may doubt that any convincing spectacle is likely to be seen in the jōruri theatre. Yet the jōruri is still capable of exerting a powerful appeal, as the following statement by the French poet Paul Claudel would indicate.

L'acteur vivant, quel que soit son talent, nous gêne toujours en mêlant au drame fictif qu'il incorpore un élément intrus, quelque chose d'actuel et de quotidien, il reste toujours un déguisé. La marionette au contraire n'a de vie et de mouvement que celui qu'elle tire de l'action. Elle s'anime sous le récit, c'est comme une ombre qu'on ressucite en lui racontant tout ce qu'elle a fait et qui peu à peu de souvenir devient présence. Ce n'est pas un acteur qui parle, c'est une parole qui agit. Le personnage de bois incarne la prosopopée. Il nage sur la frontière indécise entre le fait et le récit. L'assistance en lui voit tout ce que le vociférateur à son pupitre raconte, soutenu par le shamisen, cet instrument qui donne la vibration des nerfs pincés, et par ce camarade à son coté qui par ses cris inarticulés et ses grognements traduit non seulement l'émotion de la scène, mais le désir d'exister, l'effort pour revivre de l'être imaginaire. La marionette est comme un fantôme. Elle ne pose pas les pieds à terre. On ne la touche pas et elle ne sait pas toucher. Toute sa vie, tout son mouvement lui vient du cœur—et de ce conciliabule mystérieux derrière elle d'animateurs masqués ou non, de cette fatalité collective dont elle est l'expression. La réalité a été si habilement divisée que l'histoire se passe entièrement dans l'imagination et le rêve, sans le support d'aucune matérialité désobligeante. Par d'autres moyens, le Jōruri arrive au même résultat que le No.⁵³

Certain of Claudel's remarks show that he was familiar with the preface to Naniwa Miyage by Hozumi Ikan, written in 1738.⁵⁴ In the second part of the preface Hozumi relates a conversation he once had with Chikamatsu.

This is what Chikamatsu told me when I visited him many years ago. “Jōruri differs from other forms of fiction in that, since it is primarily concerned with puppets, the words should all be living things in which action is the most important feature. Because jōruri is performed in theatres that operate in close competition with those of the kabuki, which is the art of living actors, the author must impart to lifeless wooden puppets a variety of emotions, and attempt in this way to capture the interest of the audience. It is thus generally very difficult to write a work of great distinction.

“Once, when I was young and reading a story about the court, I came across a passage which told how, on the occasion of a festival, the snow had fallen heavily and piled up. An order was then given to a guard to clear away the snow from an orange tree. When this happened, the pine tree next to it, apparently resentful that its boughs were bent with snow, recoiled its branches. This was a stroke of the pen which gave life to the inanimate tree. It did so because the spectacle of the pine tree, resentful that the snow has been cleared from the orange tree, recoiling its branches itself and shaking off the snow which bends it down, is one which creates the feeling of a living, moving thing. Is that not so?

“From this model I learned how to put life into my jōruri. Thus, even descriptive passages like the michiyuki, to say nothing of the narrative phrases and dialogue, must be charged with feeling or they will be greeted with scant applause. This is the same thing as is called evocative power in poets. For example, if a poet should fail to bring emotion to his praise of even the superb scenery of Matsushima or Miyajima in his poem, it would be like looking at the carelessly drawn picture of a beautiful woman. For this reason, it should be borne in mind that feeling is the basis of writing.

“When a composition is filled with particles, its literary quality is somehow lowered. Authors of no merit invariably try to cast their writings exactly in the form of waka or haikai,⁵⁵ stringing together alternating lines of five and seven syllables. This naturally results in the use of many unnecessary particles. For example, when one should say, ‘Toshi mo yukanu musume wo’, they say such things as, ‘Toshiha mo yukanu musume wo ba’. This comes from concerning oneself with the syllable count, and naturally causes the language to sound vulgar. Thus, while verse is generally written by arranging long and short lines in order, the jōruri is basically a musical form, and the length of the lines recited is therefore determined by the melody. Thus, if an author adheres implicitly to the rules of metrics, his lines may prove to be awkward to recite. For this reason I am not concerned with metrics in my writings and I use few particles.

