One evening in late 2018 Daniel Green, a sixty-eight-year-old bar owner from London, sat in his office with his twin brother and business partner, George. The siblings had opened their bar more than fifteen years before; Daniel had left his office job at an insurance company and George, a self-diagnosed loose cannon recently released from prison, was grateful for the new opportunity. ‘It was our chance to start over,’ George told me. ‘We both had pasts that we’d rather forget, and our relationship had become distant, so I think opening the bar together was our way of putting things right. It was our happy ending.’
The pair were in the office for their weekly business meeting, a time to discuss everything from their stock, to operating costs, to designing new cocktails. As the more experienced of the two, Daniel usually did most of the talking. But this time was different. Daniel was quiet, confused and unresponsive. Then, apropos of nothing, he started babbling incomprehensible sentences – meaningless phrases about the ‘time of the ocean’ and ‘the sound of appointments’ – looking at George as if what he said was completely normal. Before George could work out what was wrong, Daniel collapsed. He was having a massive stroke.
At the hospital, the doctors explained to George that Daniel’s stroke had occurred in the left hemisphere of his temporal lobe, damaging the language centres of his brain. He now has what’s known as Wernicke’s aphasia: a condition in which the production of speech is unaffected, yet the meaning of words is almost completely lost. Asked how his day is, Daniel will now babble, ‘we stayed down the jango around the them there, up to spoodle with the family over the remrem towards the people, they’re taking all the moment for us today, with Jack.’ For a man who came top in his school spelling competition with the word ‘cymotrichous’ (having wavy hair), it’s a devastating loss.
In July 2021 I video-called Daniel and George at their flat in South London – Daniel has curly grey hair, green eyes and a polite, self-effacing demeanour; George resembles his brother, but is taller and looks less frail. They’d agreed to the interview to raise awareness of aphasia – an isolating disorder that many are unfamiliar with – and also to learn if I knew of any advances in research in to how the brain processes language. Naturally George led the discussion, acting now as a carer as well as a devoted brother. He said he often knows what Daniel is trying to say, and thinks of his role more as a guide than anything else. His brother’s mind was intact, but its voice lost at sea. And George was the navigator.
I asked Daniel how he was, and what he’d been up to. ‘You know that polkamoot dazzled with our friend the other day and we want to take them up the hill to see fans created by lightning,’ Daniel said. ‘It’s hard to raiserbad the flower pots since we went up there.’
I smiled and nodded and looked at George, unsure whether at least to try to respond. ‘That’s Daniel’s way of saying we went to a barbecue yesterday,’ said George.
‘Polkamoot’ is new. It might be something to do with the children there. One was wearing a Pokemon T-shirt. The ‘fans created by lightning’ is probably something to do with the barbecue itself. As for the ‘flower pots’, your guess is as good as mine.
Daniel chimed in: ‘He gets the river for the time we all need. Them, the tea over then.’ ‘Oh that’s an easy one,’ George said. ‘He’s just asking me to go make some tea. You can see how it’s manageable. He just needs someone with him a lot more now.’
While George stole out to make the tea, I asked Daniel questions about his life and career running the bar. I asked him how he felt about his condition and how it had changed his everyday life. His answers, though nonsensical to me, were fluent and thoughtful; he was clearly of sound mind in other aspects. At times, it looked as though he was trying to find the right words, even though most patients with Wernicke’s aphasia don’t know how their speech is coming across. According to George, Daniel usually thinks he’s making sense but then becomes frustrated when it’s clear that he isn’t.
When George returned, the conversation turned to deeper musings, and how Daniel is now involved in research projects to help scientists understand how the brain creates language. By using fMRI to map the language areas of his brain, scientists can examine which areas have remained intact, and whether or not they contribute to tasks such as verb generation, object naming and sentence comprehension. It’s still unclear which of these came first in evolutionary history but, by examining Daniel’s temporal lobe, scientists hope to discover the brain circuitry responsible. If they succeed, they can then look for signs of that circuitry in other primates, such as bonobos and chimpanzees, and thus help expose the evolutionary origins of language. In the epic story of human brain evolution, language was a pivotal moment.
So when did humans start talking to each other? Estimates range from 50,000 to more than 1 million years ago, for words leave no trace in the fossil record. Why did we start talking? Some think it began with our ancestors imitating various natural sounds, others that it arose from spontaneous cries of pain or surprise, or that it evolved from grunts, groans and snorts prompted by arduous physical labour. The theories are as imaginative as they are disputed.
Howsoever it arose, language became the cornerstone of human cognition. Over 7,000 languages are spoken in the world today. While other animals communicate using sounds, smells and body language, humans have gone further, creating tens of thousands of arbitrary symbols to build communities, cultures and societies. As we shall see, the stories we tell ourselves are merely adaptations for social interaction. Having evolved for emotion, sociability, memory and intelligence, the human brain could next evolve a system to connect us all.
Language probably began over 1 million years ago with Homo erectus, a supremely clever early human that made tools, lived in groups and built boats to colonise islands such as Flores in Indonesia, and Crete. ‘Erectus needed language when they were sailing to the Island of Flores,’ says the American linguist Daniel Everett.1 ‘They needed to be able to paddle. And if they paddled they needed to be able to say “paddle there” or “don’t paddle”. You need communication with symbols, not just grunts.’ Everett accepts that such language would have been no way near as sophisticated as that used by modern humans; nonetheless, he believes that Homo erectus invented symbols capable of initiating the baby steps required for grammar and syntax. In other words, there was no single moment when language crystallised. It emerged gradually when small bands of people decided to adopt culturally invented symbols.
Scientists call this the sign progression theory of language. Symbols (invented markings that represent something, such as an animal) arose from icons (pictures that physically resemble the thing, such as a cave painting of an animal) which themselves arose from indexes (something connected to the thing, such as a footprint of an animal). Over time, ancient symbols would have been combined with others to produce increasingly complex grammar and sentence structures. The language we speak today is therefore the legacy of all the words ever uttered by our species. From a long line of language-acquiring ancestors and a rich pattern of cultural inheritance, language has flowed through time like water through a canyon.
The evolutionary value of language is hard to overstate. ‘As soon as a system of symbolic communication came into being,’ the biologist Jacques Monod once wrote, ‘the individuals, or rather the groups best able to use it, acquired an advantage over others incomparably greater than any that a similar superiority of intelligence would have conferred on a species without language.’2 Or as the psychologist Cecilia Heyes put it, ‘Now capable of abstract thought and subtle communication, we became radically different from all other animals – more like gods than beasts.’3
To be sure, animals perform dazzling feats of communication. Octopuses, lacking vocal cords and unable to emit any sound, send signals with their skin colour and texture. Squid display bands of colour to produce an alphabet of patterns, said to have at least fourteen ‘letters’. Moths use pheromones, birds click their wings, dolphins whistle, honeybees dance, elephants caress with their trunks. Some animals even communicate with other species – in exchange for a share of the spoils, the honeyguide bird leads honey badgers to bees’ nests by making a distinctive trilling sound.
We may never know definitively when the first human language appeared, but it’s likely that all languages share a common origin. That origin could be the result of a bottleneck effect in which a single language then became ancestral to all the world’s languages. ‘This language would have been spoken by a small east African population who seemingly invented fully modern language and then spread around the world, replacing everyone else,’ says the linguist Merritt Ruhlen, who has searched for clues to the first human language by creating a phylogenetic tree showing the historical relationships between all the language families in the world.4 One family is the Indo-Aryan family (which includes Bengali, Hindi and Punjabi), which is a branch of the Indo-Iranian family (which includes Persian, Pashto and Kurdish), which is itself a branch of the even larger Indo-European family, said to be between 6,000 and 8,000 years old, and which includes languages as diverse as Danish, Greek, Hindi and Welsh.
Of course, language is more than a collection of symbols. It’s also an inventory of sounds (phonetics) coupled with an inventory of sound structures (phonology). Early humans exploited this fact in a remarkable way. Researchers have found that ancient cave art is often located in acoustic hot spots: spaces where sound generates echoes. These cave chambers are harder to reach than chambers that have no echo, so it’s thought they were deliberately chosen as places for secretive rituals and dramatic storytelling. Strikingly, when acoustic archaeologists study the sonic properties of these caves, they find that paintings of cloven animals such as bulls, bison and deer, appear in chambers that generate echoes and reverberations that actually sound like hoof beats. Quiet chambers, on the other hand, are painted with dots and handprints. The finding is now so consistent that some acoustic archaeologists can locate the animal paintings in pitch black, the echo of their voices being their only guide.
When our ancestors first used language they had no idea how it would shape their brains. But just as exercise changes our bodies and diet changes our gut, language does indeed change the physical structure of our brains. Neurons form durable connections in response to sights and sounds – including words – and this neuronal wiring forms the basis of the brain’s grey matter, where all information about the world is processed. As children, we pick up words and phrases unconsciously, because the sounds of our native language strengthen certain neuronal connections. For monolingual people, this means that some neuronal connections are weakened due to certain sounds being absent in their language. Which is all the more reason to teach children multiple languages: the additional input actually enhances the physical structure of their brains. Bilinguals use their anterior cingulate cortex (ACC) – a brain region important for empathy, emotion and decision-making – in a more efficient fashion than monolinguals. They possess a larger auditory cortex, making it easier for them to detect changes in pitch, locate sounds in space and understand new languages. They even encode speech sounds in their brain in a more robust manner, giving rise to a richer phonetic environment.
Changes in one brain area nearly always trigger changes in another. Remarkably, the increased brain activity of bilingualism can delay the onset of dementia by around four years, as well as help the brain cope with dementia after developing the disease. The additional neuronal pathways are also thought to enhance cognitive skills such as memory, attention, multitasking and creativity. Geographical constraints meant that our ancestors were unlikely to be learning several languages at once, but the important thing to remember is that a second language merely amplifies the cognitive benefits that occur in monolingual people. Language does for the brain what the latest software update does for your computer – it makes it faster, cleverer, more agile and better at completing tasks. For early humans, language was the software update that made better thinking go viral.
