Conversations With a Gorilla

Editor's note: Koko the gorilla, an ambassador for her species famous for learning sign language, has died at the age of 46. To honor Koko's memory, National Geographic is republishing "Conversations With a Gorilla," our October 1978 cover story written by Francine Patterson, the psychologist who taught Koko how to sign.

Current research paints a more complicated picture of primate sign language than was understood in the 1970s. We are presenting this article as originally published; the science within may not be up-to-date.

Koko is a 7-year-old “talking” gorilla. She is the focus of my career as a developmental psychologist, and also has become a dear friend.

Through mastery of sign language—the familiar hand speech of the deaf—Koko has made us, her human companions, aware not only that her breed is bright, but also that it shares sensitivities commonly held to be the prerogative of people.

Take Koko's touching empathy toward fellow animals. Seeing a horse with a bit in its mouth, she signed, “Horse sad.” When asked why the horse was sad, she signed, “Teeth.” Shown a photo of the famous albino gorilla Snowflake struggling against having a bath, Koko, who also hates baths, signed “Me cry there,” while pointing at the picture. (Aided by National Geographic Society research grants, studies of the first captive white gorilla, Snowflake, were described by Tulane University primate specialist Dr. Arthur J. Riopelle in the March 1967 and October 1970 issues of National Geographic.)

But Koko responds to more complicated motivations too. She loves an argument—and is not averse to trading insults.

At six o'clock on a spring evening last year, I went to the trailer where Koko lives to put her to bed. I was greeted by Cathy Ransom, one of my assistants, who told me that she and Koko had been arguing.

Lest I be alarmed at the thought of an altercation between this slight young woman, who is deaf, and a robust 6-year-old female gorilla, Cathy laughingly pointed to the notebook in which Koko's utterances in sign language are logged. The dispute began when Koko was shown a poster of herself that had been used during a fund-raising benefit. Manipulating hands and fingers, Cathy had asked Koko, “What's this?”

“Gorilla,” signed Koko.

“Who gorilla?” asked Cathy.

“Bird,” responded a bratty Koko, and things went downhill from there.

“You bird?” asked Cathy.

“You,” countered Koko.

“Not me, you are bird,” rejoined Cathy, mindful that “bird” can be an insult in Koko's lexicon.

“Me gorilla,” asserted Koko.

“Who bird?” asked Cathy.

“You nut,” replied Koko, resorting to another of her insults. (For Koko, “bird” and “nut” switch from descriptive to perjorative terms by changing the position in which the sign is made.)

“Why me nut?” asked Cathy.

“Nut, nut,” signed Koko.

“You nut, not me,” Cathy replied.

Finally Koko gave up. Plaintively she signed, “Damn me good,” and walked away signing, “Bad.”

“When She Is Good....”

I fully agree with Koko, if she meant that she is good even in a bad situation. I've come to cherish her lies, relish her arguments, and look forward to her insults. While these behaviors demonstrate occasional lapses from sweetness, they also provide reassuring benchmarks in the formal and controlled scientific testing that has monitored Koko's progress since I began to teach her American Sign Language in July 1972.

Of course such subjective behavior as lying is difficult to prove empirically, but when Koko uses language to make a point, to joke, to express her displeasure, or to lie her way out of a jam, then she is exploiting language the way we do as human beings. Certainly that is linguistic, though perhaps not moral, progress.

What makes all this awesome—even for me, after six years of witnessing such incidents—is that Koko, by all accepted concepts of animal and human nature, should not be able to do any of this. Traditionally, such behavior has been considered uniquely human; yet here is a language-using gorilla. (Two years ago she was joined by another of her species, a young male named Michael, who is the subject of similar study and training.)

Enrolling at Stanford in 1970 as a graduate student, I chose nonhuman primates rather than children for my research. In 1971, R. Allen and Beatrice Gardner came to speak. They were by then well-known for their success—it was an area where others had failed—in two-way communication with Washoe, a female chimpanzee.

The Gardners' breakthrough was to perceive that the chimp's difficulty in acquiring language might not be stupidity, but rather an inability to control lips and tongue. So they decided to try to teach Washoe American Sign Language—Ameslan for short—used by an estimated 200,000 deaf Americans. The language consists of gestures, each of which signifies a word or idea.

Washoe endorsed the Gardners' choice by learning 34 signs during the first 22 months. This was more than eight times the number of spoken words that the chimpanzee Viki, the subject of Keith and Cathy Hayes's six-year effort, learned to utter. After four years of Project Washoe, by 1970, Washoe had acquired 132 signs, and she used these signs in combinations similar to those employed by children during the first stages of learning to talk. (Project Washoe was supported in part by a research grant from the National Geographic Society.)

Hearing the Gardners tell their tale persuaded me that attempting to teach a chimp sign language would be to pursue the ultimate question with the ultimate animal. At that time I held no brief for gorillas.

Scientist Meets “Fireworks-Child”

My initial preoccupation with chimps changed suddenly on the day that I accompanied Dr. Karl Pribram, at that time my research adviser, to the San Francisco Zoo to talk with Ronald Reuther, then the director, about using a computer to try to communicate with the zoo's adult gorillas. We walked over to the gorilla grotto.

While Dr. Pribram and the director chatted, my eyes were drawn to a tiny infant clinging tenaciously to her mother. The infant was named Hanabi-Ko, Japanese for Fireworks-Child (she was born on the Fourth of July), but she was nicknamed Koko. Brashly I asked the director if I might try to teach Koko sign language. He said no, and quite rightly, too. Koko was only 3 months old, and Mr. Reuther did not want to separate her from her mother. Undaunted, I began to learn Ameslan, confident that one day I would have the chance to use it.

Nine months later, on a visit to the zoo, I ran into Martin E. Dias, an ebullient and sympathetic keeper. I asked about Koko. It seemed that Koko's mother had not been producing sufficient milk. As if this were not enough, the gorilla group had been afflicted with an outbreak of dysentery.

Suffering from malnutrition and dehydration, Koko had lost most of her hair, and her tiny body, racked with diarrhea, had become emaciated.

But, moved to the Children's Zoo, Koko had recovered. Perhaps, Marty suggested, the director might now look favorably upon my request to work with and care for Koko. Mr. Reuther immediately acceded. I began to get to know Koko the next day. That was in July 1972.

Gorillas are tragically misunderstood animals. In fact exceedingly shy, placid, and unaggressive, they are conceived to be ferocious, slavering man-killers. In a recent poll of British schoolchildren, gorillas ranked with rats, snakes, and spiders among the most hated animals.

On our first meeting, Koko did nothing to advance the cause of gorilla public relations. Quickly sizing me up, the tiny 20-pound gorilla bit me on the leg. But I was undeterred. People often ask if I am worried about dealing with Koko when she reaches full growth, perhaps 250 pounds. The answer is no, though at 130 pounds she already outweighs me and is astonishingly strong. While many captive chimpanzees have become difficult to work with as they mature, gorillas seem to be of quite a different temperament.

Soon after starting work with Koko, I met Carroll Soo-Hoo, the man who had donated Koko's mother to the San Francisco Zoo. Mr. Soo-Hoo brought out photos of himself, a slight man, romping with three 200-pound gorillas. That quelled whatever doubts I may have had about the danger of working with these immensely strong animals.

Pupil Begins to Learn—Reluctantly

Koko at first seemed to prefer men to women. While often contentious with me, she was beautifully behaved with Ron Cohn, my close friend and the photographer who has documented Koko's history.

Most bite attempts resulted from the method I used to get Koko to make signs—the “molding” technique the Gardners used with Washoe. The experimenter takes the hands of the subject and shapes them into the proper configuration for the sign representing an activity or an object while in its presence. As the animal comes to associate the hand movement and its meaning, the teacher gradually loosens his or her hold on the ape's hands until the animal is making the sign by itself. At first, every time I would take Koko's hands to mold a sign, she would try to bite me.

