Pathway to Japanese Literature

Among the various theories proposed to resolve the question of which indigenous race in our island nation used stone tools and left ruins for posterity, the most numerous were two schools of thought: one attributing these remains to the Emishi, and the other considering them the work of a distinct race from the Emishi—the Koropokkuru.
Other theories such as those regarding the Suksin people and Tsuchigumo existed but had relatively few supporters.
Among scholars advocating the Emishi theory, the most influential was undoubtedly Dr. Koganei; while the foremost proponent of the Koropokkuru theory was equally certainly Dr. Tsuboi.
The research of both scholars had reached a level of remarkable meticulousness.
We discussed here the similarities and differences between Emishi and Koropokkuru without presuming to comment on either doctor’s theories or attempt their critique.
We had always been deficient in archaeological knowledge and possessed almost no anatomical expertise.
Therefore concerning these points we could do nothing but revere and trust our predecessors’ theories.
However we also maintained certain convictions derived from historical records.
That is to say—establishing our argument from this perspective—we intended to contrast it with various theories already researched by our predecessors and thereby present our humble views regarding their similarities and differences.

Before presenting my humble views, we acknowledge the necessity of first examining the various theories already published concerning the similarities and differences between these two.

Dr.Koganei stated:

"When considering the degree of civilization among the Ainu people, what we call Koropokkuru or Tonchi must be understood as none other than the Ainu themselves.
That is to say—the Ainu make their livelihood through hunting beasts and catching fish; when examining all aspects comprehensively,I believe they never possessed metal-forging techniques.
From their era of shooting beasts with stone-tipped arrows and catching fish with stone-pointed harpoons onward,the Ainu have made little progress.
When we consider them remaining at such a low developmental stage—merely obtaining metal implements from other peoples—it becomes evident that these entities called Koropokkuru or Tonchi are indeed identical with the Ainu themselves."
(*Toyo Gakugei Zasshi*, No．260)

This theory was constructed from the standpoint that the Emishi constituted Stone Age people, positing that no race called the Koropokkuru existed apart from the Emishi.
Dr. Tsuboi maintained the theory that Koropokkuru were distinct from the Ainu; as his arguments had appeared in numerous papers and were believed to be universally known among readers, their citation was omitted here.
However, the first half of Dr. Koganei’s treatise on Stone Age inhabitants published in *Toyo Gakugei Zasshi* No. 259 synthesized Dr. Tsuboi’s arguments from numerous fragmented sources. When Dr. Tsuboi expressed his gratitude by stating, “Pages 151 to 163 of No. 259 primarily summarize my theories, skillfully gathering and organizing my scattered writings—for which I am deeply thankful,” it became evident that this work thoroughly encapsulated the essence of his theories. We therefore regarded it as representative of Dr. Tsuboi’s arguments.
Legal Scholar Nakada had also recently dismissed both the Ainu theory and Koropokkuru theory as extreme fallacies in his *“Ainu Divine Names Appearing in Our Ancient History”* (*Shigaku Zasshi* Vol. 17, No. 9), instead proposing that Koropokkuru constituted a branch of the Ainu people.
In a certain sense, we held opinions resembling this theory.
However, Legal Scholar Nakada further adopted the view that Koropokkuru were equivalent to earth spiders (Tsuchigumo), interpreting them as pit-dwelling adherents while considering the Ainu to be house-dwelling adherents—thus asserting that although both groups shared common origins, a clear distinction existed between them.
Yet regarding what state house-dwelling Ainu occupied during the era when pit-dwelling Koropokkuru—that is, earth spiders—had spread as far as Kyushu, his views lacked specificity.
If Mr. Nakada’s theory posited that all initially dwelled in pits, with a portion later advancing to adopt house-dwelling customs, then concerning their interrelation, this view would entirely converge with Dr. Koganei’s thesis—further augmented by the additional claim equating Emishi with earth spiders.

The one who had published opinions identical to Mr. Nakada’s prior to him was identified as Mr. Numata Yorisuke. Mr. Numata stated:

Both Dr. Tsuboi’s Koropokkuru theory and Dr. Koganei’s Emishi theory proved through contemporary research to be consistent—in other words, those who had left behind Stone Age artifacts and ruins in our nation were ancestors of the Kuril Ainu, a branch of the Emishi people, while the Koropokkuru legends preserved in Ainu oral tradition appeared to relate specifically to these Kuril Ainu.
(*New Theories on the Japanese Race* [Nihon Jinshu Shinron], p. 26)

Legal Scholar Nakada merely supplemented Mr. Numata’s aforementioned theory by asserting that Tsuchigumo equated to Koropokkuru.

Setting aside the matter of Tsuchigumo for now, regarding the relationship between the Emishi and the Koropokkuru, we are largely in agreement with Mr. Numata. And thus, regarding these similarities and differences, we intend to construct our argument using the threefold classification of Acculturated Emishi, Unassimilated Emishi, and Tsukaru as documented in *Iki no Hakotoko’s Account*, alongside distinctions such as Watarito, Karako, and Hinomoto found in the *Suwa Engi Emaki* (Illustrated Legends of Suwa Shrine).

*Iki no Hakotoko’s Account* states:

The Son of Heaven (Emperor Gaozong of Tang) inquired: “In which direction do these lands of the Emishi lie?”
The envoy respectfully answered:
“The land lies in the northeast.”
The Son of Heaven inquired again: “How many kinds of Emishi are there?”
The envoy respectfully answered:
“There are three kinds: the distant are named Tsukaru, the next Ara Emishi, and the near Nuki Emishi.
Now these Acculturated Emishi pay tribute annually to our sovereign’s court.”
The Son of Heaven inquired: “Does their land have the five grains?”
The envoy respectfully answered:
“There are none; they survive by eating meat.”
The Son of Heaven inquired: “Does their land have houses?”
The envoy respectfully answered:
“There are none; they dwell in deep mountains, living among trees.”
The Son of Heaven remarked again:
“We find the physiognomy of these Emishi to be most remarkably curious.”
...

