Monday, 25 October 2010

Symbolic communication, part 3: Origins

We may say that thinking is essentially the activity of operating with signs.
— Ludwig Wittgenstein [1]

Today, all humanity engages in the creative and complex use of symbolic representation. But there is no consensus among archaeologists about when this began. One school of thought favours a ‘long-range’ model wherein hominids have been capable of some degree of symbolic behaviour for hundreds of thousands of years. Others prefer the ‘short-range’ model, limiting it to Homo sapiens.[2] Unequivocal art appears 40,000 years ago in the cave art of France and Spain. This is very unlikely however to represent the first appearance of symbol use among humans, whose origins lie not in the fully-formed art of the caves but in cognitive capacities which must have developed earlier, perhaps much earlier, than art.

The symbolic ape

To what extent pre-sapiens human species were capable of symbolic representation is disputed. Readers will know that this blog tends to the view that art, with the possible exception of some Neanderthal objects, has only ever been created by Homo sapiens. This does not mean that symbolic behaviours cannot have awakened in our genus long before our species appeared.

Language, for example, which is communication by symbolic means, is considered by many archaeologists to have developed prior to sapiens. The Makapansgat pebble from South Africa was found in a cave where it could not have occurred naturally, inviting speculation that it was carried there by some gifted australopithecine fascinated by its resemblance to a face. This raises the possibility of a rudimentary symbolic faculty as far back as 3 million years ago (and reminds us of Terrence Deacon’s suggestion that symbolic communication may date as early as the australopithecines). From about 850,000 years ago, Homo erectus in South Africa and elsewhere seems to have collected quartz crystals that had no practical application, and 800,000 years ago in the Wonderwerk Cave, also in South Africa, we have early evidence of the collection of coloured pigments. Around 300,000 years ago we see objects like the engraved elephant’s tibia from Bilzingsleben and the alleged sculptures of Berekhat Ram and Tan-Tan.

None of these objects in my view can be considered to be true art. However, it is possible that they represent a kind of proto-art, flashes of symbolic expression developing alongside language and increasingly sophisticated tool-making. Even tools, such as highly symmetrical Acheulean handaxes, sometimes suggest an emerging aesthetic sense. The evidence is not as abundant and unequivocal as examples from the Upper Paleolithic, but nonetheless a strong case can be made for instances of symbolic behaviour in the Lower and Middle Paleolithic.

Many archaeologists argue that Neanderthals in particular, amongst non-sapiens humans, were capable of producing symbolic objects. Even if this is so, their use of symbols was clearly far more limited in quality and quantity than in our own species. According to the archaeologist Randall White, modern behaviours require “evidence of organised symbolic systems shared across space and through time”[3]; what is distinctive about Homo sapiens is that symbolic behaviours become unambiguous, widespread and a dominant part of our cultural life.

About 160–200,000 years ago in Africa, Homo sapiens emerged as a distinct species and slowly spread out across the world. What made us special was a set of new cognitive powers that separated us from our hominid ancestors, such as the creation of complex, specialised tools, ritual burial of the dead and the creation of art — all products of the widespread use of symbols.

PictogramPictogram of an animal from the Apollo 11 cave in Namibia, ca. 27,000 years old.

Our earliest evidence for symbolic behaviours amongst Homo sapiens lies in sites such as Qafzeh Cave in modern-day Israel, where red ochre appears to have been used during burials as long as 100,000 years ago. Ochre was used as a colouring in other contexts too, and pieces of perforated bone and shell were worn as beads. About 75,000 years ago, perforated shell serving as beads and a piece of ochre etched with seemingly symbolic markings were found in Blombos Cave.

How did symbolic communication evolve?

How humans’ use of symbols evolved is, unsurprisingly, disputed. Our difficulty is that brains do not fossilise and we are restricted to mostly circumstantial evidence — the sizes of brain cavities, surviving cultural artifacts, and so on. It is obvious that our unique cognitive powers began at some point in the 2.5 million years since, and possibly helping to create, the habilines. The archaeologist Timothy Taylor has argued that ‘a fractionally more complex material culture’ may account for the evolution of australopithecines to habilines.