Another early problem—before we left the zoo for more satisfactory quarters—was distraction. I found it hard to keep Koko's attention while visitors to the glass-enclosed nursery stared, knocked, and commented on the curious tableau we presented. I grew weary of smirking people (who thought we could not hear them) saying, “Which one is the gorilla?” So it was with great relief that I moved Koko into her own trailer.

Are Apes Capable of Language?

My colleagues were not very sanguine about teaching Koko sign language. Some questioned the gorilla's dexterity as compared with the chimpanzee's. Others were skeptical about the animal's intellect.

In 1959 Hilda Knobloch and Benjamin Pasamanick had reported: “There is little question that the chimpanzee is capable of conceptualization and abstraction that is beyond the abilities of the gorilla.”

My experience has been totally at odds with this assumption. While Koko certainly has been contrary at times, I believe that such brattiness may indicate intelligence rather than its absence.

In 1929 the great primatologists Robert and Ada Yerkes wrote: “It is entirely possible that the gorilla, while being distinctly inferior to the chimpanzee in ability to use and fashion implements and to operate mechanisms, is superior to it in certain other modes of behavioral adaptation and may indeed possess a higher order of intelligence than any other existing anthropoid ape.” Now, fifty years later, Koko is bolstering evidence of the gorilla's intellectual primacy.

Initially my work with Koko used many of the techniques of Project Washoe. Experts in the new field of language development in humans—part of the discipline called psycholinguistics—found little agreement about what exactly language was, or when a child could be said to have it. Linguists, however, were virtually unanimous that Washoe did not have language. But by the time I began to publish data on Koko, many early critics of the Gardners had either recanted or softened their criticisms, in part because of the mass of fresh evidence on the language capacities of apes.

At the same time as Project Washoe, Ann and David Premack established two-way communcation with Sarah, a female chimpanzee. Sarah spoke and was spoken to through plastic symbols. The Gardners and Premacks were followed by Duane Rumbaugh, who installed yet another female chimp, Lana, at a computer console at the Yerkes Regional Primate Research Center in Atlanta. Lana gradually learned to communicate by typing out statements on an arbitrarily encoded keyboard. The computer was programmed to reject grammatically improper sentences.

The weight of all these experiments helped erode the doubts that an ape could be capable of language. Certainly, the pioneering work of the Gardners, the Premacks, and Dr. Rumbaugh has richly benefited me: I have been able directly to employ methods they discovered by trial and error, and have not had to refight the battle of credibility.

Once I had established that Koko performed at least as well as Washoe—learning the signs for “drink” and “more” within the project's first few weeks—I could probe new areas of the gorilla's potential for language and thought.

Koko Becomes a Star

From the start I have daily recorded Koko's casual signing, conversations, and self-directed utterances. I have also recorded her signing on videotape and film. Grants from the National Geographic Society and other private foundations have enabled me to meet the heavy costs—especially for equipment—to keep Project Koko going.

Vocabulary development is one of the best indexes of human intelligence. Koko's vocabulary grew at a remarkable pace. Over the first year and a half, she acquired about one new sign every month. After 36 months of training, Koko was reliably using 184 signs—that is, she used each spontaneously at least once a day, 15 days out of a month. By age 4½ , she had 222 signs by the same criterion. By 6½ , she had used 645 different signs. This figure refers simply to the total numbers of signs she had ever emitted correctly, in my judgment, not signs qualified by frequency of use. Finally, I would estimate that Koko's current working vocabulary—signs she uses regularly and appropriately—stands at about 375.

Trailer Home Takes a Beating

Koko's mobile home, situated since 1974 on the Stanford University campus, came to us with normal accommodations—a kitchen, a living room, and a hallway leading to a small bedroom, bathroom, and master bedroom. Chain link panels now protect the living room windows and large sliding glass doors from Koko's enthusiastic pounding. The living room became Koko's nursery with the installation of her metal sleeping box, an exercise bar, and a trapeze. Familiar household items stock the trailer: toys, books, pots and pans, chairs, mirrors, a refrigerator, stove, sink, and bed.

After our second gorilla, the young male Michael, arrived in September 1976, we transformed the master bedroom into a second training playroom with dangling chain, swing, and bench. The bathroom became a separate kitchen for Michael.

