Where the Churches Had Been: Traces of Christianity in Japan, 1549–1873

A Portrait in the Family

How Did a 1920 Portrait of Francis Xavier Survive in Kobe?

In 1920 a painting of Francis Xavier came to light in the keeping of the Higashi family at Sendaiji, in the hills north of Osaka, where it had been handed down privately in territory once ruled by the Christian lord Takayama Ukon. Xavier held a crucifix beside a burning heart. The painter was unknown, and the picture belonged to the early seventeenth century, which is to say that it was made long after its subject was dead.

Kobe City Museum, which holds it now, records what happened next: the family sold it to the collector Ikenaga Hajime in 1935, and from his hands it passed into the city’s collections. A devotional image had become an art object with a documented chain of owners attached.

It survived because a household kept it.

Across Japan, a great deal of the Christianity introduced during the sixteenth century left its evidence on that scale and on no other. An inscribed grave preserved a baptismal name where the parish register had gone. An image in a family’s possession outlasted the institution that had taught its owners what to do in front of it. The organizations that had commissioned, taught and authorized these practices mostly disappeared, and the history was left distributed among people and places that had never been intended to serve as an archive.

The churches came to dominate the later picture, because churches are what a photograph can hold. Their towers rose above harbors and their façades supplied recognizable subjects. Yet many of the best-known buildings belonged to the nineteenth and twentieth centuries, raised either by communities whose Christian histories were much older than their masonry or by later Catholics commemorating the first missionaries. A church can be entirely authentic and still belong to a different century from the event attached to its name.

That uneven survival shaped the geography. Nagasaki preserved the sites of a missionary city; the villages around it preserved histories of concealment; collections in Kobe, Tokyo and Sendai held objects carried a long way from the places they had been made for. A journey confined to church interiors would pass over most of the evidence for the period when churches could still be built openly. A journey confined to excavations would miss the long work of the households that kept religious objects above ground.

The old word Kirishitan, from the Portuguese for Christian, covered the people the early missions had gathered, and their traces extended well past the years of direct Portuguese contact. The prohibition ran on into the nineteenth century. Its ending produced another wave of building, and that building was followed in turn by restoration and commemoration. At Sendaiji, the local Christian Relics Museum interprets the community that kept such things. The portrait itself went to Kobe. Even the opening piece of evidence required two places on the map.

The Mission Beneath the City

How Long Did Nagasaki's Mission Buildings Survive the 1614 Prohibition?

Xavier Park at Kagoshima preserves the remains of a church destroyed in the Second World War.

Xavier landed at Kagoshima in 1549. A marker on the waterfront records the landing, and the monuments that commemorate his arrival belong to the later history of remembering it rather than to the arrival itself. The distinction is easy to lose where old-looking stone stands beside the name of a sixteenth-century missionary. The stone has its own date.

What the mission became after 1549 was an infrastructure, and commemorative statuary is poorly equipped to represent one. Churches required property and maintenance. Schools trained Japanese Christians. Charitable associations cared for people whose needs did not wait for the next ship or the next superior’s appointment. In Nagasaki the church of Our Lady of the Cape, the educational establishment at Todos os Santos and the Misericórdia, the Christian charitable brotherhood, occupied separate addresses in an increasingly organized religious city, and the historical locations preserve that separation even where the buildings are gone.

The national prohibition of 1614 did not clear the ground at a stroke. The Daughters of St Paul’s historical gazetteer records that the charitable establishment stood until its destruction in 1620 and the buildings at Todos os Santos until 1619, which is to say that their functions and their fabric ended on different schedules. Shuntokuji temple later occupied the Todos os Santos area, where a monument and a well attributed to the church recall the institutions that had been there.

At Santo Domingo the ground kept more. The Dominican church was established in 1609, after the order’s earlier activity at Kyōdomari in Satsuma, and excavation during a modern school rebuilding exposed the early religious site beneath the later occupation. Inside the school complex the city’s museum preserves a stone-built underground room, paving and a drain associated with the church, together with a well from the deputy governor’s residence that came afterward. Successive occupants had left different structures in the same ground, which is what ground does.