We recognized the necessity of first scrutinizing the value of Hakotoko’s account before investigating the content of the aforementioned record. Hakotoko was a man of Empress Saimei’s court; in the fifth year of her reign, seventh month, he entered Tang China accompanying Sakaibe no Ishifu, Tsunomori no Kichijo, and others. In this year, Abe no Hirafu led a naval force to subjugate the Emishi, captured prisoners, and returned. When Ishifu and others entered Tang China, they brought along two Emishi—a man and a woman—and presented them to the Emperor of Tang. Hakotoko likely accompanied their mission as a scribe and also as an interpreter. And that which recorded his observations and experiences is none other than the so-called *Iki no Hakotoko’s Account*. Therefore, what it records was a factual account, by no means written from conjecture or supplemented later. Especially in the case of its records concerning the Emishi—compiled during this period when our generals had newly subjugated the Emishi and brought back the latest knowledge, and moreover, when they had personally accompanied two Emishi individuals on their mission—what it documents possessed sufficient value to be trusted as a direct transcription of actual conditions at that time. In the *Book of Tang*’s "Eastern Barbarians Biographies," it records the events of this time as follows:

The Emishi envoy had a beard measuring about four shaku (approximately 1.2 meters) in length and wore arrows at his neck; he would have someone place a gourd on their head and stand several dozen paces away, whereupon he would shoot without fail.

It is written.
Before the Tang Emperor, he performed this Wilhelm Tell-esque archery in full display.
He already possessed extraordinary features and moreover performed this acrobatic feat.
That the Emperor of Tang posed inquiries to our envoy, as seen in Hakotoko’s Account, was precisely as it should be.
Therefore, we recognize considerable value in Hakotoko’s Account’s records concerning the Emishi.

This valuable Hakotoko’s Account explains that there were three kinds of Emishi at that time.
They were termed Acculturated Emishi; Unassimilated Emishi; and Tsukaru.
The Acculturated Emishi and Unassimilated Emishi correspond to the Acculturated Jurchen and Unassimilated Jurchen among the Jurchen, and to the Acculturated Fan and Unassimilated Fan in Taiwan.
And it explicitly states that north of the Unassimilated Emishi there existed a people called Tsukaru.
The Acculturated Emishi paid tribute annually to the court of our country.
That is, they were the Emishi who had submitted to the court.
It was evident that the Unassimilated Emishi, in contrast to those who had submitted, were Emishi who had not yet yielded to the kingly influence.
And if we suppose that there was Tsukaru further to their north, then by what should we liken them?
We, in response to this question, first intend to propose the hypothetical theory that “these must have been Emishi at the stage of using stone tools.”
This hypothetical theory is one that I recall Mr. Numata Yorisuke and I once touched upon during our discussions concerning the Emishi when he was composing his *New Theories on the Japanese Race*, and to which I also expressed my agreement.
However, when the same book was later published, he made no mention of it whatsoever.
It seems that Mr. Numata, deeming this theory still unsettled, adopted a cautious stance and thus refrained from including it in his published work.

Incidentally, in this book, citing my theory, [Mr. Numata] included an article stating, “It appears that nearly all surnames granted to captive-allies by the court were *Kimiheguri*,” and appended words of praise: “This seems an eminently appropriate theory.” However, my argument posits that “those granted the *Kimiheguri* surname by the court were exclusively captive-allies,” whereas this book inverted this relationship. As it constitutes an inappropriate interpretation, I hereby provide correction.

To substantiate this hypothetical theory, we must first examine the etymological meanings of *Ara Emishi* (Unassimilated Emishi) and *Nuki Emishi* (Acculturated Emishi). *Ara Emishi* refers to those Emishi not yet acculturated, while *Nuki Emishi* denotes those already acculturated. And if we posit that north of the unacculturated Ara Emishi there existed yet another kind of Emishi, we must imagine that these were necessarily even less acculturated. However, to speak of “becoming acculturated” and “remaining unacculturated” is to use relative terms, for even within what is termed “unacculturated,” there naturally exist degrees. Therefore, in dividing these unacculturated groups into two categories based on degree: one would consist of those who, like the raw tribes of present-day Taiwan, while not yet submitting to kingly influence, had already established communication with other ethnic groups to conduct certain types of goods exchange and had abandoned the use of their original stone tools; the other would be those who had not yet reached this stage. And during the Matsumae clan’s rule, the Ainu of remote areas on Hokkaido’s main island were at the former level, while the North Kuril Ainu prior to Russian contact would have been in the latter category. Is the distinction between the Unassimilated Emishi and Tsukaru not also sufficient to imagine? And as for these, the Acculturated Emishi—being those who had already submitted to the Japanese government—under a government that at the time adopted policies to shift customs and reform practices, it is believed they must have gradually assimilated into Japanese ways, even in language, eventually coming to use Japanese. And thus, these Acculturated Emishi—the majority of them—would have assimilated and intermingled among the Japanese people until their traces were extinguished.

The distinction between Acculturated Emishi, Unassimilated Emishi, and Tsukaru Emishi being clearly based on their degree of civilization was a matter that now admitted no further doubt.
The dwelling places of these various kinds of Emishi likely had not been fixed across eras; just as the name Hidaka—their country’s appellation—remained in the two regions of Hitachi and Rikuzen, their boundary lines must surely have shifted northeastward in accordance with the development of our nation’s power.
And we believed that the so-called Tsukaru Emishi of Hakotoko’s time, based on records such as those of Abe no Hirafu’s subjugation campaigns, must have inhabited the area stretching from what is now the Tsugaru region to Hokkaido.
In Hokkaido’s Oshima as well, there had originally been Tsugaru District, which merged with Fukushima District to become Matsumae District.
In the *Shoku Nihongi*, there was a place-name called Tsugaru Port in Oshima.
Therefore, Tsugaru was originally a place name spanning from the northern tip of Honshu to Hokkaido; alternatively, whether the Tsugaru of Oshima derived its name from the migration of the so-called Tsukaru Emishi from Honshu—though we had yet to ascertain the full details—given the antiquity of Oshima’s Tsugaru name, it became reasonable to believe that Tsugaru Emishi had already spread to Hokkaido’s main island by Hakotoko’s time.

The majority of the Oku region coming under imperial rule was a development that occurred after the Heian period. In the preface to the *Shoku Nihongi*, there is a passage that states: “Imperial authority extended east of the Nichigawa River, and the Mao Di barbarians held their breath.” As for the Nichigawa River being the upstream river, this was explained in detail in the previous issue of this journal; and that imperial authority extended as far north as Sendai Bay was a development that occurred after the reign of Emperor Kanmu. Looking further back to Hakotoko’s era, we observe that along the Japan Sea coast, imperial authority had extended deep inland owing to Hirafu’s subjugation campaigns; yet during the Taika era, its reach on the Pacific side barely stretched beyond the Iwahune Barrier (in Echigo’s Iwafune District) to the Iwaki-Iwashiro region. Meanwhile, the Acculturated Emishi are thought to have intermingled with the Yamato people in areas south of this boundary.

As our national power developed, its borders gradually expanded northward; by the time of the Former Nine Years’ War and Later Three Years’ War, imperial authority had reached even the Rikuchū and Ugo regions, and the Emishi of this area became Acculturated Emishi—that is, captive-allies.

The Former Nine Years’ War arose from the rebellion led by the barbarian Abe no Yoritoki and his son Sadato, who commanded captive-allies, while the Later Three Years’ War originated in conflicts between Kiyohara no Takenori, a captive-ally from Dewa Province, and his kinsmen Yoshihiko (Yoshimiko) Yoshitake and others.

Due to these two successive wars, some among the Acculturated Emishi in these regions may have fled as far as the Hokkaido area.
According to what is recorded in the Konjaku Monogatari as the account of Munetō, younger brother of Sadatō, it appears that Yoritoki and his group also once planned to flee to Hokkaido.
Afterwards, when Minamoto no Yoritomo destroyed Fujiwara no Yasuhira, the entire Ōu region submitted completely.
That Yasuhira attempted to flee to Hokkaido at this time but failed is clearly recorded in the Azuma Kagami.
At that time, there can be no doubt that many of their captive-allies fled to that land.
The Acculturated Emishi who crossed over from the mainland are called Watarito.
They, indeed having once become Acculturated Emishi, primarily used our national language in matters such as speech, and differed in character from other Unassimilated Emishi.
The *Suwa Engi Emaki* from the Nanboku-chō period distinguishes the Emishi of the late Kamakura period as follows:

What the Emishi referred to as Chishima lay in the center of the great sea corresponding to the northeast of our country.
The Hinomoto, Karako, and Watarito—these three groups each inhabited three hundred and thirty-three islands.
The two islands were now incorporated into the Watarito.
Within them were small islands called Usori, Tsurugoshu, and Mandō Umamijō.
This group frequently engaged in trade along the coasts beyond Oshu Tsugaru.
Ippai referred to six thousand people.
When they gathered, their numbers reached a hundred thousand Ippai.
The two categories of Hinomoto and Karako, their lands adjoining foreign countries, had forms like yasha (demonic beings), undergoing endless transformations.
They consumed humans, beasts, fish, and meat as sustenance, yet knew nothing of cultivating the five grains.
Even after repeated interpretations, they remained nearly impossible to communicate with.
The Watarito resembled the people of Wa Country.
However, they had abundant beards and their bodies were covered in hair.
Although their language was rustic, much of it was mutually intelligible.
...

”
When considering what is termed Watarito here, judging by their description as having abundant beards and hair covering their entire bodies, they were clearly Ainu.
However, the other two categories had demon-like forms, consumed beasts, fish, and meat as sustenance, and knew nothing of cultivating crops.
Contrary to claims about communication difficulties, we observe that the Watarito resembled people of Wa Country—engaging in trade along the coasts beyond Ōshū Tsugaru with largely mutually intelligible language—while the other two categories preserved their inherent customs. These Watarito adopted clothing somewhat resembling Japanese styles, understood agricultural methods, primarily shifted to using Japanese language, and interacted with mainlanders.
Such were what we call the Acculturated Emishi.
That is to say, from the Kamakura period through the Nanboku-chō period, we know the southwestern part of Hokkaido had already become the dwelling place of the Acculturated Emishi.
These Acculturated Emishi later intermingled with the lords of various forts, came under the Matsumae clan, and ultimately reached a point where they forgot their identity as Ainu, thereby vanishing without trace.

In contrast to these Acculturated Emishi, those distinguished by their lack of shared language, ignorance of agriculture, and forms like yasha (demonic beings) must have corresponded to what Hakotoko termed either Unassimilated Emishi or Tsukaru.
They were called Hinomoto and Karako, and their land was said to adjoin foreign countries.
From these accounts, what facts can we extract?

Hinomoto was originally the name of our country.
However, that this appellation of "Hinomoto" should exist for Emishi even more primitive than the Acculturated Emishi had long been a point that scholars could not explain.
The *Hokkaido-shi* explained: "Hinomoto are a racial group of Emishi from Honshu; Karako are mixed lineages from foreign lands; Watarito are people from inner districts who crossed the sea to migrate." Yet it never detailed the foundations of this theory. We too had long harbored doubts regarding this matter, but have now arrived at a tentative understanding.
Could Hinomoto be the very Kuril Ainu?

