Language’s Ancient Switches: Human Genetic Clues Shared With Neanderthals

Language is often seen as the signature ability of modern humans, but its biological foundation may be older than the idea that it is unique to Homo sapiens. According to a study published in Science Advances, a set of ancient DNA regulatory regions that make up only a tiny fraction of the genome is significantly associated with individual differences in language ability among people today. These regions can also be seen in the Neanderthal genome, suggesting that the brain-development conditions related to language may have appeared before the two human lineages diverged.

The focus of the research was not genes that produce proteins, but regulatory sequences known as human ancestral rapidly evolved regions. This type of DNA acts more like a volume knob for gene activity, influencing when, where, and how strongly genes related to neural development are activated. The research team estimated that although these regions account for less than 0.1% of the entire genome, they carry a disproportionate share of the genetic influence associated with language ability.

The data foundation for the analysis came from language-ability assessments and saliva DNA samples collected in the 1990s from about 350 students in Iowa in the United States. Years later, researchers sequenced the preserved samples and built a polygenic score stratified by evolutionary age to distinguish how genetic variants that emerged in different periods are linked to language performance.

The results point to an intriguing depth of time: the regulatory signals associated with language did not accumulate rapidly only in the modern human lineage, but can be traced back to an earlier common human ancestor. The research team believes this means the “hardware” for language may already have been partly in place very early on. As for whether Neanderthals had language close to that of modern humans, this study cannot answer that directly; it can only say that their genome retained several regulatory conditions associated with language ability.

The study also raises an evolutionary trade-off. These regulatory regions may be involved in fetal brain and skull development. If they further increased brain capacity or head size, they could have raised the risks of childbirth. In an era without modern obstetric conditions, language-related neurodevelopmental pathways may have reached a ceiling for reproductive safety very early and therefore remained relatively stable, while other genetic pathways related to cognition, but not directly increasing the fetal skull burden, may have continued to change.

However, this remains an association study. The sample size was limited, the participants’ backgrounds were relatively concentrated, and language ability is also strongly shaped by family, education, and social environment. The research team next hopes to use the fact that the original participants have now established families to further separate genetic influences, the language environment provided by parents, and the effects formed by the interplay between the two. The real message may not lie in finding a single “language gene,” but in seeing how language draws out the long arc of the human voice through ancient regulatory switches, developmental constraints, and lived environments.