Two solid-wood doors separate Michael's domain from Koko's. With these doors open, one large common play area formed for daily exercise sessions and visits.

Because of her sharp teeth and endless curiosity about how things are put together, Koko has never had a bed with a mattress. Instead, she makes a nest using towels with a variety of underpinnings. Currently she has settled on a comfortable (I've tried it!) nest of two plush rugs draped over a motorcycle tire.

Koko rises at 8 or 8:30 in the morning, when my assistant Ann Southcombe and I arrive—that is, if she hasn't been roused earlier by Michael's morning antics. Following a breakfast of cereal or raisin-thick rice bread with milk and fruit, Koko helps with the daily cleaning of her room. She also thoroughly enjoys going over Michael's room with a sponge. Unfortunately, Koko usually rips the sponge to shreds when supervision slackens.

Then, most morning, Koko sits before the electric-typewriter keyboard in the kitchen for a thirty-minute lesson in auditory English. (More about this later.) Wearying of this, Koko asks me, “Have Mike in.”

About an hour is taken with up with Koko and Michael's tickling, tumbling, wrestling, chasing, and playing games of hide-and-seek. I usually leave during a banana-and-milk snack; then my assistant gives Koko her regular sign-language instruction.

Koko has a light meal—an egg or meat, juice, and a vitamin tablet—at 1 p.m. and a sandwich (usually peanut butter and jam) at 2 or 2:30. Most days I return at 3 and either sample Koko's signing on videotape, invite Mike in for another play session, or take them out for a walk or a drive.

Dinner at 5 consists almost exclusively of fresh vegetables. Koko's top preferences are corn on the cob and tomatoes; her lowest, spinach and carrots. She also dabbles in gourmet vegetables, such as artichokes, asparagus, and eggplant. She absolutely abhors olives, mushrooms, and radishes. If Koko cleans her plate, she gets dessert—usually Jell-O, dried fruits, a cookie, or cheese and a cracker.

After dinner Koko may engage in private monologue as she relaxes with a book or magazine (fingering a picture, she signs, “There flower”), nests with her blankets (“That soft”), or plays with her dolls (“That ear, placing the doll's ear against her own). Some evenings she asks if she may visit Michael's quarters.

Following toothbrushing and application of baby oil, both gorillas settle down about 7 or 7:30 with a “night dish.” This is a small fruit treat designed to make bedtime a more pleasant experience, for most nights Koko cries when I leave her.

“Go There....Hurry Go Drink”

On weekends, and other times when campus streets are quiet, Koko, Michael, Ann, and I all pile into my care for a drive. So absorbed are the gorillas in the passing scene that they sit quietly in their seats. Other motorists rarely notice them. Koko will occasionally engage in a little backseat driving—signing, “Go there” (so that we will not turn back toward home) or “Hurry go drink” (indicating a vending machine).

The Djerassi ranch, a 1,300-acre spread in nearby hills, is a welcome retreat on summer weekends. Here we can picnic without leashes or the threat of traffic and spectators. It's a joy to see Koko and Michael comporting themselves as free gorillas, frolicking in the trees as much as they might in their natural habitat.

Visits to my home were a thrill for Koko but strained both my nerves and the household furnishings. She would dash from room to room, slamming doors behind her. The whole house shook. I found it hard to keep my composure watching a hundred-pound-plus primate scale the walls and dangle from the ceiling moldings. When her exuberant bouncing on the bed finally collapsed it for good, I declared the house off limits.

Students Do Well on IQ Tests

With Koko's physical well-being provided for, we have every opportunity to promote and observe her mental and social progress. From the start I monitored Koko's performance on human intelligence tests. In February 1975 Koko's intelligence quotient was 84 on the Stanford-Binet Intelligence Scale. Five months later, at the age of 4, her IQ rose to 95, only slightly below the average for a human child. By January 1976 the IQ was back to 85, which is not an uncommon fluctuation. Her scores on other tests confirmed the general range established by the Stanford-Binet scale.

