Synopses of chapters from Marc Hauser's book "Wild Minds"

Learning

Different modes of learning.

• (1) One way of learning is the way the Japanese macaque learned about the method of separating wheat from sand. The macaque sees some wheat floating on the water and then deduces that this is a solution to a particular problem of foraging. Social learning is not a factor in this kind of deduction.
• (2) Another situation is the macaque discovers the technique by luck.
• (3) Another mode of learning is social. Perhaps another macaque watches the first one washing the wheat; then after the first retires into a tree, the second macaque reproduces the precise sequence of actions of the first, and thereby acquires the skill. the skill is learned by imitation.
• (4) Finally, another mode of social learning is didactic. A mother macaque takes its young down to the water and instructs the young macaque by holding its hands during the operation. In this way the young macaque acquires the skill.
• (5) Perhaps yet another kind of didacticism is one in which the skill takes a long period of persistent training in order to acquire a skill.

Although all animals learn, generally speaking, animals have different ways of learning. Some use trial-and-error. Some use imitation. Some use teaching. Deciding which tools are available to which species can be difficult.

Birdsong vs. Human language. How does a bird learn to sing as opposed to how a human learns to speak? Both are limited to the kinds of experiences they can attend to. As well, both have a range of possible expression. But both have the instinct to learn, and this instinct makes their communication learnable. Both birds and humans have clear developmental stages thoughout their process of learning to communicate. As a result, both have styles of communicating that might be referred to as dialects.

Some songbirds reproduce the same song every mating season. Others will create new variants for each new season, and then there are the mimicking birds who not only employ birdsong but will also imitate sounds in their local environment.

Birds are hardwired to imitate certain kinds of notes/syllables, but the human capacity for mimicking ranges beyond just that of vocalization. Humans copy facial gestures, hand gestures, novel sounds, etc. Furthermore, they can often decide what another person's intentions are and imitate what the other person's intentions were.

Black Rats and Guppies. There are many instances of one animal acting like another animal. There is a social learning going on here. Black rats learn the skill of stripping the seeds from pinecones from others. It also can deduce how to strip the pinecone if presented with a partially stripped pinecone. Similarly a female guppy can learn to choose its male mate based on the choices that previous females have made. It is heavily swayed by the previous choices of other guppies. The older and presumably wiser, a female is has preferential influence over a younger and more naive female. Still the question remains as to whether when an animal copies a skill it stores a mental record of the demonstrator's actions? And if they do, do they also have some idea of what the original performer's intentions or desires were?

Orangutan Firestarter. In the rain forests of Indonesia are a particular population of orangutans which have been raised among humans and then reintroduced into the wild. These "rehabilitants" have been observed to exhibit what is called spontaneous imitation by researchers Russon and Galdikas in the early 90s. Some of these orangutans reproduce an exact replica of human action, like using a toothbrush or insect repellent, without provocation. One of the orangutans, Supinah, even started a fire with kerosene and was observed to fan the burning embers with a trash can lid. No one seemingly taught Supinah this, so how did she acquire this ability? Is this spontaneous imitation? Some say yes. It is novel behavior based on the observation of humans. Others say that the copying of this behavior may not be goal-oriented. It is unclear if the orangutan understood the steps involved in copying the action or if the fire was started rather serendipitously-presented with the right materials, the orangutan stumbled on the makinng of fire. Some speculate that the orangutan may have copied all the requisite steps of the previous actor only to be astonished by the flames that broke out. The question of understanding its intent is still largely up for grabs.

Human newborns. Human newborns are hardwired to copy facial expressions shortly after birth. One experiment involved an infant and two adults. The infant was presented with two adults, one who put his tongue to the side of his mouth, the other protruded its lips. Then, each returned expressionless. In both cases the infant remembered the movement associated with each adult. When the first arrived, it moved its tongue to the side of the mouth. When the second entered the room expressionless, the infant protruded its lips.

Chimpanzees vs. Human Infants. Mike Tomasello conducted an experiment between human two-year-olds and chimps. He modeled using a rake to get food that was out of reach. Both the two-year-olds and the chimpanzees watched. However, for each species, a group watched a demonstrator using a somewhat less effective technique for getting the food. the goal of the experiment was to see if each would imitate the actions of the modeler or the goals of the modeler.

