A History of Japan: From Mythology to Nationhood/The Kofun Period

The Yamato State: Introduction

Yamato is a name with several meanings. It is an ordinary Japanese place name 山門 meaning literally "mountain gate" and more generally "in the mountains" or "surrounded by mountains," and there 5 places named Yamato that I know of and probably more. In the system of provinces that was established near the end of the 7th century AD the name Yamato was assigned to the province containing the national capital, corresponding approximately to modern Nara Prefecture. In the Imperial Japanese Navy before 1945 the principal warships were named after the old provinces, and the famous battleship Yamato was named after Yamato province. From ancient times Yamato has also been used as a name for Japan as a whole. Nihon or Nippon, the word behind the English Japan, was officially adopted as the name of the country in the early 8th century, and use of Yamato to stand for Japan has been essentially poetic since that time. However, it has become a convenience to historians to use the name Yamato when referring to Japan as a political entity in the period before the adoption of Nihon and especially before the "Taika Reform" of 645.

When the Chinese first became aware of the islands of Japan they assigned the people living there the ethnic name Wa 倭. This character is not very polite, meaning dwarf, and when the Japanese learned to read Chinese they complained about it, and the Chinese agreed to change it to 和 which is pronounced the same but has a dictionary meaning of "peace" or "harmony." The Japanese took to referring to themselves in communications to the Chinese as 大和 or "great Wa", and at some point started to pronounce this as Yamato. These are the characters used for the name of Yamato province. We do not know exactly when this happened.

In the chapter on Yayoi culture there is a discussion of the names Yamatai and Yamato in the context of the beginnings of the historical Japanese state, and I will now expand on this slightly.

When the Chinese engaged in diplomatic contact with the Wa people in the period between 240 and 266 AD they recorded that the "king of the Wa" (who happened to be female) lived at a place called Yamatai. This is so similar to Yamato that it is difficult to believe that it is a coincidence. The Chinese records read normally (meaning without expectations about what it should say because of outside knowledge you have or think you have) state that Yamatai was somewhere in northern Kyushu. Many Japanese historians over the ages have argued that that must have been a mistake because of the later usage of Yamato as the name for the Nara plain and that the "king" must have lived in Yamato province. However, it happens that there are two places named Yamato in the exact area that is the most likely spot for a Kyushu Yamatai, not far north of the modern city of Kumamoto. They are so close to one another that they are likely remnants of a name that applied to a much larger area.

In the Yayoi chapter of this book, I have given an argument for supposing that when the center of government in Japan was moved from Kyushu to central Honshu the name Yamatai/Yamato was taken along. This would have occurred, in my view, sometime not very long after 266 but before 300 when the first great "kofun" tombs were constructed in what was later to become Yamato province. This is the point in time when it becomes meaningful to talk about the "Yamato State."

The history of Japan in this period rests on just one book, Nihon Shoki, which was published in 720. It is a frustrating book in many ways because it is a combination of factual and bogus material and there have been endless arguments about which parts belong to which category. It has been subjected to close analysis since the 17th century and the total output on the topic in Japanese is enormous. Very little of this has been rendered into English. People who write on this period in English prefer to accept what the Japanese have accomplished in this area and get on with attempting to describe ancient Japanese society. I will briefly summarize the contemporary understanding of the book, and then do the same. There are a few additional materials of significance.

There are two ancient swords with inscriptions on them. There is a stone stele with an inscription of 1775 characters that was erected in 414 in the capital of the Kingdom of Koguryo that has some things to say about Wa. There is a letter from a Yamato ruler quoted in one of the Chinese dynastic histories, dated 478. And there is Kojiki, a book that is even older than Nihon Shoki, having been published in 712, but which has mainly mythological and genealogical information and almost no history. It does have some information about dates and the lengths of reigns which is mainly useful for engendering caution about what Nihon Shoki says about these things. The two books almost never agree about anything.

The situation improves a bit for the period after 572, which is approximately the time that the Japanese upper class began to read and write Chinese. There are no surviving books that are older than Kojiki and Nihon Shoki, but there are some later books that preserve old traditions, mostly about the early history of Buddhism in Japan to start with, but gradually expanding to additional topics.

Nihon Shoki can be translated as "History of Japan." This book title is the first recorded use of Nihon as the name for Japan. There is a note in the history of the Chinese T'ang dynasty that says that the Japanese made up this name because they became unhappy with Wa. It can be translated as "where the sun rises," which makes sense from a Chinese perspective because Japan was east of any other country they knew of. No one can be entirely certain how the Chinese would have pronounced this in the 8th century, but today in standard Chinese ("Mandarin") it sounds like "zhrr pun" which is close enough to Japan. This Chinese rendering was passed on to Europe via Marco Polo. His Cipangu (pronounced chipangu) stands for "zhrr pun guo" or "the country of Japan."

Nihon Shoki was a government project, entrusted to the highest officials, who quite certainly passed off most of the work to subordinates, many of whom would have been Korean ethnically, as Koreans tended to dominate jobs involving reading and writing in this era. It was from the beginning a work of propaganda more than a work of history. Its mission was to explain how Japan got to be where it was in such a way as to enhance the authority of the ruling house. It sought to combine two threads, a native thread and a Chinese thread.

The Chinese thread derived from the fact that the Chinese had been thinking about and writing about what constituted good government for centuries. It was desirable to show that Japan had become a well governed state in a way that would be approved by the Chinese. The Chinese way also had some inherent virtues from the Japanese point of view, because it emphasized the necessity to have a unified society where all important decisions are made at the center, in the court of the emperor. The emperor is an autocrat whose authority is limited by no constitution, only the certainly that if he fails to deliver good government, he runs the risk of being overthrown, opening the way to someone who can do a better job.

The authors of Nihon Shoki liked the first part but not the second part. That is where the second, native, thread comes in. The Japanese ruling house was directly descended from a god who came down from heaven to the earth almost two million years earlier, the Japanese state was established by a member of that house in 660 BC, and the same house has ruled Japan from that day to the present in unbroken succession under the protection of the gods and will always reign indefinitely into the future.

Actually, the future part is correct, so far, in that the present emperor is a member of the family that was ruling when Nihon Shoki was written. This is a remarkable fact showing that Japan is not like any other country in the world. There were at least a dozen occasions when, if Japan were China, a strong-man would have ended the old dynasty and started a new one. Only two or three times in all of Japanese history have men been accused of even thinking about that. There is no "Mandate of Heaven" here that can be fumbled away by a bad ruler. A few bad rulers have been gotten rid of, but the family remains. The gods will not permit otherwise.

Naturally, the aristocratic clans of the day had their own traditions and memories and this new history of Japan had to fit what was known of the facts. However, the main purpose was always ideological rather than scientific and if it was necessary to distort or conceal the truth to make a point, they were happy to do it. Fortunately for us, they were amateurs at fabricating history and vulnerable to close analysis. They tell us a lot more than they intended to. Unfortunately for us, close textual analysis is a tricky business and people often make mistakes, and are not infrequently themselves more interested in making propaganda than in uncovering the truth.

There is a far from perfect but still readable translation of Nihon Shoki into English by William George Aston, published in 1896. I have the paperback edition published by Charles Tuttle in 1972. It is just over 800 pages altogether. It has 100 pages of information about the gods and the creation of the world, followed by individual articles about 41 successive rulers, from Jimmu Tenno in 660 BC to Jito Tenno who abdicated in 697 AD. There is one article on a person who is not officially a ruler, Jingu Kogo, and one ruler who is on the current official list of emperors, but who is not counted by Nihon Shoki. This is Kobun Tenno who was defeated by his uncle in a short civil war in 671/72. He was added to the official list in the 1880's, probably because his father, not his uncle, was the ancestor of the current imperial family.

For the purposes of this discussion the book can be divided into chunks.

2 books of creation myths
Account of the foundation of the state by Jimmu Tenno dated to 663/660 BC
8 rulers generally considered imaginary
Account of Sujin Tenno, thought to be historical and linked to the start of the kofun archaeological monuments
3 descendants of Sujin, perhaps historical, though there is much myth and little history in this part of the text
Chuai Tenno and Jingu Kogo, probably mythical characters (Jingu = Himiko of Yamatai)
Ojin Tenno, probable founder of a dynasty, possible horse-riding foreign invader
6 historical descendants of Ojin
4 more troublesome descendants of Ojin, possibly covering up the collapse of the dynasty
Keitai Tenno, probable founder of a new dynasty, the one that continues today
3 sons of Keitai whose accounts seem confused about some things
More or less reliable history based on government records, 572 - 697

A fair bit has been written in English about the Japanese creation myths. I do not pretend to any expertise on this topic and I am going to avoid it. However, I would strongly recommend that anyone who wishes to study them should also read the corresponding myths from Koguryo, Paekche, and Silla, about which there is available information in English. This is important because the family resemblance among them is striking and means that whatever theories one wishes to erect about the formation of Japanese or Korean society require an explanation as to why their foundation myths are so much alike. There is a tendency in Japan to write about Japan as if it has no connection with any other place, especially Korea. However, these myths alone prove that Japan is intimately connected with Korea.

