The First Word: The Search for the Origins of Language

by Christine Kenneally

  5. You have something to talk about

  Twenty-two New Caledonian crows are housed in the aviary at Oxford University’s zoology department. The birds inhabit two small rooms, each subdivided by a wall. A hatch allows the crows to pass from a closed room to another where a wire mesh lets in fresh air and sunlight. The birds’ feathers are thick and glossy, and their claws are three inches from the tip to the back. Long sticks are strung up as perches from floor to ceiling in the aviary, and the birds hop constantly from perch to perch, displacing one another like three-dimensional dominoes. If their bodies are at rest, then their heads are in motion, swiveling up, down, and around, either to survey the surroundings or to preen the feathers of the crow next to them. On the island of New Caledonia in the southwest Pacific the birds are plentiful, inhabiting terrain from rain forest to mountain. They are long-lived and have complicated social lives and a wide range of different vocalizations.

  The crows are also shy, so when the graduate student Ben Kenward enters one of the rooms of the aviary, they disappear through the wall hatch. On a table in the birds’ room, Kenward places a test tube that is held sideways in a clamp. Inside the tube are tiny chunks of meat; next to it Kenward lays out small kebab sticks. When the crows return a few minutes after he has left, the most confident one sails straight from perch to table. It picks up one of the skewers in its beak and without hesitation inserts it into the tube, poking and pulling, until it manages to roll out a piece of meat. Meat in mouth, the bird returns to a perch to enjoy its food. The scene replays itself when another bird steps up, this one plying the stick until all the meat is gone. The crows love their sticks. Even after the food is consumed, they hang on to them, fly them around, drop them, pick them up again, and chew strips off them.

  The aviary’s longest-term tenant is Betty. In 2001 Betty was filmed by the lab researcher Alex Weir, then a Ph.D. student, who wanted to see if she or her aviary mate at the time, Abel, would intentionally choose a hooked tool over a straight one to get food in a test in which only the hook would work. He presented the birds with a glass cylinder inside of which was a tiny toy bucket with handle erect. Inside the bucket was meat. The only way for the birds to get the meat was to remove the bucket from the cylinder, and the only way to get the bucket was to use the hooked tool. In one of the very first trials, Abel accidentally knocked the hook away, leaving only the straight tool. Betty quickly hopped up after Abel and in a completely businesslike fashion took the remaining straight piece of wire—a material she’d never seen before—found a suitable place to wedge it, bent it into a fine-looking hook, and then used it to retrieve the bucket and then the meat.

  One fundamental idea shared by many researchers is that in order to evolve language you first have to have something to say—as opposed to, for example, going about your life, developing language out of the blue, and then finding you have a lot to talk about. The search for the origins of language thus includes a search to uncover what ultimately was so worth relaying that our ancestors began to ratchet up their communication skills in order to do so. In trying to work this problem out, it helps to know what kind of thought goes on in the heads of nonlinguistic creatures. For a long time, we have assumed that not much does.

  This conviction comes in part from the human tendency to believe that all of our complex ideas and ways of carving up the world are a result of the fact that we have language. Indeed, it can be hard to imagine otherwise. Everyone reading this book has probably experienced the odd sensation that he or she was momentarily without words, a state that mostly feels like a vacant one. Likewise, few people would claim they could remember what thinking was like before they learned language. The sense that not a lot was going on at that very young age is probably fairly common. Accordingly, we presume that our pre-linguistic ancestors had a pretty simple mental life.

  But how do we account for Betty? She is a completely languageless creature, and yet she’s no stranger to brilliant and rapid thinking. Betty not only saw that a hook was necessary to lift the bucket in order to get to the meat, she didn’t even try the straight wire first to see if it would work. She simply went about creating the tool she needed to reach her goal. Professor Alex Kacelnik, head of the Behavioral Ecology Research Group, likens her act of inspiration to the way a chess grandmaster makes a decision after viewing a given situation on a board and consciously examining only two or three moves out of the range of possible ones. Something is going on in the back of the player’s mind that leads him to reject all but a few choices, said Kacelnik. It’s a creative process that he may experience simply as an aesthetic judgment. In contrast, a computer, Kacelnik explained, must trawl through all permutations of possible chess moves. In this respect Betty is like the human, in that she must have engaged in some kind of planning that involved unconsciously discarding useless strategies and focusing on a successful one in order to spontaneously produce the hook.

  Until Betty’s invention was videotaped, no other animal aside from humans had been shown to build its own tools, substantially altering a basic material to an appropriate design. In fact, it wasn’t so long ago that we believed tool use was uniquely human. This changed when primatologists showed it was a perfectly normal activity for nonhuman primates. Apes fish for termites with sticks, crack nuts with stone and anvil, and process plants for food by whacking the tops of palm trees with large fronds. Scientists announced at the end of 2005 that gorillas in the Republic of the Congo were observed using sticks to test the depth of water before they stepped into it. (Earlier it had been assumed that gorillas were the only great apes that did not use tools at all.) In early 2007, an Iowa State University team announced that chimpanzees in southeast Senegal were observed sharpening sticks and using them as spears to hunt bushbabies. Also recently announced was the finding that dolphins off Australia’s west coast use sponges on their snouts when they probe for food on the seafloor. These animals show that tool use is not only not restricted to linguistic creatures and their close relations, it doesn’t even require two arms and two legs.1

  In 2003 the New Zealand researcher Gavin Hunt announced that New Caledonian crows in the wild create different types of tools from the island’s pandanus leaves. Like humans, said Hunt, the crows sculpt raw material into distinct tool designs with a highly regular shape. In other words, they don’t just pick up a stick that’s most likely to work for a job at hand and simply use that. Their tools are purpose-built. Additionally, the crows pass on their techniques for distinct designs to other crows.

  As the attitude toward animal cognition has changed since 1990, more and more research in the area has shown that, as with Betty, there is some pretty complicated mental processing going on in the heads of animals that do not have human language. For students of the evolution of language, these investigations help to differentiate between the mental platform that our species may have had before we had language and the kinds of thinking we do now that are shaped by the fact that we have language—the cogitation that is more directly part of the language suite.

  Sometimes ideas are turned over in science when someone stumbles across one crucial counterexample—like a gorilla using a tool. In other instances the dismantling of assumptions is a conscious effort. Such is the case with the bird brain’s recent upgrade. For a long time, neuroscientists conceived of bird brains as basic instinct machines. But in 2005 an international group of scientists calling themselves the Avian Brain Nomenclature Consortium announced a new set of terms to describe the bird brain. Their proposal was more than just a substitution of vocabulary; it represented a newly sophisticated understanding of a brain that had until then been sorely underrated.2

  Human brains have a neocortex, a relatively recently evolved sheet of neurons that encases the brain. It is in the interactions between this area and the older segments of the brain that a lot of our processing is done. Birds have no neuron sheet, and this led researchers to assume that few comparisons could be made between avian and human brains. However,
scientists eventually realized that birds possess a neural module that is functionally equivalent to the human neocortex.3

  For a scientist like Irene Pepperberg, who has been engaged in avian behavioral experiments and observations for decades, the consortium’s findings are a vindication: birds are much smarter than they are given credit for.

  Alex, a thirty-year-old African gray parrot, is the most famous resident of Pepperberg’s lab at Brandeis University. He has been filmed by camera crews from all over the world and appeared in stories in major newspapers. Pepperberg has written a book about him, and he was recently featured on Scientific American Frontiers. (“He loved Alan Alda,” said Pepperberg.) For decades Pepperberg has been teaching elements of English to Alex, who has the language capabilities of a two-year-old and the cognitive capacities of a six-year-old. He can explain his needs and wants by using language.

  Alex is about twelve inches from beak to tail, and he weighs only one pound. His companions are ten-year-old Griffin and seven-year-old Arthur. He has a clean white face, soft gray feathers in differing shades that are delicately scalloped around his face, and an intensely red tail. The lab where he lives is fairly small, about 150 square feet, with cinder-block walls painted white to cheer the birds up and newspaper spread all over the floor. The birds sit on perches in front of their cages. Otherwise there is room only for Pepperberg’s small desk and a set of shelves that store the birds’ experimental materials: plastic letters, colored stacking cups, and wooden blocks. For twelve hours the birds sleep in complete darkness, as they would in equatorial Africa. The rest of the day they always have at least one human companion to watch over them and work with them.

  All the birds eat a specially formulated pelleted diet, supplemented with shredded wheat, colored pasta, vegetables, and fruits, and when Pepperberg offered Alex a piece of a muffin, he accepted it with a “Guuurrrrrrreat!” and then “Yummy.” He calls it “banari,” a combination of “banana” and “cherry”; it is his word for “apple,” explained Pepperberg. Alex’s voice is distant and tinny, like a recording from an old-style Victrola. Because Alex has lived all over the country with Pepperberg, his “carrot” sounds Midwestern, while his “shower” is Bostonian.

  Alex can identify by word fifty different objects, seven colors, and five shapes. He comprehends numbers under ten (though he doesn’t count sequentially, he may use counting for quantities above it), and he can make distinctions between things that are the same and things that are different. Once he has learned new words, Pepperberg tests him on them. She fills a tray with blocks—maybe four green and two blue—and asks him, “How many blue?” Alternately, she’ll ask, “What color two?”

  One of Alex’s most recent accomplishments was learning to transfer his concept of “none” from the same-different study to numbers. “Folks have studied the concept of zero in chimpanzees, but never in birds,” Pepperberg explained. “None” is considered a particularly sophisticated concept for humans. “What I’m finding,” she said, “is that Alex can use ‘none,’ without training, to refer to an absence of quantity in some situations. So, if I give him a tray of two blue, four green, and six yellow blocks and ask ‘What color five block?’ he’ll say, ‘None.’” What’s most surprising about the fact that Alex understands what “none” means is that he was trained to use “none” when asked what was the same between a set of objects when in fact nothing was the same. (He was also trained to use “none” when asked what was different between a set of objects when nothing was actually different.) “Alex spontaneously used ‘none’ to denote the absence of difference in size between a pair of objects, and then also spontaneously transferred it to the ‘What color x?’ task. I had a tray of blocks and was asking, ‘What color four?’ and he kept saying ‘Five.’ I was pretty frustrated, and without thinking I finally said, ‘Okay, what color five?’ to which he replied, ‘None.’”

  Alex’s understanding of “none” is more like a child’s than an adult’s: “If I show him that nothing is hidden under a cup and ask him, ‘How many nuts?’ he is like some autistic children or like children around three years of age. He simply refuses to answer. For him there is nothing there to comment on.”

  In a demonstration Pepperberg sat Alex near her desk and showed him various trays of blocks. She got him to identify the colors and amounts. Then she put the testing aside and tried to teach Alex the color white, a new category for him. She held a square piece of paper up to him and asked, “What shape?” “Corners,” said Alex. “Yes, it has corners,” she said. “What color?” As he did with every other object she proffered, Alex beaked the paper with interest. He stalled for a few minutes, then said, “None.” Pepperberg burst out laughing. “Okay, fair enough,” she said. “In your world, this has no color.”

  I asked Pepperberg if she thought the ability we share with Alex to use these categories of number, color, and shape in making sense of the real world results from convergent or direct evolution. Is it possible this ability goes back as far as the remotely distant common ancestor shared by humans and African gray parrots? For now, it’s not clear, she said, but the amount of neurological and neuroanatomical evidence is growing. The abilities might be homologous, but at this stage the possibility is speculation only. “I think we are at an incredibly interesting point where we’re beginning to learn more about both human and animal systems,” she said. “The amount of knowledge we’re going to gain in the next years is going to be exponentially greater than what we’ve learned over the past few years.”