The acquisition of symbols was a remarkable achievement, unprecedented in the history of all known life. Symbols are extremely difficult to ‘discover’ because they draw their meaning not from immediate correlations with physical things, but from each other. This is argued by Deacon in The Symbolic Species:

The problem with symbol systems... is that there is both a lot of learning and unlearning that must take place before a single symbolic relationship is available. Symbols cannot be acquired one at a time, the way other learned associations can, except after a symbol reference system is established. A logically complete system of relationships among the set of symbol tokens must be learned before the symbolic association between any one symbol token and an object can even be determined... For this reason, it’s hard to get started. To learn a first symbolic relationship requires holding a lot of associations in mind at once while at the same time mentally sampling the potential combinatorial patterns hidden in their higher-order relationships.[4]

When a pigeon in a laboratory learns to taps a button in order to receive food, it is making an indexical relationship between the object and the delivery of food. To use symbols requires more than making rote associations between signifier and signified: we need to be able to extract abstract features from objects and observe relationships between things. Because they do not refer to things in the world through direct physical correlations, symbols work by reference to each other, drawing their meaning from a system of symbols. This doesn’t mean they have no relationship to reality at all, but it does mean the signified does not have to be physically present or, in the case of abstract concepts, to have any physical form.

How did Homo sapiens make that leap?

Given the difficulty of acquiring it, there must be something special about human brains that makes symbolic communication easier for us than for other species. (We also need to understand that the mind is not limited to our brain but is also in a relationship with our body and with the world beyond it, of which more another time.) Biological parameters are certainly important. For example, small animals are also short-lived animals, for whom the process of learning would take more time than they have available, and so their small brains would not be adapted to such strategies. This may mean that symbolic communication could only develop in a relatively long-lived, large-brained animal, such as a primate. There is however no straightforward correlation between brain size and intelligence: whales have bigger brains than modern humans, as did Neanderthals on average; even when we take the proportion of brain size to body size, the shrew comes out on top. So although the largeness of our brain seems to be important, its ‘architecture’ must be at least equally important. Deacon proposes that the prefrontal cortex, which is unusually large in humans, may be significant in helping us recognise patterns of cognition and in building the mental architecture that supports symbolic reference. This doesn’t however mean that our processing of symbols can all be found in the prefrontal cortex — any search for a ring-fenced part of the brain that processes symbolisation by itself is doomed to fail.

The leap to symbolic communication

The force that drove the acquisition of symbols is still unclear. But an obvious contender is language. Language — which is more than just speech — requires abstract thought and complex structures, as well as its role in giving advantages to our food-gathering, child-rearing, spreading into new territories, cultural production and other activities. Here is Deacon again:

The doorway into this virtual world was opened to us alone by language, because language is not merely a mode of communication, it is also the outward expression of an unusual mode of thought — symbolic representation. Without symbolisation the entire virtual world that I have described is out of reach: inconceivable.[5]

If language helped to drive the evolution of the modern human brain, this must be based upon some kind of adaptation in the brain that makes language possible, but what difference is it and by what process did it arise? Rather than ask which came first, the chicken or the egg, i.e. language or the widespread use of symbols, we should ask how the two evolved in combination — language is itself a form of symbolic communication. We have already discussed the acquisition of language by humans in a previous article, where we concluded that language probably arose from the new concepts introduced by tool-making and new forms of social organisation.

This process of intellectual development had material consequences for our brains, embodied in the famous phrase from John 1:14, which Deacon takes as a chapter heading: “and the word became flesh”. Describing the human brain as a miracle, Deacon proposes that

the changes in this organ responsible for this miracle were a direct consequence of the use of words. And I don’t mean this in a figurative sense. I mean that the major structural and functional innovations that make human brains capable of unprecedented mental feats evolved in response to the use of something as abstract and virtual as the power of words. Or, to put this miracle in simple terms, I suggest that an idea changed the brain.

