The Dolphin in the Mirror

by Diana Reiss

  Those doing the attacking included experts in training circus animals and in the arts of trickery of various kinds. One critic was the Amazing Randi, a professional magician famous for unmasking self-proclaimed psychics. It was, wrote a reporter for Science, "a celebration of deception in all its varieties."2 The title of the conference, "The Clever Hans Phenomenon," referred to a famous German horse that performed arithmetical tricks at the turn of the twentieth century. His owner, Wilhelm Von Osten, a retired high-school math teacher, claimed he had trained Hans to add, subtract, divide, multiply, and perform other mathematical and intellectual feats. He would ask Hans questions that required numerical answers; for instance, he'd ask, "If Friday is the fourth of October, what is the date on the following Tuesday?" Hans would then tap his hoof eight times, to the great amazement and delight of the audience. The secret behind this apparent phenomenon was this: as Hans approached the correct number of taps, his trainer, unwittingly, tensed up; when Hans reached the right answer, his trainer very slightly relaxed, displaying relief in his imperceptible posture change. Imperceptible to ordinary folk, that is, but not to Hans, who was able to detect head movements as minute as a fifth of a millimeter. The trainer was, unknowingly, cuing Clever Hans—Von Osten wasn't deceiving his audiences; he was deceiving himself.

  Clever Hans's talent was not in mathematics but rather in his extraordinarily keen perception. When Hans was tested in the absence of Von Osten, he could still solve the problems, provided the person testing him knew the correct answer; Hans read the subtle, inadvertent cues of his questioner. This is known as the Clever Hans phenomenon. Scientists now recognize the need to eliminate their own unconscious cues when they're testing animals, but they also recognize the remarkable sensory acuity and problem-solving abilities of their subjects. Clever Hans was clever indeed—but not in the way people thought.

  Essentially, the meeting addressed the possibility that the ape-language researchers, and others doing similar work on different species, were, like Von Osten, deceiving themselves. In truth, the people involved in ape-language research at that time had done some very powerful studies and had accumulated persuasive evidence that apes (and an African Grey parrot) could indeed learn to associate symbols with objects and actions. Whether one would call that language is another matter. In any case, they were well aware of the dangers of cuing and were more sophisticated than their critics in their understanding of what communication was all about.

  Dolphins can be terrifically subtle readers of cues. At the conference, John Prescott described a dolphin at the New England Aquarium who had been trained to respond to a visual signal, an arm raised in a particular direction, by leaping out of the water and jumping over a bar. In an experiment, the dolphin was able to do this same behavior even when it was blindfolded. The trainer assumed that the dolphin accomplished this by using echolocation through the air as a substitute for visual cues. Dolphin echolocation is usually done underwater, of course, yet this seemed to be the only explanation.

  It was, however, wrong. Prescott's colleagues eventually realized that when the trainer raised his arm as the signal to jump, he preceded it by taking a half a step forward. Unbeknownst to the trainer, that step provided an auditory cue to the dolphin, which, like all dolphins, had very acute hearing. It was the sound of the step, not the arm raising, that the dolphin understood as the signal to jump. When the trainer performed the signal he had been using for seven years with this dolphin but without taking that little step, the dolphin simply remained in front of the trainer and did nothing.

  Karen Pryor was at the conference too. I often cite her best example of cuing, which occurred at Sea Life Park Hawaii. One day one of the trainers came to her, visibly excited, and said, "The dolphins are psychic! This particular dolphin is psychic!" Karen asked her what on earth she meant, and the young woman explained that the dolphin would obey a gestural command before she actually gave it. The trainer thought the dolphin, being smart, might have remembered the order of the commands and simply done the set of responses from memory. So the trainer did what every good scientist would in those circumstances: she did the command gestures in a different order. "The dolphin still responded correctly before I gave the commands," she told Karen.