If you have read the archaeological chapters, you will already know about the theory that the historical states of Koguryo, Paekche, Silla, and Yamato all emerged from a shared historical event, the migration into the area of bands of warriors from a tribal nation the Koreans call Puyo in the era after 300 AD. This hypothetical event will be discussed below, but it should be noted that it is strongly supported by the archaeological evidence and by the mythological material in Nihon Shoki and Kojiki plus Samguk Sagi and related Korean texts. If the authors of Nihon Shoki knew or suspected that it happened, their intent was to cover it up.

An important component of Nihon Shoki is material, some of it quite lengthy, which is evidently taken from other books which were either written in Paekche or by Paekche immigrants in Japan. At least 3 different books of this kind are given names in footnotes, and other portions not given footnotes are assumed to be of the same character. For some rulers the Paekche material is much more voluminous than the Japanese material. This is important for two reasons. It gives information about ancient Korea that is not in Samguk Sagi or later Korean works, and it permits linking sections of Nihon Shoki to known Korean dates. This is vital because the dates in the earlier parts of Nihon Shoki were deliberately falsified to make Japanese history seem longer than it really was.

According to Nihon Shoki the Yamato government was founded in 660 BC. It was recognized as early as the 19th century that the authors of Nihon Shoki arbitrarily chose that founding date for astrological reasons and then had to find some way to fill that vast period of time with material, some of which they simply made up. The Chinese used then (and for some purposes use now) a 60 year cycle for routine dating outside of the formal dating by reign titles. This cycle includes the familiar 12 zodiacal signs (Year of the Ox and so on) that are combined with additional elements to get 60 years with individual names. The year 601 AD was the first year of such a cycle and 660 BC is exactly 1,260 years earlier. In the time Nihon Shoki was written the Chinese considered this amount of time, 21 60 year cycles, to be a particularly significant block of time, an era. 601 AD was an important year for the authors of Nihon Shoki because it occurred during the reign of Suiko Tenno, which they considered to be the beginning of the new era of "modern" Japan, so it was appropriate to begin "ancient" Japan 1260 years earlier.

Early in the 20th century a historian named Tsuda Sokichi proved pretty conclusively that the eight emperors listed after the founder, Jimmu Tenno, were completely imaginary, having been invented to to fill up some of the gap. Their names were constructed of elements fashionable in the 8th century but unknown earlier, and their articles are missing standard elements that are present in all of the other articles. This is common enough in East Asia. China and all three Korean kingdoms also have imaginary early rulers intended to provide an impressive antiquity. With Jimmu Tenno, the first nine emperors are assigned 562 years. However, this still only gets us to 98 BC with a long way to go. The next 7 emperors are made to cover 497 years. Some of these men are considered to be historical, but their dates are obviously false.

Help arrives with the inserted Korean material. There are several entries where it can be easily seen that the Nihon Shoki date is exactly 120 years older than the real date according to Samguk Sagi or Chinese sources. This is two 60 year cycles, and the dates in the Korean material are all cycle dates, so this means that they did not have to change the text, just assign it to the wrong cycle. This kicks in with an event with the actual date of 371 (dated by Nihon Shoki to 251) and runs into the first part of the 5th century. Then there is a section without Korean notices and then from about 450 the Nihon Shoki dates and the Korean dates are the same. This means that sometime between 371 and 450 the authors of Nihon Shoki found ways to smuggle in 120 extra years. Extremely long reigns and extremely old rulers are the visible signs of this. Even without considering problems with things that the text says, it is clear that trying to use Nihon Shoki as a real historical source for this era is essentially impossible.

The main solution to this is to avoid, by and large, attempting to use Nihon Shoki to create a chronological account of events before 572. The chances are that for much of this era the authors of Nihon shoki did not have the necessary information to give correct dates even if they had wanted to do so. Based on the way things have worked elsewhere it is quite likely that they had anecdotal material of the kind that begins "in the time of ruler so-and-so ... " that never had anything like a hard date attached to it.

If Nihon Shoki can't give us history, what can it do for us? It offers a narrative that can be linked loosely to the archaeology. With the events linked to Korean dates we are already in a timeframe associated with Ojin Tenno. Most historians will agree that it is probable that an earlier ruler, Sujin Tenno, was a historical person. It is a serious question whether he was the first ruler of Yamato or whether there were unknown predecessors. According to the Imperial Household Agency Sujin was buried in a tomb that has already been referenced in the previous chapter. It is the largest of 3 ruler-sized tombs of the "Early Kofun phase 1" period and is dated to approximately 300. This would be about 100 years before Ojin. There are many historians who believe that the Yamato state is older than this, which is why some of them seek to appropriate Himiko and Iyo and the unnamed male rulers mentioned by the Chinese to carry it back in time. However, the archaeology of the Yamato region does not offer much hope for a powerful state significantly older than Early Kofun 1. The Yayoi sites in the region are characterized by a conspicuous lack of social stratification before about 190, and the late Yayoi elite tombs dated after 190 that are within the same Makimuku group as the Early Kofun 1 tombs are not impressive when compared with what was happening at the time in Kyushu.

Nihon Shoki says that the Yamato state was founded by a conqueror from Kyushu. No one believes in Jimmu Tenno, but what if this conqueror was Sujin Tenno? This is exactly what Egami Namio has suggested. The Chinese say that on two separate occasions the Wa elevated a male ruler but he could not get the people to obey him and the solution was to replace him with a female ruler. This surely indicates a major internal disagreement about the proper way to organize society. What if the losers of the struggle after the death of Himiko decided to leave Kyushu and set up the kind of society they wanted in central Honshu? This would provide an excellent explanation of how it could be that the Chinese found a kingdom called Yamatai in Kyushu and a generation later we find one called Yamato in the Nara plain. They took the name with them. Such a move would be attractive for another reason, also. From this time onward the Kinai was always the center of Japan because it had the largest concentration of good farmland and became the wealthiest region, the one best suited to support a large aristocracy. Communications in Japan are always difficult, but this is definitely the most central location from that point of view as well. It is inconceivable that a ruler in Kyushu could have ruled all of Japan in the conditions that prevailed in ancient or medieval times. However, it was just barely possible for a ruler in the Kinai to manage this under favorable circumstances that frequently failed to materialize. Kyushu was often essentially independent and so was the far north east.

If Sujin Tenno founded a new state as a conqueror, what did this state look like? Nihon Shoki says that he spent the first part of his reign in an effort to get properly connected to the gods of the region. This material is discussed at length in an interesting article in volume 1 of the Cambridge History of Japan, which I recommend. It is clear that even in a state dominated by a military leader, one of the primary functions of the ruler was to serve as the intermediary between ordinary people and the gods, and if he could not get this correct, people would not obey him. Once he was satisfied that he was on the right track in that area, he organized four armies that were supposed to go out in four directions from Yamato and "subdue barbarians." Since we are at this point a long way from any people who can be considered "non-Japanese," the only thing that can be meant by barbarians is Japanese who have not yet submitted to Sujin's rule.

Before these expeditions could get started, however, there was a major crisis in the form of a "rebellion" by a person who is identified as a half-brother of Sujin. This led to a pitched battle between armies. Archaeologists have been struck by the fact that the site of the battle was in the southern part of Yamashiro province in exactly the area occupied by the third ruler-class tomb of Early Kofun 1. This tomb does not (traditionally) have a Yamato ruler in it. The most logical suggestion is that Sujin was not the original leader of the invasion from Kyushu and that there was a struggle over control of the new state. That would help explain Hashihaka, the third large tomb, which is 20 years older than Sujin's tomb. Nihon Shoki says that this tomb holds Sujin's aunt, who could forsee the future and who was married to a god. One must consider it possible then, that in the original conquest there was still a female ruler and it only shifted to a male ruler when she died. However, it is wise not to put too much trust in anything said by Nihon Shoki. It says that she committed suicide by stabbing herself with a chopstick, which is why the tomb is called Hashihaka, "hashi" being the Japanese name for chopsticks. Unfortunately, hashi were not adopted by the Japanese until around the time Nihon Shoki was written. Before then the Japanese ate with their fingers, as pointed out by the Chinese who visited Yamatai.

We can make some observations about this early Yamato state from Early Kofun 1 archaeology. According to Gina Barnes (2007) there are 98 known tombs from this period in Japan. They are found in three major concentrations. The three large tombs already mentioned and 8 "medium" tombs are in or near Yamato; there are 11 medium tombs in four separate groupings centered on ancient Kibi (Okayama), and there are 7 medium tombs in Kyushu, all of them on the east or north coasts. There are only three medium tombs in all the rest of Japan. One of them is in Hyogo prefecture, half way between Yamato and Kibi, one is north on the Japan Sea coast in future Etchu province, and one is in the mountains along the future Tosando near Nagano. There are scatterings of "small" tombs at the head of Tokyo Bay and in two places on the Inland Sea side of Shikoku. There is not even one small tomb in the interior of Kyushu or in the historically and archaeologically important area of Izumo on the Japan Sea coast north of Kibi.

In the Nihon Shoki invasion story (of Jimmu Tenno) the force stopped for two years in Kibi before continuing on to invade the Kinai (the general Japanese term for the area around modern Kyoto, Nara, and Osaka). It is well known that the Kibi region was always of great importance throughout the time of the Yamato state. If the invaders came from Kyushu, they must have had some base there to start from. There are no Early Kofun 1 tombs in Hyuga province (where Jimmu was supposed to have started from), though there is a cluster of Yayoi tombs. This region was part of a separate "southern" variety of Yayoi and presumably not connected with Yamatai and Himiko. The seven Early Kofun 1 tombs are in the "countries" of Na and Ito and also on the north east coast near Shimonoseki, which is where Yamato armies going to Kyushu always landed in later times. There are no Early Kofun 1 tombs in the inland area known as Tsukushi, which was the agricultural heart of Kyushu and the presumed location of Yamatai.