“The old jōruri was just like our modern street story-telling⁵⁶ and was without either flower or fruit. From the time that I began to write jōruri, first for Kaga-no-jō and then for Chikugo-no-jō,⁵⁷ I have used care in my writing, which was not true of the old jōruri. As a result, the medium was raised one level. For example, inasmuch as the nobility, the samurai and the lower classes all have different social stations, it is essential that they be distinguished in their representation from their appearance down to their speech. Similarly, even within the same samurai class, there are both daimyō and retainers, as well as others of lower rank, each rank possessed of distinct qualities; such differences must be established. This is because it is essential that they be well pictured in the emotions of the reader.

“In writing jōruri, one attempts first to describe facts as they really are, but in so doing one writes things which are not true, in the interest of art. To be precise, many things are said by the female characters which real women could not utter. Such things fall under the heading of art; it is because they say what could not come from a real woman's lips that their true emotions are disclosed. If in such cases the author were to model his character on the ways of a real woman and conceal her feelings, such realism, far from being admired, would permit no pleasure in the work. Thus, if one examines a play without paying attention to the question of art, one will probably criticise it on the grounds that it contains many unpleasant words which are not suitable for women. But such things should be considered as art. In addition, there are numerous instances in the portrayal of a villain as excessively cowardly, or of a clown as being funny, which are outside the truth and which must be regarded as art. The spectator must bear this consideration in mind.

“There are some who, thinking that pathos is essential to a jōruri, make frequent use of such expressions as ‘it was touching’ in their writing, or who when chanting do so in voices thick with tears, in the manner of the Bunya-bushi.⁵⁸ This is foreign to my style. I take pathos to be entirely a matter of restraint.⁵⁹ Since it is moving when all parts of the art⁶⁰ are controlled by restraint, the stronger and firmer the melody and words are, the sadder will be the impression created. For this reason, when one says of something which is sad that it is sad, one loses the implications, and in the end, even the impression of sadness is slight. It is essential that one not say of a thing that ‘it is sad’, but that it be sad of itself. For example, when one praises a place renowned for its scenery such as Matsushima by saying, ‘Ah, what a fine view!’ one has said in one phrase all that one can about the sight, but without effect. If one wishes to praise the view, and one says numerous things indirectly about its appearance, the quality of the view may be known of itself, without one's having to say, ‘It is a fine view’. This is true of everything of its kind.

“Someone said, ‘People nowadays will not accept plays unless they are realistic and well reasoned out. There are many things in the old stories which people will not now tolerate. It is thus that such people as kabuki actors are considered skilful to the degree that their acting resembles reality. The first consideration is to have the chief retainer in the play resemble a real chief retainer, and to have the daimyō look like a real daimyō. People will not stand for the childish nonsense they did in the past.’ Chikamatsu answered, ‘Your view seems like a plausible one, but it is a theory which does not take into account the real methods of art. Art is something which lies in the slender margin between the real and the unreal. Of course it seems desirable, in view of the current taste for realism, to have the chief retainer in the play copy the gestures and speech of a real retainer, but in that case should a real chief retainer of a daimyō put rouge and powder on his face like an actor? Or, would it prove entertaining if an actor, on the grounds that real chief retainers do not make up their faces, were to appear on the stage and perform, with his beard growing wild and his head shaven? This is what I mean by the slender margin between the real and the unreal. It is unreal, and yet it is not unreal; it is real, and yet it is not real. Entertainment lies between the two.

“In this connection, there is the story of a certain court lady who had a lover. The two loved each other very passionately, but the lady lived far deep in the women's palace, and the man could not visit her quarters. She could see him therefore only very rarely, from between the cracks of her screen of state at the court. She longed for him so desperately that she had a wooden image carved of the man. Its appearance was not like that of any ordinary doll, but did not differ in any particle from the man. It goes without saying that the colour of his complexion was perfectly rendered; even the pores of his skin were delineated. The openings in his ears and nostrils were fashioned, and there was no discrepancy even in the number of teeth in the mouth. Since it was made with the man posing beside it, the only difference between the man and this doll was the presence in one, and the absence in the other, of a soul. However, when the lady drew the doll close to her and looked at it, the exactness of the reproduction of the living man chilled her, and she felt unpleasant and rather frightened. Court lady that she was, her love was also chilled, and as she found it distressing to have the doll by her side, she soon threw it away.