The designation of our country as Hinomoto originally arose from its geographical position relative to China. During the reign of Empress Suiko, envoys were dispatched to Sui. In contrast to referring to him as the Western Emperor, they called themselves the Eastern Emperor; in contrast to calling him the Son of Heaven where the sun sets, they called themselves the Son of Heaven where the sun rises. Hinomoto is none other than the place where the sun rises and signifies the east—this is made clear by the aforementioned. Therefore, if we speak solely within our country, the appellation “Hinomoto of Oshu” already existed during the Azuchi-Momoyama period. The fact that the Pacific coast of Rikuzen was called “Sunrise Cliff” can be traced back to the early Heian period. Sunrise Cliff is synonymous with Hinomoto. Therefore, if we were to speak of this concerning Hokkaido alone, its eastern part is the place where the sun rises—that is, Hinomoto. And the Emishi term *Chupka* corresponds precisely to this meaning. Chupka is the term used by the Hokkaido mainland Ainu to refer to the Kuril Islands. And its meaning is said to signify “east” or “the place where the sun rises.” The Hokkaido mainland Ainu also call the Kuril Ainu *Chupkaguru*. It signifies “people of the east” or “people of the sunrise place.” Therefore, when the Watarito or Japanese people translated this as “Hinomoto,” was this not an appropriate appellation? Chupka extends to Northeast Kamchatka and connects to the continent. That which states “their land adjoins foreign countries”—is this not indeed capturing the truth?

However, what required attention here was the historical evolution of the scope referred to as Hinomoto—that is, Chupka.
In present times, it has become customary to call only the Kuril Islands—formerly Russian territory—Chupka; however, this likely became the case after the domain of Hinomoto—that is, Chupkaguru—came to be confined to these islands. When Chupkaguru still occupied Kunashiri, Etorofu, and other such islands, there could be no doubt that these too had been included within Chupka.
The terms *Hinomoto*, *place where the sun rises*, and *east* were inherently relative; thus, when speaking in contrast to the Hokkaido mainland, Kunashiri Island must also have been considered Chupka.
The migration of Hokkaido mainland Ainu to Etorofu Island had been occurring for two hundred years.
In Kondo Morishige’s *Tatarafuki Dan*,

Within the villages of Etorofu there were earthen pits where dwarfs had resided.
(Omitted) It was said that ever since Emishi from Kunashir migrated to this Etorofu Island eighty to ninety years earlier, those dwarfs had gradually dispersed.

... it is said.

Therefore, we can reasonably infer that prior to this period, the Hokkaido mainland Ainu crossed over to Kunashiri Island and drove out the indigenous people there.
Consequently, does this not make evident that these islands were originally situated within Chupka as dwelling places of Chupkaguru—that is, the Hinomoto Emishi?
If we trace back even further beyond that time, the eastern part of the mainland would also have been Chupka relative to the west.
The prevalence of relatively recent Stone Age ruins throughout Hokkaido has by now gained acceptance within academic circles.
Were comparative archaeological research to prove that these ruins belong to the same lineage as those in the Kuril Islands—and were these ruins shown to grow comparatively newer as one moves eastward from those in the western regions of the main island—we would then have sufficient grounds to believe that the eastern part of the mainland too once served as dwelling places for Chupkaguru during a certain era, that is to say, as Hinomoto or Chupka.
The oral traditions of the Hokkaido mainland Ainu concerning the Koropokkuru must be examined in conjunction with the account of dwarfs on Etorofu Island found in *Tatarafuki Dan* as representing identical phenomena.

Based on the foregoing arguments，we believe it reasonable to conclude that during the Kamakura-Ashikaga period，the Emishi referred to as *Hinomoto* by Japanese people or the Watarito were identical to what is termed *Chikopuka* in the Ainu language—that is，today’s Kuril Ainu—and that their dwelling areas likely extended from the eastern part of the mainland [Hokkaido] to the Kuril Islands．However，it must be noted that what we refer to here as the “eastern part” is of course a vague designation；we anticipate that future archaeological research may one day allow us to establish somewhat more concrete boundaries．Next，we shall seek to find correspondences between Hinomoto—that is，the Kuril Ainu—and what Hakotoko termed the Tsukaru Emishi．

Regarding the customs of the Tsukaru Emishi, there were almost no records that could be examined.
In the *Mutsu Fudoki*, when the Tsuchigumo of Yatsuki District encountered Yamato Takeru’s subjugation, they employed many Tsugaru Emishi as spies and repeatedly fired boar-deer bows and boar-deer arrows from their stone fortifications to resist the government forces.
Yamato Takeru took up his tsuki bow and tsuki arrows and shot them.
Needless to say, we did not believe that this oral tradition conveyed historical facts.
This must have been what Mr. Kobayashi referred to as the Tsuchigumo legend created to explain place names.
However, the fact that Tsugaru Emishi—hardly essential for explaining place names—were explicitly mentioned here provided our research with exceptionally valuable material.
For the existence of the Tsukaru faction of Emishi had been proven by Iki no Hakotoko, and those in the southern regions of Iwaki and Iwashiro—which at that time was a region where Japanese and Emishi coexisted—could not have been unaware of this.
Therefore, even if we did not accept as historical fact that the Tsuchigumo allied with the Tsugaru Emishi in Yatsuki District to resist Yamato Takeru based on the aforementioned *Fudoki* records, at least the following facts could be demonstrated through these oral traditions.

