[In the following excerpt, Butler examines the authorship and dates of creation of the Shibu text, arguing for its acceptance as the original Heike Monogatari.]

… THE AUTHORSHIP OF THE SHIBU TEXT

The starting point for all investigations of the authorship of the original Heike monogatari has been the following section of the Tsurezuregusa, a miscellany written by Yoshida Kenkō (1282-1350) about the year 1330.

Section No. 226: During the time of the Retired Emperor Go-Toba,¹ the former Governor of Shinano Yukinaga² was renowned for his learning, but because he forgot two of the virtues of the “Dance of Seven Virtues”³ when he took his turn in discourse on yüeh-fu, he was given the sobriquet of “Five Virtues Kanja.”⁴ Feeling miserable, he forsook scholarship and took Buddhist vows. The Priest of the First Rank Jichin⁵ took people even of low rank who had some artistic talent into his service and carefully looked after them, and so he gave this Shinamo Lay Priest a stipend.

This Lay Priest Yukinaga composed the Heike monogatari, taught it to a blind man named Shōbutsu,⁶ and had him recite it. Now he wrote especially well of matters concerning the Enryaku-ji. He was well informed of events concerning Kurō Hōgan,⁷ and in writing included them. But evidently not knowing much of events concerning Kaba no Kanja,⁸ he failed to record many things. As for matters concerning warriors and the arts of archery and horsemanship, Shōbutsu, being from the Eastland, questioned warriors and had it written down [by Yukinaga]. The natural voice of this Shōbutsu is studied by present-day biwa hōshi.⁹

Although Yoshida Kenkō certainly made no particular effort to include in the Tsurezuregusa only historically accurate information, research shows that a good part of what he wrote has some basis in historical fact. This is true of the above account. It is not correct in every detail, and contains a certain amount of erroneous interpretation, but some of the information given about the authorship of the Heike seems to be close to what the actual facts must have been. The three major points made by Kenkō are: (1) the attribution of the authorship of the original Heike monogatari to the Lay Priest Yukinaga; (2) the statement that Yukinaga was connected with Jichin; and (3) the dating of the Heike at the time of the retired Emperor Go-Toba (1198-1221). The first two points will be considered here, and the third one taken up in the following section.

The scholarship concerning the above Tsurezuregusa section would be tedious to relate in full, but in summary, the accepted interpretation of this entry is that Yukinaga did in fact write the original Heike monogatari, which was a text written for recitation, but that this text has been lost, and all of the preserved Heike monogatari manuscripts are revisions deriving from Yukinaga's original recited text. But when the Tsurezuregusa information is compared with evidence concerning the textual evolution of the Heike derived from study and comparison of the early manuscripts, a basic contradiction becomes apparent. Yoshida Kenkō stresses Yukinaga's renown and ability as a scholar of Chinese learning, and other historical records substantiate this. If Yukinaga did write the original Heike monogatari, it would be reasonable to suppose that he wrote it in kambun, the “literary language” of a Chinese scholar.¹⁰ But by the statement, “taught it to … Shōbutsu and had him recite it,” the Tsurezuregusa account implies that Yukinaga's original Heike was a recited text,¹¹ and there are no pure kambun versions of the Heike that were written and used as recited texts.

What seems to have happened in the Tsurezuregusa account is that while correctly attributing authorship of the original Heike monogatari to Yukinaga, Yoshida Kenkō has erred in implying that this was a recited text. In effect, he has confused the later revision of the kambun Shibu text for recitation with the original writing of the text, and in so doing has compressed at least thirty years of textual development to make it appear that the Heike monogatari was originally written for the purpose of providing a fixed text for recitation. This is a natural mistake for one of Yoshida Kenkō's period to make, since the date of the Tsurezuregusa roughly corresponds to the time Kakuichi was producting his refined recited version of the Heike. At this time recitation of the Heike monogatari was becoming so popular¹² that it obscured the fact that the original Heike monogatari prior to revision for recitation was a pseudo-historical chronicle, intended to be read, not recited.

A complete account of Yukinaga's life as it can be reconstructed from the records and diaries of the early Kamakura period is available in Japanese.¹³ Here only the highlights need be related. Yukinaga was born about the year 1164,¹⁴ the second son of Nakayama (Hamuro) Yukitaka, a man mentioned in Kujō Kanezane's diary and other records of the period as having been in charge of rebuilding the Great Buddha Pavilion of the Tōdai-ji in Nara.¹⁵ Yukitaka is mentioned many times in the Gyokuyō, and it seems clear that he was on intimate terms with Kanezane.¹⁶ Due perhaps to Yukitaka's connections, we find Yukinaga listed in the Gyokuyō entry of Bunji 6:4,10 (1190)¹⁷ as one of Kanezane's family retainers, although it is not clear when he assumed this relationship. In the Gyokuyō entry of Kenkyū 5:9,17 (1194), Yukinaga is listed as “Yukinaga, Governor of Shimotsuke,”¹⁸ and Japanese scholars have suggested that in the Tsurezuregusa account, Shinano is a mistake for Shimotsuke.¹⁹ In an entry in the Sanchōki dated Kenkyū 6:8,15 (1195), we find Yukinaga listed as again without court office,²⁰ and this is generally accepted as indicating that his tenure as provincial governor ended at about this time.²¹ The Sanchōki entry is of interest also in that Yukinaga's name is listed with that of Fujiwara Teika (1162-1241), of Shinkokinshū fame. This is understood to indicate that Yukinaga's position was of a level similar to Teika's.²² In addition to his court rank and status as a retainer of Kanezane, Yukinaga was quite well known as a Chinese poet and scholar of Chinese learning. He is mentioned in several records as having taken part in various Chinese poetry contests and such, and seems to have been above average in talent.²³ By the year 1212, Yukinaga had become a retainer of Yoshisuke (d. 1218), Kujō Kanezane's son.²⁴ Yoshisuke is referred to in the Gukanshō (p. 323) as being “unparalleled in either past or present in his ability in Chinese learning,” and it may have been due to a similarity of interests that Yukinaga became a retainer of Yoshisuke. The fact that Kanezane died in 1207, and his eldest son, Yoshitsune, preceded him in death by one year, is also undoubtedly relevant to Yukinaga's finally entering the service of Yoshisuke.

Now in the Tsurezuregusa account, it states that Yukinaga, as a result of humiliation at the time he discoursed on one of Po Chü-i's poems, had taken Buddhist vows and entered the service of the Priest of the First Rank Jichin. Jichin, or Jien, as he is better known to history, was the younger brother of Kanezane, Yukinaga's former master, and uncle of Yoshisuke, Yukinaga's master in the year 1212. This in itself tends to make plausible the Tsurezuregusa statement. There is no record giving the date or the fact that Yukinaga took Buddhist vows, but if we follow the Tsurezuregusa and consider that it was after formal discourse on the hsin yüeh-fu poetry of Po Chü-i during the time of the Retired Emperor Go-Toba (1198-1221), then there are only two possible dates for this event. One is the “Fifth Discourse on yüeh-fu” held on Jōgen 4:3,15 (1210),²⁵ and the other is the “Sixth Discussion of the Works of Po” held on Kempō 6:6,3 (1218).²⁶ If Yukinaga did in fact retire from the world after one of these two events and then write the Heike monogatari, the later date of 1218 would be the more likely, since Yukinaga's name appears in the diary of Fujiwara Teika in the years 1212 and 1213 as still a man of this world.²⁷ A further fact serving to suggest that Yukinaga may have taken vows in 1218 is that during this year his second master, Yoshisuke, also died. This, coupled with the previous deaths of Kanezane and Yoshisuke's elder brother Yoshitsune, suggests the possibility that with the loss of his secular masters, Yukinaga would renounce the world and enter the service of the Buddhist priest Jien, the remaining adult member of the Kujō Fujiwara family.

From the preceding brief account of Yukinaga's life it can be seen that he possessed the two important qualities necessary for authorship of the Shibu text: one, he was a scholar and poet of Chinese literature, hence skilled in kambun; and two, he was a retainer of Kujō Kanezane and members of his family, and therefore in a position to have had access to the Gyokuyō. When we look at the Shibu text itself, in addition to the Kanezane sesshō section, several other of its features serve also to point to Yukinaga as its author. First, of all the Heike monogatari variant manuscripts, the Shibu is most permeated with political and philosophical concepts which are traceable to Jien, the political theoretician of the Kujō family. Jien's political philosophy is presented in his work, the Gukanshō.²⁸ When various sections of the Shibu are investigated, it can be seen that the author has incorporated several of Jien's basic ideas. A “Dream Section” of maki 5, to be discussed later in connection with dating the Shibu, is one of these. Another conspicuous example appears towards the end of maki 1, in a section describing an attack on Kyoto by Buddhist monks from a temple connected with the Enryaku-ji on Mt. Hiei. Here there is a long digression, which does not appear in any of the other Heike variant manuscripts, concerning the function of the Enryaku-ji, the temple headed by Jien, as spiritual protector of the Japanese Court.²⁹ This idea was not original with Jien, but it was basic to his over-all scheme of thought and was strongly advocated by him as part of his general plan to seize political control of Japan for the Kujō family in the early part of the thirteenth century.³⁰ Since this section does not appear in any other Heike variant, it seems possible that it was included in the Shibu text by Yukinaga due to Jien's influence. Then by a process similar to that observable in the Kanezane sesshō appointment section, it was dropped out in the later recited revisions, as being disruptive to the basic narrative of the Heike story.

Another feature connecting Yukinaga with the Shibu is the treatment given his father, Yukitaka. Yukitaka was a rather minor official in comparison with the major figures who make up the principal characters of the Heike monogatari, but he is mentioned several times in the Shibu text in circumstances which suggest the author had a personal reason for including him. The longest account of Yukitaka in the Shibu comes towards the end of maki 3, and covers four full pages. This account relates that Yukitaka had been dismissed from office and was living in retirement, but that after Kiyomori, the Heike leader, achieved power, he granted Yukitaka many presents and had him reappointed to the office of ben, or “Controller.” The Shibu account reads as follows:

There was a person called the former Lesser Controller of the Left Yukitaka, eldest son of the deceased Nakayama Middle Counselor Lord Akitoki.³¹ During the time of the Retired Emperor Nijō,³² he was a personal retainer and at the time that he became a Controller, he passed over the Lesser Controller of the Right Nagafusa and attained [the position of] the Left.³³ At the time he was appointed to the Full Fifth Rank also, he passed over eight outstanding men, and [his rise] was phenomenal. But after the death of the Retired Emperor Nijō, he lost his prosperity, and on Nin'an 1:4,18 (1166) he resigned his office. After entering retirement, he for long lost all prospects; he passed fifteen years of spring and fall without changing his summer and winter clothing, and the intervals between the smoke of morning and evening [cooking fires] were many.³⁴ As he spent his days weeping sad tears, on the sixteenth [of the month (1180)] as evening deepened, there was a messenger from the residence of the Lay Priest Prime Minister³⁵ who said [for Kiyomori], “Come! There is a matter I must discuss.” Yukitaka was greatly excited, wondering what this matter was. “Many people are meeting with misfortune, what should I do? I have not been involved in anything for these [past] fourteen years, but despite this there must have been someone who slandered me and said that I was planning an uprising. He said to come immediately, but I have no oxcart,” [he said] worrying. He sent a messenger to his younger brother, the one called the former Major of the Left Military Guards, to tell him of this matter and request that he send an oxcart and court clothing. His wife and sons were sick at heart and wept, wondering what would happen. Finally he went to the Nishihachijō Residence³⁶ and met the Lay Priest, who said: “In addition to being friends with the deceased Middle Counselor,³⁷ I especially discussed all matters great and small [with him], and I regret his passing. I do not intend to neglect you. Your retirement has been long and I have regretted this, but since it was the will of the Priestly Retired Emperor [Go-Shirakawa], I have been powerless. [But] now you should be employed. I have arranged everything.” [Yukitaka] returned to his residence, and when he reported that the Lay Priest had said this, beginning with his wife, [everyone] wept together in happiness. The following morning the Gen Fifth Rank Lieutenant Suesada came bringing a small-eight-leaved cart³⁸ drawn by the Lay Priest's oxen, with driver and livery, and in addition to presenting [Yukitaka] with one hundred hiki and one hundred koku,³⁹ [he said that] [Yukitaka] would immediately be reappointed as Controller. [Yukitaka] was so happy he could not speak and did not know where to wave his arms or stamp his feet. The entire family felt it indeed must all be a dream

(3.60a1-62a1).

Like the Kanezane sesshō appointment section of the Shibu discussed previously, this section concerning Yukitaka's reappointment as ben has no connection with the narrative preceding or following it. It therefore seems to have been inserted as a result of some direct interest on the part of the Shibu author. Considered in terms of Yukinaga's probable authorship of the Shibu, the reason for this becomes apparent. By associating his father with the great events and personages related in the Heike monogatari, Yukinaga probably hoped to show that his family also had some claim to fame.

In addition to connecting Yukinaga with the Shibu text, the Yukitaka section also sheds light on Yukinaga's reason for writing the Heike monogatari and on the bias his work displays. First, although the Genji were the final victors in the Genji-Heike confrontation, in the Shibu (and subsequently in the Yashiro and Kakuichi recited variants) the Heike, rather than the Genji, are the focal point of the narrative and are depicted in a sympathetic manner.⁴⁰ Secondly, the Cloister Emperor Go-Shirakawa, although nominally the expected rallying point for the court nobles against both the Heike and the Genji warrior groups, is never developed as a character in the Shibu to the extent one would expect on the basis of his direct involvement in most of the major events related. When he does appear in the narrative, he is usually depicted from a rather hostile point of view.⁴¹ These two aspects of the Shibu would suggest that its author was probably someone with connections with the Heike and also one bearing animosity for Go-Shirakawa. In the early part of the Yukitaka section, Yukinaga's father is pictured as having had a phenomenal career under the patronage of the Emperor Nijō, but being forced to retire completely from court life after Go-Shirakawa emerged as the ruling power in Kyoto. In the latter part of the section the opposition of Go-Shirakawa is made even more explicit by having Kiyomori state that Yukitaka's retirement was due specifically to the will of Go-Shirakawa. In the latter part also, we are shown that it was Kiyomori, the Heike leader, who rescued Yukitaka from retirement and started him once again on his official career. It is perhaps not forcing the interpretation too much to assume that the aid given Yukinaga's family by Kiyomori was responsible at least in part for his motive in writing his account of the Heike, and also for his sympathetic treatment of them. Similarly, by showing that Go-Shirakawa was responsible for his family's long privations, Yukinaga indicates the basis for his animosity towards Go-Shirakawa, and this is consistent with the treatment given Go-Shirakawa throughout the Shibu text.

One further point which tends to fill in the picture of Yukinaga's authorship of the Shibu is an entry dated Jōkyū 2:4,22 (1220) in the Gyokuzui, the diary of Kujō Michiie (1193-1252), grandson of Kanezane and sesshō prior to the Jōkyū War of 1221. Michiie mentions that he sent a certain Nagatame to the residence of Mitsumori, son of Kiyomori's brother Taira Yorimori (1132-1186), to borrow and copy the many “Heike kiji” possessed by Mitsumori.⁴² There has been some question as to what sense should be given to the term kiji in this entry. The most recent interpretation is that it refers to some type of “Heike story” which was written prior to the original Heike monogatari.⁴³ The normal meaning of the word kiji is, however, “records.” On the basis of Yukinaga's probable authorship of the Shibu text, and his connection with the Kujō family, another possible interpretation of this entry would be that the “Heike kiji” were personal family records of the Heike in the possession of the one remaining branch of the Heike, that stemming from Yorimori.⁴⁴ As will be shown in the following section, the Shibu text was being written during the year 1220, and it is possible that Kanezane's grandson Michiie borrowed and had copied such Heike records for Yukinaga to use as historical source material in writing sections of the Heike monogatari.

THE DATES OF COMPOSITION OF THE SHIBU AND YASHIRO TEXTS

Internal evidence can be found in the Shibu text of the Heike monogatari which establishes with a fair degree of certainty that it was written during the years 1218-1221. These dates fall within the period when Go-Toba held the title of Retired Emperor (1198-1221), given in the Tsurezuregusa as the date for the writing of the original Heike monogatari. The terminal date of 1221 may be derived as follows. At the end of maki 12 of the Shibu text, following the account of the search for and execution of Heike descendants and retainers, which appears after the Rokudai Gozen story and the Kanezane sesshō account, there is the following passage, referring to an incident in 1199:

Mongaku,⁴⁵ being by nature a person of immoderate temperament, [felt that] the reigning Emperor [Go-Toba] gave himself over solely to pleasures, and knew nothing of the government of the world. After causing Lord Kujō's⁴⁶ retirement, [the reigning Emperor] lived at [the residence of] Kyō no Tsubone,⁴⁷ and paid no heed to the afflictions of the people. Since the Second Imperial Prince⁴⁸ alone did not neglect learning and put proper principles first, [Mongaku] planned to place him on the throne, and have him carry on the government of the world; but while the Udaishō [Yoritomo] was living it was impossible. But after [the Udaishō] passed away on Shōji 1:1, 13 [1199], when it was reported that [Mongaku] was still planning this matter, Mongaku immediately incurred the censure of the Retired Emperor; kembiishi⁴⁹ were sent to his residence at Nijō Inokuma,⁵⁰ and he was banished to Tosa province. After that Rokudai Gozen had not been at Takao⁵¹ at all, making pilgrimages …

(12.48b3-49a5).

This section presents much that is important in dating the Shibu text and connecting it with Yukinaga. First, we again have mention of Lord Kujō (Kanezane). His forced retirement from the office of sesshō was an act traceable ultimately to Go-Toba, which paved the way for Go-Toba's own attempt at forming an insei government led by himself, and one which incurred for Go-Toba the lasting enmity of Jien and the Kujō family.⁵² There is no logical reason for reference to be made to it here, however, unless the author did so in order to express obliquely his dissatisfaction with Go-Toba's dismissal of Kanezane by associating it with criticism of Go-Toba's conduct. Secondly, the Retired Emperor Go-Toba led the attack on the Kamakura bakufu, known as the Jōkyū War,⁵³ in 1221 and was subsequently banished to Oki province. No mention is made of this event in the Shibu account, although it does appear in this section in all other Heike variants. This discrepancy suggests that the Shibu account was written prior to Go-Toba's banishment in 1221. Further evidence to this effect is the circumlocution achieved by the term “the Reigning Emperor.” Go-Toba was Reigning Emperor from 1184 to 1198. Afterwards he was Retired Emperor until his banishment in 1221; and after other Emperors retired, he was referred to as Senior Retired Emperor. Following his banishment to Oki province in 1221 until his death in 1239, Go-Toba was referred to as “the Retired Emperor at Oki”. He did not receive the posthumous title of Go-Toba-in until 1242, three years after his death. The use of proper terminology in the Shibu text in referring to Go-Toba prior to 1196 only as tōkon, and after this date in 1199 as in, and the failure to use the term Oki no In or Go-Toba-in, together with the failure to mention the Jōkyū War and Go-Toba's banishment to Oki, may be adduced as evidence that the Shibu text was written prior to 1221. This becomes apparent when the treatment given the same section in the Yashiro text, which was definitely written after 1221, is considered. The corresponding Yashiro section reads as follows:

At that time the term Reigning Emperor referred to Go-Toba-in. He gave himself over to pleasures, and since all under heaven was following completely the will of Kyō no Nii,⁵⁴ the grief and affliction of the world was unending. Mongaku of Takao, viewing this, grieved over the perilous condition of the world, and since the Second Imperial Prince did not neglect learning and put proper principles first, he planned somehow to place him on the throne. However, while Kamakura no Udaishō was living, he did not, after all, mention it, and the Reigning Emperor abdicated in favor of the First Imperial Prince.⁵⁵ After Kamakura no Udaishō Yoritomo died on Shōji 1:1 [1199], when Mongaku attempted this affair, it was immediately reported, and he was arrested, and at the age of over eighty-two he was banished to Oki province. Since the Retired Emperor was fond to excess of bat-ball,⁵⁶ at the time he was punished by the Punishment Official and the Prisoner Escort,⁵⁷ and driven from the capital, [Mongaku] gave vent to many insults. Saying, “Would that in the end I meet that bat-ball scamp in the place where I am banished,” he was banished. When he arrived in Oki province he at length died of melancholy. Although he was fearful in manner, he was foolish. However, when in the summer of the third year of Jōkyū [1221], the Senior Retired Emperor attempted to attack Ukyō no Gon no Taifu Yoshitoki,⁵⁸ he was defeated in battle; although there were many places, that he should be sent to Oki province was indeed sad. Rokudai Gozen, under the name of Sammi no Zenshi was practicing religious austerities, but after …

(12.60b7-61b8).

In this section of the Yashiro several revisions and expansions of the Shibu narrative have occurred. First, the reference to Go-Toba (tōkon, “Reigning Emperor,” in the Shibu text) is made explicit by the opening statement: “At that time the term Reigning Emperor referred to Go-Toba-in,” as though the Yashiro revisor were providing a gloss to the Shibu text, which he apparently was. The use of the term Go-Toba-in indicates beyond doubt that the Yashiro text was written after 1242, the year Go-Toba was given this name.⁵⁹ Another important difference between the two accounts is that in the Yashiro there is no mention of Kanezane's dismissal from the office of sesshō. The Yashiro revisor was working at a time more than fifty years after the event, and undoubtedly had no connection with Kujō descendants, whereas Kanezane's dismissal was a very real event to the Shibu author, Yukinaga, and probably for this reason given prominence by him in his criticism of Go-Toba.

An even more significant revision in the Yashiro text of the Shibu account in terms of content, is the mention of the Jōkyū War and the change of Mongaku's place of banishment from Tosa province to Oki province. It is historical fact that after the abortive Jōkyū War Go-Toba was banished to Oki. Knowing of this, the Yashiro revisor evidently wanted to provide a more dramatic effect, and therefore changed Mongaku's place of banishment to Oki province also.⁶⁰ By the final stage of the Kakuichi text of 1371, the end of this passage appears as: “Furthermore, when in the Jōkyū period [the Retired Emperor Go-Toba] raised a rebellion, although there were numerous provinces, that he should be sent to Oki province is indeed unaccountable. It was rumored that Mongaku's ghost was roaming about in that province and frequently talked to him” (KBT 33, 421-422).

There are several other elements of the above Yashiro section, such as the addition of identifying tags to the titles of personages to make the narrative more suited to oral presentation, and various other expansions, which mark it as a revision of the Shibu account. When the Shibu section of maki 12 is thus considered in terms of this Yashiro revision, the evidence seems conclusive that it was written prior to the Jōkyū War in 1221. There are no sections of the twelve maki of the Shibu which can be shown to have necessarily been written after 1221, so that this may be taken as the approximate terminal date for its writing.⁶¹

In determining the earliest date the writing of the Shibu could have been begun, an account of a dream presented in maki 5 is of importance. In brief, it is related that a Genji retainer in Kyoto had a dream which foretold that the Genji would take over rule of Japan from the Heike, and then in turn the Fujiwara would succeed the Genji in power. This section reads as follows:

A dream had by a person serving Lord Gen Chūnagon Garai⁶² was indeed portentous. He dreamed that at a place resembling the Bureau of Worship in the Palace, many people dressed in ceremonial robes and hats were assembled in deliberation. A person seated in the place of lowest rank, [who appeared to be] a partisan of the Lay Priest Prime Minister [Kiyomori], was driven out [of the meeting]. [Then] a person of brave and distinguished appearance, seated at the head [of the gathering], said, “The sword which previously had been entrusted to the Lay Priest Prime Minister should now be entrusted to Yoritomo.” [Next] a person seated towards the middle said, “After that it should be entrusted to my descendants.” Still dreaming, [Garai's retainer] stood up and approached a person near the door and asked who was [the person] at the head [of the group]. He answered, “The person on the corridor side is Hachiman no Daibosatsu. The one who said ‘My descendants,’ is Kasuga no Daimyōjin. The one who was driven away is Itsukushima no Daimyōjin.” … After awaking from his dream [Garai's retainer] thought it was very strange, and because he told about having this dream, people broadcast it about. The Lay Priest heard of this and was very angry. He sent Yukitaka no Ason⁶³ to question Lord Garai, but nothing came of it. [Lord Garai said,] “After reporting this matter, the one who had the dream, and family, disappeared.” That the Itsukushima no Daimyōjin was a partisan [of the Heike] was well known. But Saishō Nyūdō Shunken⁶⁴ [said], “Itsukushima is a female god, so I can understand this much. But it is strange that the Kasuga no Daimyōjin said, ‘[the sword] should be entrusted to my descendants.’ Does this mean that in later years there will be a Fujiwara taishōgun?”

(5.16a6-18b1).

In this account, the sword is used figuratively to symbolize political and military control of Japan. The Hachiman no Daibosatsu was the protective god worshiped by Yoritomo, the Genji leader, while the Kasuga no Daimyōjin served the same function for the Fujiwara, and the Itsukushima no Daimyōjin was the family god of the Heike. The statement that the sword should ultimately be passed to the descendants of the Kasuga no Daimyōjin, that is, the Fujiwara, and the forecast of a Fujiwara taishōgun may be interpreted as referring to the dispatch in 1219 of a great-grandson of Kujō (Fujiwara) Kanezane to Kamakura to be seii taishōgun after the extinction of Yoritomo's line.⁶⁵ The appointment of Kanezane's great-grandson to the position of Shogun was part of the over-all bumbu kenkō (parallel rule by civil and military) scheme of Jien, which envisioned a Kujō family member as sesshō and a Kujō descendant as Emperor in Kyoto, at the same time a Kujō family member was Shogun in Kamakura. This, combined with Jien's position as the dominant figure of the Enryakuji, the Buddhist temple which traditionally was the protector of the Kyoto Court, would have given the Kujō family a monopoly on all of the important political and religious offices in Japan, and put them in a position, they felt, to usurp political power gradually from the Hōjō regents of the Genji bakufu in Kamakura.⁶⁶ This plan seemed to be on the verge of success in 1221, when the son of Kujō Yoshitsune's daughter was enthroned as Emperor, another grandson, the son of Michiie, was in Kamakura as the shogun-designate, and Michiie himself was sesshō. But the machinations of the Kujō were brought to an abrupt halt by Go-Toba's ill-advised attack on the Kamakura bakufu in the fifth month of 1221. By the time the Jōkyū War had ended, the Hōjō regents were in a position to exert directly their authority throughout Japan, and were in no need of political alliances with Court aristocrats such as the Kujō Fujiwara. Thus all prospects for a Fujiwara return to power were ended.

An awareness of this bumbu kenkō plan of Jien provides a key to understanding the Shibu dream section reference to power being passed to the Fujiwara. The Kujō had been trying to achieve political power ever since Kanezane first secured appointment as sesshō in 1186. But it was only during the later stages of the Kujō conspiracy that efforts were directed toward obtaining appointment of a Kujō as Shogun in Kamakura. Upon the death of Yoritomo's last son, the third Shogun Sanetomo (1192-1219), the Hōjō in Kamakura attempted to have one of Go-Toba's sons appointed as Shogun, apparently hoping by this move to make further inroads into the Imperial prerogatives. Go-Toba, realizing this and also undoubtedly by this time thinking of possibly attacking the bakufu, refused to allow it. Next, the infant son of Kujō Michiie was put forward as a candidate for Shogun, and it was at this time (1219) that the possibility of a Kujō Shogun suddenly materialized. Hence, if we take the reference to a Fujiwara Shogun in the dream section as referring to the appointment of Michiie's son, which by all calculations it must, then this indicates that at the time of writing this section, the Shibu author knew that his fictional prophecy “Does this mean that in later years there will be a Fujiwara taishōgun?” had already been, or was on the point of being, fulfilled. Therefore 1219, the year Michiie's son was sent to Kamakura as Shogun, may be taken as the earliest date possible for this part of maki 5. When this is combined with the previously estimated date of 1218 for Yukinaga's vow-taking, after which the Tsurezuregusa states he wrote the Heike monogatari, the year 1218 may be accepted as the approximate date that the writing of the Shibu was begun.

In the dream section of maki 5 of the Yashiro text, an account similar to the Shibu narrative is presented, but with the important difference that all mention of the Kasuga no Daimyōjin and the Fujiwara Shogun is omitted, and the account ends with the forecast of the Genji takeover from the Heike.⁶⁷ This is what one would expect from a text produced for public recitation after the year 1242. By this time the Hōjō were firmly entrenched as the legitimate representatives of the Genji bakufu, and they would have undoubtedly objected strongly, just as Kiyomori is said to have done in the dream section, to public prediction of their demise.

In summary, on the basis of the narrative of the Mongaku banishment section of maki 12, the dream section of maki 5, and the Tsurezuregusa account, the composition of the Shibu text can be dated during the period 1218-1221. In addition, because the term Go-Toba is used in the Yashiro text, it can be dated at sometime after 1242.

THE SHIBU TEXT AS THE ORIGINAL HEIKE MONOGATARI

All of the foregoing considerations tend to corroborate Yoshida Kenkō's attribution in the Tsurezuregusa of the authorship of the original Heike monogatari to Yukinaga, and to point to the Shibu as being that original text. This poses several contradictions with traditional theories of the genesis of the Heike. First, if the Shibu chronicle is accepted as the original Heike monogatari, the theory that the original Heike was written for recitation is no longer credible. Once this idea is rejected, other historical references suggesting that the Heike monogatari was first written as a read text and then revised for recitation appear to have more validity than has heretofore been granted them. Two such references are as follows:

The Heigo gūdan, a work about Heike monogatari recitation written in 1827, states that in a manuscript of the Nagato read text of the Heike, which the author apparently had seen, there was the following notation: “The twelve maki wood-block printed book called the Heike monogatari has existed from the middle ages. This is [a work] which the person called Tameie, made on the basis of Yukinaga's original work, for the purpose of biwa hōshi singing.”⁶⁸ It has been recently pointed out that this passage actually appears at the beginning of an early manuscript of the Nagato text acquired by Tenri University and that evidently this was the copy of the Nagato text viewed by the Heigo gūdan author.⁶⁹ Thus this notation probably dates from a period earlier than 1827, though just how early is not certain. It has been suggested that the Tameie referred to in this statement was Fujiwara Tameie (1198-1275), the son of Fujiwara Teika, and a Japanese poet of some note.⁷⁰ The dates of his life would place him at maturity at about the time the Shibu text was beginning to be revised for recitation, but there is no corroborating evidence showing he had a hand in such revision. In this case, like the Tsurezuregusa account, the Nagato text notation is probably only partially correct. At any rate, it does imply that prior to revision for recitation there was an original read text of the Heike monogatari authored by Yukinaga.

In the kambun work, Gaun nikkenroku, under the date of Bun'an 5:8,19 (1448), an interview with a certain Saiichi kengyō, a second generation disciple of Kakuichi, is mentioned. He is quoted as saying: “Of old, Lord Tamenaga wrote this twelve maki [Heike monogatari] while he was residing in Harima province. Afterwards, Shobutsu set this to music and recited it.”⁷¹ Here again, the relationship of both Tamenaga and Shōbutsu to the composition and revision of the Heike monogatari is unclear, but the entry does suggest that there was, among people concerned with Heike recitation in the fifteenth century, a tradition that the Heike was revised for recitation after it was first written as a read work. In addition to these two entries, there are several other records of the history of Heike recitation, dating mainly from the Edo period, which, although not accurate in details, contain statements to the effect that the Heike monogatari was first a read text and later was adapted for recitation.⁷²

A second problem arising from acceptance of the Shibu as the original Heike results from the theory that the Heike monogatari was written first in a short three-maki form, and then expanded into six maki, and finally, through a process of further revision and expansion, attained the twelve-maki form of all the major preserved recited Heike variants.⁷³ According to this concept of the genesis of the Heike, the original three-maki text and its six-maki expansion were somehow lost, and the various early variant manuscripts such as the twelve-maki Shibu and Yashiro, being later expansions, only preserve the form and content of the original to an imperfect degree.

The idea that originally the Heike was in three-maki form stems from an account in a work of uncertain date, the Heike monogatari kammonroku, which was current during the Edo period, but which was probably written earlier, during the Muromachi period (1336-1573). This work mentions six manuscripts of the Heike monogatari, none of which can be identified as any of the preserved manuscripts, ranging in form from thirty-three maki down to three maki, and assigns an author to each.⁷⁴ A good part of the supposedly factual material appearing in the Kammonroku is in error, and there is no record of any of the “authors” given for the Heike texts having had any connection with the writing of the Heike. But due to the fact that the Hōgen monogatari and the Heiji monogatari, two minor short works of the gunki monogatari genre, are in three-maki form, succeeding generations of scholars have seized on the statement in the Kammonroku that there was a three-maki Heike as suggesting that originally the Heike monogatari was in this form also.⁷⁵ Having accepted this dubious point, they then use it to postulate a three-maki recited Heike monogatari preceding all of the preserved variants. Since there is no corroborating evidence of a three-maki Heike monogatari, the Kammonroku account upon which this theory is based should be taken only for what it appears to be—a highly imaginative attempt to manufacture more or less from whole cloth a theory of Heike textual development—and undue weight should not be given to the idea that originally the Heike was in three-maki form.

The theory that there was a six-maki text which preceded all of the preserved twelve-maki Heike texts has more validity than the three-maki theory, and deserves discussion at length. This theory is based on two inserted kambun notations in the Heihanki, a diary covering the period 1132-1171 by Taira Nobunori (dates unknown). In addition to mentioning the Heike monogatari, these notations also mention a year-period change, and on the basis of this it has been established that the two notations date from the year 1240.⁷⁶ The first notation states:

The six-maki Jishō monogatari, titled Heike, is being copied. Although it is not yet finished, I have requested to see it.

The second notation reads:

The six-maki [Jishō monogatari] was expanded into the twelve-maki [version] current in the world. The six-maki [text] in many places is the same as the current (twelve-maki text]. Half of the twelve-maki [text] has places that are secret.⁷⁷

The very nature of these notations as inserted entries in a work with which they have no connection invites their acceptance as true comments concerning the texts of the Heike made by someone who had compared them. Since it has been determined that they were written during the year 1240, they are perhaps the earliest preserved comments about the textual development of the Heike. When interpreted in terms of information derived from textual comparison of the major preserved Heike variants, these notations shed a good deal of light on how the Heike developed as a written text.