The site also complicates the familiar outline in which a Portuguese Jesuit mission arrives and is then suppressed. Nagasaki accommodated several Catholic orders, each with its own personnel and its own buildings, and the Dominican move from Kyōdomari to Santo Domingo traced a displacement inside Japan rather than an arrival from outside it. Missionary infrastructure followed local permission and local expulsion as well as overseas arrival.

The city went on changing after the mission lost its public buildings. Dejima was built for Portuguese residence in 1636; the ban on Portuguese shipping followed in 1639; the Dutch moved across in 1641. Its reconstructed streets interpret that later trading post, not the Christian city underneath it. Nagasaki’s most legible historical buildings are not the ones closest in date to its first missionaries.

The Lord’s Church and the Neighbors’ Graves

What Did the Takatsuki Excavation Find Beneath Takayama Ukon's Church?

Beneath roughly two meters of later fill at Takatsuki, excavators found the dead laid out in order, adults and children together. A coffin lid carried a cross drawn in ink. Wooden rosary beads had gone into the ground with their owners.

The church those people had worshipped in is documented and gone. Takayama Ukon and his father entered the castle in 1573, and his father built a church there the following year. The Jesuit Luís Fróis described a large wooden worship hall with missionary lodging attached and a cross standing in a garden, an account that placed worship and residence inside a single establishment maintained by the ruling family.

The cemetery is the part that could still be dug. It established Christian burial beside the church independently of Fróis’s description, and it supplied evidence for a congregation whose ordinary members were rarely granted the space in the record that was given to their lord.

Such burials recorded what conversion actually required. The dead needed places inside the town as surely as the living needed a hall to pray in. An organized cemetery implies repeated use and an agreed procedure: someone had settled how a Christian grave was to be dug at Takatsuki, and it was then dug that way again. Patronage had consequences that ran from the castle’s reception rooms to the burial of children.

Kyoto kept the other half of the same arrangement, a missing building and a surviving object. The developed Nanbanji church was completed in 1576, during the period of Nobunaga’s protection, and it is gone. A bell associated with one of Kyoto’s Christian churches hangs at Shunkōin, a Buddhist temple within Myōshinji. It bears a cross, the letters IHS and the date 1577. Kyoto City’s survey separates what the bell says from where it was made: the temple tradition has it cast in Portugal, while the municipal account leaves the place of manufacture unknown.

The bell therefore carries a close connection with the missionary period without requiring the rest of its story to be settled. Its custody in a Buddhist institution also resists any tidy division of the surviving landscape into Christian and non-Christian ground. Objects moved. Institutions changed hands, or acquired things that had been made for other purposes, and the religion of a present custodian settles nothing about the original use of everything in its care.

Elsewhere the Christian lords are remembered at ruined castles, residence sites and modern statues. Hinoe preserves the landscape of Arima patronage; the Ōtomo sites in Bungo connect the mission with another regional court. Those associations explain where missionaries found protection. The graves and the portable objects explain what happened when that protection reached the people living underneath it, which is a history of adoption at a scale smaller than a domain.

Three Castles on One Peninsula

What Do Hinoe, Hara and Shimabara Castle Reveal About the 1637–1638 Rebellion?

Shimabara Castle has a keep and Hara Castle does not, and the absence is the more informative of the two. The keep at Shimabara is a modern reconstruction. Hara was pulled down on purpose.

The Shimabara Peninsula carried three fortresses whose roles collapse easily into a single story. Hinoe had been a seat of the Arima lords and of their Christian patronage. Hara became the rebels’ last refuge in the uprising of 1637 and 1638. Shimabara Castle belonged to the ruling order they had risen against.

The rebels took an abandoned fortress and made it defensible again. Hara’s headland gave them protection from the sea on several sides while its landward approach constricted an attacking army. The population that gathered inside included families as well as fighting men. Their resistance grew out of excessive taxation and famine alongside religious persecution, and the devotional objects recovered from the site establish the presence of Christian belief without explaining away the other reasons people were there.

When the shogunal forces took the position in 1638 they killed almost everyone inside it and demolished the defenses. Excavation later recovered human remains together with crosses, medals and other objects. These were the remains of a community destroyed in a siege, and they belong in that context rather than serving as interchangeable relics of persecution anywhere in Japan. The siege of Hara Castle has its own account on this site; what the wider map of sites adds is how exceptional that concentration of evidence was.

