Alex & Me - How a Scientist and a Parrot Discovered a Hidden World of Animal Intelligence--And Formed a Deep Bond in the Process

by Irene Pepperberg

  We had a very short amount of time scheduled for the demo, and the sponsors were very keen to see Alex do his stuff. I showed Alex a tray of his letters. “Alex, what sound is blue?” I asked.

  He answered, “Sss.”

  It was an S, so I said, “Good birdie.”

  He replied, “Want a nut.”

  Because we were pressed for time, I didn’t want to waste it with Alex eating nuts. I told him he had to wait, and asked, “What sound is green?”

  Alex answered “Ssshh.”

  Again he was right. Again I said, “Good parrot.”

  And again Alex said, “Want a nut.”

  “Alex, wait,” I said. “What color is ‘or’?”

  “Orange.”

  “Good bird!”

  “Want a nut.” Alex was obviously getting more than a little frustrated. He finally got very slitty-eyed, always a sign he was up to something. He looked at me and said slowly, “Want a nut. Nnn…uh…tuh.”

  I was stunned. It was as if he were saying, Hey, stupid, do I have to spell it out for you? More important, though, he had leaped over where we were with his training, which was individual phonemes, and gone on to sound out the parts of a complete word for us. Perhaps he was really saying to us, I know where you’re headed with this work! Let’s go on with it. Let’s do whole words! It was a stunning moment, and it made me wonder just how far beyond our expectations Alex was going to lead us in the years ahead.

  A few months later, the future looked even more promising when the possibility was raised that I might be able to stay at the Lab beyond the two years that had already been agreed. There was a lot of back and forth over whether I would be able to get a professorship or a long-term contract as a research scientist. It was a tense time for me, because I needed to know well before the beginning of fall term 2001, the time I was due to finish my leave of absence and return to Tucson. At one point in August I had two moving trucks reserved, one in Boston, in case I had to move back to Tucson, and one in Tucson, in case I was going to stay in Boston and needed my things there.

  Finally, at the very last minute, I received a five-year renewable contract as a research scientist, with all the financial support I would need for my work. True, I was going to have to give up a tenured job for a nontenured one. But I could not have been happier. I wrote to Tucson and told them I would not be coming back. My future could not have been more promising. I would be able to continue the cognitive explorations I had pursued for so many years, and I was going to continue to explore technological applications for them. I would not have to worry about research funding. And Alex and I would be together.

  Three months later, in mid-December 2001, I learned that I was one of thirty people whose jobs were to be eliminated at the Lab. There had been storm clouds gathering about the Lab’s financial health. The technology-heavy NASDAQ Composite index had peaked a year earlier and had begun what was to be a catastrophic decline, marking the end of the dot-com bubble. Then 9/11 happened, which exacerbated the economy’s woes. Corporate sponsors of the Lab were no longer able to keep up their previous level of support.

  When I had arrived at the Lab two years earlier, it was at the peak of its technological and financial exuberance. I had imagined a future with limitless possibilities for my research. Now I had no job and nowhere to continue my work with Alex and friends.

  Even before the December announcement, there had already been problems with where to house the birds. In September I had moved Alex and Griffin to Newton, a suburb of Boston, to live in Margo Cantor’s house. Margo’s son had been one of Alex’s MIT trainers, and she kindly agreed to look out for the birds until we sorted out housing arrangements. Wart, meanwhile, went to live with my friend Maggie Wright, in New York City. These were meant to be temporary arrangements, lasting no more than a few weeks before we would move them back to new facilities. Now I had no idea where they might go, or when, or how we would support the research. Or how I would support myself.

  Chapter 8

  The Next Horizon

  Alex was miserable and angry. Margo Cantor and her husband, Charlie, could not have been kinder in giving refuge to Alex and Griffin in their house in Newton. Alex really liked Charlie, and Griffin was very affectionate with Margo. But Margo and Charlie were gone all day long, leaving Alex and Griffin locked in their cages. It was just the situation I had always warned parrot owners to avoid. I spent my days at MIT working on manuscripts, applying for jobs, and trying to find lab space nearby to house the birds properly.

