Machiavellian Monkeys & Shakespearean
Apes: The Question of Primate Language
by Alex Hawes
Photos: Jessie Cohen/NZP
The notion of animal language sparks furious debate within and between a wide variety of fields: psychology, biology, anthropology, and, as has been noted, philosophy. The linguist Noam Chomsky cites the French philosopher Rene Descartes, who said "the word is the sole sign and certain mark of the presence of thought." Much of the argument over the use of language by animals has ultimately centered on this question: Do animals think? Linguists like Chomsky argue that the basic lack of anything resembling grammar and syntax in animal communication precludes non-human species from higher cognition. Field biologists, however, are quick to counter with evidence of complexities in the way many animal species communicate, not just vocally but with non-verbal signals. All the while, a handful of apes in scientific institutions across the United States are amassing vocabularies of hundreds of words, and using them in ways that challenge our traditional skepticism about the nature of animal thought.
Animal Linguists?
When we think of language, we think of verbs and nouns, tenses and moods. While animals in nature certainly don't have communication systems based on such grammatical ideas, some researchers have suggested that primate "talk" may show evidence of "syntax" and/or "semantics" in a loose sense. "Semantics" refers to the ability of otherwise abstract words to convey meaning, and represent objects and ideas. "Syntax" simply is the ordering of such words along grammatical lines, following rules in word order. For communication to be justifiably considered language, it must contain elements of these linguistic tools, according to many researchers. Furthermore, individuals must use language voluntarily, and communicate information, rather than just emotion. True language, linguists assert, is learned and deliberate, not innate and instinctive.
Scientists have known for years that animals can communicate in often very complex ways. Karl von Frisch, early this century, demonstrated how honeybees, upon finding a source of nectar, can communicate the distance and direction (and even quality) of this source to fellow bees in the hive through a ritualized "waggle-dance" (see What the Buzz is All About). As with humans, animals use more than just vocal signals to convey information or emotion. Animals may use body movement, odor, posture, change in color, facial expression, or other gestures to communicate anger, warning, fear, uncertainty, subservience, willingness to mate, and other messages. Does this alone suggest language use? Eugene Morton, an expert on birds, and National Zoo biologist, says no.
"They don't understand and comprehend what they're signalling," argues Morton. He criticizes certain scientists' preoccupation with the content of animal communication ("the Information Impediment," he labels it). Instead, Morton asserts, animals are simply demonstrating motivation through their signals. "[Researchers] have gone overboard in their interpretations," says Morton. In his book, Animal Talk, co-written by Jake Page, Morton urges "an understanding of animal communication as what it seems to accomplish, not what it symbolizes.'" Morton emphasizes that the use of such communication is innate to animals; they react to stimuli in "pre-programmed" ways, without real thought. Following this argument, while animals may be aware of the ends they seek via signalling, they may not be conscious of the symbolic nature of the means they use to reach those ends. However, many animals, from birds to apes, may have a thing or two to say about that.
Although genetic programming can explain much of animal communication, certain species have shown a remarkable ability to use languages novel to them, either through training or simply through imitation. Parrots, mynahs, and other birds can reproduce literally hundreds of human words. Yet most show no understanding of the words they copy. Some animals, though, seem to have demonstrated a more profound comprehension of human language. Irene Pepperberg of the University of Arizona has worked with an African grey parrot named Alex. Alex apparently is able to "request, refuse, identify, categorize, or quantify" more than 50 items, and label them according to color, shape, and material. He also has learned concepts of sameness and difference. Shown a red triangle and a blue triangle, Alex will say they are same in shape and different in color. Alex uses some syntax, always saying "green wood" not "wood green" for example. However, Pepperberg adds, "Until Alex gives talks for me, it's not really language." Nonetheless, she estimates that grey parrots are on level with a two-and-a-half-year-old human child emotionally, and a five-year-old child cognitively--an important fact to consider before buying one as a pet. "It's like buying a small child," says Pepperberg.
Louis Herman, of the Kewalo Basin Marine Mammal Laboratory at the University of Hawaii, has studied the responses of dolphins to grammatically structured hand signals. Dolphins trained to respond to commands given in syntax-like order responded with up to 95-percent accuracy, depending on the dolphin and the length of sentence given. They can apply this structure to respond to new combinations of signals. The dolphins studied reject "nonsense" sentences that are improperly ordered.