Nihon Shoki says that two of Sujin's successors led armies into Kyushu to battle the natives and both times they were met by female rulers. The campaigns always went into the heart of Tsukushi. It would appear that after getting control of the Kinai, they went back and took control of key spots on the coast of Kyushu, unless they already controlled those as being the bases from which they launched their attack on the east. It must be considered possible that the conflicts cited by the Chinese between male and female rulers were also conflicts between the north coastal "countries" and the interior. Anyway, it is important to note that this control of the coast of Kyushu explains how it happened that the Early Kofun 1 tombs nearly all contain high quality, recently imported Chinese bronze mirrors. The archaeologists believe that it was the ability to pass out these mirrors (and other elite goods) to cooperative locals rather than raw military power that explains the ability of the Yamato ruler to persuade many people to recognize him as the "king of the Wa." The Yamato presence on the Japan Sea north of Yamato gave them access to a source of high quality jasper that was used for burial objects also.

This is all highly speculative, of course, but it does bring about a certain attractive agreement between the essentials of the story as seen in Nihon Shoki and the archaeology.

Nihon Shoki says that Sujin's successor Suinin Tenno also had to face a major rebellion at the start of his reign. The archaeologists point out that the final battle, in which the rebel's "castle" was captured and burned, was in the exact district in the northwestern corner of the Nara plain which contains Suinin's tomb. They speculate that it was constructed there as a highly visible sign of dominance. It is an Early Kofun phase 2 tomb, by the way, which is characterized, among other things, by the disappearance of Chinese mirrors and their replacement by local imitations. In 291 the Chin dynasty in China fell into confusion in "The Rebellion of the 8 Princes" which led to its collapse and the overrunning of northern China by barbarian invaders. Supplying the Wa with goodies was not on the priority list after that time.

Suinin's tomb is actually inside the boundary of the later capital city of Heijokyo, alias Nara. The remaining tombs of this dynasty are near by, just beyond the northern boundary of Heijokyo. Unfortunately the rest of the material about Suinin is mythological in character. The next ruler was Keiko Tenno. He is represented as taking an army to Kyushu to put down "rebellion." He spent six years at a temporary capital in Kyushu before returning, whereupon Kyushu immediately "rebelled" again. This time he sent his son, at which point the article shifts into fully mythical mode as this son is the celebrated Yamato Takero no Mikoto, who may be termed the Herakles of Japan. The article on the next ruler, Seimu Tenno, is so short as to be logically non-existent. Then comes Chuai Tenno whom many think is invented. His article is particularly unsatisfactory and almost as short as Seimu's.

Chuai Tenno represents the point in Nihon Shoki where its authors intended to take notice of Himiko. Most historians, and all historians who subscribe to the "horse rider" theory, believe that the authors of Nihon Shoki also found it necessary to deal with the fact that the dynasty founded by Sujin now came to an end and was replaced by a new dynasty founded by Ojin.

In Nihon Shoki Chuai is the father of Ojin, but the relationship has its unusual aspects, starting with the fact that Ojin was born a full year after Chuai died thanks to the magical intervention of his mother, who was too busy to give birth at the natural point in time. Chuai Tenno found it necessary to take an army to Kyushu to deal with rebellion. Many of the details of the account of his journey match the campaign of Keiko exactly, which has raised suspicions of copying. At any rate, when he reached Kyushu he found a collection of small states with female rulers.

He had taken his empress with him and when they set up a temporary capital from which to run the campaign, she told him that she had had a vision from a god who declared that the rebels in Kyushu were too unimportant to bother with, that there was something much more important to do, specifically, conquer Korea. Chuai scoffed at this, climbing a nearby hill and looking out to sea and saying that he didn't see anything out there to conquer. A few months later he suddenly died. Nihon Shoki says in a footnote that some say he died in battle fighting against the "barbarians" of Kyushu.

His empress is known as Jingu Kogo where Kogo is "empress" in the sense of an emperor's wife, not in the sense of a ruler. There were ruling "empresses" in Japan, but their title was Tenno, exactly the same as a male ruler. "Gender" in the grammatical sense does not exist in either Chinese or Japanese so there is no problem with that usage. Jingu Kogo now took steps to delay the birth of Ojin, transformed her appearance into that of a man so she could lead the army, and invaded Korea, which she conquered (Paekche, Silla, and Koguryo, whereas neither Paekche or Silla had yet come into existence) in the space of a few months, not by combat but by awing everyone into submission. Then she returned to Kyushu, allowed Ojin to be born, and moved off to Yamato, where she had to fight her way in against sons of Chuai. She ruled to an advanced age without taking the throne and died in 270 (in Nihon Shoki time). There are four items at dates from 240 to 269 in her reign where the Chinese article on the Wa is quoted and named in Nihon Shoki. Ojin is portrayed as actively involved in affairs in her later years as the "crown prince."

This story serves two purposes, one major and one minor. The minor purpose is providing a reason why the Chinese might think that the capital of Japan was in Kyushu and why they thought that Japan was ruled by a woman. The name Himiko is never mentioned. The major purpose was to explain how it happened that an army originating in Korea and bearing the future ruler Ojin invaded and conquered Yamato without disrupting the direct succession of the Japanese ruling house from the gods. Their answer is using Chuai to make Ojin a descendant of Sujin. The proponents of the horse rider theory claim that there is no possible reason for all of this apparatus were it not necessary to explain away a successful invasion from Korea. Historians who do not accept the invasion theory simply dismiss the entire story as myth, freeing them of the need to explain it.

I don't know of any historians who do not agree that, wherever he came from, Ojin was the founder of a new dynasty. His tomb is in a completely new location, outside of Yamato, on the outskirts of modern Osaka. It is considered to represent the absolute beginning of Middle Kofun phase 6 (the ten phases of the Kofun are numbered consecutively, not restarting between Early and Middle). This is the type of tomb that in the view of the horse rider proponents proves that a completely new, very Korean, aristocratic culture was brought into Japan at this time. Ojin, his son Nintoku, and his grandson Richu are buried in the three biggest Kofun ever constructed, three of the most spectacular tombs ever constructed anywhere. It appears that only the tomb of Ch'in Shih Huang Ti in China and the pyramid of Cheops in Egypt cover a greater surface area than the tomb of Ojin (Nintoku's tomb is longer but narrower). Nihon Shoki supports the horse rider theory indirectly in the fact that the first mention of horses in the book occurs during Ojin's article, where it is recorded that the King of Paekche presented a pair to him. And, he obviously had vastly more power than any previous ruler in Japan, in order to have the resources to build that tomb.

Ojin founded a dynasty that was from beginning to end intensely involved with Korea. It ruled, in a loose sense, the parts of southern Korea that were not part of Paekche or Silla, called Imna by the Koreans and Mimana by the Japanese. It was an extremely close ally at all times of Paekche, and more than once sent troops to escort a Paekche prince living at the Japanese court to assume the throne back home. It was involved in frequent warfare against Silla and Koguryo, which were enemies of Paekche and allied with each other. The dynasty sent at least 13 embassies to southern China. Its internal politics were messy. The second ruler, Nintoku, gained the throne by murdering his elder brother, and most of his successors imitated this example. Yuryaku Tenno was such an efficient killer that there was, perhaps, no survivor of the clan other than his three sons when he died, and one of them promptly killed the other two. The exact end of the dynasty is obscure, but it seems pretty clear that Keitai Tenno represented a new one, that followed quite different policies in respect to Korea and China. The survival rate among princes went up enormously after this change.

Keitai Tenno was supposedly a 6th generation descendant of Ojin, but he was an outsider from the north who had never been involved in politics before being tapped as ruler. He was already a grown man with two adult sons. He immediately married the widow of a ruler from the Ojin dynasty and soon had a third son. There seems to have been a problem about the succession when he died, and Nihon Shoki and Kojiki disagree conspicuously about dates at this time. Officially the two older sons briefly ruled one after the other and then the third son took over. Many (but not all) historians think that the third son was the successor and that disgruntled nobles later set up the two elder sons at a rival court. This is why I put the beginning of chronologically valid history at 572 rather than a generation earlier.

The new dynasty presided over the gradual loss of the Japanese position in Korea to Silla. Keitai Tenno himself sent a large army to Korea early in his reign. It first ran into a major rebellion in Kyushu ("The Iwai Rebellion") but continued on to spend several years in Korea with no concrete accomplishments to show for it. Almost the entire Nihon Shoki article about Kimmei Tenno, the fourth ruler of this group, is actually Korean and is about the (unsuccessful) efforts of King Song of Paekche to persuade Kimmei to do something about Silla's aggression in Imna. Eventually Silla conquered Mimana and Japan was excluded from Korea. The court continued to plan attacks on Silla from time to time for another hundred and fifty years, but for one reason or another none of them actually left the country.

