Go to Lecture One or Twoor Four or Five or Six or Seven or Eight

Go to Cyber Semiotic Institute home page
Go to Course Outline


Human Communication as a Primate Heritage

Instructor: Anne Zeller

Lecture Three: Alternative Functions for Messages

In humans, language is considered the major vehicle for the transmission of information. Information is a very complex phenomenon. Its transfer involves the coding of a message in a channel that can be perceived and decoded by the receiver. Basically this is a cybernetic model which usually involves various levels of competence in coding, interference in the channel, problems in decoding and a feedback loop to indicate whether the message has been successfully received. This model presumes that the noise in the system can be filtered out so that the transmission is successful. One aspect of the model is that the coding system may be subdivided to discriminate between different senders especially on the basis of individual, age, kin group etc. These levels of differentiation must be distinguishable from coding errors or noise in the system. Usually this system must also have the potential to operate in both directions in order to be considered a communication system. Otherwise it is just an information transfer system like the speedometer on a car or a velocity meter on a wind mill.

This cybernetic model can become very complex because multiple channels of communication can be used to code the same message simultaneously and all the coding procedures must successfully transmit their messages and be decoded correctly. It is a major area of confusion when different messages are sent via different modalities simultaneously to the same receiver who must then sort out what the message "really" means. Among humans most of the actual data level information is sent in the verbal/vocal modality, but the auditory channel may detect nuances of tone, and the visual channel nuances of body language that contradict the data being received. The situation can indicate several alternative possibilities. Intentional deceit is one and confusion of intent is another. A third possibility is that the actual information is not really the important part of the interaction which may have another function entirely.

As was discussed in lecture #2 non-human primate systems are now recognized to operate at several levels. For example, the overall level predator alarm among vervets has been demonstrated to include additional information which can be classified as leopard alarm bark or snake chatter, indicating danger of specific types. Another example is the 'teeth chatter' face used as an indication of friendly approach. How the teeth chatter is done, involvement of lips, nostrils, eyebrows, ears, facial tension, tongue protrusion etc. can differentiate species, and age, sex or kin group. But layered over this is also the level of what can be called the more ultimate level of purpose in the gesture. In other words, what is the function behind making this friendly approach. It could be something as straight forward as a friendly greeting to a well known grooming partner, just before settling down to a grooming bout, or it could be an effort to establish an alliance with a less frequently encountered individual. In other circumstances the teeth chatter could be used to appease an individual who has threatened you, to approach a new infant whose mother is very nervous, to express kin bonds (between sibs or mother to offspring) or to establish a consort relationship. All of these are different functions but they rely on one basic message "I am approaching you in a friendly fashion."

This aspect of differentiating message from function applies at both the levels of communication I referred to in lecture 1 as the 'basic' and 'interactive' levels. At the basic level, the information transmitted by species and gender are not really modifiable. However age attribution has both a behaviourial as well as a morphological component and thus information about age can be modified in transmission. One common example is utilizing an infant style locomotion pattern, the stiff legged bouncing hop that infants use, to indicate a play episode. Juveniles and sub adults use this behaviourial marker especially when playing with infants which seem to make their play advances more acceptable. It functions to make them appear younger than they are. In other cases juveniles can use high pitched infant style vocalizations and insist on being carried by their mother. This may occur after an injury or sometimes just due to the idiosyncracy of the animal. One famous example is that of the old chimpanzee mother Flo and her son Flint, studied by Jane Goodall. Even after Flint was 5 years old and Flo had another baby, Flint insisted on riding on his mother's back and trying to nurse. Old Flo was hardly able to walk under the combined weight of her juvenile and her infant, but as much as she tried she could not make Flint more independent because he threw violent tantrums. When the infant died, Flint continued clinging to Flo and trying to nurse. When she died, Flint became very depressed and died soon after, in spite of the fact that he was physiologically old enough at 8 years to have survived. This was not deception but an indication of his emotional state and the message was that he was not ready to mature.

On other cases high pitched fear vocalizations can get help for a juvenile if it is harassed by others. If it screams like an infant, an adult will often come to the rescue. This is seen frequently in Barbary macaques where male care is prominent. These cases are not necessarily deception but the use of age markers in vocalization, locomotion and behaviour as a signal about self.

