readings> Vygotsky and the bifold model

The human mind appeared on earth with astonishing suddenness. Just 70,000 years - the merest eye-blink of geological time - covers our transformation from smart ape to self-conscious Homo sapiens.

On one side of the evolutionary divide stands our ancestor, Homo erectus, a clever beast with a near modern-sized brain, a simple tool culture, and a mastery of fire, yet mentally still somehow lacking. On the other stands modern Homo sapiens with not just a vastly superior ability to fashion tools but also the rituals and symbolic art - the cave paintings, beads and bracelets, decorative lamps and burial graves - that so clearly mark the arrival of a thinking, self-aware mind. Something sudden and dramatic must have happened, and it is this simple truth that should be the starting point for any attempt to explain human consciousness.

Some researchers have suggested that the brain passed a critical size threshold and just "lit up". But a more plausible reason for the great advance made by Homo sapiens was the development of rapid, modern speech - or to be precise, the internalisation of this ability to create a controlling "inner voice".

The suggestion that inner speech could be key to understanding the special attributes of the human mind is hardly new. Even Greek and Enlightenment philosophers - particularly Thomas Hobbes - made reference to the possibility. In Victorian times, the idea cropped up often. Charles Darwin wrote in the Descent of Man: "A long and complex train of thought can no more be carried on without the aid of words, whether spoken or silent, than a long calculation without the use of figures or algebra." In the 1930s, the brilliant Russian psychologist, Lev Vygotsky, tried to rebuild the whole of psychology on this insight - an attempt which, in the event, was defeated by his early death from tuberculosis and the suppression of his writings by Soviet authorities.

Yet while there has long been speculation about how inner speech may shape "trains of thought", it is only during the past ten years or so that it has begun to dawn just how deep a role inner speech may play in fashioning the human mind, going far beyond logical thought to include the entire apparatus of our higher mental powers. Much of the evidence is being gathered outside the boundaries of traditional psychology by researchers in fields as diverse as anthropology, sociology, linguistics and educational psychology. But with the findings now beginning to filter back into psychology, a new paradigm is taking shape which promises to revolutionise our very conception of the human mind.

Put in a nutshell, science is going to have to abandon the simple reductionist approach it has been taking to the mind. Instead of attempting to reduce the whole of consciousness to something biological - a collection of brain circuits or information flows - scientists are beginning to follow a more complex dual, or bifold, model. This approach correctly reduces the animal foundations of the mind to neurology, but seeks the explanation of the special human extras in a very different direction - in language, childhood socialisation and culture.

Memory research is one field that is beginning to reflect this shift in thinking. It also highlights the confusion that has been created by the traditional reductionist approach to the mind.

In the past, it has been normal to treat memory as a single, unitary, biological system. It may have been a system of many parts, such as short and long term memory, but it seemed safe to assume that all these parts evolved together and so must be constructed out of the same basic - presumably neurological - stuff. Because of this, most have believed that human memory can be treated as simply a scaled-up version of animal memory. Our bigger brains might mean we can store more, but the underlying machinery of storage and recall must be the same. Likewise, it seemed unquestionable that our capacity to remember is genetic. Our memories may be poorly developed when we are born, but they strengthen with the gradual maturing of our brains.

Recently however researchers have begun dividing the human mind along the fault-line of language to produce a quite different picture. There is indeed a core genetic capacity shared by animals and humans - the ability to recognise and associate. But layered on top of this is the learnt, language-based skill we know as recollection.

The crucial feature of the animal mind is that its awareness is locked into the present, driven purely by the flow of external events. Where the human mind is driven by an inner world of thoughts, plans and intentions, an animal's intelligence is reactive, responding solely to the clamour of circumstance. So while there is little doubt that animals have a memory, it is only a direct, associative memory that is mobilised to make sense of events as they are occurring. Cats can recognise a mouse or a dog when they see one, and this act of recognition will unlock stored knowledge about how best to react. But there is little evidence that they have an internal mechanism for bringing past experiences of dogs and mice back to mind independently of what is happening around them. However with self-addressed speech, humans discovered just such a triggering mechanism.