So why did language evolve? Was it, as acoustic caves suggest, for storytelling? Perhaps. The desire to tell and hear stories knits a community together. They foster a sense of shared values, of togetherness and cultural cohesion. Fictional narratives are especially important to society: as humans become increasingly dependent on one another, as we evolve to live in ever-bigger societies, we need such narratives to learn how to cooperate. When anthropologists visited eighteen groups of hunter-gatherers in the Philippines in 2017, they found that 80 per cent of their tales concerned morality and social justice.5 The most skilled raconteurs also proved to be the most desirable social partners, even more so than skilled foragers.
Or perhaps language evolved from shared intentions – that is, the power of human minds to collaborate to achieve a collective goal. In this context, language is merely an expression of the mind’s ability to unite with others, a veil concealing more basic objectives that evolved in tandem with cultural progress. This view is gaining significant ground in the academic community. Since languages change more rapidly than genes, this brain change was shaped mainly by culture, not biology. Consider the Romance languages (French, Italian, Spanish, Portuguese, Romanian): they diverged from Latin in less than 2,000 years. No evolutionary change works that fast.
These observations teach us arguably the most important lesson about why language evolved: it’s a social innovation. A collective act to facilitate human cooperation. Just imagine a world in which we had not evolved the capacity for language. Instructing the young and passing on knowledge of the natural world would have been very basic, not to mention building villages, planting crops and developing new technologies. As humans evolved, language became almost inevitable: with increasing contact between bands of early humans came an increased need to establish cross-group communication networks. Material culture was especially important. Trade creates opportunity and ‘with trade comes negotiation and further selection for effective communication,’ writes the anthropologist John Odling-Smee.6
Cultural evolution propelled language forward with one of the greatest inventions of all time: the printing press. In 1440, when Johannes Gutenberg first used the imposing wooden structure for books, newsheets and other printed knowledge, so too were new words and new ideas distributed throughout the world, radically expanding each language’s vocabulary and changing how thoughts were shared. The spread of science and technology in particular – much of which was based on Greek texts that were preserved by Arabic and Persian translators – led to the intellectual movement widely known as the Enlightenment. Before the fifteenth century, the world’s literacy rate was below 10 per cent. Today, it’s 83 per cent (91 per cent for young adults aged fifteen to twenty-four).
Surprisingly, given how quickly we’ve adapted to the task, reading and writing are not things brains are born to do. Certainly no hard-wired brain structure evolved for such tasks, so how does the brain do it? In 2012, neuroscientists at Stanford University showed that reading ability in young children is linked to the growth of white matter tracts in the brain.7 White matter – named after the white fatty substance (myelin) that surrounds nerve fibres – allows the rapid transmission of information throughout the brain. Importantly, studies also found that practising reading can boost white matter integrity, improving children’s reading ability and their capacity to learn more generally.
Reading changes the brain by changing its connectivity. When we read, connectivity in the left temporal cortex and central sulcus – areas important for language and the sensation of movement, respectively – is increased. Which is why reading a novel can transport you into another world. If a character you are reading about is running, for example, neurons in these areas boost their connectivity and start mimicking the physical sensation of running. Neuroscientists call this phenomenon embodied cognition, and it’s a way for our brains to incorporate bodily experiences into how we think.
In evolutionary terms, this is a major advance because it circumvents the fact that different areas of the brain have adapted for different tasks. By ramping up the brain’s connectivity, reading forces the brain regions important for attention (the temporal lobe), learning and memory (the hippocampus), and object recognition (the fusiform gyrus) to work together to produce something novel. The fusiform gyrus also helps us distinguish between words and letters, for instance. Which is pretty impressive, given that Homo sapiens have only been reading for 5,000 of our 400,000 years of existence. It’s an exceedingly new activity, and it will probably take us a long time fully to discover how it is changing our minds.
Human minds have certainly been eager to exploit language. Since modern humans first evolved, as many as 31,000 languages have come into existence. Much of that linguistic diversity has sadly perished. Of today’s dwindling 7,000 languages the ten most spoken – English (1.13 billion speakers), Chinese (1.11 billion speakers), Hindustani (544 million speakers), Spanish (527 million speakers), Arabic (422 million speakers), Malay (281 million speakers), Russian (267 million speakers), Bengali (261 million speakers), Portuguese (229 million speakers) and French (229 million speakers) – lay claim to half the world’s population. Many of the rest are concentrated in only a handful of locations. One in eight of all the world’s languages are spoken by tribes in Papua New Guinea, thought to be due to a unique landscape that allows small groups to flourish independently. In the village of Ayapa, Mexico, the pre-Colombian language Ayapaneco almost died out because the last two surviving speakers refused to talk to each other.
