Beneath the hills of Gyeongsan, in southeastern South Korea, archaeologists have spent more than four decades excavating a burial ground tied to the ancient Silla kingdom. Now, a genomic study of 78 individuals interred there has uncovered something the bones alone could never reveal: at least five of those people had parents who were closely related by blood, providing the first direct biological evidence that close-kin marriages took place during the Three Kingdoms period.
The findings, published in Science Advances in April 2026, reconstruct one of the most detailed kinship networks ever mapped for early Korean society. The study, whose corresponding author is Pere Gelabert of Seoul National University, draws on skeletal remains dating from the 4th through 6th centuries CE, a formative stretch when Silla was consolidating power across much of the Korean Peninsula.
A family vault, not a village cemetery
The Imdang-Joyeong burial complex sits near the modern city of Gyeongsan, about 20 kilometers from Gyeongju, the ancient Silla capital. According to the paper’s methods section, Yeungnam University Museum first excavated the site in 1982, returning in 1988 and 1989, recovering 259 human skeletons in total. Of that collection, 78 individuals had bones preserved well enough for ancient DNA extraction.
When the research team generated genome-wide data from those 78 people and ran kinship reconstruction algorithms, a striking pattern emerged. The burial ground did not represent a broad cross-section of the local population. Instead, the interred individuals were overwhelmingly related to one another, spanning multiple generations of the same extended family or cluster of families. Close relatives were buried near each other, suggesting the site functioned more like a family vault than a communal cemetery.
The genetic fingerprint of close-kin unions
Five of the 78 individuals stood out. Their genomes contained unusually long stretches of identical DNA inherited from both parents, a pattern geneticists call “runs of homozygosity,” or ROH. In practical terms, ROH segments grow longer and more numerous when someone’s mother and father share recent common ancestors. The levels detected in these five individuals are consistent with parents who were first or second cousins, or possibly even more closely related.
The team used the widely adopted 1240K capture panel, which targets roughly 1.2 million informative genetic variants and has become the standard tool for ancient human DNA comparisons. Raw sequencing data have been deposited in the European Nucleotide Archive under accession PRJEB104733, and supporting analysis scripts, genotype files, and intermediate outputs are stored in a public Dryad repository. That transparency means any qualified laboratory can independently verify the results.
A PubMed-indexed summary of the study emphasizes that the five high-ROH genomes stand out against an otherwise diverse but interconnected burial community, reinforcing the interpretation that the cemetery served a restricted lineage rather than a random sample of the surrounding settlement.
What the DNA cannot tell us
Confirming that close-kin unions happened is not the same as explaining why they happened. No surviving Silla-era legal code or literary text addresses rules around consanguineous marriage for this period, so the genomic signal stands without a matching written explanation.
Several scenarios remain plausible. The marriages could reflect a deliberate strategy by an elite lineage seeking to consolidate wealth and political influence, a pattern well documented in other ancient societies from Ptolemaic Egypt to Habsburg Europe. Alternatively, they could stem from the demographic constraints of a small, relatively isolated community with a limited pool of marriage partners. Or the answer could involve elements of both, shifting over the two centuries the cemetery was in use.
The five individuals with elevated parental relatedness represent roughly six percent of the sampled group. Whether that proportion holds for the full 259-skeleton collection, or for Silla-period populations beyond Gyeongsan, is unknown. Ancient DNA recovery rates depend on burial conditions, soil chemistry, and skeletal preservation, so the 78 genomes may not perfectly represent everyone interred at the site.
A finer-grained picture could come from cross-referencing each genome with the grave goods and tomb architecture recorded during excavation. Were the five high-ROH individuals buried with richer offerings or placed in more prominent tombs? The published genetic analysis does not systematically tie individual genomes to detailed artifact catalogues, leaving that question open for future work.
Why ancient DNA from Korea matters
For researchers studying East Asian genetic heritage, the Imdang-Joyeong project fills a stubborn gap. Ancient DNA studies have produced spectacular results in Europe, Central Asia, and the Americas, but the Korean Peninsula’s warm, humid climate breaks down genetic material far faster than the cold, dry conditions of Siberian permafrost or Scandinavian bogs. Extracting usable genomes from 78 individuals at a single Korean site represents a significant expansion of the available dataset for this part of the world.
The study also demonstrates how genomic tools can add dimensions that traditional archaeology cannot reach on its own. Decades of excavation and osteological analysis established the cemetery’s chronology and physical layout. DNA then revealed the biological ties linking individuals across tombs and generations, turning a collection of skeletons into a family tree.
Whether future excavations will uncover similarly well-preserved remains elsewhere in Korea, or whether Imdang-Joyeong proves to be unusually favorable for DNA preservation, remains to be seen. Comparative data from other Silla-period sites would help clarify whether the close-kin marriage pattern observed here was a local practice, a class-specific custom, or something more widespread.
Five ancient genomes and the questions they open
For now, the Imdang-Joyeong results are best understood as a detailed case study rather than a sweeping statement about Silla society. The genomic data convincingly document a multi-generational kin group in Gyeongsan, including a small number of individuals whose parents were closely related. That finding is robust, reproducible, and grounded in a site with more than 40 years of archaeological attention.
What it cannot yet do is tell us whether the couple who produced a high-ROH child saw their union as a strategic alliance, a family obligation, or simply the most practical option in a tight-knit community. Answering that will require more genomes, more sites, and perhaps the discovery of written records that have so far eluded historians. Until then, five ancient genomes from a hillside in Gyeongsan offer a rare, biologically grounded glimpse into how family, power, and burial practice intertwined during the rise of one of Korea’s most consequential kingdoms.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