Ronald Schusterman, of the University of California at Santa Cruz, has worked with a female sea lion named Rocky. As with Louis Herman's tests on the dolphins, Schusterman and fellow UC-Santa Cruz biologist Robert Gisiner taught Rocky a language-like system of signals, and then tested how she responded to unfamiliar combinations. Rocky too could respond to syntactic structures properly after grammar training. The authors emphasize the importance of reinforcement in Rocky's coming to understand the structure of the commands given. Rocky's "language" ability, in their opinion, comes from the "general learning abilities" of animals rather than any innate language comprehension.
However, it seems reasonable from an evolutionary point of view that if animals can learn a structured language taught by humans, then they must have evolved this ability as a result of there being a selective value of it in the wild. If we accept this reasoning, what then is the adaptive benefit driving complex communication between animals? Most scientists point to one underlying factor: social living. Roger Fouts, the researcher who has studied the sign language abilities of the chimpanzee Washoe for nearly 30 years, argues that language is primarily a social behavioral mechanism that allows animals to form relationships between one another. "You must form relationships in order to mate," Fouts points out. If you can't do that, then you lose, genetically speaking.
Members of social groups must communicate with each other to relay strategies for foraging or hunting. Information about intentions must be communicated delicately between rival males, lest all encounters result in violence (which often benefits neither one). Communicatory signals may also be used to deceive fellow group members, leading individuals living in social groups to become "natural psychologists" in order to decipher between honest and dishonest signals from others.
However, complex communication does not a language make. Biologists, psychologists, and linguists alike emphasize that language requires rules that guide how signals are combined to relay meaning. Donald R. Griffin, one of the founders of the study of animal cognition, and the author of the book Animal Minds, writes that "rule-governed combinations of words convey a much wider array of meanings than would otherwise be possible if each word were entirely independent, and its relationship to the other words did not convey any additional meaning." Few studies of animal communication in the wild have attempted, and even fewer have succeeded, in showing such a rule-bound system in operation among animal societies. Short of proving the existence of such structured arrays of communicative signals, we cannot comfortably say that a given species of animal uses "language."
When Monkey Talks, People Listen
If any animal group shows the traits that might select for language, it is the primates. Members of most primate species live in close-knit, highly structured societies. Group members forage together, alert each other to dangerous predators, and compete with each other for positions in social hierarchies, and for potential mates. Social communication is essential for survival both in a hostile external environment, and a competitive internal society. While primates are not the only animals with these characteristics, they are most closely related to us, and may give us insight into our own communication. Thus, these "Machiavellian" individuals have served as ideal targets for testing the extent to which animals are able to attain language mastery. Hoping to hear what these close relatives of ours can tell us about the nature of their vocal repertoire, researchers have packed up their sound equipment, and gone out into the field to eavesdrop.
R.L. Garner, a biologist with the then newly created National Zoo, set out in 1892 with recently invented recording equipment to capture the speech of primates in the jungles of West Africa. Safe inside a wire-mesh cage for three and a half months, Garner deciphered about 25 to 30 distinct chimp "words." In addition to studying these apes, he also wrote The Speech of Monkeys, a book that was decades ahead of its time in investigating the vocalizations, as well as general cognitive abilities, of monkeys.
Since then, technology has advanced, and so has researchers' success in understanding what primates are communicating. Dorothy Cheney and Robert Seyfarth, a husband and wife team from the University of Pennsylvania, are in their third decade of listening to and analyzing the vocalizations made by vervet monkeys (Cercopithecus aethiops) in Kenya's Amboseli National Park. Cheney and Seyfarth have spent a great deal of time, it seems, messing with monkeys' minds. Using recordings made of these Amboseli vervets, they have tested how individuals react to playbacks of other vervets' calls.
It is the alarm calls vervets make in reaction to predators that have really gotten the attention of Cheney and Seyfarth. Vervets have different calls for eagles, for leopards (and leopard-like predators), and for snakes, each eliciting different behavioral responses from their group. An eagle alarm call sends vervets scrambling down trees to hide in the brush, while leopard calls send them scurrying up into the trees, while snake calls cause the vervets to stand up on their hind legs and survey the area for the intruder. Later research has shown that vervets aren't alone in having differentiated alarms--so do certain other primates (such as lemurs), as well as ground squirrels, prairie dogs, and starlings.
