For roughly two centuries, people from the same extended family returned to a burial ground on the Mongolian steppe, yet they were not treated alike in death: some members of ancient Xiongnu lineages occupied the cemetery’s central spaces and elaborate tombs, while others were pushed toward its edges, and unrelated people could sometimes receive similarly prestigious treatment. By combining ancient DNA with grave architecture, burial goods, spatial patterns and computational analyses, researchers have reconstructed a social landscape in which family mattered—but wealth, status, alliances and symbolic affiliation mattered more.
The Tamir necropolis does not immediately look like the kind of place where archaeologists would expect social rules to be complicated.
Its graves preserve people who belonged to extended biological lineages stretching across five to six generations. Genetic analysis identified two principal lineages, called A and B, containing 10 and nine individuals, respectively. Another 28 individuals did not belong to either lineage and were placed in a separate group, C.
There are also familiar signs of family life in the arrangement of the dead. Spouses were buried together, while young children could be placed near grandparents. Members of lineage A were concentrated in the center of the cemetery, while members of lineage B were distributed across different areas, sometimes separated by more than 250 meters.
Taken together, those patterns could easily encourage a simple interpretation: this was a family cemetery, and biological descent determined where people belonged.
But the researchers behind the new study wanted to test that assumption rather than accept it.
Their question was not simply who was related to whom. It was whether genetic relatedness actually explained the organization of the cemetery better than other factors, particularly socioeconomic status.
That distinction turned out to matter.
The graves were not merely records of family relationships. They were also records of hierarchy.
Tamir preserves the burial ground of an elite community
The study focused on the eastern sector of the Tamir necropolis in Mongolia, a cemetery located on a promontory west of the Tamir–Orkhon river confluence and overlooking the surrounding floodplain.
The full necropolis contains 398 tombs, divided between a western sector with 344 graves and an eastern sector with 54. The researchers concentrated on the eastern sector because its large tombs represent an elite burial area dating from roughly the first century BC through the end of the first century AD.
Some of the tombs were substantial constructions. Stone circles could exceed 20 meters in width, while quadrangular burial pits could reach more than 6 meters in depth. Many graves had been looted, sometimes more than once, meaning that precious metals are probably underrepresented in the surviving archaeological record. Even so, excavators recovered iron objects, Chinese lacquerware, pottery and animal remains.
The cemetery began around 100 BC, after military defeats inflicted on the Xiongnu by Han forces between 129 and 119 BC. Those defeats were followed by a retreat north of the Gobi Desert into the region where Tamir is located.
The burial ground then remained in use for approximately two centuries.
Its final phase coincided with the defeat of the Northern Xiongnu by the Han in AD 89. The eastern sector appears to have been abandoned around AD 100, and the last tombs were notably poorer than some of the earlier burials.
That long timeline gave the researchers an unusual opportunity.
They could follow families across generations while also watching the material signs of status change as the political and economic circumstances of the community changed.
The graves carried a visible language of wealth
The researchers divided the burials into broad wealth categories using the objects and construction associated with each grave.
The wealthiest individuals were buried in tombs about 4 meters deep, often with two coffins and at least two prestige objects. Their graves could contain traditional Xiongnu prestige goods, Chinese lacquerware, precious stones such as jade and jasper, and metals including gold, silver and bronze.
At the other end of the spectrum, poorer individuals were generally buried about 2.5 meters deep, often without a coffin and with few or lower-quality grave goods. Intermediate burials were around 3.5 meters deep and generally included one prestige object. Horse bones occurred exclusively in wealthy graves, while goat offerings appeared only in graves belonging to group C.
These differences gave the researchers a way to separate two questions that can otherwise become tangled together.
A person could be wealthy because of their social position, and they could also belong to a powerful family. If family and wealth always traveled together, then the genetic lineages should closely match the archaeological patterns.
But if they did not, the differences could reveal something about how status operated inside the community.
The researchers therefore combined the genetic relationships with statistical and computational analyses of the burials.
Ancient DNA supplied the family tree
The genetic evidence was based on Y-chromosome and mitochondrial DNA markers, which help trace paternal and maternal lines, together with short tandem repeat data used to confirm familial relationships.
The final analytical sample contained 47 individuals from 44 graves. There were 24 males and 23 females, and only seven people were younger than 20 when they died. Two were younger than five.
Nineteen individuals belonged to the two major genetic lineages.
Lineage A formed a relatively concentrated group in the central area of the cemetery, while lineage B was more dispersed. The remaining 28 people formed group C, although the genetic evidence indicated that some of them might have had more distant relationships to one another.
