The First Word: The Search for the Origins of Language

by Christine Kenneally

  Intention movements are the beginnings of an actual movement, like a raised fist to indicate a threat in humans.

  The process by which these gestures evolve in individuals, Tomasello explained, goes like this: “I’m really doing something, you come to anticipate it, I notice your anticipation so I only make the beginnings of the movement.” Male chimpanzees, for example, make a penis-offer gesture to propose sex. They sit back on their haunches and repeatedly thrust their pelvis, pushing their erect penis in the direction of another chimpanzee. “In papers we call it the penis offer,” Tomasello said. “Between ourselves, it’s called ‘dirty dancing.’”

  Mimicking another intention movement, Tomasello rolled his arms over his head, like a chimp barrel-hitting a companion. The move is reminiscent of the way that humans feint at each other to make a point without actually following through. Cats and dogs make a similar movement when they raise their paws and bat them, as if they are about to strike another animal, so the gesture is not restricted to primates. “It’s typical mammalian play,” Tomasello explained. “Remember,” he said, invoking the tree of life, “it’s not a ladder; it’s a tree. It’s not a ladder; it’s a tree.”

  Another gesture researcher, Joanna Blake at York University in Canada, directly compared the gestures that infants make when they are learning language with the gestures made by apes, which have a lot in common. Both apes and children make a lot of request gestures—begging for food, raising their arms to be picked up and carried—and they extend their whole hand to point. Children and apes likewise make the same gestures of protest, pushing someone away or turning away themselves while shaking their heads. They also emote in the same ways, stamping their feet, flapping their arms, and rocking, and when they want someone to do something, both take a person’s or an ape’s hand and place it on the object to be manipulated, or they proffer objects that they want someone to manipulate. Clearly there is a close family relationship between human and ape gesture, confirming that it is an ancient trait that precedes the existence of modern humans and of language.

  Janette Wallis, who has been watching primates since she was an undergraduate at the University of Oklahoma, is drawn to the more subtle aspects of primate communication. She used hidden cameras to capture evidence of a baboon gesture she calls the muzzle wipe—a quick pass across the bridge of the nose with the hand. The muzzle wipe typically occurs in situations in which a baboon may be nervous or conflicted for some reason. As with many human gestures, there’s no evidence that the wipe is intentional, but it’s likely that other animals read it as a signal that reveals information about the wiper.1

  Wallis presented videos of the muzzle wipe at the Leipzig gesture conference. Although most early studies of baboons, she said, hardly mention the gesture, her films showed baboons doing it in captivity and in the wild. The gesture rarely lasts longer than a few seconds, so it is not easy to see, yet once Wallis told the audience what to look for, the muzzle wipe was clearly evident. Nervous baboons could be seen constantly putting their hands to their faces in difficult situations. She noted that monkeys make a similar move and that a chimpanzee will often put its wrist to its forehead in similar contexts. Could this overlooked gesture be some kind of precursor to comparable gestures in humans? asked Wallis. Humans do put their hand to their face when nervous, and indeed, as she pointed out, psychiatrists and law enforcement officials often interpret a hand-to-face gesture as evidence of uncertainty or even deception.

  Once Wallis convinced the audience that the muzzle wipe existed, she showed a video of George H.W. Bush. The ex-president was speaking at a press conference about his son the president of the United States. He discussed what was at the time headline news—George W. Bush’s having been arrested in his youth on a drunk-driving charge. “Unlike some,” said the older Bush in a tone of complete confidence, “he accepts responsibility.” He then raised his hand to the bridge of his nose and scratched it.2

  Only ten years ago researchers were unanimous in their agreement that pointing was unique to humans. Even now many stand by that claim. In fact, apes and many species of monkeys that are much more distantly related to humans do point as well, though they typically do so with their whole hand.3 (Scholars of gesture complain that pointing with the hand has been treated as a second-class kind of pointing, even though it is common in many human groups.) Usually, apes make this gesture only for humans, not between themselves. They point at objects and alternate their gaze between the object that is pointed at and the human they are pointing for. The animals learn how to point without explicit training, and simply pick it up from humans.

  Although there is only one anecdotal report of a bonobo’s pointing with its index finger in the wild, some apes have been shown to do so in captivity. William D. Hopkins, a researcher at the Yerkes National Primate Research Center at Emory University, and his colleague David Leavens, a professor of psychology at the University of Sussex, showed a videotape at the gesture conference of a chimpanzee pointing. In the video, Leavens is in a white lab coat and a surgical mask while a chimpanzee stands eating on the other side of a wall of wire mesh. When the ape drops some food through the mesh, it points its index finger through the wire to indicate the food and looks at Leavens, who picks it up and returns it. “I submit,” Leavens said, “that there is a well-trained primate in this video, but it is not the chimpanzee.”

  At the Leipzig conference Tomasello was skeptical that apes could point and, if they did, that it actually meant anything. But he began to wonder about it and later said, “Many of the aspects of language that make it such a uniquely powerful form of human cognition and communication are already present in the humble act of pointing.”