Health and sexual state are two other states which are generally projected as information at a basic level. The animals colour, smell, presence of obvious wounds or markers of estrus are involuntary messages transmitted to all who understand the code. However, what is important has more to do with how others respond. State of estrus in baboon females will elicit quite different response in young low rank females and from mature, high rank proven breeders. Different consort partners, different intensity, different levels of interest or focus of attention can result. Estrus swelling, which should indicate potential for fertile mating, can occur in already pregnant female chimpanzees especially in captivity, where potential for aggression is higher and more prolonged sexual contact will tend to reduce the potential for aggressive attack. This is also seen in bonobos (once called pygmy chimpanzees) who maintain state of estrus swelling more than half the days of a month. Sexual behaviour and the markers which indicate sexual receptivity and elicit sexual interest from the males have been transformed from an indicator of reproductive readiness to an indicator of social readiness. This transformation is further expressed by the extensive amount of same sex sexual contact usually called G-G rubbing (genital rubbing) in females, and occurrences of males mounting each other. Same sex contact is not rare in many primate species, but in bonobos it is an extremely common everyday occurrence with immature as well as mature animals indicating that sexual behaviour has become widely separated from the reproductive function that estrus swelling was evolved to indicate.

As far as health goes animals can transform the information presented in some particular cases. Usually these do involve individual efforts at misrepresentation. In one case among captive chimpanzees a high ranking male was injured in the hand by another who had just taken over the alpha position. Every time the injured animal walked by the new leader or saw that he was observed, he limped pitifully, although when out of sight he did not. The interpretation put on this behaviour is that the new leader did not tend to attack injured individuals and thus the "injured" male gained a week of respite before having to engage in social interaction. The message was not just 'I am hurt' but 'I am no danger to you' a much more ideational and social message that is an interpretation of the information on health actually being transmitted. (Byrne and Whiten 1988, Machiavellian Intelligence.)

Moving on to the level of interactive communication, another episode from this same captive group of chimpanzees (referred to in Chimpanzee Politics, deWaal, 1982, and Machiavellian Intelligence, Byrne and Whiten, 1988) suggests how social status can be utilized as a mechanism to manage potentially agonistic situations. Two of the chimpanzees mothers were sitting under a tree watching their infants play when the infants began to fight. Lying between the two mothers in the shade of the tree was a third very high ranking female who was asleep and oblivious to the growing commotion around her. As the tension levels increased one of the two mothers poked the sleeping female till she woke up and indicated the fighting infants. Immediately the older female got up and vocalized and arm waved at the infants who then ceased their quarreling and the old female lay down again to sleep. The level of the message was 'stop fighting' sent from old high ranking female to the infants, but the function of this communication was to prevent a fight developing between the mothers of the two infants. If one of them had threatened the infants to stop the altercation it is very likely that the other would have become involved to protect her own infant and the two mothers would have ended up fighting. The communicative complexity of this is that both the mothers and the old female recognized this, and the old female's high status was enough to prevent either mother from attacking her when she threatened their infants. She did not have to send a specific message about her status. It was a well accepted social fact that was implied by the deliberate involvement arranged by the one mother in waking her up.

Other cases in which the meaning of the message went far beyond the information being transferred included situations in which peaceful animals minding their own business are suddenly attacked by other group members for seemingly no reason. It is only when researchers are aware of the interplay of behaviours over the course of the day or a few days previous that the rational is clear. In baboon groups matrilines are an important structuring element with both young and grown offspring of a particular female remaining in a close and supportive relationship with her. Also, in baboon groups some males have supportive alliance relationships with particular females who they sit with, groom with, eat with and often baby tend for. If a conflict occurs in which, for example, a high ranking female attacks a female of a different matriline, that female's male friend may attack a member of the first female's matriline latter in the day or even a day or two later. This attack will not seem to be precipitated by any act of hers, but is purely on the basis of her kin relationship to the female who made the first attack. The message of aggression looks the same, but the actual function of the attack is retaliation for the initial aggression (Smutts, 1985 Sex and Friendship in Baboons). This type of communication can also be considered a message to the high ranking female instigator not to continue with aggression towards that male's friends. Sometimes, these interactions occur much more closely in time in which case the second attack is often called re-directed aggression and is often considered to have an emotional basis. However, the choice of who to re-direct the aggression to is rarely random. The choice of the recipient may indicate either 'I'm still higher ranking than you are' or 'if I get attacked someone in the attacker's network of allies will suffer for it'. I realize that these seem like anthropomorphic observations on my part, but these are not rare occurrences and a skilled observer can often predict who out of all the animals present at the scene will be the one attacked. If there is predictability and patterning there is more than a message of anger and frustration being sent.

