The Gwanggaeto Stele

The stone is 6.39 metres tall and carries about 1,775 characters on all four sides; around 150 of them can no longer be read.

Most of what we know about the Three Kingdoms comes from two books written in the Goryeo period, centuries after the events. The Gwanggaeto Stele is different: it was carved in Goguryeo itself, in 414, two years after the king it honours had died. The National Institute of Korean History calls it the most important inscription of ancient Korean history. It stands in Ji’an, Jilin province, China, near the old capital Gungnae-seong. In Korea it is usually called the Gwanggaeto Stele or Gwanggaeto Daewangneung-bi; in China it is known as the Haotaiwang stele.

A rough stone pillar

The stele is a four-sided pillar of breccia tuff, a stone made of volcanic ash with angular fragments in it. The faces are uneven, so it looks more like a natural rock than a cut block. It is 6.39 metres tall. It stands in a groove cut into a granite base about 3.35 metres long and 2.7 metres wide; the base is now broken into three pieces. Face 1 looks south-east.

The carvers ruled a fine grid on every face so that the characters would be evenly spaced. There are 11 lines on face 1, 10 on face 2, 14 on face 3 and 9 on face 4, 44 lines in all. Most lines hold 41 characters. The script is mostly the clerical style of the Han dynasty, and each character is about 9 to 12 centimetres tall.

What the text says

The inscription has three parts. The first tells the founding story of Goguryeo, the line of the first three kings, beginning with the founder it calls Chumo, and the life of King Gwanggaeto. The second, which caught the most attention from the start, lists the king’s campaigns year by year, counted in his own reign years, called Yeongnak. The third part, from face 3 line 8 to the end of face 4, is practical: it lists the households assigned to guard the royal tombs, where they came from and how many came from each place, and the penalties for breaking the rules.

Forgotten, then found

Making rubbings was difficult because the surface was so rough. Early copies traced the outlines of the characters and filled in the background with ink. One such copy reached Japan in 1883 through Sakō Kageaki, an officer of the Japanese army general staff. As demand grew, a local father and son who sold rubbings, Chu Tianfu and Chu Junde, coated the surface with lime to make clean copies easier, and some characters were changed in the process. For a while some scholars accused the Japanese army of altering the text on purpose; the field study by the Chinese scholar Wang Jianqun concluded that the lime was the rubbing sellers’ work. Rubbings made in the late 1880s, before the lime, are now treated as the most reliable.

One disputed sentence

The best known debate concerns one sentence placed between the entries for Yeongnak year 5 and year 6, about the sinmyo year, 391. Early Japanese scholars read it to say that Wa, the people of the Japanese islands, crossed the sea and made Baekje and Silla its subjects, and they used it to support the claim that Wa had ruled southern Korea. From the 1930s Korean scholars, beginning with Jeong In-bo, argued that the grammar allows a different subject and a different meaning. Because some characters are damaged, the sentence is still debated, and it is one reason the original rubbings matter so much.