Most people have felt the need to warn companions at a party that the dreaded character X has just entered the room and is heading their way. In such a situation, you don’t simply point or shout, even though those are natural inclinations. No, you raise your eyebrows at your companion, jerk your head a bit in the direction of the approaching X, and maybe purse your lips to warn your companion to be quiet. Clearly, humans point in multiple ways.
A chimpanzee named Nikkie once communicated with me through the same subtle technique. Nikkie had gotten used to my throwing wild berries to him across the moat at the zoo where I worked. One day, while I was recording data about the apes, I totally forgot about the berries, which hung on a row of tall bushes behind me. Nikkie hadn’t forgotten. He sat down right in front of me, locked his red-brown eyes into mine, and -- once he had my attention -- abruptly jerked his head and eyes away from mine to fixate with equal intensity on a point over my left shoulder. He then looked back at me and repeated the move. I may be dense compared with a chimpanzee, but the second time I turned around to see what he was looking at, and spotted the berries. Nikkie had indicated what he wanted without a single sound or hand gesture.
That simple act of communication went against an entire body of literature that links pointing to language, and that therefore has little room for nonlinguistic creatures like apes.
Pointing -- making deictic gestures, in more scholarly language -- is defined as drawing another party’s attention to an out-of-reach object by locating the object in space for the other party. There is no point to pointing unless you understand that the other hasn’t seen what you have seen, which involves realizing that not everyone has the same information.
Some scientists and philosophers believe that humans alone can grasp that fact. They consider other animals prisoners of their own experiences -- unable to see beyond what they themselves know, and thus without the inclination or the capacity to intentionally modify what others know. The word “intentionally” is critical. Even though a bird’s alarm call informs other birds, common wisdom holds that the caller is simply reacting automatically to some frightening stimulus. Similarly, the honeybee’s dance, which directs other bees to far-away nectar sources, is not considered intentional.
Thus, defenders of human uniqueness have surrounded pointing with heavy theoretical artillery, designed to keep other creatures at bay. Some investigators, like George Butterworth and Lesley Grover of the University of Sussex, have focused on the typical human gesture of pointing with outstretched arm and index finger. That gesture has been linked to symbolic communication, which calls up a picture of early humans walking around on the savanna pointing and assigning words to objects: “Let’s call the animal over there a zebra, and call this your belly.” Yet doesn’t such a scenario imply that our ancestors understood pointing prior to the evolution of language? If so, the idea that our nonlinguistic primate relatives point should not upset anyone: It is to be expected.
I knew it was time to explore the subject of pointing further when the December 1999 issue of the Journal of Comparative Psychology featured three separate papers on it: one on whether dolphins understand human pointing, one on the use of manual pointing by captive chimpanzees, and one on a chimpanzee’s ability to spontaneously direct people to hidden objects that the ape couldn’t reach. The new data on pointing force us to rethink the idea that pointing is inextricably linked to language.
The first step is to move away from silly anthropocentric definitions, like the one requiring an outstretched index finger. Some animals don’t have arms and hands, let alone fingers, but that is no reason to declare pointing beyond their abilities. For example, bottle-nosed dolphins can develop a perfect understanding of human pointing, as Louis M. Herman and his colleagues at the University of Hawaii-Manoa have demonstrated. In one series of experiments, a person stood next to a dolphin’s tank in which three objects floated at different angles and distances from the dolphin. The dolphin lifted her head out of the water, watched the person point at one of the objects, and -- with very few errors -- swam to retrieve that object. Initially, the dolphin had a problem with items that were so far away they were out of sight, but after the human used exaggerated motions for such items, the dolphin retrieved them correctly as well.
Admittedly, the experiment demonstrates only that dolphins comprehend pointing, not that they do it themselves. But Herman speculates that their abilities may relate to their natural tendency to gather information about the environment from “listening in” on one another’s sonar signals. That may also permit them to draw others’ attention to objects. If animals can indeed extract information from the echolocation others engage in, the door is open to forms of pointing well outside the human experience.
