Table of Contents
( Note: All signed wordsthose made in American Sign Languageare indicated in italics. )
Project Koko began in July 1972, the day after I received permission from the San Francisco Zoo to attempt to teach Ameslan to an infant gorilla. I had had my eye on this gorilla for nine months. In fact, I had begun planning Project Koko the day I first saw little Hanabi-Ko, or Koko, as she was nicknamed. And months before I first saw Koko, I had decided that I would devote my graduate education to the study of the language abilities of animals.
I was inspired by a lecture delivered at Stanford University by Allan and Beatrice Gardner, the comparative psychologists who first succeeded in teaching language to a great ape. This was in September 1971, five years after the Gardners had begun their work with Washoe, a chimpanzee, and ten months before I was to begin working with Koko. I had read some material on the Gardners' research, and wanted to hear them describe their methods and their accomplishments and see the films of Washoe conversing in sign language with her human companions.
As the Gardners described how they got the idea to teach sign language, their search by trial and error for a proper teaching method, the elaborate controls they developed to ensure that their data were reliable, and finally Washoe's willing response to their efforts to teach her language, I felt increasing excitement. Clearly there might be untapped language abilities in other animals as well. Although the Gardners delivered their lecture soberly, I felt that I was hearing about something from the realm of myth or fable: Animals were capable of telling us about themselves if one knew the proper way to ask them.
This lecture gave focus to my lifelong interest in animals. I started planning to try to find an ape and the funding that would permit me to pursue research along similar lines, and I enrolled in a course in American Sign Language. My inclination was to work with chimps because they were noted for their tractable, gregarious nature. At first I did not entertain the idea that it might be possible to try to teach language to a gorilla. But I would have leapt at the chance to work with any great ape.
I did have some background working with primates. I had entered the doctoral program in psychology at Stanford in the fall of 1970, after receiving a BA in psychology from the University of Illinois and traveling west with Ronald Cohn, a molecular biologist and close companion who has devoted all his free time to Project Koko since its inception. My interest in psychology came from my father, who is a professor emeritus in educational psychology at the University of Illinois and has published several books on counseling and psychotherapy. For me, however, graduate work in psychology was attractive because it would permit me to work with animals.
To the nonresearcher, the idea of a behavioral scientist "working with animals" often conjures up an image of the horrors of vivisection. This was not what I had in mind. I count myself among a "new breed" of behavioral scientists who would rather observe an animal than take it apart. We are more interested in understanding animals in their own right than in seeing how they might be used to understand and cure human problems. Indeed, the most delightful aspect of my work with Koko is that language allows us to see the complexities and subtleties of the gorilla's mind.
In effect, my career in psychology has been one of climbing the primate ladder - if in fact we can consider one primate higher than another. I began at Stanford working on a study of attachment behavior in rhesus monkeys under the guidance of Karl Pribram, a leading theorist on neuropsychology. In this study, infants were separated from their mothers (briefly) to prove what seemed to me the obvious point that they would prefer their mother to a peer and a peer to an empty cage as a source of comfort in an anxiety-producing situation.
Next I became involved in a study of self-recognition in gibbons. Simply put, this means that I was trying to see whether a gibbon knew whom it was looking at when it saw its image in a mirror. I found this study more intriguing because it would indicate whether the ape has any consciousness, a quality that had proven chimps capable of self-recognition, and the purpose of the study of gibbons was to help to determine how far down the evolutionary scale this ability might extend. The six months of the study produced no signs of self-recognition in the gibbons.
It was shortly after I began work on the self-recognition study that the Gardners came to Stanford to speak. From that moment onward, I began looking for opportunities to work with a chimp, or failing that, any great ape. Thus I agreed instantly in September 1971 when Karl Pribram suggested that I accompany him to San Francisco to look at the gorilla colony there. Dr. Pribram was toying with the idea of constructing a sturdy console with an encoded keyboard connected to a computer, which he would then use to teach the gorillas to communicate by pressing different keys.