Vervets also grunt to each other in passing. With the aid of sound spectrographs, Cheney and Seyfarth discovered that what seemed to the naked human ear as only one vocalization actually had quite subtle distinctions. Vervets grunt differently to subordinate individuals, to dominant individuals, to approaching individuals, and to members of other groups.
This evidence of vervet alarm calls and grunts has led scientists to assert excitedly that vervet lingo has some of the trappings of language; their alarms seem to refer to specific objects or ideas, and their grunts are contextually dependent. Equally important is the fact that both the signal sender and the audience exercise voluntary control: Low-ranking individuals sometimes choose not to emit alarms after having noticed a predator, and group members will often ignore alarm calls given by unreliable individuals (especially juveniles). This apparent intentionality on the part of vervets argues against the notion that animals respond automatically to stimuli without any thought.
While Cheney and Seyfarth's ground-breaking work with vervets demonstrates a degree of semantics in the vocalizations of these monkeys, other scientists have investigated the subject of syntax in animal communication ("zoosyntax"). John Robinson, of the New York Zoological Society, has trained his microphones upon wedge-capped capuchin monkeys (Cebus olivaceus) in central Venezuela. These monkeys emit five categories of vocalizations: squaws, chirps, trills, whistles, and screams. Nearly 40 percent of the capuchin calls were part of compound "sentences"--two or more calls in a row. These compound calls seem to have a syntactic quality; chirps are frequently followed by squaws, but never vice versa; trills and whistles often lead into chirps, but rarely the reverse. Capuchin combinations seem to allow them a greater capacity for refined expression. For example, trills communicate uncertainty with another's proximity, while chirps indicate a willingness for physical contact. However, a subordinate individual, desiring contact with a dominant troop member but not wanting to appear too "forward," may combine the trill and chirp to indicate its humble intentions.
John Mitani and Peter Marler of Rockefeller University have
looked to gibbons, relatively small-bodied apes from Southeast
Asia that fall closer to humans on the evolutionary tree than
monkeys. As early morning visitors to the National Zoo (and
neighboring residents of Woodley Park and Mount Pleasant)
probably know, gibbons have elaborate songs audible up to
two-thirds of a mile away, which they sing out each morning
to announce and establish their respective territories. These
versatile ballads consist of distinct elements ("notes"),
which are combined to form individualized songs. Mitani and
Marler recorded the songs of the agile gibbon (Hylobates
agilis), and rearranged their notes, to compare how these
gibbons react to phony and authentic songs. While their gibbon
subjects approached playbacks of rearranged songs, only authentic
song recordings elicited "squeaks" from males, behavior
exhibited in circumstances of territorial contests between
males. Apparently, the sequences of notes were crucial in
conveying meaning.
Although these various findings suggest that animal communication may employ a certain measure of syntax and semantics, few would argue from this evidence that these animals are utilizing true language. What makes language so powerful is its being "open and productive." That is, human speakers can produce sentences that have never before been uttered, but make perfect sense nonetheless. That ability has not yet been seen in the communicatory systems of any other chattering primate. However, these research projects and others show that what for years appeared as simple behavior may not be so. Perhaps there are unimaginable intricacies in animal communication that continue to escape our discovery and translation.
Ape Bilinguists?
The difficulty of decoding primate vocalizations has led some scientists to try a novel approach. Like a stereotypical American tourist abroad, researchers have decided that if you can't understand what they're saying, make them speak a language you know. Being our closest relations in the animal world--they share up to 98 percent of our DNA--and having large brains relative to other primates, great apes (chimpanzees, bonobos, gorillas, and orang utans) have been the prime focus of scientists interested in exploring the limits of animal language.
Originally, researchers thought they might be able to teach some unsuspecting apes spoken language. The most successful attempt was made in the 1950s by Cathy and Keith Hayes of the Yerkes Laboratory of Primate Biology in Atlanta. However, after several years of having her lips pulled around and her mouth squished and rearranged, their juvenile chimp, Vicki, could only muster a few words ("Mama," "Papa," "cup," and "up"). Chimps and other non-human primates simply don't seem to have the necessary anatomical apparati to produce human speech (see First Words).
