Foreign Relations in Early Modern Japan: Exploding the Myth of National Seclusion

Until relatively recently, the Edo period (1603–1868) was so inextricably associated with the concept of national seclusion—sakoku in Japanese—that most people familiar with the outlines of Japanese history would have regarded the title of this article as an oxymoron. Nagasaki, famous for the Dutch factory(*1) on Dejima, was widely regarded as early modern Japan's one and only portal to the wider world. But our understanding of Japan's interaction with the rest of the world during this period has changed dramatically in the past two or three decades. Today most Japanese historians regard the Edo period as a time when Japan maintained active, if indirect, ties with the larger world through not one but four portals, under a system that allowed the nation to develop and eventually emerge as a modern state.

The term "four portals" (yottsu no kuchi) is now virtually standard terminology in Japanese history textbooks, but it was a foreign concept when I introduced it some 30 years ago in an effort to overturn the prevailing image of early modern Japan as a country shut off from the rest of the world. Such a basic change in premise necessitated a complete reexamination of foreign relations in early modern Japan, including the role played by Nagasaki. In 1983, I proposed that the concept of sakoku be replaced with the concepts of kaikin (maritime restrictions) and a Japanese version of ka-i chitsujo (Sinocentric world order). More recently I have analyzed how the notion of closed and open countries took hold and helped shape the identity of the Japanese in the modern era.

In this way, the "four portals" concept became the springboard for a wholesale deconstruction of early modern foreign relations that has occupied me for almost 35 years.

In the following, however, I will restrict myself to a fairly general overview of Japan's foreign relations in the early modern era via the four portals, together with a discussion of Nagasaki's changing role as a center of Chinese and Dutch trade during the Edo period.

Overview: Foreign Relations through the Four Portals

To assist in visualizing this system, I have prepared two diagrams illustrating Japan's interaction with other countries during the Edo period. Both deal with the period traditionally associated with "national seclusion" in early modern Japan: from 1641, when the Dutch trading post in Hirado was transferred to the island of Dejima off Nagasaki, to 1853, when the arrival of Matthew Perry's ships forced a major shift in the shogunate's foreign policy.

Figure 1 provides a schematic overview of Japan's political (primarily diplomatic) relations with neighboring states and peoples in East Asia during the period. In addition to China and Korea, these neighbors included (clockwise from the top) Santan, the Japanese designation for a Northeast Asian coastal people (predominantly Orok but also Nivkh and Oroqen) engaged in maritime trade; Ezo, the land of the Ainu, covering all but the southernmost area of present-day Hokkaidō; and the kingdom of Ryūkyū (present-day Okinawa). Ezo and Ryūkyū, though they later became integral parts of Japanese territory, were at the time still regarded as foreign lands, inhabited by non-Japanese peoples with their own national identity. This understanding is basic to my theory of the four portals.

Nagasaki and the Three "Gateway Daimyō"

As I first explained in 1978, each of the four portals played an integral, structural role in the conduct of Japan's foreign relations during the early modern era. In essence, the Tokugawa shogunate delegated the conduct of international affairs to the daimyō of three outlying domains—Satsuma in southern Kyūshū, Tsushima off the northwestern coast of Kyūshū, and Matsumae in southern Hokkaidō—and the special shogunal trading city of Nagasaki. Each of the three "gateway daimyō" had exclusive jurisdiction over relations with a particular country or territory: The Shimazu clan of Satsuma was in charge of relations with Ryūkyū;  the Sō clan of Tsushima handled affairs with Korea; and the Matsumae clan of the Matsumae domain had jurisdiction over dealings with Ezo. Nagasaki differed somewhat in that it had official jurisdiction over foreign relations in general as the seat of the Nagasaki bugyō (governor), a key shogunal official in charge of port administration.

At the heart of these arrangements was the feudal relationship between lord and vassal. As vassals to the shōgun, Nagasaki and the three daimyō oversaw relations with foreign entities as part of the military service they owed the shōgun. As compensation, and to support these functions, they were granted trading monopolies, which were regarded as a special type of fief (more commonly granted to the shōgun's vassals in the form of domains). In this sense, foreign trade was a privilege. But because it helped supply goods that played an important role in Japanese society, it was also regarded as a service; the proof of this is that the shogunate was known to issue reprimands when trade was languishing. In the following, we will examine Japan's relations with the rest of the world via each of these portals.

Relations with Korea and Ryūkyū

Korea and Ryūkyū were considered vassal states of China, and as such they sent regular tribute missions to the Qing emperor. But the Joseon kings of Korea also sent missions to the Tokugawa shogunate—referred to as tsūshinshi (diplomatic missions) from the fourth mission on—and the kings of Ryūkyū sent similar emissaries under the names of  shaonshi (gratitude missions) and keigashi (congratulatory missions). Altogether, the Tokugawa shogunate received 12 missions from Korea (beginning in 1607 and ending in 1811) and 18 from Ryūkyū (from 1634 to 1850). The shogunate sent no direct envoys to either kingdom, although the Tsushima domain did send envoys, to Korea in the shōgun's name.

The expense of feeding, lodging, and protecting the missions from Korea and Ryūkyū  was borne by the shogunate and the relevant daimyō. Japan also paid all expenses associated with the care and return of shipwrecked mariners from either country. Likewise, Korea provided money and food for Tsushima's envoys, domiciled in Busan, under various budgetary headings. Together with Japanese merchants in Korea, these envoys stayed in a special compound known as the wakan, built by the Korean government. As this suggests, bilateral ties in every area were built on the principle of mutually beneficial, reciprocal ties of friendship between national sovereigns (kokuō, referring to the Tokugawa shogunate as well as the king of Korea).(*2) This type of relationship was later classified as tsūshin (roughly, diplomacy), the same term applied to the ties established with the Western nations via the amity and commerce treaties concluded in 1858.

Relations with the Chinese and the Dutch

Relations with China and the Netherlands via the Nagasaki portal were relegated to the "private sector" to be conducted between the merchants of Nagasaki and their Dutch or Chinese counterparts without direct government involvement. Nonetheless, from the 1630s, the director, or opperhoofd, of the Dutch factory on Dejima was required to make the long journey to Edo once a year, ostensibly to thank the shōgun for the privilege of trading at Nagasaki. From the shōgun's perspective, these visits had the same status as the New Year's courtesy calls paid by officials from Edo, Osaka, and other cities directly under the control of the shogunate. This meant that the costs were shouldered entirely by the Dutch, who were furthermore expected to bring valuable gifts for the shōgun, his immediate family, and his senior advisors and ministers.

Chinese merchants were not required to send yearly emissaries to Edo,(*3) but they were expected instead to lavish expensive gifts on the Nagasaki bugyō and other local officials when they arrived in the port. In addition, a representative of the Chinese factory was expected to convey his gratitude to the shōgun by way of the Nagasaki bugyō each year on the first day of the eighth month, an important holiday for the Tokugawa clan.

In other ways as well, the Dutch and Chinese paid heavily for the privilege of trading at the port of Nagasaki. They were charged rent for their lodgings in their designated compounds and were required to pay a variety of taxes and other charges on incoming and outgoing cargo. They also paid for all food and supplies needed to carry on daily life in the factory, as well as the services of local prostitutes. In addition, the Dutch and Chinese (unlike the Koreans and Ryūkyūans) were charged for the care and return of shipwrecked mariners.

Both the Dutch and Chinese factories were built by order of the shōgun at the expense of  Nagasaki's citizens, to whom the Dutch and Chinese paid their annual rent. (The initial construction of Dejima was financed directly by 25 local wealthy merchants, while building of the Chinese factory was carried out with a loan from the shogunate.) As the "landlord" that rented out these facilities, the city of Nagasaki also bore responsibility for their supervision and control; specifically, each factory was assigned an otona (supervisor), as well as a crew of official interpreters.(*4) While the Nagasaki bugyō, as the representative of the shōgun, had nominal oversight over Nagasaki's Chinese and Dutch trade, business affairs were basically in the hands of the Chinese and Dutch traders on the one hand and Nagasaki's privileged merchants on the other, with the city of Nagasaki functioning as a kind of intermediary. Under the conceptual framework later developed by the shogunate, these arrangements (closely resembling the Canton System developed by the Qing government to control trade with the European nations in the second half of the eighteenth century) fell under the category of tsūshō, or commerce.

Relations with the Ainu

The Ainu people inhabiting the land then known as Ezo did not send envoys to the shōgun in Edo. In place of such missions, they carried out two ceremonies: the uimamu, in which a representative of the Ainu paid his respects to the lord of the Matsumae domain in southern Hokkaidō each year; and the omusha, a tribute ceremony carried out at the border between Ezo and Japanese territory with one of the shogunate's traveling inspectors in attendance. The rationale for this arrangement was that Ezo was a "land with no sovereign," meaning that, in the shogunate's view, the inhabitants had no political sovereignty of their own. The rituals were regarded as an expression of thanks for the privilege of "living under the protection" of the shogunate in the person of the Matsumae daimyō, a relationship referred to as buiku.

The Formation of a Japanocentric Order

The system outlined above, whereby Japan used four portals to carry out three categories of foreign interaction (tsūshin, tsūshō, and buiku) was the basis for a Japan-centered regional order that gradually took shape in the early modern era—in essence, a Japanese version of the Sinocentric world order, with the Tokugawa shogunate at the summit.

The resumption of official relations with Ming China was the single biggest foreign-policy challenge facing three successive Japanese rulers spanning the late sixteenth and early seventeenth centuries: the warlord Toyotomi Hideyoshi, the first shōgun Tokugawa Ieyasu, and Ieyasu's successor Tokugawa Hidetada. Indeed, Hideyoshi's invasions of Korea (1592–98), the arduous peace negotiations following that ill-fated adventure, and the 1609 invasion of Ryūkyū by Satsuma forces were all aimed at this larger goal. However, by the early 1620s, the Japanese government was forced to recognize that all its efforts to resume ties with China had been in vain. Working from the premise of this failure, and from the relationships it had rebuilt with other regional countries through trial and error, it turned its attention to constructing a hierarchical regional system modeled on the Sinocentric model, but with the shogunate at the center. This process (which can only be narrated here in its bare outlines) began in the second half of Hidetada's rule and continued under the third shōgun, Tokugawa Iemitsu, spanning such events as the suppression of the Shimabara Rebellion (1637–38), the ban on Japanese maritime travel in the China seas (1635), the exclusion of Portuguese ships (1639), and the fall of the Ming dynasty (1644). After the defeat of the remaining loyalist Ming rebels at the hands of Qing forces in 1684, this Japan-centered system took root as a fact of international society in East Asia. The establishment of this order heralded Japan's political independence from China, the first step in the process by which the Japanese state achieved full autonomy in the course of the early modern era.(*5)

To be sure, it was not until Japan came under pressure from the Western powers to deregulate its trading system that the shogunate explicitly adopted and set forth this formal classification of foreign interaction based on the three categories of tsūshin, tsūshō, and buiku, and some have argued as a consequence that the entire system I have described, based on four portals and three classes of interaction, did not exist before that time. But the thinking underlying this system had existed from long before the early modern era. Indeed, it constituted the foundation—the regional grammar, as it were—on which international relations were traditionally built in East Asia. The striking similarity between the system developed by the Tokugawa shogunate and the Canton System by which China sought to control trade with the countries of the West attests to the common grammar on which both systems were built.

Kaikin in Japan, Korea, and China

The centerpiece of the Japanese policy that has been erroneously called "national seclusion" was kaikin, that is, government restrictions on maritime travel and trade. As the table below indicates, China, Korea, and Japan all had comparable restrictions in place, although they differed in their particulars. In fact, the Japanese borrowed the term kaikin from the Chinese haijin, which in turn derived from a provision in the Ming Code.

Japanese, Chinese, and Korean Regulations on Maritime Activity

Notes: Information on Korea is from Taejon hoetong (Institutes of the Korean Government); for Japan, Tsūkō ichiran (Survey of Foreign Intercourse), etc. For China, I have relied on a secondary source in the form Masui Tsuneo's article "Kaikin" in Shukusatsu Tōyō rekishi daijiten (Abridged Dictionary of Oriental History) (Rinsen Book Co.), which in turn cites the Da Qing huidian (Collected Statutes of the Great Qing Dynasty) and Da Qing huidian shili (Statutes of the Great Qing Dynasty with Precedents).
Source: Arano Yasunori, Kinsei Nihon to Higashi Ajia (Early Modern Japan and East Asia), University of Tokyo Press, 1988.

