Mother Tongues

a journey through language


The Evolution of Language: Three Things We’ve Learnt

A coupla hundred years ago, our good ol’ mate Darwin thought up this totally radical idea called evolution. This had two noticeable results:

  • the church got their knickers in a twist,
  • and it revolutionised our understanding – of biology, humanity, pretty much EVERYTHING-y.

Cheeky Darwin and his pesky evolution theory got the Church’s knickers in a twist.

Since then, all manner of disciplines have co-opted the evolution concept: stellar evolution! Directed evolution! Music evolution!

Linguistics is no different. In my last post, I briefly compared biological evolution to language evolution. We also touched on this topic in our foray into the origins of language.

This time round, we’re gonna get fully sucked into the mind-bending twister of language evolution. Here’s three wicked things we’ve learnt from investigating how our mother tongues have transformed.



1. What Ancient People Were Like

By tracing the evolution of languages back in time, we’re essentially creating a knowledge-wormhole to ancient societies and their cultures.

Take the Proto-Indo-Europeans, for example. We talked a lot about their zombie PIE language in the previous post – but what I didn’t mention is that we can learn a lot about PIE society based on this reconstructed vocabulary. The words used by people are a mirror to their culture – in the PIE case, a ~~time travel mirror~~.

We know that the PIEs lived in a location with snow because they (and all their daughter languages) used the word sneigwh-. Similarly, we reckon they lived inland, because they didn’t have a general word for sea. We even know they worshipped a sky god and had domesticated both horses and cattle. You can read more about PIE society here.


Proto-Indo-European society didn’t have a specific word for “sea” so they probably lived inland.

2. How Humans Populated the Corners of the Earth

By examining how languages have evolved and diversified, we can also map out migration patterns across the globe.

For example, researchers built a phylogenetic tree of around 40 indigenous languages from North America and Central Siberia. This allowed them to construct a model of how the languages diffused across the two continents.

The results suggest that dispersal first occurred down the western coast of North America, followed by a “back-migration” into Siberia and then further diffusion throughout inland North America.

That is, there wasn’t a one-way migration from Asia to the New World via the Bering Land Bridge – human migration is more complex than we first thought. This lends support to the “Out of Beringia” theory.


Map showing the dispersion of people and their languages from Beringia. Reproduced from this paper.

Another migration-mystery that has bewildered researchers for aaaages is the “Austronesian Expansion” – or how the heck the Polynesians made it to all those tiny Pacific islands.

There are two main theories: a gradual settlement over 30,000 years (the “slow boat”) and a more recent “pulse-pause” scenario with ancestors migrating from Taiwan some 5,000 years ago.

To test these hypotheses, linguists constructed an evolutionary tree of Pacific languages. But they weren’t the only ones investigating this conundrum: biologists constructed a classic phylogenetic tree of a species of gut bacteria. By combining the evidence from both approaches, the scientists were able to discern the probable migratory routes – and they matched pretty well with the Taiwan-origin hypothesis. The “pulse-pause” pattern of migration was also replicated in the linguistic tree – it suggested one migration pulse of 7,000km over just 1,000 years.

Pretty cool, hey.


Linguistic evidence and genetic evidence from gut bacteria solved the Pacific migration mystery.

3. Language Evolution Really is like Biological Evolution

Evolution as a scientific theory has only been comprehensively analysed in its biological habitat. But it seems as though this concept may operate in a cultural domain too. This makes intuitive sense: cultures (and their languages) do change over time.

But the similarities go deeper than just a simple “shit changes” sense. Language is a complex adaptive system: it is an interconnected network of small, related parts (words) that work together to adapt to changes (cultural and environmental context) to ensure the survival of the collection as a whole (language).

Recently, some scientists have been explicitly adopting a Darwinian approach to culture.

In one study from 2008, linguists observed the evolution of an artificial language from random to highly structured, in a controlled laboratory setting.


The artificial language becomes more precise and more ordered over time – it evolves! Figure reproduced from this paper

These experiments revealed that, “languages transmitted culturally evolve in such a way as to maximise their own transmissibility.”