  Each afternoon I drove the eight miles from Cambridge to Newton. I tried to greet the birds cheerily, though my mood was very dark. Alex often stuck his beak in the air and turned his back to me, punishing me for abandoning him. He sometimes refused to come out of his cage, quite uncharacteristic for him. I stayed with him and Griffin until Margo came home, around six. I then went back to the Lab to cram in a few more hours of work. Both Alex and Griffin were very subdued during this trying period. They showed their stress by plucking feathers as the temporary housing arrangement stretched from an anticipated few weeks to an eventual five months.

  Even before I’d been let go from the Lab, I’d begun searching for lab space. The room the birds had shared with Ben and Spencer was needed for other projects. I was fortunate that Bob Sekuler, a friend and vision physiologist from Northwestern who was now at Brandeis University, had offered to help me secure space there. Brandeis University was relatively close by. I managed to obtain a room in the animal care facility of the psychology department. All it really needed was a coat of paint. It was mine pretty much as long as I could pay the rent.

  Eventually, when I had a lab manager and a group of students, the annual tab to Brandeis grew to $100,000. I had an unpaid adjunct position with Brandeis and no research grant. Therefore The Alex Foundation had to pay the bills. Fund-raising became a constant refrain, and strain, in my life. But at least I had somewhere to continue my work.

  Alex and Griffin moved to Brandeis in mid-January 2002, and Wart joined them shortly afterward. Wart had had a ball in New York City. My friend Maggie worked at home a lot, so he’d had her company as well as that of two young female Greys. He had been top bird for those five months; now he was going to have to get used to being at the bottom of the hierarchy again.

  Space was again tight in the new quarters. The room measured no more than ten by fifteen feet. With three large cages, cupboards, bookshelves, a small refrigerator, a sink, T-stands for the birds, the lab manager’s work desk with a computer, and a chair, it was crowded. Add one or two students and the stools they sat on during training, and, well, you get the picture. Fortunately, however, Arlene Levin-Rowe was my lab manager from late fall 2002 onward. Not only is Arlene a superb organizer and a natural with the birds, she is also one of the kindest and most even-tempered people I know. Life at the Brandeis lab would be hard to imagine without Arlene to make everything run harmoniously.

  The cramped quarters affected the birds, particularly Alex. In Tucson, each bird had his own room, where he was trained and tested and where he slept. They shared the large communal lab space for chill-out times. Even at the Media Lab, where the communal room was small and crowded, the birds had had their own separate sleeping facilities. Now they shared one room for training, testing, chilling out, and sleeping. Alex, always the boss of the lab, became yet bossier. He was the unrivaled “big man on campus,” and he made it known.

  Alex would subject new students to an endless stream of requests—“Want corn,” “Want nut,” “Wanna go shoulder,” and so on—effectively ensuring that the new person knew his lexicon. He’d always done that, but now there was more urgency. He also tried to trick new students into giving him extra corn in the afternoons, when he’d already had his noontime ration. His higher-octane bossiness was most obvious when we were trying to test Griffin on labels and concepts. In Tucson, Alex’s opportunities to butt in were relatively rare; now they were constant. When Griff hesitated with his answer, A
lex marched to the edge of his cage top and piped up with it from the back corner of the room. Alex occasionally even chimed in from inside his cardboard box on top of his cage.

  If Griffin answered at all indistinctly, Alex would admonish him, “Say better.” If I asked Griffin, “What color?” Alex might butt in with “No, you tell me what shape.” Sometimes Alex gave the wrong answer, thus further confusing the already unsure Griffin. Alex was, to put it bluntly, a pain. Wart, meanwhile, was content to stay in his cage, playing enthusiastically with his toys.

  The avian hierarchy that had always existed became even more sharply defined, with Alex insisting his seniority be properly acknowledged. He always had to be top bird. Literally. There are several photographs of the three birds and me in the new lab, always apparently portraying a harmonious “family.” In reality, because Griffin liked to be on my shoulder, I had to arrange a more prominent perch for Alex, well in front of the other birds, but closest to my face—otherwise he refused to cooperate. Wart was usually on my hand, lowest of all. He was fine with that.