We should take a broad view, therefore, of what constitutes pointing. Monkeys, for example, often point with their whole bodies and heads when they recruit allies during fights. If monkey B threatens monkey A, A may walk over to his usual protector, C. Sitting next to C, he then looks at him, jerks his head with grunts and threats toward B, and repeats that back-and-forth movement many times, as if telling C: “Look at that guy -- he’s bothering me!” Among macaques, an aggressor points with lifted chin and staring eyes at the opponent, in between conspicuous glances at the ally. Among baboons, the same behavior is so repetitive and exaggerated that fieldworkers call it “head flagging.” In all of those cases, the goal is to make absolutely clear to the ally where the adversary is.
In a classic study at the Delta Primate Center (now the Tulane Primate Center), in Louisiana, the primatologist Emil Menzel found that chimpanzees used their whole bodies to point. Menzel would take a juvenile chimpanzee with him into a large outdoor enclosure, and hide food or a frightening object, like a toy snake, in the grass. When he let other chimpanzees join the juvenile in the enclosure, the newcomers quickly grasped whether the hidden object was attractive or frightening, and its approximate location, by watching the first chimpanzee’s body language. Menzel commented: “A quadrupedal chimpanzee orienting in a given direction is, if anything, a more accurate pointer, especially for an observer in a tree or on a laboratory tower, than a bipedal human being pointing in the same direction manually. One good reason that chimpanzees very seldom point manually is that they do not have to [emphasis in the original].”
A widely used criterion for intentional pointing is that the pointer checks the results of his actions. The pointer should make sure, by looking back and forth between the object he is pointing at and his partner, that the partner is paying attention and the pointer is not pointing for nothing. A number of recent experiments have applied that criterion to great apes. The researchers have looked at manual pointing -- not because that is the most natural way for apes to point, but because captive apes readily learn that the gesture is most successful in producing a response from humans.
Here at the Living Links Center, part of the Yerkes Primate Center at Emory University, William Hopkins and his students have worked with chimpanzees in an environment in which many people walk by the apes every day. It is logical for the apes to learn how to draw a human’s attention to things they want, like a piece of fruit that has dropped out of their cage. David Leavens, one of Hopkins’s students, has investigated whether apes spontaneously attract people’s attention to out-of-reach food.
It turns out that 68 percent of more than 100 chimpanzees in Leavens’s study gestured to the experimenter. A few did so by stretching out an open palm, which is an ape’s way of begging. Most, however, used the whole hand to point at a banana outside their cage, although no one had ever explicitly trained them to point in that way. A few even pointed with an index finger. And there were clear signs that the apes monitored the effect of their gestures in exactly the same way that scholars have evaluated whether pointing in children is intentional. An ape would make eye contact with the human and then point while alternating its gaze between the food and the human. One chimpanzee, perhaps afraid to be misunderstood, pointed first with her hand at a banana and then with a finger at her mouth.
In a second, continuing study at Living Links, an experimenter holds a banana while either facing the ape or turning away. No self-respecting chimpanzee should point when the human isn’t looking, because those gestures would go unnoticed. On the other hand, producing noise -- like calling or banging on the bars of the cage -- makes sense because that draws the human’s attention. All indications so far are that that is exactly what the apes do: They gesticulate more when the experimenter can see them, whereas they vocalize when the experimenter turns away.
For anyone who still has doubts, the literature contains additional intriguing tidbits: Apes will point in the wrong direction when confronted with an unfriendly experimenter, who tends to appropriate food for himself, and apes trained in American Sign Language sign to themselves while pointing at pictures in a magazine. In the latter case, apes monitored by video were pointing in the absence of people; hence, they were not acting for the benefit of a human audience.
One criticism remains, however: The apes in the studies of pointing are familiar with human behavior. Would they ever have developed pointing in the absence of a species that readily uses and responds to pointing? Show us the cases where apes point things out to one another, the skeptics say. Unfortunately, they mean manual pointing, not Nikkie’s pointing with his eyes, or the whole-body point noted by Menzel -- methods that are far more common. But we have two reports of manual pointing.