In non-wanky-science-speak: the language itself evolves in order to increase the chance it will be passed on (and stay “alive”). It’s like language is a cultural parasite, using human hosts to reproduce.


A language parasite and a human host? (Based on the real-life tongue-eating louse)

Evolutionary theorists have also used biology techniques to reconstruct language family trees using vocabulary “genes.” (Just like we talked about with the reconstruction of Proto-Indo-European.)

Using statistics, they were able to figure out that 10-30% of all words emerged during language splitting events – i.e. word “speciation” occurs in bursts of change, not gradually. This mirrors what we observe in biological evolution: about 22% of all genetic differences arise from bursts of sudden change.

Perhaps these “word explosion” events may reflect groups of people trying to establish a unique identity through their own distinct language. Kinda like those kids in primary school who thought they were so cool and different speaking Gibberish.

One Last Thing

If we take a step back and look at all these cool language evolution-y things, we can see how diverse this field is: from neuroscience to linguistics to evolutionary biology to archaeology (and more).

We’ll continue exploring this cross-disciplinary complexity in posts to come. Keep checkin’ in for more rad yarns! (And more shitty illustrations.)





PIE: The Zombie Language

It’s 6,000 years ago and you’re one hot Neolithic mama, gossiping with your bestie about the latest tribal happenings. Perhaps, you’re discussing which Bronze Age sugar daddy has the most cows (highly desirable) or what kind of troublemaking the warrior-class hotties are up to.


PIE babes love boys with cows

But what language are you speaking? ‘Cause it certainly isn’t English.

Yeah, turns out 10,000 BC wasn’t historically accurate. Who woulda thought?

So it’s not English. And it’s not some prehistoric grunt-speech. It turns out you’d be speaking a weird ol’ language called Proto-Indo-European (or PIE, for short). The thing is, this language existed before humans had invented writing. So how the hell do we know about it?

This week, I put on my detective hat to investigate this lingo-enigma.

How do we know about PIE?

Just like plants and animals, languages evolve. They change over time and diversify. Similarly, just as chimpanzees and humans share a common ancestor on the tree of life, modern languages have descended from a common ancestor too. You may be familiar with Latin, the granddaddy of the Romance languages (like French, Spanish and Italian).


A linguistic evolutionary tree. Reused with permission from this Nature Reviews Genetics paper by M. Pagel.

In the 19th century, linguists noticed similarities between Latin, Ancient Greek and Sanskrit. They wondered, “Could a GREAT-granddaddy ancient language have spawned all these?”

The answer was yes, and the great-granddaddy was Proto-Indo-European. But PIE was dead, so linguists had to do a lil Easter-Jesus-style trick.

Here’s how we raise languages from the dead:

First, compare specific words. If words that have similar meanings also have similar sound structure, then it is possible the languages are related. A good starting place for comparison is numbers – I’ve used Spanish, Italian and French for illustrative purposes.


If it means the same, has similar phonetic structure and sounds the same, the languages might be related.

But you also have to look for sounds that correspond in a regular pattern – if you can find lots of these “correspondence sets”, then it’s pretty certain your languages are related.


Correspondence sets: a “k” sound in Spanish and Italian corresponds to a “sh” sound in French in a regular pattern. The correspondence set is k:sh and we write the ancestral sound as *k. (In this case we know the predecessor – the latin “c-” which has a “k” sound.)

Linguists also discovered that sounds evolve in specific ways. By applying these rules in reverse, they can reconstruct ancient languages.

This type of linguistic reconstruction is called the comparative method. Nowadays, computer programs can do most of the fiddly work.

So PIE is a zombie language. Of course, we don’t know for sure that we’ve got it 100% correct – but it’s a pretty good guess.


PIE is a zombie language

It’s PIE time

Most linguists reckon that PIE was spoken around 3,500 BC – that’s roughly one thousand years before the Egyptians starting building pyramids (WOAH). But other hypotheses place the emergence of PIE as far back as 10,000 BC – when woolly mammoths were still kickin’ it.

