"What Do the Ten Characters Mean?" Academia Abuzz Over 3.93 kg Palm-Sized Stone Fragment: "Is It Goguryeo or Silla?"

Inscribed Stone Fragment Unearthed at the West Side of Wolseong in Gyeongju

Ten Decipherable Characters, Countless Interpretations

From "Built by Goguryeo" to "No Conclusive Answer"

Scholars Offer Divergent Theories Over a 16-cm Stone

Width 27.19 cm, height 16.58 cm, weight 3.93 kg. A stone fragment just slightly larger than two palms together has attracted significant attention in academic circles. More precisely, these are two fragments. One was discovered in 1937 on the west surface of Wolseong by Choi Namju, then an employee of the Gyeongju Branch of the Joseon Government-General Museum. The other was unearthed in 2020 from a disturbance layer of a drainage channel at the excavation site of the Wolseong moat. Separated by 83 years and found by different hands, the two pieces were confirmed to be a single object during a joint investigation by the National Research Institute of Cultural Heritage in Gyeongju and the Gyeongju National Museum in May last year. The broken surfaces fit perfectly together without any gaps.

Gyeongju Wolseong West Side Unfinished Uninscribed Tablets / National Research Institute of Cultural Heritage in Gyeongju · Gyeongju National Museum

According to each scholar’s count, 15 to 17 characters remain legible on the stele fragment, and of these, less than 10 can be read with any certainty. There is no continuous phrase forming a sentence. The only clue is the script style, noted to closely resemble that of the Goguryeo 'Gwanggaeto Stele.' As a result, the academic community has started to envision two radically different histories. One side interprets it as a symbol of conquest embedded in the heart of Silla’s royal capital by King Gwanggaeto in the year 400. The other side sees it as the very root of Silla calligraphy, inscriptions penned by Silla people themselves.

The academic symposium "Diverse Perspectives on the Unfinished Stele Fragment from the Western Side of Wolseong," organized by the National Research Institute of Cultural Heritage in Gyeongju and the Korean Society of Ancient History on August 13 at the Gyeongju National Museum, became the focal point where these two interpretations clashed head-on. Ki Kyungnyang, a professor of Korean history at Catholic University specializing in Goguryeo history, identified the writer of the stele fragment as a person active in the late 4th to early 5th century. This conclusion is drawn by comparing the shapes of four characters—jon (存), ching (稱), baek (白), and bul (不)—with those on the Dongjin Chanboja Stele (405) and the Former Qin Guangmu General Stele (368). If the stele was created during this period, his reasoning is that it is unlikely that Silla figures would have taken the lead in making it.

Professor Ki also cited the inscription on a roof tile found in the Jeongan Cheonchuchong tomb as evidence. He interpreted the partially preserved character as 'rak' (樂), and confirmed the phrases "Lelang (樂浪) General Zhao (趙將軍)" and "General Hu (胡將軍)"—interpreting these as Chinese-Koguryeo people active around Pyongyang. After the area was annexed by Goguryeo in 313, it became concentrated with people familiar with continental culture and stone monument production. Professor Ki speculates that these people likely participated in King Gwanggaeto’s 400 AD campaign as administrative scribes and stelae craftspeople.

'白 (left)' of the southern excavation of Wolseong in Gyeongju, the 3D scanned '白 (center)', and the '白 (right)' from the Kaneko edition rubbing of the Gwanggaeto Stele

Ko Gwangui, Senior Research Fellow at the Northeast Asian History Foundation and an expert in epigraphy and Goguryeo script culture, advanced the discussion further. Recalling August 2024 when he first saw the stele fragment, he remarked, "I immediately recognized it as a Goguryeo-style epigraph very similar to the script of the Gwanggaeto Stele." At the site, he made a first reading of four characters—baek (白), cheon (天), gong (貢), and bul (不)—and requested a rubbing for detailed analysis.

The conclusion of Ko’s research is unambiguous: the Wolseong stele fragment is material evidence that the 'Gwanggaeto Stele Script,' the national script of Goguryeo, was forcibly implanted in the Silla royal capital. As a key piece of evidence, he points to the technique of leveling the surface as smooth as marble—a type of finishing rarely seen in early Silla stelae, which typically used natural stones with minimal surface processing until the early 6th century. In contrast, this kind of sophisticated technique is found only on major Goguryeo monuments such as the Jiban Goguryeo Stele, Chungju Goguryeo Stele, and Baekamseong stele fragments.

The stone itself is granite from Namsan in Gyeongju. Ko further interprets this as evidence of overwhelming Goguryeo control. He reasoned that, rather than hauling the heavy stone from the homeland, the conquerors likely conscripted local stone and labor from the occupied territory. As for the combination of five inscribed characters—“gwe(?)”, baek (白), cheon (天), gong (貢), and bul (不)—Ko reads them as language signifying a vertically subordinate relationship, where one party kneels and vows tribute to a superior.

