Speaking of Words
By MICHAEL FERBER
How did human language get started? We have no idea. End of column.
Well, all right, there are a few ideas, based on a few facts, but they don’t amount to an explanation, not yet anyway. It used to be taboo among linguists even to discuss the origin of language at conferences, but for the last fifty years it has gotten more respectable as new theories have taken hold about what language is, and research on the brain, DNA, and primate evolution has made strides. It is possible to foresee a day when a plausible answer to our question may be offered. Right now, however, the very existence of human language is a mystery.
To get anywhere on this project we need to have a clear notion of what human language is. When it was thought to be a skill like any other skill, such as chess or basketweaving, then it followed that language was something we teach to infants, at first through memorization and imitation and lots of practice. The first humans to acquire language, it was assumed, must have imitated other animals, who made sounds in their mouths and throats and communicated information, warnings, etc, to their kind. This view has been called the “bow-wow” theory, and it seems very unlikely; even if it were true, moreover, it would only put off the mystery by a step. We would still be scratching our heads over how chimps or dogs acquired language.
Noam Chomsky has argued, I think convincingly, that human language is fundamentally different from all other animal languages, if we can even call them languages, and that it is biologically built into our minds or brains. We have a unique innate language faculty, some of which has been located through brain scans. Animals and birds don’t have it. They have something else, various faculties that may be very subtle and difficult for humans to understand, and perfectly adequate to their needs and intentions. There has been little success, however, in teaching chimpanzees or dolphins human language, and not only because their vocal tract rules out imitating our sounds; they can learn a lot of words, but their command of syntax, or the rules of word order, is elementary, even after years of training.
Not only do human babies pick up syntax rules very quickly, simply by listening to other speakers, but they also quickly pick up phonemes, the set of distinctive sounds each language adopts to make words out of. A very small set of phonemes, even fewer than twenty, is enough to generate many thousands of words. American English has forty phonemes, which is higher than the average, and with them we have room to create tens of thousands of words that are not yet in the dictionaries. We recently coined “bling” and “fomo”; if we need them, “fing” and “blomo” are available, waiting in their pigeon holes.
Babies, then, by the time they are just a few months old, have an eager brain, equipped with a syntax machine, a phoneme detector, and a memory compartment devoted to vocabulary. They devour words and phrases at a rate perhaps of one every hour, until puberty or very early teens, after which their power weakens and more or less shuts down—just about the time in America when we normally start teaching them a foreign language!
If this is true of human infants, then it would seem that some genetic mutations took place during the fairly recent history of hominids, perhaps only in the homo sapiens branch, though it is not unlikely that neanderthals also had language. One gene so far has been located, known as FoxP2. In the speech of families where that gene has been damaged, some grammatical features are lacking, though their grammar is not disabled severely. There must be more genes involved, or perhaps a particular gene, difficult to identify, that enables several other genes to work together, a meta-gene. We don’t know yet.
Could language, through a group of mutations, have arisen all at once? Mutations, we believe, take place randomly, though there are new theories about the “epigenome” that have suggested that they may not all be random, or that the “expression” of genes in the genome may be encouraged or suppressed by changes in the rest of the human body, such as stress, illness, or diet. Perhaps three or four mutations were enough to make the human language faculty, and perhaps the neanderthals, who could interbreed with us homo sapiens types, had all but one of them.
The problem comes down to this. We are used to thinking of evolution as mainly gradual, coming about through great numbers of mutations and through natural selection of the better ones over great spans of time. Human eyes, for instance, which used to be held up as evidence against evolution, so complex and perfect do they seem (not mine, however), can be traced back hundred of millions of years in thousands of different animal eye types, or proto-eyes. But human language seems not to have had a long history of proto-languages. Every language on earth today seems all of a piece. A language whose words are made up of phonemes must have words made up of phonemes all the way through, surely, though it is true that most languages also have a modest repertory of imitative sounds and interjections not fully based on its phonemes. The rules of syntax, too, must govern all words and phrases, not just some of them. It would seem that either you have a human language or you don’t; it is difficult to imagine what a proto-language or intermediate language would be like.
In the silly 1981 movie Quest for Fire, early humans, some still rather ape-like, had two ways of speaking. One resembled chimpanzee sounds and gestures, during which the people would squeal and hop around and wave their arms extravagantly. The other sounded like Proto-Indo-European—the word for an antlered stag was teerdondra (I’m not sure of the spelling), which is simply a compound of German Tier (“animal”) and Greek dendra (“trees”)—and they spoke it like Roman senators in the forum. That was hilariously unlikely, but it underscored how different human speech is from chimpanzee speech. Something big happened not long ago, and it may have been abrupt.
It is also possible that human language arose in tandem with music and dance. This is what Jean-Jacques Rousseau thought in the 1750s, and some recent anthopologists and linguists have considered the idea anew. Music, song, choral song, and choral dancing all seem distinctively human, though there are partial parallels among some animals and birds. Music seems deeply embedded in the human brain, though there are enough cases of congenital tone deafness or amusia (about 4% of the population) to suggest that music is not quite as deeply embedded as language is. Beat deafness, or an inability to keep time or respond to rhythm, is rarer. When Alzheimer’s ruined my mother’s mind to the point where she could no longer speak, she could still sing old songs. The musical part of her brain preserved her words longer than all the other parts. Did music, then, get human language going in the first place? If it did, we don’t know how it arose either, so we may be only compounding one mystery with another.
Ancient DNA research has exploded in the last decade, so perhaps we will soon discover a crucial gene or two, both for language and for music. But even if we find them, we may need bold new ideas to account for how they led to human language, that unique and wonderful power we all deploy brilliantly without knowing how we do it.
I am happy to hear from readers with questions or comments: mferber@unh.edu.
Michael Ferber moved to New Hampshire in 1987 to join the English Department at UNH, from which he is now retired. Before that he earned his BA in Ancient Greek at Swarthmore College and his doctorate in English at Harvard, taught at Yale, and served on the staff of the Coalition for a New Foreign Policy in Washington, DC. In 1968 he stood trial in Federal Court in Boston for conspiracy to violate the draft law, with the pediatrician Benjamin Spock and three other men. He has published many books and articles on literature, and has a deep interest in linguistics. He is married to Susan Arnold; they have a daughter in San Francisco.