The most popular theory contends that PIE was the language of people who lived in the Pontic-Caspian Steppe – a region roughly spanning Eastern Europe and Central Asia.


As the Proto-Indo-Europeans migrated, the PIE language spread and diversified into many languages spoken today. Image by Dbachmann (Own work, CC BY-SA 3.0) via Wikipedia

While PIE is a prehistoric language, that doesn’t necessarily mean that it’s simpler than modern-day lingos. In fact, PIE was pretty complex: it had tenses and other modifications for verbs and nouns could be modified to indicate number, case and gender. PIE may be a zombie, but it’s an elaborate zombie!

What did PIE sound like?

In 1868, a German linguist by the name of August Schleicher wanted to answer this exact question. So he wrote a little tale about a sheep who meets some disagreeable horses. Here’s a recent version, courtesy of Archaeology magazine:

Sounds pretty trippy, hey.

Here’s the English translation:

The Sheep and the Horses

A sheep that had no wool saw horses, one of them pulling a heavy wagon, one carrying a big load, and one carrying a man quickly. The sheep said to the horses: “My heart pains me, seeing a man driving horses.” The horses said: “Listen, sheep, our hearts pain us when we see this: a man, the master, makes the wool of the sheep into a warm garment for himself. And the sheep has no wool.” Having heard this, the sheep fled into the plain.

Of course we don’t know what written symbols, if any, speakers of PIE used. So linguists use special symbols and combinations of letters to represent different sounds. Here’s the written text that tells you how to pronounce it in PIE:

H2óu̯is h1éḱu̯ōs-kwe

h2áu̯ei̯ h1i̯osméi̯ h2u̯l̥h1náh2 né h1ést, só h1éḱu̯oms derḱt. só gwr̥hxúm u̯óǵhom u̯eǵhed; só méǵh2m̥ bhórom; só dhǵhémonm̥ h2ṓḱu bhered. h2óu̯is h1ékwoi̯bhi̯os u̯eu̯ked: “dhǵhémonm̥ spéḱi̯oh2 h1éḱu̯oms-kwe h2áǵeti, ḱḗr moi̯ aghnutor”. h1éḱu̯ōs tu u̯eu̯kond: “ḱludhí, h2ou̯ei̯! tód spéḱi̯omes, n̥sméi̯ aghnutór ḱḗr: dhǵhémō, pótis, sē h2áu̯i̯es h2u̯l̥h1náh2 gwhérmom u̯éstrom u̯ept, h2áu̯ibhi̯os tu h2u̯l̥h1náh2 né h1esti. tód ḱeḱluu̯ṓs h2óu̯is h2aǵróm bhuged.

Pretty unintelligible, unless you have a background in linguistics. This particular recital isn’t fixed either – it may change over time as linguists and archaeologists discover more about PIE speakers.

Can I learn PIE?

If the above recording hasn’t put you off this tricky language, then you can definitely give learning it a go. The video below can teach you how to say another fable in PIE – but be warned, some of the sounds required are very difficult for English-speakers to enunciate properly.

Video via Xidnaf on Youtube

I reckon I have an okay knack for languages, so I decided to give it a shot. Here’s my attempt:

Wow, PIE is frickin’ hard.

If you’d like a more rigorous PIE learning experience, check out Dnghu, a non-profit organisation dedicated to resurrecting the Indo-European language. These enthusiasts even want to teach Indo-European as a second language for all European citizens, and have the EU adopt it as an official language.

Good luck with that, Dnghu – I think you’re gonna need it.




And then, there was Language…?

What better place to start than the beginning? That is, the ~~Beginning of Language~~ 

Consider this your crash course on this mysterious genesis.

You and I probably imagine a bunch of shaggy cavemen huddled around a fire, grunting like gorillas (or me early in the morning). We envisage that, gradually, these gorgling snurts morphed into “Would you pass the peas, please, love?”

Caveman Jesus doesn’t want to share his peas.

As it turns out, the emergence of language didn’t really happen like that.

In fact, pinning down this prickly issue is like pinning down the identity of Banksy – it’s a persistent mystery. At one point, linguists were even banned from discussing this question ’cause it’s so damn difficult.

Let’s try to unravel some of this spaghetti-complexity and dig down into the roots of language.

It’s tricky

The origin of language is a messy topic. Perhaps the single biggest factor contributing to this intellectual shitfest is the lack of direct evidence. It’s not like we can dig up the fossilised hollering of hunter-gatherer hominids.

If only it were this easy.

How can we figure it out then?

Investigating this puzzle is multidisciplinary, drawing on fields like linguistics, biology and archaeology. Here are a few methods scientists use:

  • examining the anatomy of fossils – which early Homo species had a larynx of the right shape for vocalising?
  • comparison to modern apes – how do differences in our brain structures give us different language capabilities?
  • finding and dating cultural artefacts – evidence of symbolism and rituals are often used to infer that their makers used language too
  • analysing language diversity using statistics and computer simulations

Although we are nowhere near a comprehensive answer, these sneaky strategies have unearthed a few clues.

When did language make an entrance?

By careful analysis of ancient hominid fossils, researchers have suggested that vocal speech may have arisen one million years ago. But it’s unlikely Homo erectus would have communicated via anything that resembles today’s languages.

Taking a different approach, a linguist analysed the structure of languages around the world. He found that African languages contained more distinct “sounds.” This suggests that older languages have a greater number of different sounds.

We can estimate how quickly new sounds develop, and based on this, some researchers say that language evolved roughly 350,000-150,000 years ago in sub-Saharan Africa. Incidentally, this coincides with the emergence of our own species, Homo sapiens.

Homo erectus is down with the lingo. Image adapted from Flickr user Tim Evanson.

Language: a sudden sensation?

Some peeps think that a single genetic mutation suddenly produced the power of language. The seed of speech was virgin-birthed into the human brain – and voilà! Thanks to a single mutant generation, we could talk.

Clearly this hypothesis is a bit wacky – how could your everyday Joe Bloggs caveman converse with a speaking genetic freak, if they themselves didn’t possess the capacity for language? Logic, man.

So it’s no surprise that the consensus lies more with a gradual approach. First, our ancestors could utter discrete speech sounds. “FOOD.” Over time, they began to string words into messages. “ME TARZAN, YOU JANE.” Then grammar developed. “I AM TARZAN AND YOU ARE JANE AND I NEED FOOD.” Seems pretty common sense.

But some other scientists think it’s not really gradual or sudden. It’s kinda a bit of both. They suggest that some revolutionary social transformation occurred, which liberated our (pre-existing but dormant) cognitive potential for language. They argue that “words are cheap” (i.e. they’re easy to fake), so there had to be some structure of mutual public trust for language to develop as our primary mode of communication.

Finally, there’s some people who think this is just a dumb question. They say, you can’t separate language from the wider development of human symbolic culture. It’s like trying to explain Instagram without first explaining smartphones and apps and all that shizz.

The virgin birth of language?

A change in our brain’s wiring

Language requires us to do some pretty funky cognitive tricks – like refer to things that are not in our immediate vicinity. So what happened to our brains? Other animals can vocalise, so it’s tempting to conclude that our language ability stems from a souped-up vocalising part of the brain. Interestingly, this doesn’t seem to be the case.

The region of the brain that allows apes to ooh-ooh-aah-aah is only used in humans for involuntary vocalisations, like screaming in pain. We humans use a different brain-bit for language, containing software that is responsible for both gestures and speech. This has led some to suggest that we pointed at stuff before we could talk about it. It might also explain why sign languages are equally as complex as spoken ones.

One last word

Clearly there’s no definitive, satisfying answer to the question of when and how and why language began, and there probably won’t ever be one.

But it’s one of those curly questions that strikes at the heart of what it means to be human – so scientists and scholars will continue to argue and hypothesise and experiment. And language will keep on evolving. And that’s pretty frickin’ cool.