A Marxist’s initial response to this is probably suspicion. It sounds like idealism: abstract thought leading the physical world, or consciousness determining being. In fact, Deacon’s idea sits comfortably with the dialectical materialism of Marx and Engels. His argument is that the acquisition of symbolic reference by our ancestors influenced the ways in which natural selection affected our evolution.

To this end he recruits the extension of Darwinian theory proposed at the turn of the last century by the American psychologist James Mark Baldwin, sometimes called the ‘Baldwin effect’ or ‘Baldwinian evolution’. Baldwin suggested that learned and cultural behaviour can affect a species’ evolution, because behaviour can influence the environmental context that selects for future generations. Natural selection will “move along the channel already cut by culture, thereby converting learned behaviours into genetic adaptations or, alternatively, supporting learned behaviours by related genetic adaptations”[6].

Seasonal migration, for example, might allow some members of a group to enter new evolutionary niches at the very edge of the group’s normal habitat. As the sub-group spends more time in these new conditions, physical adaptations such as a thicker coat or hibernation would be selected for. An example amongst humans is lactose tolerance, which is strongest in areas where people have cultivated animals for their milk for the longest — i.e. a cultural change, the domestication of animals, has led to a genetic change. Deacon proposes:

More than any other group of species, hominids’ behavioural adaptations have determined the course of their physical evolution, rather than vice versa. Stone and symbolic tools, which were initially acquired with the aid of flexible ape-learning abilities, ultimately turned the tables on their users and forced them to adapt to a new niche opened by these technologies. Rather than being just useful tricks, these behavioural prostheses for obtaining food and organising social behaviours became indispensable elements in a new adaptive complex. The origin of ‘humanness’ can be defined as that point in our evolution where these tools became the principal source of selection on our bodies and brains.

Engels, in his essay The Part Played by Labour in the Transition from Ape to Man (1876), placed great stress on the human hand — labour and tool-making perfected the hand, which led to advances elsewhere in the organism. Engels admitted that the correlation was not yet understood, and his emphasis on the hand, based upon the scientific level of the time, oversimplifies the evolutionary process. Yet his statement in general terms is correct: our evolution was a consequence of our material engagement with the world.

The significant increase in brain size associated with the beginning of the genus HomoHomo habilis — roughly coincides with the Oldowan technology dating to approximately 2.5 million years ago. Perhaps, it was not that greater intelligence led our ancestors to start using tools. Rather, it was the very first use of tools that encouraged a leap of intelligence to our own genus: “implicit in their stone tools and social-ecological adaptations,” writes Deacons, “were the seeds of future human characteristics.” He argues, “symbols have literally changed the kind of biological organism we are... I think we operate fundamentally different at a cognitive level because of this.”[7]

This is something Marxists have been arguing for a very long time. In the venerable phrase of Gordon Childe: ‘man makes himself’.

Signs would have expanded our ability to communicate and survive as well as the cognitive powers required to create art. An animal may make an association between, say, a particular sound and a particular event. But associating things that have no correlation in the physical world through an infinite number of representations: that is unique to human beings. In addition, language organises our thoughts and perceptions to the point that we cannot be sure where our perceptions end and their interpretation by language begins.

Deacon suggests that the very earliest symbol use may date back to the australopithecines, with larger brains, opposable thumbs, bipedality and other key changes developing as a result of emerging symbolic systems rather than the other way around. The use of symbols then strengthened as a solution to the particular situation in which Homo sapiens found itself. This is the kernel of his theory that the brain was changed by ideas.