1. The region of Yatsuki District (within Shirakawa County) possessed an oral tradition that at least during the Nara period, the Tsugaru Emishi once came to this place.

2. Consequently, the Tsugaru Emishi resided in a region that allowed for easy communication with the Shirakawa area during a certain period.

3. That region must have been closer to the Shirakawa area than to present-day Tsugaru in Aomori Prefecture or Tsugaru in Hokkaido.

4. Consequently, Tsugaru was originally the name of a tribe (this can be verified from Hakotoko’s writings); the later place name Tsugaru must have been derived from the name of the Tsugaru tribe becoming fixed to their dwelling place.

Incidentally, it was said in one theory that Tsukaru was none other than Okōro and that they were migrants from the continent.

Until such time as future research in comparative linguistics, archaeology, ethnology, and related fields establishes sufficient connections between our Stone Age ruins and those of North Korea and Manchuria, we must retain our doubts and append this note accordingly.

5. Whereas the Imperial Army employed tsuki bows and tsuki arrows, the Tsugaru Emishi used boar-deer bows and boar-deer arrows.
Although there exists a theory interpreting *ika* as meaning "severe," it is most appropriate to understand *inokoya* [boar-deer arrows] as arrows that used boar tusks, deer antlers, or bones for their arrowheads.
As for the boar-deer bow, although its nature remains difficult to ascertain, given its contrast with the mention of tsuki bows and tsuki arrows, it may have been a term used rhetorically in conjunction with "boar-deer arrows."
Alternatively, there may have been such a designation for bows decorated with animal horns, bones, teeth, and the like.

From the above facts, although we did not obtain definitive proof that the Tsugaru Emishi were in a state of using stone tools, we came to know that their weapons differed from those of the Imperial Army and that they used arrowheads made of bone or horn rather than metal.
This Emishi, by the Heian period, had already retreated and are believed to have been in what is now the Tsugaru region.

In the second year of Gangyō (878), a rebellion by the Emishi captive-allies in Dewa occurred.
At this time, the captive-allies joined with government forces and appear to have assisted in the subjugation of the Emishi captives.
Captive-allies are the acculturated Emishi, while Emishi captives refer to the unassimilated Emishi.
At this time in Dewa, the boundary between the Emishi captives and captive-allies lay at the Akita River.
And the Tsugaru Emishi were located even deeper inland in the Tsugaru region.
The official documents of the time state:

The Tsugaru Emishi: their factions were of many kinds, and how many thousands they numbered remained unknown. They were by nature brave and robust, ever trained in warfare. Were they to ally with rebels, their vanguard would prove difficult to withstand.

Furthermore, it states:

"The Tsugaru Emishi captives are innately wild and untamed.
If they were to ally with violent factions, it is said that they would indeed be difficult to control."

Here they noted their great numbers and spoke of their fierce nature.
There could be no doubt that they constituted a formidable force not to be underestimated.
Therefore, unless one imagines this people subsequently became completely extinct, the most reasonable view would be to conjecture that during the era when ordinary Emishi retreated northward to become the Hokkaido mainland Ainu, the Tsugaru Emishi withdrew even further into the hinterlands and became the Kuril Ainu—that is, the so-called *Hinomoto*.
And based on this interpretation, we believe that the Kuril Ainu—referred to as *Hinomoto* during the Kamakura-Muromachi period—inhabited not only the Hokkaido mainland but also resided as Tsugaru Emishi in the northeastern part of Honshu until a certain historical era.

Having reached this point, we must now proceed further to examine the relationship between Tonchi or Koropokkuru—preserved in the oral traditions of the Hokkaido mainland and Sakhalin Ainu—and the Kuril Ainu.

According to Ainu oral traditions, the Koropokkuru fled into the hinterlands due to contact with the Hokkaido mainland or Sakhalin Ainu.
And according to the *Tatarafuki Dan*, the pit-dwelling dwarfs of Etorofu Island were dispersed nearly two hundred years ago due to the arrival of Hokkaido mainland Ainu from Kunashiri Island.
Where, then, did they flee to the hinterlands? Where did they scatter to?
This is the problem we are currently seeking to investigate.
In *Hen’yō Bunkai Zukō* (authored by Kondo Morishige in Bunka 3 [1806]), the words of Akkeshi Chiefs Ikotoi and Ichiyangemushi are recorded as follows:

The Kurumuse Emishi are descendants of Toichisekocchiyakamui.
According to ancient Emishi accounts:
In antiquity, there existed one called Toichisekocchiyakamui in the Emishi lands.
Their stature was remarkably short; all dwelled in pit dwellings.
As the Emishi territories were opened up, they gradually withdrew into deeper regions until at last, leading their entire tribe, they embarked on rafts and journeyed to Racco Island in the Eastern Seas, where they founded their settlement.
Moreover, in Kamchatka too there existed a variety of Kurumuse.

.”
Racco Island, according to Japanese understanding at that time referred to Urup Island; however, the Ainu maintained it lay east of Etorofu and Urup—an island visible from afar at sea on clear days.
In any case, it remained within the Kuril archipelago.
Regarding Kurumuse matters, *Ezo Kyūbun* (authored by Suzuki Yoshinori in Ansei 1 [1854]) records:

In antiquity there existed a people called Kurumuse in Emishi territories.
Later, as settlement expanded through Emishi lands and populations grew concentrated, these Kurumuse Emishi ultimately migrated to Kamushyatsuka where they established their community.
Consequently, Kamushyatsuka’s indigenous inhabitants showed no distinction from mainland Emishi in physical characteristics—both possessing black hair and eyes.

it states.
Considering these two accounts together, it goes without saying that Toichisekocchiyakamui was Koropokkuru, and we find this tribe recorded as having departed for the Kuril Islands and Kamchatka.
Indeed, the Kuril Ainu dwelt in pit dwellings, and according to Mr. Torii’s research, there were traces indicating they obtained glass bottles and fashioned them into arrowheads when the Russians first arrived.
That is to say, it was clear they existed in what we call a Stone Age state.
Moreover, when synthesizing facts such as the Kuril Ainu not preserving oral traditions about the Koropokkuru, that the Kuril Ainu must indeed be the Koropokkuru—as Dr. Koganei and other senior scholars had already advocated—was something anyone would readily acknowledge.