“In view of this we can see that if one makes an exact copy of a living being, even if it happend to be Yang Kuei-fei,⁶¹ one will become disgusted with it. Thus, if when one paints an image or carves it of wood there are, in the name of artistic licence, some stylised parts in a work otherwise resembling the real form; this is, after all, what people love in art. The same is true of literary composition. While bearing resemblance to the original, it should have stylisation; this makes it art, and is what delights men's minds. Theatrical dialogue written with this in mind is apt to be worthwhile.”⁶²

These remarks by Chikamatsu are of paramount importance in understanding all of his writings in general, and The Battles of Coxinga in particular. We must first note Chikamatsu's insistence on the special demands of the puppet jōruri. Many of the features which may puzzle us as we read The Battles of Coxinga become clear when we realise that Chikamatsu was consciously attempting to “impart to lifeless wooden puppets a variety of emotions”. The effects an actor can achieve by his facial expression, the modulations of his voice, and the general manner in which he moves about, are almost impossible to duplicate in the puppet theatre. The puppet, as Claudel stated, is the incarnation of the dramatist's words, incapable of expressing more than has been written. To aid the puppets there is the narrator, who can inform us, in case we were not sure, that the speech we have just heard was delivered in anger or in jest, but by and large the life of the play must come from within the words.

A certain amount of exaggeration and emphasis is required if the puppet's actions are to acquire any of the force of an actor's. But beyond the exaggerations required to give the semblance of life to the puppets, there are those which are permissible solely because the work is conceived for the puppet stage and not for actors. However skilful an author may be in imparting life to the puppets, however cleverly they are manipulated by the operators, the audience can never forget that they are not human beings. Because the structure of a jōruri play is thus built on a basis of unreality, the audience will accept exaggerations and fantasy such as would be intolerable in an ordinary stage presentation. In a somewhat similar fashion, the unreality of a Wagnerian opera will permit dragons and giants to people its stage. The audience will accept not only such curiosities, but even the spectacle of a stout middle-aged woman playing the part of a radiant young goddess, because of the initial suspension of disbelief in a world where people sing instead of talk.

Many of the scenes in The Battles of Coxinga which appear fantastic and foolish to the reader relied for their effect on a similar suspension of disbelief on the part of the jōruri audience. For example, the battle between Watōnai and the tiger in the Bamboo Forest of Senri would be ludicrous on the stage, for the spectator would be conscious of the two men inside the tiger skin, and could not take seriously Watōnai's wrestling with so unwieldy a creature. Only a very generous spectator would suspend disbelief at that point. A reader of the play is in a similar position. In the puppet theatre, on the other hand, the tiger is no less realistic than Watōnai, and there is no reason why a spectator who accepts the initial unreality of a puppet performing as a man should be unable to accept a puppet tiger. The jōruri dramatist is thus at liberty to write scenes of great extravagance of language and action whenever they suit his purpose.

The purpose of Chikamatsu in writing the story of the battle between Watōnai and the tiger is made apparent by his remarks on the importance of having sadness in a play felt of itself without recourse to expressions like “it is sad”. In a similar way, when Chikamatsu wanted to show the audience how brave his young hero was, he chose to give an example which would show Watōnai's courage better than any statement about it the author could make, Instead of saying, “Watōnai was not afraid to pit his strength against a tiger”, Chikamatsu gave the audience a few minutes of exciting battle which would convince it most effectively of the fact. Later, to show the power of the Japanese gods he has Watōnai's mother subdue the tiger with the aid of a charm from the Great Shrine of Ise, which was the dramatic way of saying, “Even though Watōnai was the most powerful of men he could still not overcome the tiger, but this was no difficult matter for the gods”.

If the effect on the reader of the battle in the Bamboo Forest of Senri or of the miraculous cloud bridge in the fourth act is not entirely convincing, it is further proof of the importance of seeing the play performed by puppets, for the requirements of the puppet theatre were Chikamatsu's first consideration. Other scenes, such as the one in which Kinshōjo on the wall recognises her father down below, may exert a more powerful appeal on the reader, but even in them the element of the unreal is never absent. It was impossible, for example, that the image in Kinshōjo's mirror would have been clear enough for her to make out her father's features, yet we are willing to allow this unreality in the interests of the story.⁶³

In his attempt to make the puppets live and to make their qualities apparent of themselves, Chikamatsu was forced to sacrifice any subtlety of portrayal that would have been possible with actors. Chikamatsu declared, as we have seen, that it was important that the different social stations be distinguished in performance by their appearance and their speech. He did not add that it was important that different members of the same social class be differentiated. This may have been because the concept of the individual was still not sufficiently developed in Japan for individual traits to be studied, or it may have been because the puppets were incapable of detailed character portrayal. The personages of The Battles of Coxinga are types, and even extreme types. Watōnai is brave beyond all human bravery, Ri Tōten is the complete villain, and all the other characters are almost as quickly recognisable by their types as by their names. Chikamatsu is careful to distinguish between the countrified Komutsu and the lady-like Kinshōjo in their speech and attitudes, but he did not attempt to make either one of them come alive for us as a person. He might have regarded the reduction of the characters to their outstanding features as “stylisation” which he favoured. There is one important element yet to be mentioned in connection with the structure of The Battles of Coxinga, the element of surprise. Surprise is almost the governing factor of the play, whether it is surprise on the lowest level, where it takes the form of puns and plays on words, or surprise of a major nature over developments in the plot. The great danger of a play in which the recitation is done by one narrator at a time is that it will become monotonous, and to avoid this danger Chikamatsu filled his jōruri with surprises.

The action of The Battles of Coxinga may be traced in terms of its surprises, which are so frequent as to be the rule. Each act is composed of passages of varying length linked or separated by surprises, and the individual acts form a pattern of surprises and contrasts. There are also little surprises in the sentiments themselves, for those familiar with traditional Japanese usage. For example, when Go Sankei tells his infant son whom he is about to kill not to regret leaving his parents “in this world of illusion” we are surprised, because we would expect at this point some expression by Go Sankei of his sorrow, and the surprise makes the scene more touching.⁶⁴

Any attempt to make a full literary analysis of the play, however, must be hindered by the non-literary elements. A jōruri text has been compared to a motion picture script; it might also be likened to the piano score of an opera. All the themes of the play and all the turns of phrases are apparent from the text, but we miss the accompaniment—the actions of the puppets, the music and the decor, as a piano version of the Miserere would communicate only a fraction of its power. A person endowed with an especially vigorous imagination might be able to restore to the text, even in translation, some of the elements that have been lost. It is the hope that, even for those not so gifted, at least the minimum fraction of the excitement and colour that made The Battles of Coxinga the most famous play of its day will make itself felt across the many obstacles.

Notes

1. The play is critically described in Brewster: Aaron Hill, p. 95. It was presented in February, 1716, or about two months after The Battles of Coxinga.

2. Dramatic Works of Aaron Hill, Volume I, p. 151.

3. See above, Chapter I, p. 6.

4. Cf. Takahashi: Chikamatsu no Gikyoku Taiyō, p. 92.

5. It should be noted that the possibility exists that some book may yet be found that will prove to be a much closer source of Chikamatsu's play than any yet uncovered. The comparative lack of interest by Japanese scholars in Tokugawa literature has meant that much valuable source material of the period still lies neglected in the great Japanese libraries.

6. For a list of Chinese works on Coxinga written before Chikamatsu's play, see Hsieh Kuo-chen: Wan-ming Shih-chi K'ao, chuan 13, pp. 1-6.

7. The importance of the Minshin Tōki is described in Noma: Kokusenya Gozen Gundan to Kokusenya Kassen, etc., p. 625 ff.

8. Chinese sources all agree on this point. Cf. Huang: Ssu-hsing Shih-mo, p. 1a.

9. Nishikawa: Ka'i Tsūshō-kō (written 1695) p. 300. Arai: Shinsho, p. 685, states that Coxinga was in his eighth year rather than his seventh.

10. A biography of Li Tzu-ch'eng may be found in Hummel et al.: Eminent Chinese of the Ch'ing Period, Volume I, p. 491.

11. See Chapter VI, note 14.

12. Ross: The Manchus or the Reigning Dynasty of China, p. 151.

13. Coxinga was actually married to a Chinese woman of the Tung family. See Chapter III, p. 46.

14. See Chapter I, p. 6.

15. Cf. Hummel et al.: op. cit., Volume II, pp. 877-8, for life of Wu San-kuei.

16. This work was actually written four years after The Battles of Coxinga, but appears to reflect an independent tradition about the loyalty of Wu San-kuei. The author seems to have gathered his information from Chinese living in Nagasaki. Chikamatsu may have read an earlier work of similar content.

