The world today has roughly 7,000 languages. Two thousand years ago, a new study argues, it may have had ten times as many, Scientific American reports. If so, human linguistic diversity has collapsed dramatically, and the study offers a theory of why.
What the study claims
The headline figure is an estimate, arrived at indirectly, and its scale is striking.
According to Scientific American, the research, published in the journal Science, estimates that humans spoke somewhere between 20,000 and 75,000 languages around 2,000 years ago, roughly ten times the number spoken now. That is a claim about a "golden age" of linguistic variety in the deep past, followed by a long decline to the present.
How you could possibly estimate that
The obvious objection is that no one recorded the world's languages 2,000 years ago, so the method matters.
Scientific American reports that the lead author, Damián Blasi, and his team at Harvard reconstructed ancient linguistic diversity by modeling rather than counting. They began about 12,000 years ago, when the global human population was between 4 and 7 million and, they estimate, there were 5,000 to 6,000 languages. From that starting point they used computational modeling to trace thousands of possible trajectories of how the number of languages could have changed over time, arriving at the figures for 2,000 years ago and the decline since.
That is worth being clear about: this is an inference from models and assumptions, not a headcount. The strength of such work lies in its method and its plausibility, not in direct evidence, which is exactly why it is contested.
The proposed explanation
The interesting part is not just the number but the story of its fall.
Scientific American reports that the researchers attribute the decline largely to the rise of large, dominant states, with the Roman Empire as an example. Blasi put it vividly: "At some point, a few of these [languages] explode, and they wipe out everything that's around them." The idea is that expansive empires and states spread a small number of languages over huge territories, displacing the many local tongues that had existed before. The same forces, extended through later history, would help explain how the world got to its current, comparatively concentrated 7,000.
The skeptics
Responsible science reporting means giving the doubters their say, and here there are prominent ones.
Scientific American reports that not all linguists accept the conclusion. Lyle Campbell, of the University of Hawaii, called the estimate "highly speculative." Johanna Nichols, of the University of California, Berkeley, raised a subtler objection: that many of the ancient communities counted as speaking separate languages may in fact have spoken mutually intelligible dialects, which would mean the true number of distinct languages was lower. Those are serious challenges, both to the raw figure and to what counts as a "language" at all.
We report the study's claim as a claim, and the skepticism alongside it, because that is where the science actually stands: an eye-catching estimate, a plausible mechanism, and genuine disagreement about whether the numbers hold.
Why it matters
Even with the uncertainty, the underlying point is not really in dispute.
Whether the peak was 20,000 languages or 75,000, and whether it was 2,000 years ago or another date, the broad picture, that human linguistic diversity was once far greater and has fallen sharply, aligns with what is well established: languages are disappearing, and the process has deep historical roots. Today a large share of the world's languages are endangered, many with only a handful of elderly speakers left. A study like this reframes that loss on a longer timescale, suggesting the world we live in, dominated by a relatively small number of widely spoken languages, is the thinned-out remainder of something once much richer. That is a claim worth debating precisely because what is at stake, the diversity of human thought and culture that languages carry, matters well beyond linguistics.