That is:

1. What Hakotoko termed the Togaru Emishi.

2. The so-called *Hinomoto* of the Kamakura-Ashikaga period.

3. The so-called Koropokkuru in the oral traditions of the Hokkaido mainland and Sakhalin Ainu.

4. The so-called Chupkaguru of the Hokkaido mainland Ainu—that is, the Kuril Ainu.

These four can be stated to be one and the same.

Herein lay the remaining problem: whether the Hokkaido mainland Ainu and the Kuril Ainu were of the same race or not.
To contribute to resolving this problem, we first had to examine the “Karako” among the three types of Emishi appearing in the *Suwa Engi*.

The *Suwa Engi* records a distinction among the Emishi, classifying them into three types: *Hinomoto*, *Karako*, and *Watarito*. It positions *Karako* between *Hinomoto* and *Watarito*. Furthermore, it does not mention that there exist other types beyond these. Now, that the Watarito are the acculturated Emishi dwelling in the southwestern regions, and that *Hinomoto* being Chupkaguru—that is, the eastern Emishi—is something we have already discussed. What, then, corresponds to the remaining *Karako*? It must be sought nowhere else but among the Ainu inhabiting the intermediate regions of Honshu and Sakhalin. The Hokkaido mainland Ainu and the Sakhalin Ainu both possess identical oral traditions concerning the indigenous people. Regarding their similarities and differences, beyond Dr. Koganei’s assertion that the two are identical (the Doctor also states that the Kuril Ainu are identical to them), to my limited knowledge, I have yet to hear many opposing theories. Therefore, following Dr. Koganei and adopting the perspective that the two are identical, we find that this view not only aptly applies to the *Suwa Engi*’s statement that “their lands connect to foreign countries,” but also that the name *Karako* seems largely explicable. *Research on Northern Barbarians* (*Research on Northern Barbarians* [Takahashi Kageyasu]) states:

To the north of Sōya in the Ezo region lay a vast land separated by a strait, which was called Karafuto.
Karafuto were Tang people.
The ignorant populace of our country had generally referred to foreign lands as *Kara*.
The term *futo* was a corruption of *hokujin hito* (northern people).
As for why it was called Karafuto, there were those who brought Han-made goods from there and engaged in trade with the Emishi of Sōya for many years.
The items they brought were so-called *dangire*, insect amber, smoking pipes, and various similar kinds.
They gradually spread these to the local region.
Since these were brought by people different from our Emishi kind, the merchants of Esashi and Matsumae, hearing this, came to vaguely call it Karafuto, until it finally appeared to have taken on the semblance of a place name for those northern barbarians.

it states.
The name Karafuto has become something that people today struggle to comprehend.
However, when considering the aforementioned explanation together with the name “Karako,” it appears quite reasonable.
Karako signifies Tang people.
Household members are called “house children,” and slaves are called Yatsuko (meaning “house children”)—both share the same underlying concept.
In ancient times, when referring to people with slight contempt, they appeared to use the term “child.”
Therefore, since the Gilyaks and others in the Karafuto region engaged in trade with the Santan region, bringing continental goods through the Hokkaido mainland Ainu to supply the Watarito or Japanese people, these individuals came to call the aforementioned Ainu “Tang people”—that is, Karako.
Later, this name fell into disuse on the mainland, surviving only among the Ainu who migrated to Karafuto thereafter, until it ultimately became the name of the island where they resided.

From these considerations, we came to understand that the Karako—comprising the Hokkaido mainland Ainu and Sakhalin Ainu—must have existed in a somewhat more developed state than Hinomoto, namely Chupkaguru, due to their positional relationships: bordering the Watarito to the west and interacting with the Santan people to the north.
They indeed obtained tools through trade with Japanese or Santan people, thus no longer needing to use stone tools and likely ceasing pottery production as well, becoming what we previously termed the unassimilated Emishi, thereby creating a great divide between themselves and the Tsugaru Emishi—that is, the Chupkaguru of Hinomoto—until finally being driven out as an entirely separate race called Tonchi or Koropokkuru.
To further allow us to expand our imagination, we could posit the hypothetical theory that the current Sakhalin Ainu, much like how the Hokkaido mainland Ainu migrated as far as Etorofu Island after driving out the Koropokkuru, themselves migrated from the mainland at a relatively later period, such that the relationship of the Sakhalin Ainu to the Hokkaido mainland Ainu would be analogous to that between the Kunashiri-Etorofu Ainu and the Hokkaido mainland Ainu.

In any case, there could be no doubt that a close relationship existed between the Sakhalin Ainu and the Hokkaido mainland Ainu. Herein, we returned to the previous problem and observed the relationship between the Hokkaido mainland Ainu and the Kuril Ainu.