However, this failure did not deter others from seeking alternate strategies to teach apes language. Two psychologists from the University of Nevada, Beatrix and Allen Gardner, in 1966 became the first researchers to start teaching chimps sign language. Their first student, Washoe, has since been transferred to the tutelage of Roger and Deborah Fouts of Central Washington University. Washoe, nearing age 30, has now used more than 1,000 taught signs in American Sign Language ("Ameslan"), and since instruction of new signs stopped in 1970, she has learned many more by her own observation. Washoe has mastered more than 200 of these signs under the Fouts' "reliable" criteria; a sign is reliable if three different observers see a chimp use a given sign in the correct context on each of fifteen consecutive days.
Washoe has surprised the Gardners and Foutses, and the scientific world, not only by learning hundreds of words, and effectively using them in complex "sentences," but by doing this usually without prompting, in a way eerily similar to the way we ourselves use language. Even with such an extensive vocabulary, Washoe occasionally finds the words she knows lacking, and will simply invent new words when she doesn't know an appropriate one. Such innovations include labeling a fur coat as "hair coat" and a candy bar as "candy banana." A peer chimp, Lucy, came up with "cry hurt food" when she didn't know the word for radish.
According to Roger Fouts, Washoe can communicate empathy and sorrow. She signs "hurt" and "sorry" in addition to offering physical comfort when others seem upset. "It shouldn't come as a surprise," says Fouts, "that apes have emotions." Chimpanzees and other apes show significant development of the limbic region of the brain, the area thought to control emotional responses. It would be bizarre if other animals didn't have emotions, Fouts adds.
However, we shouldn't expect chimps to communicate all of their emotions through semantic language; we humans certainly don't. Gesture, expression, tone, and other non-semantic features often communicate much more than just words alone. In the wild, Fouts suggests, chimps may take these subtler forms of "language" to the extreme. When other chimps fail to respond to Washoe's signed pleas, she knows to resort to physical communication. Once Washoe spotted a snake near a fellow chimp. She frantically signed "Come hug, come hug" to the chimp, but to no avail. Washoe finally rushed to her friend's side, and led the chimp away from danger by hand.
The lab chimps also are able to express another area of their psyche: their curiosity and inquisitiveness. While Washoe only rarely asks questions using proper interrogative pronoun signs (what, where, etc.), she indicates questioning by gestural signals, such as raising her eyebrows, and holding the sign in its position for a long time. This usage is not improper; many deaf humans use similar gestures to make signed sentences into questions.
Perhaps the most impressive use of Ameslan by the chimps, according to Fouts, is their use of "private signing." Hidden cameras have recorded the chimpanzees signing things to themselves, even in instances of imaginary play. On one occasion, a juvenile chimp, Loulis, grabbed Washoe's magazine and ran off. Washoe then signed to herself, "Bad, bad, bad!" Fouts notes that chimps will occasionally sign to themselves "Gotta be quiet." Such uses counter arguments that apes only use learned language when "cued" by human trainers, and only for some sort of reward.
Washoe and her companions have fulfilled another requirement of language--that of teaching and learning. The young chimp Loulis learned to use Ameslan by observing others, and from others (especially Washoe) molding his hands to form proper signs. Loulis even abides by many of the grammatical rules of Ameslan, without any formal instruction. However, before proclaiming true language use by Washoe and Company, we mustn't forget the sine qua non of language: syntax. Fouts states that Washoe does know certain (but not all) fundamental rules of syntax, such as the difference between "Roger tickle Washoe" and "Washoe tickle Roger." Imperfect syntax should not negate the importance of this and other ape language studies though. As Fouts argues, language in apes is not black and white; we should strive instead to understand the degree to which apes use language, both in the lab and in the wild.
