Language may be much older than previously thought

Posted on Tuesday, July 30th, 02013 by Austin Brown
link Categories: Long Term Science   chat 0 Comments

Homo_Models

A recent study brings together archaeological, biological and linguistic research to posit that spoken language may be much older than previously thought. Authors Dan Dediu and Stephen C. Levinson argue that emerging research indicates ancestors of modern humans as far back as 500,000 years ago may have been capable of spoken language, in contrast to most current estimates limiting the age to between 50,000 and 100,000 years.

A significantly different picture of language evolution emerges if we are to imagine that it’s been around for up to 10 times longer than previously thought. First and foremost is that it would predate biologically modern humans. Moreover, it would reach back prior to the evolutionary divergence of modern humans and Neandertals, raising the question of Neandertal speech capacity:

In sum, the evidence points to modern speech capacities in the common ancestor of Neandertals and modern humans. The auditory specializations for speech on the modern bandwidth are present, the morphology of the larynx looks modern, and air sacs have been replaced by a finely controlled pulmonic airstream mechanism for vocalization. In addition, the gene that is known to be involved in the fine motor control necessary for speech, FOXP2, has its modern form (although possibly not all of its modern regulatory environment).

In fact, the authors argue strongly for the likelihood that Neandertals had significant linguistic abilities and may even have influenced modern human language in detectable ways, mainly in the observed differences between Eurasian and African languages:

But if modern humans exiting from Africa interacted and interbred with Neandertals (and later, on their way through Asia, with Denisovans), then their contribution, we propose, might have shaped modern linguistic diversity.

Not only is it interesting to imagine that we may hear echoes of Neandertal in our own speech, but as the authors explain, the increased age of and broader biological and cultural context in which language must be placed implies a wider possibility space for linguistic diversity. In other words, linguists have observed around 7,000 spoken languages on Earth and attempted to derive universal characteristics from this dataset, but are likely working with only a fraction of the possible languages and language types the hominin mind and body make possible. That fraction – given these new developments – just got significantly smaller, calling even more into question any claims of universality in language characteristics.

  • Ogonix

    “auditory specializations for speech on the modern bandwidth are present, the morphology of the larynx looks modern,”
    That’s good for vocal language alright.

    Why assume there weren’t sign languages even before that? It’s like cognition wants to break out in one form or another– and until the vocal+acoustic stuff was fine tuned, I think the neural machinery was well in place to have sign languages: extremely articulate hands, gesturing (to a degree), and brains wonderfully tuned to be constantly asking “what’s that guy doing with his hands?”.
    The success that researchers have had in teaching (simplified) sign languages to other hominids is a strong indicator that something special is going on.

    In us modern H. sap., vocal language is universally favoured over sign languages (by which I mean simply: it’s the default), but sign language always seems ready to jump up at the slightest motivation.
    Notably: there’s the universality of gesture across cultures (but with wide variance as to what gestures there are and what they mean); and there’s the fact that sign languages are as easily learned as spoken ones. I.e., deaf children don’t have to struggle and fumble as if their brains were slowly rewiring so they can use arm-waving as some *desperate substitute* to vocal speech– To the contrary, they (and adults) take to sign languages as a duck takes to water. (In fact, it’s so easy that it makes the other language thing– learning to read and write– look vastly more difficult.)

    Lots of my friends are doing the “baby signs” thing, where, with their very young kids, they use a small-vocabulary simple sign language for quite a while until the kids’ vocal articulation is finally sufficient for switching to that as the main language medium.
    When I see the situation of “we can’t do vocal yet, so let’s use sign– then when vocal is available, we can switch”, I can’t help thinking: Am I watching a recapitulation of the historical development of language in the human species, and possibly our closer cousins?

  • Austin_B

    Yes, another line of inquiry. I didn’t mention this in the blog post, but the authors of the paper definitely see what you’re describing as a possibility. They also point out, though, that as vocal communication took off, selection pressures seem to have fine-tuned the vocal tract even further, whereas the hand doesn’t show evidence of similar derivation. (They don’t elaborate on what that might look like.) So, it would seem that if there were a period of gesture-based communication, it didn’t last long. The takeaway is that, to clear this picture up, we need to look for evidence further back in the fossil record that we have been.

  • Aaron

    And here we thought it was because of ancestral usage of Psilocybin expanding the pre-frontal cortex in early humanoids. That Terrence McKennafeller had us convinced.


Next Article navigateright