  The first year at Brandeis was very difficult, and we didn’t achieve a lot, mainly because I was distracted with administrative details and with job applications. Gradually, though, we settled into a productive routine. Because Alex was always butting in with Griffin, we decided to enlist him as one of Griffin’s trainers, as we had attempted at Tucson. This he did enthusiastically. For the first time he was willing to question Griffin, something he had refused to do in Tucson.

  Alex certainly tried to be helpful. At one point we were teaching Griffin the label “seven.” Griffin gets very self-conscious when he can’t produce what we want. His pupils get small. His body language broadcasts his discomfort. Sometimes he stops trying. Alex saw Griffin’s difficulty and kept saying “sss,” “sss,” trying to prompt him. It was endearing, really. We hoped Griffin might learn faster with another Grey as a trainer. In the wild, after all, Greys learn vocalizations from each other. In fact, Griffin did make his first attempts faster after working with Alex, but then he had a more difficult time polishing his pronunciation.

  It was sometimes amusing to figuratively step back and observe the two parrots: Alex saying, “What color?” and Griffin responding, “Blue,” or whatever was appropriate. The fact that Griffin eventually spoke exactly like Alex—same tone, same inflection, same everything—added to the amusement.

  We eased our way back into a work program by reviewing previously mastered tasks—colors, shapes, bigger/smaller comparisons, and so on. But then I embarked on what would turn out to be a remarkable series of studies with Alex in the challenging realm of numbers and mathematical concepts, some of which we had started in Tucson but then dropped. He was about to give new meaning to Woody Allen’s line in Annie Hall: “Must be smart, went to Brandeis.”

  When we started the new series of number trials in the fall of 2003, Alex already knew the numbers one through six. He hadn’t learned them in that order. He learned three and four first, as in “three-corner wood,” for a triangular piece of wood, and “four-corner paper,” for a square piece of paper. He later learned two, then five, then six, and lastly one. We now wanted to find out if Alex truly comprehended the number labels he used. When you ask a child below the age of three, “How many?” when holding up four objects, she will likely answer correctly, “Four.” Ask the same child to give you four marbles from a dish of them, and she will just scoop up a handful and give them to you. As with words, number production does not necessarily mean comprehension.

  Alex’s test was quite straightforward. I would show him a tray on which there would be, for instance, two green keys, four blue keys, and six red keys, then ask, “What color four?” The correct answer, in this case, would be “Blue.” Over a period of several days, Alex answered correctly in eight trials. I was impressed. Smart bird!

  Suddenly he refused to continue with the trials for two weeks. He stared at the ceiling, gave responses of colors or object labels not on the tray, or fixated on an inappropriate label and repeated it endlessly. He preened. He gave every answer except the correct one. He asked for water or various foods, or said, “Wanna go back.”

  Then, for no obvious reason, he ended his strike in a remarkable way. I showed Alex a tray of objects, on which there were now two, three, and six blocks of different colors. I asked, “What color three?”

  He answered very purposefully, “Five.” Something was different about his attention and the tone of his response this time, unlike his previous indifference and inattention.

  I asked again, “What color three?”

  “Five,” he shot back.

  “No, Alex, what color three?” By now I was both puzzled and increasingly impatient. Why say “five”? There is no set of five on the tray.

  “Five,” he said again firmly.

  Let’s turn this around, I thought. I said to Alex, “OK, smarty pants. What color five?”

  Without hesitation he said, “None.”

  I was astonished. Is that what he meant? Years earlier Alex had transferred the term “none” from a “same/different” study—telling us about the absence of a similarity or difference in shape, color, or material for a pair of objects—to a study on relative size. When first shown two objects of different colors but the same size and asked, “What color bigger?” he used “none” to signify that they were the same. He did it on his own, without training. Now, he apparently was using “none” to indicate the absence of a set of five objects, that is, using “none” to mean “zero,” the absence of existence. To make sure this wasn’t a fluke, we did six more trials in which trays lacked a set of one, two, three, four, and so on, and asked “What color one?” “What color two?” and so on. Of six trials, he was correct on five. His one error was to label a color not on the tray. Alex really did seem to have a concept of something like zero.