The Ming system, which prohibited private citizens from overseas travel or interacting with foreigners, had been effective in controlling the so-called Wokou pirates in the fourteenth century, but that was not its sole purpose. In the Ming Code, the kaikin statute comes under the heading of guanjin, that is, laws pertaining to the control and governance of maritime and land transportation. This grouping confirms the idea that the law was intended to control the flow of people, goods, and information in and out of the country as a basic prerogative of the state (or the head of state), on the principle of "no diplomacy between subjects." While this principle originated in China, it was also operative in Japan from ancient times and should be considered a basic element of the common "grammar" governing international relations in East Asia. The kaikin policy seen throughout East Asia is a systemic manifestation of that concept and thus fundamentally different in its historical and traditional significance from the Dutch meaning of national seclusion on which the term sakoku is based.

Of course, kaikin encompasses a broad range of forms and historical phases, including the situation in medieval Japan, where the government's prerogative was officially espoused but practically ignored; the early Qing government's policy requiring coastal residents to move inland (the Great Clearance, 1661–83); the Japanese and Korean practice of using a few "portals" to strictly control movement in and out of the country; and the more sophisticated customs and border protection systems employed by modern states in the region.

Be that as it may, at the end of the seventeenth century, the rulers of Japan, Korea, China, and Ryūkyū all enforced kaikin policies that banned or strictly controlled the private movement of people and goods in and out of the country, while simultaneously building a network of official ties—in essence, international relations. Through the exchange of diplomatic missions, trade in goods, agreements regarding the return of shipwrecked mariners, and other understandings, this system functioned effectively to avert serious conflicts between states, as indicated by the long peace that persisted in East Asia until the second half of the nineteenth century.

The Four Portals and the Evolution of a Japanese Trading Area

Figure 2 provides a schematic overview of foreign trade under the four-portal system. To the east, west, and south of the four portals we can see the region's three major markets: Japan, China, and Southeast Asia, centered in Batavia (the capital of the Dutch East Indies, corresponding to a part of present-day Jakarta). The key role performed by the Chinese junks and Dutch trading vessels that called at the port of Nagasaki was to link these three markets directly and indirectly.

Chinese Junks at Nagasaki

Until around 1685, the Chinese junks that entered the port of Nagasaki were all trading without the authorization of the Chinese imperial court, and were thus viewed by the Chinese government as pirates; this was the main reason the Tokugawa shogunate relegated relations with the Chinese to the realm of private commerce. (This situation fueled the rise of Chinese merchant settlements in Kyūshū, discussed below.) Moreover, in the second half of the seventeenth century, the Chinese trading fleet at Nagasaki consisted mainly of 20–30 junks under the control of the Taiwan-based Ming loyalist regime of Zheng Chenggong (Koxinga), which carried on the Han Chinese manners and customs familiar to the Japanese. The last of these loyalist forces surrendered to the Qing in 1683, and in 1684 the Qing government scrapped the Great Clearance. From the following year, the number of Chinese trading vessels soared to more than 100, and the Chinese merchants visiting Nagasaki bore the stamp of the Manchu dynasty that had come to control China. In response to this change, the authorities established trade quotas and built a separate compound for resident Chinese merchants. Over the next few decades, however, rampant smuggling and unregulated trade caused profits from trade at Nagasaki to fall off dramatically, and the city fell into decay. The situation was finally salvaged and relations with China stabilized thanks to Arai Hakuseki's Kaihaku Goshi Shinrei (New Regulations on Ships and Trade) of 1715, recognized by the Qing government in 1716.

The goal of the new policy was to simultaneously salvage the dire trade situation at Nagasaki and rebuild the city itself. To maintain the trade system on which the city of Nagasaki depended, and vice versa, Arai Hakuseki decided to ensure its sustainability by keeping the volume of imports to a level consistent with Japan's export potential. Thanks to this policy, both the trading at Nagasaki and the government's system for managing and controlling foreign relations functioned smoothly until the final years of the Tokugawa shogunate.