Reading of the unearthed wooden tablets from the western side of Wolseong, Gyeongju, presented by Kwang Lee, Senior Research Fellow at the Northeast Asian History Foundation.

The most striking claim concerns the estimated time of breakage. Ko focused on the marks left on both fractured surfaces and asserted that the stele fragment was intentionally and systematically smashed, not broken by accident. He pinpointed the act to immediately after the “rooster hunt” in 464—a massacre of Goguryeo troops stationed in Silla—citing the 450 AD murder of a Goguryeo envoy by Silla as an irreconcilable precursor to the breakdown of relations between the two kingdoms. He claims that Silla intentionally shattered the symbol of subjugation that once stood in the heart of the royal capital and discarded it in the moat. Ko even invoked the Roman concept of 'Damnatio Memoriae,' the destruction of monuments to erase enemies from collective memory, suggesting, "This was Silla’s declaration of restored sovereignty."

Jeong Hyunsook, research professor at the Institute for Calligraphy & Culture at Wonkwang University and an expert in the history of calligraphy, offered a different view. She meticulously compared each character on the stele fragment—jon (存), chon (村), ching (稱), baek (白), cheon (天), gong (貢), bul (不), do (渡)—with their counterparts on the Gwanggaeto Stele. The characters on the fragment generally have looser spacing and taller, rectangular shapes reminiscent of standard script (haeseo), whereas the Gwanggaeto Stele displays denser, flatter regular script, typical of clerical script (yeso). Professor Jeong emphasized that while both stelae are in the “Goguryeo-Clerical” style, their structural arrangement and brushwork patterns are distinctly different.

So where did the script style of the Wolseong stele fragment originate? Professor Jeong pointed decisively to the ‘Pohang Jungseong-ri Silla Monument’ erected in 501 as a clue. The use of abbreviated dots for hooks and sweeps in the calligraphy is a highly distinctive and uncommon feature shared by the two monuments, making coincidence an unlikely explanation. Based on this, she believes the Wolseong stele predates the Pohang monument and should be viewed as a fifth-century inscription. She reasoned that Silla people absorbed and adapted the clerical script of the Gwanggaeto Stele and developed their own calligraphic style. "We have now traced the root of sixth-century Silla calligraphy back to within Silla itself," she concluded.

Pohang Jungseong-ri Silla Monument / National Gyeongju Cultural Heritage Research Institute.

Lee Jaehwan, professor of history at Chung-Ang University, started his analysis by observing that the stele is damaged on all sides, leaving no extant phrase with more than two or three consecutive characters. This makes it difficult to determine the stele’s creators or its era. Nevertheless, by tracking the usage of character pairs in Chinese and Korean historical texts, he suggested that ‘gongbul (貢不)’ is most likely an abbreviation of 'gongbulgweol (貢不闕),' meaning "the tribute is never omitted." This phrase also closely mirrors the words of Emperor Taizong of Tang, who described Silla as "never omitting tribute" in an imperial edict to Goguryeo.

He briefly considered the possibility of a late seventh-century Tang-Silla context, but ruled it out as there would have been no reason to use clerical script at that time. He instead argued that the most likely context is the relationship between Goguryeo and Silla in the late 4th to early 5th centuries. "Given the current state of evidence, the most reasonable position is to withhold interpretation and instead remain silent," he added.

The subsequent discussion saw scholars tempering the more assertive arguments. An Jeongjun, professor of Korean history at the University of Seoul, criticized the very notion of a ‘Goguryeo state script’ as a logical leap linking calligraphic similarity with political dominance. He also urged caution in tying the date of breakage to specific historical incidents. Baek Seungok, secretary-general of the Gaya Tumuli World Heritage Center, questioned the readings themselves, stating that among the eight characters Professor Ki presented, only seven are agreeable, and he expressed doubt about the character ‘rak’ (樂) used as evidence for the reading of General Zhao and General Hu. Kwon Inhan, professor of Korean language and literature at Sungkyunkwan University, suggested that the character read as 'chon (村)' by Professor Jeong could also be 'jae (材),' further casting doubt on the basis for these interpretations.

Western side of Gyeongju Wolseong Sueup Bipyun (left) and its 3D scanned grayscale shading effect photo (right) / National Gyeongju Cultural Heritage Research Institute·National Gyeongju Museum.

Who inscribed this monument and when remains unresolved. Perhaps it is precisely this distance—the uncertainty—that makes the artifact so valuable. Paradoxically, it is because fewer than ten characters can be read that it is possible to conjure up tales of dynastic triumph, expressions of surrender, or proud historical records. It is possible that a missing fragment, still buried somewhere, will one day be miraculously found and bring these academic debates to a conclusion. Until then, the stone fragment from Wolseong, no bigger than a palm, will remain as a testament embracing multiple conflicting histories at once.

This content was produced with the assistance of AI translation services.