Gifu Castle: An Invitation Up the Mountain

In 1569, the Portuguese Jesuit Luís Fróis was received on the summit of the mountain above Gifu. His escort was Shibata Katsuie, a commander serving Gifu's ruler, Oda Nobunaga. The missionary's account, retold by Gifu's cultural-property office, describes tea and food in a lavish room, with Nobunaga himself serving. Fróis raised his tray overhead in gratitude; his host cautioned him about spilling the soup.^[1]

It is an unexpectedly domestic detail from a fortified height. There had to be somewhere to prepare that meal, somewhere suitable to receive it, and a way for both guest and provisions to reach the room. The mountain was difficult ground on which to attack a ruler. It was also ground on which that ruler expected selected people to visit him.

Gifu Castle's history turns on the coexistence of those purposes. How could a steep mountain serve as a place of defense, residence, and political encounter? The answer lies in the relationship between its lower palace, its slopes, and the buildings above—and in the work that made those relationships possible.

Mount Kinka rises about 329 meters above sea level beside the Nagara River and the northern edge of the Nōbi Plain. Its castle occupied a commanding position, but elevation alone is an incomplete explanation of its usefulness. The same climb that hindered an attacker complicated access for anyone with legitimate business. Holding this mountain required ways to use it in circumstances other than a siege.^[2]

Before Nobunaga, the stronghold was known as Inabayama Castle. It belonged to the Saitō rulers of Mino, the province around present-day Gifu. Nobunaga defeated Saitō Tatsuoki in 1567 and renamed the place Gifu. He acquired an existing castle and town, not an unoccupied summit awaiting his imagination.^[2]

Research into Saitō documents makes that inheritance more substantial. Historian Ishikawa Misaki identifies legal proceedings among the functions of the mountain residence. The summit had been a place of government as well as defense before Nobunaga took it.^[3]

That changes how his lavish hospitality should be understood. He did not discover that a military lord might need to receive people. He enlarged and reworked a setting already used for exercising authority. His visitors entered a landscape in which earlier rulers had also made the climb worth undertaking.

The distinction is easily lost when a castle is identified with one famous building. Gifu's national historic-site designation, made in 2011, instead encompasses roughly 209 hectares, including the foot residence and the mountain itself. The protected historical subject is the connected landscape: approaches, fortified positions, occupied terraces, and the terrain between them.^[4] A reception at the top belongs to that whole, rather than to a tower floating above an otherwise empty slope.

Fróis's account of the lower residence provides a sequence rather than a measured plan. In extracts published by Gifu's education authorities, he describes stone boundaries, a long stair, spacious rooms divided by folding screens, and a passage beside five or six gardens. He moves through several levels, including a room used for tea, and reaches places overlooking the town.^[5]

These details convey an inhabited palace encountered through movement. A stair changed the visitor's level; a screen divided an interior; a gallery brought rooms into contact with gardens. The interest was not confined to the building's exterior. What a guest could see depended on where he was taken.

Fróis wrote as a missionary reporting an encounter with a powerful lord, not as an architect recording dimensions for a future reconstruction. His letter of 12 July 1569 and his later history preserve related accounts; their descriptions have to be compared with one another and with archaeology. A phrase about several floors cannot, by itself, settle the form of every building on a terraced hillside.^[6]

There was another complication: the residence he visited was still being worked on. A recent official discussion draws attention to his mention of construction workers and retainers carrying materials. This is evidence of work continuing during the visit, not proof that the entire palace was unfinished or that Nobunaga had devised a special preview for guests.^[7]

It does, however, remove the need to imagine a single unveiling of a completed masterpiece. Rooms could be used while other parts of a large residence were being altered. A guest's description caught the place at one point in a process. Excavated remains may belong to work completed before that moment, changes made afterward, or the buildings of earlier occupants.