Both used the rake to get the food, but there was a difference in their technique. The children used the precise technique of the modeler whereas the chimps used a more random approach, whatever worked for them. They were less imitative. Monkeys used in the same experiment eventually retrieved the food but only after many trials. The experiment suggested that chimps did not imitate the exact gestures of the adult modeler.

Tomasello postulated that maybe the chimps had the capacity for imitation but lacked the proper conditioning to trigger this capacity. He showed that about 40% of chimps imitated a human modeled action on an object. The rate for 2-year-olds was slightly higher. Naturally reared chimps failed to imitate at all. This showed that enculturation was an important factor in the chimps' imitation of human behavior.

However, another researcher, Andrew Whiten disputed Tomasello's findings. He fond that 3 out of 4 naturally reared chimps were able to remove fruit from an artificial plexiglass trap after demonstrations were performed by a human. But not all aspects of the opening technique were copied exactly by the chimps. Whiten concluded that chimps do imitate even if they are not enculturated.

Pedagogy. The skilled use of a tool often takes many years, but among chimps in Western Africa, Christophe Boesch observed chimps wh use a hammer an anvil technique to crack palm nuts. He observed that chimps start to crack these nuts at about the age of three but do not become fully competent until many years later. Mothers usually allow their young to watch them crack nuts, and they may leave the materials around so that the young chimps may experiment with the hammer and anvil technique. This is laissez-faire teaching at best though.

However there are two cases of mother chimps correcting their young's attempts at cracking the palm nuts by moving the hands of their young during the technique. This suggests chimps have the potential to be pedagogical but rarely do this.

Interestingly, the cheetah on the Serengeti takes a much more proactive approach. The females bring their cubs along in distinct steps with regard to hunting. At three months of age the female begins to bring back prey maimed (but not killed). The cubs have their first opportunity to pounce on their prey and knock it down. When the cubs are a few months older, the females release about 1/3 of their prey in the presence of the cubs and allow the cubs to kill the prey. At eight months the prey is held down while the cubs rip apart the flesh. By ten months the femlae releases almost 1/2 of her prey in the presence of the cubs and allows them to finish the kill. This teaching technique provides opportunity, but does not guide the cubs through process which is much different than much of human pedagogy which is much more tutorial.

Elephant Shaman. On the Masai Mara Game Reserve in Kenya Holly Dublin observed an extremely pregnant elephant moving very quickly. the elephant walked several miles outside of her normal range when she stopped at a tree and proceeded to strip the tree of its leaves. Her daughters watched. The next day the pregnant elelphant gave birth. Dublin then went to inquire what kind of tree it was the elephant stripped. When she arrived at the local village, the villagers told her that the tree was used to induce labor during difficult deliveries. This zoopharmacognosy is puzzling. How is this knowledge passed on? Is there some sort of collective knowledge on the part of female elephants or is it something learned from one's mother (even though one's mother might not have had a difficult delivery?)

Deception

A competitive foraging test using a follower and a leader. A submissive chimpanzee watches an experimenter bait one of several sites with a banana. A second, more dominant, chimp is allowed to watch the submissive chimp, but it cannot see what is going on outside. When the chimpanzees are let outside, the submissive chimp races for the banana. The more dominant chimp can only watch since it does not know where the banana is hidden.

After a dozen or so trials, the dominant chimp starts to catch on. The dominant chimp begins to shadow the submissive one, and as a result, the dominant one gets the banana as soon as the submissive one indicates the direction where the banana is.

However, the submissive chimp is smart and greedy and it catches on to the dominant animal. Once the submissive chimp determines that the dominant one is possibly looking away [the experiment was not conducted to measure whether the submissive chimp actually takes the perspective of the dominant chimp into consideration which could have been done by putting a blindfold or a bag over the dominant chimp's head] it makes a beeline for the banana. But after a while the dominant chimp catches on to even this and it tracks the submissive chimp more closely.

Then the experimenter makes it more interesting. He places some lettuce at one site as well as a banana which chimps usually prefer. The submissive chimp will now lead the dominant chimp to the lettuce site and then head for the banana site where the submissive chimp can eat the banana in peace.