The genetic relationships were therefore strong enough to reconstruct family continuity across several generations.
But the researchers did not stop with the family tree.
They wanted to know whether the objects and architecture of the graves would reproduce that same structure.
The first test showed that family and wealth overlapped—but not perfectly
The researchers began with linear discriminant analysis, a statistical method that can classify individuals into predefined groups based on their characteristics.
The analysis was able to distinguish the major genetic groups with high accuracy. Only two individuals were misclassified: one person from group C resembled lineage B, while one member of lineage B resembled group C.
The features associated with the lineage groups included lamps, faunal remains—especially horse bones—richer grave goods and single-grave arrangements. A machine-learning approach using XGBoost reached 70 percent accuracy and identified large, deep tombs, dual coffins and prestige objects among the traits contributing most strongly to the distinction between lineage members and unrelated individuals.
So there was clearly some relationship between family and the way people were buried.
But something else stood out.
The archaeological traits that helped identify the lineages were heavily connected with wealth and elite burial treatment.
That meant the same evidence could be read in another way: perhaps the cemetery was not organized primarily around family, but around social rank, with important families disproportionately represented among the wealthy.
The wealth analysis provided a particularly strong clue.
Wealth was easier to predict than family
When the researchers used their LDA model to predict wealth category, it achieved 100 percent accuracy.
That result does not mean that every family member was wealthy or that every wealthy person belonged to one of the two principal lineages.
Indeed, the study found substantial variation among couples. Of five identified couples, two were wealthy, one was intermediate, one was poor, and one consisted of a wealthy male and a poor female. The researchers suggest that the mismatched couple may relate to the economic decline occurring toward the end of the cemetery’s use.
The timing strengthens that interpretation.
Using radiocarbon dates, genealogical information and Bayesian chronological modeling, the researchers estimated the collapse of the Xiongnu Empire at around AD 85, just before the final two generations represented at Tamir. That estimate closely matches the historically reported date and supports the possibility of a progressive economic decline in the cemetery’s final generations.
The poorest final burials therefore were not necessarily signs of a family suddenly losing its identity.
They may instead reflect a wider deterioration in the economic conditions surrounding the community.
This was an important distinction.
Family relationships persisted through time, but the material ability to express status changed.
The cemetery divided people into social groups that did not follow the family tree
The researchers next used K-means clustering, an unsupervised statistical method that groups individuals according to similarities without first assigning them to a particular family.
The analysis produced an optimal solution of five clusters.
More than half of the members of lineages A and B fell into two wealthy clusters, one primarily male and the other primarily female. That indicates that kinship did contribute to some of the similarities between individuals. But several people were not grouped with their genealogical relatives.
The remaining clusters reflected different characteristics.
One included people associated with cranial-removal practices. Another corresponded to economically disadvantaged individuals. A fifth contained older males who also showed evidence of cranial removal. Several members of the principal lineages—including individuals in Tombs 03, 11, 19, 21, 32, 34, 36 and 40—fell outside the clusters expected from their family relationships.
The pattern was difficult to reconcile with a cemetery organized primarily by biological kinship.
The clustering instead supported a different explanation: socioeconomic status played a greater role than genetic relatedness in determining the layout and cultural organization of the necropolis.
The family tree was there.
It simply was not the whole map.
Some people were allowed near the center without being close relatives
The distinction becomes especially clear in individual burials.
Tomb 22 contained a woman who had no close genetic relationship to the people buried around her. Yet her son shared the same Y-chromosome lineage as the men belonging to lineage A.
The researchers propose that this pattern may indicate symbolic or social adoption. They do not present adoption as a proven explanation, but the genetic and spatial evidence shows that biological relatedness alone cannot explain the woman’s position within the cemetery.
Another unusual case appears in Tomb 19.
The grave contained a woman from lineage A who may have continued her family line because her father had no male heir. She was buried beside her husband in Tomb 23, a man carrying the same Y-chromosome haplogroup found in lineage B.
The researchers interpret this as potentially representing a redefinition of descent that gave the woman a distinctive status.
The cemetery therefore contains cases in which social relationships appear to have mattered alongside biological descent.
That becomes even more apparent when the researchers look at people who were physically separated from the central burial areas.
Being related did not guarantee a central burial
The researchers identify individuals who belonged to the broader community but did not represent the direct line of descent as being placed toward the cemetery’s margins.
One example is the brother in Tomb 48, who did not represent the direct line of descent and was interred at the periphery. Although he had access to the cemetery, his position appears to have remained symbolically marginal.