17. Nishikawa: Nagasaki Yawa-sō, pp. 258-62.

18. It is interesting that in Vondel's drama Zungchin (1667) Wu San-kuei (Uzangueius) also figures as a model of loyalty.

19. Noma: op. cit., p. 629.

20. Ibid., p. 635. The original story of the shrike and the clam is in Chan Kuo Ts'e (Szu Pu Pei Yao edition), Chapter 30, p. 10a.

21. Noma: op. cit., pp. 635-7.

22. Yamaguchi: Kokusenya Kassen no Beni-nagashi ni tsuite, pp. 237-8.

23. Noma: op. cit., p. 638.

24. Shuzui: Article in Nihon Bungaku Daijiten, Volume I, pp. 917-8, describes the play.

25. Ibid., loc. cit.

26. Tsubouchi et al: Chikamatsu no Kenkyū, p. 304. Kokusenya Takenuki Gorō is written; Oshimodoshi is.

27. Kitani: Introduction to Kokusenya Kassen in Dai-Chikamatsu Zenshū, Volume III, p. 106.

28. Other plays on the theme include Hiragana Imagawa-ō (1732) (described in Tsubouchi: op. cit., p. 305); Tōjin-mage Ima Kokusenya (1825) (described ibid., p. 343); Wakoku-bashi (1863) (in Kawatake: Kawatake Mokuami, p. 152); and Kokusenya Rihatsu no Sugatami (1887) (in Iizuka: Kabuki Saiken, p. 242).

29. Ihara: Meiji Engeki-shi, p. 189-90. The play is called in the original Kokusenya Sugata no Utsushie.

30. Takeda: Chūshingura, tr. Inouye, pp. 156-7.

31. Koizumi: Article in Nihon Bungaku Daijiten, Volume II, p. 103, describes the work, Kokusenya Gozen Gundan.

32. Keiseiya Gundan may be found in Kokusho Kankōkai: Kinsei Bungei Sōsho, volume 5. The title is untranslatable, being a combined play on the names of Ki no Kaion's work and of the novel Kokusenya Gozen Gundan.

33. In Teikoku Bunko, Kiseki Jishō Kessaku-shū, Volume I. The title is written.

34. Ibid., pp. 173-78.

35. Ibid., p. 202.

36. Kanagaki: Kokusenya Ichidai-ki, p. 3b.

37. Ponsonby-Fane: Koxinga, p. 72. It was probably not coincidence that in the same year, 1898, the tale Watōnai was added to Iwaya Sazanami's great collection of children's stories.

38. Cf. Akizato; Shūi Miyako Meisho Zue (written 1787), p. 105.

39. Lee: Studies of the Eighteenth Century in Italy, pp. 171-2.

    Burckhardt (in Civilization of the Renaissance, p. 192) related how ballets and pantomimes served as interludes when classical plays were presented in the Renaissance courts. “… That while the play was going on everybody was longing for the interludes is quite intelligible, when we think of the picturesque brilliancy with which they were put on the stage. There were to be seen combats of Roman warriors, who brandished their weapons to the sound of music, torch dances executed by Moors, a dance of savages with horns of plenty, out of which streamed waves of fire—all as the ballet of a pantomime in which a maiden was delivered from a dragon.” Wakatsuki (in (Ningyō) Jōruri-shi Kenkyū, p. 591) described the less elaborate Japanese counterparts of the Italian interlude entertainment.

    It is interesting to note what J. Middleton Murry (Shakespeare, pp. 189-90) wrote about The Merchant of Venice. “We may say that it is, as nearly as possible, a pure melodrama or tragi-comedy, an almost perfect example of the art-form which being prior to art itself, most evidently and completely satisfies the primitive men in us all … The Merchant of Venice is the type of entertainment the theatre should supply—villain discomfited, virtue rescued, happy marriages, clowning, thrills, and a modest satisfaction of the general appetite for naughtiness.” This description might almost as well apply to The Battles of Coxinga. The difference in the nature of the two plays is that Chikamatsu's is at times deeply pathetic, at times even horrifying, and at times low comedy, while, as Murry points out, The Merchant of Venice is essentially a fairy-tale throughout, and has a uniform tone which The Battles of Coxinga lacks.

