The First Word: The Search for the Origins of Language (26 page)

Scientists say that in humans and songbirds, the gene is 98 percent the same. FoxP2 (nonhuman versions) appears to play a significant role in the learning and expression of song in birds like the zebra finch; its expression increases in certain brain areas at the developmental stage when the birds are learning how to sing. In addition, the expression of FoxP2 in canaries varies seasonally and correlates with a change in song.

The mouse and human versions of the gene are even more alike than the human and songbird versions, and it’s recently been demonstrated that FoxP2 affects the vocalizations of mice.
⁷
Scientists at Mount Sinai Hospital in New York showed that while mice with only one normal FoxP2 had some general developmental delays, more strikingly their patterns of vocalization were affected.
⁸
Typically, if a mouse pup is separated from its mother, it will produce cries that are above the range of human hearing. (It was only a few years ago that we learned mice produce sound in the ultrasonic range. In 2005 scientists at Washington University discovered that male mice sing to females in the ultrasonic range.) The purpose of the pup’s ultrasonic cries is to alert its mother to its whereabouts. Mice with only one working copy of FoxP2 produced far fewer vocalizations when separated from their mothers than normal mice. FoxP2 seems to play a role in both learned and innate vocal production. (The Mount Sinai researchers found that mice with disrupted versions of both of their FoxP2 genes had severe motor difficulties, lacked crucial vocalizations, and died prematurely.)

Even though language ability is not contained in one or two genes and somehow generated out of them, the FOXP2 work is compelling evidence that we need certain genes to have structured communication—and that human communication, of which language constitutes a huge part, depends in some measure on the same genetic foundations that animal communication does.

Gary Marcus, a professor of psychology at New York University and author of
The Birth of the Mind,
has worked closely with Simon Fisher, one of the geneticists known for FOXP2 research. Marcus explained FOXP2 by way of comparison to another gene, PAX6:

 

PAX6 is what is called a master control gene—a gene that achieves great influence by guiding the actions of other genes. Strictly speaking, what PAX6 does is the same sort of thing that any other gene does: it gives a template for building a particular protein, and information about when and where that protein should be built. But the protein that PAX6 governs influences the expression of other genes, telling them when and where other genes do their thing. And because it’s atop (or at least close to the top) of a hierarchy, PAX6 can have a huge influence.

One experiment showed that by switching on PAX6 in the right place on a fruit fly’s antenna, the fly can grow a whole extra eye in an entirely new place. FOXP2 may or may not be so high up the food chain, but like PAX6 it clearly does modulate other genes; if it’s not a CEO, it at least seems to be an important middle-level manager. The broader lesson is that all genes work as parts of hierarchies or cascades. PAX6 isn’t “the eye gene.” It’s a gene that can spawn an eye by influencing thousands of other genes. FOXP2 isn’t “the language gene” but it may have a profound influence by regulating the actions of many other genes.

 

After the discovery of FOXP2’s language effects, Steven Pinker hailed the possibilities for a new science: cognitive genetics. Vargha-Khadem and her colleagues called it neurogenetics. Whatever this new field ends up being named, the next century will be an exciting time of determining the closeness of the weave of genes, brains, and behavior. The old nature-versus-nurture debate will finally be shucked off and left behind.

In working out the way genes build linguistic brains, one of this new science’s greatest challenges is determining how experience affects the spread of job specialization across the brain. The dynamic interplay between genes and experience as it propels a creature through conception, development, sexual maturity, parenthood, and eventually death is greatly complicated by brain plasticity—which must itself, presumably, be underwritten by genes. Solving the mystery of language and its evolution will involve working out what is innately specified and what alternative routes to processing the same kind of data are enabled by plasticity.

As the field progresses, we will discover more about the reach of FOXP2. In his most recent book,
Toward an Evolutionary Biology of Language,
Philip Lieberman notes that in addition to vocal learning, “humans possess more cognitive flexibility than other species.” He argues that FOXP2 also underlies this trait, which itself gives rise to creative thinking, language, voluntary motor control in speech, and, perhaps, dancing.

The different research projects reviewed in this book do not line up perfectly with one another; still, much of this work inhabits the same intellectual space, and together it promises to explain at least some of the larger language evolution story.