Are we really to conclude that the chimp is deceiving with regard to the dominant chimp's beliefs and desires? The submissive chimp must be aware that the dominant prefers bananas to lettuce [shows recognition of intent--a mental state].

This has only been observed once so it could be that the lettuce was closer than the banana so the submissive chimp went there first. It could be that the submissive chimp prefers lettuce. Perhaps the submissive chimp saw some early indication that the dominant one would discover the banana, so it went to the lettuce first.

The Two Trainers. David Premack and Guy Woodruff designed an exeriment with chimpanzees that would show whether they are able to reflect on another's intentions, consider their own desires and beliefs and then use this to decieve someone. They wanted to see if the chimps would share information about where a piece of food was hidden with a trainer who was kind and cooperative and withhold this information with a mean and uncooperative trainer who would eat the food as soon as its location was disclosed.

Three of four chimps did not discrminate between the two trainers. They indicated where the food was regardless of how they were treated. The fourth chimp, however, indicated where the food was to the kind trainer but did not indicate where the food was to the mean trainer.

Several months later the experiment was repeated. At that time two of the four chimps withheld information about where the food was hidden, even indicating the wrong area when prompted. However, this behavior was seen only after several hundred trials.

In the third phase of the experiment the roles were reversed. The chimps were supposed to follow the lead of the the trainers. All four chimps initially followed the instructions of both trainers. After about 40 trials one chimp figured out that the mean trainer did not provide reliable information. One chimp never caught on and the other two took about 150 trials.

After one year all three phases were repeated and all had improved. They all either indicated where the food was or followed the instructions of the kind trainer but did not do so for the mean trainer.

What had the chimps learned? They seemed to have learned to discriminate between two trainers because of each trainer's different means of reinforcing the chimps' behavior.

Measuring the Gaze. Following the eyes of another individual is a good way for children to ascertain knowledge about their environments. Children are adept at this. Similarly, chimps and some monkeys will look in the same direction as someone else.

When an actor and a chimp stand on opposite sides of a tube, covered at the chimp's end, open at the actor's end, and the actor places some food in the tube, then stares at it, the chimp tends to follow the actor's gaze in order to find the food. From this it would appear that chimpanzees have some inkling of visual perspective, but this behavior can be explained another way. First, the chimp may be using the head orientation of the actor as a signal about where to search.

If chimps use head orientation over eye orientation to determine where the food is, this demonstrates that chimps do not understand the relationship between seeing and knowing.

However, there is conflicting evidence about whether chimps can track eye movement separately from head movement.

At the neurophysiological level, there is some evidence that neurons in the brains of macaques fire when the macaque is looking at something (independent from the way the head is oriented) as opposed to other neurons which fire when the head is oriented in a particular direction (independent of the direction of looking). Several years later the chimps were tested again but showed no ability to differentiate between the two.

On the other hand, Povinelli and Eddy have shown that chimps are just as willing to beg for food from an actor who can't see a source of food (the actor is either oriented in the opposite direction of the food, is blindfolded or is wearing a bag over the head) as they are to beg for food from an actor who can see a source of food. They make no differentiation with regard to which one can see the food or not.

General consensus. The general consensus in the field is that primates lack beliefs, desires and intentions and therefore anything resembling what is conceived of as mind. But the experimental data is still relatively inconclusive. For many, the notion of whether animals possess a mind should be determined by the resemblance of human language to animal communication.

Communication

Philospher Ludwig Wittgentstein argued: even if lions could speak, we wouldn't understand them. Can humans translate the utterances of animals into something meaningful?

Let's consider a classic problem in the philosophy of language and mind. Let's say you are visiting a foreign country where you are the first person to come into contact with a particular group of people. A rabbit runs by. A man from the group says "gavagai". Our first intuition is to conclude that the word means rabbit, but we may be jumping the gun. The utterance could mean anything. This example illustrates the fundamental difficulty of understanding the meaning of an utterance from the behavior associated with it.

Children's babble, Children typically tend to produce their first words at about 10 to 11 months. These seem to always refer to solid, bounded objects. Their first utterances are object-oriented. When the child hears a new word in the presence of a familiar object they already know, they tend to assume that the new word is an attribute of the object. Shortly thereafter they begin to produce word strings like Me baby, me want, me eat.