  In fact, what the dolphin had been picking up was not the arm gestures, or the trainer's thoughts, but a slight shift of the trainer's body in a particular way before she moved her arm. Each arm gesture had a subtle body movement associated with it that preceded the gesture itself, and this was what the dolphin was detecting and responding to. Like Clever Hans, this dolphin had supersensitive visual perception (perhaps there was an acoustic cue as well) enabling it to pick up tiny body movements that the trainer was completely unaware of and that were imperceptible to most spectators.

  When we think about dolphins being smart, we tend to think about intellect in the realms with which we are familiar, things that we humans do. These incidents of cuing tell us that these creatures are smart in ways that ordinarily we don't even consider. I will give animal-language critics credit for one major point, however: whether chimps, gorillas, parrots, or dolphins can be said to have any language ability, in the sense that humans use the term language, is a very complex question. We know they communicate, and we know they are intelligent. But should we expect them to have a sense of nouns and verbs? Or—in my case—should we let ourselves be open to whistles or to something complex yet fundamentally different from human syntax?

  ***

  A few years after the conference, I spent the night at my lab and pondered some extraordinary sounds as I lay on a small cot in my lab, just feet from the dolphin pool. Hydrophones picked up whistles from the water as Terry, Circe, and Gordo felt the need to break their silence, and these somewhat eerie sounds were pumped into my room. Often, the three dolphins were quiet. Dolphins are most vocal when they are at a distance from one another, and when in close proximity, they apparently turn to other means of communication. At night, the distinctive guttural roar—aaouu ... aaouu ... aaouu—of nearby lions added to the acoustic tableau. All this took place against a constant background of chirping crickets at Marine World Africa U.S.A. in Redwood City.

  The research pool at Marine World was in the middle of a raised wooded mound in what was effectively a six-hundred-acre wildlife theme park. Aside from the nearby lions, there were two water buffalo, Waldo and Wilma, close by, housed just behind me in a large enclosure. They looked fearsome, these giant bovines, but they were softies, really. My path to the pool each morning took me past them, and when I stopped at their fence they would come over to me, eager for the apples I always brought for them. I rubbed their huge, curved horns and patted their big black noses. They were adorable. Off to another side was a chimpanzee enclosure, from where we occasionally heard excited hoots when food arrived. An exaggerated mural of the African savanna covered a long wall nearby. So, although a good deal of excellent research was done at Marine World by local scientists, the area didn't give the appearance of the usual research environment.

  One day soon after I began my work, I was by the pool feeding Circe. I happened to look up, and I saw a group of people by the wall, staring at me. That wasn't especially unusual; this pool wasn't accessible to the public, and I was used to people peering over to see what I was doing. But this morning was different. They weren't staring out of curiosity. They had looks of mounting horror on their faces, and some of them started to run. I looked at them, wondering what was so alarming, and then I felt a gentle nudge at my back. I turned, and there was Waldo staring at me, as friendly as could be. Someone must have left the gate to his enclosure open. I scratched Waldo's nose, got out my walkie-talkie, and said to security, "We have a Houdini"—the code many zoos use when an animal has escaped. I added, "Get me some apples." Waldo was very easily led back to his area.

  ***

  Ironically, Terry and Circe were thought to be "loser" dolphins, as both research subjects (according to none other than John Lilly) and
performers (according to Marine World trainers). Lilly had heard about my program at Marine World, and he'd contacted the owner and negotiated his own use of two pools in an area behind the scenes. He set up shop and began a new but short-lived research project, Project Janus, trying again to train dolphins to speak English. He brought two female dolphins with him. One died due to a complicated pregnancy. The other, Terry, seemed too aggressive for the project and was transferred into our research pools. Terry actually turned out to be a sweetheart, and each time I arrived at the pool she greeted me by swimming rapidly around, porpoising occasionally, and then coming to where I stood and looking at me with big eager eyes. Terry was older than Circe, more mature, and she behaved like the matriarch. I worked gently with her to get her to trust us, and eventually she did. She was also very affectionate.