Hinoe holds the earlier half of the peninsula’s history, its terraces and masonry belonging to the political arrangement under which Christianity had enjoyed support. Hara holds the later half, a population defending a prohibited faith inside a revolt against its rulers. The Arima Christian Heritage Museum brings the two sites into one interpretation, which allows patronage and suppression to be examined together instead of alternately. A castle ruin on its own locates power. It takes the finds and the written accounts to establish how that power was used.

The peninsula’s other persecution sites survive in other forms. At Unzen the hot springs used to torture Christians remain an active natural landscape with memorials added to it. At execution grounds along rivers and inside settlements, markers identify places whose earlier appearance has changed beyond recognition. Those markers acknowledge in public what happened without pretending to be the structures in which it happened, which is more than can be said for a reconstructed keep.

Objects in Official Custody

How Did Confiscated Christian Objects End Up in Tokyo and Sendai Museums?

Tokyo National Museum holds images that people were compelled to tread on, and images that people hid in order to pray to, and the collection owes that pairing in part to the Nagasaki magistrate’s storehouse.

Objects confiscated in Nagasaki went into official storage. Images used to test suspected believers stayed among the instruments of administration. Some of each eventually reached Tokyo, so that the collection now brings together the things people had worshipped with and the things officials had used against them.

Among the latter are fumi-e, images on which people were required to step as a test of religious affiliation. The museum also holds Chinese porcelain figures of Kannon, the Buddhist bodhisattva, believed to have been worshipped by some Christians as representations of Mary, and its description retains the qualification. The identity of a figure, its use in a particular household and the circumstances of its confiscation are three separate questions. A resemblance to familiar Marian imagery suggests an interpretation. Provenance is what gives it substance.

In northern Japan, official custody produced a different kind of survival. Hasekura Tsunenaga left in 1613 on the Keichō embassy sponsored by Date Masamune, crossed the Pacific and went on to Catholic Europe. The expedition was a project of trade and diplomacy that included the proposed arrival of Franciscan missionaries, and the objects it brought home put the Christian encounter a long way from the western ports through which the Portuguese had entered.

Sendai City Museum records that those objects were confiscated during the lifetime of Hasekura’s son and kept under domain control. The same custody that separated them from worship is what carried them into a museum. The museum’s account also records the death of the Portuguese Jesuit Diogo de Carvalho and other Christians under water torture in the Hirose River in 1624. The domain that had sponsored an embassy seeking friars was killing Christians in its own river eleven years after that embassy sailed.

The ship at Ishinomaki belongs to the modern interpretation of the voyage rather than to the voyage. The Miyagi Sant Juan Bautista Museum’s dock exhibition uses a quarter-scale reconstruction together with components kept from an earlier full-size replica, and its value lies in explaining a vessel. The surviving diplomatic objects have an entirely different relationship to the seventeenth century. Neither of them depends on a reconstruction being called original.

This dispersal changed what counts as a Christian site. A museum in Tokyo could preserve an object confiscated in Nagasaki. Sendai could hold the evidence of a Catholic court on the other side of the world. The geography of the persecution included its places of custody as well as its places of execution. Survival sometimes followed a family’s refusal to give an object up. Sometimes it followed an official’s decision to keep one.

The Fields That Carried the Faith

How Did Kuroshima's Farmland and Registers Preserve Hidden Christian History?

On Kuroshima, off Sasebo, the Christian settlers registered at Kōzenji temple and trod on the image at the local office attached to the headman’s residence.

The official heritage account links the island’s nineteenth-century settlement to the redevelopment of former domain pasture. Christian migrants joined that movement and established themselves inside the island’s existing order, which meant inside its paperwork. The surviving heritage landscape is cultivated land, the sites of the leaders’ houses, a cemetery and the later church. The map of concealment is also a map of ordinary administration.

That arrangement required participation in institutions whose public account of the population left no room for Christianity at all. A household’s registered affiliation did not necessarily describe everything practiced inside it. Nor did the existence of concealed devotion make every relationship with a Buddhist temple or a local shrine an empty performance. Communities lived inside those relationships across generations, and what their later religious choices showed is how much had grown up within them.

Behind Kuroshima lay the reason people were moving at all. After the loss of public churches and resident clergy, the transmission of Christian practice fell increasingly to households and local communities, and the villages of Sotome, north of Nagasaki, became important both as places of persistence and as places people left. Their later churches marked the public reorganization of communities that had already existed through generations of prohibition.

Migration gave that history a larger map. Families from Sotome moved to islands where land could be cultivated and new settlements established, and the faith traveled with the skills and the relationships a household needed to keep itself alive. An island was not automatically outside government supervision. Migrants still had to live within a domain’s arrangements for land, registration and religious affiliation, which is why the most revealing sites include fields and offices alongside places of prayer.

Nozaki preserved a different outcome of the same movement. At Nokubi and Funamori the settlement traces include house sites, former agricultural land, paths and cemeteries. Nokubi’s later church remained after the village had stopped functioning as a village. Funamori’s church disappeared. The contrast arose on one island, and an account that counted only surviving churches would describe the two communities very unevenly.

These places make the domestic scale of transmission visible. Land provided food and work; paths connected houses; burial grounds held the relationships with the dead. None of it looks unmistakably Christian, which is the whole of the difficulty and most of the interest. Their religious history becomes legible through the association between material remains and community records. A field could be ordinary agricultural ground and evidence of the settlement that sustained a prohibited practice, and it did not need a hidden symbol to do both.

Sacred Ground Without a Steeple

Why Was No Catholic Church Ever Built at Kasuga?

No Catholic church was ever built at Kasuga.

The village lies on western Hirado among terraced rice fields, graves, household sacred places and routes toward Mount Yasumandake, a mountain that already belonged to a religious geography before Christianity entered it. Under the prohibition the community maintained inherited Christian practices among Buddhist and Shinto associations. When Catholic missionaries returned in the nineteenth century, Kasuga’s believers did not join the restored Church, and the landscape has no conventional architectural ending as a result.

Nakaenoshima, offshore, holds another place in the same geography. Christians had been executed there, and the island became a sacred source of water in inherited religious practice. Its importance never depended on a chapel: the shore, the island and the act of collecting water connected remembered deaths with later observance. A church-centered account can register it only as an absence, when what it did was positive and ritual.

On nearby Ikitsuki, the Shima no Yakata museum interprets the continuation of practice outside the re-established Catholic Church, including the recitation of orasho, prayers transmitted from the early mission. It sets that history inside the island’s wider life, alongside fishing and the other livelihoods, which gives the religious objects neighbors of their own.

The terminology carries the distinction. Heritage accounts use senpuku Kirishitan, concealed Christians, for the communities of the prohibition, and Kakure Kirishitan for those that continued inherited practices apart from the restored Catholic Church. English usage flattens both into “Hidden Christians” and loses the fact that ending concealment and joining a Catholic institution were two different decisions. The communities described in the history of Hidden Christian practice did not all make the same one.

Sotome’s Karematsu Shrine resists classification in its own way. Local tradition connects it with San Juan and with concealed prayer near a rock. The present shrine building is later. The Japan Tourism Agency’s interpretation presents the prayer-rock account explicitly as legend, which leaves the shrine’s history of veneration, reconstruction and shared remembrance to be taken seriously without being converted into a surviving clandestine Catholic chapel.

What the landscape preserved was never a faith waiting unchanged for foreign priests to come back. It recorded practices handed down inside families, attachments to particular dead, and relationships with places that had acquired their authority locally. A restored parish could continue one part of that inheritance. A community outside the parish could continue another.

Twelve Mats on Hisaka Island

How Many Believers Were Confined at Rōyanosako on Hisaka Island in 1868?

At Rōyanosako in 1868, more than 200 believers were shut into a space that Gotō City describes as approximately twelve tatami mats.

The municipal account records thirty-nine deaths in custody and three more after release, and attributes them to hunger, disease and torture. The memorial church raised on the prison site in 1984 marks the confined area out with a change in the floor covering, so that a visitor can see where its edges were. The dimensions explain what an aggregate death toll does not.

That was three years after the encounter usually treated as the end of the hiding. On 17 March 1865 believers from Urakami had approached the French priest Bernard Petitjean at Ōura in Nagasaki, in a church built for the foreign settlement and dedicated to the Twenty-Six Martyrs executed in 1597. That encounter and the deportations that followed it are told at length elsewhere on this site. What a map of surviving evidence needs from them is narrower. Ōura was itself a late-Edo building, belonging to the reopening of contact more than three centuries after Xavier came ashore at Kagoshima; the discovery it is famous for was a discovery from the returning mission’s side of the door, since for the families involved the practices had been part of their lives all along. What changed was access to priests. The prohibition did not change at all.

Then the fall of the Tokugawa regime in 1868 failed to end it. Urakami believers were deported to places beyond Nagasaki, and communities in the Gotō Islands faced arrest and confinement. The map extended along routes of forced movement, which added inland places to a history usually drawn along the western coastline.

Tsuwano received Urakami exiles under the new government, and the later Maria Chapel at Otome-tōge commemorates the imprisonment and the deaths at the former detention site. Hagi preserves the graves and the memory of exiles alongside its own earlier Christian martyr history. These places belong to the final persecution, the nineteenth-century one. To read their graves as remains of the seventeenth-century suppression would erase the regime under which those people actually suffered.

The prohibition notices came down in 1873. That supplies a practical stopping point for the long concealment, although communities emerged from it with different histories and different losses, and public church-building followed. The towers that became the most recognizable evidence of hidden Christianity were made possible by the end of the conditions under which it had been hidden.

The Churches They Could Build

Why Did a Buddhist Architect Design Japan's 1918 Egami Church?

The wooden church at Egami was finished in 1918 and designed by Tetsukawa Yōsuke, who was Buddhist.

The prefectural account connects its construction with a period of prosperity in the local fishery and describes its adaptations to humidity and storms. A community emerging from Christian concealment had given the design of its public church to a craftsman who did not share its religion.

The churches of the restored mission went up where the congregations and the money were, and their dates record decisions rather than continuity. At Shitsu the French missionary Marc Marie de Rotz was appointed in 1879 and had a church consecrated in 1882, its low roof and later extensions answering the needs of the returned mission, and around it welfare and vocational facilities tied religious organization to the work of keeping people fed. The older community supplied the people. The new buildings recorded what changed once clergy could serve them in public.

Nearby Ōno used what was on the hillside. Its walls incorporate basalt and a binding mixture, the latter described in the prefecture’s architectural account, while its roof employs Western trusses. That combination came out of the encounter between missionary building knowledge and techniques already in use for houses in the district. Its modest scale does not make it a clandestine survival. It makes it a local one.

Buildings moved through the developing parish network like anything else people owned. The wooden church now known as Former Gorin Church was built at Hamawaki on Hisaka in 1881 and transferred to Gorin in 1931, so that its construction date and the date it reached its present site record two separate events. When a replacement later made it redundant, preservation stopped it being pulled down. It had served successive needs before it became a heritage object.

Urakami accumulated the most violent sequence of all. The community’s monumental church, begun in 1895, was devastated by the atomic bombing in 1945; the cathedral was rebuilt in concrete in 1959 and faced with brick tiles in 1980, and bomb-damaged sculptures and the fallen belfry preserve material from the building that was destroyed. That sequence has a history of its own, told separately. What it contributes here is the plainest version of the rule: at Gorin a congregation received a building from elsewhere on its own island, and at Urakami parish life continued in a replacement standing beside the fragments of its predecessor.

The Village Around the Monument

Why Did UNESCO List Twelve Nagasaki Sites as World Heritage in 2018?

Twelve places in the Nagasaki region went onto the World Heritage List in 2018. The Xavier portrait in Kobe was not one of them.

UNESCO inscribed the Hidden Christian Sites in the Nagasaki Region as a serial property whose subject is the transmission of a religious tradition under prohibition, and the developments surrounding its emergence. Hara Castle and Ōura belong to that account alongside village landscapes and sacred places, and the scale of the inscription required attention to settlements rather than treating every named component as a church building.

The selection could not stand for the whole Japanese Christian past. The portrait in Kobe, the graves at Takatsuki and the embassy objects in Sendai answer questions outside its regional and thematic boundaries, and Nishizaka and Ikitsuki remain central to the history while sitting outside the twelve components. An international designation identifies a particular group of places for protection. It does not relieve the rest of the evidence of its work.

Kasuga’s inclusion was the useful anomaly. Its community had stayed outside the restored Catholic Church, so the village prevents the inscription from being read as a sequence in which every concealed group eventually returned to the same institution. Its fields and its sacred places preserve a history for which a newly built parish church would have been the wrong concluding object. World Heritage recognition could take in a community whose answer to the returning mission differed from those of its island neighbors.

Recognition also changed the practical circumstances of survival. A sacred place acquires explanatory signs and visitors. A house site becomes part of a mapped route. Conserving a landscape requires decisions about what may change around it; keeping a church in repair requires labor and money. None of those arrangements can by itself maintain the households or the ceremonies that gave a place its earlier purpose. The survival of an object and the continuation of its use remain separate matters.

Nozaki makes the separation visible. At Nokubi the church stands within the remains of the settlement that supported it, and the prefectural record identifies the former houses, the disused farmland and a path the inhabitants made. Comparison with a map from around 1877 helps establish continuity in the layout, which supplies a more exact history than calling the island untouched: the ground retained recognizable arrangements after the people’s use of them had changed.

At Funamori, the island’s other former Christian settlement, the church has gone. The site retains the traces of habitation and burial that connect it to the island’s longer history. The two settlements entered the same heritage account with different things left standing. At Nokubi the church remained above the former village, and the path its inhabitants had made still ran through the ground around it.

Sources & Further Reading

Gotō City. Rōyanosako Martyrs Memorial Church. Gotō City, updated 13 March 2024. The municipal account supplies the confinement conditions, distinguishes deaths in custody from deaths after release, and dates the present memorial church.

Hirado City, Shima no Yakata Museum. Kakure Kirishitan: Exhibition Room 3. Shima no Yakata, n.d. The museum explains Ikitsuki’s inherited religious practice and its continuation apart from the restored Catholic Church.

Japan Tourism Agency. Karematsu Shrine. Ministry of Land, Infrastructure, Transport and Tourism, prepared in fiscal year 2020. The interpretation distinguishes the later shrine building from the traditional account of concealed prayer near the rock.

Kobe City Museum. Portrait of Saint Francis Xavier. Kobe City Museum, n.d. The collection record dates the painting, records its discovery in 1920, and traces its movement from the Higashi household into public collections.

Kyoto City. Christian Historical Sites in Kyoto. Cultural History 16, Kyoto City, n.d. This municipal survey documents the lost churches and the surviving objects, with a specific qualification about the bell’s presumed Portuguese manufacture.

Miyagi Sant Juan Bautista Museum. Dock Building. Miyagi Sant Juan Bautista Museum, n.d. The current exhibition record distinguishes the quarter-scale ship from the components preserved out of the earlier full-size reconstruction.

Nagasaki City. Santo Domingo Church Site Museum. Nagasaki City, updated 11 November 2024. The municipal record confirms the preservation and display of early-Edo church remains and excavated objects within the modern city.

Nagasaki Prefecture. Orasho: Churches and Christian Historical and Cultural Heritage of Nagasaki. Nagasaki Prefectural Government, n.d. The individual records for Kasuga, Nakaenoshima, Nokubi, Shitsu, Ōno, Egami and Former Gorin Church establish the different histories of settlements, sacred places and later buildings.

Sendai City Museum. Materials Relating to Hasekura Tsunenaga, Part 5. Sendai City Museum, updated 1 February 2024. The account explains the confiscation and custody of the embassy objects and records the domain’s persecution of Christians after Hasekura’s return.

Takatsuki City. Site of Takayama Ukon’s Takatsuki Church. Takatsuki City, updated 31 March 2025. The municipal history distinguishes Fróis’s account of the church from the excavated burial evidence, including the cross-marked coffin and the wooden rosary.

Tokyo National Museum. Researching and Conserving Artifacts of Early Japanese Christianity. Tokyo National Museum, 2026. The exhibition record identifies confiscated objects and image-trampling materials formerly held by the Nagasaki magistrate, and retains the qualification about the Christian use of Kannon figures.

UNESCO World Heritage Centre. Hidden Christian Sites in the Nagasaki Region. UNESCO, inscribed 2018; live property record. The statement of significance establishes the scope of the twelve-component inscription and its emphasis on religious transmission during the prohibition.