The clearest demonstration comes from the large garden excavators call Area A, north of the stream running through the lower residence. A 2014 excavation handout describes a substantial pond below an exposed rock face. Two falls brought water down the rock; stone paving received it before it entered the garden's water system. A long retaining wall enclosed the western side. Stairs approached the terrace from below, and remains on its southern edge were interpreted as supports for a bridge across the stream.^[8]

A bare inventory—pond, waterfall, wall, bridge—misses the construction problem. The garden occupied unequal levels beside running water. Its makers needed to hold terraces in place, channel the flow, and connect the places from which the scene would be encountered. Engineering did not happen somewhere out of sight before decoration began. The supporting walls and the altered rock face contributed to what the garden presented.

The excavators' drawings propose buildings around the pond and a crossing toward the principal residential area. Those roofs are reconstruction, while the recovered stairs and possible bridge supports are archaeological observations. We need not accept every drawn elevation to recognize the important relationship: access to the garden involved a change in height and a crossing, rather than a door opening onto a flat rectangle of ornamental ground.^[8]

Garden researcher Igawa Shōko offers a broader reading of the lower precinct. As summarized in the city's conservation plan, her work treats its separate garden areas as parts of a larger composition connected by the stream, bedrock, and changes in level. Nobunaga's builders substantially altered a landscape inherited from the Saitō.^[9]

This explains why counting gardens is less revealing than understanding how they joined. A visitor could experience different settings without leaving the larger mountain-and-water composition. Natural rock provided scale that no portable ornament could supply; constructed routes determined how that scale appeared alongside a room, a terrace, or a pond.

The result was demanding terrain made usable without being made flat. At Gifu, the mountain's height was not only a military advantage above the palace. A smaller-scale version of the same physical condition—rock rising beside occupied space—had been worked into the residence below. Guests encountered the mountain well before they began the ascent to its summit.

The lower residence also produced small objects that give Fróis's language of magnificence a material counterpart. Excavators recovered decorative roof tiles with gilded floral designs. Two pieces could be substantially reconstructed: roughly square plaques about 28 centimeters across, one with a chrysanthemum and another with a flower provisionally identified as a peony. The gold covered the raised floral decoration, not the whole surface. Surviving nail holes indicate how the plaques were attached.^[10]

These fittings adorned the roof ridge. They are not evidence for an entire roof made golden, still less for the color of every vanished wall. Their limited surfaces are part of what makes them informative. Gold could be concentrated where an architectural detail would distinguish a building, rather than spread uniformly across it.

The objects also shift the scale of the story. The garden enlisted a rock face; the roof fittings depended on shaped clay, applied gold, and a means of fastening each piece in position. Large landscape and small manufactured surface belonged to the same residence. Neither had to perform the other's task.

An account of splendor might be dismissed as a visitor's enthusiasm. A gilded fragment establishes an actual costly finish, but cannot tell us who looked up at it or what they thought. Together, description and object support a more precise conclusion: the palace made deliberate use of conspicuous materials, while the experience of those materials depended on buildings and routes that are only partly recoverable.

The means of presenting wealth are better documented than their effect on each guest. Admiration cannot simply be converted into political obedience.

The lower palace and summit offer an easy diagram: residence below, fortress above, a line between them. Investigation has complicated that division. At roughly 150 meters elevation, survey and excavation identified another group of terraces, stonework incorporating large rocks, and earthenware dishes. Something substantial occupied the middle of the mountain, although the attribution and functions of that complex remain under discussion.^[11]

This matters even without attaching a famous name to the remains. The intermediate zone cannot be reduced to the distance between two destinations. It contained spaces that people built and used. A mountain residence could have more than one center of activity and more than one stage of approach.

Routes require similar care. Historical maps and documents have helped identify several lines of ascent, but parts of the modern paths were subsequently diverted. A walkable trail today is not automatically the route followed by a sixteenth-century guest or servant.^[12]

The practical questions raised by Fróis's meal therefore remain useful questions, rather than opportunities to invent a supply routine. Food, water, fuel, utensils, and people had to reach occupied rooms. The distribution of terraces makes a picture of an isolated summit less convincing, but it does not locate a particular kitchen or establish a standard delivery schedule. Archaeology recovers the framework within which those arrangements operated more securely than their daily choreography.

There is uncertainty at the top as well. Gifu's official heritage discussion warns against locating Fróis's 1569 meal inside the tower familiar from later reconstructions. One archaeological interpretation dates the tower to around 1571; on that reading, his reception preceded it. Another summit platform, known as the Upper Kitchen, has been proposed as the setting, but no identification is settled.^[7]

The reception can be located on the fortified summit; its room cannot yet be securely named. Gifu could receive an important guest before the particular tower we associate with that reception had necessarily been built.

Fróis was not the only visitor whose reception entered the written record. In 1569, the Kyoto courtier Yamashina Tokitsugu asked to see the mountain castle and was invited to dine and inspect it. In 1574, the Sakai merchant and tea practitioner Tsuda Sōgyū received a private tea gathering at the lower residence; the city's account of his visit includes Nobunaga personally replenishing his rice.^[13]

The encounters involved people with different reasons to matter to a lord: a missionary, a courtier, and a merchant accustomed to elite tea culture. The evidence does not require them to have followed one ceremonial script. Indeed, their variety helps explain why Gifu needed more than a single audience chamber performing a single task.

Hospitality made an encounter particular. A defended entrance could admit someone without welcoming him warmly. A meal or a tea gathering added an exchange in which the host could single out the guest. Nobunaga's own participation was memorable precisely because the food could otherwise have been served by someone else. The sources preserve that attention, not a measurable return on it.

It would be equally mistaken to treat the palace's gardens as camouflage for politics or to turn every remembered courtesy into proof of a benevolent personality. Enjoyment and calculation need not cancel each other. Rooms, food, tea, and scenery offered actual experiences through which relationships could be conducted. They did not determine the future loyalty or beliefs of those who received them.

Here the mountain added something distinctive. To be welcomed in its upper rooms was to be admitted into a defended place ordinarily difficult to approach. The invitation temporarily changed the meaning of that difficulty. It remained the host's mountain, but the visitor was given a place within it.

Nobunaga moved his principal residence to Azuchi in 1576, leaving Gifu to his son Nobutada. Gifu continued under later lords after both men died in 1582. In 1600, Nobunaga's grandson Hidenobu held it for the western coalition during the struggle that culminated at Sekigahara. Forces on the opposing side took the castle before that decisive battle; its subsequent abandonment ended the mountain's life as an active ruling residence.^[2]

The change was not a discovery that hills had ceased to be defensible. After victory, Tokugawa Ieyasu established the regional stronghold at Kanō, on lower ground to the south. The political arrangement connecting lord, castle, and surrounding territory had changed. Gifu's powerful location could not compel the new rulers to keep their residence there.^[14]

Nor should the later history of the summit be folded into Nobunaga's architecture. The first modern reconstructed tower was built in 1910 and burned in 1943. Its replacement, completed in 1956, is reinforced concrete. Architect Kido Hisashi consulted documents and a drawing associated with Kanō's former three-story turret; citizens and business donors helped finance the project.^[15]

That building belongs to the history of Gifu's civic identity. It supplies neither Fróis's dining room nor a surviving Saitō residence. Its architectural genealogy involves modern research and decisions made for a city that wanted a tower again.

Separating those histories actually restores more of the mountain. The modern silhouette can identify the place from a distance. It cannot stand in for the lower palace, the garden's waterworks, or the intermediate compounds. Recognizing what the reconstruction is allows those less conspicuous remains to carry their own evidence.

The relationship between the lower residence and upper castle was not a neat division between comfort and war. Government had taken place on the mountain before Nobunaga; formal hospitality took place above as well as below. Between the better-known precincts lay additional occupied ground. At the foot, the terrain itself had been incorporated into the pleasures of a visit.

These connections give Gifu an explanation more exact than either an impregnable fortress or a magnificent palace. Its builders made selected parts of a difficult mountain habitable, connected, and presentable. Their work allowed the same topography to restrict an approach, frame a garden, separate activities, and reward an invited ascent with a room and a view.

Fróis remembered the person serving him. Excavation makes it possible to recognize some of the setting and effort behind that apparently personal moment. The tray could arrive because people had made places to receive it. At Gifu, hospitality was conducted through a landscape that had been built to admit, accommodate, and distinguish its guests.