Have Ancient-DNA Studies Found Close Relatives Who Lived Hundreds of Years Apart?

DNA from excavated remains can reveal family relationships. Radiocarbon dating can estimate when each person died. What happens when those two measurements are compared?

Research report · 10 August 2026

Tested · no secure example found

Scientists can recover DNA from excavated bones and teeth and use it to determine that two people were close biological relatives. The same individuals can sometimes be dated by radiocarbon. Put those findings together, and an unusual question becomes possible: are any two people separated by only a few generations within a family, but by hundreds of years in time?

This study searched the public record for that combination. Such a pattern could be consistent with the small, exceptionally long-lived family proposed by this site. It could not identify that family as Noah’s descendants, but it would be a reason to investigate further.

How buried remains can answer the question

DNA does not always reconstruct the exact family tree. It may securely show that two people were related at about the second or third degree without showing every parent between them. For that reason, the study used conservative limits for each kind of relationship. A parent and child, for example, can die many decades apart; siblings can be separated still further. The question was whether the full radiocarbon-date ranges left a substantial chance that even these generous ordinary-human limits had been exceeded.

The comparison concerns the years between the two deaths, not the physical age assigned to either skeleton. Ordinary skeletal ageing estimates cannot reveal that someone aged unusually slowly without an independent chronological measurement.

Why this bears on the proposal made here

This site proposes that Noah and seven relatives entered an already populated world near Cudi Dağı around 2900 BC. It allows that this small family may initially have had exceptionally long lives and delayed maturity, then became more numerous, dispersed and married into surrounding populations.

If that happened, the family would not necessarily produce a detectable ancestry signal in the first generations. Much later, however, a close family relationship might occasionally connect people whose dates are surprisingly far apart. This study looked for that possible consequence. It did not ask whether extreme longevity is biologically possible, and it could not identify a genetic family as Noahite.

The primary period was 2200–1000 BC. A wider sensitivity analysis began at 2370 BC, when this site places the beginning of the dispersal era, and another began at approximately 2033 BC, the proposed end of Peleg’s life. The study was therefore not restricted simply to “after Babel.” It concentrated on the later period because a family of eight would need time to grow and become visible among the much larger surrounding populations.

What was available to examine

The audit combined reported family relationships in the Allen Ancient DNA Resource with longer-range family connections from the public ancIBD study. It covered the southern Caucasus, Anatolia and the Aegean, northern Mesopotamia and Syria, Iran, the Levant, Egypt and the Upper Nile within the frozen periods.

The broad count is not 232 independent families. Many pairs belong to the same connected family groups, and the public record is geographically uneven. Of the nine pairs suitable for the primary test, eight came from one infant family grave at Mygdalia in Greece. The ninth was a sibling pair from Oylum Höyük in Turkey. The usable evidence therefore represents only two family groups.

What the comparison found

For each primary pair, the analysis used the complete calibrated radiocarbon distributions rather than subtracting two midpoint dates. It calculated the probability that the people’s deaths were separated by more than the conservative limit for their reported relationship.

No primary pair reached the frozen 20% threshold for source investigation. The highest result was the Oylum Höyük sibling pair, with a 15.1% probability of exceeding the study’s generous 135-year sibling limit. No pair approached the stringent 95% threshold, and no anomaly survived the multiple-comparison rule.

Those failures are informative. An archaeological date attached to a tomb or cultural phase is not the same as a radiocarbon date measured from the person. Group tombs may be reopened for generations; bones may be moved or mixed; compiled databases may preserve an older estimate after a paper has revised it. Apparent gaps created in these ways cannot measure a human lifespan.

The one group worth dating again

DNA connects four people from two Armenian burial sites, Noratus and Beniamin, into one extended family group. For three of the connections, the central estimates of the current dates differ by about 322–325 years. That is the kind of result the audit was designed to notice.

But the dates producing the gap are not four independent measurements. One person has a date inferred from close relatives, two retain only the archaeological date of a group burial, and only the Beniamin individual has the direct date needed for a secure comparison. The exact family relationship has also not been published for two of the connections. Reuse of the Noratus tomb, movement of remains or an incorrect context assignment remain ordinary explanations.

What the result does and does not say

The responsible conclusion is therefore limited: public evidence presently provides no positive support for unusually long post-dispersal generations or extreme longevity. One context-dated family group merits new direct measurement, while the independent high-quality sample is far too small to exclude a rare lineage.

Study status and analysis materials

The questions, periods, regions, quality tiers, relationship limits and positive-result rule were frozen before new pair-level outcomes were opened. Relevant cases already seen in earlier work were disclosed in advance. This makes the work a prospectively frozen exploratory audit, not an independent registered confirmation. One execution-order deviation affecting the matched-control comparison is recorded in the technical report; it does not alter the primary pair results.

The full report and supporting files preserve the detailed thresholds, source checks, sensitivities and limitations behind this public summary.

This is an original exploratory analysis for this site. It has not been peer reviewed. The relationship-specific ordinary-human limits follow Sedig et al., Journal of Archaeological Science 125 (2021), 105452.