Korean Sources related to Nihon Shoki

As stated above Nihon Shoki is known to contain material that was originally Korean. It mentions the titles of three books in footnotes. They are Kudaraki 百済記, which covers events from the time of Paekche King Kunch'ogo through the death of King Kaeru (475), Kudara Shinsen 百済新撰, which covers events from the accession of Paekche King Piyu (427) to the accession of King Munyong (501), and Kudara Hongi 百済本記, which starts in the middle of the reign of Munyong and continues to the accession of King Widok (554). Kudara is the ancient Japanese name for Paekche. No one knows where it came from. There is an interesting Internet article on this topic by an Italian professor, in imperfect but readable English. See Anselmo, Valerio in the bibliography.

The oldest of these entries in Nihon Shoki date to the "reign" of Jingu Kogo and there are also a lot of them assigned to Ojin Tenno. Then there are few except for notices of events like the death of Paekche kings. However, the greater part of the information associated with Keitai Tenno and Kimmei Tenno is Korean, including some quite lengthy sections. Then it ceases. I will include in this section two additional items of some importance, the inscription on the sword in the possession of the Isonokami Shrine and the Koguryo stele commemorating the career of King Kwanggae'to.

There were 6 Paekche kings in the 6th century, and according to Shinsen Shojiroku, an ancient Japanese book on genealogy, there were noble families in Japan descended from 5 of them. A considerable number of other people from Paekche emigrated to Japan during the 300 years in which Paekche and Japan were close, nobles and commoners both. A number of Korean families other than the royal relatives were also admitted to the Japanese aristocracy. Shinsen Shojiroku counts 1065 aristocratic families and 324 were Korean, though they were nearly all given relatively low status within the aristocracy. Most of them worked in the lower levels of the bureaucracy, and some of them were involved in writing Nihon Shoki. It is not known whether the Paekche historical materials were written especially to provide sources for Nihon Shoki or whether they had had an independent existence, nor is it known whether they were written in Korea or in Japan.

The first block of materials is dated between 366 and 375 (real dates, the Nihon Shoki dates being 120 years earlier). It is assigned to the "reign" of Jingu Kogo, but the future ruler Ojin is frequently mentioned as the "crown prince." In the context of these dates the death of Jingu is assigned to 389 and the accession of Ojin as ruler to 390. The presentation of horses mentioned earlier is also noted in Kojiki as "during the reign of Ojin King Kunch'ogo of Paekche presented two horses." Kunch'ogo died in 375.

The texts in Nihon Shoki are more than a little obscure, but appear to describe the initial establishment of diplomatic relations between Paekche and Japan, which was close to the time of the first Paekche diplomatic mission to China. It is thought that Paekche came into existence some years earlier (it appears in a list of names in a Chinese source in 342), but that it had been a dependency of Koguryo, and was now asserting its independence. In 369 King Kogugwon of Kogoryo launched an attack on the Paekche capital near modern Seoul. This was defeated and a Paekche army commanded by the king's son (and eventual successor) Prince Kusu pursued. In 371 Kusu stormed the Koguryo military base at the former site of Lo-lang and Kogugwon died in the battle. Paekche was now a fully independent nation.

The Nihon Shoki material mentions nothing of this. It merely says that the first contact between Paekche and Wa occurred in 368, that a Wa army was on the frontier of Silla, then still a dependency of Koguryo, and that troops from Paekche joined them in capturing territory both from Silla and from "savages" in the area, and this cooperation was followed by a formal alliance. Several embassies went back and forth and one presented a "seven branched sword" and a "seven horned mirror" to the Japanese ruler. The Isonokami Shrine sword appears to match the first, and the Boston Museum of Fine Arts has an excellent example of the type of mirror that is meant, a mirror which it is thought was stolen from the tomb of Nintoku Tenno when it was repaired after a partial collapse in 1872. In the context of mentions of "barbarians" in Japan in Nihon Shoki, I think that the "savages" here were Han Koreans who had so far managed to escape rule by Puyo warriors.

The Isonokami Shrine was in ancient times associated with the Mononobe clan, whose hereditary responsibility was the manufacture of weaponry, and it was used as an arsenal for the storage of weapons. There is a record of this function as late as 804. The administrator of the shrine starting from 1873 was a well-known historian named Kan Masatomo. He had previously written a book about Chinese descriptions of the Wa. He wrote a book about the collections of the shrine that said nothing about the sword. However, he revealed its existence in a later publication.

The most crucial information about the sword is the date in the inscription. In Kan's original publication he read it as 泰始四年*月十*日丙午 (asterisks are characters that cannot be read), but he mentioned that the second character 始 is hard to read and represented a guess. In a later publication he gave the date as 泰初四年六月十一日丙午. For a long time the shrine refused to let other scholars examine the sword, but after WWII it was finally permitted and the consensus reading is now 泰和四年*月十六日丙午, or 4th year of 泰和, * month, 16th day 3rd year of the Horse in the Chinese year cycle. That gives 3 different readings of the second character of the "nengo," or Chinese style reign title used for the official dating of documents.

泰始 is the first nengo of the Western Chin dynasty of China and the fourth year is 268. The same nengo was also used by the Liu Sung dynasty, where the 4th year would be 468. 泰和 belongs to an impossibly late dynasty and would be 1204. 泰初 is unknown. The current theory is that 泰和 is an error for 太和, the reason offered being that they would be pronounced the same way, "taiwa" in Japanese. 太和 is a nengo of the Eastern Chin and the 4th year is 369. Everyone likes the 369 date because the inscription on the other side of the sword names Prince Kusu of Paekche, who was alive at that time. However, using 369 depends on deciding to assume that the actual date written on the sword is a mistake (which many discussions omit mentioning, jumping to the good part which is the attractive year 369).

No one appears to pay any attention to the cycle date, 丙午, which last occurred in 1966. Its only occurrence in the 4th century was 346. That was a time of deep confusion in northern China with many short-lived regimes. It would not surprise me to find that nengo were proclaimed that did not make it into surviving records.

However, I do trust the naming of Prince Kusu on the sword. It can be inferred from Samguk Sagi that even though his father was still on the throne, he was in control in Paekche as of 371 and 372. Conventionally one would expect such an inscription to mention only the king, but in this case the king was effectively retired and his son was in charge.

Anyone who reads W. G. Aston's translation of Nihon Shoki will see that before the Isonokami sword became known no one had any idea of what a "seven branched" or "seven bladed" or "seven pronged" sword might have looked like. Aston admitted in a footnote that he didn't understand this term at all. A hypothetical forger had a free hand to come up with anything. There are people who believe that the sword is a forgery. However, a sword has been excavated from a kofun at Kisawa in Tochigi prefecture that is, if not exactly identical, clearly made on the same pattern. The sword itself proves nothing. It is entirely logical that "Wa" and Paekche would be in diplomatic contact by 371. However, if the sword is authentic, then it proves that by 371 people in Paekche were capable of composing an inscription that named a prince of Paekche. No one in China knew anything about Prince Kusu.

After the date the inscription says "A seven bladed sword made of multiply forged iron able to subdue a myriad of soldiers. It is suitable for a duke or a king. Made by [4 illegible characters]." On the other side it says "This sword of a character heretofore unknown was made by The King of Paekche and his heir Prince Kusu for the King of Wa to be passed on to later generations." It appears certain that these two inscriptions were made at different times because their characters look quite different. The first one looks Chinese and the second does not, being much clumsier, as if engraved by someone who was copying characters he could not read.

In 372, the year of the Nihon Shoki record of the presentation of a sword, Paekche sent an embassy to the Eastern Chin dynasty of south China and the King was granted the Chinese military title of Chinto Shogun (Japanese pronunciation) and Governor of Lo-lang. This would be after his capture of the former Lo-lang from Koguryo (though Paekche did not attempt to hold it, setting the frontier further south). This would mean, if we trust the record in Nihon Shoki, that the sword was presented before the return of the first Paekche embassy to China (at the end of the year), so that it is impossible that the Chinese presented the sword to Paekche on this occasion. They got it some other way, perhaps as loot when they captured Lo-lang, in which case it could have been quite old, and it could have been made in Kogoryo or in Manchuria, and the nengo was not necessarily Chinese at all.

Officially the sword is authentic. It is a registered National Treasure. I have some good pictures of it (both sides plus a close up of the date) in an art book. The shrine does not permit it to be seen, but there are replicas in several museums in Japan, including the National Museum in Nara, and one in South Korea, and it is easy to find pictures on the Internet.

There is one more source to consider. In 414 the King of Koguryo erected an inscribed stone stele at the Koguryo capital on the Yalu River (called "Northern Capital"). The stone celebrates the career of King Kwanggae’to (391-412), who revived the fortunes of Koguryo after the disaster of 371. It describes all of the king’s wars, including those against Paekche. In it the Wa are mentioned several times. Kwanggae’to claims to have captured the Paekche capital and forced Paekche to submit to him in 396, only to have the Wa interfere and undo everything, and he also claims to have defeated a major Wa attempt to conquer Silla in 400. Nihon Shoki knows nothing about any of these events. It does record a Paekche embassy to Wa in 397.

In places the inscription is hard or impossible to read. I here give a translation of those parts that are of interest for Japanese history which is based on a translation to Japanese by Enoki Kazuo.

Paekche and Silla were originally people subordinate to Koguryo and have since ancient times paid tribute. However, in 391 Wa crossed over the sea and defeated Paekche, [unreadable], and Silla and made them their vassals.