When examined as a whole, the studies presented here signal a profound change of mood in the scientific community. In most disciplines the focus used to be on the separateness of animals and humans, that gulf being marked most strikingly by language. But over the last few decades, the emphasis has switched to investigating the continuity of life
in addition to
clarifying the boundaries that lie between species. We no longer have a sense that we are standing apart from all animal life and that language is a discrete, singular ability that isolates us.

Despite the initial controversy connected with examining the mental life of nonhuman animals, once this research began (in every field in which it’s been approached), it didn’t take scholars long to discover that thinking is a widely spread characteristic of many forms of life. In addition, in many animals there is some lexical ability, a capacity for simple, meaningful structure, elements of culture, and the ability to imitate and learn. In animals closely related to us, the rudimentary beginnings of vocal control are evident. Although language evolution is a relatively new field, it has brought together this research from many disciplines in a completely new way.

Part of the field’s struggle is that the very language used to get at these ideas does not serve it well. Language evolution research has illuminated a complicated geometry of species, traits, and relationships, and in the face of this newly defined space words like “uniqueness,” “innateness,” and “instinct” have come to mean everything and nothing. Those terms are still bandied about to explain the disagreements between people working on language evolution, but in fact everyone agrees there is linguistic innateness, and everyone agrees there is
something
unique about language.

Language has to be partly innate, simply because human babies are born with the ability to learn the language of their parents. While this can justifiably be called a language instinct, there is no one gene compelling us to produce language. Instead, a set of genetic settings gives rise to a set of behaviors and perceptual and cognitive biases, some of which may be more general and others of which are more language-specific.

Language is unique in that there are no other animals with which we converse, no matter what language we are speaking. And yet the miracle of this research has been the realization that what is unique from one perspective may be constructed of mostly old parts from another.

All the work in genetics, neuroscience, ethology, biology, and linguistics has emphasized both the undeniable separateness and the powerful continuity of language. We are not the only animals that live within a world of meaning. And yet no other animal mimics in quite the way we do, no animal gestures like we do, no other animal is able to produce such an ordered flurry of distinct and meaningful bites of sound, and certainly no other animal puts all of this together and communicates it in the same way we do.

In their completely different approaches, by building on the work begun by Noam Chomsky, Sue Savage-Rumbaugh, Steven Pinker and Paul Bloom, and Philip Lieberman, most researchers described in part 2 have emphasized the same important fact of evolution—having evolved means that you are less a creation than an accretion. You are a piled-up assemblage of systems and organs (some of which work better than others), and because of this, focusing only on sameness or only on difference doesn’t take us very far.

Like biology, language is constituted of an aggregate of different traits and processes that have developed over time. There was no one moment at which humans became definably human, just as language did not appear suddenly from the ether. As important as the shared traits that we use as the basis of language are, so too are the parts that are different. In the end, you have to be human to have human language.

Investigating the language suite helps us identify the way these traits and behaviors are wondrously assembled by evolution into an ability to learn language. What this research does not explain, though, is how language itself came to exist. Indeed, humans won’t speak or produce language unless they are taught to do so, which means that our remarkable capacity doesn’t amount to much at all if someone isn’t there to provide a model for how to use it. In order to understand this conundrum, you have to look at how the old parts, the shared parts, and the new parts have wound together in humans to produce this novel ability to learn language.

W
hen you consider language for any length of time, you come to realize that for its users, language is the operating system of the world.
¹
And after barely two decades of research, it is now undeniable that many of the traits implicated in the learning and use of human language are much older than humanity itself. So how did these particular abilities coalesce to produce language as we have it today?

Evolution is a slow and dirty process, and it’s difficult to see how something so complex can arise from something so unpremeditated. But it is only because of the opportunistic zigs and zags that biology took through time that we now have words and rules and their infinite permutations, from tedious political speeches to information-packed instruction manuals, from the irresistible oomph of “WAR! (What is it good for?)” to the dirty word someone once whispered in your ear.

The mechanics of evolution mean that humans became the linguistic species through a purposeful but not perfect process. The purpose was not to create modern language per se, but to provide an advantage in staying alive. Nature selected your father and mother and their parents for survival. It selected their parents before them, their ape parents before them, and their lizard parents before them. The long line of specific individuals that precedes you was not fated at birth to survive and pass on a selection of its genes, but somehow it managed to do so. And here you are—the language-rich result of the haphazard, mostly wordless path they fashioned.