The ships allowed into the port of Nagasaki consisted of tōsen, or Chinese junks, and orandasen, or Dutch ships. Tōsen were further divided into three categories based on the distance they traveled: the kuchibune, or short-distance ships, originated from nearby Jiangsu and Zhejiang in eastern China; the naka-okubune, or mid-distance ships, sailed from Fujian, Guangdong, and Guangxi; and the okubune, or long-distance ships, set out from Southeast Asian countries such as Vietnam, Cambodia, and Thailand. As this suggests, the term tōsen was used rather broadly, to cover all foreign ships of East Asian origin. In fact, the orandasen generally set out from Batavia (Java), navigating to Nagasaki directly or sometimes by way of other Southeast Asian ports. Thus, their range overlapped closely with that of the okubune.

Trade with Korea, Ryūkyū, and Ezo

Trade with Korea, Ryūkyū, and Ezo was of two basic types. One involved the direct exchange of any of these regions' products for commodities produced in Japan. The other involved the use of these regions as intermediaries for the exchange of Chinese and Japanese commodities. In other words, these three portals were more than a conduit for imports from Korea, Ryūkyū, and Ezo; they also facilitated the import of Chinese goods, such as silk, and gave the Chinese market better access to Japanese commodities, such as silver. In this way, the Chinese and Japanese markets were linked not only by the port of Nagasaki but also by these other portals via Ryūkyū, Korea, and Ezo.

Let us briefly examine the development of this regional trading system. An East Asian trading network centered on the exchange of Chinese silk for Japanese silver sprang up by the beginning of the sixteenth century, as a result of economic development in the region. At the same time, the tributary system by which the Ming government sought to control overseas trade grew increasingly dysfunctional, until in 1557, officially sanctioned trade between China and Japan came to a halt. This suspension of official relations left a vacuum that was quickly filled by indirect trade (via Korea, Ryūkyū, etc.) along routes already established by this time, by pirates and other unauthorized private traders, and by Portuguese and Spanish trading vessels.

The pirate raids continued to escalate, until in 1567 the Ming government relaxed its kaikin policy and began to allow private Chinese trading ships to travel to Southeast Asia. As travel to Japan was still prohibited, ports in Taiwan, the Philippines, and other parts of Southeast Asia soon emerged as centers for the exchange of Chinese raw silk and Japanese silver. Japanese traders traveled to these ports to do business, and eventually Japanese settlements sprang up. In the early seventeenth century, Tokugawa Ieyasu moved to secure and control these routes through the "red seal" system, under which the shōgun granted exclusive trading privileges in the form of permits bearing a vermilion seal.

Meanwhile, merchants of Chinese origin had been settling in Japan to trade for silver in defiance of the Ming government's prohibition. This led to the growth of numerous Chinese settlements and neighborhoods known as tōjin machi (Chinatowns), particularly in Kyūshū (figure 3). Indeed, the immigration of Chinese merchants was a driving force behind the development of Nagasaki itself, even though the immediate impetus was the opening of the port to Portuguese ships in 1571.

In 1635, the Tokugawa shogunate instituted a policy restricting the entry of Chinese nationals and ethnic Chinese to the port of Nagasaki as part of a sweeping effort to consolidate control of the country. After that, Nagasaki siphoned up the population and functions of Kyūshū's smaller tōjin machi, and the city emerged as the key portal through which the shogunate managed trade and other foreign relations via the China seas.

The potential of the Japanese market spurred intense rivalry among the various powers initially involved in the entrepot trade in the China seas. By restricting access to Nagasaki to Chinese and Dutch traders the shogunate hoped to secure a steady supply of imports while averting international conflicts, shutting out Catholic influences (deemed subversive), and ensuring loyalty to the shogunate. The red-seal trade was abolished at the same time out of concern that it provided an opening for the spread of Christianity and had the potential to spark international conflicts.

Almost all of the documents from this era that historians have traditionally referred to as "seclusion edicts," or sakoku rei, were in fact administrative directives addressed by shogunal ministers to the Nagasaki bugyō. In this context, it becomes clear that what was previously interpreted as a nationwide prohibition on overseas travel by Japanese was intended to apply only to the area under the jurisdiction of the Nagasaki bugyō (i.e., around Southeast Asia). This interpretation is bolstered by the fact that upwards of 500 Japanese could be found residing in the wakan at Busan at any given time during the Edo period, and considerable numbers were recorded in Ezo and Ryūkyū.