This appears to refer to the Japanese role in the death of King Chinsa and his replacement by King Asin as mentioned in Nihon Shoki, though not in Samguk Sagi. Nihon Shoki says that King Chinsa was "disrespectful" to Japan and that two officials were sent to remonstrate, but before they could reach the capital, the people overthrew the king and killed him. The Japanese then sent for a prince who was residing at the court in Japan who became King Asin. Samguk Sagi says only that King Chinsa died suddenly while on a hunting trip in the countryside. It appears that Paekche princes quite often resided in Japan. At least 3 future kings are known to have done so. Nihon Shoki tends to present them as "hostages", but it is equally likely that the Paekche kings wanted to keep them away from palace politics.

Then in 396 King Kwanggae’to personally led a fleet and defeated Paekche taking 54 fortresses [all named in the inscription] and reaching the Paekche capital. The villains refused to surrender and dared to come out and fight, so the King brilliantly crossed the Ari River and attacked four fortresses. The King of Paekche was sorely distressed and offered 1000 slaves and 1000 bolts of fine cloth and surrendered to King Kwanggae’to and swore to be the king’s servant from that time forward through eternity. The king accepted this and seized 58 fortresses and 700 villages and took 10 high officials including the King of Paekche’s brother as hostages and settled them at his capital.

In 398 the King sent an army and removed 300 people from [unreadable]. As a result, since then they have paid tribute.

In 399 Paekche broke the treaty and allied with Wa and took a hostile position, so the king moved to Pyongyang where he took up a defensive position. According to an embassy from Silla that came to him there, the Wa had invaded Silla, destroyed fortresses, and liberated slaves and made them free men and then allied with them. They requested the king to do something to help them. Kwanggae’to praised the loyalty of Silla and dispatched a commission to report on the situation there. Then in 400 he assembled an army of 50,000 men and sent it into Silla. Wa forces were spread over Silla all the way from 男居城 to the Silla capital. When the Koguryo army arrived the Wa villains retreated and the Koguryo army chased them to Kara in Imna and that fortress surrendered, however the men of Anra attacked the Silla capital and [unreadable] and Wa soldiers also retained many fortresses in that area.

Kara and Anra were the two principal states in the Mimana region that was under Japanese control in the 5th century. This sounds like an admission that the offensive against the Mimana and Japanese forces was not entirely successful. There is also no claim of a return of Paekche to obedience.

In 404 the Wa were so villainous as to cross the Koguryo frontier at T'ai-fang. [an unreadable section occurs] The king was ferociously attacked, but the Wa were defeated and countless numbers were slain.

In 407 an army of foot soldiers and cavalry amounting to 50,000 men was fought, and 10,000 suits of armor and countless other military supplies were captured, and also two fortresses were taken.

This section has many illegible parts, but the name of Pyongyang appears, so it is possible that this fighting was near there. It does not explicitly say that this fighting involved Wa.

The stele was erected in 414 at Kwanggae’to’s tomb by his son and successor. It is unquestionably a contemporary document, but it is royal propaganda and there are a lot of problems with it. Many things that it says are contradicted by other sources, most particularly Samguk Sagi, the oldest surviving Korean history of Korea, and Chinese diplomatic records, neither of which knows anything about the surrender of Paekche claimed for 396. Also, it does not match up well with the materials derived from Paekche that are quoted in Nihon Shoki. The stele is not mentioned in any historical source before its discovery in the Meiji period by a Japanese army officer who was surveying the frontier between Korea and Manchuria. There has been considerable work investigating whether it might have been a forgery, but this theory is, I believe, no longer entertained. Some restoration work was done during the period of Japanese control of Manchuria, but the inscription is not flattering to Japan and there is no reason to think that it was changed. There have recently been discovered some old rubbings of the inscription that were supposedly made by Chinese before the stele was "discovered" by the Japanese that support the authenticity of the text.

There are a number of reasons why many historians are not happy with the dates assigned to Ojin and his descendents, even when the early ones are adjusted by 120 years (most of which would come by shortening Ojin's and Nintoku's reigns considerably). The main problem is the information in the southern Chinese dynastic histories about Japanese embassies. This leads us to yet another perplexing topic that has attracted a great deal of attention, though not much in English.

The Five Kings of Wa

The records of the Liu Sung dynasty of 420-479 name 5 Wa Kings. They are 讃, 珍, 済, 興, and 武. The rulers’ names as given in Nihon Shoki are polysyllabic and there is no possibility that these Chinese words could represent them phonetically. This has not prevented dozens of scholars from trying to make them do it. Either they work from the meaning of the character, or they select one syllable of the Japanese name that sort of fits a character’s Chinese pronunciation. None of this seems productive.

The Chinese say that 珍 was the successor of 讃 and was his younger brother. This is encouraging, because succession by brothers was the norm in Japan at this time, although contrary to Chinese ways of thinking. They did not say how 済 was related to the first two, but do say that 興 and 武 were his sons. The Nihon Shoki genealogy for the period in question has three sons of Nintoku reigning in succession, Richu, Hanzei, and Ingyo, followed by two sons of Ingyo, Anko and Yuryaku. This actually fits. Unfortunately, the dates are badly out of synch with what Nihon Shoki and Kojiki have to say. And Nihon Shoki and Kojiki also usually disagree with each other. Here is a little table on that.

Nihon Shoki
Ojin	270-310	age 110	d. 394	age 140
Nintoku	313-399	not given	d. 427	age 83
Richu	400-405	age 70	d. 432	age 64
Hanzei	406-410	not given	d. 437	age 60
Ingyo	412-453	not given	d. 454	age 78
Anko	453-456	not given	d. ?	age 56
Yuryaku	456-479	not given	d. 489	age 124

The Kojiki dates are given as Chinese cycle dates, so that it is necessary to guess which cycle they are in. The table shows the most reasonable choices. They look better than the Nihon Shoki dates (which is not hard). Unfortunately, that doesn't make them right, and the ages are not exactly what we would hope to see. There is no reason to believe that these dates were in the underlying, probably oral, material because the Japanese were probably not using the Chinese calendar when it was created. They give every sign of having been inserted much later.

According to the Sung dynasty records, the king who sent the first embassy they received (there was one earlier to a previous dynasty, but it doesn't give a name for the king) in 421 had a successor who was his younger brother. This means that it must have been Richu, Hanzei, or Anko. However, according to Kojiki, Nintoku was still on the throne, and according to Nihon Shoki the ruler was already Ingyo. Somebody has to be wrong. There is more. The 3rd king sent two embassies, in 443 and 451. In 462 the Chinese report that he died (date unknown) and his "crown prince" sent an embassy to announce the fact. The next embassy was not until 478 and everyone enthusiastically endorses the idea that this was Yuryraku. However, that means that the 462 embassy was sent by Anko, which doesn't match any Japanese dates. The reason that Yuryaku is thought to be 武 is because one Japanese word linked to 武 is "takeru" and this was a part of the Japanese name of Yuryaku (remember that the name Yuryaku was only created in the 8th century and is not used in Nihon Shoki). This is the only one of the "kings" whose name works in this way. However, the Chinese of this era were unlikely to know "takeru" at all, or the fact that it means the same thing as the Chinese 武. The Chinese pronunciation of 武 is Wu.

The bottom line here is mainly that these Chinese entries prove that the Nihon Shoki dates are wrong, which means that everything that Nihon Shoki says has to be viewed with skepticism. And, the most important fact is that Nihon Shoki does not even know that any of these embassies ever occurred. This proves absolutely that Nihon Shoki was not working from anything like government records for this period but other kinds of materials. Nihon Shoki does mention two contacts with China in this era, rather mysterious ones. Ojin Tenno was supposed to have sent two agents to China in quest of expert seamstresses. Then there is mention of the return of such a party, with seamstresses, during the reign of Yuryaku, at least one hundred years later.

If you read the Nihon Shoki articles on these kings you will see that politics during this period were quite violent. The majority of these rulers reached the throne over one or two corpses. You will also notice that Nihon Shoki has a three year interregnum after Ojin and two years after Hanzei. That is because there were civil wars. Anko was assassinated by the son of a prince he had murdered to gain the throne, and Yuryaku killed the assassin. The violence was not limited to succession quarrels. There was a tendency to follow Stalin's principle "no person, no problem". If you had doubts about anyone, kill him. This raises the possibility that the official list that made it into Nihon Shoki is not necessarily complete. There may well have been men who functioned as ruler for a period of time but were written out of the record because they were ultimately killed by a rival. One or more persons of this sort could have easily sent an embassy to China, nicely messing up both the dates and the implied genealogy.

The end result is that we have to acknowledge that our information about this period is scanty and move on. I do not believe that making an effort to try work out who the 'five kings of Wa' might have been is a worthwhile activity. There is one benefit from the Chinese embassies, however, that is probably more valuable all by itself than everything that Nihon Shoki has to say about Ojin and his successors. This is the letter that accompanied the 478 embassy, presumably sent by Yuryaku. This tells us much about the 5th century Japanese rulers that Nihon Shoki knows nothing about, or at any rate doesn't tell us about. Here is the Chinese account.