Between the Hokkaido mainland Ainu and the Kuril Ainu, there existed a great divide at present.
The Hokkaido mainland Ainu had indeed driven out the Kuril Ainu as a different race.
(Or perhaps they had departed of their own accord.) However, if my theory—that the Hokkaido mainland Ainu were the unassimilated Emishi and the Kuril Ainu were the Tsugaru Emishi—was not mistaken, one could not imagine them as entirely separate races.
Moreover, given that the Kuril Ainu currently dwelled in pit dwellings and the Sakhalin Ainu—who should have been of the same kind as the Hokkaido mainland Ainu—were also in a state of pit dwelling, how could there not be a connection between them?
Furthermore, according to what Dr. Koganei cited, Mr. Schrenck’s work contained a report by a Russian named Golovin—who had been captured by Japanese people in Hokkaido a hundred years prior—stating that the Hokkaido mainland Ainu too had dwelled in pit dwellings during winter, just like the Sakhalin Ainu.
If this indeed proved to be the case, the relationship between them would seem to deepen even further.

Having reached this point, we could not help but further recall the considerable similarity between the Sakhalin Ainu's custom of dwelling in winter pit dwellings and the customs of the Emishi in Honshu during their subjugation by Yamato Takeru. According to the *Keikōki*, Takeuchi no Sukune investigated the eastern barbarians and reported back, stating: "Among the eastern barbarians lies the land of Hitakami. The men and women of this country all tie their hair in topknots and tattoo their bodies; as people, they are fierce and bold. All of these are referred to as Emishi." Later, Yamato Takeru subjugated the Emishi and reached this Hitakami Province. The same text records the customs of the Emishi of that land,

Among the eastern regions, the Emishi were the most formidable.
Men and women lived together, and fathers and sons were not separated.
In winter they lodged in holes; in summer they dwelt in nests.
They wore fur as clothing and drank blood; siblings distrusted one another; they climbed mountains like birds in flight and moved through the grass like wild beasts.
When receiving kindness, they would forget it; when perceiving a grudge, they would invariably repay it.
They would hide arrows in their topknots and wear swords under their clothing.

It states.
Though it remains difficult to believe these recorded Emishi customs fully reflected actual conditions during Yamato Takeru's era, they were undoubtedly accepted as truth among Japanese people at the time of the Nihon Shoki's compilation.
Furthermore, since the Japanese of that compilation period possessed substantial knowledge about the Emishi, their records—like Hakotoko's accounts—held credible value comparable to that found in the Illustrated Account of Northern Ezo...

The Emishi of this island included those who dwelled in pit dwellings when winter arrived.
However, this practice depended on regional temperature variations; not all Emishi engaged in it.
Those who inhabited pit dwellings did so out of sheer necessity, being unable to withstand the cold's severity.
Around October, when snowfall had already begun, they would construct these dwellings and enter them, remaining until the second or third month of spring when they emerged before the snowmelt to reside in their ordinary houses.

When compared with these accounts, the records concerning their dwellings seemed to tally exactly.

We believed that through the records of the *Hitachi Fudoki*, one could furthermore ascertain that the Emishi who inhabited the ancient Hitachi region had also dwelled in pit dwellings.

The Hitachi Fudoki states:

"In ancient times, there were the Mountain Saeki and Field Saeki of Kunisu (colloquially called Tsuchikumo or Yotsukahaki), who dug pit dwellings throughout the land and constantly resided in holes. When people approached, they would immediately retreat into their burrows and hide. When those people departed, they would emerge into the open fields and roam freely...."

Here, let us first compel ourselves to examine the name of the Saeki tribe.
When examining how scholars have ordinarily interpreted this matter until now, they often appear to regard Saeki as identical to Kunisu.
Such cases include Mr. Kobayashi citing this in his Tsuchigumo theory as evidence that Kunisu equates to Tsuchigumo (*Shigaku Zasshi* Volume 17, Issue 12, p. 47), and Mr. Yagi Shōsaburō proposing to name so-called Yayoi-style pottery as “Tsuchigumo pottery” or “Saeki pottery”—all of which exemplify this tendency.
However, upon closely examining this text, we find three other instances mentioning Kunisu and four instances mentioning Saeki within it, all clearly distinguished in their documentation without any conflation.
From the perspective of textual structure, if one were to regard Kunisu and Saeki as identical, then an author who goes so far as to annotate Kunisu with colloquial terms such as [so-and-so] ought to have included a note in the preface to their commentary along the lines of “alternatively called Saeki”—yet no such remark exists.
However, in this book, Kunisu and Saeki are recorded with equal treatment in their writing style.
One should discern the author’s intent.

Now, upon reconsidering, that the Saeki were Emishi had become beyond dispute through records such as the *Nihon Shoki Shinsenroku* and others.
That is to say, Yama no Saeki (Mountain Saeki) and No no Saeki (Field Saeki) corresponded to what appeared in the *Shoku Nihongi* as mountain barbarians (*yamamorodo*) and field barbarians (*tamorodo*).
Therefore, while an ordinary *fudoki* compiler would have simply written these as "Emishi" or perhaps "mountain barbarians" or "field barbarians," the author of the *Hitachi Fudoki* was quite the modernist scholar—one who frequently employed four-character phrases reminiscent of Chinese *fu* poetry in his prose, who went so far as to avoid the vulgar appellation "Tsuchigumo" in favor of the unconventional designation "Kunisu" (which would remain incomprehensible without annotation)—and thus must have deliberately eschewed the common term "Emishi" to instead use the name "Saeki."
That annotations were provided for Kunisu while none existed for Saeki stemmed from Saeki being distinct from Kunisu as the name of a renowned tribe already widely known among contemporaries.
Therefore, these accounts in the book were,

The elders said: "In ancient times, there were Tsuchigumo, Mountain Barbarians, and Field Barbarians; they commonly dug pit dwellings and constantly dwelled in them."