Do animals have sounds that correspond to words? If so, are they referring to whole objects, parts of objects, events, etc.? What motivates them to communicate? Do they intend to convey accurate information? Do they try to alter the beliefs of another animal (see deception above)?

Claims of identity. All vocal species seem to produce signals that encode their identity. Acoustic analyses of these signals allow scientists to make these claims. Yet animals tend to include these acoustic signatures automatically and subconsciously. Therefore, there is no indication of planning or calcualation on their part. However, generally another animal can make out information such as, species, group, sex, and individual identity, from these utterances. These utterances have communicative value.

Yet some young animals emit singular kinds of automatic vocalizations when they are under emotional stress, like when they have lost contact with their mother. What could be the underlying motive for these distinctive calls? Does the parent recognize a particular state of unhappiness in the offspring or is it just reacting to a stimulus which it has been conditioned to understand as leading to particular consequence?

Segmentation. One of the most difficult things about learning a foreign language is determining the boundaries of words--when one word begins and another one stops. This is called segmentation. This same difficulty is encountered with animal communication with the additional problem that we, as humans, are limited by our own auditory apparatus. We tend to hear sounds within a particular frequency range and differentiate fluctuations in tone within a specific span of time. Other animals are not limited in the same way as we are, particularly birds, fish, reptiles and insects.

Vervet words. In Amboseli National Park in Kenya very few vervet infants live past the first few months of life. This is due to the great number of predators. Therefore, vervets have developed a rather extensive number of alarm calls. There is an alarm for predators from the air (certain kinds of eagles), an alarm for leopards, and a soft nearly inaudible alarm that indicates the deadly mamba snake is nearby. the intensity and frequency of calling usually corresponds to the proximity of threat. So, the calls are clearly identifiable for each kind of threat and the increased intensity corresponds to the amount of fear.

Vervet infants produce adult-sounding calls from birth, but they must learn how to use them appropriately. Often they mistake other birds or leaves falling to the ground as a threat and sound off the eagle alarm. Similarly, the infants will mistake sticks or other ground creatures and sound off the leopard alarm. But they do not get mixed up and sound the alarm for eagles for a threat percieved on the ground and vice versa. Slowly the vervets winnow down the number of false stimuli until it appropriately sounds each alarm for the corresponding correct stimulus. In this way, they resemble human infants who may use "dog" to refer to all four-legged creatures at first, but soon learn to use the word only if reference to a specific kind of four-legged animal.

The vervet alarm system shares three similarities with human words. 1) The relationship between the object referred to and the sound of the alarm emitted is arbitrary, that is there is nothing inherently meaningful about the sound used to indicate the object referred to. The vervets could just as easily have used another sound [this is smilar to the way different languages can use different words to refer to the same object--cat, gato, chat, katze]. 2) Vervets learn to use the alarm signals appropriately as they develop. 3) Vervets can determine what the call is referring to even in the absence of the vervet that is sounding the call or absent any vision of the events responsible for sounding off the call. When the animal knows the call, all that is necessary for meaning to be transferred is for the perceiver to know what the call refers to.

Rhesus calls for food. The next question we must ask is whether animals may have distinctive calls for the same object, in essence whether they have a rich vocabulary just as we may have more than one word for the same object. Rhesus monkeys have five distinct calls for food: three for high quality food and two for low quality food. Experimenters wanted to test whether these three vocalizations that referred to the high quality food did indeed refer to the same object. The experiment used a taped vocalization of one vocalization called a "warble". The test animal habituated to this call; that is it did not respond when the call was made. When suddenly a second call was introduced, a "harmonic arch," the rhesus monkeys did not respond. Presumably this was because the harmonic arch referred to the same thing as the warble. Vervets, baboons and diana monkeys also seem to classify vocalizations according to what is being referred to. However, there is no data among any of these species to suggest that this kind of rich vocabulary applies to other situations besides food like predators, friendships or aggression.

Intentional control. Do animals seem to have any control over their signal when they communicate? Do thier vocalizations have the intention of achieving some goal? Or are they just hardwired to respond, like a traffic light?

