All Languages Linked To Common Source

Old Wolf writes "A New Zealand evolutionary psychologist, Quentin Atkinson, has created a scientific sensation by claiming to have discovered the mother of all mother tongues. 'Dr Atkinson took 504 languages and plotted the number of phonemes in each (corrected for recent population growth, when significant) against the distance between the place where the language is spoken and 2,500 putative points of origin, scattered across the world (abstract). The relationship that emerges suggests the actual point of origin is in central or southern Africa, and that all modern languages do, indeed, have a common root." Reader NotSanguine points out another study which challenges the idea that the brain is more important to the structure of language than cultural evolution.

Comments

Re:Phoneme counts

[snowgirl]

I read this earlier, and at first glance it's counter-intuitive. Why would older languages have more phonemes and not less?

This is a good question and in fact it's right on the money as a way to argue against this study. Languages change, this is true, but they don't change in a monotonic way. Some languages gain phonemes, some languages lose phonemes. That's how linguistic change works. In the same way, some languages have a complex synthetic syntax, and some have a relatively simple creole-like isolating syntax. When languages become too simplified, children learning the language create novel complications to fill out niches.

As an example Hungarian has only about 11 irregular verbs, but this is because their verb system is complex and unwieldy, meanwhile English with its incredibly simple verbal patterns has numerous (and in fact no single authoritative count) of irregular verbs.

Chinese has a limited syllable construction pattern, and as a result has picked up tones to make distinctions between words, while Japanese with a similarly limited syllable construction pattern uses longer words, and Hawai'ian with even more strict syllable construction rules and phonemes has gone for yet longer words. (I was surprised to realize that "Meli Kalikimaka" is literally "Merry Christmas" pounded into the strict Hawai'ian phonemic rules.)

So, while I think his ideas might have interest, and could be intriguing, there is also the fundamental problem that he's making a deep assumption of monotonic language "growth" that is not supported by reality. I imagine it's similar to measuring which animal has more evolutionary change by it having more teeth. But everyone wants to be the person to prove that all the world languages are related, right?