  As for the younger female, Circe, the Marine World trainers had tried to work with her but described her as standoffish. She didn't like to be approached. Initially I had to keep my distance. I actually found this quite intriguing, and I thought of her as a kind of Asperger's syndrome dolphin. We gradually negotiated a relationship and eventually bonded very closely. The wall of the pool extended about three and a half feet above the ground, up to my waist. I could almost be face-to-face with Circe when she came to the edge. Many times we would gaze eye-to-eye for minutes on end, and I could sense her relaxing as we did this. Then she would roll over and present her belly to be stroked. When I did, her whitish skin would become rosy-colored, a sign of arousal. She liked me to rub her tongue, which seems to be a special source of pleasure for dolphins (Circe in France and other dolphins I've been with seem to like this). I have to admit that Circe was my favorite, and when she gave birth to Delphi, it was like my having a grandchild. My tight bond with Circe carried over to Delphi, and he became my muse. I loved these animals dearly, and I proudly insist that this does not get in the way of doing good science. In fact I think it helps.

  And then there was Gordo, a Pacific bottlenose dolphin, a big male, eight hundred pounds, and the gentlest of creatures. I thought of him as a couch potato of sorts. In the spring of 1984, almost two years after I started at Marine World, Gordo's health began to decline rapidly and for no reason that the multitudes of vets that examined him could discover. He was very large. His problem may have been hormonal; his blood numbers looked bad. We rigged up a giant sling so that he would be supported in the water. He was weak for some days, and I could see he was slipping away from us. I was very worried and upset. I was in the pool with him often during these final days, usually along with a couple of trainers, giving him comfort, holding up his body when he was not in the sling. As I stood by him, I kept thinking of the ancient myths and modern news accounts of dolphins rescuing drowning people, holding them up, staying by, taking them safely to shore.

  One evening, after I'd been in the water with Gordo for five hours, rubbing his skin to comfort him, the phone in my office rang. An assistant shouted to me that I really should take the call. I was reluctant to leave Gordo, but I did. As I was getting out of the pool to get to my office just fifteen feet away, I looked back, and I could see him straining to look in my direction. I made the call short, explaining the situation, but when I returned and placed my hands on him again, Gordo was gone. Dolphins can live to be over sixty years old, but the average lifespan in the wild is unknown. In my thirty years of research and rescue work, I have become close to many dolphins, of which a few have died. Every death is like a family member's passing; I mourn the loss of each deeply.

  ***

  The big question when I arrived at Marine World in the fall of 1982 was, Where will the money to build a research facility for a newly minted PhD who has almost no research experience come from? With the brash confidence of youth, I assumed the money would, as Gandhi once put it, "come from where it is now." I closed up my apartment in downtown Philadelphia and moved to an apartment in the Pacific Heights District of San Francisco. I landed a part-time teaching job in the Speech Department at San Francisco State University. A second part-time job doing biomedical research at Stanford Research International, in Menlo Park, boosted my tenuous financial situation, but it meant that I had to get to SRI by five in the morning, work until ten, and then go to Marine World to pursue the task of establishing a research facility. It was a tough schedule, and it didn't provide sufficient funds for building and equipping the kind of lab I had in mind.

  I had an audacious idea. "Hello, I'm Diana Reiss," I said when Dr. Bernard Oliver answered the phone. "I just got my PhD. I'm starting a dolphin lab with a research program on communication here at Marine World, and I would love to talk to you about my research." Barney Oliver was founder and director of the Hewlett-Packard laboratories, and he'd played a key role in developing the HP-35 pocket calculator. He was also deeply involved in NASA's Search for Extraterrestrial Intelligence project, or SETI. That was my motive for contacting him. We had a lot in common, the SETI group and I, I reasoned. The people at SETI were trying to detect signals from an alien intelligence and find a way to decode them, and I was planning to record signals from a nonterrestrial intelligence and find a way to decode them. The only difference between us was the source of those signals. "Your work sounds very interesting, Dr. Reiss, but I am very busy right now," Barney said politely. In retrospect, I suppose I was obnoxiously pushy; I said, "Could I just come down and have lunch with you?" He was very kind and said, "Okay. How about tomorrow?"