The researchers interpret the spatial arrangement as evidence that access to the most central funerary spaces reflected a combination of lineage continuity, status and legitimacy.
That distinction is crucial.
A person could belong to a family without occupying its most privileged social position.
And someone without close genetic ties could still become part of the cemetery’s social world.
The boundary was therefore not simply “family” versus “outsider.”
It was a hierarchy.
Women reveal another layer of the social system
The placement of women adds another complication to a strictly patrilineal interpretation.
Several women were buried alongside their husbands even though women would not normally form part of a strictly patrilineal funerary complex. Other women—including adults in Tombs 36, 34, 40 and 32, and young daughters in Tombs 21, 11 and 03—were buried at the cemetery’s periphery.
The researchers cautiously propose that this peripheral treatment could reflect marginality associated with gender, age or symbolic role.
At the same time, the woman in Tomb 19 appears to represent a possible exception to a strict male line of descent.
These cases do not produce one simple rule governing women’s burial.
Instead, they reinforce the study’s broader conclusion: several social principles could operate simultaneously.
Marriage could determine proximity.
Lineage could establish legitimacy.
Status could determine access to prestigious space.
And gender, age or social role could influence how an individual was positioned within that hierarchy.
Cultural traditions did not all travel through families
The researchers then asked a different question.
Even when families remained connected across generations, did their burial traditions remain unchanged?
To investigate that, they built a cultural phylogeny using 14 traits. Rather than tracing DNA, this analysis examined the transmission of cultural characteristics, allowing the researchers to explore whether particular funerary practices were passed down within families or spread through broader social networks.
The results showed that different traditions followed different paths.
Features of tomb architecture, including volume, depth and the number of coffins, were predominantly transmitted vertically. In other words, they tended to be passed from one generation to the next along familial lines.
The same was true of the inclusion of Xiongnu artifacts, prestige objects and other indicators of wealth.
Those patterns suggest that families could preserve recognizable ways of displaying status.
But not every tradition behaved that way.
Tomb orientation, the abundance of pottery, lamps and swords, and practices such as cranial removal appeared to be transmitted horizontally through broader social mechanisms.
Other characteristics, including inferred social roles and the presence of Chinese imports, showed vertical transmission but also strong social mediation.
Culture, in other words, moved in multiple directions.
Some things were inherited.
Others were learned, adopted or reshaped through wider social interaction.
The cultural family tree did not match the genetic one
The consensus cultural phylogeny produced three broad cultural groupings.
One clade contained nine of 11 related individuals and was closest to the earliest reference individual, Tomb 24. A second consisted mostly of unrelated people with culturally divergent characteristics. A third, intermediate clade contained both related and unrelated individuals.
The researchers interpret this structure as evidence for three cultural pools: an elite familial group, a general familial group and an unrelated group.
That pattern again shows why biological ancestry alone cannot explain the cemetery.
If genetic kinship had been the dominant force behind cultural organization, close relatives should have consistently resembled one another in their funerary traditions.
They did not.
The researchers therefore compared different evolutionary models to determine which factors best explained the cultural patterns.
The result was consistent across the analyses.
Models based on family relationships were repeatedly outperformed by models involving economic variables, the cladistically derived structure and even random models. Economic factors provided the strongest explanatory power.
The conclusion was not that kinship had no influence.
Rather, the evidence indicated that economic factors structured the cultural organization of the cemetery more strongly than biological kinship did.
The cemetery may have preserved shifting political power
The researchers also considered what the cemetery might reveal about the larger political organization of the Xiongnu.
The Xiongnu confederation had a dual political structure divided into Right and Left branches, with branches typically led by sons or other close relatives of the Shanyu, and the Left branch often associated with the heir.
The authors emphasize that detailed historical records for the specific social organization at Tamir are lacking. Because of that, they present the connection between the cemetery’s two lineages and the political branches as an interpretive hypothesis, not a demonstrated identification.
Under that hypothesis, lineage A could correspond to the Left branch and lineage B to the Right.
The proposed reconstruction is particularly interesting because the two lineages appear to have followed different trajectories.
Lineage A seems to have maintained a strict patrilineal pattern until the fourth generation. When no male heir was available, a socially modest son-in-law may have been incorporated.
Lineage B, by contrast, appears to have reinforced its patrilineal identity through a wealthy alliance involving an individual connected to Chinese elites in Tomb 13B.
The researchers propose that this might reflect a shift in political influence from one branch to another.
If that interpretation is correct, it could help explain why lineage B continued burying male descendants in the cemetery for two additional generations while lineage A eventually lost its central position.
