It is our pleasure to welcome Simon Prentis, author of SPEECH! How Language Made Us Human, as our featured author for September. In his book, Simon offers a bold new theory for the emergence of language from animal communication, showing how a simple yet radical change was the key to everything that distinguishes humans from other species. He explores the growth of consciousness and the development of culture, religion and identity, and shows how the logic of language ultimately allows us to escape the inevitable traps they set for us.
Interact with Simon on our AoM forum here
“I couldn’t stop reading until I finished it. This book should be widely read!” –James Lovelock
“Crisp and clear – I agree with your hypothesis.” -Desmond Morris
“I look forward to reading it.” -Sir David Attenborough
“Bravo! A compelling read.” -Yoko Ono
The Mystery of Mysteries
We all love a good mystery. From the origin of the universe to the enigmatic traces of ancient civilizations, our curiosity about how we come to be what we are and our desire to unravel the secrets of the unknown is virtually insatiable – it’s the driving force behind everything we humans have achieved. Yet as mysteries go, none is more mysterious than the one you’re experiencing right now: the extraordinary mystery of language. For just by arranging and rearranging a handful of simple marks on this page, I can put thoughts into your head that you’ve never had before – though you don’t know who I am, have never met me, and quite likely never will.
When you stop to think about it, that’s real magic – happening right in front of your eyes. Without it, we wouldn’t be writing – or even thinking – about any mysteries at all. Yet here’s the thing: no one seems to know how we’re able to do it, or even when language begani. Given how important it is to us, that’s pretty astonishing.
It’s particularly astonishing because – as far as we know – none of the millions of other species on our planet have learnt the trick. Animals do communicate with each other, of course. Dogs are pretty good at showing us what they want, and can probably understand quite a few of the things we say to them. In the wild, we know that many species make distinctively different calls that warn each other of threats from particular predators. Bees do wiggle-dances that tell their hive-mates where to find food. But they certainly aren’t talking about Göbekli Tepe – or quantum physics. They’re not even discussing football. Or Love Island. It’s a grim world out there.
Why should that be? In almost every other respect, we’re pretty similar to our animal cousins. We all of us breathe, eat, drink, sleep, poop and make whoopee in much the same way. Our genome is almost 99% the same as the genome of chimpanzees, our closest relatives, nearly 85% the same as mice, and we share as much as 60% of our DNA with fruit flies. We even share 50% of our genes with bananas! So how come we’re the only ones who are living in vast, glittering cities and flying helicopters on Mars? What is it about humans that makes us so different?
One explanation might be that we are the result of some kind of extra-terrestrial interaction – exogenetic manipulation of our DNA by an alien life form that caused our brains to become bigger and better, bestowing the gift of language. That, after all, is the essence of the traditional religious explanation: ‘God’ gave us language to let us rule over the earth. Even the currently accepted theory for the origin of language is consistent with that possibility: Professor Noam Chomsky, the foremost theoretical linguist of our times, believes that humans have language thanks to a single genetic mutation specific to our species, one that probably occurred quite suddenly, around 70,000 years agoii.To expect evolution to have delivered that in one go is a big ask.
But it also presumes language must be the consequence of genetic changes to the brain. And what I’m going to argue here is that there’s a completely different, much simpler explanation – one that’s been staring us in the face all the time – but like all such obvious things, we haven’t seen it because we’ve been looking in the wrong place. Yet once you realise that language is more of a trick than a trait, you can see the fingerprints of its origin in the languages we still use today – and if I’m right, it also provides us with a persuasive answer to the ongoing puzzle of why our early hominid brains grew so rapidly in such a short time (in evolutionary terms)iii, even as our muscles grew puny and weak. But this is not accepted theory. Not yet. So strap yourselves in for a bumpy ride. It’s up to you to decide if I’m making sense.
