How Language Began

by Daniel L. Everett

  Figure 3: Primate phylogenetic tree

  But the story of primate evolution begins in earnest with the first-known true primate genus, the Teilhardina, which were the precursors of all other primates, including us. The Teilhardina (named after the Jesuit theologian and palaeontologist Pierre Teilhard de Chardin) were small creatures, similar in size to modern marmosets. As these creatures developed and found separate niches, they formed varieties and eventually new species. From the new species came new genera.¶

  Many primates break down into genera according to the shape or conditions of their noses. For example, the wet-nosed or strepsirrhine primates today are limited almost exclusively to the lemurs of Madagascar. Humans are counted among the haplorhini or ‘dry-nosed’ monkeys. The haplorhini in turn break down into the tarsiiformes and simiiformes – simians, or anthropoids – from which humans come. There is further subdivision into catarrhines (long noses) and platyrrhines (flat or ‘downward’ noses). All monkeys native to the New World have nostrils that point sidewards, while all Old World primates, including humans, have nostrils that point downwards. Figure 3 summarises the classification of humans among other primates.

  Figure 3 shows the primate tree with humans as part of the infraorder catarrhini, as previously described. As well as being catarrhines, humans are also, along with all monkeys and apes, simians. The platyrrhines of the Americas evolved differently from their Old World relatives, producing no apes, only monkeys. No humans therefore evolved in the New World.

  All primates can climb trees. Though I confess that I cannot climb as well as a woolly monkey, all humans possess, as primates, bodies designed at least partially for tree climbing. Some also stand out from other mammals for more important reasons than their climbing prowess: humans have large brains, a reduced sense of smell and stereoscopic – 3D – vision, all traits that have helped them become the rulers of a very real planet of the apes.

  Evolution is the great procrustean bed of life, where species are stretched, chopped and otherwise modified to fit their niche. As evolution brought the primates forth from the ‘great mammalian surge’ that followed the end of the dinosaurs during the Cretaceous–Paleogene (K–Pg) extinction event some 66 million years ago, it launched another in a long line of cognitive leaps. Eventually, this long mammalian advance produced the hominids.

  Most modern scientists use the term hominid, the singular form of the Latin word hominidae, men, to refer to all the great apes (the English form of the plural is hominids). There are a few who still prefer to use the word in its older sense, to refer not to all the great apes, but to the genus Homo. But most modern scholars reserve hominid to describe collectively the great apes and hominin (hominini), humans and all of their direct ancestors, beginning from the time that chimps and humans split into separate branches of the primate tree. Using two terms, hominid and Homo, does not increase the perceived complexity of human evolution; it simply helps to avoid ambiguity.

  By contrast, Darwin’s theory of evolution by natural selection is much less complicated. Indeed, part of its attraction and elegance is its very simplicity. It consists of three postulates:

  First, the ability of a population to expand is infinite in principle, though the ability of an environment to support a population is always finite.

  Second, there is variation within every population. Therefore, no two organisms are exactly alike. Variation affects the ability of individuals to survive and multiply as some produce more viable offspring in a given niche.

  Third, somehow parents are able to pass their variation along to their offspring.#

  Together, these three postulates have come to be combined as ‘descent by natural selection’. Darwin noticed during his famous trip aboard the HMS Beagle (1831–1836) that animals seemed adapted to fit their environment. And throughout his scientific career he also noticed how different creatures (such as his famous Galapagos finches) adapt rapidly to even small changes in their environments.

  Ironically, while Darwin was proposing natural selection, a Czech monk, Gregor Mendel, was laying the foundations of genetic theory. Working with garden peas for seven years (1856–1863), Mendel developed two important principles of genetic research:

  First was the principle of ‘segregation’, which is the idea that given traits are broken into two parts (known now as alleles) and only one of these parts passes from each parent to their shared offspring. Which allele is passed along is random.**

  Second came the principle of ‘independent assortment’. By this principle the pairs of alleles produced by the union of the parents’ haploid cells form new combinations of genes, that are not present in either the mother or the father.

  Though Mendel’s work eventually gained dramatic acceptance, there were still problems that were overcome only with the advent of modern genetics. First, as Thomas Morgan showed in his now famous work on fruit-fly mutations, genes are linked; that is they work in tandem in many traits. This is contrary to Mendel’s idea that each gene behaves independently, an idea that many geneticists had embraced prior to Morgan’s work. Morgan’s findings meant that the independent assortment of genes wasn’t right – contra Mendel. Morgan also brought to his research original and long-term interests in cell structures. Through this work it became clear that chromosomes were real entities in cells and not merely hypothetical gene vehicles.1 Second, Mendel suggested that variation is always discrete, but in fact it is usually continuous. If one mates a 6'5" mother with a 5'2" father one will not get simply either a 5'2" vs a 6'5" child, but any height between those at minimum, among other possibilities. In other words, many (in fact most) traits blend. This is not captured in the simple interpretation of independent assortment from Mendel’s work on peas, in which all traits he worked on were discrete and, for the purposes of his research, unrelated.

  Another crucial fact about evolution is that the targets of natural selection are phenotypes (externally visible physical and behavioural attributes, resulting from genes and environment), not genotypes (the genetic information that is partially responsible for the phenotype). Thus natural selection operates on (selects for survival) creatures based on their behaviour and overall physical properties. Genes underlie these properties and behaviours, but the phenotype is more than genes: it is partially produced by histones, environment and culture. The histones control the timing of the unfolding of genetic information and thus how the genes produce the phenotype.

  When non-biologists think of evolution, they often conjure up ideas of new species, but, though that is one by-product of evolution, looking only at speciation as evidence is misleading. If a creation ‘scientist’ says that creatures can’t transform into others, and that evolution must therefore be false, what they are actually disputing is macroevolution – evolution on a grand scale. However, macroevolution is not the only form of evolution. In fact, macroevolution is usually the accumulation of smaller evolutionary changes, perhaps as small as the mutation of a single allele, known as microevolution.

  While microevolution is, by definition, less discernible, especially to observers with short human lifespans, it is where the real action takes place. As a result, if one can explain the small changes, the larger changes will follow by and large. Evolutionary scientists seek to understand biological change over time (evolution) in all of its forms. Macro- and microevolution are simply points along a vast continuum of modification by natural selection.

  One of the ways in which micro- and macroevolution are stimulated is via mutation. Many mutations are neutral. Other mutations are fatal. But some mutations provide a survival advantage to their host organism. A change favoured by natural selection in a particular environment is advantageous if the mutated creature produces more viable offspring than creatures lacking the mutation.

  Neutral mutations are important for evolutionary theory even though they are by definition neither harmful nor helpful to the survival of their host. As Linus Pauling, the only person in history to win two unshared Nobel
prizes, one for chemistry and one for peace, and Emile Zuckerkandl, a pioneer in genetic dating, proposed in 1962, neutral changes occur at a constant rate over time. This constancy works like a molecular clock, one that can help determine when two related species diverged. Today it has become a vital tool in understanding evolutionary differences among creatures, even though they are not themselves responsible for those differences.

  Mutations favoured by natural selection, however, are not the only way that evolution works. For anything as complicated as the whole of life on earth, it should not be surprising to learn that no one concept explains it all. There are other sources of micro- and macroevolution than natural selection. One of these is known as ‘genetic drift’. Technically, in genetic drift there is a reduction in a population’s genetic diversity. Imagine that the population of all humans is one thousand individuals from one hundred families. Now assume that the genes that produce photopigments in five families from this one thousand, say fifty individuals, are deficient. These individuals are ‘colour-blind’. Next, imagine that these colour-blind individuals come to be shunned by the majority as undesirable for some cultural reason and that all fifty of them therefore decide to move elsewhere. Finally, suppose that the original population, the non-colour-blind individuals, are wiped out by disease or natural disaster after the departure of the colour-blind individuals. The colour-blind people are unaffected. This improbable but possible chain of events will result in a state where the only genes left among the species are the genes that produce colour-blindness. The colour-blind community may grow over time, producing many offspring and descendants, founding entirely new populations of humans. This scenario would result in significant changes to the human species, independent of natural selection.

  Genetic drift is a naturally occurring reduction in genetic diversity that is produced by Mendel’s principle of randomness in the selection of alleles. Again, this is not caused by natural selection because fitness plays no role in the result.

  A special case of genetic drift is known as a population bottleneck. A population bottleneck is an alteration in the allele ratio produced by external causes, as in our example of ostracised colour-blindness sufferers. Such a bottleneck can include things like migration, where a migrating population is some sample of the main population wherein there is a different allele ratio than that found in the population as a whole. Population bottlenecks include any reduction of the genetic diversity of a population caused by external events. Take a disease that kills off one member of each family. Chances are that reduction left a different gene distribution in the overall population, thus producing a population bottleneck. This too can lead to a ‘founder effect’ – a subpopulation with a different distribution of alleles than the original population that in turn produces generations of viable offspring. In other words, if the original population of Homo erectus that left Africa had a different allele ratio than the population that remained in Africa, the former and latter populations would be separate founder populations for the ensuing generations.

  Another form of evolutionary change is that effected by culture, a form known as the ‘Baldwin effect’, and particularly relevant to the evolution of human language. The Baldwin effect, first proposed in 1896 by psychologist James Mark Baldwin, was an important conceptual advance in evolutionary theory for at least two reasons. First, it underscored the importance of phenotypes (visible behaviours and physical characteristics) for natural selection. Second, it demonstrated the possible interaction of culture with natural selection. As a hypothetical example, let’s suppose that a population of Homo erectus enters Siberia in the summer only to discover later that Siberia is cold in the winter. Now assume that everyone learns how to make winter garments from bear fur and that the most effective stitching of these furs requires manual dexterity that is extremely challenging or unavailable to the community as a whole – except for one lucky person who has a genetic mutation that allows him or her to bend their thumb to their forefinger in such a way as to produce a more effective, long-lasting stitch for bear fur coats. They therefore make more effective coats for their family. This in turn allows the members of their family to produce more offspring than the families of the stitch-challenged. Eventually, the mutation will increase the chance that the original mutant’s ‘dexterity genetics’ will reproduce through their offspring who, in turn, out-survive (in winter at least) the offspring of the less dexterous stitchers. Over time, the dexterity gene will spread throughout the population.

  The same genetic mutation in another environment would not have propagated throughout the population because it might not have provided any survival benefit. It would be in another environment simply a neutral mutation, of which there are many. This might happen if the coat-making phenotype is neutral in a warmer climate, such as Africa. We can say, therefore, that culture can turn neutral mutations into positive mutations. The Baldwin effect, also known as dual inheritance theory, brings culture and biology together and seeks to explain those evolutionary changes that can’t be explained by either one on their own.

  Now using our imagination once again, let’s suppose that a woman is born during the time that humans are developing language. We will call this woman Ms Syntax. While the rest of the community says things like, ‘You friend. He friend. She not friend,’ Ms Syntax says, ‘You friend and he friend but she not friend.’ Or while everyone else is talking like, ‘Man hit me. Man bad,’ Syntax Lady might say, ‘The man who hit me bad.’ In other words, the syntax master has the ability to make complex sentences while the rest of the population can only form simple sentences. Could the entry of complex sentences into human language have been a mutation, spreading via the Baldwin effect or some other mechanism, such as sexual selection? This is unlikely. Language presents a different case than genes for physical skills.

  The first reason for doubting that a mutation for syntax could spread through a population or be favoured by the Baldwin effect is that it is unlikely that complex sentences would provide a survival advantage, especially in light of the fact that there are languages spoken today, as we discuss later on, that lack such complex syntax. These latter languages have survived in the same world as languages that do have complex syntax. Moreover, even if it were discovered that the languages currently claimed to lack complex syntax did show such syntax in some cases, this discovery would only underscore the fact that speakers of these languages survive fine in an environment 99 per cent free of complex sentences.

  More importantly for Ms Syntax, in order to be able to interpret complex sentences, one would need to be able to interpret complex syntax. Uttering complex sentences in a population that lacks the ability to do this – where, in other words, you are the only person that is able to interpret or produce complex utterances – would be like yelling a warning to a deaf, mute and blind person. One could argue that non-human primates can already do this, since they seem to be able to respond effectively to requests using complex syntax (the bonobo Kanzi comes to mind). But this is a far cry from actually being able to fully understand complex sentences. Following instructions given in recursive sentences, for example, might be a first step towards acquiring or evolving recursion, but only that. Ability to think in complex ways must precede talking in complex syntactic constructions or no one would be able to fully understand those utterances.

  But how might such thinking arise? How could someone think in ways that they do not speak? One possibility is, perhaps, by planning events within events via images within images or even, as many speakers in the world seem to do today, by thinking in larger stories that, although they use simple sentences, weave together complex thoughts:

  John fishes.

  Bill fishes.

  John catches fish.

  Bill stops.

  Bill eats John fish.

  Bill returns.

  John returns same time.

  This story, completely composed of non-complex sentences, says that John went to fish and later, or at the s
ame time – depending on which is inferred by the context – Bill went to fish. John caught fish before Bill. So Bill stopped fishing and ate fish with John. Bill decided to stop fishing and return home. John returned home with him. There are, in fact, many languages in which simple sentences are woven together in complex stories just like this.

  Another example of complex thinking without complex sentences is intricate task performance or planning without talking at all, such as in weaving a basket with many parts. As the hypothetical fishing story has just shown, complex sentences are not required for complex thinking or storytelling. Complex thinking might make it possible to utter and interpret complex sentences, but it doesn’t require them. The reverse, however, is not true. One must be able to compose complex meanings in order to interpret a complex sentence.

  On the other hand, it is possible that complex syntax would spread through a population by sexual selection. Members of the opposite sex might like to hear the melodic cadences of complex sentences and so might mate more frequently with Ms or Mr Syntax, spreading the syntax genes. But this is unlikely. Complex sentences normally require words that indicate that they are complex. Those words, however, are largely unintelligible outside of the complex syntax they have arisen to signal. For example, ‘John and Bill went to town in order to buy cheese’ is a complex sentence, because of its clause-within-a-clause, ‘in order to buy cheese’, but also because of its coordinate subject noun phrase, ‘John and Bill’. The word ‘and’ is not understandable apart from being able to think in complex syntax. Complex sentences themselves also require complex gestures and pitch patterns that would need to have come about separately and that are unlikely to have arisen owing to a single genetic mutation.

  What seems more likely is that complex thinking was favoured by natural selection and thus was a genuine Baldwin effect because it enabled complex planning. It might – and probably would have – shown up later in the form of complex sentences in some languages. In any case, the fascinating conclusion is that natural selection would quite possibly have treated a gene for syntax in language as a neutral mutation, not subject to natural selection.