Certain researchers have looked beyond sign language as the lingua franca between humans and chimps. Georgia State's Duane Rumbaugh and Sue Savage-Rumbaugh, originally of the Yerkes Lab in Atlanta, have devised an elaborate automated machine that parcels out goodies to chimps, and shows them slides and video tapes. To operate the machine, and ask it for treats, chimps punch "lexigrams" (symbols composed of various shapes and lines) on a large keyboard. Each lexigram represents a different word. Two closely related species, chimpanzees (Pan troglodytes) and, later, bonobos, or "pygmy" chimps (Pan paniscus), have been pupils of this invented language, which lab workers have labelled "Yerkish." (see Chipping Away)
Two-way communication is possible through the machine. In fact, two chimps, Sherman and Austin, learned to use the machine to ask each other for tools the other possessed in order to open containers of food. Sherman and Austin, and the first student of "Yerkish," Lana, were tested for comprehension of categories of objects. Shown food items and tools, and the lexigrams representing each one, Sherman and Austin were able to arrange new objects into the proper general categories of "food" and "tool" perfectly (except for one instance when Sherman labeled a sponge "food," probably because he occasionally nibbles on them). Poor Lana, however, had seemingly not acquired these abstract concepts.
Recently bonobos have shown particular success with Yerkish, perhaps because they forage in larger parties on average than chimpanzees, where social communication between each individual is necessary to maintain group cohesion. Instead of simply responding to questions and commands posed by the machine, bonobos are now using their new language to express themselves spontaneously. Kanzi, a male bonobo, learned Yerkish entirely by watching human trainers and his own mother use the system. Now Kanzi can understand, Savage-Rumbaugh estimates, between two and three thousand spoken English words, as well as all of the 250-plus keys on the Yerkish keyboard. Kanzi is apparently trilingual, if you count bonobish.
Kanzi and friends seem to understand many of the complexities of human language. When presented with a tomato, a bonobo will respond to the command "Go and get the tomato that is inside the microwave" by correctly retrieving the distant tomato rather than the one that is visible. Bonobos use Yerkish to discuss things they will do hours in the future, as well as things they did yesterday. They know, for example, the word "later," and use this term to convey activities they plan on doing at another time.
They also comprehend the possibility of deceit in language. One bonobo was sitting with Savage-Rumbaugh and another trainer, Liz, when Savage-Rumbaugh placed some M&Ms in a bag, in plain view of the other two. When Liz left the room, Savage-Rumbaugh took out the M&Ms and replaced them with a bug. The bonobo signalled "bad" in light of her trainer's chicanery. When Liz came back and looked through the bag for the candy, Savage-Rumbaugh asked the bonobo what Liz was looking for. It signalled "M&M," despite the fact that they were not in the bag. Human children up to the age of four or five would likely answer instead "bug," not being able to separate the knowledge of the duped assistant from their own knowledge.
As for the dreaded "s" word--syntax--the chimps taught Yerkish apparently understand spoken English syntax perfectly well (as the tomato example shows). In his signed "utterances," Kanzi employs what Savage-Rumbaugh labels "proto-grammar." Like Washoe, he seems to have rules he has worked out on his own. However, Savage-Rumbaugh maintains that the ability to communicate complex ideas and messages, rather than the specific order of the messages' component parts, is what indicates a profound comprehension of language use. It is "extraordinarily anthropocentric," says Savage-Rumbaugh, to expect apes to use language in exactly the same way we do.
One initial skeptic of these studies was Herb Terrace of Columbia University. He trained a young chimp, Nim, in Ameslan. He then performed an exhaustive study of Nim's use of word order. Nim rarely used any regular pattern in combining words, and when he did, he seemed to be merely copying the previous statement of the trainer. Most researchers in primate language have not taken this critique to heart; first of all, Nim was only a juvenile, and even juvenile human children have trouble with word order. Also, Nim had dozens of different human trainers; this instability may have impeded his progress in acquiring proper grammar.
Moreover, it is the way chimpanzees use language as a whole, rather than their use of specific syntactic relations, that most impresses scientists. Their ability to manipulate language for the purposes of deception may most effectively prove that they are using language in thoughtful ways. Anecdotal evidence from field studies has demonstrated duplicity in primate behavior in the wild. Jane Goodall has commented on numerous instances when individual chimps, foraging in a larger group, will come across a source of food, not give a food-location call, and later circle back when the others are out of sight to eat it up.