Then 興 died and his brother 武 came to the throne. In the 2nd year of Sheng-ming during Shun-ti’s reign (478), 武, signing himself as King of Wa and General who Maintains Peace in the East commanding with battle-ax all military affairs in the seven countries of Wa, Paekche, Silla, Imna, Kaya, Chin-han, and Ma-han, sent an envoy bearing the following memorial:

Our land is remote and distant, it lies far out in the ocean. For generations our forbears have clad themselves in armor and helmet and gone across the mountains and rivers, sparing no time for rest. In the east they conquered 55 countries of hairy men, and in the west 66 countries of barbarians. Crossing the sea to the north, they subjugated 95 countries. The way of government is to keep harmony and peace, thus establishing order in the land. Generation after generation, without fail, our forbears have paid homage to the court. Your subject, ignorant though he is, is succeeding to the throne of his predecessors and is fervently devoted to your majesty. Everything he commands is at your disposal. Thinking to travel by way of Paekche, distant though it is, we prepared boats, but Koguryo criminally schemed to seize them. Borders were violated and many deaths have occurred. We have been delayed time after time and missed favorable winds. Every time the way seemed clear, Koguryo was rebellious. My deceased father became indignant at this marauding foe who blocked our path to the court. Urged by a sense of justice he gathered together a million archers and was about to launch a great campaign. But, because of the death of my father and my brother, the plan could not be carried out. Mourning required the laying down of arms. Inaction does not bring victory. Now, we have again set our armor in array and carry out the ambitions of our predecessors. The fighting men are ready, civil and military officials are ready, and none have fear of sword or fire.

Your majesty’s virtue extends over heaven and earth. If through it we can crush this foe and put an end to our troubles we shall ever continue our loyalty to the throne. I therefore beg you to appoint me as commander of this campaign with the status of minister, and to grant others ranks and titles so that loyalty may be encouraged.

By imperial edict, 武 was made King of Wa and General who Maintains Peace in the East commanding with battle-ax all military affairs in the six countries of Wa, Silla, Imna, Kaya, Chin-han, and Ma-han.

The Kwangae'to stele says nothing about this because it says nothing about anything that happened before Kwangae'to came to the throne of Koguryo in 391. However, here the "king of Wa" claims himself to be a bitter enemy of Koguryo, something which does not appear in Nihon Shoki, where the enemy is always Silla, which controlled all of Korea when Nihon Shoki was written and was hostile to Japan. It is worth noting that in 475 Koguryo seriously defeated Paekche, capturing its capital and killing King Kaero. The new king relocated the capital in a mountain location well to the south, though the boundary between Koguryo and Paekche remained approximately the same as before. King 武 didn't say anything about this, but it must have been a factor in the overall situation. Nihon shoki says that Japan assisted Paekche at this time by transferring some lands from Mimana to support the new capital.

Note should be taken of the list of countries claimed by King 武 and the list granted by the Chinese. King 武 claimed seven but was only awarded six. Several of the 5th century Japanese embassies asked the Chinese to name the King as in control of Paekche and the Chinese always refused. They also always awarded the King of Paekche a Chinese military title that was one step higher than the one given to the King of the Wa. However, the Chinese did not object to listing Silla. This was because Silla at this time had never sent an embassy to any southern Chinese dynasty. Also listed are Imna and Kaya, whereas Kaya was a country in Imna. And, there is Chin-han and Ma-han which were ancient names for the parts of southern Korea that eventually became Silla and Paekche. Clearly, the Chinese didn't have much knowledge of the political geography of the region (and they did not send embassies to either Korea or Japan so that they had little opportunity to find out).

It is difficult to find anything to say about the domestic side of the Ojin dynasty. Nihon Shoki offers nothing to suggest a structured government or any system for administering the countryside other than assigning hereditary rulers everywhere. In the article about Ojin it says that he assigned the area of Kibi (Okayama Prefecture) to a certain person, dividing it into five districts to be inherited by five of his children. This happened at the beginning of his reign when Kibi presumably needed a new ruler after the invasion. There are four noble clans who are most frequently mentioned in the time of the Ojin dynasty.

The Kazuraki clan provided several wives to rulers and also at least one military commander assigned to duty in Korea. It is not mentioned after the end of the dynasty.

The Wani clan had a territory in the Yamato area and did survive, although the name did not as it divided into several successor clans, some of which were still prominent in the 8th century.

The Otomo clan differed from the first two in that their name had a functional character instead of a geographical one. "Tomo" means follower in the sense that soldiers are followers, and the Otomo constantly appear as military commanders, presumably leading troops who were followers of the ruling clan, not men belonging to the Otomo clan itself.

And, the Mononobe clan also had a functional basis. One would expect Mononobe to refer to groups of commoners who specialized in making "things", specifically iron weapons. However the Mononobe clan were not themselves blacksmiths but rather charged with managing the workers who made weapons and also running the arsenals containing the ruler's weapons. They also often appear to function in situations where something like police would be used in our society. A striking case occurs at the time of the Iwai Rebellion. This occurred as the head of the Otomo clan was leading an army to Korea. He was ordered to ignore the rebellion and continue on to Korea, and the head of the Mononobe clan led a second force to Kyushu to suppress the rebellion.

The Collapse of the Ojin Dynasty

This topic has been discussed much by Japanese historians but has been pretty much ignored in English. According to Nihon Shoki Yuryraku Tenno was a particularly efficient killer of relatives who might have a claim upon the throne. He is credited with a bizarre death-bed speech, the only one in Nihon Shoki, in which he expressed regret that he was leaving the country in such bad shape. This speech is stolen almost word for word from one in Shih Chi related to the fall of the Ch'in dynasty of China. The 478 embassy attributed to Yuryaku by most Japanese historians is only one year before the date of his death according to Nihon Shoki. It was also the last recorded Japanese embassy. The Sung dynasty fell in 479 and was replaced by one called Ch'i which lasted for only 23 years, followed by Liang. The Liang history says that in 502 they renewed all of the titles granted to King 武. However, it mentions no embassy and this does not necessarily mean that he was still alive.

Yuryaku had three sons, and the customary conflict ensued upon his death, leaving the two oldest dead and the youngest on the throne. This is Seinei Tenno who was childless. Allegedly, during the reign of Seinei it was fortuitously discovered that there were two surviving princes, sons of a man killed by Yuryaku, who had been hiding in the countryside under false names. These two reigned one after the other and the second of them (who was the older) had one son who died childless ending the dynasty in the immediate sense. Otomo no Kanamura, the most powerful official at court, suggested a distant relative who might serve, but when a party was sent to contact him, he assumed that their purpose was to kill him and he ran off into the hills and was never seen again. So, Kanamura proposed a second candidate who showed more courage and listened to the message and became Keitai Tenno.

Nobody likes this material. The article about Yuryaku's son Seinei is very short and records essentially nothing beyond the discovery of the two princes. The story of Kenzo Tenno and Ninken Tenno reads like a fairy tale from beginning to end. The last ruler, Buretsu, is essentially Caligula, a classic "last emperor" used in Chinese historiography to justify the end of a dynasty as punishment for his moral failings. About all that Kojiki says about him is that he reigned 8 years, was childless, and it names his tomb. Its article about Kenzo says the same three things and the name of the tomb is the same in both cases and many think that Buretsu did not exist. It has also been suggested that if the "hidden princes" story is real they were not discovered while Seinei was still alive but after his death. Nihon Shoki makes it plain that a sister of Yuryaku led an interim regime that lasted for a year or so after the death of Seinei. In the Nihon Shoki version the delay occurred because the two princes could not agree on which of them should take the throne, each deferring to the other. Many scholars think that if they existed at all they were imposters, probably run by Otomo no Kanamura.

The overall feeling is that there was probably a somewhat long period of confusion when it might have been unclear whether there was a ruler at all before the emergence of Keitai. There is a note in Nihon Shoki that after the death of Ninken Tenno a noble named Heguri no Matori no Omi attempted to rule Japan as if he were king, but he was attacked and killed by the prince who became Buretsu Tenno. Many scholars are convinced that the story of the "discovery" and invitation of Keitai is also false, and that he was the hereditary leader of the Koshi region of Japan, which is to say modern Ishikawa Prefecture, and that he simply showed up one day with an army and took over. He ruled for 20 years before he first moved into Yamato province. He mainly ruled from Yamashiro, from locations that are inside the modern city of Kyoto (which did not yet exist). This is taken to suggest that it took a long time to secure acceptance from the main clans in Yamato so that he felt safe to set up his palace there.

The one thing that is certain is that the evidence is so sparse that the matter will never be resolved. However, most historians would agree that it is unlikely that Keitai was actually a direct descendant of the Ojin dynasty. It is said that his father was a fifth generation descendant of Ojin, which made Keitai sixth generation. According to the law in effect at the time that Nihon Shoki was written that would mean that he could not claim membership in the ruling clan, five generations being the cutoff point, with a fifth generation prince having strictly limited privileges. It may be that the point was that he was descended from the Puyo invading group.

He was new and the policies followed by his successors were noticeably different in important respects than those of the Ojin dynasty. Whereas the political roots of the Ojin dynasty were in Korea and it was always heavily involved in Korea, the subsequent dynasty for the next two hundred years consistently showed a special connection with the regions to the east and north east of Yamato. The rulers were always careful to keep considerable numbers of eastern warriors called toneri at court. These were apparently younger members of the rural clans who served a term at court and then went home again. Keitai's successors allowed the Japanese position in Korea to be dismantled by the Kingdom of Silla without putting up much resistance. They also sent no embassies to China. And, there soon occurred a considerable shakeup among the leading aristocratic clans.