When we arrived at the San Francisco Zoo, we met the director, Ronald Reuther, and then walked over to the gorilla grotto, a large, rocky, cement area separated from onlookers by a dry moat. While Dr. Pribram and Mr. Reuther discussed the pros and cons of the proposed experiment, I became absorbed watching the gorillas idly pass the day. The tableau was a study in lassitude, broken only by a little struggle between a mother gorilla and her infant. The tiny gorilla was clinging ferociously to its mother, who kept pushing the baby up onto her back, only to have the baby slide off each time. The sight of the infant brought my mind back to my quest. I was not that interested in Pribram's proposed experiments because I had already concluded from my reading on the subject that a sign language was the most productive way to study ape language abilities. As I watched the infant I thought, "Well, Pribram can have his experiment, and I will just have mine with this baby." It did not turn out to be so simple.
When I made a proposal to the zoo director, I was turned down. A primary goal of the zoo was to breed endangered species such as the gorilla, and Mr. Reuther, sensibly enough, felt it would not advance that purpose to separate the infant from its mother at the tender age of three months. Undaunted, I continued my study of Ameslan and resolved to find another gorilla or wait until this infant was older. I tried to find out what I could about the baby gorilla and her circumstances at the zoo.
The infant was Koko. The mother who had so peremptorily placed her daughter on her back was Jacqueline, nicknamed Jackie. Poor Jackie had previously suffered the indignity of being thought to be a male. In fact, she had been purchased from the Brookfield Zoo in Chicago to be the mate for Missus, one of the San Francisco Zoo's female gorillas. Jackie came to San Francisco courtesy of Carroll Soo Hoo, a philanthropic businessman, who donated the money to purchase Jackie - then named Jacob - and another gorilla. The zoo expectantly closeted Jacob with Missus and nervously wondered why the couple did not hit it off and raise a family.
Ultimately, the zoo discovered their error and, with some embarrassment, decided that the cause of breeding gorillas might be better served if Jacob was put in with a male. The zoo changed her name to Jacqueline, and undoubtedly to her vast relief, Jackie was introduced to Bwana. They did mate, and a female gorilla was born on the Fourth of July in 1971. The zoo held a contest to choose a name for the infant. The winning entry was Hanabi-Ko, Japanese for "Fireworks Child."
The zoo's plan to keep Koko with her mother did not work out as they had hoped. Shortly after Dr. Pribram and I visited the gorilla grotto, Koko's health began to deteriorate. Jackie was a good mother, but the San Francisco Zoo is not the jungle. Jackie's milk was not sufficient to keep Koko nourished, nor could Koko supplement her mother's milk with forage as infant gorillas are reported to do in the wild. She became undernourished, and when an outbreak of shigella enteritis swept through the gorilla compound, she almost died. Suffering from malnutrition, racked with diarrhea and septicemia, hairless, and dehydrated, Koko was a pathetic 4 pounds 14 ounces - the average birth weight of gorillas - at the age of six months. At that point, just before Christmas, Koko was separated from her mother and taken to the Animal Care Facility of the University of California Medical Center in San Francisco for a few days before being taken into the Reuther household for two weeks. With round-the-clock care, she recovered sufficiently to be transferred to the house of Deedee and Landis Bell, manager of the Children's Zoo, on the Children's Zoo grounds. After six months in the Bell's care, the zoo felt it was time to put Koko back on permanent display, and installed her in the nursery of the Children's Zoo. Subsequent examinations determined that she had suffered no discernible lasting harm as a result of her illness.
At about this time, I made another trip to the zoo. I had come up to photograph gibbons as part of the self-recognition study. I ran into one of the keepers, Marty Diaz, who told me about Koko's illness. He suggested that the zoo might now listen more favorably to a proposal, if I still wanted to work with Koko.
Marty Diaz was most sympathetic to my desire to work with sign language, and he offered to speak to Mr. Reuther on my behalf. That same day, I asked my advisor for permission to switch to a language project with Koko. Mr. Reuther and my advisor both granted their permission, and the very next day, with no funding, few private resources, and a yet no formal project design, Project Koko began.
Too excited to be tired from a night sleepless with anticipation, I drove from Stanford to San Francisco with Ron Cohn to meet Koko on a foggy Wednesday morning on July 12, 1972. When I entered the nursery of the Children's Zoo, Koko left the arms of her caretaker, Debbie Lee, for mine. She pushed her soft face close to mine, smelling me and looking me over. Then Debbie put the 20-pound gorilla, all black save for a white rump patch, onto the nursery floor and I signed, Hello (a gesture somewhat like a salute). Koko put her hand on her head and patted it and then promptly pulled my hair as I sat down.
The glimpse I had caught of her sleeping serenely in her basket the day before did not prepare me for this interaction - she was a real dynamo and seemed much bigger this day. While Debbie was in the room with us, Koko responded to my beckoning come gesture, but later, when alone with me, she went on about her play with her toys as if I wasn't there. Whenever I stood up, however, she rushed to my feet and started to scale my legs- evidently she thought I was leaving. At one point Koko became excited and played a game of peekaboo behind a door with Ron when he and Debbie joined me in the room for a quick photo session. Later, while Debbie and I chatted, Koko bit me a couple of times. Taking this as a sign that we had perhaps overstayed our welcome, Ron and I departed for the day.
The next morning I arrived at 9:00 A.M. with a wading pool for Koko. She cautiously put her nose up to it, touched it, and nibbled on the edge. When Debbie placed Koko in it, she immediately ran her fingers over the upraised bubbles on the bottom of the pool. She delighted in running in and out of it and splashing in a few inches of water. Excited by the pool, she nipped me several times, but by now I was learning to anticipate and divert these testy assaults.
While the zoo volunteers performed the morning chores I joined Koko in the nursery. She still ignored me often, but when the horses, goats, and sheep were let out into the zoo yard and stampeded by the nursery window, Koko scrambled over to me and briefly clung to my clothes. Then the whirring of the blender to mix her formula of similac and strained cereal set her into a frenzy of activity: She vigorously banged her toys around, and repeatedly pounded on and rolled herself over a rubber dog. She interrupted her wild play only to peer under the door to the adjoining room where her bottle was being prepared and to hammer on the door periodically. I asked the zoo volunteers to sign drink before feeding Koko her formula and up before picking her up.
Initially, Koko seemed to prefer men to women. During the first week, she was more inclined to interact with Ron and my office mate, John Bonvillian, than with me. She took to John very well - examining his beard closely, sniffing, fingering, tasting, and yanking it. She climbed all over him jungle-gym style and rode on his back. Ron also got the jungle-gym treatment, and Koko was very responsive to him. She imitated his twisting of a knob on her toy clock, and his clapping. When my male friends were present, Koko interacted very little with me. I also somewhat enviously noted that she never attempted to bite them. After a couple of weeks, though, she seemed to conclude that I was a reliable, and likely to be a permanent, fixture in her life. She attempted to bite me less and less frequently, and she also began to show a preference to be held by me rather than by a man when she had an alternative. Her first response when frightened was to jump into my arms and cling tenaciously.
From the beginning of Project Koko I had a dual role: I was a scientist attempting to teach a gorilla a human sign language, but I was also a mother to a one-year-old infant with all an infant's needs and fears. My initial problem was to establish rapport with Koko, who was, perhaps because of the unsettling events that had marked her short life, at first suspicious of this strange blonde human.
Each morning before the zoo opened to the public I would carry Koko for walks through the Children's Zoo. I felt it was important to get Koko out of the confines of the nursery at every opportunity. At first I had no need to restrain her with a leash; for one thing, it is normal for an infant gorilla to stay on or near its mother for the first year and a half of life, and for another, Koko was terrified of the large animals (particularly a baby elephant who was fond of trumpeting every morning) and wouldn't venture from my side. The only large animals that Koko could intimidate at that age were a herd of surpassingly stupid llamas. They would congregate at the fence when we passed, apparently under the impression that we were zoo goers bearing llama food. Koko would rush at them threateningly and enjoy with evident satisfaction the stampede she precipitated.
