What Made a Place Part of Japan?

For three days in August 1875, representatives of the Ryukyu Kingdom argued with a Japanese official about a relationship they were being ordered to end. Matsuda Michiyuki demanded that Ryukyu break its ties with Qing China. The Ryukyuan representatives refused. A damaged record of the meetings, preserved from the Ryukyuan side, survives in the University of the Ryukyus Library.^[1]

The Japanese government had already designated Ryukyu a han, or domain, in 1872. Yet a designation issued in Tokyo had not resolved who could conduct the kingdom's external affairs. The argument continued because the two sides did not accept the same consequences of that designation.^[2]

A present-day outline of Japan makes such uncertainty difficult to see. We recognize islands, then tend to imagine a country filling them. The historical problem runs the other way: how did particular authorities establish what they could demand from particular places and people?

Three settings make the question manageable: the ancient court's northeastern frontier on Honshu, the Tokugawa order's relationships with Ainu communities and Ryukyu, and Meiji incorporation in the north and south. They are not a complete territorial history. Together, they show why a military foothold, a claim of subordination, and an administrative boundary cannot simply be treated as different names for membership in the same country.

In 762, a stone inscription at Taga, near present-day Sendai, recorded distances from the government establishment there. The imperial capital was 1,500 ri away; the boundary of the Emishi country was 120. Other entries located the boundaries of Hitachi and Shimotsuke provinces. The inscription also commemorated Taga's establishment in 724 and its subsequent reconstruction.^[3]

The capital in this account was Nara, not Kyoto. More important than converting the old distance units into modern kilometers is the difference between the places being measured. Some belonged to a provincial geography through which the court organized government. The Emishi boundary marked another relationship.

Emishi was a designation used in the court's description of northeastern peoples. The inscription's expression “Emishi country” should not be turned into proof that one unified state stood beyond a surveyed international frontier. What the stone establishes is that the officials who commissioned it distinguished an area beyond their provincial order while locating that area in relation to their own headquarters.^[3]

This was a frontier within the main island, not a national boundary conveniently supplied by the sea. Nor was it a division between inhabited Japan and vacant land. People already lived on both sides of the court's expanding reach. The city’s historical account places the early province of Mutsu in a more limited part of the northeast and identifies Taga as a base for bringing additional territory under the statutory government.^[4]

The inscription makes the headquarters the point from which relationships can be stated. It joins distance to institutional memory: this establishment was founded by named officials, rebuilt under named authority, and positioned in a larger political world. Its words present the court's ordering of that world. They do not contain the response of everyone living in it.

That asymmetry matters. A frontier can be described confidently by the power seeking to extend itself even when its authority remains uneven on the ground. Reading the inscription as an official claim, rather than the voice of the landscape, preserves both what it tells us and whose perspective it preserves.

Taga was more than an inscription announcing possession. Archaeological investigation identifies a formal government precinct, working offices and associated residential areas. A temple stood outside the enclosure, while settlement developed to its south and west. This was an administrative and military establishment around which an inhabited center formed.^[5]

The combination changes what incorporation means. A commander could pass through a district. A government that expected to remain needed places where officials could receive business, maintain supplies and arrange continuing service. An installation became consequential through the repeated activity it supported.

The distinction is visible in Taga's formal center and the less ceremonial spaces around it. The court could make its authority recognizable in buildings, but buildings also concentrated people able to carry out its demands. Establishing an office in the northeast shortened the institutional distance between an order and the people expected to answer it. The capital did not need to handle every local matter through a fresh expedition.

Settlement around the compound further complicated any simple inside-and-outside picture. The archaeological account describes an urban environment connected with the government establishment, not merely an isolated military box. Administration acquired neighbors, local requirements and a material presence that could persist between individual commands.^[5]

This did not make expansion consensual. It identifies the additional work involved in turning a foothold into a continuing provincial center. The distinction between conquest and administration is not a distinction between force and its complete disappearance. It is a distinction between different ways in which authority becomes capable of recurring.

Taga's later history also resists a picture of institutions advancing forever without changing their form. The city’s archaeological summary places the end of its principal role as a provincial headquarters around the middle of the eleventh century. Later references to a Taga provincial government cannot automatically be located in the same ancient compound.^[4]

A place could therefore lose a particular administrative establishment without the surrounding region returning to the political position represented by the inscription of 762. The fate of one headquarters was not the fate of every relationship it had helped create. Territorial incorporation, once developed through institutions and settlements, could outlast the original premises through which it had been organized.