But the uncertainty remains important.
The cemetery can reveal patterns in descent, wealth, burial treatment and spatial organization.
It cannot, by itself, prove the exact political identities of the people buried there.
The final generations became poorer as the Xiongnu world changed
The chronological evidence provides another connection between social status and political change.
The eastern sector of Tamir was used from the early first century BC until roughly the end of the first century AD. The earliest phase followed the Xiongnu defeats by the Han, while the final burials occurred around the period when the Northern Xiongnu were defeated in AD 89.
The cemetery appears to have been abandoned around AD 100.
The last graves—Tombs 18, 32 and 11—were notably poorer.
The researchers’ Bayesian chronological model places the collapse of the Xiongnu Empire around AD 85, just before the final two generations represented in the cemetery. That close correspondence supports their hypothesis of a progressive economic decline during the cemetery’s final phase.
The change therefore appears in both history and archaeology.
As the political world surrounding the community shifted, the material expression of status within the cemetery changed as well.
That makes the graves more than static snapshots of social rank.
Together, they form a sequence.
The study’s sample is small, and some graves remain difficult to interpret
The researchers were able to analyze 47 individuals, but the full archaeological record was larger.
Ten graves were excluded from the analysis: six came from earlier excavations, two had incomplete information because of later disturbance, and two belonged to the much later Turkic period rather than the Xiongnu horizon.
Looting also complicates the archaeological record. Many tombs had been disturbed, meaning that valuable objects may no longer be present even when they were originally deposited.
Those limitations matter when interpreting wealth.
The researchers therefore do not treat every absence of a grave good as proof that the object was never present.
The study also contains individual cases whose meaning remains uncertain.
The person in Tomb 11, for example, was the only identified representative of the sixth generation. He was returned to the cemetery in a secondary burial and placed near his grandparents. The researchers interpret his inclusion as appearing more likely to reflect emotional considerations than symbolic or economic ones.
Such cases are a reminder that no statistical model can turn every ancient decision into a single rule.
People could make choices for reasons that leave only faint traces in the archaeological record.
What the cemetery ultimately reveals is a hierarchy layered over a family tree
The Tamir necropolis began to make sense only when the researchers stopped treating kinship as the complete explanation.
The genetic evidence clearly reveals extended families.
The spatial evidence clearly shows that some relatives were concentrated together.
The archaeological evidence clearly records inherited differences in tomb architecture and wealth.
But none of those patterns alone explains the whole cemetery.
Some relatives were separated.
Some unrelated people shared the cultural characteristics of elite burials.
Some women crossed the boundaries implied by patrilineal descent.
Some traditions stayed within families for generations.
Others spread through wider social networks.
And the strongest statistical models repeatedly identified economic factors rather than biological kinship as the better explanation for the cultural organization of the site.
That makes Tamir an unusually revealing case study in what ancient DNA can—and cannot—tell us about past societies.
A genetic connection establishes biological relatedness.
It does not automatically establish social identity.
People can become connected through marriage, adoption or alliances. Conversely, people who share ancestry can occupy very different positions within the same social system.
The researchers’ combined approach was designed to make those distinctions visible.
The dead were arranged according to more than who they came from
The most compelling feature of Tamir is therefore not simply that archaeologists found ancient family lines.
It is that those family lines did not determine everything.
The cemetery preserved a social order in which wealth, lineage continuity, status, political authority, marriage alliances and symbolic affiliation intersected.
The central spaces appear to have been especially important. Members of lineage A were concentrated there, while wealthy men from lineage B could be buried with their wives near the lineage A cluster. Women who married outside their lineage were placed in the northern sector, while individuals without direct descent connections could remain at the margins.
The pattern is neither purely familial nor purely economic.
It is layered.
Family could provide legitimacy.
Wealth could amplify status.
Marriage could create new connections.
Political alliances could reshape the position of a lineage.
And cultural practices could move beyond the boundaries of biological descent.
The researchers ultimately describe Tamir as a funerary landscape organized along socioeconomic lines, with kinship playing a secondary or legitimizing role.
That conclusion changes the way the cemetery’s graves can be read.
They are not simply places where related people were buried together.
They are records of a society deciding, generation after generation, which relationships mattered most.
And the answer was never just blood.
Publication details
Juhyeon Lee et al, Genetic population structure of the Xiongnu Empire at imperial and local scales, Science Advances (2023). DOI: 10.1126/sciadv.adf3904. www.science.org/doi/10.1126/sciadv.adf3904