When did language begin? Its foundations can be traced back to our common ancestor with primordial lizards. Yet what we recognize as language today took shape sometime in the last six million years. It’s clear just from the distribution of elements of the language suite over different animal species and throughout the lineage of the human species that language did not evolve overnight, turning us from
animals
into
people.
The mere fact that different traits are shared with different animals suggests that language came together in bits and pieces, step by step.

Just by looking at what is shared and what isn’t, you can start to glean the outline of a trajectory through time from less linguistic to more linguistic. How old is gesture? At least as old as our shared ancestor with other great apes. How old is simple syntax? Perhaps as old as our common ancestor with monkeys, which lived forty-five million years ago. In addition to the fact that traits have changed, it is useful to examine the different ways they have changed over time. When you look at how these pieces evolved, you can start to narrow down why they might have evolved.

T
he beauty of comparing the minds and behaviors of humans with those of other animals is that it illuminates a past so distant that it is almost unimaginable. For a trait like language, which leaves behind no fossils, this method serves us particularly well. Even though our common ancestor with chimpanzees lived as much as six million years ago, the extensive research that has been conducted on chimpanzees today enables us to make useful inferences about traits this creature may have had. Chimpanzees and bonobos don’t seem to have changed a lot in the last six million years (certainly not as much as we have), so when Kanzi demonstrates the ability to produce or comprehend language at the level of a young human child, it suggests that humans and their closest relatives have been bequeathed a common set of skills that could be used to produce something like language as we know it.

The same is true of monkeys; our common ancestor with the putty-nosed monkey (which utilizes a simple rule to create new meaning out of two separate sounds) lived about thirty million years ago. Indeed, we can trace a common heritage with all of our monkey cousins—the baboons, with their one-sided social syntax, and the Diana monkeys, which make such savvy use of Campbell’s monkey calls. It makes sense to assume that even those many million years ago, some animal evolved the trick of combining sounds—sounds it heard or maybe sounds it produced—to create meaning. It may be that our common ancestor with monkeys had only some very limited form of this ability. Regardless, the many language-related abilities that monkeys possess suggest that language as a whole didn’t simply spring intact from the head of
Homo.
Its foundations were around long before we were. Alex the parrot, Lou Herman’s dolphins, humpback whales—all these creatures help us retrace the long and winding trail from wordlessness to linguistic meaning.

Yet even though the common mental platform we share with other animals turns out to be both deep and wide, a lot has happened since our ancestors split from the ancestors of modern-day chimpanzees and bonobos. Humans have much greater control over the muscles of the face and mouth, the brain has developed along an unusual trajectory, we use a special mental device called a word, and we are exceptional at tracking structure through time. Many of these accomplishments and their consequences—thinking, speech and complicated syntax, rhyming couplets, corny jokes, self-help manuals—were significantly refined in the last six million years.

Unfortunately, there are no animals alive today with which we have a common ancestor from this time range, which means there are no living creatures more closely related to us than chimpanzees with which to make comparisons. We did once, however, have many closer cousins. The creature that was first cousin to the grandparents of all modern-day chimpanzee and bonobos, the creature from which we eventually descended, spawned a number of different species. Unfortunately, all of these branches of the family have died—some relatively recently.

The only evidence we have for the existence of these closer cousins and their relationship with us comes from the fossil record. For the most part, we have identified these relatives by comparing their skeletal features with ours. We imagine who they were and what their lives were like by measuring traits like the volume of the cranium. Cranial size is a clue to brain size, and given our knowledge about how brains work, brain size means we can make some guesses about neural organization. Changes in leg and pelvic bones, as well as the curvature and orientation of the spine with respect to the skull, indicate whether the bone’s owners were partly or fully bipedal. We can develop the picture further by examining the artifacts that accompany their remains, if any. We have techniques for modeling the weather of prehistory, so we can establish whether our ancestors lived in warmth or in cold. We also take into account the animal bones and fossils found from the same time period. Because many of them bear telltale signs of having been hunted and consumed, we know a lot about what our various grandparents ate (or what ate them). It is also possible to detect traces of hearths long burned, which tells us who was using fire on a regular basis.

Most of our understanding of human genetic history has come from comparing the DNA of humans across the globe and tracing it back to common ancestors. Occasionally, we can analyze DNA from nonfossilized bony remains. In at least one case, this has told us whether a humanlike fossil came from a direct ancestor. Altogether, this evidence points to many different human relatives—our common ancestors with these creatures lived five, four, two, and even just a half million years ago. As recently as twenty-eight thousand years ago, there lived creatures that were much closer kin to us than chimpanzees are—so close that if you were standing near one of them on a New York City subway platform you might not look twice.

Some obvious general patterns can be discerned in the family tree over the last half-dozen million years. As you’d expect, the further back you go and the closer you get to our common ancestor with chimps, the more apelike our own human ancestors and their cousins are. One of the best candidates for the most distant human ancestor that is not shared with chimpanzees and bonobos is called
Sahelanthropus tchadensis,
or Toumai. Toumai had widely spaced eyes and a small, chimpanzee-sized brain; its face was flatter than a chimpanzee’s and pushed outward more than a human’s. Toumai’s canine teeth were smaller and more humanlike as well. Toumai had a huge browridge, another primitive hominid characteristic, and it lived in the forest on the edge of Lake Chad. So far, only pieces of Toumai’s skull have been found, so it’s not clear whether it was bipedal or not.

Another distant ancestor (whether direct or more like a great-aunt we don’t yet know) is a six-million-year-old Kenyan species known as
Orrorin tugenensis,
thought to have walked on two legs for at least some of the time. The remains of
Orrorin tugenensis,
like those of most of our more recent ancestors, are fragmentary and far-flung. Only twenty-two traces of the
O. tugenensis
family have ever been found, mostly teeth and pieces of limb bone. Despite how different they seem to us today, these animals were closer to us than modern chimpanzees are.

The further along the branches you go and the nearer you get to modern times, the more recognizable the members of the family tree become. Around four million years ago, our ancestors left the forests and moved out to the savanna, where they remained for a very long time. This emigration marks the birth of the fully walking ape. All other apes, even if they can walk bipedally for short periods, primarily move about on four limbs. In 1978, Mary Leakey, of the famous family of paleoanthropologists, stumbled across a line of footprints in Tanzania that date to 3.6 million years ago. Bipedal apes are called hominids, and it’s possible to see from the trail that they left that three hominids once strolled together across wet volcanic ash.

This period is characterized by an extensively populated branch of the australopithecines, including
Australopithecus anamensis, Australopithecus africanus, Australopithecus boisei,
and
Australopithecus afarensis
—the famous Lucy, whose close kin lived from around 3.6 million to 2.9 million years ago. Many of these species dwelled side by side in Africa, where they all remained. Lucy’s skull was more chimpanzee-like than human, but her canine teeth, though more pointed than a human’s, were much smaller than those of other apes. In size, shape, and relative proportions, her leg and pelvic bones were clearly more human, and she was bipedal. Another important
Australopithecus afarensis
discovery was announced in 2006. In Ethiopia, not too far from where Lucy was found, the remains of a small child, dubbed Dikika (“nipple” in the local language, after a nearby hill), were discovered. Scientists pieced together Dikika’s face (with a full set of milk teeth, as well as unerupted adult teeth), a hyoid bone, complete rib cage, some fingers, and parts of her legs, including knees and a foot. She was three when she died 3.3 million years ago—the most complete skeleton of her species ever found. Researchers think that Lucy’s and Dikika’s family are very distant, direct ancestors of modern humans. (The other australopithecines are cousins to
A. afarensis,
parallel lineages that eventually died out.) Not long after the time that Lucy and Dikika walked the earth, our ancestors and their cousins began to develop tools; the first in the australopithecine-hominid lineage that we know of were simple stone flakes.

Between 2.5 and 1.8 million years ago the first species that we would call
Homo
are detected. Around this time, there were at least four branches of the family—
Homo habilis, Homo rudolfensis, Paranthropus boisei,
and
Homo ergaster. Homo ergaster
is our grandparent; the others are our aunts and uncles. (While
Homo habilis
was not a human ancestor, many
habilis
remains have been found with stone tools. Tool use is therefore a more general family trait than a feature of the specific lineage that produced modern humans.) Some hominids had a brain almost twice the size of the australopithecines’.
H. ergaster
is the first creature in the line from the chimpanzee-human ancestor to have a basically modern human body form—tall and upright.
H. ergaster
was also the first human ancestor to start traveling beyond Africa; its fossilized remains have been found in China and Java, where it lived about 1.8 million years ago.

Turkana Boy is the best-known example of a
Homo ergaster
specimen that looks like a human as we know it. Almost his entire skeleton has been recovered. He lived about 1.6 million years ago, and he was about twelve years old when he died. His brain size was double that of chimpanzees, although it was still significantly smaller than ours. Technological innovation had pretty much remained the same since the flake appeared some 2.5 million years ago, but approximately 1.5 million years ago, not too long after Turkana Boy’s death, the hand ax was invented, and it appears from the archaeological record that it was the dominant tool for about a million years.

H. ergaster
gave rise to
Homo heidelbergensis,
who invented the prepared-core tool, in which a stone was shaped and then struck once to produce a finished tool. The significance of the prepared-core technique is that it involves using a mental template with which to create many copies of the same tool. Both the Neanderthals and humans descended from
H. heidelbergensis,
though for a long time it was thought that humans descended from Neanderthals. In 1997, however, a team of geneticists announced that they had sequenced the mitochondrial DNA of a Neanderthal bone (mtDNA is passed from mother to child and is used to track female ancestors) and found that, for at least this part of the genome, there was so much difference between us and the Neanderthals that they could not be our direct forebears.

In fact, Neanderthals came from the branch of the family that left Africa long before our more immediate ancestors did. They lived across Europe and in western Asia for at least 200,000 years. They were excellent stone workers and survived well in cold, harsh climates. The Neanderthals buried their dead, at least some of the time. They created stone-tipped spears and hunted large game, killing animals as big as rhinoceroses. For most of their time on earth, Neanderthal culture was fairly static; the same tools were used for thousands and thousands of years. (Imagine if clay tablet and stylus were the only ways to write that we’d invented for 2,000 years.) Scattered evidence suggests that toward the end of their time, after living a few hundred thousand years in the same way, they began to use fire, they possibly made flutelike musical instruments, and they even fashioned ornaments like pendants from the teeth of bears, wolves, and deer.

Our direct ancestors left Africa around sixty thousand years ago and, after taking thousands of years to reach Europe, coexisted with the Neanderthals there, until the latter died out twenty-eight thousand years ago. The Neanderthal extinction is generally attributed to either too much competition (clever, aggressive
Homo sapiens
outmaneuvered them) or too much loving (we interbred with them and eventually swamped their genome with our far larger population).
¹
It’s also been suggested that in the same way that Western Europeans caused the decimation of indigenous populations when they made first contact in the last few centuries,
Homo sapiens
may have brought new diseases into the Neanderthal world and thus contributed to their decline. Most recently, Paul Mellars, a University of Cambridge archaeologist, and colleagues found evidence that a sudden climate shift, when the temperature dropped as much as 8°C, precipitated the demise of the species.

As the perceived mental gap between humans and animals has narrowed, so it has for modern
Homo sapiens
and their recent ancestors. For a long time Neanderthals were considered brutish, unintelligent creatures with no language or symbol use. But as the anthropological evidence has accumulated, they have undergone something of an image upgrade. It’s become clear in recent years that even though Neanderthals didn’t have as rich a culture as we did, what they eventually developed before they went extinct was fairly sophisticated. It’s worth keeping in mind that although the Neanderthals disappeared soon after humans arrived, our species has yet to prove that it has even half the longevity outside of Africa that they did.

Another cousin, thought to have descended from
Homo ergaster,
was discovered only in 2003. The scientific world was shocked by the news that on the island of Flores in Indonesia a team of scientists had unearthed the remains of a creature they called
Homo floresiensis.
Prior to this no one had suspected that humans had once had relatives as closely related as Neanderthals. Nicknamed the hobbit,
H. floresiensis
is an interesting contrast to the Neanderthal. While the northern branch of the family was large, thickset, and reputed to have had larger brains than ours, the hobbits were mini-hominids, reaching only a meter’s height at adulthood. The remains of seven different
H. floresiensis
individuals were found; they died between 95,000 and 13,000 years ago. Flint blades were discovered alongside the remains, indicating that the hobbits were also tool users. Scientists say the
H. floresiensis
tools are stylistically similar to a cache of 800,000-year-old
Homo ergaster
tools that had been found nearby, suggesting they inherited the technique from their
H. ergaster
ancestors.
²
The last of the hobbits disappeared at the same time that a nearby volcano erupted, so it’s believed that this catastrophe led to the species’ extinction. The individual that died 13,000 years ago was our last, closest cousin.
³