  Tomasello had already established in previous experiments that apes know what other apes are seeing, and it was clear that they gesture easily and creatively for one another. More recent experiments have shown that chimpanzees will cooperate with one another in situations where collective help is needed (in order to get food, for example), and in quite simple tasks they’ll also assist without the prospect of a reward—like picking up a dropped object and handing it to someone. While the Hopkins and Leavens video showed they are capable of pointing, why, Tomasello asked, do apes point only for humans and not one another? The answer he arrived at is both simple and far-reaching: it is because humans respond. Apes don’t point referentially for other apes, because they will be ignored.

  Human children learn to point at a very young age. Tomasello and his colleagues have videotaped many instances of children spontaneously pointing in a helpful manner. In one experimental setup, a very young child was placed on her mother’s lap. Mother and child sat across a desk from a woman stapling papers together. The woman left the room for a moment, and while she was away a man entered, took the stapler, and placed it on a cupboard behind the desk. When the woman returned she made a great show of looking for the stapler. The infant watched her for a while, and then, unprompted, pointed to where the stapler had been moved so the woman could find it. In other examples, a child and adult played together until for some reason (the ball dropped, the toy fell) the game stopped. Without prompting, the child looked at the adult and pointed to the problem, clearly requesting that the game begin again. In other cases, the child pointed at an object or proffered it merely to show it to the adult in order to elicit a reaction.

  Tomasello first started to consider how much this kind of shared, cooperative attention mattered at dinner in a restaurant one night. He was watching a mother and child play together. The mother blew a raspberry on the child’s arm, then the roles were reversed, and the baby followed suit. Why did it happen this way? wondered Tomasello. Why did the child reciprocate the gesture rather than simply imitating the action on himself?

  The answer, he believes, is that humans are particularly cooperative in the way they communicate.4 Reciprocation is fundamental to the interactions of our species. Offering is not instinctive for humans, but is taught by parents to childr
en, who learn it very easily. And crucially, we offer not only food and other objects but information and experiences as well. Children, says Tomasello, want you to look at what they are looking at and to emote in response. In many theories of evolution, human altruism is treated as an anomaly. But Tomasello thinks of it as an evolutionary strategy that has served us incredibly well.

  Chimps don’t spontaneously point in this fashion, and Tomasello believes it is due to a fundamental difference in the balance of cooperation and competition within the species. Chimpanzees lack the set of skills and motivations that underlie our pointing. Tomasello conducted an experiment with Brian Hare, then a doctoral student, in which two barrels were set up in a room. Food was placed in one, while the other was left empty. Hare stood on one side of the barrels as a chimpanzee entered the room. In one run-through, Hare pointed helpfully at the barrel with the food in it. But, said Tomasello, the chimpanzee would look at the finger, and then look at the barrel, and then look at the other barrel, and then it would choose completely randomly between them. It did not comprehend that Hare was being helpful and telling it where the food was located. In another run-through of the experiment, the chimpanzee would come into the room, and instead of pointing to the food, Hare would reach for the barrel, as if to grab it and the food in it. The chimpanzee understood this gesture without any problem, and it would head for the appropriate barrel. The movement Hare made was essentially the same in each case—a basic arm extension—but his intention was clearly cooperative in the first instance and competitive in the second.5

  Tomasello and his colleagues’ gesture work demonstrates both a continuum that connects human and ape communication and significant differences between them. In our evolutionary history some individuals must have been born with a greater inclination and ability to collaborate than our common ancestor with chimpanzees. These individuals were more successful and bred more offspring with those characteristics, Tomasello said. What we have evolved into now is a species for whom an experience means little if it’s not shared. Chimpanzees took a different path. In their communication, there is never just plain showing, where the goal is simply to share attention. While they do share and collaborate and understand different kinds of intentions, they don’t have communicative intentions. We do, said Tomasello, and it’s in this shared space that the symbolic communication of language lies.

  Tomasello’s conclusions resonate deeply with observations made by Sue Savage-Rumbaugh. Before Kanzi, Savage-Rumbaugh worked with two apes called Sherman and Austin. The apes had successfully acquired many signs and used them effectively. There didn’t seem to be anything odd about their language use until one day they were asked to talk to each other. What resulted was a sign-shouting match; neither ape was willing to listen. Language, wrote Savage-Rumbaugh, “coordinates behaviors between individuals by a complex process of exchanging behaviors that are punctuated by speech.”6

  At its most fundamental, language is an act of shared attention, and without the fundamentally human willingness to listen to what another person is saying, language would not work. Symbols like words, said Tomasello, are devices that coordinate attention, just as pointing does. They presuppose a general give-and-take that chimpanzees don’t seem to have. For this reason, Tomasello explained, “asking why only humans use language is like asking why only humans build skyscrapers, when the fact is that only humans, among primates, build freestanding shelters at all…At our current level of understanding, asking why apes do not have language may not be our most productive question. A much more productive question, and one that can currently lead us to much more interesting lines of empirical research, is asking why apes do not even point.”