At one level, the power of language is self-evident. Words are mere noises - puffs of air - but because we learn to associate them with specific sets of ideas, the act of saying or hearing a word will unlock a wealth of images, associations and experiences in our heads. Hearing words such as "alligator" or "hovercraft" will sting our brains into action, creating a surrogate experience that is almost as good as having an actual alligator or hovercraft before our eyes. Words are like handles with which we can haul ideas through the biological "display area" of consciousness. But it is at the next level, the question of how we use our inner voices to organise higher mental activities such as remembering, reflecting and planning, where the questions start (see below, extra on circular process or circular argument?).

Some researchers - such as a group centred around James Wertsch at Clark University in Massachusetts - are pursuing the developmental approach pioneered by Lev Vygotsky in the 1930s. Vygotsky observed that children go through a phase between the ages of two and seven when they spend a lot of time talking aloud to themselves. Closer analysis showed that what the children were doing was rehearsing the habits of planning and organising that they would later internalise as inner speech. A boy sitting down at a playschool art table might be heard to say: "I want to draw something. Let's see. I need a big piece of paper. I want to draw a cat."

Vygotsky tried to show how a similar use of self-addressed speech underpinned the human ability to recollect. To begin with, Vygotsky argued that simply learning a vocabulary imposes an abstract structure on our memories. As a child masters words like cat, draw and paper, it creates a series of culturally-defined categories to which new experiences can be assimilated. At first, a child may call every furry animal a "kitty", but eventually the word will come to stand for a sharply defined set of images and ideas.

Developmental psychologists have found many examples of how children's memories come to be linguistically organised, but perhaps the most obvious is in the act of drawing. When kindergarten-aged children are asked to draw something like a house, their pictures are surprisingly stereotyped - a square box with two windows, a door in the centre, a triangle for the roof and a chimney blowing smoke. What the children put down on paper is not a remembered image of an actual building but a word-based representation - a house followed by a list of all the elements, like windows, doors, curtains and chimneys, which they have learnt to associate with the concept of a house. It is only with a lot more training that children can be taught to draw simply "what they see".

But Vygotsky went further, saying that language not only gives our memories a culturally-based order, it is also the mechanism by which we call memories back to mind. Through skilled self-questioning, we prod old memories free.

Most people are aware of the kinds of effort they need to make to flush out a recollection. For example, when asked to recall what they had had for breakfast the previous day, normally people report going through a similar process of inner questioning and suggestion. They may try thinking of the usual kinds of things they have for breakfast, saying to themselves something like: "Coffee? cornflakes?...no, I was out of milk...toast, it was toast!". The image-stimulating power of words would be used to set up a series of "target" experiences, then when their thoughts strayed close enough to the dormant memory trace, the natural knowledge-mobilising powers of the brain would take over and there would be an confirming flash of recognition, a mental aha! telling them they were correct. And once the correct patch of memories had been aroused, associations would begin to flow. Extra details about the breakfast - such as who they were talking to, what was on the radio, or whether they were in a hurry - would emerge unprompted.

Circumstances meant that Vygotsky could only gather sketchy evidence of this inner use of speech. But during the 1980s, a number of researchers tried to demonstrate the kind of "verbal protocols" people might use by asking subjects to think aloud as they carried out a memory task. One classic experiment involved people trying to recall the names of all their classmates from their teenage schooldays. With as many as 600 classmates to be remembered, some of the tape-recorded sessions went on for a total of ten hours!

The recordings showed the mix of self-prompting and confirmation that might be expected from a process of word- based suggestions leading to answering flashes of recognition and association. A typical report went: "It's like I want to think of, sort of, prototypical situations and then sort of examine the people that were involved in those. And things like PE class, where there was...ah...Gary Booth. Umm, and Karl Brist, were sort of, we always ended up in the same PE classes, for some reason. Umm...I can think of things like dances. I guess then I usually think of...of girls (chuckle)."

While the developmental approach pioneered by Vygotsky represents the most direct attack on traditional reductionist models of memory and the mind, by the late 1980s, others had begun to reach a similar conclusion having started at the other end of the equation - in the fields of sociology and anthropology.