First, the record that a text called the Jishō monogatari (The Tale of the Jishō Period, [1177-1180]), existed prior to the twelve-maki version of the Heike may be interpreted as referring to the Shibu text. The full title of the Shibu text as we have it today is Shibu kassenjō daisamban tōjō, or “Campaign number three of the four-part battle record.” This title is evidently connected with, or derived from, a section in the Heike monogatari kammonroku which begins, “This Heike is a battle record in four parts …”⁷⁸ Following this statement is an explanation to the effect that the account of the battle of the Hōgen period [the Hōgen monogatari] may be termed the first battle record of Japan, the account of the battle of the Heiji period [the Heiji monogatari] may be termed the second battle record of Japan, the account of the battle of the Jishō period [the Heike monogatari] may be termed the third battle record of Japan, and finally, the account of the battle of the Jōkyū period [the Jōkyūki] may be termed the fourth battle record of Japan. The significant point of all this in terms of the Shibu text is that it was written prior to the battle of the Jōkyū period in 1221, hence prior to the Jōkyūki, the fourth battle record of Japan. Therefore, at the time it was written it would have been impossible for the author to have given the Shibu the title of “Campaign number three of the four-part battle record,” since he probably did not know yet of the Jōkyū battle, and certainly did not know of the Jōkyūki, which was written later during the Kamakura period.

The fact that the Shibu text bears the title of Shibu kassenjō daisamban tōjō indicates that as a result of its various copyings over the centuries it has undergone a certain amount of superficial change, and that at sometime after the appearance of the Jōkyūki, and possibly after the Kammonroku (the work that seems to be the first to have established the four categories of battle records), someone attached the present title to the Shibu text. But the Shibu must have had some title from the start. The first part of the name “Campaign number three of the four-part battle record” is a circumlocution for what in analogy with the Hōgen monogatari and the Heiji monogatari, both named after year-periods, would have been termed the Jishō monogatari. Since the Heihanki notation of 1240 refers specifically to a Jishō monogatari being in existence prior to this date, it seems probable that the text presently known as the Shibu kassenjō daissamban tōjō is in fact the original Jishō monogatari text, which has merely had its title changed to conform to its being the third in a series of four battle records. The only major flaw in such a theory is that the Heihanki notation states that the “Jishō monogatari titled Heike” was a six-maki text, rather than a twelve-maki text as is the Shibu. This apparent contradiction might be explained, however, by the fact that after the appearance of the twelve-maki Yashiro and Kakuichi recited texts, the twelve-maki form came to be accepted as standard for the Heike monogatari. Perhaps some copyist of the Shibu, influenced by this, revised the six-maki Shibu (Jishō monogatari) text to make it also conform to this form.⁷⁹

There remains one further problem arising from the Heihanki notations. In addition to the reference to the Jishō monogatari, it is stated that from the six-maki work there developed a twelve-maki version of the Heike. If it is accepted that the Shibu text is the Jishō monogatari, then it would seem logical to suppose that the twelve-maki text referred to would be the recited Yashiro text, since it has been shown that the Yashiro is an early revision for recitation stemming from the Shibu. But the Heihanki notations were written in the year 1240, and it will be recalled that internal evidence shows that the Yashiro text was written after 1242. Thus it is hardly likely that the twelve-maki text of the Heihanki notation is the Yashiro text. It is therefore necessary to look elsewhere among the preserved early variants of the Heike for a text to fit the Heihanki notation's description. …

Notes

1. Go-Toba-in (1180-1239) was Retired Emperor from 1198 to 1221.

2. Shinano no Zenji Yukinaga. See pp. 18-19 above for an outline of his life.

3. Ch‘i-te-wu, a hsin yüeh-fu poem by Po Chü-i predominantly in seven-word lines. See Takagi Shōichi, Haku Kyōi (Tokyo: Iwanami, 1958) [Chūgoku shijin senshū, 12], i, 19-24 for the text and commentary of the poem and a list of the seven virtues.

4. Kanja, a title for persons of the Sixth Rank who did not hold office.

5. Jichin kashō (1155-1225), more commonly known as Jien. He was a Tendai abbot, author of the Gukanshō, and younger brother of Kujō Kanezane.

6. Shōbutsu is generally taken as the first reciter of the Heike monogatari (see Atsumi, 34-35), but there is no acceptable corroborating evidence of this fact.

7. Minamoto Yoshitsune (1159-1189), the younger brother of Yoritomo who led the Genji warriors in their defeat of the Heike.

8. Minamoto Noriyori (d. 1193), a younger brother of Yoritomo and a Genji general in various battles with the Heike.

9. Hōjōki Tsurezuregusa (Tokyo: Iwanami, 1958) [KBT 30], 271-272.

10. The revision of the kambun Shibu text for recitation was one of the principal factors in the development of the wakan konkō style of writing. Prior to this revision, a mixed kanji-kana style such as that of the Yashiro and Kakuichi texts did not exist. A possible exception to this statement would be the mixed kanji-katakana style of the Konjaku monogatari, but this style developed from the practice of inserting katakana into Buddhist kambun texts for shōdō lecture purposes; and while this is another factor accounting for the wakan konkō style, the Konjaku method of writing is inherently different from that of the Yashiro and Kakuichi Heike texts. Kambun was still the standard written language at the time the original Heike was written, especially among people like Yukinaga who were concerned with Chinese studies.

11. For what has been the standard Japanese interpretation of the significance of the Tsurezuregusa implication that the Heike was originally a recited text, see Takagi Ichinosuke, “Heike monogatari no jojishiteki kanren”, Heike monogatari koza (Tokyo: Sogensha, 1957), i, 5-39, esp. 15-19. This interpretation is followed in Atsumi, 26-27.

12. Yamada Yoshio, “Heike monogatari kō zokusetsu”, Kokugakuin zasshi (April, 1933), 61-62, quotes two sources which mention recitation of the Heike by biwa hōshi in the years 1315 and 1321.

13. One of the most recent and most accurate and comprehensive accounts of Yukinaga's life appears in Ishida Yoshisada, “Heike monogatari to Shinkokinshū, Yukinaga sakushasetsu no kentō” Bungaku 30 (1962). 52-64. Atsumi also has an account and discussion of Yukinaga's life and quotes many of the references to him in historical sources, although her conclusions differ from those presented here (Atsumi, 25-35).

14. Ishida, 56.

15. Gyokuyō, Juei 3:1,5 (1184), iii, 2.

16. Gyokuyō, Juei 3:6,23 (1184), iii, 24; Bunji 2:3,17 (1186), iii, 173.

17. Gyokuyō, iii, 605.

18. Shimotsuke no Kami Yukinaga, Gyokuyō, iii, 891.

19. Atsumi, 29.

20. Quoted in Ishida, 57.

21. Ibid.

22. Ibid.

23. One such reference is in the Meigetsuki, Fujiwara Teika's diary (Tokyo: Kokusho kankōkai, 1911), i, 278 (Kennin 2:8,23 [1202]). In this entry Yukinaga is referred to as Saki no Shimotsuke no Kami (“the former Governor of Shimotsuke”). Yukinaga also took part in the Japanese and Chinese poetry contest held at Go-Toba's residence on Genkyū 2:6,15 (1205). See Ishida, 57.