Testing Koko's IQ has not been easy. There is, for instance, a cultural bias toward humans that shows up when tests are administered to a gorilla. One quiz asked the child, “Point to the two things that are good to eat.” The depicted objects were a block, an apple, a shoe, a flower, and an ice-cream sundae. Koko, reflecting her gorilla tastes, picked the apple and the flower. Another asked the child to pick where he would run to shelter from the rain. The choices were a hat, a spoon, a tree, and a house. Koko naturally chose the tree. Rules for scoring required that I record these responses as errors.

Koko has made numerous other “errors” that offer insight into the personality of an adolescent gorilla. One day my associate Barbara Hiller saw Koko signing, “That red,” as she built a nest out of a white towel. Barbara said, “You know better, Koko. What color is it?” Koko insisted that it was red—”red, Red, RED”—and finally held up a minute speck of red lint that had been clinging to the towel. Koko was grinning.

Another time, after persistent efforts on Barbara's part to get Koko to sign, “Drink,” our mischievous charge finally leaned back on the counter and executed a perfect drink sign—in her ear. Again she was grinning. Sometimes Koko will respond negatively, but without a grin—leading me to believe her intent is not to joke but to be disobedient.

She seems to relish the effects of her practical jokes, often responding exactly opposite to what I ask her to do. One day, during a videotaping session, I asked Koko to place a toy animal under a bag, and she responded by taking the toy and stretching to hold it up to the ceiling.

With Koko in a contrary mood I can almost program her actions. For example, Ron Cohn got her to stop breaking plastic spoons by signing, “Good break them,” whereupon Koko stopped bending them and started kissing them. On such occasions, Koko knows that she is misbehaving, and once when I became irritated with her negativity, she quite accurately described herself as a “stubborn devil.”

Koko's Pet Hate: Alligators

Koko reserves an equally expressive lexicon of insults—”rotten stink” and “dirty toilet,” in addition to “bird” and “nut”—for people besides herself who are unmannerly. She has referred to me as “Penny toilet dirty devil” during a fit of pique.

One sign in Koko's vocabulary comes close to having totemic associations for her. That is “alligator,” a sign made by snapping the two palms together in imitation of an alligator's closing jaws. Heres, as in other instances, Koko modulates signs. For a large alligator, she makes a huge motion with her fingers. Aklthough Koko has never seen a real alligator, she is absolutely petrified of toothy stuffed or rubber facsimiles.

In an intriguing exchange with Cindy Duggan, my assistant, Koko appears to have used the “alligator” sign as a threat. Cindy had prepared a snack for Koko, but was slow delivering it. Koko reacted by signing, “Alligator chase lip.” (“Lip” is Koko's idiosyncratic term for a girl or woman.) A puzzled Cindy asked, “Alligator?” “Alligator do that hurry,” replied Koko, indicating the plate of food. Cindy reflected, and then laughed—Koko seemed to be resorting to verbal scare tactics to expedite the treat.

I have exploited Koko's irrational fear of this reptile by placing toy alligators in parts of the trailer I don't want her to touch. Consequently, at first glance a visitor might suspect that it is the temple of some obscure religious cult.

Remembrance of Events and Emotions

A cardinal characteristic of human language is displacement, the ability to refer to events removed in time and place from the act of communication. To learn whether another animal has this ability, we try to find out if the animal uses its sign vocabulary merely to label the events of its world, or if it is framing propositions that re-create a particular event. Does the animal use its symbols to refer to events earlier or later in time?

Koko and I had a revealing conversation about a biting incident. My try at cross-examination—three days after the event—went much as follows:

Me: “What did you do to Penny?”

Koko: “Bite.” (Koko, at the time of the incident, called it a scratch.)

Me: “You admit?”
Koko: “Sorry bite scratch.”

(At this point I showed Koko the mark on my hand—it really did look like a scratch.)

Koko: “Wrong bite.”

Me: “Why bite?”

Koko: “Because mad.”

Me: “Why mad?”

Koko: “Don't know.”

