The Child that Books Built

by Francis Spufford

  It is good to know that real woods, even in England, hold very small areas that may not have changed since the wildwood. In leg-breaking gullies, on precipitous slopes, there are reservoirs of ancient vegetation perhaps a few feet square that have never been cleared. From here, lime seedlings or snake’s-head fritillaries replenish the wood around. These too are doors.

  *

  If the forest is where we go when, in Bettelheim’s words, we ‘lose the framework which gave structure to our past life’, we can go there at any age. Any time can be the time when structure collapses and the tangle of roots and branches surrounds us. The order we are living by ends without warning or after long struggle, it crashes or it dies of exhaustion. And for the third or fifth or umpteenth time the leaves are under our feet again. ‘In the middle of the journey of this life, I was travelling through a dark wood,’ begins the Divine Comedy. Dante’s neat balance on a medieval life’s midpoint makes him thirty. From the dark wood hell gapes, heaven beckons, purgatory promises refining fire. But however many times we return, there’s a first time for the forest. It is also the place we begin, as individuals: which perhaps explains the permanent temptation to line up childhood and primitiveness together, to try and combine the early time of one human and of a society. It is the place a baby is, before the developing mind has built up a model of things that it can rely on. It is the place we are before structure. Before we master speech, and can wield the power of names to distinguish the elements of the world – before we know for sure that our self has a boundary, and does not exist in a warm, milky continuum with everything else – we are in the forest. We don’t so much enter the wood as find ourselves there: knowing a little more that it is a wood, in fact, with each success at naming and placing. Gradually we recognise the dark uprights around us as trees. This first time, the forest doesn’t represent an emotional state. Dante’s dark wood is grown of emotions not disentangled, a spiritual state not articulated in the literal sense of not being split into the separate, hinged units of feeling that make us able to take the sad compound conditions of adulthood in a train, in a sequence, and so understand them. But for the infant coming to him-or herself in the wood for the first time, the problem is cognitive. Think of the cross-sections through the brain that a medical scanner produces: those false-colour images of activity, red and violet, yellow and orange. They map a greenwood inside our heads. When we’re born there are few paths established yet; the dendrites linking our neurons are a random tangle; the world is all to learn. From this forest, stories help to lead us out. The first stories we encounter, that is. The fragmentary sequences that first make a wavering, story-like sense, like the ‘story’ of a letter arriving through a letterbox. The stories in picture books read aloud to one-year-olds and two-year-olds. Eventually, the fairy tales which themselves are spiky with pine needles.

  *

  The only accounts of very young children are given by adults observing them. Not only can a child of one not speak to describe the time before language: they can’t remember it when they have become adults of thirty. Usually a person’s earliest memories go back to when they were three or four. Memory development is closely linked to language development. Not all memories are verbal – some of the most potent are gestalts of colour and mood that would be completely irreducible to words – but the ability to store absent sensation calls on the same faculties as the ability to represent the world in symbolic form. So the knowledge of the pre-linguistic forest has all been arrived at by adults working like anthropologists. They look at the strange tribe of infants, and try to deduce an inward order from the behaviour that they can see.

  The scope for fantasy on the adult’s part is great. In Through the Looking Glass Alice enters ‘the wood … where things have no names’. She meets a fawn. Without names, neither knows what they are.

  So they walked on together through the wood, Alice with her arms clasped lovingly round the soft neck of the Fawn, till they came out into another open field, and here the Fawn gave a sudden bound into the air, and shook itself free from Alice’s arm. ‘I’m a Fawn!’ it cried out in a voice of delight. ‘And, dear me! you’re a human child!’ A sudden look of alarm came into its beautiful brown eyes, and in another moment it had darted away at full speed.

  Anyone who has ever seen an eighteen-month-old in pursuit of a house cat knows that the real forest of unknowing is not the peaceable kingdom where the lion lies down with the lamb. Lewis Carroll was seduced by the adult wish for an alternative to the cynicism of adult self-consciousness. He describes, with cynicism and amusement, a place where there is neither, because in the absence of words everything is at rest. As a fantasy which reverses a fall it has a lot in common with Peter Pan. But early childhood is not a stable surface on which to project the dream of innocence – any more than later childhood can be fixed comfortably at a stage from which a real child would want never to grow up. The only child who never grows up is the child that adults imagine themselves being. A real child perpetually changes, has new experiences, has the same experiences in new ways; constantly moves on; declares that they are a subject in their own right. And the forest before language is neither innocent, nor silent. A toddler who has not yet spoken, or a baby who has not yet walked, is still reaching and receiving, ceaselessly sorting out the world, in physical ways that are also cognitive explorations. There is no point of origin for human beings, if by that we mean some stationary moment at which nothing is going on. From the moment when the fertilised cell starts dividing, there is always something going on.