24. Ishida, 58.

25. Rakufu mondō goban. Ichidai yōki, Jōgen 4:3,15 (1210). Quoted in Ishida, 59.

26. Hakushi monjū rongi rokuban. Hyakurenshō (Tokyo: Kokushitaikei kankōkai, 1929), 151-152 (Kempō 6:6,3 [1218]).

27. Meigetsuki, ii, 147 (Kenryaku 2:2,12 [1212]) and ii, 286 (Kempō 1:6,11 [1213]).

28. The Gukanshō was written during the period 1220-1221. Although it is cast in the form of a record of Japanese history, in the later part of it Jien presents a view of Japanese history slanted to establish the Kujō claim to dominance of the Fujiwara, and advocating the sesshō form of government with a Kujō at its head. He also includes a theoretical argument for the validity of his bumbu kenkō (parallel rule by civil and military) plan, which envisioned eventual political control of Japan by the Kujō. See especially the concluding furoku section, 287-329 (translated by Johannes Rahder, “Miscellany of Personal Views of an Ignorant Fool,” AO 15(1937).173-230).

29. Shibu 1.57a5-60a6. This passage is quoted in Atsumi, 138-139, where it is accepted as representing Jien's thought. The passage is cast in the form of a statement by Chōken, the father of the probable first revisor of the Shibu, Shōkaku. See below, pp. 34-35.

30. For a discussion of Jien's concept of the protection of the state by Buddhism, see Taga Munehaya, Jien (Tokyo: Yoshikawa Kobunkan, 1959) [Jimbutsu sōsho, xv], 135-142.

31. For the genealogy of the Hamuro Nakayama family, see Heike monogatari, Takagi Ichinosuke et al., eds. (Tokyo: Sanseidō, 1958) [Kokugo kokubungaku kenkyūshi taisei ix], 52. This work quotes all of the early historical references to the Heike (47-69) and gives a complete bibliography of Heike studies through 1957 (275-304). In the present article some of the more obscure references to the Heike are keyed to this source, hereafter abbreviated as Takagi.

32. Nijō no In (1143-1165). He retired in 1165 and died the same year. Here the reference is to the reign of Nijō prior to his retirement, 1158-1165.

33. The Controller of the Left ranked above that of the Right.

34. I.e., he lacked official position and therefore passed his days in poverty.

35. Nyūdō Shōkoku, the title of Kiyomori.

36. Nishihachijō-dono, the residence of Kiyomori on Nishihachijō (street) in Kyoto.

37. Ko-chūnagon, a reference to Akitoki, Yukitaka's father.

38. Kohachiyō no kuruma, a cart with small eight-leaved figure decorations on its sides, used by people of the Fourth and Fifth Rank.

39. Hiki, koku, measures of silk and rice respectively.

40. The neglect of the Genji in the Shibu text is evidenced by the almost immediate composition in Kamakura of the Gempei tōjōroku, a revision in the read division of textual development, which added much detail about the Genji and remedied what to them was a serious defect in the Shibu account. The transmission of the Shibu text from Kyoto to Kamakura may have resulted from the appointment of Kujō Michiie's grandson as Shogun-designate in 1219. For a study of the Gempei tōjōroku as a revision of the Heike made by eastern warriors associated with the Genji, see Yamashita Hiroaki, “Gempei tōjōroku kanken”, Kokugo to kokubungaku (August, 1961).24-40.

41. See Yamashita, “Katari to yomi,” 39, 44, and Ishimoda Shō, Heike monogatari (Tokyo: Iwanami, 1955), 75. Ishimoda acknowledges that Go-Shirakawa is treated unsympathetically and suggests that the cause was in part the original Heike author's opposition to the insei system and support of the sesshō system of rule. Yukinaga would also qualify as author in this respect. See also Yamashita, “Katari to yomi,” 35-36 for a brief discussion of the sympathetic treatment of the Heike in the Shibu vs. the Genji bias of the Gempei tōjōroku.

42. Quoted in Takagi, 51.

43. Atsumi, 59.

44. Yorimori was a brother of Kiyomori who took no direct part in the Gempei battles and therefore survived.

45. See note 28 above.

46. Kanezane. He was dismissed from the office of sesshō in 1196.

47. Kyō no Tsubone (1155-1229), daughter of Fujiwara Norikane, and Go-Toba's wet-nurse during his early life. Also known as Kyō no Nii. After Go-Toba became Retired Emperor she was one of his closest advisors and influenced him greatly.

48. Ni no Miya, second son of the Emperor Takakura, who had been passed over at the time Go-Toba was made Emperor.

49. Kembiishi, also read kebiishi, “police.”

50. Nijō Inokuma, the residence of Mongaku at the intersection of Nijō and Inokuma streets in the northern section of Kyoto.

51. Takao, the mountain just north of Kyoto where Mongaku's temple, the Jingo-ji, was located.

52. Kanezane's dismissal is discussed in Gukanshō, 247-248. After being appointed sesshō, Kanezane had his daughter Taeko made consort of Go-Toba at the time of Go Toba's manhood ceremony at the age of eleven in 1190. He hoped his daughter would bear a future emperor. Then as sesshō and maternal grandparent of the Emperor, he would be able to establish a new Fujiwara Regency with his branch of the family at its head. But upon reaching actual manhood, Go-Toba became determined to continue the insei or direct rule by Retired Emperor, started by the Retired Emperor Shirakawa in 1086 when he wrested control from the Fujiwara. The removal of Kanezane in 1196 was the first step towards this goal. After Kanezane's dismissal, his daughter was removed from her position as consort, and Jien also resigned his position as Abbot of the Enryakuji. The Kujō family was then forced into total political eclipse. Later Jien was able to persuade Go-Toba to appoint Kanezane's son Yoshitsune as sesshō in 1202, again setting the stage for the Kujō family to attempt to seize political power. These details are related in Taga, Jien, 46-78.

53. Also read Shōkyū. On this war see William H. McCullough, “Shōkyūki: An Account of the Shōkyū War of 1221,” MN 19 (1964) 1/2.163-215; 3/4.186-221.

54. See note 101.

55. Daiichi no, the Emperor Tsuchimikado, eldest son of Go-Toba. Reigned 1198-1210.

56. Gitchō, a game played either on horseback or on foot, using mallets and a wooden ball. See Sakai Kin Nihon yūgi shi (Tokyo, 1933), 456-463.

57. Chakushi; ryōsōshi.

58. Hōjō Yoshitoki (1163-1224), son of Tokimasa (father-in-law of Yoritomo) and second Regent (shikken) of the Kamakura bakufu (1205-1224). He was head of the bakufu at the time of Go-Toba's attack in 1221.