The entire conversation concerns a past event and, equally significant, a past emotional state. It is not a discussion one would expect to have with an animal whose memories were dim, unsorted recollections of pain and pleasure. Of striking import to me was that Koko knew she could not remember or express whatever it was that had prompted the bite.

Koko Learns to Lie

Perhaps the most telling, yet elusive, evidence that a creature can displace events is lying. When someone tells a lie, he is using language to distort the listener's perception of reality. He is using symbols to describe something that never happened, or won't happen. Evidence I have been accumulating strongly suggests that Koko expresses a make-believe capacity similar to humans'.

At about the age of 5 Koko discovered the value of the lie to get herself out of a jam. After numerous repeat performances I'm convinced that Koko really is lying in these circumstances and not merely making mistakes. One of her first lies also involved the reconstruction of an earlier happening. My assistant Kate Mann was with Koko, then tipping the scales at 90 pounds, when the gorilla plumped down on the kitchen sink in the trailer and it separated from its frame and dropped out of alignment. Later, when I asked Koko if she broke the sink, she signed, “Kate there bad,” pointing to the sink. Koko wouldn't know, of course, that I would never accept the idea that Kate would go around breaking sinks.

Some of Koko's lies are startlingly ingenious. Once, while I was busy writing, she snatched up a red crayon and began chewing on it. A moment later I noticed and said, “You're not eating that crayon are you?” Koko signed, “Lip,” and began moving the crayon first across her upper, then her lower lip as if applying lipstick.

A Sense of Past and Future

Gradually Koko is acquiring signs that make reference to past and future. One day during a filming session she signed, “First pour that,” as I was preparing milk for her. “First that yes!” I exclaimed, delighted that she had used the sign “first.” Just as I began to sign, “Then you drink,” Koko signed, “Later Koko drink.”

More recently she has begun to use the sign “later” to postpone discussion of possibly unpleasant subjects. “Tell me about what you did,” I demanded one day. “Later. Me drink,” was Koko's reply. She understands other words referring to the future. One bright morning that followed weeks of rain, I told Koko that if it was still sunny during the afternoon, I would take her out. When I arrived at three o'clock, she looked out at the still-bright weather and collected her gear to go outside.

In sign-language experiments with chimps, the animals learned to draw on different gestures to describe a new object or event. Dr. Roger Fouts, at the University of Oklahoma, noted that chimps could describe objects for which they had no sign: Washoe, for example, once called swans “water birds.” Koko, too, has generated compound names to describe novelties. She referred to a zebra as a “white tiger,” a Pinocchio doll as an “elephant baby,” and a mask as an “eye hat.”

A memorable joke turned on one of Koko's cleverer associations. Last winter, Cindy Duggan was holding a jelly container when Koko signed, “Do food.”

“Do where, in your mouth?”

“Nose.”

“Nose?”

“Fake mouth,” said Koko, opening her mouth and then licking the jelly container.

“Where's your fake mouth?” asked Cindy.

“Nose,” repeated Koko.

The next day I asked Koko what was a fake mouth, and she said, “Nose.”

Koko displays remarkable mental gymnastics in merging different signs to create compound or composite words. For instance, she has made the sign for “Coke” superimposed on the sign for “love.” For grapefruit—which she doesn't like—Koko simultaneously made the signs for “frown” and “drink,” executing “drink” in the position of the sign for “fruit.”

A Lot Still to Learn

Having worked only with Koko and Michael, I'm not in a postiion to rank chimp and gorilla in sign-language ability. However, by such indicators as range of vocabulary, frequency of utterance, and mean length of utterance, Koko must be considered at least the intellectual peer of the chimp.

After meeting Koko, Eugene Linden, author of Apes, Men, and Language, commended, “Compared to the sign-language-using chimps, the gorilla is calmer and more deliberate. Koko seems to resort to the sign language more often to express herself, and she discusses a wider range of activities.”

Even had I not come to know and love Koko as a witty, sweet, and trusting personality, I cannot foresee terminating Project Koko. Nothing indicates that Koko has reached the limit of her learning capacities. We have a great deal yet to learn from her.