One animal Koko was particularly afraid of was the gorilla. When I took Koko on a trip to see her parents at close quarters inside the gorilla compound, her relatives gathered quietly to examine the little gorilla. Bwana, the dominant male and protector of the group, was upset when he first saw us approach; he barked, followed us, and threw feces at us. Frightened, Koko squirmed and defecated in my arms. We left in a hurry.
With the beginnings of our rapport, the problem was to focus Koko's attention on hands. Koko was, after all, only one year old, and when not asleep, she was constantly moving and exploring. I would construct little games to divert her and show her the utility of her hands. I breathed fog onto the glass of the large windows in her room and then drew stars and simple faces on the misted surface. Koko loved these games and would attempt to draw as well, although what appeared were amorphous squiggles.
It was impossible within the confines of Koko's display cage to seal her off from spoken languages (the glass was hardly soundproof and some zoo visitors seemed to take it as a sacred obligation to make remarks to Koko and whoever was in with her). Consequently, I decided to make a virtue of necessity by adopting a method known as "simultaneous or total communication." This simply means that the speaker accompanies his signing with the spoken equivalent of the message.
The ambitions of the project were quite modest at first. On July 22, Karl Pribram and I spoke with Ronald Reuther about the amount of time I was to be allowed for the project. Mr. Reuther's idea was to reunite Koko with the other gorillas as soon as possible, which he thought would be in about six months. On the other hand, Landis Bell, the director of the Childrenžs Zoo, thought Koko should not be put back with other gorillas for about three years. I was a bit disappointed at this point, since I hoped to carry on my work with Koko for as long as the Gardners had worked with the Washoe - four years. On the other hand, Dr. Pribham felt that I should concentrate on teaching Koko only three or four signs. I thought she could probably handle more than that, but decided to begin by molding and shaping drink, food, and more.
I would divide Koko's bottle into two portions, and would sign drink before giving her each portion. The drink sign is made by shaping one's hand somewhat like a hitchhiking gesture, and then placing the extended thumb to the lips. While preparing and offering the bottle, I made this gesture, and then attempted to get Koko to make the gesture. Koko, being a one-year-old, had few thoughts other than getting her hands on the bottle, and then the bottle into her mouth.
Although I tried for a strict routine, we were frequently interrupted when children came up to the glass, and then, when she discovered they were out of reach, she would pound on the glass in frustration. Her principal amusement those first few weeks was to close her eyes and spin wildly around the cage - something gorillas do in the wild. As Koko grew older, she embellished this game by pulling a blanket over her eyes, generally when she had some mischievous intent, such as giving a playful smack to a human companion. Possibly Koko felt that by pulling the blanket over her eyes she became invisible. Indeed, she was perpetually surprised to find herself accused of these petty assaults.
During the first few months of the project, the Children's Zoo volunteers who had looked after Koko before my project began continued to look after her when I could not be with her. At the end of the first summer, these volunteers had to go back to school during the week, but I was able to fill the gaps with two new volunteers who offered their services. One was a deaf woman, the mother of my sign language teacher. The other was Barbara Hiller, a docent at the zoo. Barbara cared for Koko from the time she was in diapers and is sill with the project today. Later in the fall, the Stanford psychology department provided salary money that permitted me to hire Hank Berman, an assistant whose native language was sign.
As Project Koko got underway, I had the advantage of surveying the trial-and-error approach to teaching language used in previous experiments with chimps. These experiments also produced a great fund of information against which I might judge Koko's performance - if, in fact, she learned language at all. In 1972, when I began Project Koko, there were a great number of scientists who disputed that the chimps' achievements had any linguistic significance. Project Koko began during turbulent times in the behavioral sciences, and it was only because of previous pioneering work with chimps' that I had any chance of being taken seriously. My cause was not helped by the fact that the subject of my experiment was not a chimp, but a gorilla.