Another set of situations in which the function and the message can be quite different in primate communication occurs when the message only serves as a vehicle for the function. This is rather similar to human cocktail party conversation in which the information transferred -- about the weather -- your vacation -- boss -- co-workers etc. -- is not nearly as important as being seen in close proximity to and animated interaction with an attractive potential sexual partner or an important company executive. Primates use communicative ploys to gain proximity to and interaction with desirable partners. In one case from my own experience, two females were eyeing each other but a little nervous to approach closely. Then a young infant wandered in between them. One way of proceeding would have been for both females to groom the infant and thus gain proximity to each other and eventually end up grooming each other. In fact, in this case one female hit out at the infant, and it vocalized suddenly. Next the other female threatened the first who jumped back quickly, then began to grimace and teeth chatter, which are submissive and friendly affiliative gestures. As she did this she slid forward until she was quite close to the other, but on an angle. The other female then lay down in a 'present for grooming' position and the first female moved to her back and began to groom her. I recorded the whole sequence on video and in slow motion it is fascinating to see how each glance and move by one female is mirrored or responded to by the other. The initial infant was totally ignored by the two females after it had served as a pretext for the beginning of the interaction. The initial aggression towards it was not 'real' in the sense that it had done anything to deserve it, and the function of the aggression, threat, submission, appeasement sequence was the final social goal of a grooming session. I'm sure that many of you can think of examples of this indirect approach in human interactions, especially in courtship situations. Sometimes these tactics can work in reverse as in situations where an initial good impression can be spoiled by aggressive actions or responses to an intruding individual. One rather intolerant macaque male I saw who was being groomed by a female hit out at her infant when it returned to her from a play bout with another. Immediately she stopped grooming him, picked up her infant and would not respond to more solicitations to groom from him for the rest of the observation period. Instead she groomed with a group of other females and infants while he tried a number of ways to renew the grooming interaction. This is also recorded on video and it is fascinating to see how she kept her face averted, and her back to the male as much as possible as he moved around her. Eventually he went and roughed up another juvenile and left the vicinity. A day or two later I saw them again and this time the male was much less aggressive towards the infants, made appropriate friendly faces to the females, and eventually got a grooming partner. This was an old, very high ranking breeding male, who if the standard literature is to be believed should have been a prime grooming associate. Even he had to learn to modify the messages he sent in order to gain the functional goal he was seeking.

Thus there are several ways in which the information being transmitted may not be the same function as the communicative impact. The impact can modify an individuals's perceived abilities and social persona as in manipulations of perceived age-whether intentional or not. Social modification of particular signals, such as the extension of estrus/sexual interaction from a reproductive to a social mechanism can occur both for individuals and for whole species. It can go so far as in bonobos to characterize their whole group social organization, where sex is used as a general anxiety reduction and conflict resolution mechanism.

Animals can utilize social situations and realities to reduce levels of aggression towards themselves or within the group with or without the complicity of others. The functions of some interactions can be retaliatory, or socially adhesive without regard to the nature of the message actually being sent. This can occur due to positive action on the part of the sending individual who can manipulate the outcome as desired, or can be a negative outcome when messages are sent that are not appropriate to the social end that an individual is trying to achieve.

These primate examples all have analogues in the human communication systems which will be the topic of the next lecture.

References Cited

Byrne, R. and Whiten , A. 1988. Machiavellian Intelligence , Oxford Science Publication, Clarendon Press. Oxford.

Smutts, B. 1985. Sex and Friendship in Baboons. Aldine Publishing Co. New York.

de Waal, F. 1982. Chimpanzee Politics. Johnathan Cape. London.


copyright 1999 Anne Zeller
Send comments or question to Anne Zeller: azeller@artspas.watstar.uwaterloo.ca
Go to Lecture One or Twoor Four or Five or Six or Seven or Eight
Go to Cyber Semiotic Institute home page
Go to Course Outline