SCIENCE: NUDGE NUDGE, WINK WINK

THE SHORTEST correspondence in history took place in 1862. Victor Hugo - famous for The Hunchback of Notre Dame - had gone on holiday following publication of his novel Les Miserables. But Hugo could not restrain himself from asking how the book was doing, so he wrote the following letter to his publisher: "?"

His publisher was not to be outdone and replied fully in keeping with the truth: "!"

As the Guinness Book of Records says of this reply, "the meaning was unmistakable" - certainly to Victor Hugo, and Les Miserables was indeed an enormous success.

It is fun to guess at what preceded the formulation of these two letters: on holiday, Hugo was surely wondering anxiously whether his great work would be appreciated by the public. Countless concerns led him to contact his publisher, but instead of writing, "Come on, damn it, tell me whether my book is selling!" he made do with that discreet question- mark. His publisher was probably ensconced in sales figures, reviews and accounts, from which he could have served up endless statistics, but he was tactful enough to know that they would not have helped; what Hugo wanted to know was simple. An answer such as "." could have ruined his vacation.

Undoubtedly, considerable thought preceded the letter-writing. Measured in units, a question-mark isn't much for a letter home. But it was not the number of units transmitted here that was decisive, but the context of the transmission. For Hugo and his publisher, the fate of Les Miserables was foremost in their minds. It filled their consciousness. The correspondence refers to a plethora of information - otherwise it would not be so full of meaning. Both messages represent many considerations - thoughts, feelings, and facts - which are not present, but nevertheless are.

This applies to any correspondence, of course. Before words are written, a considerable amount of mental work takes place. Not all of it is present in the words, yet it is present. So actual information in a correspondence, at face value, refers to a mass of information that is implicit.

In writing his question-mark, Victor Hugo refers explicitly to information the publisher is not told about in any way apart from the reference itself. Before the question-mark was put on the paper, Hugo discarded a mass of information that had flown through his consciousness. He refers explicitly to this information without including it in his letter.

Hugo's question-mark is the result of an explicit discarding of information. He has not simply forgotten it; in fact, he refers explicitly to what he has discarded, but it is still discarded. We will call such explicitly discarded information "exformation".

From the information content of a message alone, there is no way of measuring how much exformation it implies. Only the context can tell us that. The sender fashions the information contained in the message so that it refers to information he had in his head.

The puzzle of communication is how it can be possible: how can we refer, in information that we pass on, to a quantity of information that we discard? How can we chart our mental state in the form of some information? This is remarkable in itself, but is made no less remarkable by the fact that others have to use this chart in order to picture the terrain for themselves.

A good communicator does not think only of himself: he also thinks about what the receiver has in his head. And it is not enough just to refer to information in the sender's head if that information does not somehow lead to the correct associations in the receiver's.

The idea of transmitting information is to cause a state of mind to arise in the receiver's head that is related to the state of mind of the sender, by way of the exformation referred to within the information transmitted. The information transferred must elicit certain associations in the receiver.

Take the word "horse", for example. When an author writes "horse" he draws on a huge amount of personal experience. He's seen horses, he's read about horses, he's watched horses on television; he knows that people variously associate horses with beauty and sensuality, betting wins, and manure. From his memory he can summon up a vast amount of information related to horses.

Out of context, he can't expect that what is in his mind when he writes "horse" will have much to do with what you think when you read the word. But if he uses the word in a passage about the history of horse-racing, he can be pretty sure that he and his readers will have the same thing in mind.

"Cow." It is already apparent that we are not talking racetracks or symbols of wealth. We are talking about domestic animals: big, frightening, fascinating, nuzzling, amiable animals.

The author has excited a different space of association in your head. It does not take much for him to spark off associations in your head. But he has to think about what he is doing, and so do you. The transfer of exformation requires attentiveness.

Transfer of exformation ...? Has it not been discarded prior to communication? So surely it cannot be transferred during communications? How can something be transferred that is by definition absent from information nominally transferred? By writing "I did it my way" and "Frank Sinatra", how can an author strike up a very specific mood in your head and set emotions flowing through your mind and body? Likewise with "yesterday", "Christmas", "tax return"?

He can do so because he shares a vast number of experiences with his readers. They have all heard the same hits on the radio, taken part in the same rituals, and filled in tax returns. They are part of a context communicated through language. When the author writes a word, it is the result of an inner activity where lots of experiences flash through his consciousness. The reason he selects a particular word is that he senses it will arouse some of the same associations in you.

But he cannot be certain. Nor do you know what was in his mind when he wrote "Christmas". Perhaps he was just looking for a word he was fairly sure most people would respond to. Perhaps there was not much depth, not much exformation, in the word at all.

That is the risk of communicating: a receiver never knows how much information the sender has discarded. You never know how much exformation a given piece of information implies. It could be a bluff, or intellectual snobbery. Or indifference. Or the nine o'clock news. There is no guarantee that people listen to what they themselves are saying. Volumes of words can churn forth from people's lips (or fingers) without their "being there". If you make considerable effort to listen to them, you soon get cross - not necessarily because what they are saying is uninteresting, but because what you really wanted was to obtain a picture of what was going on in their heads, and you cannot do that when they supply information but no exformation.

The least interesting aspect of good conversation is what is actually said. More interesting are the deliberations and emotions that take place simultaneously in the heads of the conversers.

The words are merely references to something not present. Not present in the words, that is, but present in their heads. The idea of conversation is to elicit related states in another's mind and then exchange the events that take place. You don't believe it, you sympathise, you oppose, you are carried away, you remember, you love it, you love them, you miss them, you get ideas ...

Exformation is perpendicular information. Exformation is what is rejected en route, before expression takes place. Exformation is about the mental work we do in order to make what we want to say sayable. Exformation is everything we do not actually say but have in our heads when or before we say anything at all. Information is the measurable, demonstrable utterance we actually come out with.

The information content of a conversation is demonstrable, expressed, explicit. But the whole point of this explicitness is to refer to something else, something implicit, something unexpressed. Not just not present, but explicitly not present.

There is no conflict between information and exformation, but neither is there a link. A very brief message can contain enormous depth. A very long chat can contain enormous depth. But brief and extended messages can also be very superficial.

As concepts, though, they are linked. Exformation is the history of the message, information the product of that history. Each is meaningless without the other: information without exformation is vacuous chatter; exformation without information is merely discarded information.

In most contexts, it is difficult to decide what the exformation in a piece of information actually is. We can tell in the case of very precise messages: "I know somebody who has a rotary cultivator." In this case, the sender is obviously thinking about a digging job that would be easier if done by mechanical means and about a person who might be prepared to lend his implement. There is no reason to say a great deal about the person concerned; as long as the other digger understands that this person is sufficiently friendly to lend his rotary cultivator, that will do.

But we have no idea of the exformation in most of the messages we hear. We guess and sense and suspect - but we do not know. On the phone, it is harder to judge someone we don't know than it is face-to-face. But it can be done.

If we view conversations from without, as information exchanged, they are not especially rich. But if we see them from within, as exformation, they can be tremendous fun. If you do not know the context, they can be dull. It is very boring to lis- ten to people talking about someone you do not know. Such conversations tell you very little. But it is fun to talk about people you do know, whether personally or as public figures.

Information is not very interesting. The interesting thing about a message is what happens before it is formulated and after it has been received.

This does not apply merely to speech and writing but also to what happens when one listens to music: the way a composer converts something spiritual or intellectual into a score that can be played by fingers on keys, leading to sound waves that are sensed, experienced, and transformed into music in the ears of the listener.

The main thing in music is not the sound waves, it is that the composer converts a number of mental states into a pattern which evokes the same (or different) mental states in the listener. If we want to understand Bach or the Beatles, what we need to look at is not so much the information that is conveyed by the notes as the exformation that led them, the exformation that the notes evoke.

This line of thought is widespread in the study of the perception of music. David Hargreaves, a psychologist, has developed a theory of musical preference, described in New Scientist as follows: "The theory has its base in information theory, but the important insight comes from the distinction between this conception of 'information' and its psychological counterpoint. Fundamentally, the coding of physical information contained in a musical composition, as in information theory, predicts very little of interest, but coding the information in 'subjective' terms predicts a lot. Whether a person likes a piece or not depends on the information they are able to take out of it, rather than the information that is already 'in there'."

When we listen to music, certain states are created in our minds. They may be related to the state of the composer's mind when he wrote the music, but not necessarily. Music can provide access to happy states. Not necessarily because "they're playing our tune"; it may simply be that a kind of music or tune happens to put us in a good mood.

Blood-pressure readings and measurements of the electrical conductivity of the skin show that we really are affected by music: studies have even demonstrated that the same places in a score can affect different people in the same way.

Music is a method of conveying emotional states from a composer to an audience via sound. It can arouse mental states of which we may prefer not to be reminded, either because they are unpleasant or because we get depressed when we think about them. Music can arouse wonderful states: it can inspire energy, eroticism, calm, pensiveness, freedom, rebellion, sorrow, the urge to dance, pride, laughter, irritation, a feeling of belonging.

During live performances, the transfer goes both ways; an interplay takes place between musician and audience. The emotional states evoked in the audience affect the player (as, for example, their breathing, postures, and facial expressions change). For rare, glowing moments, a reverse coupling can thus take place in which the player expresses his state of mind through the music and sparks states in the audience related to his own, which is thereby enhanced.

But where does it come from, this ability to reconstruct information not directly present in the information we receive? And how do we ever learn to communicate in the first place? Children learn to talk - and to understand. It takes a few years, and nobody has any idea exactly how it happens. But we've all done it: we've learned to understand what a horse is. We've learned to listen to stories about horses and to picture what they're about.

Children's stories are a good example. Children love having stories read to them. They love hearing them again and again, while the grown-ups sit with them, reading.

Children love hearing the same stories over and over again because they are practising understanding. Along with the grown-up, they are learning the noble art of associating. Of guessing at the mental state of the author, packed with horses as it was, when he wrote the words.

But how can the child guess its way to more information than is present in the narrative? How can a tiny bit of information set off an avalanche of the stuff? How can exformation in the sender become the recollection of old information on the part of the receiver? Information (from previous experiences of horses) not present in the recei-ver's consciousness here and now, but which is nevertheless recalled?

How can information previously discarded from the consciousness during association to an idea be excited again, so that a narrator's exformation provokes memories of previously discarded information in the receiver? How can one couple the sender's information to recalled, excited exformation in the receiver?

The only real answer is "ask the kids". They are the only people capable of carrying out the unfathomable process of acquiring this ability. But we have all done so, we were all kids once. So even if we have forgotten how we acquired it, just as we've forgotten how we once learned to ride a bike (but not how to ride one!), perhaps we can reconstruct part of what must have happened.

At any rate, we can say this much: there must have been more information present during the process than that in the words that were actually spoken. Otherwise we would never have been able to guess what we were meant to think about upon hearing words. After all, we surely cannot produce information in our heads just because we hear a word we have never heard before - such as "erecacoexecohonerenit" - can we?

But children do learn. There must be something else present, something more than just the text, when it is read aloud. And indeed there is: a grown-up. Little children can learn from grown-ups. Over and over. There must, then, be something else present, something more than just words, in a context that can teach a child to speak. More than mere verbal information.

We must face another question: when we talk to one another, the talking is what we are aware of. It fills our consciousness. But if most of a conversation takes place beyond the talking, and the rest takes place in our heads, why are we not aware of it? How do our thoughts get sorted out before they emerge as speech? Is there a demon for sorting information? Is consciousness only the tip of a mental iceberg? Is consciousness just as heartrendingly meagre, and in all its self-importance just as helplessly comical, as information?

The answers must lie in the units. !

BOOK OFFER

Tor Norretranders' The User Illusion, from which this is extracted is available to IOS readers at the special price of pounds 18 (RRP pounds 20) inc p&p. Send credit card details or a cheque/postal order payable to Penguin Books Ltd to: User Illusion Offer, Penguin Direct, Bath Road, Harmondsworth, UB7 0DA. UK residents only. Allow 30 days for delivery. Offer closes 30 Sept