Just where did language come from and what does it take for a language to develop? If a group of children grew up isolated from the rest of the world, without any access to human language, would they create one of their own and, if so, what would it look like and how long would it take? These questions have fascinated humans for centuries and speculation on the matter dates back at least as early as ancient Greece.

Some early thinkers believed that without outside influence, a child might default to an ancient or ancestral language. Depending on the location and time period, Hebrew, Greek, Egyptian, and Sanskrit were all suggested as possibilities. This might sound farfetched to us today, but ancient scholars gave the idea serious consideration. Several accounts (of varying credibility) describe attempts to isolate children from linguistic input and observe what language they would eventually speak.

The earliest such account comes from Herodotus, who described a supposed experiment by Psamtik I (who ruled Egypt from 664-610 BCE). Psamtik is said to have placed two newborns in the care of a shepherd with instructions to raise the children in isolation, with goats to provide milk as needed and no exposure to human speech. The goal was to determine whether the Egyptians or the Phrygians were the “eldest of all men” by observing whether these isolated children grew to speak the Egyptian or Phrygian language. According to legend, when the children were brought before Psamtik, they held out their hands and cried “becos”, the Phrygian word for bread. This was taken as evidence that the Phyrgians, rather than the Egyptians, were the oldest people. It is unlikely that these events unfolded exactly as described by Herodotus, but it does show that our fascination with the origins of language dates back thousands of years.

Another experiment said to be conducted by King Frederick II of Sicily was described in the 13^th century by Franciscan friar Brother Salimbene.

…he made linguistic experiments on the vile bodies of hapless infants, bidding foster-mothers and nurses to suckle and bathe and wash the children, but in no wise to prattle or speak with them; for he would have learnt whether they would speak the Hebrew language (which had been the first), or Greek, or Latin, or Arabic, or perchance the tongue of their parents of whom they had been born.

However, it is uncertain whether this experiment actually occurred. If it was conducted, it appears it was a failure, judging from Brother Salimbene’s description of the outcome. “But he laboured in vain, for the children could not live without clappings of the hands, and gestures, and gladness of countenance, and blandishments.”

Other accounts describe more extreme methods taken to ensure sufficient isolation. King James IV of Scotland is said to have left two infants, in the care of a mute nanny, on the isolated island of Inchkeith. Akbar the Great, a Mogul emperor of India, reportedly locked away a small group of children in a dedicated building known as the Gang Mahal, or “dumb house” where any form of speech was forbidden.

Unfortunately, in addition to the obvious ethical concerns, documentation of these early experiments is sparse, making it difficult to determine whether they actually occurred as described or draw any conclusive results. In modern times, deliberate isolation of helpless infants to satisfy scientific curiosity is obviously no longer considered an acceptable approach, but “natural experiments,” often resulting from strange and sad circumstances, have continued to provide insight into our innate language capacity. Case studies of children who grew up in isolation, often due to abandonment or abuse, suggest that, perhaps unsurprisingly, these situations are not exactly ideal for language development.

Perhaps the most well-known example of this is the case of Genie, a child who was found at the age of 13 after spending almost her entire life locked in a small room with minimal human interaction and almost no exposure to speech. At the time she was removed from this abusive situation, she did not speak. She quickly began imitating words and within the first year made impressive progress in learning to speak and understand English. However, she continued to show difficulty in fully acquiring English grammar, even many years later.

Similar language difficulties have been observed in other children isolated from human contact for extended periods of time. They do not generally develop strong language skills and, especially in the case of older children, many have difficulty learning language even if they are later exposed to it. This has been proposed as evidence for the Critical Period Hypothesis, the idea that fully acquiring language is difficult or impossible after a certain age. However, in these cases, it’s pretty much impossible to distinguish the effects of extreme social isolation and lack of linguistic input.

It is often reported that twins develop their own secret languages for communication with each other. It is true that twins often converse with each other using speech unintelligible to others, including close family members. However, it wouldn’t be quite correct to refer to this “twin speak” as a language in the strictest sense. In cases where the private speech of twins has been studied, the vast majority of vocabulary can be traced back to the languages the children have been exposed to, with changes to the sounds and simplified grammar. In fact, many children go through a phase where their speech is unintelligible, with all but the closest of family members. For twins (and sometimes other children close in age being raised together), the difference is that they have a readily available partner at a similar phase in speech development. Add in the fact that the twins will copy each other’s speech just as readily as that of adults and the feedback loop can quickly result in speech that sounds like gibberish to anyone but the twins themselves.