  Whether you are human or another kind of ape, one of the ways that gesture becomes ritualized and communicative is in being passed on by learning. As humans, we observe a gesture, and then we reproduce it by imitation. Imitation is crucial to the learning process, and we are not the only imitators in the animal world. Lori Marino, one of the researchers who explored the ability of dolphins to recognize themselves in mirrors, said that “imitation is an everyday behavior with dolphins.” They are very good at shadowing, imitation in real time. “If you make certain hand gestures in front of the tank in a captive facility, they will be able to follow your hand, even when you’re moving your hand back and forth in different ways. They also seem able to pick up patterns very well and anticipate patterns, so if you set up a certain pattern going and then you stop, they seem to anticipate what the next step in the pattern is.”

  Frans de Waal speaks of the difficulties of measuring fleeting and ephemeral behaviors like imitation. “A lot of the cognition studies are on technical cognition, like: Can they count? How do they use tools? Do they understand gravity? Social intelligence is more difficult,” he said.

  Particular difficulties arise with imitation studies, as de Waal explained:

  What people do, for example, in these imitation studies is they put an experiment in front of the chimpanzee and they show how to do something, and then they see the chimp imitate. But I think imitation also requires that you identify with the person and that you like the person actually. If you look at humans who imitate, children who imitate, they imitate the people they know and they like, and they want to be like Mom or they want to be like Dad or their big brother or whatever. They’re not imitating a random person. It’s very selective. I think the scientists who have failed to come up with these social learning tasks on chimpanzees, to some degree, have worked with the wrong paradigm. They put a human in front of the animal, which is already a different species, and the human may not have much of a relationship with them. I think we can only resolve these issues by focusing on behavior among animals themselves.

  De Waal has been studying the ways that capuchins imitate one another. The experimenters train one capuchin to perform a task, and while other monkeys watch it, they attempt to determine if any imitation is taking place. De Waal is also probing the relevance of who gets imitated—if a capuchin is more likely to imitate its mother, for example, than an unrelated male.

  Sue Savage-Rumbaugh’s experiences with Kanzi back up de Waal’s observation about laboratory experiments. She noted that Kanzi’s mother, Matata, had two other children who never got the amount of attention from human caretakers that Kanzi did. She believes it was the significant relationships with humans in the period in which Kanzi was most sensitive to acquiring language that enabled him to pick it up.7

  Other research suggests that imitation can be affected by who the original performer is. One recent study described the way a population of dolphins off the coast of western Australia passed on a tradition of tool use. These dolphins learned from adults in the pod to use sponges to forage on the ocean floor. But they didn’t just acquire the skill from any of the adults: the tradition seemed to be passed down solely from mother to daughter.

  The combination of gestural communication and imitation can be as powerful as vocal communication. In human hunter-gatherer groups, such as the Ngatatjara of western Australia and the northern Déné of the Canadian subarctic, the transmission of knowledge about the environment and how to survive in it is achieved by observation and experimentation rather than by verbal explanation. Moreover, studies have shown that a group learning how to manufacture a stone flake (such as those used by Stone Age societies) from a teacher who only gestured took no longer at the task than, and were as good at it as, a group in which the teacher gave precise verbal instructions on how to make the flake.8

  In modern humans gestures come in a variety of types. There is here-and-now pointing (this book, right here!), action gestures (she picked it up with one hand!), abstract pointing (and another thing!), and metaphorical gestures that make symbolic reference to people, events, space, motion, action. Most gestures are initiated with the right hand. They typically occur slightly before or at the same time as speech.

  Gestures that accompany speech typically amplify the meaning conveyed by the sp
eaker. Sometimes, gesture communicates information that isn’t explicitly stated in the verbal message it accompanies. For example, a speaker may move his fingers stepwise in a spiral while saying, “I ran all the way upstairs.” The listener can infer that the staircase was spiral even though the fact was not stated.9 While gesture doesn’t break up into wordlike segments, there are rules about the way gestures can be combined. And as obvious as the meaning of many gestures is when they are used by people while they are talking, listeners can usually guess at the meaning of a gesture without sound only 50 to 60 percent of the time. (Think about gesturing while saying, “I had a big ball” and “The guy had a huge hot dog.”)

  For a long time gesture was more or less ignored in linguistics, and elsewhere it received little attention. Researchers considered it paralinguistic, meaning that it was merely supplementary to language, perhaps useful in terms of emphasis but ultimately a secondary and unimportant phenomenon. People assumed that gesture was only for the benefit of the listener and justified removing it from serious consideration for the simple reason that it could be removed. It is possible, after all, to hear and understand someone even if you don’t look at him. (In the same way, structure in language has been treated as separate from meaning, because you can go a long way analyzing both of them without reference to the other. Similarly, intonation has been largely ignored within Chomskyan linguistics.) The assumption was that because you could separate them in analysis, they worked independently in the body and they therefore evolved independently of each other. But even though you can discover much about speech and language without worrying about gesture, the fact is they usually occur together in the real world. Speech is disembodied only on the phone or radio, and in evolutionary time these types of communication have not been around very long.