The New Dynasty

At the start of Keitai's reign we find that the two most powerful men at court were Otomo no Kamamura and Mononobe no Arakabi. In 512 Paekche sent an embassy which proposed that four districts in the northwestern part of Mimana be transferred from Japan to Paekche. The reason offered was that they were close to Paekche and were in daily contact with Paekche, whereas they were far from the Japanese base on the Naktong River in the east of Mimana and had few dealings with that area. Otomo no Kanamura studied this application and recommended that it be accepted. Mononobe no Arakabi strongly opposed. The ruler's eldest son, the future Ankan Tenno, was away on business when this occurred, but when he found out he also was opposed. However, Keitai agreed with Kanamura. This conference and its decision is referred to repeatedly subsequently in Nihon Shoki and eventually led to the fall of the Otomo clan from prominence.

In the original notice Nihon Shoki says that some thought that Paekche had bribed Otomo. In short, the enemies of the Otomo claimed that this decision was the direct cause of the loss of Mimana. In 515 a section of Mimana rebelled and a Mononobe commander was sent in command of a fleet to restore order. However, he was defeated by the rebels and was apparently then rescued by a Paekche army. In 527 a large army was dispatched to Kyushu on the way to Mimana when its progress was blocked by a rebellion by the most powerful local ruler in Kyushu, named Iwai. We can be pretty sure that he was the most powerful because his Kofun, constructed while he was still alive, is the biggest one in the whole island. It's identification is pretty certain as it had certain highly distintive features mentioned in Nihon Shoki and confirmed archaeologically.

Otomo Kanamura recommended that the army continue on to Korea and that Mononobe no Arakabi take charge of the rebellion. This was agreed, and in 528 Mononobe fought Iwai in a large scale battle and killed him. Iwai's son surrendered and was pardoned on payment of a fine. These events were presumably aftermath of the replacement of the Ojin dynasty by a new one, but Nihon Shoki tells us nothing about the politics behind them. In 530 a delegation from Mimana came to the court to complain about the behavior of the general dispatched in 527. Basically, he was behaving like a tyrant and not doing much of anything to protect them against Silla. An officer was sent to recall him, but he refused to leave, so the officer requested military support from Paekche, which besieged him in his castle, but could not take it. The court sent additional officials and this time the general decided to return, but died en route. Keitai died only a few months later.

No one is quite sure how to deal with the records of the next three reigns because of suspicions that Nihon Shoki is once again trying to conceal what happened. A number of theories have been proposed and all of them have been vigorously attacked. There is not really enough evidence to be sure of anything.

According to Nihon Shoki Keitai died in 531 and was succeeded by his eldest son as Ankan Tenno. He died in 535 and was followed by his brother as Senka Tenno, who died in turn in 539, when he was followed by his half brother as Kimmei Tenno. However, it is possible to find in ancient sources three different dates for the death of Keitai. Kojiki says that he died in 527, the year of the Iwai Rebellion. Nihon Shoki has a footnote which says that there is an alternate version that he died in 534. The authors say that they decided to go with the 531 date because it matches an entry in Kudara Hongi. However, this entry simply says that a Paekche army operating in Mimana learned that the King of Kuguryo had been assassinated and that the king of the Wa and his heir and a second prince all died "at the same time." No names are given. In its main article Nihon Shoki says that Keitai summoned his heir, the future Ankan, and had him proclaimed ruler before his own death, the only time such an event is reported for this era. However, the article on Ankan says that he assumed the throne in 534. There are also two different dates given for the enthronement of Kimmei Tenno. Nihon Shoki says he took the throne in 539 on the death of Senka and died in 571 in the 32nd year of his reign. However, another ancient book says that Kimmei reigned for 41 years, implying that he took the throne in 531.

The first historian to tackle this was Hirako Takurei in 1905. He pointed out that Nihon Shoki says that King Song of Paekche presented a Buddha image to the Kimmei court in 552, but all other early sources of the Nara period say that this happened in 538. Hirako also noted that one of the Nara books that mentions the presentation of the statue says that the year was the 7th of Kimmei’s reign, which points to 531 for his enthronement. Hirako found room for the reigns of Ankan and Senka by proposing that the Kojiki date of 527 is correct for the death of Keitai.

Then, Kida Teiki, writing in 1928, proposed that the date of 531 is correct for Keitai’s death and that Kimmei was immediately elevated to the throne at that time. Ankan then established a rival court in 534 and was followed by Senka. The statement that “the crown prince” was elevated to the throne before the emperor died refers to Kimmei, not to Ankan.

Next, Hayashiya Shinsaburo proposed that the Iwai rebellion and the collapse of the Korean expedition caused the credibility of the Keitai court, dominated by the Otomo clan, to collapse. The leader of this criticism was Soga no Iname. The Kudara Hongi item quoted by Nihon Shoki means that Soga assassinated Keitai and at least one of his sons, presumably including the designated heir. Soga then placed Kimmei on the throne. However, the Otomo faction responded in 534 by setting up Ankan as a rival emperor. This standoff lasted for 7 years but was finally resolved with the political destruction of the Otomo, who are not mentioned after 539.

The last version I will offer is by Inoue Mitsusada. He called attention to an event of the first year of Kimmei's reign when a conference was called on the question of whether to send an army to Korea. Mononobe no Okoshi said that sending only a small force would accomplish nothing and said that the main cause of the problem was that Otomo no Kanamura had transferred the four districts to Paekche, which angered Silla, which has been aggressive ever since. Kanamura was not at this conference, claiming illness. A messenger was sent to inquire about this and he said that he knew he was going to be criticized and didn't want to come.

Inoue says that, first, it is really pretty clear that when Nihon Shoki reports the hastly elevation of the crown prince before Keitai's death Ankan is meant since he is identified using a short form of his personal name. Inoue explains the item in Kudara Hongi as a mistake. The courtiers refused to recognize the enthronement of Ankan and elevated Kimmei instead and this fooled the Koreans into thinking that Ankan, who they knew was the heir, must have died also. When they deposed Ankan they probably stripped Senka of his status as well, explaining the reference to two princes dying at the same time as the ruler. Inoue argues that if Soga had killed Keitai he would certainly have killed the two princes and probably Otomo no Kanamura as well.

In the end, whatever the details, it is strongly suspected by most that Keitai died and Kimmei was enthroned in 531 and that Ankan established a rival court in 534, and that after the death of Senka in 539 the Otomo clan no longer held high status at court. The Soga clan replaced them in prominence.

In 537 the Senka court ordered Otomo no Kanamura to send an army to Korea commanded by two of his sons. One son was to take the army to Korea and the second to control supplies from Kyushu. This force actually went to Korea (it is mentioned in other ancient writings) but it is not known what it may have accomplished.

By far the largest part of the Nihon Shoki article on Kimmei is actually Korean. It is a relatively readable account of extensive and ultimately unsuccessful attempts by King Song of Paekche to persuade Kimmei to do something serious about defending Mimana from Silla.

The Japanese position in Korea was completely lost by 562. This was in part the consequence of a major series of wars among the three Korean kingdoms that started in 548 with a powerful attack on Paekche by Koguryo. Silla unexpectedly sent troops to the aid of Paekche and the attack was driven off. Paekche and Silla then formed an alliance to seek to recover various lands in the north that had been gained by Koguryo in recent decades. Koguryo was defeated in 551 and a large territory was captured, but Silla then seized it all. King Song of Paekche decided to stake everything on an all out invasion of Silla, but when travelling near the battle zone with a small force he was ambushed and killed in 554. In both China and Japan it was thought for a time that Paekche would collapse and disappear, but that did not happen. However, Silla was now unquestionably the most powerful kingdom in Korea. It had undertaken strenuous reforms on the Chinese model starting with the accession of King Chijung in 500. With Paekche out of action at least temporarily, Silla was able to quickly complete the conquest of all of Mimana. Apparently there was not much local resistance as the locals were fed up with Kimmei's refusal to do anything to help them earlier. It is possible to suggest that they had loyalty to the Ojin dynasty rather than to Japan.

The functioning of the Yamato State

Lars Vargo's monograph, listed in the bibliography, is particularly useful here. Unfortunately, it may be rather hard to find. It is basically an annotated bibliography which summarizes the views of a large number of Japanese historians as of 1982, and in the course of performing this task it quotes essentially every relevant scrap of information in Nihon Shoki. The most striking result is the fact that hardly any of the historians agree with each other about anything significant. The problem is the same as with the other topics discussed here, the amount of real information is too small to reach definitive conclusions, leaving enormous room for speculation. Nihon Shoki shows no interest at all in telling us how the Yamato state worked. It drops crumbs of information from time to time in passages that are actually about something else. These crumbs fall into five categories. There is information about territorial organization, about aristocratic clans, about aristocratic titles, about those groups of commoners who were most important to the rulers, and about a mysterious institution known as "miyake."