One famous example of this is that of Virginia and Grace Kennedy, also known as Poto and Cabengo in the private speech they used with each other. The two girls had minimal contact outside their family and were not sent to school. In the early years of their life, they spoke almost exclusively in a unique style of speech that no one understood but them. After the girls were referred to speech therapy around the age of 6, their speech was found to be largely derived from a mixture of English and German (the two languages spoken in their household), with some invented words thrown in. The relative social isolation the girls experienced is believed to have been a contributing factor to the development of their unique way of speaking.

Deaf children without early access to sign language provide some insight into the case of linguistic deprivation without the same degree of social isolation. In the absence of an established sign language, deaf children and their families often develop a form of communication known as homesign. These systems of gestures allow for some communication with family members, but the gestures typically vary from family to family and lack the complex, regular grammar found in established sign languages. Early exposure to language input, including this kind of complex grammar, appears to be a stepping stone to later language learning. Lack of early exposure to language makes it more difficult for children in this situation to later learn any language, spoken or signed. Interestingly, this means that deaf children who have early access to sign language actually end up showing higher proficiency in not only signed, but also spoken language as adults.

New languages can also form when two or more languages come into contact. In these situations. speakers may develop a pidgin in order to communicate with each other. These often combine material from multiple languages into a sort of simplified “emergency” language that people who don’t share a language can use to communicate. In some situations, children may grow up in an environment where a pidgin is the main language spoken around them. In these environments, the language can evolve into a creole. Unlike pidgins, creoles are spoken by children as a native language and have the full functionality of a natural language. This process can happen very quickly, often within a generation or two. That is, even with input that lacks some of the usual features of language, kids can fill in the blanks and add complexity to the system quite quickly.

In the past, creoles were often viewed as corrupt or imperfect versions of languages they developed from. Haitian Creole developed through contact between French colonists and African slaves in the 17^th and 18^th centuries. Much of the language’s vocabulary is drawn from French, but its influences also include languages from West Africa and the Caribbean as well as Portuguese. Despite being the first language of most residents of Haiti (around 9 million people), it was only recognised as an official language in 1987. Even today, French is still considered more prestigious in many contexts and school is most often taught in French. However, in Haiti and elsewhere, creoles are gradually gaining prestige and acceptance. Far from being lesser forms of their parent languages, creoles are some of the best evidence we have of children’s remarkable ability to create and modify language to suit their needs.

While ethical standards are much more strict than in the past, scientists have continued to look for creative ways to observe the process of language creation. In the late 1970’s, one researcher, Derek Bickerton, made a serious proposal to take six families with young children, each speaking a different language, and leave them all isolated on an island for a year. The idea was to observe how the young children learned to communicate with each other over the course of the experiment. The adults would farm coconuts and communicate using a simple system of easy to pronounce words provided by the researchers while their children (presumably) kept themselves busy inventing a new language. No problem at all, right? The idea may sound like a wacky reality show pitch, but Bickerton actually made it as far as scouting out an isolated island in the Philippines for the experiment and getting ethics approval from the University of Hawaii. However, funding ultimately fell through and the experiment never took place.

There are still many unanswered questions, but humans (or our ancestors) must have invented language somehow. So, how did we go from no language at all to the 6000+ languages spoken around the world today? While languages show a dizzying variety of features, they also have many striking similarities. The degree to which language and its rules are genetically innate remains a source of heated debate, but it is clear that humans are well-equipped for language and have at least some innate ability for language learning. Even deaf infants babble initially (and can also babble with their hands) and newborns already recognize some characteristics of their mothers’ language due to exposure before birth. Anyone who has watched how quickly toddlers’ vocabularies can explode to hundreds of words, often before they are even out of diapers, can attest that humans are veritable language-learning machines. It’s hard not to wonder how long it would take for children to create a language of their own if given the opportunity.