  Jean Piaget, who was the great modern pioneer explorer of the purposefulness of young children, also insisted on the radical difference of their minds. After the first era of life when all knowledge is body knowledge – the ‘sensory-motor period’ in his terminology – a child who has discovered speech becomes ‘pre-operational’. Pre-operational children, he concluded from his work in the 1920s and 1930s on Swiss three-and four-year-olds, see the world in some distinctive ways. They are egocentric, for example, as revealed in the common belief that the sun and moon follow them when they go for a walk. They think that almost everything that moves is alive including clouds and cars, only gradually restricting life to those things that move of their own accord. (There’s an exception: right from the start, anything broken is dead.) Conversely, they believe that the natural world was all arranged as it is for human benefit – was artificially constructed to suit. A large part of the pre-school population of Geneva, it turned out, thought that Lake Geneva had been dug deliberately after the city was built, by men with spades. All these ‘distortions’, put together, make up a world neither coldly material nor alive with a life that is going about its own business indifferent to the child, but instead magically responsive, surrounding the child at its centre like the ring of painting on the inside of a zoetrope. But underlying these elaborate beliefs, Piaget argued, were difficulties in grasping reality’s elementary truths, the most basic laws governing the behaviour of things. The magical aspects of the world of a three-year-old did not present a fundamental challenge to the adult understanding. But these differences in perception down at the roots almost defied imagination.

  It is a law of the world for adults that the physical properties of the environment around you only change in predictable ways, by predictable processes (barring events arriving from completely outside the ordinary context of things, such as meteor strikes or earthquakes). For adults, the world is dynamic, but intelligible, at least at the level of everyday sense-perception. Pre-operational children, according to Piaget’s experiments, seemed not to know that things existed in dependable quantities. They had learned to recognise that an object kept its identity even when its appearance changed, so Daddy in a hat was still Daddy: that was one of the cognitive achievements they needed to graduate from the sensory-motor period. But they weren’t out of the woods yet. They didn’t yet know that physical attributes were also ‘conserved’, so that if the attributes changed, there was alway
s compensation for a shrinking in one dimension by growth in another. Show a pre-op child a row of buttons, then spread them out and the child would say that now there were more of them. Line up two identical wooden rods, and the child would agree they were the same length. Slide the lower one to the left, and suddenly to the child it became ‘longer’. Roll a ball of modelling clay out into a sausage, and Presto! It was ‘bigger’.

  It’s difficult to remember thinking like this, because as the child learns to conserve the different attributes of material things, in a predictable sequence leading up to the conservation of volume, which is hardest of all, each item once mastered becomes natural and inevitable, and impossible to un-know again. These are the ‘operations’ a child must master in Piaget’s model, before moving on to the next stage of development at six or seven: the traditional age of reason in many cultures, he noted. The operations were procedures of logic, but not as in mathematical logic. They were logic’s atoms, its ground-rules from which all else followed; its simplest moves, that let you deal with the changing world by knowing that if A is bigger than B, B must be smaller than A.

  One of Piaget’s most famous experiments was designed to demonstrate the pre-op child’s inability to handle another of the operations, ‘class inclusion’. Class inclusion is when one category of things nests entirely within a larger category of things. The class of ‘women’, for example, fits inside the class of ‘people’. The problem is to understand the relationship between part and whole. Piaget tested it with beads. He laid out a row of (say) nine brown and three white beads – enough for the imbalance between the two colours to be immediately striking. Then he asked: ‘Are there more beads, or more brown beads?’ Virtually without exception, pre-operational children replied that there were more brown beads. The older his test subjects were, and the closer they had come to the next era in their development, the more troubling they found the test, the more aware they were that something was awry, though they couldn’t put their finger on it. Piaget took this as proof that a child younger than six or seven is cognitively incapable of relating a whole to its parts. If this is true, it is not a trivial point of logic. Knowing how to fit phenomena into categories of different sizes is as fundamental to understanding the world as being able to tell what is bigger and what is smaller. It isn’t a question of young children not being able to say that a whole is greater than a part. This is about them not even being able to know so. If Piaget was right, the child who listens to a fairy tale doesn’t know there is more weather than there is sunshine, more pets than there are cats, more cutlery than there are spoons. If Piaget was right, then the magical forest in which a three-year-old will find it natural for animals to speak, is very dense indeed; is a thicket; because the child literally cannot tell the trees from the wood, the wood from the trees.

