The Ghost in the Machine
* A participant in the Stanford Seminar.
I have deliberately chosen these examples of improvisation by insects because the flexible skills of the higher mammalians are more familiar. Even fishes, according to Thorpe, can change their habits: 'If their normal behaviour-pattern is continually interfered with, quite large modifications in the normal instinctive orientation may be made.' [3] As for birds, in some species the male, who normally never feeds the young, starts doing so in the absence of the female. Lastly, I must briefly mention Lindauer's study of the honey-bee. We all know about von Frisch's discovery of the dance-language of the bee, but this is something different. Under normal conditions, there is rigid division of labour in the hive, so that each worker is occupied on different jobs in different periods of her life. During her first three days she cleans the cells. For the next three days she feeds the older larvae with honey and pollen. After that she feeds the younger larvae (who require an additional diet). From the age of ten days she is engaged in building cells; at twenty days she takes over guard-duties at the entrance of the hive; finally she becomes a forager, and remains one for the rest of her life.
That is, if all goes well. However, if any of the specialised age groups is taken away from the colony by the experimenter, other age groups take over their duties 'and thus save the superorganism. When, for instance, all foragers are taken away -- usually bees of twenty days or over -- young bees of scarcely six days old, who normally would feed the larvae, fly out and become foragers. If all building workers are taken away, their task is taken on by older bees who have already been builders before, but who had gone on to the stage of forager. To this end they not only change their behaviour, but also regenerate the wax-glands. The mechanisms of these regulations are not known.' [4]
Thus at one end of the scale we find fixed action-patterns and rigid compulsive rituals; at the other end surprising improvisations, and the performance of feats which seem to go far beyond the animal's repertory of habitual skills.
The Mechanisation of Habits
In man, innate instincts are merely the foundation on which learning will build. While learning a skill we must concentrate on every detail of what we are doing. We learn laboriously to recognise and name the printed letters of the alphabet, to ride a bicycle, to hit the right key on the typewriter or on the piano. Then learning begins to condense into habits: with increasing mastery we read, write, type 'automatically', which means that the rules which control the performance are now applied unconsciously. Like the invisible machinery which transforms inarticulate thoughts into grammatically correct sentences, so the canons of our manipulative and reasoning skills operate below the level of awareness, or in the twilight zones of awareness. We are obeying the rules without being able to define them. In so far as our reasoning skills are concerned, this situation has its obvious dangers: the axioms and prejudices built into the canon act as 'hidden persuaders'.
There are two sides to this tendency towards the progressive mechanisation of skills. On the positive side, it conforms to the principle of parsimony or 'least action'. By manipulating the wheel of the car mechanically I can give all my attention to the traffic around me; and if the rules of grammar did not function automatically, like a programmed computer, we could not attend to meaning.
Mechanisation, like rigor mortis, affects first the extremities -- the lowest subordinate branches of the hierarchy. But it also has a tendency to spread upward. To be able to hit the right key of the typewriter 'by pure reflex' is extremely useful, and a rigid observance of the laws of grammar is an equally good thing; but a rigid style composed of clichés and prefabricated turns of phrases, although it enables civil servants to get through a greater volume of correspondence, is certainly a mixed blessing. And if mechanisation spreads to the apex of the hierarchy, the result is the rigid pedant, slave of his habits -- Bergson's homme automate. First, learning has condensed into habit as steam condenses into drops; then the drops have frozen into icicles. As v. Bertalanffy wrote: 'Organisms are not machines, but they can to a certain extent become machines, congeal into machines. Never completely, however, for a thoroughly mechanised organism would be incapable of reacting to the incessantly changing conditions of the outside world.'
One Step at a Time
Thus the mechanisation of habits can never transform even an 'organisation man' into an automaton; but conversely, the conscious ego can interfere to only a limited extent with the automatic functioning of the subordinate units of his body and mind. The driver at the wheel can control the speed of his engine, but has no power to interfere with the order in which the cylinders fire, the valves open and close; and the conscious ego is in a similar position. It has no control whatsoever over functions on the sub-cellular or cellular level. It has no direct control over smooth muscles, viscerae and glands. Even the co-ordination of 'voluntary' skeletal muscles is only to a limited extent under conscious control: one cannot alter at will one's characteristic gait, gestures, handwriting.
We have seen that when a conscious intent is formed at the apex of the hierarchy, such as 'Unlock that door' or 'Sign that letter', it does not activate individual muscle contractions, but triggers off patterns of nerve impulses which activate sub-patterns, and so on, down to the single motor units. But this can only be done one step at a time. The higher centres in the hierarchy do not normally have direct dealings with lowly ones, and vice versa. Brigadiers do not concentrate their attention on individual soldiers, and do not give them direct orders; if they did, the whole operation would go haywire. Commands must be transmitted by what the army calls 'regulation channels' -- i.e., step by step down the levels of the hierarchy. Attempts to short-circuit the intermediary levels -- to turn the focal beam of awareness on the obscure and anonymous routines of lowly holons -- usually end in the paradox of the centipede. When the centipede was asked in which precise order he moved his hundred legs, he became paralysed and starved to death, because he had never thought of it before, and had left it to the legs to look after themselves. We would share a similar fate if asked to explain how we ride a bicycle.
The paradox of the centipede derives from a breach of what one might call the 'one step at a time rule'. On the face of it it looks trivial; but it leads to some unexpected consequences, if we try to go against it. Thus the pseudo-explanations of language as the manipulation of the vocal cords or the chaining of operants leave a gaping hole between thinking and spelling, between the apex of the tree and its terminal branches. The rule also has some applications to psychopathology -- from the awkward condition we call (by a misnomer) self-consciousness, to psychosomatic disorders. Self-consciousness (gaucheness, stage fright) results when conscious attention interferes with routines which under normal conditions are performed unconsciously and automatically. More serious disorders can result when attention is concentrated on physiological processes which function on even more primitive levels of the hierarchy, such as digestion and sex, and which must be left 'to look after themselves' if they are to function smoothly. Psychological impotence or frigidity, and spastic colons, are distressing variations of the paradox of the centipede.
Loss of direct control over processes on lower levels of the body hierarchy is part of the price paid for differentiation and specialisation. The price is of course worth paying so long as the individual lives under fairly normal conditions, and can safely rely on his more or less automatised routines. But conditions may arise when this is no longer the case, and it becomes imperative to break with routine.
The Challenge of Environment
This brings us to a point of vital importance which I have so far not mentioned: the influence of the environment on the flexibility or rigidity of behaviour.
If a skill is practised in the same unvarying conditions, following the same unvarying course, it tends to degenerate into stereotyped routine, and its degrees of freedom freeze up. Monotony accelerates enslavement to habit; it makes the rigor mortis of mechanisation spread upward in the hie
rarchy.
Vice versa, a changing, variable environment demands flexible behaviour, and reverses the trend towards mechanisation. The skilled driver on the familiar road from his home to his office hands over to the automatic pilot in his nervous system, while his thoughts are somewhere else; but if he gets into a tricky traffic situation, he will suddenly concentrate on what he is doing -- the man takes over from the computer. However, the challenge of the environment can exceed a critical limit where it can no longer be met by skilled routine, however flexible -- because the customary 'rules of the game' are no longer adequate to cope with the situation. Then a crisis arises. The outcome is either a breakdown of behaviour -- 'when in danger or in doubt, run in circles, scream and shout'. The hierarchy has disintegrated. The alternative possibility is the sudden emergence of new forms of behaviour, of original solutions -- which, as we shall see, play a vital part in both biological evolution and mental progress.
The first possibility is demonstrated by the cat which, unable to comply with the strict rules of its canon of hygiene, goes through the pointless motions of trying to bury the mess under the hard kitchen tiles. Human beings in a crisis are capable of equally senseless behaviour, repeating the same hopeless attempts to get out of it.
The alternative possibility is demonstrated by the unexpected improvisations of the digger-wasp, the reorganisation of labour in the mutilated beehive -- or a chimpanzee breaking a branch from a tree to rake in a banana beyond the reach of its arm. 'Original adaptations' of this kind, to meet challenges of an exceptional nature, point to the existence of unsuspected potentials in the living organism, which are dormant in the normal routines of existence. They foreshadow the phenomena of human creativity, to be discussed in Chapter XIII.
Summary
On successively higher levels of the hierarchy we find more complex, flexible and less predictable patterns of activity, while on successively lower levels we find more and more mechanised, stereotyped and predictable patterns. In the language of the physicist, a holon on a higher level of the hierarchy has more degrees of freedom than a holon on a lower level.
All skills, whether derived from instinct or learning, tend with increasing practice to become mechanised routines. Monotonous environments facilitate enslavement to habit; while unexpected contingencies reverse the trend, and may result in ingenious improvisations. Critical challenges may lead to a break-down of behaviour or to the creation of new forms of behaviour.
The higher echelons in a hierarchy do not normally communicate directly with lowly ones, but through 'regulation channels', one step at a time. A short-circuiting of intermediary levels may cause disorders of various kinds.
Part Two
BECOMING
IX
THE STRATEGY OF EMBRYOS
Benjamin Franklin's reply to a lady who queried the usefulness of his work on electricity: 'Madam, what use is a new-born baby?'
The classical Darwinian answer to the question how man was created out of a blob of slime is much the same as Watson's answer to the question how Patou creates a gown out of a piece of silk: 'He pulls it in here, he pulls it out there, makes it tight or loose at the waist. . . . He manipulates his material until it takes on the semblance of a dress. . . .' The evolutionary process is supposed to operate by similar random manipulations of its raw material -- pulling it in here, pushing it out there, putting a tail on here, putting an antler there -- until 'a pattern is hit upon', fit to survive.
Flat-earth science explains mental evolution by random tries, preserved by selective reinforcement (the stick and the carrot), and biological evolution by random mutations (the monkey at the typewriter) preserved by natural selection. Mutations are defined as spontaneous changes in the molecular structure of genes, and are said to be random in the sense that they have no relation whatsoever to the organism's adaptive needs. Accordingly the great majority of mutations must have harmful effects, but the few lucky hits are preserved because they happen to confer some small advantage on the individual; and given sufficient time, 'anything at all will turn up'. 'The hoary objection', Sir Julian Huxley wrote, 'of the improbability of an eye or a hand or a brain being evolved by "blind chance" has lost its force' -- because 'natural selection operating over the stretches of geological time' [1] explains everything.
In fact, however, the hoary objection has been steadily gaining in force during the mid-century decades -- so much so that there is hardly a prominent evolutionist alive who has not expressed some heretical views concerning some particular aspect of the orthodox doctrine -- while staunchly rejecting the heretics of others. Although these criticisms and doubts have made numerous breaches in the walls, the citadel of neo-Darwinian orthodoxy still stands -- mainly, one supposes, because nobody has had a satisfactory alternative to offer. The history of science indicates that a well-established theory can take a lot of battering and get into a tangle of absurdities and contradictions, yet still be upheld by the Establishment until an acceptable global alternative is offered.* But historically the only serious challenge to neo-Darwinism came from Lamarckism; and Lamarckism had much valid and scathing criticism, but no constructive alternative to offer.
* See Thomas Kuhn's thesis of 'Paradigm-Change' [1a], and the chapter on 'The Evolution of Ideas' in The Act of Creation.
Indeed, for nearly a hundred years, the theorists of evolution have been fighting an embittered Civil War of Lamarckian Roundheads versus Darwinist Cavaliers. The actual dispute was of a complex, technical character; but it was highly charged with metaphysical, emotional and even political implications. In the Soviet Union, the Darwinian Cavaliers were summarily sent to labour camps under Stalin, and the survivors summarily rehabilitated under Khrushchev: an episode known as the 'Lysenko affair'. The main issue -- over-simplified and put in a nutshell -- is this: Lamarck believed that the adaptive modifications of physique and ways of life, which an animal acquires to cope more effectively with its environment, are transmitted by heredity to the offspring ('inheritance of acquired characteristics'). Thus if a boxer develops strong muscles by training, then his son, according to Lamarck, ought to be born with strong muscles. This would provide a sensible and reassuring view of evolution as the cumulative result of learning through experience and training for a better life; but unfortunately, as so often happens, the commonsense view turned out to be inadequate. To this day, in spite of great efforts, Lamarckism has failed to produce conclusive evidence to prove that acquired characters are transmitted to the offspring; and it seems fairly certain that, while experience does affect heredity, it does not do so in this simple and direct way.
But the failure of Lamarckism in its primitive form does not mean that the monkey at the typewriter is the only alternative to choose. Random mutations, preserved by natural selection, without doubt play a part in the evolutionary process -- just as lucky coincidences play a part in the evolution of science. The question is whether this is the whole truth, or even the most important part of the truth.
A number of corrections and amendments to neo-Darwinian theory have been proposed by evolutionists over a number of years; and if these were to be put together, there would be little left of the original theory -- as amendments to a Parliamentary bill can reverse its emphasis and intent. But, as already said, each critic had his particular axe to grind, with the result that ''Tis all in pieces, all coherence gone' -- as John Donne lamented when medieval cosmology was landed in a similar crisis. In this and the next three chapters I shall collect some of these bits and pieces, and attempt to fit them together.