Move forward to the Tokugawa period, when a warrior government based in Edo presided over numerous domains. The northern problem was no longer the same as the ancient court's Mutsu frontier. Ainu communities had their own histories and connections, extending beyond the island later named Hokkaido. They must not be treated as simply the people on the other side of Taga's stone, unchanged across a thousand years.^[6]

Historian Arano Yasunori emphasizes that the Tokugawa government's external relationships were administered through differentiated arrangements. Matsumae handled the northern connection; Satsuma handled Ryukyu. The responsible lords received privileges but also obligations. Their position depended on managing a relationship, not merely occupying an interchangeable segment of a national perimeter.^[7]

For Matsumae, this distinction was substantial. David Howell's account describes seventeenth-century military subordination followed by the growing effects of commercial fisheries on Ainu livelihoods. Trade and wage labor increasingly bound communities to Japanese enterprises during the eighteenth and nineteenth centuries.^[6] Political and economic intervention preceded the Meiji government's renaming of the region.

Yet growing dependence did not make Ainu communities ordinary villages within a uniform administrative system. Arano's interpretation specifically warns against projecting modern territorial categories backward onto the northern and southern relationships. An authority could assert superiority over people whom it continued to classify as belonging to a different realm.^[7]

That arrangement contained a practical advantage for an intermediary. Where relations required a recognized broker, the broker's special position remained valuable. Replacing every differentiated relationship with direct administration was therefore not the only imaginable way to exercise power. Maintaining a distinction could help sustain the office that claimed to manage it.

This is an interpretation of the institutional arrangement, not a claim that every Matsumae official followed one unchanging strategy. The narrower point is sufficient: subordination and administrative sameness were different conditions. People could experience force, commercial dependence and official intervention without being governed through precisely the same machinery as inhabitants elsewhere.

The contrast with Taga is useful. In one setting, a provincial establishment supplied a means of extending the court's working presence. In the other, a lord's privileged management of relationships helped connect the northern world to the Tokugawa order. Calling both “Japanese control” is only a beginning. It leaves unanswered what the controlling authority actually controlled.

Ryukyu makes the difference especially concrete because its royal government continued after military defeat. Satsuma invaded in 1609. The kingdom subsequently operated under Satsuma's domination while maintaining relations with China. Amami, however, came under direct Satsuma administration. The same campaign thus produced different governing arrangements in different islands.^[8]

The distinction prevents two opposite mistakes. The survival of a king did not mean the absence of Japanese coercion. Conversely, Satsuma's intervention did not immediately turn the remaining kingdom into an ordinary district governed without its own court or external relationships.

Nor was Satsuma's authority only a distant ceremonial assertion. In 1628 it established a resident office in Naha. The local historical record describes officials dealing with affairs between Satsuma and Ryukyu and supervising trade. The office continued until the administrative changes of 1872.^[9]

A royal government and a supervising external office could therefore occupy the same political landscape. Their coexistence was not a failure to decide where either building stood. It was part of the way authority had been distributed.

For the royal government, relationships with China remained a component of governing as a kingdom. For Satsuma, the continuing Ryukyuan institution existed within a subordinating relationship. Neither description can be replaced by imagining two modern states jointly administering an overseas province on identical terms. The forms of authority were different, and their implications could be disputed.

The physical history of Satsuma's Naha office offers an unusually clear bridge to the next period. After the establishment of Ryukyu han, the premises served Japanese central-government offices; after the creation of Okinawa Prefecture, they became an early prefectural seat.^[9]

That continuity of working ground highlights a change in institutional purpose. An office maintaining a lord's oversight of a kingdom and a prefectural office administering territory directly for the central government could occupy the same address. A photograph of the premises, even if one existed for each stage, would not by itself explain the difference.

This is also why the arguments of 1875 cannot be dismissed as a delayed acknowledgment of something fully settled in 1609. The earlier arrangement had allowed a subordinated kingdom to continue particular relationships. The Meiji demand required those relationships to end. It altered the meaning of incorporation rather than merely repeating an old claim more loudly.

In 1869, the Meiji government renamed Ezochi Hokkaido and designated eleven provinces and eighty-six districts. These were administrative names imposed on a region with existing inhabitants, place-names and earlier relations with Japanese authorities. The decree did not create the island or begin its human history.^[10]

The naming formed part of a larger undertaking. The Kaitakushi, established that year, combined the functions of a central development agency with responsibility for local administration. Hokkaido's archives describe its work in attracting and settling migrants, building infrastructure and promoting industries. Its position differed from both an ordinary domain and a single provincial office.^[11]

This combination helps explain why renaming mattered without making it sufficient. A common set of territorial categories gave officials units within which to organize business. An agency with staff and resources could pursue settlement and production within those categories. The written designation and the administrative organization addressed different requirements of the same project.