The main factor affecting linguistic diversity, though, is not culture but climate. The anthropologist Robert Lee Munroe observed that people in warmer countries rely more on vowels because they’re easier to hear outside than consonants. Speakers in colder climates, on the other hand, use more consonants because most communication occurs indoors. It’s called acoustic adaptation, and is seen across the animal kingdom. Birds such as sparrows, blackbirds and the great tit sing at higher pitches in cities in order to penetrate the noise produced by traffic and human activity. Woodland birds sing at lower frequencies than birds living in the open in order to cut through branches, which deflect high-frequency sounds. In 2015, the linguist Ian Maddieson and his colleagues took 633 languages and explored the ecology and climate of where each language developed.8 (Languages such as English, Mandarin Chinese and Spanish were excluded from the study since they are no longer restricted by geographical location.) The pattern that emerged gave resounding support to Munroe’s observations: languages in hotter, humid climates were vowel-rich and mellifluous; languages in colder, drier climates were consonant-rich and guttural.
Social experience is crucial for language development, hauntingly demonstrated by the existence of feral children: children who grow up with no human contact and thus no experience of human language. Such children – usually confined by their own parents – can only speak a few words and find it tremendously hard to learn a language after being isolated for so many years.
There have been over a hundred reported cases of such children. There was Oxana Malaya, a Ukrainian girl whose parents left her in a dog kennel behind her house. When the authorities found her five years later, she couldn’t speak, ran on all fours and barked. There was Victor of Averyon, a French boy who lived alone in the woods in the late 1700s. At twelve years old, he was caught by local hunters and found to speak no recognisable language. And there was Prava, a seven-year-old Russian boy who communicates only by chirping after his mother raised him as a pet bird in her aviary. Dubbed ‘bird-boy’, Prava doesn’t understand any human language and flaps his arms when he knows he is not understood. But perhaps the most famous case of all is that of a Californian girl nicknamed Genie.
Genie was born in 1957 to dust bowl migrants from Oklahoma. Her father, a machinist who worked on an aircraft assembly line during the Second World War, didn’t want children and hated Genie from the start. A deranged and tyrannical man, he labelled her retarded when she was twenty months old and decided to keep her as isolated as possible. For the next thirteen years he kept her either strapped in a straitjacket that was tied to a toilet, or bound in a wire-mesh covered crib with her arms and legs restrained. If she cried or made a sound, he would growl at her like a dog and then beat her with a baseball bat.
Genie’s nightmare existence may never have been apprehended were it not for a chance occurrence one day in October 1970. Her mother – a loathsome figure who was mostly indifferent to her daughter’s plight – had cataracts and decided to take Genie with her to apply for disability benefits for the blind. Because of her poor eyesight, though, she accidently stumbled into the wrong office: the office for social services. Upon seeing a hunched, withered and shaking girl beside her – a girl they guessed was six but were shocked to learn was in fact thirteen – the social workers immediately called the police. When Genie’s father was charged with child abuse, he shot himself. ‘The world will never understand,’ he said in his suicide note.
Genie was moved to the Children’s Hospital Los Angeles. She weighed just fifty-nine pounds, had deformed hips and an undersized rib cage. She was unable to straighten her arms and legs, and moved with what the nurses called a ‘bunny walk’, holding her hands in front of her like claws. According to one doctor, she was ‘the most profoundly damaged child I’ve ever seen.’ Her rehabilitation team included paediatricians, psychologists, linguists and scientists from around the world. Among them was Susan Curtiss, a linguist determined to teach Genie English. Although Genie could speak a few words, she couldn’t produce grammatical sentences. A transcript of one of her first attempts reads: ‘Father hit arm. Big wood. Genie cry… Father hit big stick. Father angry. Father hit Genie big stick. Father take piece wood hit. Cry. Me cry.’
After a few months, Curtiss and her colleagues saw improvements in Genie’s vocabulary. She could learn new words and would occasionally put two or three together. ‘She was smart,’ said Curtiss. ‘She could hold a set of pictures so they told a story. She could create all sorts of complex structures from sticks. She had other signs of intelligence. The lights were on.’9 To psychologists, Genie’s case was the perfect chance to settle a debate. There are two models for language development: the nativist and the empirical. The nativist posits that language is biological and depends on circuitry installed in the brain at birth, a so-called ‘language acquisition device’ that allows children to learn language quickly and easily. The empirical model describes language as a product of the environment, with a ‘critical period’ of learning that begins at birth and lasts until the onset of puberty.
Genie’s inability to learn language proved the latter to be true. In the years following her release, separate research showed that a child’s brain develops much of the machinery for language before the age of ten: regions of the auditory cortex connect to the frontal lobes to sprout pathways for speech production, speech recog-nition, phonological memory and lip reading. The ability to learn language fluently, to apprehend its grammatical complexity seamlessly, persists until the age of eighteen and then tails off in adulthood. This doesn’t mean that learning a new language in adulthood is futile, but it’s hard, and Genie’s extreme isolation and abuse made it impossibly so.