Working with language-trained chimps, David Premack, recently retired from the University of Pennsylvania, performed an experiment in which chimps were instructed to indicate to two different trainers where food was hidden in their enclosures. One trainer would always share the food with the chimps, while the other one would always eat all the food himself. Gradually, the chimps learned to withhold accurate information from the "greedy" trainer, and would often mislead him or her with their signals. Roger Fouts notes that, while discerning chimp intentions is essentially impossible (as it is with humans), chimps do demonstrate as wide a spectrum of deceit as is found in human behavior. (Perhaps they will soon be given roles on "Melrose Place.")
"Fireworks Child"
Ever since King Kong scaled the Empire State Building, the public's view of gorillas has been unfairly tainted. But gorillas (Gorilla gorilla) are not the monsters we make them out to be--they are, rather, "peace-loving vegetarians" according to Francine Patterson of the Woodside, California-based Gorilla Foundation. Patterson's acquisition of a baby gorilla named Hanabi-Ko (Japanese for "fireworks child"), or "Koko" for short, has done much to turn back the tide of gorilla bad press. Koko has since used American Sign Language to show the world her sensitive, humorous, and often stubborn personality.
Patterson has had a hard time keeping track of Koko's vocabulary--the gorilla has used many hundreds of different signs (matching and then surpassing the chimps' performance) with varying frequency since Project Koko's inception in 1972. Patterson stresses, however, that having a huge vocabulary isn't all that necessary; fluent human Ameslan users can get by easily with 500 to 1,000 signs, compounded by gesture and expression. An English dictionary may have as many as 200,000 words, while an Ameslan version has only 2,000. English may be unnecessarily wordy, verbose, prolix...(you get the point). A study of graduate-level essays at Stanford University found that students used on average only 300 different words per essay--and graduate students are renowned for being excessively prolix.
Although getting Koko to use her knowledge of sign-language has proven difficult at times, her stubborness has also provided insight into her grasp of Ameslan. Once when Patterson was drilling Koko on anatomical terminology, the gorilla signed, "Think eye ear eye nose boring." One undoubtedly thinks of human schoolchildren when reading that statement.
Like Washoe, Koko has created new terms for unnamed objects: "elephant baby" for a Pinnochio doll; "bottle match" for a cigarette lighter; and "eye hat" for a mask, to name a few. Koko has displayed her comprehension of the representative nature of words additionally through "overgeneralizations"--a term for when one learned sign takes on new meanings. Patterson notes that, having learned "straw," Koko applied this sign as a name for plastic tubes, hoses, and even cigarettes. "Tree" similarly came to signify asparagus and scallions.
Responding to spoken English as well as Ameslan, Koko also can understand the relationship between spoken and signed words. This link is most evident in her interest in rhyme. Patterson reports that Koko can use Ameslan signs to "rhyme" words that would rhyme if spoken aloud in English. She once signed "flower stink fruit pink," and when Patterson remarked about her rhyme, Koko continued, "Love meat sweet." Koko's a poet, and apparently she knows it.
Koko also is a comedian, according to Patterson. One target of her humor, and ridicule, is a researcher named Ron. When asked, "Who Ron," Koko responded, "Stupid devil." When later asked "What is funny?" Koko demonstrated her mastery of sarcasm: She signed "Koko love Ron" and gave him a kiss, grinning the whole time.
This gorilla has also mastered a less honorable technique: lying. Once, when playing roughly, Koko nipped a trainer, and then signed "Not teeth." Reprimanded by the trainer, "Koko you lied," Koko meekly admitted "Bad again Koko." On another occasion, a trainer discovered Koko eating a crayon. When the trainer cried, "You're eating crayon!" Koko made the sign for "lip" and pretended to be applying lipstick.
Readers may recall a traumatic episode in Koko's life, when her pet kitten, All Ball, escaped and was run over by a car. Koko openly expressed her grief by signing "Sad" repeatedly. Moreover, she seems to understand concepts of life and death surprisingly well. When asked where gorillas come from, she pointed to her abdomen, and when asked when gorillas die, she signed "Trouble old." Where do they go once dead?: "Comfortable hole." It shouldn't surprise you too much now that, on various intelligence tests given to Koko over the years, Patterson reports that Koko has recorded I.Q. scores from 70 to 95 (not too far below the human average of 100), and that's including deductions for answering "flower" for "What is good to eat?" and "tree" for "Where do you go to sleep?"