They also directed attention toward particular futures for the land. Where a development agency promoted farms and industries, officials had to decide which uses would receive support and which claims would be recognized. The region was not simply being added to a list of places. It was being reorganized through decisions about how people would live and work there.

The agency itself was temporary. When it ended in 1882, local administration passed to the prefectures of Sapporo, Hakodate and Nemuro, while industrial responsibilities went to a separate central-government organization.^[11] Ainu communities did not cease to exist during these changes, and the end of the agency did not undo the territorial project.

There is a revealing parallel with Taga, but not an identity. Both settings involved establishing institutions able to make authority operate at a distance. The Meiji agency additionally made large-scale settlement and economic development explicit parts of its assignment. Its eventual division of responsibilities shows how the administrative machinery could change while the broader claim continued.

Hokkaido did not follow Ryukyu's route. There was no single Ainu monarchy equivalent to the court at Shuri whose removal could stand for the whole transformation. The northern changes reached communities through administration, settlement and the regulation of livelihoods. In the south, abolishing a continuing royal government was itself a central act. Incorporation had no single institutional recipe.

The consequences for Ainu households become particularly visible in the Hokkaido Former Aborigine Protection Act of 1899. The English title translates the statute's historical classification; it is not a description adopted here for the people concerned. Its provisions identified a population within Japanese administration and subjected that population to a distinct set of rules.^[12]

Article 1 authorized grants of land to households already engaged in farming or wishing to begin. Article 2 limited what recipients could do with that land: transfer other than by inheritance was forbidden, as were mortgages and certain other property interests. Article 3 required the confiscation of granted land still uncultivated after fifteen years.^[12]

These clauses make the arrangement more precise than the statement that the government “gave the Ainu land.” Eligibility depended on an approved occupation. Receiving ownership did not confer every power ordinarily associated with disposing of property. Keeping the grant depended on carrying out a specified transformation of the ground.

The statute also distinguished newly granted land from land already owned. Its restrictions on earlier holdings included provisions allowing some transactions with the Hokkaido governor's permission. That distinction is important: the text itself does not describe all Ainu property as something created for the first time by this law.^[12]

Read together, the clauses show administration entering the relationship between a household and its resources. The central issue was no longer whether an external lord could claim a privileged northern connection. Officials could define a permitted path to ownership, set conditions for continued possession and decide when permission was necessary.

Howell interprets the agricultural emphasis as an effort to remake Ainu livelihoods within the state's preferred model of social membership. His study also reports that by 1909 land-receiving households had generally obtained less than the statutory maximum.^[6] The legal ceiling cannot be used as though it were the actual allotment of every family.

The distinction between territorial inclusion and equal treatment is therefore visible in the law itself. A population could be brought within a state while continuing to be singled out by that state's administration. Becoming governable as part of Japan did not erase every internal boundary; it could relocate the distinction into property rules and official classifications.

Nor does adopting a livelihood or accepting a grant settle a person's identity. The statute records the government's categories and conditions. It cannot tell us, without other evidence, how every household understood its own membership, choices or future. An official answer to “who is inside?” could remain different from the answers of people living there.

Return to the southern negotiations. In 1872 the Meiji government had designated Shō Tai as ruler of Ryukyu han and incorporated him into the Japanese nobility. The title altered the government's formal description of the relationship. It did not, as the surviving 1875 record shows, secure agreement to all the demands that followed.^[1][2]

In March 1879, Matsuda returned with armed support. At Shuri he announced the abolition of the domain and the establishment of Okinawa Prefecture. His demands included the departure of Shō Tai for Tokyo, surrender of the castle, and transfer of the kingdom's land, people and documents. On April 4, the change was proclaimed nationally.^[13]

The documents belong beside the castle in that list. Taking over government required more than occupying a recognizable seat of authority. Records were resources through which the successor administration could identify and manage what it claimed. A command to hand them over exposes the practical distance between announcing incorporation and assuming the business of rule.

Across the cases, “part of Japan” consequently describes several changes that must be dated and explained separately. A provincial center could extend the court's working presence. A domain could dominate a relationship while preserving a differentiated political counterpart. A national government could replace that counterpart, or impose household-level rules within territory already claimed.

These were not successive steps taken by one population toward a predetermined destination. The institutions differed, and so did the people confronting them. Nor can a country's later territorial outline make those earlier decisions inevitable or supply the consent missing from their records.

What made a place part of Japan was the establishment of particular claims and governing arrangements, backed by the means to sustain them. What that membership meant depended on which authority was acting, which powers it acquired and which distinctions it retained. The argument in 1875 survived precisely because a name issued by one government had not yet become an accepted answer for the other.