  What I didn't know at the time was that John Lilly had gone down this same path of reasoning more than twenty years earlier, and the SETI people had been very receptive to it. SETI was effectively established at the now famous 1961 meeting in Green Bank, West Virginia, at which the astronomer and astrophysicist Frank Drake introduced his famous equation calculating the number of civilizations in our galaxy that could potentially communicate with us. (It was known variously as the Drake equation, the Green Bank equation, and the Green Bank formula, and the solution is around ten thousand.) The meeting's attendees had come together to talk about the possibility of detecting intelligent life in the Milky Way, and they included astronomers, physicists, biologists, social scientists, and industry leaders. This group later came to be known as the Order of the Dolphin, in recognition of the common goal of its astrophysicists and biologists: to establish communication with "alien" intelligence. When I arrived on the scene, SETI was located in Mountain View, just thirteen miles southeast of Marine World. Very convenient.

  The outcome of the lunch I had so audaciously pushed for with Barney Oliver was the evolution of a symbiotic relationship between me and SETI people, which included Frank Drake, Carl Sagan, and Jill Tarter, director of the SETI Institute and the woman on whom Jodie Foster's character in the film Contact was modeled. I soon found myself on discussion panels with eminent astronomers, astrophysicists, and other scientists so notable that I never would have imagined I'd even meet them, let alone sit side by side with them debating issues of mutual interest in an atmosphere of intellectual collegiality. I contributed articles to publications on exobiology, one of which was a paper with the title "The Dolphin: An Alien Intelligence."3 The title was a bit of a mistake. It should have been "A Nonterrestrial Intelligence," but "Alien Intelligence," a phrase I used from time to time, mostly as a joke tailored for that community, slipped through. It didn't matter, because the issue was the same: How do you decode a signal that is completely foreign to the knowledge and experience of your species?

  In addition to the intellectual relationship that was established that day at lunch, a more practical relationship began as well. Barney generously donated some of his personal HP stock, which allowed me to get my lab up and running and was sufficient to keep it operating for several years. Despite this, I was still on a very tight budget, but I was lucky with my subject of research; it made raising funds easier than it would have been if I were studying, say, the digestive system of liver flukes. I was frequently on the phone saying something like "Hello. I am doing r
esearch on dolphin communication here at Marine World, just down the road from you. We have very limited funds and I was hoping you might consider donating [some piece of equipment] in support of the research program." The answer was usually "Yes, willingly." For instance, I was given a generous supply of audiotapes from Ampex Corporation, a neighbor in Redwood City. With the bounteous support of Ampex, several other Silicon Valley companies, and the U.S. Geological Survey, I was now ready to embark on my research program proper.

  ***

  Terry, Circe, Pan, and Delphi were the first dolphins with whom I tried keyboard studies. The experimental setup was fairly simple, too simple probably to allow the dolphins to show us what they were really capable of, but I was constrained by engineering issues. I was lucky to have a terrific engineering student, Bill Baldwin, who helped me construct my equipment, a twenty-one-by-twenty-four-inch underwater keyboard with nine positions on it in a three-by-three matrix.* Designing an underwater keyboard for dolphins is not a trivial engineering task, given the need for safety, sensitivity, and speed of switching in a corrosive, saltwater environment; water is a hazardous environment for most electronic equipment. Our solution was to power the keyboard by light, via fiber-optic cables, so that when a key was pressed by the dolphins, a light beam was broken behind the key. Since my lab was generously supported by Barney Oliver, the founder and director of Hewlett-Packard laboratories, he paved the way for his company to donate the fiber-optic cables.

  When I began the study, I had only three labeled keys: a circle, which represented "fish"; a triangle, which represented "ball"; and an H, which represented "rub"—that is, one of us would rub the dolphin. Bill had cleverly engineered the keyboard system and software in such a way that if a dolphin touched, say, the triangle key, a specific computer-generated whistle would sound in an underwater speaker for the dolphins, and in my headphones I would hear the word ball produced by a computer-generated voice. My job would then be to give the dolphin the ball. The same thing happened for the fish key and the rub key. I would give a fish or a rub, depending on what I heard in my headphones.