A Smoking Gun
To understand why language is more of a trick than a trait, we’ll need to think a little bit about how language works. In simple terms, all languages have two key parts: words and grammar. Words seem straightforward enough – all you need is a dictionary, and you can start to have a basic conversation. But grammar takes more time to learn, as different languages can have very different strategies. Take the simple sentence Their god was a dog. In some languages, German and Japanese for example, you say Their god a dog was. Some languages even say A dog their god was. And then you have the more complex aspects: conjugation, declension, agreement and so on. That’s where most people give up. It’s also where most linguists begin.
They begin there because these are the aspects of language that are hardest to understand: how do we use these grammatical rules so effortlessly if we don’t know how they work? This is the question that first led Professor Chomsky to propose (and most professional linguists to agree) that there must be some genetically determined ‘Universal Grammar’ function in our heads, that allows us to do it as if by instinct. But there’s a problem, Boston. Two of them, in fact. The first is that even after decades of intensive research, no one can agree on what this elusive Universal Grammar might look like, or even if it actually existsiv. The second is this: language cannot have begun with grammar because you can’t have grammar without words.
If that seems too obvious to need saying, think again. Compared to the sounds that animals make, words are not as simple as they seem. Animal calls are mostly just what they are: unique holistic noises that can vary in quality according to the feeling they express – louder, softer, harsher, smoother – but cannot be broken down into parts. Words are different. All words are combinations of vowels and consonants, the units of sound that linguists call phonemes. Take the word dog. It’s made up of three phonemes: /d/, /o/ and /g/. Reverse them, and you get a completely different word: god. You can’t do that with the moos, hoots, screeches or squawks that animals make, and which our ancestors must once have made too. If we want to understand how language got started, we need to know how we came up with words.
A vital clue is found in a paper published in Science in 2011v, which showed that the number of phonemes used in languages around the world is not constant, but varies in what is known as a ‘cline’ – meaning that there is a relationship between the number used in any given language and its geographical location. Though there are many exceptions along the way, the general pattern that emerges shows that overall, the number reduces with distance from Africa. To take the two extremes, the largest number of phonemes is found in the Khoisan languages spoken by the bushmen of Southern Africa – known from genetic studies to have one of the oldest lineages on earth – and the smallest is among in the Pirahã tribe in the Amazonian jungle: one of the last places to be inhabited as we slowly spread out from Africa round the globe.
Why is that important? Well, it means that the number of sounds used in languages as they changed over time seems to have decreased as they evolved – and this is our smoking gun: a clear fingerprint of the origin of language. For if the number of phonemes has dropped away over time, the first languages must have used more of them – and if it’s possible to communicate just as well using a much smaller range of sounds (where English uses 44, Japanese has only 20; ǃXóõ has 144, Pirahã only 11), it means that many languages are still using more phonemes than they require. To see why, we’ll have to do some maths: but stay with me – it’s not too demanding.
The Trick of Speech
All words are made up of syllables – and at its simplest, a syllable is no more than a combination of a vowel and a consonant. English has around 20 vowels and 24 consonants (we don’t have enough letters in the alphabet to represent them all, which is why English spelling is so complicated). This means, at a rough approximation, that English can generate a minimum of 20 x 24 = 480 unique single-syllable words. That may not seem very many, but the magic of maths means that when it comes to two syllables, it’s possible to create as many as 480 x 480 = 230,400 unique words – way more than any one person could ever need, or hope to use.
Now let’s see how things stand at the two extremes. The Khoisan languages can have over 40 vowels and 100 consonants, which means that at least 4,000 unique words are available with just a single syllable, and an astonishing 16 million (4,000 x 4,000) with two – which is why most words in these languages are monosyllabic: they don’t need any more. Yet though the Pirahã language has only three vowels and eight consonants, and can only make 3 x 8 = 24 single-syllable words, once you get to three syllables there’s the potential for 24 x 24 x 24 = 13,824 words (three syllable words are common in most languages; think of ‘syll-ab-le’ in English). And by adding on just one more syllable, the number goes to over 330,000.
It’s this huge potential that’s so important. You may not know how many words you know. Most people don’t – and that includes most linguists. It’s actually a rather hard thing to measure – academic estimates vary greatly – but the consensus seems to be around 20-30,000 words. That’s not the same thing as the number of words that exist or the number of words you actively use on a daily basis – but it seems to be the number most people need to know in any language. This means that it’s easy to make enough words to enable the full spectrum of human communication with even quite a small number of phonemes.
So why would languages use more sounds than they need? And now we come to the crux of the matter. Most animals use sound to communicate. Yet as we have seen, they’re just simple noises – analog representations of feeling, like oral emoji. But there’s a limit to the number of unstructured, random sounds you can remember. Without a system of some kind, it’s hard to retain unrelated information. So if our earliest ancestors were communicating simple emotions in this way, slowly building up a stock of calls with different meanings, they would eventually have come up against a memory limit. And the easiest way to have more sounds without increasing the memory burden would be to combine existing sounds.
If you were already using the sound Aah! to mean ‘snake’, and Eeh! for ‘bird’, for example, you could use Aah-Eeh! for ‘elephant’ and Eeh-Aah! for ‘tiger’. Or Aah-Aah! for sex and Eeh-Eeh! for food. Suddenly you have four new words from just two sounds. Combining three sounds gives you 3 x 3 x 3 = 27 possibilities, four gives you 256 and so on. It’s exponential. Think of numbers. All languages have different words for the first ten numbers, but after that, we break them down into units of ten, allowing us to remember much larger numbers than we otherwise could. Take the number 1348. That’s a quantity that’s easy enough to grasp when expressed that way, but imagine we had a different, unrelated word for every number beyond ten (like ‘eleven’ and ‘twelve’, but continuing on endlessly.) You’d soon lose track of them. It’s the same thing with sounds and words: for the big secret of human language is that it’s digital.
A Human Singularity
What does that mean? Well, don’t ask a linguist – with a few honourable exceptions, they don’t recognise that description, any more than biologists think of DNA as digital. But it’s accurate enough for all that. Like numbers (which are digital by definition; all numbers are made up of simple combinations of just ten digits, 0-9), words are made by combining a fixed number of sounds specific to their particular language. Whether we count them as syllables or phonemes, all words are unique ‘digital’ combinations of sounds we already know – which is why it’s so easy to remember them. So if the transition from animal communication to human language came about through a switch from analog sounds to digital words, we’d expect it to have begun with a relatively large number of sounds, which slowly reduced over time as we realised we didn’t need so many in combination. And that’s what the evidence seems to show.
If that’s the explanation, though, surely we’d expect to see signs of that happening with other species, too? Well, it turns out that we do. As we start to study the calling strategies of wild animals in more detail, we begin to see that quite diverse species – among them Campbell’s Monkeys, prairie dogs and babbler birds – are already combining soundsvi, using existing calls that mean one thing on their ownvii to mean something quite different in combinationviii. That too is a fingerprint of language – one that’s clearly consistent with the start of the process. Because grammar would only emerge much later, driven by the need to organise words once there are enough of them – allowing the relationships between them to be expressed. If you’ve ever had to learn a foreign language, you’ll recognise the problem. Words come first.
But if that’s the case, why haven’t any other species learnt to speak? Well, let’s just consider the effect of language. Being able to talk, however simply, means we can share ideas – once someone has figured something out, words allow it to spread. We don’t all have to be smart. It may have taken forever to invent the wheel – but once we did, the idea was shared so fast that archaeologists can’t agree where it first appeared. Or take the telephone: in just a few generations since it was invented, there are more mobile phones on our planet than people – even if most of us have no idea how they work. That’s the fruit of language. It’s a tool for linking minds – for the ability to exchange ideas means we are effectively all part of a giant, species-wide brain.
Which means that language wasn’t just another evolutionary add-on, an incremental step like an extra toe or colour vision. Being able to talk would have brought about a rapid and transformative change. Ray Kurzweil, Google’s futurist-in-chief, talks of what he calls The Singularity – the moment when AI becomes smarter than we will be able to understandix. But with language, we humans have already been through our own Singularity: as far as other animals are concerned, from the moment we learnt to speak we left all other species standing. We can’t see that, because even the longest individual human life is a mere speck in the large scale of evolution – but in the big picture, language took us to warp drive. Other animals may be stumbling toward that moment, but by getting there first, we cornered the market in communication.
The Brain Gain
And there’s more. If language emerged the way I have outlined, by slowly digitising analog sounds to create words, it helps us explain another deep mystery of the human species – for there’s a further fingerprint of language to be found in the very shape of our bodies. Why do we have such big brains, and why did they grow as quickly as they did, evolutionarily speaking? The view accepted both by linguists and anthropologists is that language somehow emerged as a natural consequence of our big brains, which just happened to become big enough to enable us to speak. What’s never explained is how, or what might have happened to drive that sudden change.
For brains are expensive to run: those of most mammals, including the other great apes, are no more than a third of the size of ours, relatively speaking, and consume about 8% of the energy they get from food. But our brains gobble up over 20% of our available energy – a huge increase in demand. Not only that, the change happened very quickly (there’s no precedent for brains tripling in size over the course of two million years as ours did) – so there would have to have been a huge benefit to doing so. And there’s something else: the fossil record shows our ancestors’ large jaws and brawny muscles actually shrank as their brains grew.
That’s especially odd, as it was happening just as we’d have needed more food to service our growing brains. Why would evolution select for weaker bodies when we’d need all our strength to get more food? As things stand, we have neither a consistent nor a convincing explanation. But suppose it coincided with the start of our journey with language. Far from being the cause of language, our big brains could far more plausibly be the result of it. For if language is a tool that allows us to share ideas, a community that can do that – at however primitive a level – is much better off than one that can’t, because everyone can benefit from the best ideas of a few.
Which would mean that it would start to be more useful to be smart than strong – creating a strong evolutionary pressure for the survival of those with bigger, better brains. And that’s not all: language lets individuals plan and work together, so we could harness the power of collective action – useful in hunting, for example. Without language, physically strong individuals can dominate their pack; with it, a bunch of smart nerds can use their combined intelligence to outwit and defeat them. And with language putting a premium on intelligence, it’s no surprise that our brains should have continued to grow: indeed they kept going until a natural limit for childbirth was reached – humans are still more likely to die in childbirth than other species.
Then there’s the question of consciousness, the idea of the soul, and how language contributes to all that – and how the ability to discuss ideas raises questions that lead us down predictable avenues: the blind alleys of culture, religion and identity among them. These false gods have caused us much trouble through history, and still do today – hampering our ability to come together as a species, and overcome the problems we now face collectively at a global level. Because beyond these tribal concerns, the real fruit of language is science and the technology it enables. Whether on this planet or any other, any advanced civilization will have to have developed a method of storing and sharing information, and though the physical forms may vary, the broad process is likely to be similar. These are the issues I explore in more detail in my book, for the emergence of language was just the beginning – what really defines us as human is the use we have made of it.
This may not be the kind of mystery you expected to read about when you came to this site, and I’m grateful to Graham for allowing me to air my ideas. They may not help us determine how and when the earliest civilisations first appeared on our planet, but I believe they are important nonetheless. For one thing, you wouldn’t be reading or talking about anything at all if we didn’t have language, and understanding how we have it and how it works is an important part of what it means to be human. And for another, by showing how we may have been using language for much, much longer than we’ve previously thought – even for millions of years – it offers a way to understand how our earliest ancestors may have been able to spread out around the globe and begin to civilize, by themselves, much earlier than is currently supposed. It’s food for thought, anyway. And isn’t that why we’re all here?