Ghost DNA and Missing Ancestors: Another Evolutionary Narrative Built on Unseen Foundations

A recent University of California, Berkeley study published in Science reports on a new computational method called TRACE. In it researchers claim that modern humans carry DNA from two previously unknown “ghost” lineages in addition to Neanderthal and Denisovan contributions. One ghost lineage is said to contribute roughly 0.5 to 1 percent of the genome of every person alive today and to have entered the human line in Africa more than 50,000 years ago. A second, even older “super-archaic” lineage, supposedly diverging around 1.8 million years ago, left traces that reached us indirectly through Denisovans, especially in Oceanian populations. Altogether, the study concludes that archaic DNA accounts for about 2 percent of the modern human genome.

The headlines present these findings as discovery. Closer examination shows they rest on a series of interlocking assumptions that turn statistical patterns into historical claims about extinct species we have never seen and whose DNA we have never sequenced.

Every date in the narrative (800,000 years for the ghost split, 1.8 million years for the super-archaic lineage, 50,000 years for the out-of-Africa expansion) depends on molecular-clock calibrations that presuppose long ages and gradual mutation rates calibrated against the evolutionary timeline itself. TRACE reconstructs ancestral recombination graphs from present-day genomes and flags segments whose ancestry appears unusually deep. Those segments are then interpreted as introgressed material from unknown hominin groups. No bones, no tools, and no ancient DNA sequences from the supposed ghosts exist. The method detects real patterns of deep coalescence; the leap to separate extinct species is an interpretive choice driven by the evolutionary tree model.

Successful, fertile interbreeding between modern humans, Neanderthals, Denisovans, and these proposed ghosts further undercuts the claim that they were distinct species. In ordinary biology, sustained gene exchange producing viable offspring is strong evidence that the populations belong to the same kind. The evolutionary narrative treats them as different species when convenient for the tree and close enough to exchange genes when the DNA data require it.

A more straightforward reading of the same data

When we set aside the deep-time assumptions and instead use the biblical timeline, measured mutation rates, and the actual limits of the evidence, a coherent alternative emerges that requires no ghost species and no millions of years.

Scripture gives the framework. God created humanity as one kind from one man (Acts 17:26; Genesis 1–2). All people alive today descend from Noah’s family after the global Flood roughly 4,500 years ago. The dispersion at Babel then produced isolated populations that rapidly diversified under selection, founder effects, and the adaptive capacity God placed in the human genome. Within that short history the observed genetic patterns make sense.

Pedigree studies that count actual de novo mutations in parent-child trios consistently report rates on the order of 1.0 to 1.5 × 10⁻⁸ mutations per base pair per generation. These are direct measurements, not rates inferred by forcing the data to fit a multi-million-year human-chimpanzee divergence. When these observed rates are applied to the number of differences between modern human lineages, the time back to a common ancestor shrinks dramatically into the range of a few thousand years once population bottlenecks and rapid post-Flood expansion are taken into account. The same principle applies to the segments TRACE flags as deep. Their apparent age is largely an artifact of the slow molecular clock the evolutionary model requires. Use the rates we actually measure in living people, and the need for 800,000-year or 1.8-million-year divergences disappears.

The “ghost” and “archaic” segments themselves are residual variation within the human kind. TRACE detects regions whose genealogical history looks deeper than the average modern human coalescence. In an evolutionary framework those regions are assigned to unknown species. In a biblical framework they are simply alleles that were present in the original created human population or that arose early after the Flood and then became unevenly distributed as families dispersed and isolated after Babel. Incomplete lineage sorting, population structure, and differential retention of ancestral variation all produce exactly the kinds of patterns TRACE is designed to find. No separate species is required.

Neanderthals and Denisovans fit the same picture. Their genomes are overwhelmingly human. The differences that exist fall well within the range of variation we see among living populations. They interbred successfully with the ancestors of present-day people because they were people, post-Flood groups adapted to particular climates and environments. The small percentages of their DNA retained in non-African genomes today are the expected result of limited intermixing among related human populations, followed by selection and drift over the subsequent millennia. The “super-archaic” signal TRACE finds enriched in Denisovan-related segments is likewise better understood as deeper residual variation within the same human family tree rather than evidence of a third lost species.

The study reports that roughly 0.5 to 1 percent of the genome of every modern person carries these flagged segments, and that the contribution is shared by African and non-African populations alike. That universality is precisely what we expect if the variation predates the final major post-Flood dispersals. The fact that some of these segments sit in regions previously called Neanderthal deserts does not prove a separate introgression event. It simply shows that different ancestral alleles were retained or removed in different genomic contexts under selection. Enrichment in immunity and metabolism genes is consistent with rapid post-Flood adaptation to new pathogens and food sources as families moved into new environments after Babel.

Unbiased science stays with what is actually observed. We have sequenced Neanderthal and Denisovan genomes. We have never sequenced a ghost genome. TRACE is a sophisticated statistical tool that reconstructs genealogies from modern DNA. It does not recover fossil DNA from unknown populations. The leap from statistical depth in a recombination graph to the historical existence of two extinct species is an interpretive choice driven by the evolutionary tree model. When that model is set aside, the same numbers describe ordinary human genetic diversity shaped by a recent bottleneck (the Flood), rapid expansion, and subsequent isolation and selection. No additional species are needed to explain the data.

This reading aligns with the low overall genetic diversity of humanity, the shallow coalescence times produced by observed mutation rates, and the clear biblical testimony that we are one blood. The patterns TRACE recovers are real. The ghost-species story built on top of them is not required by the data. It is required only by the deep-time assumptions that the evolutionary narrative brings to the table.

God designed the human genome with remarkable capacity for variation and adaptation. What we see today is the outworking of that design across a few thousand years of real history, not the leftover DNA of species that never appear in the fossil record and whose existence is inferred only when the data are forced into a multi-million-year framework. The more carefully we measure actual mutation rates and the more honestly we stay with what has been sequenced, the clearer the biblical picture becomes.