Japan as an Independent Trading Area

The composition of Japan's imports and exports during the period under consideration changed over time (see figure 2, under the heading "Goods traded"). The first major shift occurred as a result of domestic economic development and the decline in Japan's silver production. Since the export of silver was suspended, Japan was unable to import such luxury goods as premium-grade silk and ginseng from China. The focus of Japanese exports subsequently shifted to copper and (in the case of China-bound exports) marine products, while the composition of imports shifted to more widely accessible consumer goods, such as sugar and medicinal substances. This shift put an end to the trade structure that had defined international relations in East Asia since the early sixteenth century.

The second shift came about as a result of the emergence of a mature Japan-centered economic bloc encompassing Ezo, Ryūkyū, and Korea and essentially independent from the Chinese market. This phenomenon was recorded by Philipp von Siebold, who lived and traveled in Japan from 1823 to 1829 and wrote extensively about the country, including its relations with other countries and peoples in the region. Siebold's observations regarding Japanese trade (Nippon, 1832–51) center on three points: (1) Japan is a world unto itself that can maintain its prosperity even without trading with the European powers. (2) Japan's trade with China is "hardly worth mentioning," though it allows Japan to maintain contact with the "old world," and its people to obtain certain necessities. (3) Japan carries on a vigorous trade with neighboring lands and dependencies that function essentially as colonies. Siebold's first observation, that Japan is "a world unto itself," corresponds to my own concept of a "Japanocentric regional order." His second observation supports the notion that the Japanese economy had by this time become virtually independent from that of China. And his third observation corroborates the idea that the foundation of this independence was the maturation of a Japanese economic bloc, within which Japan engaged in active trade with its neighbors.

One major change that occurred in the eighteenth century, between the first shift and the second shift, was a growth in Japan's domestic production capacity, as the Japanese learned to produce goods that they were no longer able to import from China. (This eventually led to a collapse in the price of the sugar imported from China and Holland, as well as that of the black sugar produced in Ryūkyū and the Amami Islands.) Meanwhile, the domestic Japanese market became increasingly dependent on Ezo as a source of herring used for fertilizer, and on Ryūkyū and the Amami Islands (under the administration of Satsuma) as sources of the black sugar that the common folk used for sweetener. As a result, these regions came under tighter control. In this sense, it cannot be denied that the maturation of a Japan-centered trading area was in large part an outcome of Japan's exploitation of the Ainu of Ezo, the Ryūkyū people, and particularly the inhabitants of the Amami Islands.

The Limited Influence of the Dutch

For many people, any mention of foreign relations in the Edo period automatically conjures up picturesque images of the Dutch traders on Dejima. However, in terms of the volume of trade flowing through Nagasaki, the Dutch component amounted to only about a third of the Chinese share. Furthermore, the actual presence of Dutch traders in Nagasaki peaked in the seventeenth century at several dozen inhabitants, and after 1700 the population of Dejima hovered at 20 or so—including the African and Southeast Asian servants the Dutch merchants brought with them.

Although Japan was a cash cow for the Dutch East India Company until the middle of the seventeenth century, it became less profitable after 1662, when Zheng Chenggong expelled the Dutch from their trading post on Taiwan and deprived the VOC of a key hub for trade in the China seas. Moreover, as the flow of silver out of Japan was halted in the early eighteenth century, the profitability of the VOC's Nagasaki-based trading operations fell to the point where the company began to consider abandoning the Japanese market altogether. The journals of the opperhoofd of the Dutch factory are filled with complaints about the meager proceeds from trade. It is unclear why the Dutch nevertheless remained on Dejima for approximately 150 years, but it would seem that the VOC saw sufficient potential for profit to keep the trading post open.

In truth, the Netherlands did not begin to exert a significant influence on Japanese culture until around the turn of the 19th century, when the shogunate came under pressure to open its ports to trade with the other Western powers. The fact that the word sakoku—coined to translate a Dutch term—entered into the Japanese language around this time (with the publication of Shizuki Tadao's 1801 work Sakoku ron), is one ironic indicator of this belated influence.