Vervet monkeys suppress certain kinds of vocalizations about predators based on who their audience is. For example, in an experiment conducted by Robert Seyfarth in the San Fernando Valley, adult female vervets were more likely to suppress an alarm call about a human predator when in the presence of unrelated juvenile females than they were in the presence of their daughters. This showed that they were selective with regard to their audience and that they were not simply hardwired to respond. But were they exactly intentional?

From the experiments discussed in the deception section, it is clear that macaques and other monkeys do not take into consideration the state of knowledge of another monkey before they vocalize. Therefore animal vocalizations are not intentional based on what another monkey knows or believes or wants. They are blind to such mental states in others. Even though such vocalizations have been designed to perform a specific goal, they are not intentional in the way humans can be intentional.

Grammar. All humans have the ability to rearrange words into different order to come up with novel combinations. We possess this unique ability to produce an unlimited range of meaningful expressions, or do we?

In field studies by Roger Payne and Katie Payne they described the songs of humpback whales on their breeding grounds as consisting of subphrases, phrases, themes, songs and song sessions (like jam sessions). They may last up to 10 hours in length and each breeding season shows changes in the structures of these elements. There appear to be conventions for how the notes are arranged and how they may change over time. But whale songs differ from human sentences in the way that all whales in a group produce the same song at a given time. There is no individual expression. Also, the songs don't seem to be referring to anything. They serve simply to differentiate members of a group from one breeding season to the next.

The most detailed work about the presence of syntax in animal communication has been done on black-capped chickadees. They give off a four note call referred to as A, B, C and D. They are always produced in that sequence, and the D is always found at the end. The variation occurs when one or more of the elements are either repeated or omitted. Two common calls, for example are AD and BCD. Still, there are some sequences that never occur, and Cs are never followed Bs or As. The chickadees seem to have an open-ended combinatorial system with some overarching rules, and it appears that this would fulfill at least one of the requirements for possessing a syntax.

However, one must also ask for what purpose has such a syntax evolved. At present there is no meaningful evidence that the chickadee syntax conveys any meaningful information. On the other hand, this does not mean that such information is not conveyed.

One answer to the question of why syntax has developed might be relevant to the idea that social pressures are responsible for the development of syntax. Research among capuchin monkeys in the rain forests of Venezuela suggests that individuals can organize their vocal repertoires around a set of rules. Capuchin calls range in a continuum from submissive to agressive, or they range in a continuum from contact-seeking to contact-avoiding. Capuchins generate compound calls from these two categories; however, there are no discrete rules about which must go first, nor about which one seems to be modifying the other. The combination calls seem to mix together to create a kind of average of the constituent element calls. In this way it resembles the way red and white paint can be mixed together to create pink paint. Unlike the chickadee, there seems to be no formal structural rule.

Animals clearly have rules they use to combine sounds, but these combinations do not seem to generate any new content which refers to anything new. Yet it is possible that such a system lies dormant in the brains of animals waiting to develop under the right conditions.

Perhaps they are like the "Wild Child of Aveyron" who did not develop any ability for human language because he was raised in the wild without the privilege of listening to human language.

Teaching ASL to animals. Because the vocal apparatus of most creatures is severely limited, it may be asking too much for vocalizations to resemble human language as in speech. One way scientists have gotten around this limitation is to use American Sign Language to communicate with animals.

In several species scientists have been successful in using this approach. The first trick was to see if animals could understand the notion of reference, to see if they understood that certain signs corresponded to objects or symbols. Researchers have reported varying degrees of success. Some animals have been able to learn signs for hundreds of symbols. But one must ask the question if these learned symbols are nothing more than learned gestures as a result of conditioning. Or do the animals use the signals in open-ended communication? Are they not restricted to a particular context? Washoe, a chimp, has reportedly used the sign for "open" in response to several novel situations such as a closed tupperware container.

However, there are other aspects of sign usage to consider. Do animals use signs for objects or symbols that are out of sight, that have yet to occur or that are planned actions? Most of the evidence here is anecdotal. There are some reports of animals signing for food requests or a favorite trainer when these things are absent.

There are even some cases where a novel sign was produced in a novel situation. Koko the gorilla signed "white tiger" when she was presented with a toy zebra doll, an bject for which she lacked a sign.

The most wondrous of all cases is Sue Savage-Rumbaugh's experiments with a bonobo named Kanzi. Kanzi uses a symbolic keyboard, and often takes it with him when he goes out into his forest enclosure. There he uses the keyboard like a notepad, commenting on the things he sees or the places he intends to go. All of this occurred without the prompting of his trainers, so the notion of its being a learned response can be ruled out.

Even more extravagant were Savage-Rumbaugh's claims that Kanzi began to develop an understanding of spoken English. Kanzi reacts to spoken words of relatively equivalent sound and intonation the way a dog responds to its name, but he can't differentiate slight variations in the beginning consonant (the same way, for example, my dog comes when I call Lucy or Goosey or Juicy). Kanzi can differentiate these slight variations. Kanzi's mother, Matata, however, developed no such ability, and Savage-Rumbaugh attests this to the fact that Kanzi was raised in the training lab whereas Matata began to train as a young adult. From this, she hypothesizes that there may be a critical window for language development in monkeys just as there is in humans.

Kanzi also seems to be attending to some sense of word order. When asked to throw one of his balls (he is very possessive of his toy balls) into the river, he obliged. However, it could also be assumed that by just hearing "throw","ball"and "river" in the same sentence Kanzi was able to intuit the correct response. Because one can not throw a river into a ball, this was not a difficult test. Savage-Rumbaugh then asked Kanzi to throw a potato at a turtle. In this situation, without some inkling of the importance of word order, either the potato or the turtle could be thrown. But Kanzi passed these tests remarkably well. However, Kanzi made significant errors when asked to negotiate more than two objects. When he was asked to give Sue the potato and the hat, he would hand over the potato but not the hat. Similar findings have been reported with other young bonobos.

One may ask why there has been no indication that these bonobos are using their understanding of such rudimentary syntax to communicate with each other.[ Savage-Rumbaugh reports no indication of this occurring.] Perhaps it is because scientists have not looked hard enough for this, but also one must entertain the notion that they do not transfer these abilities within their own framework of communication.

Morality

According to Hauser, there are six issues to be dealt with concerning the question of whether animals are moral creatures.

1) We must assess whether animals experience the moral feelings of guilt, shame and embarrassment. As with deception, the question of empathy comes up. To Hauser the answer seems to be no, but he also holds out the possibility that further experiments may show that animals are indeed empathic. If so, Hauser believes that another question must be asked along these lines. Do animals place some sort of value on these moral emotions? In other words, are they viewed positively or negatively?

2) Animals must also exhibit some kind of inhibitory mechanism, a mechanism that will allow them to suppress desire. Experimenter Sarah Boysen devised an experiment with chimps which would measure their ability to inhibit desire. Two chimps were chosen. One was selector. The other was receiver. The selector chose between two food wells. However, whichever one she chose would be received by the receiver. The selector would actually receive the other one. Two quantities of food are placed in the wells. If the chimp is able to inhibit desire, she will choose the lesser quantity and therefore receive the greater quantity. After dozens of trials, however, the selector chimp still chose the greater quantity.

Two other experimenters, alan Silberberg and Kazuo Fujita, argued that the reason the chimps could not inhibit themselves was that the dowside to their choice was not great enough (even if they chose wrong they got some food albeit not as much). Silberberg and Fujita reproduced the results of Boysen, but they went further to give no reward when the selector chimp chose the wrong hand (in this version the food items were held in the open hands of an experimenter rather than a food well). The chimps quickly learned the rule of "pick the hand you don't want". Thus, they exhibited some rudimentary ability to inhibit their desire if their is great enough cost associated with their choice.

Conceptual change. Related to the issue of whether animals can inhibit desire is whether they can undergo a paradigmatic shift in thinking about how the world works. This ability is important in light of changing mores. Scientists believe that animals are hardwired to conceptualize both the way the physical world works and the way behavior is to be interpreted. If an animal is unable to readdress its assumptions about the way the physical world works, then it is unlikely to readdress its assumptions about behavior.

The experiment that seemed to illustrate this rigidity of mind was one in which a rectangular frame was fitted with three short pipes (A, B, C) and three boxes down below (1, 2, 3). Then an experimenter attached an opaque tube which connected A to 3. When three-year-olds and an adult tamrins were prompted to find the ball (or food as the case may be), both were unable to make the connection, searching box 1 for the reward item. Even after 20 to 30 trials they did not catch on. When they finally did discover the reward item in box 3, a reconfiguring of the tube (from say B to box 1) saw both groups make similar numbers of errors in trials. Similarly, when a transparent tube is used, both groups find the reward item easily, but when the opaque tube is replaced, the same kinds of errors recur. The results suggest that both three-year-olds and tamarins cannot override their strong assumptions about how the physical world works. They expect all falling objects to fall straight down. Therefore, Hauser says, "animals have no role to play in shaping the moral community because they have an impoverished capacity for inhibition and conceptual change."

3) Do animals consider the beliefs, desires and needs of other animals when they are planning an act? At present, there is no convincing evidence that this is the case (see section on deception). Though animals react to another's behavior, it is not convincingly clear that they take the beliefs or desires of others into consideration when they make their decisions. Surely animals can cooperate, fight and make up afterwards, but it is unclear whether they understand any moral component is associated with this kind of behavior.

4) Do animals have the capacity to understand how their own actions might impact others (cause pain or happiness) in the future? Do they understand the risks associated with their behavior for themselves and others?

Macaulay. The author observed a vervet in Kenya in the 1980s named Macaulay who took over the lead male position among a group of female vervets whose males had all succumbed to predation. Several males had come and gone within this group of females without a male. Macaulay stayed even though he had lived his entire life with a group which lived in an area where there was a lot to eat and a lesser threat of predators. After some time with his new group, Macaulay made some trips back to his old stomping grounds; then he headed back over several hundred yards of treacherous open grasslands where he was vulnerable to attack. Eventually he led his new group back to the old, richer habitat. There he encountered his old group and with the help of the females of his new group, he even attacked his old group back. One day, though, Macaulay left on a trek back to his old territory and never returned (he may have been eaten by one of the many predators in the grasslands). Though his new group clearly knew the way back to the richer habitat, without Macaulay, they never returned on their own even with other males at the helm. Certainly Macaulay seemed to have a special character as a fighter and a leader.

5) Do animals understand duties and responsibilities to each other? If they are altruistic and then burned by this act of altruism, do they consider this act of disrespect the next time they are faced with a decision to act altruistically with the same creature? Let's look at a model system.

Blood Regurgitation in bats. Vampire bats live in large social groups. They must find a blood meal in order to survive. Often, though, that blood meal is hard to find, and a vampire bat can only survive approximately 60 hours without a blood meal. A bat with a stomach full of blood can regurgitate some to another in need, an insurance policy against starvation. Evolutionary biologist Gerry Wilkinson studied the pattern of blood exchange among the bats. Individuals were most likely to regurgitate to those they had regurgitated to in the past, whether that individual was related or not [Note: this undermines the selfish gene theory of Richard Dawkins who posited the notion that all altruistic acts--exclusively with close relatives--were essentially selfish ones because they preserved the donor's own genes or similar genes of a close relative]. Wilkinson's study supports the notion that reciprocation is the main element when an animal decides to act altruistically. Wilkinson also observed that bats seem to punish cheaters. They seem to do this by examining the distension of the stomach during grooming and determining the capacity for regurgitation. Thus, vampire bats seem to pass the test for reciprocal altruism, but more research needs to be done in order to determine their thoughts and emotions. For example, does punishment work? Are cheaters aware of the consequences of cheating? Do cheaters develop reputations? Do bats feel sympathy for those on the brink of starvvation? Is there room for compassion within their system for the elderly or an immature bat which may have trouble in procuring a blood meal?

6) Is there any sense that animals understand the norms of a society and when their rights have been violated? Does a subordinate redefine these norms if he/she forms a coalition to overthrow a tyrannical leader. Even though there are many instances of submissives overthrowing dominants, there is no case to support an active reengineering of social norms by the new dominants when he/they come into power. There are no animal revolutions.

All above accounts are summaries taken from Marc Hauser's book Wild Minds: What Animals Really Think. Henry Holt and Company : New York. 2000.