A Fourth Party is Heard From...
It was only a matter of time before scientists gave the fourth
and final great ape,
the orang utan (Pongo pygmaeus), its fair shot at language
mastery. Orangs actually have fared as well or better than
chimps and gorillas on various cognitive tests. The first
attempt to teach an orang a human language was made by Lyn
Miles, of the University of Tennessee. Miles taught Ameslan
to a male orang named Chantek. Learning his first word in
less than a month, Chantek went on to learn and use more than
150 signs, and can form simple sentences. Chantek too can
create novel words, labeling contact lens solution "eye
drink." Deceit? Chantek shows that too--lying an average
of three times a week, calculates Miles.
Chantek has also invented some of his own grammatical rules. If referring to an action in conjunction with an object present ("play ball" for example), Chantek places the action first, whereas if the object is not visible, he places the object sign ahead of the verb ("Ball, play").
In the wild, orang utans lead mostly solitary lives. Stable groups are formed only between a female and her young. This apparent lack of sociality has caused scientists to wonder whether the orangs' perceived language ability really derives from social interactions between individuals, as has been theorized for other apes. Miles suggests another possibility, "the extractive foraging origin." Orang utans are fruit-eaters, and because of this food source's scarcity, they must be able to remember over the course of several years the location of various reliable fruiting trees within a vast territory, and routes between such sources.
Another cognitive capability orangs show is expertise with tools--the most complex tool use witnessed in captivity by a great ape. Chantek, observes Miles, can put together a chain of at least 22 different tool tasks to attain a goal (for example, using a key to open a box to get wire cutters to open a cage to get a hammer, for up to 22 commands). It was only recently that researchers discovered intricate tool ability by orangs in the wild. The general cognitive ability of orangs, seen in their foraging strategies and their tool use, may then underlie their capacity to learn human language.
We can't completely write off the social living theory though, cautions Miles. "Orangs may have been more social in the not-too-distant past, and have only recently resorted to solitary living." Why then would they have "gone solo?" Environmental degradation, Miles theorizes. Before human disturbance drove orang utans from most of their original habitat across southeast Asia into only a few pockets of Borneo and Sumatra, they may have lived in social groups more similar to those of chimps and gorillas. However, with a less than ideal environment, group living may have gradually disappeared, as competition within groups prevented individuals from finding enough food for themselves to survive.
Towers to Babble
Politicians are no longer the only ones struggling to communicate their ideas inside the Beltway these days. Orang utans at the National Zoo have now entered the fray in the heated debate over apes' language capabilities. Under NZP biologist Rob Shumaker, the Orang utan Language Project (OLP), three years in the making, was unveiled October 27 at the opening of the Zoo's newest exhibit, Think Tank (see Mindful of Thought). National Zoo orangs cross overhead between the Great Ape House and the new Think Tank building via tall towers and long vinyl-covered cables ("the O-line") to study their new language. The OLP picks up where the other investigations have left off: It not only involves teaching orangs a language composed of abstract symbols, but focuses on how their knowledge of this language will be transferred, and possibly taught, to other orangs not instructed by trainers.
The OLP's language of choice is a computer-based form, designed by Shumaker, and centered on abstract symbols that each represent a word (see illustration below). Words are broken down into seven categories: food, activities (verbs), adjectives, non-food objects, human names, orang names, and numbers. Each category has a distinct exterior shape (such as diamond, circle, rectangle) and words within each category are specified by interior markings. For example, food items are characterized by rectangles: The symbol for "apple" is a rectangle with a straight line and a dot in the center, while "banana" is conveyed by a rectangle with a horizontal wavy line. Verb activities, however, are all defined by diamond shapes forming the symbols' exterior. Seventy words have been developed in the system so far, of which the orangs have only been shown a few. However, there is room for an infinite amount of further additions.