This is because the brain becomes fine-tuned for language upon contact with others, an idea first proposed by the Soviet psychologist Lev Vygotsky in 1930. As far as public intellectuals go, Vygotsky was as influential as Sigmund Freud and Ivan Pavlov. The Communist Party censored much of his work and so he only became known to Western scholars in the 1970s. In publications spanning a ten-year period, Vygotsky made the following novel argument: language – indeed much of human development – evolves from the interaction between individuals and society. Society can change people as much as people can change society. Children learn language using what he called a ‘zone of proximal development’, a gap separating knowledge from ignorance that can easily be filled by a ‘more knowledgeable other’: a teacher, a parent, a friend. ‘Through others,’ he wrote, ‘we become ourselves.’
Today we live in a world that takes Vygotsky’s reasoning for granted. Qualified teachers, good school attendance and sound parental care are enshrined in law. Yet Vygotsky’s claim that other minds literally sculpt our own when learning language is one of the most staggering ideas in neuroscience. It suggests that language evolved to link our brains together physically: to create, in effect, a hive mind. Like the densely packed society of neurons in our head, each communicating using a chemical web of neurotransmitters, language binds human societies together using a collection of sounds, symbols and meanings. Language is the interconnected web of experience writ large.
In fact, society changes language in a process very similar to evolution by natural selection. Words, like genes, survive and produce offspring based on how adapted they are to the environment. The word aspirin for instance was invented in 1899 by a German chemist and derives from the opening letters of Acetylirte Spirsäure (acetylated spiraeic acid); it survives because it’s easy to say. Shakespeare, a prolific neologist, invented words such as critic, swagger, lonely and hint. Other new species include fragrance and pandemonium (John Milton), universe and approach (Geoffrey Chaucer), valediction and self-preservation (John Donne), anticipate and atonement (Sir Thomas More). English speakers are adding a thousand new words a year to their lexicon. Some of my favourites include blog, doublethink, quidditch, podcast and McJob. The word selfie has even spawned its own offspring: there’s helfie (a selfie of one’s hair), welfie (a selfie during a workout) and drelfie (a selfie when drunk).
Words, like species, can also go extinct. In 2009 the biologist Mark Pagel and his colleagues at Reading University used supercomputers to apply the theory of natural selection to the family of Indo-European languages. The results suggested that a number of commonplace words such as ‘dirty’, ‘bad’, ‘turn’, ‘guts’, ‘throw’ and ‘stick’ may become extinct or be permanently replaced by new words over the next 700 to 1,000 years.10 Pagel’s analysis showed that the most stable words were those essential to daily conversation, such as ‘I’, ‘we’, and the numbers one, two and three. Adjectives and adverbs were found to be particularly susceptible to dying out. In fact, half of the words we use today, Pagel points out, would be unrecognisable to our ancestors 2,500 years ago. As the noted philologist Max Müller observed in 1870, ‘A struggle for life is constantly going on among the words and grammatical forms in each language.’
The Internet is almost certainly accelerating this process. From fashionable punctuation to emojis to repurposing other symbols (the # was recently named UK children’s word of the year), the way we talk online is changing fast enough to make even the most progressive baulk. Trying to stop this is futile. In a recent study, Joshua Plotkin, a mathematician and biologist at the University of Pennsylvania, charted how grammar changes over long stretches of time.11 By examining over 100,000 texts published over the past 200 years, Plotkin and his team found that languages evolve by random chance as well as natural selection. An example of natural selection would be when verbs are regularised: the word ‘smelt’ changing to ‘smelled’. They’re easier to learn, thus more adapted to their environment. Verbs changing by random chance are those that became irregular: ‘dived’ changing to ‘dove’, ‘lighted’ to ‘lit’, ‘waked’ to ‘woke’. There’s no reason they should have changed; they just have. In the natural world this kind of change is called genetic drift, the second major force of evolutionary change, when genes become more or less frequent by pure fluke. And so if, like me, you want to move with the times but also keep some of those cherished rules of grammar and syntax, prepare to be disappointed.
As children, we practise language through pretend play. From about eighteen months onwards, all children, in all cultures, engage in pretend play: having imaginary conversations with made-up characters, talking into a banana as if it were a telephone. Childhood psychologists encourage pretend play because it appears to have a positive effect on a child’s ability to learn. The pretence is often full of rich collaborative dialogue and set in an environment, usually the home, that reduces the pressure of learning. Additionally, neuroimaging studies show that the same brain circuitry underlying childhood pretend play is essential for adult creativity. Neurons in the prefrontal cortex, dorsal and ventral striatum motivate children to engage in pretend play; and then, as adults, the same population of neurons become vital for creative thinking. The link between language, pretend play and creativity makes perfect sense: language is built on social rules, and when children pretend to be something else, they are experimenting with the rules, thinking creatively to dream up novel visions for the future.
None of this is to say that biology isn’t important for the development of language. In the relentless nature versus nurture debate, it is accepted that language evolved as a composite of both. Several brain regions are responsible for language. The most studied are Broca’s area (left hemisphere, frontal lobe) which is important for speech production, and Wernicke’s area (left hemisphere, temporal lobe) which is important for speech comprehension. Less understood are the angular gyrus and insular cortex, regions thought to unite the essential supplements of language – hearing, vision, attention, emotion and motor control – into a cohesive whole. Strikingly, none of these regions is dedicated exclusively to language. Some can even transfer their roles to other regions; Broca’s area can be destroyed without affecting language, for instance. The borders within the human brain are dazzlingly blurred.
Some believe the secret of human language lies in a region within Broca’s area: a bundle of nerve fibres called BA44, which is underdeveloped in chimpanzees, weak in new-borns, but fully formed and strong in adults with language ability. ‘This fibre tract,’ says Angela Friederici of the Max Planck Institute of Cognitive Neuroscience in Germany, ‘could be seen as the missing link which has to evolve in order to make the full language capacity possible.’12 We don’t yet know if Homo erectus or another ancestor had BA44. Many researchers, therefore, hold to the notion that Friederici’s thesis will have to remain a striking observation.
At the genetic level, scientists have long been intrigued by the FOXP2 gene in language evolution. FOXP2 made international headlines in 2002 when researchers found evidence that a special version of the gene spread to all Homo sapiens about 200,000 years ago. Songbirds use a version of FOXP2 for vocal learning, as do chimpanzees, whales, dolphins and bats. FOXP2 controls the muscles used in speech, and mutations in the gene can lead to severe speech and language problems. One example is a condition known as apraxia, which affects the sequence of sounds, syllables and words by damaging the brain regions controlling the movements of the mouth, namely the parietal cortex and the corpus callosum. The parietal cortex, remember, is one of the four major lobes of the cortex, sitting at the back of the skull, just behind the frontal lobe. The corpus callosum (or ‘tough body’ in Latin) is the thick bundle of nerves separating the entire left from right hemispheres.
It’s thought that human predecessors used a more ancient FOXP2, meaning they had less speech control, thereby giving Homo sapiens the evolutionary upper hand. This seductive narrative ran aground in 2009, however, when scientists discovered that Neanderthals also had the human form of FOXP2. That pushed our possible language gene back by at least 700,000 years. Did Neanderthals speak like Homo sapiens? It’s certainly possible as Neanderthals did have an anatomically human hyoid or lingual bone – a horseshoe-shaped bone in the neck crucial for speaking. So they were probably capable of speech. Whether they actually spoke or not is a mystery.
Some scholars think that language evolved not from vocal calls, but manual gestures. Dubbed the gestural theory, the idea is that primitive sign language slowly became integrated into the circuitry of the brain, which then freed the hands for other functions. Eventually, the brain organised this sign language into a hierarchy of phonemes that were grouped together to form words, which formed the basis for sentences. Is there evidence for it? Absolutely. Primates use gestures to communicate with each other: in 2019 video footage of chimps living in Uganda’s Budongo Forest recorded fifty-eight types of playful gestures. The most frequently used gestures were brief, staccato-like bursts, while longer sequences comprised short, syllable-like gestures. This may be why children use gestures to communicate before they have the ability to speak: our brains are hardwired to gesture by evolution.
In Nicaragua during the 1980s, a community of deaf children offered perhaps the strongest evidence for the gestural theory. The children, born to parents who could hear and having had little contact with each other until the government opened a school for the deaf in 1977, each had their own simple sign system at home and thus needed to develop a common sign language when brought together at the school. Which is precisely what they did. Over the course of thirty years, successive generations of children have created an entirely new sign language out of hand gestures, known today as Nicaraguan Sign Language (NSL). It was learnt unbidden, with no influence from any existing language, yet has rules found only in other languages. What surprised scientists the most, though, was how closely it followed evolutionary logic: the first signals were crude, producing only a few gesture traits that survived into the second generation, who in turn produced more sophisticated gesture traits that survived into the third generation, and so on.
The Nicaraguans aren’t alone. In the Negev desert of southern Israel, some 150 deaf members of the al-Sayyid Bedouin tribe are in their seventh generation of evolving their own sign language, the al-Sayyid Bedouin Sign Language (ABSL). Stretching back over seventy years, ABSL has a longer history than NSL and evolved in a more natural setting than a school. But like NSL, ABSL is a totally unique language. It shares no characteristics with Bedouin Arabic, modern Arabic, Hebrew or Israeli sign language. And it possesses a curious complexity in that it combines simpler words to achieve a different meaning: ‘sweat’ and ‘sun’ mean summer, for instance; ‘pray’ and ‘house’ mean mosque. Tribe members use ABSL to discuss everything from their dreams and desires to national insurance and local construction projects. Linguists are now keen to see how (and if) the language continues to evolve; unfortunately, the tribe has become isolated because it is stigmatised by other Bedouin tribespeople.