Social scientists have, of course, always seen the shaping hand of culture in human attitudes and habits of thought. They have also had their own inspirational figures in the anthropologist, George Herbert Mead, the linguist, Benjamin Whorf, and the philosopher, Ludwig Wittgenstein. But it is only recently that they have pushed the argument further and begun to analyse the role that language and society might play in moulding human mental abilities.

One flourishing school is social constructionism which, by comparing ways of thought in different societies and at different times in history, attempts to show how language acts as a "genetic code" for the beliefs and customs of a culture. This approach has had a particular impact in the field of human emotions where anthropologists have been able to show that higher feelings, such as loyalty, righteousness, and even romantic love, are really culturally-invented attitudes. A word is used to stand for a set of ideas, characteristic physiological states and behaviours. In this view, a higher emotion is not a pure mental state but something more akin to a script we have internalised and learnt to act out.

Recently, the social constructionist approach has been extended into the field of memory research. Psychologists, such as David Middleton and Derek Edwards of Loughborough University, have adopted modish techniques like discourse analysis in which samples of everyday conversation are put under the microscope to see how social situations often control people's efforts at recall. A typical discourse analysis experiment would involve recording a group of people discussing the film "E.T." and observing how a collective version of events gradually emerges as ideas and associations are bounced about.

But it has to be said that while the blossoming of developmental and social approaches to memory over the past few years suggest that a fundamental change may be afoot, the mainstream of psychology is still firmly reductionist in outlook. The majority of cognitive psychologists remain wedded to the assumption that recollection and recognition are simply different parts of the same biological system - a system which should eventually be reducible to a collection of neural circuits or computational modules. The doyen of memory researchers, Alan Baddeley of the MRC Applied Psychology Unit in Cambridge, perhaps summed it up when he recently dismissed the discourse analysis approach, saying: "a great deal of the study of memory is precisely concerned with avoiding confounding memory with the language in which a particular recollection is phrased."

Given this outlook, it is unsurprising that when the social, language-based aspects of the human mind do rear an ugly head, cognitive psychologists tend to cloak them in mechanistic jargon. For example, when describing the act of recollection, cognitive psychologists prefer to talk about "cue generation" under the guidance of a "central executive", with information being held in "articulatory loops" and "visuo-spatial scratchpads" - all terms with the reassuring ring of being computational processes.

Turning to the burning question of consciousness itself, what new model of awareness would emerge from taking a dual reductionist approach? Probably the most important change is that it lifts a huge weight off the shoulders of neurology, making consciousness something far simpler to explain. Instead of having to account for the elaborate superstructure of the higher mind, including such language-based constructs as reflective awareness and the controlling ego, neurology only has to consider the naked animal machinery of awareness.

The tendency to treat the mind as a package, a closed biological system, means that normally we think of self- awareness as being an inherent property of consciousness. But really the ability to introspect is a trick of language-controlled memory. The animal mind is locked into the present, its contents being a direct reflection of whatever happens to be occurring at the moment. So an animal would be aware, but have no special awareness of being aware. But humans can use inner speech to call to mind previous states of awareness; previous thoughts, feelings and actions. We can live the moment, then relive it, creating the eerie illusion that consciousness is also conscious of itself.

The true "nakedness" of consciousness only becomes apparent when we do something so fast and reflexive, such as hitting a tennis volley or catching a falling glass, that we do not have the usual time to both act and review the act as it is unfolding. What follows is that, contrary to common assumption, humans are not automatically self-aware beings. We have to learn the trick of reflection - and undoubtedly we learn the skill with varying degrees of success, some people becoming more skilful in, or devoted to, the ways of introspection.

Furthermore, if we want to understand how self-awareness is formed in children, we will have to look to linguistics and social psychology for an explanation. It is already clear from the work of Vygotsky and other noted thinkers of the 1930s, like Mead and Whorf, that much of the necessary training is implicit in our culture. The very fact that children need to learn to handle words like "I" and "you" in conversation forces them into a realisation of their own existence. Then society's demand that they learn to become self-directed individuals, able to give a reasoned account of their actions, places further pressure to develop a habit of introspection and a sense of self.

Once the concept of consciousness has been stripped of the many puzzling qualities we normally associate with it, such as introspection and deliberation, science is freed to chase a much more specific target. Consciousness boils down to nothing more than a single, core brain process - the mapping of sensations which creates the primary field of experience, followed by the recognition and association processes that make this field seem meaningful and understood.