  Alex’s talents demonstrate that not only is the ability to understand and act on general conceptual categories like color and shape and number not human-specific, it’s not specific to apes, or even to mammals. Alex can use those categories in the comprehension of complicated labels, and in the larger meaning created by stringing some of these labels together, like “What color five?” We may have words for these concepts, but it’s clear that you don’t have to have language to understand them and to be able to act on that understanding.

  Other sophisticated forms of cognition include awareness of oneself and the ability to generalize. Gordon G. Gallup started exploring self-awareness in animals in the late 1960s, when he began to look at the way animals make use of reflection. Different animals interact in different ways with mirrors. Some ignore them entirely. Others use mirrors to locate things in space; parrots, for example, can find hidden objects that are visible only by their reflection. Other animals, like monkeys, engage with their reflections as if the reflection were another individual entirely. Gallup was the first to show that chimpanzees recognize that the image they are looking at in a mirror is themselves, an ability that was previously thought to be human-specific.4 When Gallup announced his findings, many researchers were shocked and unsuccessfully tried to disprove them.

  In 2000 Diana Reiss, at Osborn Laboratories of Marine Sciences at the New York Aquarium in Coney Island, and Lori Marino, a senior lecturer in the Neuroscience and Behavioral Biology Program at Emory University, applied the test to dolphins. Like all other whales, dolphins have traveled a radically different evolutionary trajectory from ours. Their closest land relatives are the ungulates, like the hippo. The researchers marked the dolphins with a nontoxic black marker on parts of their bodies that couldn’t be seen without the use of a mirror, and then watched and recorded their behavior at a mirror attached to the outside glass wall of their pool. Once the dolphins had been marked, they swam to the reflective surface and used it to examine the ink marks, showing clear awareness of themselves. In fact, Reiss pointed out, the test didn’t just expose the capacity for self-awareness: it also demonstrated that the dolphins were motivated to view themselves.

  In 2006 it was announced that elephants are able to recognize themselves in mirrors. This work was also conducted by Diana Reiss, with Frans de Waal and Joshua Plotnik. In a similar fashion to the dolphin experiments, the researchers marked three Asian elephants and gave them access to mirrors. They noted that, c
ompared to dolphins, elephants have the advantage of being able to touch most of their body with their trunks. Accordingly, after being marked the elephants spent a significant amount of time in front of the test mirror, repeatedly touching the experimental marks (but, crucially, not touching invisible marks that had also been made).

  Reiss’s work with dolphins has also provided evidence for the ability of nonlinguistic animals to generalize. Dolphins instinctively eat only live fish, so in captivity they must be taught to consume prey that is already dead. Reiss had to cut each fish she fed them into three parts. A dolphin would happily eat the head and the middle, but it would eat the tail only if the fins were cut off. If the dolphin misbehaved during feedings, Reiss gave it a time-out. This involved getting up from where she knelt at the side of the pool, walking back about twenty feet, and looking at the dolphin but not interacting with it in any way for a minute or so. “It let her know something was not right,” explained Reiss. One day Reiss accidentally let an untrimmed tail slip into the dolphin’s food. The dolphin responded by swimming to the opposite side of the pool and then rising out of the water in a vertical position, just looking at Reiss for a minute or so. This feels a lot like a time-out! thought Reiss.

  She decided to test the dolphin, and a few days later she let an uncut fish tail slip through on purpose. The dolphin did the same thing, giving her another time-out. Reiss repeated the experiment three additional times, each with the same result. Dolphins are natural imitators, said Reiss, and imitation is an important part of the ability to learn. They are what Reiss calls “contingency testers,” forever probing and exploring objects, and extremely adept at recognizing and generating patterns. The intentions behind their actions can be as obvious as our own.