    I have chosen not to call The Battles of Coxinga a melodrama because the word generally has a pejorative sense, although if used as in the quotation above from Murry's work it comes fairly close to describing Chikamatsu's play.

40. Wakatsuki: JK, p. 618, quotes an account which reads, “Up to and including the work Jitō Tennō Uta Gumpō [presented three months before The Battles of Coxinga] there were puppet comic pieces in between the acts because the jōruri plays were short. This practice stopped with The Battles of Coxinga.” Another reason why the practice may have stopped was that it may have been felt that the comic interludes were already incorporated in the text itself.

41. Cf. Sonoda: Jōruri Sakusha no Kenkyū, p. 101.

42. Ibid., p. 102.

43. It was the tradition of the old jōruri to begin acts with certain set expressions and to end them with the praise of some great figure, such as the first Tokugawa shōgun, Ieyasu (cf. Wakatsuki: JK, pp. 150, 217). This fact in itself gave a degree of independence to the form of each act.

44. An example of a truncated version of the play may be found in Atsumi: Zoku-zoku Gidayū Kyōgen-shū, pp. 494-522. This is an acting version dating from about 1835. After a short introductory scene on the beach of Hirado, where Watōnai witnesses the battle between the shrike and the clam, an abbreviated version of the third act is given. The play ended with the death of Kinshōjo and of Watōnai's mother.

45. See glossary.

46. Cf. Tsubouchi et al.: Chikamatsu no Kenkyū, p. 299.

47. Hozumi in Naniwa Miyage, p. 3, made this statement.

48. In this connection, the writings of Takayama Chogyū might in particular be cited. In Chikamatsu Sōrinshi, p. 11, he wrote, “A time has now at last come for our Chikamatsu, who has been likened to England's Shakespeare and Germany's Goethe, to be appreciated”. Although Chikamatsu's plays were popular on the stage and were widely read, they were not considered worthy of consideration by serious literary critics until it was discovered that Chikamatsu's Western “equivalent”, Shakespeare, was so highly regarded.

49. Wakatsuki: JK, p. 735.

50. Tōgi Tetteki: Ongaku-jō yori mitaru Kokusenya Kassen, p. 318. Extracts from the music are reprinted in this article.

51. Ibid., p. 319.

52. An excellent example of how a musical accompaniment consisting of a stringed instrument playing melodies which have no direct relation to the scenes they accompany may serve to heighten the emotional intensity of a film is found in The Third Man. Presumably the effect of the samisen at a jōruri performance would have been much the same.

53. In Miyajima: Contribution à l'Etude du Théâtre des Poupées, pp. xii-xiv.

54. Particularly those about “la frontière indécise”. The attribution of the Naniwa Miyage to Hozumi Ikan is almost universal, but the Kokusho Kaidai, p. 1524, credits the work to Konoshita Kisei.

55. A waka has come in modern times to mean the same as a tanka. The latter is a poem written in 31 syllables arranged in lines of 5, 7, 5, 7 and 7 syllables. Haikai are poems of 17 syllables arranged in lines of 5, 7 and 5 syllables.

56. Saimon, here probably short for saimon-yomi, the name given to persons who made a living by chanting ballads about the miracles of the gods. These were chanted to the tunes of popular songs. By the time of Chikamatsu, however, the only thing to distinguish the saimon-yomi from others who sang or told stories in the streets to gain a few coppers was the style of chanting.

57. For additional information about Kaga-no-jō and Chikugo-no-jō (Takemoto Gidayū) see Chapter II, p. 29.

58. Bunyabushi was the name given to the style of jōruri practised by Okamoto Bunya. It was noted for its sentimentality.

59. The word I have here translated as “restraint” is giri. The word normally means “propriety” or “duty”, but in the context, “restraint” seems to be the meaning indicated, although it is not a recognised meaning of the word. The meanings of “propriety” and “restraint” are actually not so far removed, however. If one acts in accordance with the principles of propriety one will not gush over into uncontrolled emotion but will be restrained. I think that is what Chikamatsu meant.

60. The word Chikamatsu used, rikugi, refers to the six types of poetry recognised in the prefaces to the Kokinshū (cf. Ueda: Chikamatsu Go-i, p. 371). The word is here used to indicate all parts of the art of jōruri including music, words, etc.