This should be translated as follows:
Hitachi bordered Mutsu to the north and was not far from the Emishi's dwellings at that time.
Therefore, the author of the Fudoki must have possessed considerable knowledge concerning the Emishi and recorded the circumstances of their pit-dwelling practices.
These accounts do not warrant placing full confidence in them.
Claims such as "blocking holes with thorns" amount to mere folktales fabricated to explain Ibaraki's toponymy and deserve no serious consideration. Similarly, descriptions like "when people came, they would hide in caves; when those people left, they would emerge into the outskirts and roam freely"—though Mr. Kobayashi rightly noted their ambiguity—merely characterize rat-like thieves who concealed themselves upon others' arrival and resumed mischief after their departure.
Furthermore, while Mr. Kobayashi doubted the Hitachi Fudoki's account of "digging holes to create fortresses"—arguing pit dwellings inherently lacked defensive utility—the Hokkaido Ainu indeed constructed such pit fortifications until relatively recent times, as evidenced by the so-called Chashi ruins still found throughout Hokkaido.
Therefore, the Hitachi Fudoki's accounts concerning pit dwellings need not be doubted.

Based on the above considerations, we believed that even in the Hitachi region there had been an era when the Emishi dwelled in pit dwellings, and consequently became inclined to accept that in the Ōu region too the Emishi had once inhabited such structures.
There existed one further fact compelling this conclusion.
When examining pit dwellings reportedly discovered along the Yoneshiro River in Ugo during Bunka 14 (1817), their pit depth, roof height and construction methods showed striking resemblance to Sakhalin Ainu dwellings documented in works like Hoku Ezo Zusetsu and Kyumeikoki.
The complete absence of nails or metal fittings along with walls and roofs made from split planks without saw cuts presumably indicated dwellings built by people unacquainted with metallurgy.
Among items found within stood a wooden rod over one shaku long bearing calendrical symbols and measurement markings.
These measurements closely matched Tang dynasty standards.
This object likely functioned as a calendar device obtained by the inhabitants from Japanese traders.
Though its exact dating remained unclear, only Emishi interacting with Japanese during Tang measurement usage could have occupied this area.
We therefore considered these pit dwellings Emishi relics though some might question their sophistication for supposed 'barbarian' housing.
Yet one could hardly imagine Tang-era Japanese builders achieving such construction without nails or saws.
Should this reasoning prove flawed, one must conclude Japanese built them after all.
But then they would necessarily have copied local Emishi designs for cold protection.
Thus from every angle it became clear that during Japan's Tang measurement period Emishi here maintained pit dwelling customs.
Critics might counter:
The Yoneshiro River dwellings could belong neither to Japanese nor Emishi but some unknown race.
Yet historical records revealed no peoples besides Emishi interacting with Japanese here during Tang measurement times.
This conclusion drew support not only from national histories but Heian legal scholars' commentaries.
The Fuyakuryō contains tax labor provisions for barbarian categories.
The Gikai explains: "Barbarians mean foreign tribes; miscellaneous categories denote barbarian subtypes."
Shūkai's commentary states: "'Barbarians' specifically indicates Eastern Barbarians; other types may be inferred by analogy."

“The term ‘miscellaneous categories’ refers solely to the miscellaneous categories among the barbarians.”
Furthermore, the ancient record they cite states:
“The term ‘various categories of barbarians’ refers to groups such as the Kebito, Hito, and Amamibito.”
And further, [they] posed the question of whether the “various categories of barbarians” constituted one or two [groups], stating: “They are fundamentally one but differentiated into two in their branches—for example, the ancestral lands of the Hayato and Kebito are what we call Emishi.”
Those who intermingled with Hua Xia are referred to as miscellaneous categories.
“One states: ‘They are one kind without distinction.’”
Although these explanations do not all say the same thing, in essence, they do not acknowledge the existence of other races that interacted with Japanese people in the Ōu region.
Therefore, the term “miscellaneous categories” likely refers to groups such as the Shamo people of Hokkaido during the Matsumae clan’s rule—those who intermingled with Japanese settlers. Consequently, it is difficult to recognize the aforementioned pit-style dwellings as relics of a people distinct from either the Emishi or the Japanese.

Therefore, we are inclined to believe that, barring any compelling counterevidence, the Hondo Ainu also originally existed in a state of pit dwelling, once used stone tools, and had been in the condition of what is called the Tsugaru Emishi—that is, the Hinomoto Emishi—namely, the Kuril Ainu.
If we posit that they, upon coming into contact with more advanced peoples, first became unassimilated Emishi—that is, Karako Emishi—and further transitioned to the state of acculturated Emishi, namely the Watarito, then the Kuril Ainu and Hondo Ainu—in other words, the Emishi and Koropokkuru—should be regarded as no more than distinctions in their degree of advancement, just as Hakotoko recognized Tsugaru Emishi and unassimilated Emishi as distinct subtypes within the same ethnic group.

The findings we obtained through research from historical records are as outlined above.
As for theories employing accounts such as the giant legend related to Ōkushioka Shell Mound in *Hitachi Fudoki*, or records in *Shoku Nihon Kōki* and *Sandai Jitsuroku* describing stone arrowheads raining upon Dewa—which assert that those who left these artifacts were not Emishi—such arguments merely demonstrated that Emishi of that era and region lacked these customs. This point had already been conclusively refuted by Dr. Hamada Kōsaku (*Journal of the Anthropological Society*, Issue 200, p.70), leaving us no need to expand upon his argument.

The above constituted the views we had formed primarily based on records.
As a result of archaeological research—were all Stone Age ruins across the nation truly remnants left by the same race?
Was there room to acknowledge that ruins of other ethnic groups were intermingled, or whether hope existed to distinguish these ruins as being limited to certain regions? Ascertaining such facts was of utmost importance for understanding the migratory directions and distribution patterns of indigenous tribes; however, this research naturally belonged to a separate inquiry.
For now, we confined ourselves to distinguishing between the Emishi and the Koropokkuru and laid down our brush.

(Meiji 40 [1907], February 25th Draft)