One report derived from my own observations; it appeared in Chimpanzee Politics, a book I wrote more than two decades ago. I was discussing how a female chimpanzee may enlist male support:
“The threatened female challenges her opponent with a high-pitched, indignant bark, at the same time kissing and making a fuss of the male. Sometimes she points at her opponent. This is an unusual gesture. Chimpanzees do not point with a finger but with their whole hand. The few occasions on which I have seen them actually point have been when the situation was confused; for example, when the third party had been lying asleep or had not been involved in the conflict from the start. On such occasions the aggressor would indicate her opponent by pointing her out.”
The other account is the only one involving wild apes. It appeared in a brief article in a specialized journal about bonobos in dense forest in the Democratic Republic of the Congo (formerly Zaire), studied by a Spanish team led by Jordi Sabater-Pi, from the University of Barcelona. In this case, one ape alerted other bonobos to the presence of hidden scientists:
“February 24, 1989. 13:09 h. Noises are heard coming from the vegetation. A young male swings from a branch and leaps into a tree. ... He emits sharp calls, which are answered by other individuals who are not visible. He points -- with his right arm stretched out and his hand half closed except for his index and ring fingers -- to the position of the two groups of camouflaged observers who are in the undergrowth (30 m apart). At the same time he screams and turns his head to where the other members of the group are.
“13:12 h. The same individual repeats the pointing and calling sequence twice. Other neighboring members of the group approach. They look towards the observers. The young male joins them.”
In both examples, the context was entirely appropriate (the apes pointed at objects that others had not seen), and the behavior was accompanied by visual checking of its effects. Also, the pointing disappeared once the others had looked or walked in the indicated direction. There is every reason to believe, therefore, that apes spontaneously guide the attention of others while taking into account the others’ lack of knowledge.
Additional evidence comes from what may be the most telling study of apes’ referential signaling -- that is, signaling that explicitly refers to external objects or events -- conducted by Charles Menzel (the son of Emil) at the Language Research Center of Georgia State University. Menzel let an 11-year-old (hence barely adult) female chimpanzee named Panzee watch him from a distance while he hid an object in an outdoor area. Sometimes, Menzel had Panzee watch him hide something after all other people had left for the day. That meant that Panzee, who was housed alone, had no chance to communicate with anybody until more than 12 hours later. When the caretakers arrived the next morning, they were unaware of what Menzel had done. That was the beauty of the study: An uninformed person will not deliberately trigger communication. Panzee had to initiate the exchange, and provide information about the object to a person who didn’t know what she knew, and who at first had no clue what she was “talking” about.
Menzel notes that caretakers generally have a higher opinion of apes’ mental abilities than most of the philosophers and psychologists who have written on the subject -- few of whom have ever studied apes themselves. All the people recruited by Panzee to retrieve a hidden object said that they were at first surprised by how she acted, but immediately understood what she was trying to get them to do. Instructed by her pointing, beckoning, panting, and calling, they had no trouble locating the item. Panzee never pointed in the wrong direction, or to any of the locations used on previous trials, and she often touched her board containing lexigrams to identify the hidden object. The result was intentional communication about a past event, present in the ape’s memory, to people who knew nothing about it and thus were unable to give her any clues.
The new evidence on pointing shows that language is not synonymous with the capacity to share knowledge about events that are distant in time and space. Our species uses both language and pointing to share knowledge, but pointing is not unique to us. Under the right circumstances, other animals also use it to refer deliberately to the outside world. Like us, they understand that everybody may not know the same things, and that one individual can communicate knowledge to another.
Frans B.M. de Waal is a professor of psychology and the director of the Living Links Center at Emory University. His latest book, The Ape and the Sushi Master: Cultural Reflections by a Primatologist, will be published this spring by Basic Books. He writes regularly for The Chronicle.
http://chronicle.com Section: The Chronicle Review Page: B7