Two crucial motor systems evolved to enable speech: the mesencephalic periaqueductal gray (PAG) and the caudal medullary nucleus retroambiguus (NRA), both of which control neurons involved in vocalisation and our ability to modulate vocalisation into words and sentences.
The brain’s PAG generates human speech in a mysterious way. Occupying a column of brainstem roughly half an inch long, the PAG can elicit vocalisations in animals when it is poked and prodded in a laboratory setting. Experiments on (I’m sad to say) cats found that four different sounds emerge when the PAG is electrically stimulated – mews, howls, cries and hisses – depending on where in the PAG the stimulation occurs. In humans, lesions in the PAG result in absolute mutism; we completely lose our ability to produce sound.
While the PAG’s underlying neurophysiology remains unclear, many believe that it somehow weaves together breathing and laryngeal muscle patterns to generate a code essential for speech and song. Remarkably, that code was recently decoded into written text by researchers at the University of California, San Francisco.13 Using a group of epilepsy patients about to undergo surgery for their condition, the researchers placed electrodes directly onto their brains and asked them to read out a list of twenty-four responses to nine sets of questions. This data was then used to build a computer model that matched patterns of brain activity to the patients’ spoken words. Though the accuracy wasn’t perfect (76 per cent) and the technology still fairly basic, the researchers now want to attempt to decode ‘imagined speech’, the inner voice that only we can hear.
Much less is known about the brain’s NRA, only that it receives signals from the PAG and then sends its own signals to the diaphragm and other muscles controlling breathing. Unlike the PAG, the NRA does not generate speech patterns. Each of its neurons produces a specific motor action or tone, combinations of which can assist vocalisation, vomiting, coughing, sneezing, even mating posture and child delivery. The brain chooses the action based simply on what is needed in the circumstances. In this way, the NRA has been likened to a piano and the PAG the piano player. Together they exhibit a little-understood brain change that was critical for human speech.
In fact, it’s now thought that monkeys and other apes cannot speak because they didn’t evolve the PAG and NRA. Given how closely related we are, one would think that our primate cousins would at least be able to outspeak a parrot. For a long time it was thought that monkeys’ and apes’ vocal anatomy meant that they could not: if only they had more flexible vocal cords and larynx muscles, they would be able to mimic some human speech. Darwin wasn’t so sure. He thought apes couldn’t talk because of the architecture of their brains; and he was right. In 2016 a team of researchers led by Tecumseh Fitch at the University of Vienna and Asif Ghazanfar at Princeton University in New Jersey filmed long-tailed macaques with an X-ray scanner while they made various calls and sounds.14 Then, using X-ray stills from the video, they made a computer model to show all the sounds the monkeys could make when air flows through their vocal anatomy.
They found that the monkeys do, in theory, have the anatomy to produce vowel sounds, including those in the English words ‘bit’, ‘bet’, ‘bat’, ‘but’ and ‘bought’, and the vowel-heavy phrase, ‘Will you marry me.’ The simulated monkey voice sounds a little artificial, but it’s undeniably intelligible. ‘A monkey’s vocal tract would be perfectly adequate to produce hundreds, thousands of words,’ Fitch said in an interview.15 So why don’t they? Because they lack the proper wiring in their brains; specifically, the wiring needed for the PAG and NRA. ‘In short,’ Fitch and Ghazanfar conclude, ‘primates have a speech-ready vocal tract but lack a speech-ready brain.’
Not everyone agrees that monkeys have speech-ready vocal tracts, however. Philip Lieberman, a cognitive scientist at Brown University, Rhode Island, has argued that even if monkeys possessed the cognitive hardware to speak, their vocal range is simply too low to produce human-like speech. ‘If monkeys had brains capable of learning and executing the motor commands involved in human speech, their “monkey speech” would not be as robust a means of vocal communication as that of fully modern human beings,’ he writes.16 Part of the problem is that monkeys cannot produce the ‘quantal’ vowels (i, u and a, and vowels in the words ‘see’, ‘do’ and ‘ma’), which are present in nearly all human languages and are thought to be critical to the robustness of human speech. Fitch and Ghazanfar retort that monkey speech would obviously not sound the same as human speech, but that nevertheless, ‘a monkey vocal tract would be able to produce clearly intelligible speech.’
For now, there are no firm answers in this debate. Perhaps monkeys are very far down the evolutionary path to language, but just not far enough. Perhaps ‘monkey speech’ would sound very strange, but still be intelligible. Perhaps the Simian Tongue will one day be decipherable and found to be the most extraordinary language of all.