  Who knows what went through his mind with that first series of “five” responses? He likely had become bored with the tests, which explains why he went on strike. After two weeks it was as if he said to himself, OK, how can I make this interesting? I know—I’ll label what’s not on the tray. Boredom is a powerful emotion for children in school and many adults. It surely isn’t unique to humans.

  Alex’s use of “none” in this context is important for several reasons. First, zero is a highly abstract concept. A label for zero entered Western culture only in the 1600s. Second, Alex’s use of “none” in this case was entirely his invention. We didn’t teach him to do it. He figured it out for himself.

  Shortly before I left Northwestern for Tucson, I had had a conversation with Tufts University philosopher Dan Dennett. He wondered, “What if you asked Alex, ‘What’s green?’ when there is nothing green on the tray? Would he say ‘none’?” I was a little hesitant about pursuing this idea but tried it anyway. I showed a tray of different-colored objects to Alex and asked, “What’s purple?” Nothing purple was on the tray. He looked at me and said, “Want grape.” Grapes are purple.

  He’s outwitting me. He’s not doing what I want him to do. He seems to be doing something clever. But how do I know when he’s being clever and when he’s wrong? This is harder than I supposed. I dropped the idea.

  In the end, it was Alex who thought it up himself. This parrot, with his teeny brain, seems to have come up with a concept that had eluded the great Greek mathematician Euclid of Alexandria. Alex’s use of “none” was as impressive as his leap to putting together the separate phonemes “nnn,” “uh,” and “tuh” to make a complete word. Probably more so. What would he do next?

  In June 2004, the month we completed the number comprehension work, we started a study on addition. I hadn’t planned such a study. Instead, it grew out of Alex’s habit of butting into Griffin’s sessions. We were teaching Griffin the number two by having him listen to two computer-generated clicks and asking, “How many?” Griffin didn’t respond; he hunkered down and looked awkward. I generated two more clicks. “How many, Griffin?”
No answer.

  Then, from the top of his cage, Alex offered, “Four.”

  “Pipe down, Alex,” I said testily. “I’m asking Griffin.” I thought his response was random; after all, I had clicked twice.

  Two more clicks. Still nothing from an increasingly anxious Griffin.

  “Six,” said Alex.

  Six? Did he add up all the clicks and get six?

  Psychologist Sally Boysen had investigated counting abilities and addition in chimps some time earlier, but she had used physical objects rather than sounds. I decided to do the same thing with Alex. Coincidentally, I had just received a one-year fellowship from the Radcliffe Institute, at Harvard, starting that fall. The fellowship was designed to allow me to collaborate with colleagues at Harvard who were studying number concepts in children.

  We tested Alex’s math skills by presenting him with a tray on which were two inverted plastic cups. Under one would be, say, two nuts, and under the other three nuts. We lifted the first cup and said, “Look, Alex,” and then replaced the cup. We did the same with the second cup. We then said, “How many total nuts?” Alex’s accuracy in a series of such tests over the next six months was well above 85 percent. He indeed could add. This performance put Alex on a par with small children and chimpanzees.

  What if there were no nuts under either cup and we asked, “How many total?” Would Alex say “none”? We tried it eight times. On the first four he said nothing. Instead, he looked at me as if he were thinking, Hey, haven’t you forgotten something? He didn’t say “two,” as he might if he had confused the number of cups with what we actually wanted. On the next three trials he said “one.” On the final test he again said nothing. Interestingly, in similar tests, chimps made the same “mistake” of saying “one.”

  Alex’s performance indicated to me that his concept of zero was not as sophisticated as in humans. He didn’t have “none” as the beginning of a number string, going up to six. When he responded “one,” it was as if, like the apes, he knew he had to get as low in the number string as he could. (Later we started to train him on “none corner wood,” for a circle, which he got.) But good as he was, he wasn’t that good—somewhere between Euclid and the seventeenth century, so to speak.