One of the mechanisms for acquiring symbols may have been ritual. Ritual demands a shift of attention from immediate indexical relationships of signs to objects to a set of relationships between signs, i.e. from concrete relationships to the abstract. Ritual fulfils this role by repeating actions over and over again until they become automatic, releasing the mind from the individual details and concentrating instead on a kind of ‘higher order’ meaning. Rituals such as initiation at puberty, or marriage, define group members according to symbolic roles such as child, adult, husband, wife, using vocalisations, gestures, and objects.

Objects routinely used in the ritual become symbolic of the ritual themselves — a wedding ring becomes symbolic of marriage. “Symbolic culture was a response to a reproductive problem that only symbols could solve: the imperative of representing a social contract.” We know from the evidence at Qafzeh Cave that ritual burials were being performed 100,000 years ago, where the participants seem to have been making a symbolic connection between a coloured pigment and the concept of death.

weddingA modern Western marriage ritual. All elements of a marriage ceremony — the location, the clothes worn by either sex, the actions they perform — are symbolic conventions determined by culture.

The jump to symbolic communication was so difficult and so significant that it must have offered advantages in terms of natural selection. Once humans had acquired their very first proto-language, many aspects of human behaviour would have been improved by its acquisition. This is why there are so many theories about what particular aspect kicked things off. Language would have enhanced everything we did, from rearing babies to felling mammoths, enabling the sharing of information, the building of social bonds, and so on. Symbols would have made possible the creation of contracts between social groups, as a contract is itself a kind of symbol — not based on an indexical relationship between things, but on abstract concepts such as peace or marriage.

There is another reason why humans might have made this cognitive leap. Hominid species experienced a remarkable swinging of the world climate from cold to hot and back again, sometimes within a single lifetime. Richard Potts, an anthropologist, has argued that we were adapting to a constantly changing environment: the use of symbols and abstract thought gave us the flexibility to diversify our behaviour and take control of our environment.[8] This plasticity allows us to put concepts together in countless creative ways. In my view, however, this adaptability should be seen as a benefit of our cognitive development rather than a cause.


One does not have to agree with Deacon on everything, but for a Marxist, his emphasis upon relationships and process, rather than fixed templates and the gradual acquisition of individual facts, is very compelling. His argument for the evolution of symbolic communication is in my view compatible with Engels’ emphasis on human evolution as the work of humans themselves, thus demonstrating that although Marx and Engels were both born nearly 200 years ago [9], their thought continues to find support not only in economics but in modern neuroscience.

Whatever the particular cause for the original emergence of symbolic communication, it probably spread into other social and practical functions. We see everything in symbolic terms, down to seeing our lives as stories, or interpreting the sunrise as the work of a deity. The ability to overcome one’s immediate interest for a ‘higher order’ meaning is what makes it possible for humans to enact all sorts of unique behaviours, some of which represent the best and worse of our species — bravery or self-sacrifice on the one hand, torture and atrocity on the other.

The ubiquity of symbols based upon arbitrary associations means that human beings are caught in a network of ideology and conventions. We have already quoted Voloshinov in the previous post: “Wherever a sign is present, ideology is present, too”. Some of this network has a strong basis in material reality, other aspects are illusions or downright nonsense. This too is an inescapable aspect of what it means to be a human being.

[1] Wittgenstein, The Blue Book, The Blue and Brown Books (1958).
[2] For a discussion, see Robert Bednarik, ‘Beads and the origins of symbolism’ (2000). Archaeologist Bednarik is an exponent of the long-range model, insisting that symbolic behaviour is very ancient.
[3] Cited in Michael Balter, ‘What Makes Modern Humans Modern’, Science (February 2002).
[4] Terrence Deacon, The Symbolic Species: The Co-evolution of Language and the Brain (1997).
[5] ibid.
[6] Bruce H. Weber and David J. Depew, Evolution and learning: the Baldwin effect reconsidered (2003).
[7] In an interview available at
[8] Richard Potts, Humanity’s Descent: The Consequences of Ecological Instability (1997).
[9] To be precise, Marx was born in 1818, Engels in 1820.

No comments: