Ten-story Stone Pagoda from Gyeongcheonsa Temple Site

A pagoda of marble, 13.5 metres high, standing on a three-tiered foundation that projects in four directions like a cross - the shape Korean sources describe as an "a-ja-character" plan. The lowest three storeys repeat that cruciform plan; the seven above are square. Each panel of the tiered base is carved with Buddhas, bodhisattvas, flowers and arabesque ornament.

The upper storeys do something odd and deliberate. They carry railings, hip-and-gabled roofs, eaves and carved courses meant to read as roof tiles: the marble is worked to create the illusion of a wooden building. This is a stone monument imitating the timber architecture of its own period, which makes it a source for Goryeo wooden building as well as an object in its own right.

It has ten storeys. It is very commonly said to have thirteen, because the three-tiered foundation reads as three more levels to anyone counting from a photograph. Getting the count right is not pedantry: the name of the object in Korean and in every catalogue is the ten-storey pagoda, and a source that says thirteen has miscounted rather than described a different thing.

The carving is not decorative filler. According to the National Museum of Korea's own curatorial account, the lowest tiers carry scenes connected to the Xuanzang pilgrimage story cycle later known as Journey to the West, the second through fourth storeys carry Buddhist ceremonial scenes, and the fifth through tenth storeys carry a bodhisattva progressing through practice toward enlightenment. A senior museum researcher, Shin So-yeon, has described the sequence as depicting the path of all living beings seeking enlightenment, read from the base upward.

The pilgrimage story predates the Ming-dynasty novel of the same English title by more than two centuries; a Yuan-era popular cycle of Xuanzang tales was already current in the 1340s, which is the tradition the carvers had to be drawing on, not the later book. That the earliest instalments of an East Asian narrative programme this famous should survive carved on the lowest tier of a 1348 Korean pagoda is a fact the tourist-facing description undersells.

Korean stone-carvers worked in granite. It is the material of the peninsula's pagodas and of most of its sculpture, and it is extremely durable. This pagoda is marble, and marble is why almost everything that has happened to it since 1907 has happened.

Marble takes the fine undercutting that produces those imitation eaves and tiles; granite does not, at least not at this scale of detail. The choice buys the illusion of timber. It also buys a stone that acid attacks, and the rain of a twentieth-century industrial city is acidic.

So the material explains the object's ambition and its fragility together. The damage became serious enough that the Korean authorities took the monument apart again and, in 2005, moved it indoors, which is not a step any of Korea's granite pagodas has required.

The inscription on the first storey gives the year: the fourth year of King Chungmok, 1348, recorded in the Yuan reign-name Zhizheng rather than a Goryeo era name alone. Goryeo was then in the last decades of its existence and had spent nearly a century under Mongol Yuan overlordship, with its royal house tied to the Yuan court by marriage and its elite serving in Dadu.

The Academy of Korean Studies' Encyclopedia of Korean Culture and a 2026 Korean press account both state what the object was for: it was commissioned by pro-Yuan Goryeo aristocrats, and its dedication prayed for the longevity of the reigning Yuan emperor. That is a documented statement of patron intention, not a modern reading back onto the stone - a court faction advertising its Yuan allegiance in permanent marble, at a scale no granite pagoda in the country matched.

The pagoda stood at Gyeongcheonsa, a temple at the foot of Mount Buso in Gwangdeok-myeon, Gaepung County, Gyeonggi Province. The temple is recorded as lost. Gaepung lies in the northern part of the old Gyeonggi Province, on the far side of the line that has divided Korea since 1953, which means that the findspot of a South Korean National Treasure is not accessible from the country that holds it.

That is worth sitting with, because it changes the meaning of the 1918 return. The pagoda was returned to Korea, and Korea was then one country under Japanese rule. It was never returned to its site, and since 1953 it could not be.

In 1907, during the Japanese occupation of Korea, Tanaka Mitsuaki, Japan's Minister of Imperial Household Affairs, had the pagoda dismantled and smuggled to Japan. Korean-language historical journalism, principally a detailed history column in the Kyunghyang Shinmun, has since reconstructed how: Tanaka delegated the operation to Kondo Sagoro, an antiques dealer working in Seoul, who assembled some 120 to 130 armed men at the site on 4 February 1907, with tents and supplies for an extended operation. When local Korean officials and even a Japanese interpreter present objected, the workers responded with gunfire and drawn swords. The pagoda was dismantled within the day into upward of 140 blocks and carted to the railway station at Gaeseong.

The same reporting states that Tanaka's justification - that Emperor Gojong had granted the pagoda to Japan - was false, and was exposed as such; other accounts report that he wanted the monument to decorate his own garden. Neither claim gives him a legal title to the object.

The year matters on its own terms regardless of the mechanics. 1907 is the year the Japanese Resident-General forced the abdication of Emperor Gojong, disbanded the Korean army and reduced the Korean state to a formality. Formal annexation followed in 1910. A senior Japanese court official removing a Korean national monument in 1907, by armed force, against the objections of Korean officials on the ground, was operating in a country whose government could not stop him and whose sovereignty was three years from being extinguished.

Popular Korean historical writing also attaches a piece of folklore to the case: that Tanaka's target had been coveted three centuries earlier by Kato Kiyomasa, the Japanese commander of the 1592-1598 invasions, and that Tanaka completed what Kato was said to have wanted and failed to take. This appears only in historical-interest journalism, not in a primary record read here, and it is reported as legend rather than as an established fact; it is included as what is being said about the object now, not as history that has been verified.

The publicity is documented in more detail than the earlier version of this record credited. Homer Hulbert, an American missionary and journalist, accompanied the Korean emperor's delegates to the Second Hague Peace Conference in 1907 - the same mission remembered in Korean history as the Hague Secret Emissary Affair - and used the international press attention it drew to hold a press conference denouncing the plundering of Korean cultural property, the pagoda among it. Ernest T. Bethell, an Englishman, ran a parallel campaign in the Korean and English-language press. Both men were foreign nationals and so could not simply be silenced by the Resident-General's censorship of the Korean press, which is precisely why their campaign carried internationally where a Korean-run one could not have.

Later Korean historical writing on the case adds an interpretation, not a document: that the colonial administration eventually judged a Korean National Treasure visibly restored to Korean soil, under Japanese rule, served the appearance of orderly, legitimate colonial administration better than the same object sitting in a minister's private garden in Japan under continuing international criticism. That is an argued reading of motive, attributed here to the historians who make it, not a claim this record treats as proven.

In 1918 the pagoda was returned to Korea, on 15 November, reportedly in exactly the dismantled and packaged state in which Tanaka's crew had left it eleven years earlier - no pieces confirmed lost, though the stone had deteriorated across those years in ways that complicated the eventual 1960 reassembly. That is a restitution which predates the 1970 UNESCO Convention by fifty-two years and predates almost every framework restitution arguments now use, and there was no law to invoke.

What remains undocumented in the sources consulted is a single Japanese administrative order or ministerial decision authorising the return. The record now available is a sequence - sustained international press pressure, a change in the administration's calculation of what served it, and a shipment back - not a cited legal instrument, and this record does not manufacture one.

The pagoda came back in 1918 and was placed on the grounds of Gyeongbokgung Palace. It was not put up. The colonial authorities intended to install it at their own museum; the Korean independence movement broke out shortly afterwards - the March First Movement of 1919 - and the plan was shelved.

It stayed down for four decades. The reassembly was finally carried out in 1960 under Lim Cheon, a museum researcher and architect, at Gyeongbokgung.

A record that says returned in 1918 and stops there has told about a fifth of the story. For forty-two years the object existed as a stack of numbered marble on palace ground, through a colonial period, a liberation, a partition and a war. The restitution of 1918 restored possession. It did not restore the monument, and nothing in the campaign that won it obliged anyone to.

Standing in the open in central Seoul, the marble was attacked by acid rain. The surface that carries the Buddhas, the bodhisattvas and the imitation roof tiles is precisely the part a dissolving stone loses first.

The pagoda was dismantled again in 1995 for conservation, and the National Research Institute of Cultural Heritage carried out a decade-long conservation programme; in 2005 the restored monument was moved indoors, to the new National Museum of Korea at Yongsan, where it stands inside one of the main halls. It is the only way this object can now survive, and it is the second time in a century that it has been taken to pieces and put back together.

Two dismantlings, then: one by a foreign official's hired crew in 1907 and one by its own conservators from 1995, both requiring every joint to be opened and every block handled, so the pagoda standing in Yongsan is a modern assembly whichever way the history is read. There is no published figure in the sources consulted for how much of the original jointing and surface survives the two campaigns, and that is a real gap in the record of a National Treasure.

About 120 years after this pagoda was made, the Joseon royal court commissioned a closely similar one for Wongaksa Temple. It was built in 1467, it is marble, it is twelve metres high, it has ten storeys, it stands in what is now Tapgol Park in central Seoul, and it is National Treasure of Korea No. 2. Since 1999 it has been enclosed in glass to protect it, for the same reason this one was moved indoors.

Two marble ten-storey pagodas, both National Treasures, both in Seoul, one modelled on the other and separated by 119 years and 1.5 metres of height. They are confused constantly, in photograph captions above all, and the confusion runs in both directions. Popular Korean historical writing pairs them further still, attributing to Kato Kiyomasa an interest in taking both to Japan in the 1590s; that pairing is repeated as legend and is not independently documented here.

The distinguishing marks are easy once known. This pagoda stands indoors, in a museum hall, unglazed, on a cruciform base. The Wongaksa pagoda stands outdoors, in a park, inside a glass case. Any photograph showing a pagoda behind glass in the open air is the Wongaksa pagoda and is not this object.

The pagoda is not a static exhibit. The National Museum of Korea runs "Lighting the Sky Pagoda," a twelve-minute media-facade projection onto the monument that animates the carved narratives - the Journey to the West scenes and the life of the Buddha among them - shown on Wednesdays and Saturdays in the evening, free and without reservation, alongside a daily augmented-reality programme that lets visitors examine the carving more closely on their own devices. The museum's own material calls it one of the institution's most recognisable highlights.

None of this changes the object's ownership, designation or physical condition, and this record notes it for the same reason it notes the conservation history: a National Treasure that draws crowds for an evening light show is still living an institutional life, seven centuries after it was carved and more than a century after it was taken apart at gunpoint.

There is no sale, no auction record, no valuation and no insurance figure in the public domain. The pagoda has been Korean state property since its return and has carried the highest national designation since 20 December 1962. It cannot be sold and never has been.

The only transaction in its history was not a sale either. In 1907 it changed hands by armed removal under a pretext later shown to be false, and no consideration passed to anybody in Korea. The most useful financial statement that can be made about the object is a negative one: it establishes that a monument of this rank has no market and therefore no price, and that the entire commercial apparatus of provenance - hammer, premium, estimate, underbidder - is inapplicable to it. What it has instead is a chain of custody, and that chain is unusually complete.