Saturday, 23 October 2010

Symbolic communication, part 2: Animals and symbols

“Human behaviour is symbolic behaviour; symbolic behaviour is human behaviour. The symbol is the universe of humanity.”
— Leslie White

Symbols are a part of human cultural life only — no other animal normally uses symbols. Yet the assertion of the socialist archaeologist Leslie White above [1] may seem too exclusive. We often assume that animals have a simple understanding of sign and signifier, even of words. After all, we know that animals can be trained to do things. We tell our dog to sit and it sits: isn’t this evidence that it understands the word ‘sit’ on some level by associating a sound with an object? The alarm calls of social species such as vervet monkeys vary according to the type of threat detected: doesn’t that equate to using signifiers for different predators in a way that, in essence, is like language?

Firstly, signs are not unique to language. Both animal and human communication use signs. Rather, what matters here is the kind of reference being used: animals do not use reference in the same way as humans do with words. This has been persuasively argued by Terrence Deacon in his book The Symbolic Species. When we tell our dog to sit and it sits, it has learnt to associate a sound with an action, but that is different to the dog understanding what the word ‘sit’ means. Deacon uses the example of laughter, which can’t normally be said to ‘mean’ something the way the word ‘laughter’ does. Instead it indicates a certain emotion, relying on a physical co-occurrence between the signifier and signified.

By contrast, a word’s meaning is not intrinsic to it but requires “an appropriate symbolic interpretive response”. The animal’s response to the ‘sit’ command is inflexible, rather like rote learning. Thus, to refer back to the discussion in the previous post, the sign functions on the level of an index, not a symbol: a certain smell may indicate to a dog that a female is on heat, or the vervet’s alarm call may indicate the presence of a certain predator. This is quite different to the symbolic communication used by humans, wherein things are associated by a social agreement. Two people will use the word ‘dog’ even when a dog is nowhere to be seen, and no person has ever seen a ‘snark’. In addition, the discussion of a dog will call into our mind all sorts of other associations and concepts, such as ‘fur’, ‘pet’, ‘companionship’ and so on, for which no physical or correlative object need be present. When we tell the dog to sit, the dog only understands the word ‘sit’ indexically, while the human understands it symbolically.

Symbol use in apes

To say that no other animal normally uses symbols is not to say that all animals are incapable of using them.

A great deal of research is being conducted with our nearest living primate relatives to see what can be learned about the effects of language and culture on early human development. Some clever apes can be trained in sign language or Lexigram boards and show a capacity for using symbolic communication, including abstract concepts such as ‘good’ and ‘bad’ and a conception of time. One of the most famous of these apes is a bonobo or pigmy chimpanzee named Kanzi, who along with his sister Panbanisha understands spoken English — although they don’t have the vocal apparatus to even try and respond in kind, they do communicate with humans using symbols.


Kanzi using a Lexigram board.

Kanzi can do more than simply match a symbol to a word. By touching a symbol, he can communicate his own needs, wants and feelings.

Researchers such as Sue Savage-Rumbaugh, who works with Kanzi, suggest that he is not necessarily exceptional. His and Panbanisha’s ability to understand symbols was only unlocked through intense human intervention, but the potential clearly existed in the bonobos already. Perhaps this is not especially surprising in hindsight. Chimps and bonobos are separated from humans by about six million years of evolution, but we share a common ancestor [2] and a primate biology. It seems that at some point in our common hominid evolution, conditions favoured a symbolic capacity which only Homo sapiens brought to maturity. Although animals cannot create art, we know that apes can display a rudimentary aesthetic sense, as when the famous ‘painting apes’ match brush marks to a pictorial field.

The abilities of Kanzi and Panbanisha are astonishing for someone who hasn’t come across them before, and they cause difficulty for nativist arguments for language, i.e. that human language ability is based upon a module, ‘Language Acquisition Device’ or similar development unique to Homo sapiens. Some scientists, including Steven Pinker, have been highly critical, arguing that Kanzi does not really understand symbols and is reacting to rewards given him by his handlers. If one does not wish to go that far, the human capacity for symbolisation is nonetheless of a different order. Kanzi can communicate approximately on the level of a two-year old child, or to look at it differently, of one of our hominid ancestors. He may help us understand how symbolic communication got started in an early human such as Australopithecus. But he has no physiological capacity for speech, depends entirely upon a human-made ‘gifted environment’ for his abilities, and is severely lacking in sophistication (he is unable to ask a question, for example).

Language is not simply a ‘more complex’ version of animal communication, it is a different order of communication. It demands that signs acquire culturally determined meanings that are learned, not biologically inherited. This is not to deny that rare cases like Kanzi can be taught to use it to some extent, or that no other species on any planet could possibly acquire them. The achievements of apes like Kanzi prove that it is incorrect to claim that only humans are capable of using symbols, and show that language acquisition depends upon a culture of language. However, thoroughgoing symbolic communication, such as art, demands cognitive skills which only humans possess. We wait in vain for Kanzi to write a poem. It is very difficult to say for sure, but our own evolutionary lineage seems to have acquired some degree of symbol use (tools, language) before it acquired art. Language is sometimes attributed to early hominids, e.g. by Steven Mithen, whereas full-blown art — leaving aside the arguments over the Berekhat Ram object, the creations of the last generation of Neanderthals, etc — only appears with Homo sapiens. Kanzi has proved capable of taking the first steps on the symbolic journey, but unlike humans has not experienced the many millennia of subsequent biological and cultural evolution necessary for art.

Lacking the advantage of intense intervention from a symbol-using species, our own ancestors had to make the leap to using symbols for themselves. It was probably an uneven process, with flashes of symbol use emerging and dying here and there before it finally became a universal human trait. How it might have arisen is the subject of our next post.

Further investigation
Watch video documentaries about Kanzi at http://kanzi.bvu.edu/Kanzivideos2.html (two videos, length approx. one hour each).



[1] Leslie White, ‘The Symbol: The Origin and Basis of Human Behaviour’, ETC: A Review of General Semantics, vol. 1 no. 4 (1944).
[2] One contender is Ouranopithecus, an ancient hominid species of which a skull fragment was discovered in Greece.

No comments: