Linguists have long known that when cultures collide, languages rub off on one another. We borrow words, swap sounds, and even reshape grammar. But charting those exchanges across centuries and continents is hard when the written record is patchy or nonexistent.
A new study flips the problem on its head: instead of starting with history books, the researchers read our DNA to reconstruct past human contact – then asked what happened to the languages.
DNA traces human language history
A team led by Anna Graff at the University of Zurich pulled together genetic data from more than 4,700 people in 558 populations. They paired this with two of the world’s largest linguistic databases, which catalogue everything from word order to consonant inventories across thousands of languages.
Genetic “admixture” – the telltale signature of populations mixing – stands in for a reliable, globally comparable record of contact. With that proxy in hand, the team could identify more than 125 instances where groups clearly met and mingled, even if historians never wrote the encounters down.
What they saw was unambiguous: when people mix, their languages tend to move closer together. Unrelated languages spoken by populations with genetic contact were four to nine percent more likely to share features than expected.
That might sound modest, but across entire grammars and sound systems, it’s a strong, consistent nudge toward similarity.
One surprise was how steady that effect proved to be. Whether the contact was vast and recent – say, colonial movements between continents – or ancient and regional, like Neolithic migrations within Eurasia, the degree of linguistic convergence was strikingly similar.
“No matter where in the world populations come into contact, their languages become more alike to remarkably consistent extents,” said senior author Chiara Barbieri, now at the University of Cagliari.
That consistency cuts against a common assumption that only intense, prolonged encounters reshape grammar and sound systems, or that small-scale neighborly contact leaves barely a trace.
The genetic lens suggests languages are sensitive instruments, registering social touchpoints across the full spectrum – from trade and intermarriage to conquest and diaspora.
Not all grammar is transferable
Of course, not every part of a language is equally malleable. Some features seem to travel more readily. Word order patterns and certain consonant sounds were more likely to converge than deeper layers of morphology or prosody.
But here, too, the study counsels caution. The researchers didn’t find a one-size-fits-all rulebook that says, for example, “Nouns move; verb endings don’t.”
That challenges decades of “borrowability hierarchies” that rank which features can be shared across languages. Instead, the authors argue, social dynamics – prestige, power, identity, and the practicalities of multilingual life – can override structural inertia.
If a community prizes a dominant group’s speech, it may adopt conspicuous elements quickly; if it resists assimilation, the most “borrowable” features might barely budge.
You can see the everyday version of this push and pull in familiar loanwords. English picked up “sausage” from French after the Norman Conquest; centuries later, French borrowed “sandwich” from English.
Vocabulary swaps like these are the visible tip. Beneath the surface, the new study shows, subtler shifts in sound and syntax can ripple outward whenever people share space and stories.
The team even found the mirror image of borrowing: divergence on purpose. In some places, features became less alike after contact, not more.
That happens when communities lean into linguistic differences to mark who they are – tightening vowel systems, restoring older word orders, or reinforcing local pronunciations as a badge of identity. In other words, contact doesn’t always melt boundaries; sometimes it sharpens them.
This duality – convergence alongside intentional distancing – is central to how languages evolve. It helps explain why some neighboring tongues grow steadily more alike, while others cling to distinctiveness despite centuries of cohabitation.
DNA tells story of language
Methodologically, the study’s move is elegant. Historical documents and oral traditions can be rich but patchy, and for many regions and eras they simply don’t exist.
Genes keep a different kind of ledger. When populations intermix, they leave a durable statistical imprint that persists for thousands of years.
By aligning those genetic signals with language structures, the authors could quantify contact and its linguistic consequences on a global canvas – from Amazonia and the Sahel to the Pacific and the Arctic.
The approach also helps disentangle coincidence from contact. Two unrelated languages might independently develop similar features for internal reasons. But when that similarity lines up with clear genetic admixture between the speakers, the balance of probability shifts toward historical interaction.
Languages evolve in real time
Beyond satisfying our curiosity about how English, Hausa, Quechua, or Hmong came to be what they are, the findings carry a warning for the present. Contact has always been part of human life, but the pace and scale are accelerating.
Globalization, urbanization, climate-driven displacement, and land-use change are bringing communities together – and pushing others apart – in new ways. As that churn intensifies, we should expect languages to converge more in some respects and to splinter or vanish in others.
It also reframes the stakes of language loss. We often focus on shrinking vocabularies and disappearing oral literatures.
This study suggests that deeper, structural layers – sound patterns, grammatical architectures, the hidden wiring of a language – can erode under sustained contact, even when a language survives on the surface.
Protecting linguistic diversity, then, isn’t only about counting how many languages remain; it’s about preserving the full range of their internal variety.
DNA and language share histories
The research tells a familiar human story with fresh evidence. When people meet – through trade or conquest, migration or marriage – we share more than DNA. We pass around technologies, beliefs, recipes, and ways of speaking.
Some of those exchanges are voluntary, others are imposed; some knit communities together, others spark efforts to stand apart. Our languages, like our DNA, carry the scars and gifts of those encounters.
By treating genes as a time machine, this study gives linguistics a new global baseline. It doesn’t replace fieldwork, historical scholarship, or typological theory.
It makes them sharper, pointing to where contact likely mattered and how deeply. And it leaves us with a clear takeaway for the century ahead: as our lives intertwine ever more tightly, the sounds and structures we use to make meaning will change with them.
The study is published in the journal Science Advances.
—–
Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates.
Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.
—–