  As a set of metaphors, Piaget’s sequence of stages is unquestionably valuable. That’s how I use them throughout this book: his progression from pre-operative thought, to concrete operations, to abstract operations, gave me a loose conceptual map for my whole history as a reader. But for the last twenty or thirty years developmental psychologists have increasingly challenged Piaget on such crucial details as the class-inclusion problem, and their disagreement focuses on language. For a start, ‘Are there more beads, or more brown beads?’ is a question that could well catch out an adult, without implying that they hadn’t mastered class-inclusion. For Piaget, the only difficulty of the sentence is the underlying cognitive difficulty. To understand it correctly requires you to notice that ‘beads’ and ‘brown beads’ belong on different levels in a ladder of categories, which is what Piaget intended. But it also requires you to discount, or override, the very strong expectation set up by the form of the sentence itself that you are going to be comparing parallel categories of things, that is categories on the same level of the ladder. Children who get Piaget’s question wrong are clearly behaving as if he had asked a more natural question, one which goes with the grain of an ‘Are there … or are there …’ question, rather than against it. They are telling him whether there are more brown beads or white ones. Recent psychologists are not as comfortable as Piaget was with this deceptive linguistic cue.

  More fundamentally, they disagree over the role language itself plays in the mind’s development. Piaget thought that language only tracked the growth of understanding. In the Piagetian system, spoken language was the great achievement of the first two years of life, and by letting the child manipulate the world through a set of shared symbols, it immensely accelerated learning in every other area, so in that sense it was deeply implicated in every later achievement. But Piaget took it as common sense that a child learned something first, then was able to say it. Words expressed what you already knew. They didn’t contribute on their own account – or lead intellectual development in particular directions. If this were true, if language were transparent like this, then the form in which the bead question was cast (barring obvious difficulties of vocabulary) should not make any difference to whether a pre-operational child could answer it. In the early 1970s James McGarrigle set out to experiment with multiple versions of the class-inclusion problem, some altering the wording, some changing the other attributes of the test that might strike a child. He discovered straightaway that you could increase significantly the number of children who got the test right if you put in an extra adjective drawing attention to what the whole class had in common, as opposed to its black/white subsections. Lay out toy cows lying down, and ask if there were more black cows or more sleeping cows, and the proportion of right answers almost doubled. He also discovered – a finding that goes beyond the role of language but bears vitally on it – that the more the wording and the other aspects of the test were adjusted together to make the situation familiar, and the clearer the intention of the adult asking the question consequently grew to the children, the more able they were to direct their minds to the right contrast – black/total rather than black/white. Young children could pass the class-inclusion test if it was presented to them in particular ways; which proved that they were not defeated by the part-whole relationship down at its cognitive root, with the bizarre consequences for their inner worlds implied by Piaget, because that kind of complete inability would not have been altered by any alterations in the testing. What they could not do, it seemed, was easily comprehend the kind of language represented by ‘Are there more beads, or more brown beads?’

  McGarrigle’s colleague Margaret Donaldson, whose Children’s Minds became one of the most important statements of the revisionist case, pointed out that to answer Piaget’s original question a child had to ignore a set of things. They would have to ignore their own sense of what question it was probable you would ask about some white beads and some brown ones; and overlook the familiar sentence haunting the unfamiliar sentence that the questioner actually spoke; and reject any instinctive guess about what the questioner’s intention was likely to be. In short, the child had to ignore every piece of data except the exact wording of the question. This proved to be the very hardest thing for children to do. The work of Donaldson, McGarrigle and many others showed that small children understood language best when it was, so to speak, fused with a situation, and all the non-linguistic clues to hand could be brought to bear. They called this ‘embedded’ language; it is speech ‘embedded … in the flow of events which accompany it’. An example would be an adult saying ‘Come here’ while beckoning and smiling. A child would have to be very young, just on the cusp of articulate speech, to need non-verbal clues in that instance, but even in adulthood whether speech is embedded continues to affect how easy it is to understand. A proposition in the abstract terms of logic becomes strikingly easier for most people to follow once it is translated into words that refer to natural, concrete situations – although the conceptual content remains the same.

  This permanent inclination in humans to grasp words quicker when events sustain them, directs us back to a time when words and events ran much c
loser together in our minds, and we were ‘reading’ what happened around us without making hard-and-fast distinctions between what people said and what it seemed they wanted. To rip a question completely out of its context, and to consider it purely as a mechanism of words interacting to produce one precise meaning, requires a three-year-old or a four-year-old to go against this whole current in the mind. It is either very difficult, or actually impossible. ‘The child is not able to pay scrupulous attention to the language in its own right,’ concluded Donaldson. She pointed out that in the extreme case, if they’ve come from a family where they’re never encouraged to play with the spoken language, it is possible for pre-school children not even to know that separate words are what they are speaking. They can perceive language as a flow, a continuum, instead of a succession of units. Of course, the word as a unit is intrinsic to speech: our ability to talk at all, at any age, is dependent on a mental grammar that splits up our thoughts into distinct chunks of meaning that play different grammatical roles. But the process can be completely unconscious for a small child, since the last stage of utterance for a human being – the part you actually notice – is the further conversion of the grammatical string into sounds that do indeed flow continuously from throat, tongue, and lips. We float our words on ‘a river of breath’, as one scholar of linguistics put it, and we leave no audible spaces between them. So it may only be the river that a child hears.