The first stage of the process, the mapping of a field of sensations, is already reasonably well documented. The development of medical brain scanners which can "photograph" patterns of neural activity and blood flow as they sweep across the brain has revealed in detail how the images striking our eyes are recreated as matching patterns of nerve firing in the brain's visual centres. A cat's cradle of connections is tugged into existence, forming a living model of the world inside our heads. But it is the next stage of the process which has been rather neglected - the forging of a link between these fleeting sensory maps and the stored memories needed to make sense of them. An image of a cow or a car is just a clutter of shapes and colours until viewed against the backdrop of experience. This means that many times a second, the brain must be finding the connection between each newly arrived image and an answering patch of memory. It is out of this rapid sequence of matches that a coherent stream of consciousness emerges.

With consciousness pared down to a simple cycle of activity - a cascade of mapping, recognition and association processes - it begins to look like the orderly flow of neural events that reductionist science always believed it to be. Neuroscientists have already dubbed the 1990s the Decade of the Brain because of the promise of the discoveries to be made with new technologies, such as medical brain scanners. Yet there is a big difference between finding patterns of brain activity and interpreting them. The mind sciences are as much in need of a conceptual revolution as a technological one. Encouragingly, the current surge of interest in the writings of Vygotsky and others who are looking at the mind as part biological machine, part cultural habit, suggests such a change may now be in the air.

a circular argument?

There are two major difficulties with the argument that humans use inner speech to shape thought. First, the position seems circular - how could we know what words to use to create a thought unless we had already thought about what to say? For instance, how could we decide to say the word dog in order to conjure up a mental image of a dog unless we had already got the urge to do so from somewhere? The second puzzle is why is it that stroke victims can almost completely lose the ability to speak without apparently affecting their ability to think?

An easy answer to the accusation of making a circular argument is that really we are describing a circular process. The process we dub thought is a loose feedback cycle in which words prompt associations and associations, in turn, prompt further words. Some association must have led us to mention the word dog in the first place. Then having called up a dog image, further associations will lead us to the next words we say. If often words and images seem to arrive together in our minds in a tangle, this is just because the cycle of thought moves so quickly it becomes a blur.

However, although this is almost certainly part of the explanation, the full story must be a lot more complex. Much of the interaction between language and the natural powers of the brain take place at a deeper level, in "out-of-sight" sub-conscious processes, and what seems conscious and deliberate about the process is just the tip of the iceberg. This ties in with the second puzzle. It is true that many stroke victims can think but not speak, but the reverse is also true. The Soviet neurologist, Alexander Luria - Lev Vygotsky's closest colleague - described many patients with damage to their frontal lobes who could repeat words spoken to them but could not generate any conversation of their own. These patients would lie immobile on their beds, apparently robbed of all thought or will. Luria concluded that the frontal lobes must be where the planning of a speech act begins and it is only when this is knocked out that inner speech, and hence thought, is paralysed.

how language evolved

Originally, language would have developed as a tool of social organisation. All apes use expressive cries and gestures to help order their social worlds and early hominids, being particularly large-brained and social animals, undoubtedly made a sophisticated use of expressive noises. Over the course of a million years, they probably even developed a protolanguage - a system of single word utterances elaborated by the use of intonation and gesture.

However recent fossil findings and computer simulations suggest that hominids like Homo erectus lacked the physical equipment to take these first dabblings with language any further. It was only with the appearance of Homo sapiens, about 100,000 years ago, that biology caught up with behaviour and we became especially adapted for speech. The palate arched and the tongue and voice box shifted to allow for a more precise articulation of sounds. We also evolved specialised brain centres which could decode the particular sounds of speech at a rate of more than 20 a second - a "handling rate" two or three times greater than for any other kind of sensation, including non-speech sounds.

Having perfected speech as a tool of social communication, our ancestors found they could turn the organising power of language around on themselves, using silent, self-addressed speech to prompt and stimulate what went on inside their own heads. From this simple feedback trick flowed all the higher mental abilities that distinguish humans from animals, such as self-awareness, recollective memory, the civilised emotions, selective attention and reflective thought.

home> back to readings