Now there is the challenge of new areas of language use. Professor Patrick Suppes and his colleagues at Stanford's Institute for Mathematical Studies in the Social Sciences have designed a keyboard-computer linkup that permits Koko to talk through a speech synthesizer by pressing buttons. Simultaneously, all her utterances are transferred to a computer data file.

I noticed early that Koko responded appropriately to things I said in English, and often spontaneously translated spoken phrases into sign. For example, when asked in English, “Do you want a taste of butter,” Koko responded, “Taste butter.”

Now with the auditory keyboard, which produces spoken words when she presses keys, Koko can talk back as well as listen. The 46 active keys bear the usual letters of the alphabet and numbers. But in addition, each key is painted with a simple, arbitrary geometric pattern in one of ten different colors representing words for objects, feelings, and actions, as well as pronouns, prepositions, and modifiers.

If I place, say, an apple before her, she may push the keys representing “want,” “apple,” “eat,” and the computer-generated female voice speaks those words. Thus Koko can produce the spoken English for objects, ideas, and actions already banked in her sign vocabulary.

Typing usually with the index finger of her right had, but always reserving one hand for signing, Koko can sign and speak simultaneously. As she signs, she can type out an identical or complementary phrase, and the synthesizer will vocalizer her message. An ambidextrous and bilingual gorilla!

Koko responds to hundreds of spoken words independent of the auditory keyboard, but her vocabulary of spoken English that she can generate (it will surely expand) is now restricted to 46 words. A major objective is to evaluate the gorilla's sense of spoken word order.

Then there is Michael, the 5½-year-old male gorilla we acquired in September 1976 as a companion for Koko. Michael has been receiving sign-language instruction from Ann Southcombe—and from Koko, who has taken it upon herself to coach Michael's execution of the signs for “Koko” and “tickle.” So far, Mike's vocabulary is only about 35 signs, and he doesn't always sign fast enough for Koko.

Early this year Mike was fumbling for the right sign to convince Ann to let him in to play with Koko. After Mike signed, “Out,” Koko, waiting in her own room, began to get impatient. She signed to Mike through the wire mesh, “Do visit Mike hurry, Mike think hurry,” imploring him to come up with the right sign. Then she said, “Koko good hug,” and it finally dawned on Mike to say, “Koko.” A relieved Koko signed, “Good know Mike,” and then, “In Mike.”

Now the godmother of two gorillas, I weigh my responsibilities to this threatened species. I have set up the Gorilla Foundation to protect the future of Koko and Michael. My fondest hope is to establish Koko and Michael, myself, and my associates in a place set aside for the study of gorillas and for their preservation in circumstances of relative freedom. It is sad that the gorilla's best present prospect for survival is under the active protection of man. Yet it would be tragic should these animals disappear before we fully understand them.

“Fine Animal Gorilla”

That understanding enlarges as Koko grows ever more flexible and sophisticated in communication. Her recent progress is nothing short of astonishing.

Koko is defining objects. “What is a stove?” I ask her. She points to the stove. “What do you do with it?” “Cook with.”

“What is an orange?” “Food, drink.”

I ask Koko, “Tell me something you think is funny.” She signs, “Nose there,” pointing to a bird puppet's tongue. “That red,” showing me a green plastic frog we had talked about. When I put a stethoscope to my ears, Koko smirks and puts fingers over her eyes.

She perceives right and wrong, but is touchy about blame. During a videotaping session, when I turn away, she tries to steal grapes fro ma bowl. I scold her. “Stop stealing. Don't be such a pig. Be polite. Ask me. Stealing is wrong, wrong, like biting and hurting is wrong.”

Then I ask, “What does Penny do that's wrong?” Koko says, “Break things, lie, tell me 'polite' [when I'm] hungry pig.”

Koko is ill, a mild respiratory disorder. I ask her, “Where do you hurt?” Koko signs, “Underarms.”

Finally, Koko is learning self-esteem. A reporter asks about Koko as a person. I turn to Koko: “Are you an animal or a person?”

Koko's instant response: “Fine animal gorilla.”