The Chinese said that Japan was divided into "countries," and the archaeologists tell us of status differentiation as displayed by tombs that permit us to identify a tomb suitable for a village chief, a tomb suitable for the chief of a district, and a tomb suitable for the "king" of a "country." These grades map to territorial names and titles in the Yamato state. For the Kofun period we see essentially two territorial titles, agata nushi for the man who controlled a relatively small district and who would be buried in a "small" tomb, and kuni no miyatsuko, for a much higher status person who would be buried in a "medium" tomb. Most historians think that an agata was a real unit and the agata nushi was the man who saw to it that the wealth needed by the aristocrats was extracted from the commoners. Agata had names and could presumably be pointed to on a map, and many of their names carry forward to become districts within the provinces and quite a few survive as local district names today. The kuni no miyatsuko is more amorphous. Few have any confidence that a kuni was an identifiable territory with a boundary, most think that the title of kuni no miyatsuko simply indicated a powerful man who controlled a number of agata nushi. The layout of their authority might be highly changeable over time. However, they unquestionably had real power because they built much bigger and more elaborately equipped tombs than other people.

These aristocrats were organized into clans known as "uji." In theory all clans were descended from one clan founder, and much of the text of Nihon Shoki is devoted to stories telling how a particular clan got its start. They had names which were most commonly place names, indicating the district which they controlled. Most clans were territorial. A minority had names indicating that they had a functional role closely linked to the ruling house rather than a territorial base. These clans appear to have had units scattered across the country, and some think that these units may not have been related to each other by blood. In the hypothetical horse rider invasion scenario, there would have been a large body of warriors, divided into a hierarchy corresponding to military units, with a smaller number of persons who were closely associated with the ruling group. The former became the main body of the rural aristocracy, the kuni no miyatsuko and agata nushi, and the latter took up special responsibilities in assisting the ruler to maintain control over the whole.

For the Ojin dynasty, typical examples are the Kazuraki clan, a territorial clan which furnished many rulers' wives, and the Heguri, a similar clan with a similar role. However, there were also the Otomo clan which appears to have had a role as hereditary leaders of the ruler's military forces and the Mononobe clan which was in charge of the manufacture and storage of the ruler's weapons and armor. These clans had a nationwide distribution.

Each clan had its own special god and one important function of the clan chief was to serve as the god's chief priest. In later times, when Shinto shrines came to be built as a response to Buddhist temples, members of the uji founded their hereditary priesthoods. Nearly every ancient Shinto Shrine is traditionally linked to a particular uji.

There is a word written using the standard Chinese character used today for a government department 部 which can be pronounced "be" (as in berry). This is thought to derive from the government of Paekche where there were 12 departments responsible for supplying all of the needs of the royal palace. The be in Japan were responsible for supplying the needs of the ruler also. It would seem that these were groups of commoners who were placed under the control of an aristocratic family permanently, as a hereditary status that in their case substituted for a territorial assignment. The members of a be had to eat, of course, and nothing is said anywhere about how that might be accomplished, whether they grew their own food or were paid for their other services. The be themselves were responsible for things such as supplying pottery for the palace table and haniwa for the royal tombs (hajibe), running the kennels with the royal hunting dogs (inukaibe), setting up sites for religious ceremonies and then cleaning up afterward (hafuribe), guarding royal buildings (yukehibe), and so on. The royal clan definitely had its own territories where it owned the farmers and collected their crops into its warehouses. There was a special category of be that consisted of the immigrants from Korea. These produced high class goods like silk fabrics and writing paper.

Miyake are the only institution mentioned that looks like something that only a ruler could do. The rural aristocrats probably had their own local versions of be. A miyake is a named location that was controlled by the ruler via an official known as a tomo no miyatsuko. "Tomo" is a term that implies a military unit, and tomo no miyatsuko could mean something like "captain." The Keitai dynasty certainly had bands of warriors that were specially loyal to the clan that they could call on and these were the core of their army. In peacetime they were simply rural aristocrats living on an agata somewhere. The word miyake consists of "mi-" which is an element meaning royal and "yake" which means a house or other sizeable building. A miyake in early times was evidently a physical warehouse or group of warehouses that would contain grain. Apparently, when a miyake was established the local aristocrats were compelled to keep it filled, so that when a royal military unit came through it would be able to draw grain for rations. Miyake were the only thing we can see that looks like a form of taxation on the aristocracy.

It is noticeable that the formation of miyake is often mentioned in the aftermath of a rebellion, though that may be because Nihon Shoki normally only mentions rural areas in connection with rebellions. Two hundred years later the word miyake seems to have been used briefly for a private landed estate and this has caused a lot of confusion. The clearest example I know of relates to the establishment of a miyake in 535 at the port of Nanotsu, the most convenient place in Kyushu to transship men or supplies heading from central Japan to Korea. It was to be initially supplied by transferring grain from miyake in the Kinai and then its regular supply would come from the suppression of three miyake in northern Kyushu that were in less useful locations. The grain formerly sent to them would now go to Nanotsu. Then, although much later, in 607, a project was inaugurated calling for irrigation works in three provinces, and a miyake was to be set up in each province to feed the workers.

The national administration is as amorphous as everything else. There is mention periodically of two special titles that appear to designate the most powerful men at court, "o-omi" and "o-muraji" where "o-" means great. During the latter part of the Ojin dynasty and the early part of the Keitai dynasty the only man ever called "o-omi" was the head of the Otomo clan. There were sometimes two "o-muraji" simultaneously, but one of them was always the head of the Mononobe clan. Numerous other people had the title of "omi" or "muraji". These are the kind of titles that are called "kabane" in technical discussions. They were hereditary or traditional status designations such that a particular clan might be called "omi" but never "muraji" and vice versa. It appears that clans titled omi were considered to be collaterals of the royal house and therefore eligible to marry with princes and princesses. The muraji clans were not allowed this privilege.

Kabane are complicated (there are articles about them in the bibliography above), but they show that the court was primarily about status and not about jobs. In Paekche the 12 departments that supported the palace were balanced by 12 other departments that ran the country outside of the palace. The Yamato state had no hint of any such thing. The O-omi and the o-muraji were, essentially, the men who were allowed to have the last word in discussions, but they held no special portfolio. The ruler, in theory anyway, apparently could choose which of the men entitled to omi would be designated o-omi and which of the muraji would become o-muraji, but he could not give a clan traditionally called muraji the title of omi.

The only thing we can say for sure, and it continued to apply in the historical period, is that whenever there was a serious issue there was a meeting of the senior nobles at which everyone was entitled to speak, lowest ranking first, highest ranking last. If there was a consensus, the ruler would go along with it. Even today it is a characteristic of Japanese government and businesses alike that it is difficult to proceed with anything if there is not a consensus among the key people. There have been few autocrats powerful enough to disregard this way of thinking. The way you get things done is to only call meetings when you know what the result will be because you have thoroughly consulted with everyone first. If one important person won't go along, you may have to change your plans. The purpose of the meeting is to allow everyone to see that there is a consensus, so all can go forward together.

Within the ruling family, at least, each adult prince (and often princesses, too) would have his own house somewhere in the countryside. It was the custom that when a man died his house was burned and his heir built a new one in a new location. These were evidently rather flimsy constructions not intended to last for generations. It was not uncommon for a ruler to change the location of his palace. Usually no reason was given. Sometimes it appears to be a move from a location where he felt insecure to a safer one, but it may often have been to equalize the burden on the commoners, much as medieval European rulers found it more convenient to circulate from estate to estate than to attempt to move supplies on mostly theoretical roads.

I have already mentioned that there was considerable violence within the aristocracy. We know mainly about the ruling family, but there is evidence of similar habits elsewhere. The favorite method of fighting an internal aristocratic war was a surprise attack on the rival's house in the middle of the night. Open battles between organized forces were rare. Assassination also occurred, often while hunting. It appears that it was the custom that the women of the defeated rival were taken as booty.

All in all it would be fair to say that the level of social development in this era was rather like that shown by the Celtic tribes known to the Romans of the era of Julius Caesar, though the construction of the kofun tombs shows that the rulers had rather greater power to gather wealth from the commoners than their European counterparts. As of the reigns of Keitai and Kimmei there is no sign of direct influence from China nor is their any sign that the experience of the Korean kingdoms was causing the Japanese to think that their own society needed to be reformed. The Korean kingdoms were at least nominally literate, but it is believed that they were only beginning serious efforts to replace tribal methods of rule with Chinese style bureaucratic institutions supported by formal taxation in the 6th century. It is clear that Paekche attempted such a reorganization only after its disastrous war with Silla between 551 and 554. Before then, it is believed, "civilization" existed only in the royal capital and the countryside was administered by traditional rulers much like those found in Japan. Silla raised itself from the weakest kingdom to the strongest in the period of a generation by using Chinese methods to increase the power and wealth of the state, and Paekche and Koguryo were forced to respond in order to survive. In the next generation Japan began to feel the same pressure.

For that reason, this ends the discussion of the Yamato State, and the period from 572 onward is assigned to the next chapter.

Japanese History

Introduction
Prehistory through the Jomon Period – The Yayoi Period – The Kofun or Yamato Period – The Asuka Period – The Nara Period – The Spread of Buddhism in Japan – The Early Heian Period – The Middle Heian Period – The Late Heian Period – The Kamakura Period – The Kemmu Restoration – The Nanboku-chō Period – The Muromachi Period (Ashikaga) – The Warring States Period – The Azuchi–Momoyama Period – The Edo Period – The Meiji Restoration – The Meiji Period – The Taisho Period – The Rise of Militarism – World War II – The American Occupation of Japan – Post-War Japan – Japan Today
Further Reading