How Language Began

by Daniel L. Everett

  Schleicher’s and Jones’s work inspired others to think deeply about the relationships between languages. It became apparent that, using the method of constructing language trees that began to be developed in India, Germany, France, England and elsewhere, we could look back into time in our search for where and when specific languages originated. It was eventually discovered that Indo-European was the mother of most European languages. And it was then discovered that this was also the mother of non-European languages such as Farsi, Hindi and many others. The question thus naturally arose whether we could discover the mother of Indo-European itself. We know now that Indo-European began to split into the modern European languages approximately 6,000 years ago. Can we go further? Ten millennia? A hundred? Could we actually use the methods of comparative and historical linguistics to reconstruct the first language ever spoken?

  Most contemporary linguists respond to this question with a firm ‘No’. The methods Jones’s work called attention to seem to hit the wall at about 6,000 years. To go deeper we will need the methods of other fields, such as palaeontology, archaeology and biology – and we’d need what we probably can never have, preserved samples of languages.

  But the question remains. If we were able to travel back further than those 6,000 years, where would we end up? Would the quest of Jones, Schleicher and others take us back to a single language at the root of one enormous tree of human languages? Some people think so. The late Stanford professor Joseph Greenberg claimed that we could trace all human languages back to a single source, which he and his followers labelled proto-sapiens. But other scholars say no. They maintain that there are many trees all going back to different prehistoric communities of hominids. Greenberg and his disciples believe in monogenesis, the hypothesis that there is only one beginning – one mother tongue – for all human languages. Others advocate polygenesis, that there are multiple evolutionary beginnings for modern human languages. These people argue that the ancestors of modern humans left Africa speaking different languages. Different communities of speakers developed different languages that in turn are the sources of all modern languages. Choosing the best hypothesis, monogenesis vs polygenesis, is just one of myriad problems we face in trying to reconstruct the evolution of human languages.

  We know that other methods, other sciences beyond linguistics, can take us further back in time. But can they transport us to the beginning of human language? Can we know anything about who told the first story? Or who first said, ‘I love you’? The romance and the science come together in this story of the search for human language origins. It is a tale replete with scientific controversy and marked by frustratingly slow progress towards the ultimate goal of knowing how humans, but no other species, moved from mere communication to language. Although historical linguists believe that the methodology of their field is unable to reveal much once we go beyond 6,000 years ago, its major insight, that languages change over time due to a combination of cultural and linguistic reasons, is essential to an understanding of language evolution.

  Historical (or ‘diachronic’) linguistics, the field virtually launched by Jones’s work, is the field dedicated to understanding how languages change over time. For example, English and German were once the same language (‘Proto-Germanic’), as were Spanish, Portuguese, Romanian and French (Latin). And we know that Latin and Proto-Germanic were themselves one language some 6,000 years ago, Indo-European. The science of how languages drift apart as these languages have is one of the oldest branches of the study of language and it is relevant to the field of language evolution. After all, if Homo erectus underwent evolution to become Homo sapiens, maybe the language of Homo erectus also changed into the languages presently spoken by Homo sapiens. Any change in erectus languages, however, would be beyond the science of historical linguistics. That is because erectus lived much longer than 6,000 years ago. Even one of the main tools used by historical linguists for dating when one language in all likelihood split to become another, glottochronology (literally, ‘tongue time’; referred to by some as ‘lexicostatistics’) is of no assistance to here. Glottochronology, invented by linguist Morris Swadesh, assumed that there were some vocabulary items (such as parts of the body, words for sun, moon and others) that were less likely to be borrowed. He therefore came up with a list of two hundred words or ‘lexical items’ which he considered represented the words least likely to change. A mathematical formula was proposed and developed, based on the rate of change of the words in his list, to predict the rate at which these most resistant-to-change words might in fact change over time. The formula was tested and deemed to have 87 per cent accuracy in known cases, such as the Indo-European languages. Though many linguists still are highly sceptical of the method, it does seem to be useful. But it is not able take us back further than the 6,000 years wall. So it is not a tool for language evolution.

  What it and the entire field of historical linguistics do show, however, is that languages continue to change. In fact, linguists recognise that change in modern languages is largely the result of a form of linguistic natural selection that would have certainly been operative in the first languages ever spoken. All languages change all the time. They change because of geographical separation (think ‘genetic drift’), or from differing preferences of age, economy, race and many other factors. And these forces in one form or another mean that the languages of Homo erectus began to change as new communities were formed. Much of historical linguistics boils down to the idea ‘You talk like who you talk with.’ Once you stop talking to a group of people, you will eventually stop talking like them. Or at least your group will. This is why each time we reach a major river or mountain range in Europe, we are likely to find distinct languages on each side that were once the same language. As for English and German, English was born from German after the crossing of the English Channel by the Saxons.

  Because language is a cultural artefact, we must understand what culture is in order to understand language. So what is a culture, then? Is a culture like a football team? Or perhaps like an orchestra? Or is culture simply the overlap in values, roles and knowledge of individuals who live together and talk together? The larger issue is how cultures hang together at all. In other words, in what sense might the American motto E pluribus unam * describe American culture? Since I claim that culture is an abstraction, it can only be found in the individual. It is the result of a ‘gestalt’. From its various individual members, a culture emerges which is greater than the sum of its parts. Understanding culture has profound implications for understanding language evolution.

  I have in my own work developed a theory of culture in which the individual is the bearer of culture and the repository of knowledge, rather than the society as a whole. I want to look at the effects of culture on the nature of national and local societies and individuals and their languages, via examples such as the role of the teacher in the classroom, the organisation of businesses and the organisation of societies. The three ideas of my work that are most important for language evolution are values, knowledge and social roles.

  As a further illustration of the importance of culture in language, consider the following interactions between two linguists:

  A. ‘Colourless green ideas sleep furiously.’

  B. ‘They sure do.’

  The general population of English speakers may have no idea what A’s utterance means. But if A and B are members of the culture of linguists, then they know that this is a famous example sentence in Chomsky’s early writings that is designed to show that a sentence can be grammatical yet meaningless. For the two linguists, A’s sentence is an insider joke and B’s response a humorous rejoinder. The function of the exchange might be largely phatic, simply to say, ‘Hey, we are both linguists.’ What is often overlooked, though, is that B’s reply shows that ‘Colourless green ideas sleep furiously’ is not, in fact, meaningless. It tells us that, whatever green ideas are, they sure do sleep furiously. In other words, because of the cooperative
principle, all people will believe that an utterance has meaning and will work to attribute meaning to it, regardless of what words it is composed of.

  Now consider the following. Persons C and D are watching the New England Patriots play the Miami Dolphins. The Patriots take the lead. C and D both yell, ‘Yes!’ and ‘high five’ one another. In this joint action they show knowledge that there is a game of football, knowledge of how this game is scored, shared value-ranking for the Patriots relative to the Dolphins, knowledge of what ‘high fives’ are like and what they are for, knowledge that they are both ‘rooting’ for the same team and reinforcement of all of the above.

  From such culture activities come knowledge, community belonging and shared communication. These exemplify the role of living in a culture and speaking a language in constructing our identities and our societies. From these actions the individual assembles his or her own experiences and an ability to understand the actions and speech of their fellow culture members. As an example of how unspoken much of culture is, it is worth reviewing one of the many failed treaty attempts between North American indigenous peoples and their European-immigrant conquerors.

  A historical incident, the famous Treaty of Medicine Lodge, signed in 1867 between the Arapaho, Kiowa and Comanche peoples and the US government at the Medicine Lodge Creek in Kansas, was simply one of many failed communications between European and indigenous communities based on misunderstandings caused by the tacit cultural information that is required to interpret what language leaves unsaid. A serious and war-threatening misunderstanding grew out of two distinct cultural interpretations of this deceptively simple-looking treaty. The indigenes expected one thing. The government expected another. And both were right according to the language. This is a common source of misunderstanding between individuals, across cultures and internationally. Such misunderstandings boil down to culture’s role in filling in the gaps that language is missing and underlying the interpretation of language itself. I want to look at the Medicine Lodge misunderstanding in more detail.

  For more than a century, anthropologists have bickered about the definition of culture. The members of a given family, community, society, or nation clearly share some knowledge, some values and some relationships. They may talk alike. They may dress alike. They may show disgust at similar things. They may all drink coffee from their saucers.

  So the question arises naturally, ‘What is culture?’ Culture is the tacit knowledge and overt practice of social roles, values and ways of being shared by a community. Each of us has many different roles. I am a father, a teacher, an administrator, a husband, a shopper, a patient and a researcher. Each one of these roles is recognised by most members of my community. To the degree that they are recognised, my community shares knowledge of these components of culture with me. Culture distinguishes and shapes us, even when our roles may seem universal, such as ‘father’, and so might appear independent of culture on the surface. But although there are Italian fathers and American fathers, the concept of ‘father’ is not identical in every culture. It seems likely that between any two cultures, fathers will have overlapping but never identical roles. Even fathers of ostensibly the same cultures vary in the nature of their roles at different times.

  Some societies may believe that fathers should support their families. In such a society, it may be assumed that fathers have a responsibility to provide food, clothing and shelter for their children. And, in Western societies at least, both the society and many fathers themselves believe that it is good for fathers to help their children with schoolwork, heavy lifting and tasks in general too difficult for children to do alone. Fathers of other generations may share exactly these beliefs and values. But these values are not identical across different cultures. A Pirahã father will not often pick up a child that has injured itself to offer comfort. He will expect the child to work hard and not complain on long treks through the jungle and will not offer assistance in many cases that the American father would. And his individual values emerge partly from the values of other members of his society.

  And, of course, across different generations, fathers may differ profoundly. Values shared by many of my father’s generation included corporal punishment, the expectation that women did the bulk or all of the housework, the belief that their wishes and orders would be carried out without question and the attitude that their children were not deserving of respect or of a voice in family affairs. These fathers might regularly side with teachers against their own children in disputes. They considered the child and all its resources as mere extensions of themselves or their possessions. The fathers of the generation of my children, on the other hand, usually avoid corporal punishment, see their family as a unit of equals, know they ought not to believe that their desires should be the only or even the main ones heard, often help clean the house, would almost always take their children’s side in a school dispute and so on. Being a father in the 1950s was considerably different from being a father in the twenty-first century. This is because the cultural role of ‘father’ is defined by shifting cultural values.

  If my quick summary of the evolution of fathering over the past few years is on the right track, then the fact that the changes affect entire generations similarly indicates sharing of values – culture. This is part of what it means for a group to have a culture. All cultural roles show similar diachronic, geographic, economic and other shifts across time, space, or populations. If we move from roles to beliefs or from beliefs to shared concepts, to shared phenotypes (a phenotype is the visible appearance and behaviour of an individual), shared food and shared music, we will find many examples of shared knowledge producing distinct cultures.

  In part these shared mental items emerge because over the course of one’s life, each accumulates experiences, lessons and relationships. These are all in a sense assimilated into our bodies and minds. People who grow up in the same community have similar experiences – climate, television, food, laws and values (such as fat is bad, honesty is right, hard work is godly). Episodic and muscle memories hold our various experiences together as cultural experiences embed themselves within us. Arguably our ‘self’, or at least our ‘sense of self’, is no more than this accumulation of memories and apperceptions.

  How does one recognise another as part of the same culture? Indexes are readable by members of any culture, in fact by members of most species. They are clues to the environment necessary for survival across many life forms. And thus it is known that the ability to connect a representation to a form is an ancient ability of the genus Homo. Humans were never without it. Icons, on the other hand, require more. Whether making an icon or simply collecting one, the reader of the icon must understand that it physically resembles what it represents. Understanding, whether indexes, icons, or symbols, is an intentional act, directly or indirectly. (This is because to understand requires at least tacit recognition of connections between the sign and the thing it refers to.) An index is itself, however, non-intentional. One doesn’t plan for a footprint to be connected to man. It just is.

  The ability to interpret cultural information comes slowly. Everyone is born outside of culture and language. We are all partially aliens as we emerge from our mothers’ wombs (‘partially’ because learning about our new culture and language begins in the womb). When we are born, we are outside our mother’s culture looking in. Our senses provide information. But it takes time to interpret what we are seeing, feeling, tasting, touching and hearing.

  Examples of misunderstandings caused by culture are plentiful in the history of the Native American and ‘manifest destiny’ in the nineteenth century. Communication between them failed many times due to an inadequate appreciation for each other’s culture, just as communication between governments in the twenty-first century often do.† An example alluded to earlier is the 1867 Treaty of Medicine Lodge.

  This treaty was ineffective from the beginning. For once at least an official treaty with the Indians was invalidated not bec
ause of dishonesty on the part of the government but because the signatories failed to realise that language, whether spoken or written in treaties, is merely the visible portion of an invisible universe of understanding that derives from the values, knowledge and experiences – the cultures – of individual communities. Though people might read the same words in a treaty, as in all communication our interpretations are slaves to our assumptions, based on background beliefs and knowledge that the literal meaning of the words rarely conveys.

  In this case, the treaty called for the government to provide food to the Indians so that they could feed their families through the winter months. The Indian agency was responsible for providing the food. Congress was responsible for ratifying the treaty that was signed. Each in turn depended on other cultural institutions, all with their own deadlines and priorities. The Indians gave no thought about ratification. But they should have, because when they arrived to collect their provisions, prior to ratification of the treaty, the pantry was bare. The government had provided nothing because the treaty had not yet been approved. Regardless of the reason, the Indians felt betrayed.

  On the other side, the government thought that the Indians, when they had agreed to live in the reservations, would now consider themselves bound to stay there in perpetuity and to forever abide by the ‘law’. Perpetual obligations to anyone other than their own families were foreign to the Indians’ values and understanding of the way the world worked. They could never have legitimately made the commitment expected of them. It made no sense. The government could not have cared less about Indian interpretations rooted in their very different cultures. But it should have. Comanche chief Quanah Parker, present at this ill-fated gathering, at least learned from the experience. In his future dealings with whites he learned to respect the importance of the dark matter of the unsaid. He subsequently inquired about every potential assumption that he thought whites might be making before signing future treaties (though no one outside a culture can ask all the right questions).