It might seem that such questions would forever remain in the realm of idle speculation, but around the same time as Bickerton’s proposed experiment, the unique ingredients needed for the birth of a new language came together in an unexpected place— a school in Managua, the capital city of Nicaragua.

Prior to the late 1970’s Nicaragua did not have an established Deaf community and few educational resources for deaf students existed. This was a tumultuous time in Nicaragua, with a devastating earthquake in 1972 and a violent revolution in 1979.  Lack of access to education was a problem across the country, and not just for deaf children. At the time of the revolution, it was estimated that as few as one in five rural Nicaraguans were literate. It was against this unlikely backdrop that Nicaraguan Sign Language was born.

In the late 1970s, a special education school with a program for deaf children in grades 1-6 was opened in the capital city of Managua. Within a few years it had around one hundred students. In 1980, the new government opened a vocational school, also in Managua. Instruction at these schools didn’t include any sign language. As was common practice at the time, the focus was on teaching the children the spoken and written language of their community, in this case Spanish. Unfortunately this method, known as oralism, tends to result in impoverished language input and generally does not lead to native-like language acquisition, so these children did not become fluent speakers of Spanish either.

These early attempts at language teaching may not have been successful, but, unknown to their teachers, the children were taking matters, quite literally, into their own hands. Gestures were not allowed in the classroom, but outside of class, the children were busy doing what children everywhere do—playing, socializing, and interacting with their peers. On the playground and on the bus to and from school, the children were spending time together every day. Teenage students were hanging out together after school and beginning to date each other. Within a few years, teachers began to notice that the children were gesturing prolifically among themselves, although no adult was able to understand the gestures. In 1986 they brought in outside researchers to help them figure out what the children were saying. What they found was remarkable.

Older students at the vocational school had all brought different homesign systems with them, but had converged on a system that let them communicate with each other. In many ways, this system resembled a pidgin, the kind of emergency, makeshift language that lets speakers who don’t share a common language communicate. Younger children at the elementary school, however, signed more fluently and had expanded on this system, adding rules, regularity and complexity. For example, they had developed a system of using location in space to indicate agreement between a verb and its argument.

What was remarkable was that no outsider taught them this. The youngest children were using the most complex structures and signing the most fluently, so they could not have learned this from their older classmates. The structure of their signs was also completely different from that of Spanish. It appeared that these children had simply invented the language, more or less out of thin air. In less than a decade they had created themselves what no adult had managed to teach them–a language.

Returning to our original question, would a completely isolated group of children be able to invent their own language? Assuming their basic survival needs were met, it seems likely they would come up with some way of communicating. Social interaction and development across generations appear to be important ingredients as well, so there would need to be a large enough group of children and it’s possible that, much like in the development of pidgins into creoles, a second generation would add complexity, making something more like a fully-fledged language. We may never know for sure what communication would look like in a population raised in complete isolation, but the natural experiments in language formation show us that humans have an innate drive to communicate, a capacity for creating language, and that seemingly with the current state of our brains, it doesn’t take long for us to develop a fully featured language.

If you liked this article, you might also enjoy our new popular podcast, The BrainFood Show (iTunes, Spotify, Google Play Music, Feed), as well as:

• The Origin of Language
• How Far Back in Time Could a Modern English Speaker Go and Still Communicate Effectively?
• Where Did the F-Word Come From?
• The Contested Title of the Person who Speaks the Most Languages
• Did English Speakers Really Not Use Contractions in the 19th Century as Depicted in True Grit?

Bonus Fact:

Even in songbirds, it appears that learning from an adult is often a necessary step in developing the ability to vocalize and communicate. Baby male Zebra Finches deprived of an adult “tutor” will still learn to sing, but not the in same way as their peers (In case you’re wondering, just listening to a recording doesn’t work as well. Yes, scientists have tested this.) Like many songbirds, male Zebra Finches utilize their singing skills in order to attract a mate. If they don’t get the chance to learn from an older mentor, their resulting sub-par singing skills may even mean they are less likely to attract the attention of the lady Zebra Finches. Interestingly however, one study found that when isolated Zebra Finches were allowed to raise young of their own, successive generations gradually altered the song. Within 4 or 5 generations, the descendants of these isolated birds were producing songs very similar to those of their wild counterparts.