The orangs, in their "smart room" enclosure, have
two video monitors--the "response monitor," a touch-frame
screen, and the "stimulus monitor," over which visuals
are broadcast. Among other possibilities, the apes are posed
questions via the stimulus monitor, and they respond by pointing
to possible choices on the response monitor. Stimuli and responses,
however, can take all forms; a photograph, painting, or iconic
drawing of an apple may appear, and the subject tries to match
that image with the corresponding "apple" symbol
from a choice of several on the response screen. Conversely,
the stimulus screen may show a language symbol signifying
"apple," and the orang will pick from photographs
of fruit and/or other objects displayed on the response screen,
or even displayed physically by a trainer. From there, lessons
can get more and more complex depending on the progress of
the orangs. Just as the questions posed will ultimately be
limitless, so too will be the orang utans' possible responses.
We can only wait and see, literally, what they will have to
say.
Most intriguing from a research point of view is the OLP's focus on observational learning. Of the three pairs of orangs involved (Azy and Tucker; Bonnie and Kiko; Indah and Iris), only the dominant (and usually older) individual in each pair is actively instructed in the language by trainers. The subordinate ones can only pick up on the symbols and the use of the language system through observation, or through their partners' teachings. However, on a regular basis each individual is tested on this knowledge and use of the language. Researchers quantify and compare the success of instructed and non-instructed individuals. This study promises to offer us interesting clues into how tool use, communication, and other behavior may be transmitted culturally between primates in the wild.
Perhaps the most exciting part of all this, from a zoogoer's perspective, is that the action takes place in plain view of the visiting public. Three mounted monitors display the same video "feed" that the orang is working with, as well as live footage from cameras focused on different areas of the orangs' enclosures, and on the O-line. Both orang students and human visitors can watch the action of orangs frolicking elsewhere whenever either gets bored.
Shumaker strongly emphasizes that the machine is simply a means through which the orangs can communicate. Shumaker himself controls, through a computer hooked into the language program, the questions posed to the apes. He hopes that, when the orangs become more proficient with the use of the system, he will be able to build upon the intricate relationships he already has with his primate pupils. Shumaker mentions one of the project's hopes and goals down the road: to allow Zoo visitors to sit down in front of the computer, and not only try their hand at the increasingly difficult puzzles challenging the orangs, but in fact to use the language system to communicate directly with the orangs themselves. That might represent the ultimate experience a zoo can offer.
Planet of the Apes (Too)
"We are looking at the ability of apes to use abstract symbols in a concrete way," explains Shumaker. The fact that apes possess the ability to think abstractly may represent the most significant finding of the various ape language projects. While orang utans and other primates may not have developed a system of communication in the wild complex enough to be labeled as a language, they may, however, be thinking about their social and physical world in abstract ways. As Shumaker cautions, we should not be overly obsessive about the specific performances of apes learning new languages, for the projects are only "applying [ape] communication in a way more meaningful to us."
Linguists have been quick to deny any language ability in wild or tutored apes, shouting cries of "Syntax!" and "Semantics!" left and right. They have ignored, however, the proven intricacy and power of non-verbal communication, such as gesture and expression, in humans as well as in our primate brethren. Although there may be a gap between the structured way we use language and the way primates combine subtle non-vocal cues with seemingly representative signals, this divide has been all but bridged by great ape "students" once given the chance to learn. Perhaps the gap between wild primates and ourselves does not reflect primate stupidity, but "just gaps in [researchers'] scholarship," as Roger Fouts suggests. One thing that is certain is that these studies have caused us, as Lyn Miles puts it, "to wonder how far boundaries of human uniqueness really go."
Miles, who says she has always been interested in the possibility of life on other planets, adds that "There is extraterrestrial intelligence--right here on Earth, in the orang utans." And the gorillas, and chimps, and bonobos, and who knows how many other species. "The most important point of all this," she continues, "is that we're not alone. We share this planet with other intelligent animals." Unfortunately, as mankind threatens the habitat and ultimate survival of these brilliant creatures, we may never understand the full complexity and potential of their communication--and their minds. Hopefully, humans will learn to show the animal world what it has clearly shown us: thoughtfulness.
(ZooGoer 24(6) 1995. Copyright 1995 Alex Hawes. All rights reserved.)
First Words (evolution of human speech)
Chipping Away (tool use)
Mindful of Thought (National Zoo's Think Tank)
