Sankin-kōtai: The Origins, Implementation, and Effects of the Alternate Attendance System

• Published: 12 May 2025 • • Last updated: 12 May 2025 • • Written: 4 November 2024 •

(This entry is part of an ongoing series of papers I've written for various assignments. It is not necessarily representative of my beliefs; only my effort.)

Sankin-kōtai: The Origins, Implementation, and Effects of the Alternate Attendance System

Toyotomi Hideyoshi was the second great unifier of Japan, whose reign “defined a workable relationship between daimyō and court” (50). However, at the time of his death in 1598, Hideyoshi’s son and heir, Toyotomi Hideyori, was only a child. And while Hideyoshi entrusted his lieutenants to cooperate to secure the boy’s reign and continue his own legacy, each lieutenant claimed to be “acting on [Hideyori’s] behalf” (52) for his own benefit. The rising tensions between potential successors culminated in the Battle of Sekigahara, a bloody conflict waged in the year 1600 in the wake of the freshly-opened power vacuum.

Emerging victorious from the battle was Tokugawa Ieyasu, “the greatest among [Hideyoshi’s lieutenants]” (50), and the founder of the Tokugawa shogunate. Ieyasu understood that, to avoid a repeat of the Battle of Sekigahara that had been predicated on the ambiguously legitimate claims of each daimyō to the seat of shogunal power, loyalty to the shogun’s bloodline and centralization of power around the shogunate capital of Edo needed to be established.

Part of Ieyasu’s feelings of urgency surrounding the consolidation of power stemmed from the fact that, even after his victory at the Battle of Sekigahara, many daimyō that had fought against him remained powerful. These individuals were called “tozama daimyō” (lit. “outside daimyō”) (108), and they were insulated from the direction of Edo by the size and intactness of their estates and armies, and sometimes, geographical distance. And while some tozama daimyō were killed, or had their land seized and their titles stripped during the Battle of Sekigahara or in its aftermath, others were reintegrated under Tokugawa control, albeit with restrictions. In the southwest, for example, the Satsuma domain was designated tozama, but, being one of the largest domains in the region and also quite integral to Japanese trade with the Ryūkyū Islands, Ieyasu allowed Satsuma to “restore [their] trade role” (77), but “ordered [daimyō] in the southwest to surrender all ocean-going vessels in excess of 500-koku capacity and forbade construction of any new ones” (77).

Such material restrictions, however, were not enough for the Tokugawa. A sufficiently powerful tozama daimyō was under no serious threat if he decided to stage a rebellion. In the 1630s, Tokugawa Iemitsu, grandson of Tokugawa Ieyasu and third shogun of the Tokugawa shogunate, decided to standardize the “daimyō hostage practice that had operated haphazardly since Ieyasu’s day” (108). Under Iemitsu, this practice was formalized into a tradition where “lords and large numbers of retainers moved between castle towns and Edo every year,” (108) as an opportunity to show their obedience to the shogun. Iemitsu even went as far as to assign his vassal daimyō “sankin-kōtai schedules” (110), determining at what intervals their subjects would make their pilgrimage.

Additionally, the “wives, heirs, and suitable number of attendants” of the daimyō were required to live in a mansion encircling the shogun’s castle in Edo (108). Under this system, the wives and children of the daimyō were well-cared for and entitled to entertainment and enjoyment of Edo—they just couldn't leave the city walls. And so was born the "alternate attendance system,” sometimes called the "alternate attendance and hostage system,” which served as a deterrent against the fomentation of rebellion—if a daimyō tried to raise his armies to challenge the Tokugawa, the first to be in harm's way would be his wife (or wives) and children.

Iemitsu’s regularization of sankin-kōtai “spurred the growth of Edo” in the wake of the population boom that moving the wives, childrens, and attendants of the daimyō had created (111). However, this was achieved at the cost of the rural farming backbone of the country: Edo was “headed by a military elite still capable of compelling rural areas to yield up food supplies on demand, even if that left the local folk hungry” (111). Additionally, commoners were required to attend the pilgrimages, if for anything, to serve as attendants and porters for nobles and samurai, which starved the countryside of laborers and reduced crop output. Totman notes that “[this] combination of factors reducing food output and availability just when demand for it was sharply increasing…produced hard times” (111), resulting in peasant revolts, although they amounted to little compared to later crop failure and mass livestock deaths, which raised the cost of food and left people “starving along the highways” (112).

Although the material consequences of sankin-kōtai, especially on poor and farming populations were dire, the alternate attendance system did usher in a period of Japanese history that, although rife with growing pains and political disputes, was relatively free from conflict between daimyō domains, and between vassal domains and the Tokugawa. In this light, the sankin-kōtai system succeeded in ensuring the loyalty of the daimyō to Edo, and the succession from one Tokugawa heir to the next for almost three hundred years to come.

Bibliography

Totman, Conrad D. “The Politics of Pacification.” Early Modern Japan: Conrad Totman. Berkeley, University of California Press, Ann Arbor, Michigan: University of Michigan Library, Scholarly Publishing Office, 1995, pp. 39-58.

Totman, Conrad D. “The Economics of Pacification.” Early Modern Japan: Conrad Totman. Berkeley, University of California Press, Ann Arbor, Michigan: University of Michigan Library, Scholarly Publishing Office, 1995, pp. 59-79.

Totman, Conrad D. “The Politics of Order.” Early Modern Japan: Conrad Totman. Berkeley, University of California Press, Ann Arbor, Michigan: University of Michigan Library, Scholarly Publishing Office, 1995, pp. 105-139.