59. The text of the Gukanshō presents another example of a contradiction in referring to Go-Toba by his posthumous name. It has been demonstrated fairly conclusively that the work was written by Jien in 1220-1221, just prior to the Jōkyū War. In the main body of the text (maki 3 through maki 6, plus the furoku appendage), the name Go-Toba is not used, he being referred to as in (the Retired Emperor). When, however, at the end of maki 2 he is listed as the eighty-second Emperor, he is referred to as Go-Toba. Japanese scholars feel that the latter part of maki 2 is a later addition written after 1221. See the introduction to the Iwanami edition, Gukanshō, 3.

    The posthumous title Go-Toba-in also appears in the Ujishui monogatari (story no. 159) and has been used in establishing that this part of the Ujishui must have been written after 1242. It is still not clear, however, whether or not this story was inserted after the original collection had been made. See Ujishui monogatari (Tokyo: Iwanami, 1960) [KBT 27], 12-15.

60. There is no historical evidence that Mongaku was ever banished to either Tosa or Oki. He was, however, banished several times to Sado, Tsushima, and Izu because of his stubborn efforts to secure contributions for rebuilding the Jingo-ji. The Shibu reference to Tosa may be a copyist error for Sado. … For a discussion of Mongaku, see Hoshino Kō, “Heike monogatari Gempei jōsuiki wa gobyū ōshi”, SZ 9 (1898).1-17. This article is quoted in full in Takagi, 80-90. See p. 86 for the reference to places where Mongaku was banished.

61. As possible internal evidence which might aid in dating the Shibu, Tomikura Tokujiro mentions an entry appearing in maki 11 (11.48a1-2) in his study, Heike monogatari kenkyū (Tokyo: Kadokawa, 1964), 245. Here the person Kura no Kami Nobumoto is glossed as being the grandfather of Norisuke and the father of Chikasuke. Norisuke died at the age of forty-three in 1234, during the reign of Emperor Shijō (1232-1242), and Tomikura suggests that the inclusion of his name can be taken as evidence that the Shibu was written after the reign of Emperor Shijō. He then combines this fact and the mention of the posthumous title Go-Toba-in in the kanjō no maki to date the entire Shibu text at sometime after 1242. In the present article I have suggested that the kanjō no maki is a later addition to the twelve maki of the Shibu. My reasons are the complete dissimilarity of writing styles and the use of the posthumous title Go-Toba-in. I therefore prefer to date the Shibu only on the basis of internal evidence in its twelve maki. As for the appearance of Norisuke's name in maki 11, he would have been approximately twenty-eight years old in 1220, and I see no reason to prohibit the use of his name in this way by Yukinaga in 1220.

62. Gen Chūnagon Garai no Kyō, a member of the Murakami branch of the Genji.

63. Ason, also read asomi. A title of respect attached to the surname of persons of the Third Rank and above, and to the personal name of persons of the Fourth Rank. The usage here would indicate that Yukitaka held the Fourth Rank at this time, and the use of his name is another instance of Yukinaga's attempt to weave his father into the Shibu narrative.

64. Saishō Nyūdō Shunken. The details of his life are unknown. In the Kakuichi text Shunken appears as Nariyori.

65. Atsumi discusses this dream section in these terms as it appears in the Kakuichi text. She mistakenly states, however, that the reference to the Fujiwara Shogun does not appear in the Shibu text, an error that throws off her entire consideration of dating the Heike on the basis of this section. She also mistakenly gives 1221 as the date of the last Genji Shogun Sanetomo's death (1219) and thus further confuses the issue (Atsumi, 33-35). The vocabulary of this dream section is quite similar to that of references to the appointment of the Fujiwara Shogun in Jien's Gukanshō, 305, 317.

66. The theoretical basis of Jien's bumbu kenkō plan is related in the Gukanshō appended furoku section, 287-329. See also Taga, 142ff. Jien's ultimate objective was a unification of civil and military under a sesshō form of government (Taga, 149).

67. Yashiro 5.11b8-13a6. The Kakuichi account, written in 1371, reintroduces in the dream section the part about the sword passing to the Fujiwara, apparently because it makes a better story and because in 1371 there was no longer any fear of reprisal by the Hōjō (KBT 32, 342-343).

68. Quoted in Gotō, Senki monogatari no kenkyū, 101. See also 526-527.

69. Tomikura Tokujiro, “Heike monogatari no seiritsu ni tsuite,” Kokubungaku (November, 1956).13.

70. Gotō, 101.

71. Quoted in Takagi, 54.

72. Atsumi, 49, 15, mentions these other works.

73. Tomikura, “Heike monogatari no seiritsu,” 9.

74. The relevant passage of this work is quoted in Takagi, 57-62. Atsumi discusses it in Kisoteki kenkyū, 47-49.

75. See Yamada Yoshio, Heike monogatari (Tokyo: Hōbunkan, 1933), 41-45, for a concise statement of this view, which has been followed by later scholars.

76. Yamada Yoshio, “Heike monogatari kō zokusetsu,” 63. I have not translated the line about the year-period change. The title Heihanki is also read Hyōbanki.

77. These notations are quoted in Takagi, 51, and Atsumi, 41. The Gempei wars began in 1180, the last year of the Jishō period (1177-1180), hence the name Jishō monogatari.

78. Quoted in Takagi, 59.

79. There has been a tendency to assume that if the Jishō monogatari was in a six-maki form, the narrative of the later twelve-maki texts must be twice as long as this (Atsumi, 44). That such an assumption is not necessarily valid is easily shown by the existence of a later expanded text in the read division, the Enkyō text, which is divided into six divisions, but runs to 1007 pages of fine print as compared with the 733 pages of the standard printed twelve-maki version (Ōei shosha Enkyōbon Heike monogatari [Tokyo: Hakuteisha, 1961]). Scholars feel that parts of the narrative of the Enkyō text are close to that of the original Heike. The organization of the Enkyō into six divisions (ichi, ni, san, etc., the term maki is not used) probably also reflects the organization of the original Heike text. See the photographically reproduced text of the Enkyō: Enkyōbon Heike monogatari (Tokyo: Koten kenkyūkai, 1964). It is interesting to note that the six divisions of the Enkyō text are further subdivided into twelve sections as follows: dai-ichi hon; dai-ichi sue; dai-ni hon; dai-ni chū; dai-ni sue; dai-san hon; dai-san sue; dai-shi; dai-go hon; dai-go sue; dai-roku hon; dai-roku sue. In copying a text such as this, it would be a simple matter to revise these twelve sections into twelve maki, and it was probably through a process similar to this that the Shibu original Heike monogatari text was changed from six to twelve maki. The divisions of the twelve maki of the Shibu correspond roughly to the twelve divisions of the Enkyō. See the comparative charts given in Atsumi, 378-427 and especially p. 406, which shows that the beginning of maki 7 of the Shibu corresponds exactly with the beginning of dai-san sue of the Enkyō and maki 28 of the Gempei jōsuiki, the two texts of the read division which are thought to preserve some elements of the structural division of the original Heike.