The Shimabara–Amakusa Uprising and the Siege of Hara

In 1637 famine sharpened the effects of heavy taxes and Christian persecution in Shimabara and Amakusa. Under the Matsukura and Terazawa lords, collection and religious investigation reached communities with longstanding Christian ties. Accounts differ over the immediate incident that sparked revolt, but both pressures shaped its spread.

Rebels attacked Shimabara and Tomioka castles, then joined at the disused Hara Castle. Armed villagers occupied the site together with women, children and older people. The young Amakusa Shirō became their leader and religious symbol; his personal biography remains uncertain. His prominence did not make the enclosed population one uniform body.

The shogunate sent Itakura Shigemasa, who died in an unsuccessful assault. Matsudaira Nobutsuna took command and maintained a blockade. The Dutch supplied military assistance, including cannon fire from the ship Ryp, without directing the besieging armies. Attempts to undermine Hara by tunneling also met resistance. As food ran short, the defenders’ ability to hold out declined.

Hara fell in 1638. Most occupants were killed, including noncombatants. Excavated bullets, Christian devotional objects and human remains preserve different parts of the siege’s material history. Matsukura Katsuie was stripped of his domain and beheaded; the Terazawa house lost Amakusa. Hara’s fortifications were dismantled and buried.

Religious investigations intensified, and Portuguese ships were banned in 1639 within an already developing maritime restriction policy. Dutch trade continued. The uprising brought both mass killing and changes in local rule. Letters, survivor statements, archaeology and later tales leave an uneven record: besieging forces produced far more surviving documentation than the people destroyed inside.

Taxation, Harvests and Christian Communities

Shimabara and Amakusa had substantial Christian communities under earlier lords. The Arima supported missions in southern Shimabara, while Konishi retainers and Christian villagers had ties across Amakusa. Changes in lordship left these associations in place under new rulers.

The Matsukura government built Shimabara Castle and demanded heavy taxes and labor contributions from its territory. In the 1630s successive poor harvests reduced what villages could provide. Officials continued collecting dues while also enforcing the prohibition on Christianity.

Religious investigation included violent pressure to renounce belief. In 1637 famine added to the crisis. Communities faced both loss of subsistence and punishment over affiliation, making the same local administration responsible for several forms of coercion.

Historical disagreement · What sparked the Shimabara revolt?

Accounts disagree over the opening incident: some emphasize violence against religious observance, others collection from people unable to pay.

The revolt spread through southern Shimabara and across to Amakusa, where the Terazawa administration also imposed demanding obligations. Participation was uneven; domain-side records also describe people drawn in under coercion by rebel groups.

The new lords governed people whose connections had formed under the Arima and Konishi. Former retainer ties and Christian affiliation could exist within farming households; they were not separate populations that can be counted simply as “warriors,” “peasants” and “Christians.” Castle work and dues required resources from those same communities. Poor harvests reduced the margin available for meeting demands, while religious pressure threatened relationships through which belief had been maintained.

Early accounts describe a population of tens of thousands eventually enclosed at Hara, but do not provide one agreed total or a consistent division between armed people and those accompanying them. The scale of the crisis included households, rather than only a field army. Some joined an organized resistance, and others were drawn into it under pressure. Neither the taxes nor Christian persecution alone explains that mixture.

Castle Attacks, Hara and the Blockade

The rebels approached Shimabara Castle and besieged Tomioka in Amakusa. Neither became their lasting base. Amakusa groups crossed the water to Hara, where they joined the Shimabara forces.

Hara was an Arima castle that had ceased to function as a lordly seat. Its plateau, sea-facing sides, western wetland and surviving fortifications provided a defensible enclosure. Villagers entered with family members; armed positions and household refuge occupied the same site.

Amakusa Shirō was named leader, with experienced villagers and former retainers contributing to the forces around him. Village groups were organized within the defenses. Accounts provide more evidence for this organization than for the boy’s personal direction of every operation.

Itakura Shigemasa arrived as the shogunate’s commander. A letter to Kamei Koremasa reported preparations to attack Hara. Koremasa was guarding the transfer of Matsue Castle and was receiving news, rather than serving in the siege army. Itakura died during a subsequent unsuccessful assault.

Matsudaira Nobutsuna maintained the siege and cut supplies. Dutch merchants provided cannon support through the Ryp. Their commander Couckebacker returned to Hirado before Hara’s final fall. Mining specialists also attempted a tunnel; defenders dug toward it and used smoke and waste to frustrate entry.

Food shortage increasingly weakened resistance. The final attack followed the blockade. Correspondence among besieging lords also records demands for distinguished service: Hosokawa Tadatoshi urged his son Mitsunao in the field to exceed ordinary performance. Service under central command still involved individual houses seeking recognition.

Excavation has identified rows of small, partly sunken buildings inside Hara. Their regular arrangement led the investigators to suggest use by family units grouped by village. Burned floors, ceramics and human remains were found together. These are traces of organized accommodation within the defenses, rather than evidence that everyone inside fought or entered voluntarily. Maintaining a refuge required provisions as well as a perimeter; the blockade attacked that requirement even where the ruins remained defensible.

The change in command did not turn the besieging forces into a single household army. Itakura’s failed assault was followed by Nobutsuna’s sustained restriction of access and supplies. Mining and Dutch cannon support were additional means of attack, not substitutes for the surrounding domain forces. As hunger deepened, remaining inside Hara became less sustainable. Accounts of approaches to surrender cannot be reduced to one freely made decision by its whole population.

The Fall of Hara and Changes in Local Rule

Hara fell in 1638, and most of those inside died. The killing reached women, children and older people as well as armed defenders. Surviving accounts disagree on totals, so no single figure establishes the exact number of occupants or deaths.

The shogunate punished the previous local rulers. Matsukura Katsuie lost his domain and was beheaded; Kōriki Tadafusa became lord of Shimabara. Terazawa Katataka lost the Amakusa territory. The region’s government changed after the destruction of the rebel stronghold.

Hara’s defenses were systematically broken and buried. Excavation later exposed disrupted entrances and stonework, together with shot, devotional objects and human bones. Destroying the military site also buried much of the material evidence now used to study it.

Hosokawa correspondence records arguments over who entered first and how retainers should be rewarded. These disputes concern the victorious forces. Yamada Uemonsaku’s surviving statement supplies a different position from inside Hara, but his account of his own conduct and survival requires particular caution.

Punishment of Katsuie and transfer of the lordship addressed the government of the territory, rather than restoring the households destroyed at Hara. The shogunate could remove a daimyō whose administration had become part of the crisis. That intervention followed the mass killing and did not relieve the besieging forces of responsibility for it. The people who had suffered under local rule did not thereby acquire control over the new administration.

The written aftermath also separates participants unevenly. Victorious houses could argue over service and rewards in letters; far fewer occupants of Hara left accounts. A devotional object establishes Christian practice at the site without identifying the motives of every person around it. Bones and demolished structures preserve the scale and violence of the end, while many individual names and choices remain unavailable. The destruction of resistance was not the disappearance of all Christian practice from the wider region.

Religious Enforcement, Maritime Policy and Memory

The shogunate renewed orders for religious investigation after the uprising. Controls on affiliation had preceded it, but suppression gave them renewed urgency. Temple certification associated households with a Buddhist temple, while religious investigations involved village officials and records of inhabitants. In regions where image-trampling inspections were used, people were required to step on Christian devotional images as a public test. These were related but different ways of detecting prohibited affiliation, and their development did not begin with one order immediately after Hara’s fall.

In 1639 Portuguese ships were barred. This followed earlier restrictions on Japanese overseas movement and foreign trade; the uprising intensified official suspicion of the Portuguese Christian connection. Dutch commerce remained authorized and its factory moved to Dejima in 1641.

Memorials also preserved differing judgments. Suzuki Shigenari’s 1648 monument used hostile language for Christians and portrayed Shirō as misleading people. Later collections of bones and memorial rites at Hara addressed the dead through other local practices. Commemoration did not produce one agreed account of the revolt.

The surviving record combines timely letters, domain memoranda, later compilations and military tales containing fiction. Archaeology adds objects and structures that written categories omit. Together these sources establish local coercion, organized resistance and catastrophic suppression, while leaving many individual experiences inside Hara unrecoverable.

The later administration in neighboring Ōmura shows how several measures worked together. After a separate discovery of Christians in the 1650s, the domain established religious-investigation officials in 1658. They collected jointly signed household-group pledges and temple certificates, checked them against population registers prepared by village heads, and conducted image-trampling inspections. This was a later regional implementation, not a procedure that can be assigned unchanged to every village in 1638. It made temples, household groups and village record-keepers answerable for detecting prohibited practice.

Some communities nevertheless maintained faith secretly while outwardly complying with required affiliations. The term senpuku Kirishitan is used for those practicing under the ban. After the prohibition was lifted, some returned to the Catholic Church, while others continued rites developed during the long period without regular Catholic contact; these later traditions are often called kakure Kirishitan. Regional histories and usage differ, but the distinction matters: the wider history did not consist of one unchanged community carrying the same practices forward from Hara.

The Portuguese exclusion likewise concerned a developing policy on maritime access. Portuguese traders and Hara’s occupants were not one body, even when officials associated Christian connections with a security threat. The continued Dutch factory, later at Dejima, shows how suppression and authorized commerce could coexist. Local government, religious classification and foreign access were tightened through different acts over time.