Bringing together the threads of how and why we evolved to speak, read and write teaches us something profound about language: it shapes our thoughts and changes the way we perceive reality. Consider how we experience colour. The Dani tribe of Papua New Guinea recognise only two colours: ‘mili’ for cool dark colours such as blue, green and black, and ‘mola’ for warm light colours such as red, yellow and white. Russian speakers lack a word for blue: instead shades of light blue are ‘goluboy’ and dark blue, ‘siniy’. Pink is just a light shade of red, yet it’s treated as its own colour in English. Australian aboriginals have no words for left and right: their language uses only north, south, east and west. Some languages, such as German and Spanish, assign every noun with a gender; the word ‘bridge’ is feminine in German, masculine in Spanish. When asked to describe a bridge, German speakers are more likely to use conventionally feminine words such as ‘beautiful’ and ‘elegant’, whereas Spanish speakers are more likely to use conventionally masculine words such as ‘strong’ and ‘sturdy’. The examples are endless. One is tempted to speculate that with 7,000 languages come 7,000 cognitive realities.
For the time being, the language that dominates the human mind is unequivocally American English. Nearly 400 million people speak it as a first language and more than 1.6 billion as a second. It serves as the official language of fifty-nine countries and is the universal language of business, diplomacy, the Internet and science. The literary critic Jonathan Arac has remarked, ‘English in culture, like the dollar in economics, serves as the medium through which knowledge may be translated from the local to the global.’17
Historically, English spread due to the expansion of the British Empire between the seventeenth and nineteenth centuries, and the emergence of the United States of America as the world’s superpower in the twentieth century. A dark history of conquest and colonisation accompanied this expansion, leading many to view English as a form of linguistic imperialism. To this day, English continues to erode the native languages of postcolonial settings such as India, Pakistan, Uganda and Zimbabwe.
Like all empires, the English language may eventually start to decline. If it does, Chinese will probably take its place, a wondrous language which appeared over 3,000 years ago and is spoken by 1.1 billion people in the world today. Business experts have pointed out that Chinese is already becoming the language of choice in global markets. And as China’s economy continues to outpace America’s, Chinese could soon supplant English as the most widely used Internet language.
The Chinese language has a fascinating history. Belonging to the Sino-Tibetan family of languages, it is the oldest written language in the world, with a dictionary containing some 40,000 characters. Symbols that bear a striking resemblance to these characters have been found inscribed in 8,600-year-old tortoise shells at Jiahu in the Henan Province of western China; they appear to anticipate the Chinese characters for ‘eye’, ‘window’ and the numerals 1, 2, 8, 10 and 20. Latin didn’t appear for another thousand years, well into the seventh century BC. As the Chinese language evolved, it transformed into a collection of languages that spread far and wide – to Malaysia, Singapore and Indonesia. The main Chinese language is Mandarin; others such as Yue Chinese (Cantonese), Min Chinese and Hakka Chinese are now among hundreds of mutually unintelligible varieties.
Would the fall of English and the rise of Chinese or another language shape our thoughts and change the way we perceive reality? Almost certainly, writes linguist Nicholas Ostler, who chairs the Foundation for Endangered Languages (the world loses a language every two weeks). Ostler believes that by 2050 English will cease to be the world’s lingua franca, changing much more than how we speak to one another:
If other centres emerge, the result may be more mixed, asserting local Islamic, Buddhist or Hindu traditions. Evolving translation technologies may make languages largely interchangeable, pushing national cultures into the background. Whatever, there will be no special deference to the current English-speaking tradition.18
Whatever happens in the future, we first have to remember that language evolved for one reason – to unite us. By creating abstract mental terms for what a hunter-gatherer ‘sees’, ‘wants’, and ‘knows’, language allows them to build a relationship with others and form a social group. For most of human history this was done by small pockets of people that were totally isolated from one another. Neighbouring tribes were thus seen as a threat to a group’s social identity, and a competitor for scarce resources. Cue social conflict and a blood-drenched history of human confrontation.
But like groups, languages are living creatures that adapt to the times. As human populations grew, mixed and coalesced into ever-expanding cities, languages branched into different dialects and families. Some of these would have been understood by multiple sets of speakers – they had what linguistics call mutual intelligibility. But over time, as groups spread to the point where languages collided, and trade and religion required a way for humans to communicate with each other, mutual intelligibility gave way to translation.
Translation dates back to earlier than 1000 BC; in a written document from the Chinese Zhou dynasty, the imperial scholar Jia Gongyan declared, ‘translation is to replace one written language with another without changing the meaning for mutual understanding.’ Translation helped make the world a global village, and abstract terms for what a human ‘sees’, ‘wants’ and ‘knows’ can now be decoded in order to build relationships between social groups. This act of transference ushered in international cooperation and the rapid advancement of ideas and social progress.
Language matters. Human minds are organs that use words to connect, sharing our experiences and transmitting them across generations in the form of ancient texts, books and, increasingly, social media. We may never fully understand how language arose – how ape-like communication gradually transformed into the poetic prose of Vladimir Nabokov or the graceful Arabic of One Thousand and One Nights. But ‘how’, in this case, is less important than ‘what’. Because more than anything, language represents the culture of a particular group by codifying their values, beliefs and customs. It’s a social phenomenon; indeed its social function is its evolutionary raison d’être. But as powerful as language is, it would not have been possible without the brain change that knitted all our thoughts together – a phenomenon that some believe evolved only a few thousand years ago: the illusion that is consciousness.