61. Yang Kuei-fei (d. 756) was the most famous of Chinese beauties.

62. Hozumi: Naniwa Miyage, pp. 5-11.

63. This device was imitated in Chūshingura, see Chapter IV, p. 82.

64. Cf. Aeba Kōson (in Tsubouchi et al.: Chikamatsu no Kenkyū), p. 292.

Bibliography

Japanese works

Akizato Ritō: Shūi Miyako Meisho Zue (in Nihon Zue Zenshū), Tokyo, 1928.

Arai Hakuseki: Shinsho (in Vol. VI, 3rd series; Nihon Zuihitsu Taisei), Tokyo, 1930.

Atsumi Seitarō: Zoku-zoku Gidayū Kyōgen-shū (Vol. XXXVII of Nihon Gikyoku Zenshū), Tokyo, 1932.

Hozumi Ikan: Naniwa Miyage (ed. Ueda Kazutoshi, Tokyo, 1904.

Ihara Toshirō: Meiji Engeki-shi, Tokyo, 1933.

Iizuka Tomoichirō: Kabuki Saiken, Tokyo, 1926.

Kawatake Shigetoshi: Kawatake Mokuami, Tokyo, 1925.

Kitani Hōgin: Introductions in Dai-Chikamatsu Zenshū, Tokyo, 1922.

Koizumi Tōzō: Articles in Nihon Bungaku Daijiten, Tokyo, 1933.

Nishikawa Joken: Ka'i Tsūshō-kō (in Takimoto: Nihon Keizai Daiten), Tokyo, 1928.

Nishikawa Joken: Nagasaki Yawa-sō (in Iwanami Bunko), Tokyo, 1943.

Noma Kōshin: Kokusenya Gozen Gundan to Kokusenya Kassen no Genkyō ni tsuite (in Kinen Rombunshū) Kyoto, 1934.

Shuzui Kenji: Articles in Nihon Bungaku Daijiten, Tokyo, 1928.

Sonoda Tamio: Jōruri Sakusha no Kenkyū, Tokyo, 1944.

Takahashi Hiroshi: Chikamatsu no Gikyoku Taiyō (in Bungaku, Vol. II. No. 4), Tokyo, 1934.

Takayama Chogyū: Chikamatsu Sōrinshi (in Chogyū Zenshū), Tokyo, 1905.

Tōgi Tetteki: Ongaku-jō yori mitaru Kokusenya Kassen (in Tsubouchi et al.: Chikamatsu no Kenkyū).

Tsubouchi Shōyō et al.: Chikamatsu no Kenkyū, Tokyo, 1900.

Ueda Kazutoshi and Higuchi Yoshichiyo: Chikamatsu Go-i, Tokyo, 1930.

Wakatsuki Yasuji: (Ningyō) Jōruri-shi Kenkyū, Tokyo, 1943.

Yamaguchi Gō: Edo Bungaku Kenkyū, Tokyo, 1933.

Yamaguchi Gō: Kokusenya Kassen no Beni-nagashi ni tsuite (in Edo Bungaku Kenkyū).

Chinese Works

Hsieh Kuo-chen: Wan-ming Shih-chi K'ao, Peiping, 1932.

Huang Tsung-hsi: Ssu-hsing Shih-mo in Li-chou I-chu Hui-chi.

European Works

Brewster, Dorothy: Aaron Hill, New York, 1913.

Burckhardt, Jakob: Civilization of the Renaissance, Oxford, 1945.

Hill, Aaron: Dramatic Works, London, 1760.

Hummel, A. W., ed.: Eminent Chinese of the Ch'ing Period, Washington, 1943.

Lee, Vernon: Studies of the Eighteenth Century in Italy, London, 1880.

Miyajima, Tsunao: Contribution à l'Etude du Théâtre des Poupées, Osaka, 1928.

Murry, J. M.: Shakespeare, London, 1948.

Ponsonby-Fane, R. A. B.: Koxinga, Chronicles of the Tei Family, (TPJSL, Vol. XXXIV) London, 1937.

Ross, John: The Manchus or the Reigning Dynasty of China, Paisley, 1880.

Takeda, Izumo: Chushingura, (tr. Jukichi Inouye), Tokyo, 1910.