Chapter 4: Free Will in a Hierarchic Context by Arthur Koestler, Responses by Charles Hartshorne and Bernhard Rensch

Mind in Nature: the Interface of Science and Philosophy
by John B. and David R. Griffin Cobb, Jr.

Chapter 4: Free Will in a Hierarchic Context by Arthur Koestler, Responses by Charles Hartshorne and Bernhard Rensch

Arthur Koestler has published 30 books, ranging from political novels to theoretical works on creativity, evolution and parapsychology. His works are now being republished in a collected edition (The Danube Edition) by Hutchinson in London and by Macmillan and Random House in New York.

It is a common experience that you hear or read a sentence which remains implanted in your memory like a seed beneath the snow. Such an implantation happened to me many years ago when I read Bergson’s phrase: "The unconsciousness of a falling stone is something quite different from the unconsciousness of a growing cabbage." Comparing this to the Cartesian brand of dualism which restricts consciousness to human thinking, we are faced with two diametrically opposed conceptions: the first implies pan-psychism in one form or another, a continuum stretching from the rudiments of sentience in the cabbage to human self-awareness, the second places a kind of iron curtain between matter and mind.

This curtain seemed to be raised by a few inches in the nineteen twenties, in those heroic days when de Broglie and Schroedinger de-materialized matter like the stage magician who makes the lady vanish from the box, while Heisenberg (1969) eased her out of the straitjacket of determinism and proclaimed that the principle of complementarity agreed "very nicely" with the mind-body dualism -- the implication being that the particle aspect of the electron was analogous to the body, its wave aspect to the mind.

This led to a quasi-schizophrenic split in the Zeitgeist to which David Bohm has also alluded in his paper: I mean the contrast between the developments in physics on the one hand, and of academic psychology on the other. While the dominant school in physics emphasized the unpredictability of the fundamental processes in nature, the dominant school in psychology proclaimed its programme "to predict and to control human activity" (Watson 1925, p. 11). At the time when Heisenberg (1969, pp. 63-64) wrote that "when we descend to the atomic level, the objective world in space and time no longer exists" -- at the same time Skinner (1953, pp. 30-31) proclaimed that "since mental events. . . . are asserted to lack the dimensions of physical science, we have an additional reason for rejecting them." And when Eddington coined his celebrated aphorism "the stuff of the world is mind-stuff," the Behaviourists declared the mind to be a dirty four-letter word.

Now the Zeitgeist seems to be moving away from naive Reductionism, but as yet no coherent alternative system seems to be in sight. The indeterminacy of the micro-level cannot be transferred, through a short-cut, to the macro-level. God may be playing dice with the world, but He sees to it that it should be a fair game where the capers of the dice must conform to the Gaussian curve or the Poisson distribution. This may be an even more miraculous achievement than to control the balls on the Newtonian billiard table -- John von Neumann called the laws of probability sheer black magic. But they nevertheless prevent rash extrapolations from the indeterminacy on the quantum level to the robust determinacy of our everyday world.

A possible way out of this cul-de-sac is offered by the concept of multi-leveled hierarchic organisation. It seems to provide the missing link between pan-psychism and Cartesianism. The hierarchic model replaces the pan-psychist’s continuously ascending curve from protoplasmic to human consciousness by a series of discrete steps -- a staircase instead of a slope; and replaces the single Cartesian discontinuity by a series of transitions.

In his paper on biological hierarchy, H. H. Pattee (1970, pp. 117-136) wrote that hierarchic organisation is the central problem whether we consider the origin of life, philogeny or ontogeny. He went on: "It is the central problem of the brain, when there appears to be an unlimited possibility for new hierarchic levels of description. These are all problems of hierarchical organisation. Theoretical biology must face this problem as fundamental, since hierarchical control is the essential and distinguishing characteristic of life." Or, to quote Paul Weiss (1969, pp. 3f.): "The phenomenon of hierarchical structure is a real one, presented to us by the biological object, and not the fiction of one speculative mind." It is at the same time a conceptual tool, a way of thinking, an alternative to the linear chaining of events torn from their multi-leveled context -- like a thread extracted from a Persian carpet and suspended in a vacuum. One might say that hierarchical thinking relates to linear causality as the theory of algebraic functions relates to elementary arithmetic.

At this point I must introduce a bit of jargon, essential to the particular conception of hierarchy that I have described elsewhere (1967) and summarised in a paper read at the symposium "Beyond Reductionism." If I may quote very briefly from it (Koestler 1969, pp. 1 92f.):

The organism in its structural aspects is not an aggregation of elementary parts, and in its functional aspects not a chain of elementary units of behaviour; it is to be regarded as a multi-leveled hierarchy of semi-autonomous [i.e., self-regulating] sub-wholes, branching into sub-wholes of a lower order, and so on. Sub-wholes on any level of the hierarchy are referred to as holons. Parts and wholes in an absolute sense (as envisaged respectively by Behaviourists and Gestaltists) do not exist in the domains of life. The holon, as its name indicates, derived from the Greek holos, whole, with the suffix on, indicating a particle or part, is meant to be descriptive of a Janus-faced entity which displays both the quasi-autonomous, self-governing properties of wholes and the dependent properties of parts. The term holon may be applied to any stable sub-whole in a biological, social or cognitive hierarchy which displays rule-governed behaviour and/or structural Gestalt constancy.

I must refrain from going further into the cognitive framework of this particular model and confine myself to those basic aspects of the hierarchies of behaviour which are relevant to our present context.

All instinctive and learnt behaviour may be described in terms of hierarchies of sub-skills or subroutines, i.e., behavioural holons. The activities of these holons are rule-governed, but also adaptable within the constraints imposed by the rules. At this point it is essential to make a categorical distinction between (a) the fixed rules, and (b) the flexible strategies which between them control all organised activities of animal and man, regardless whether we consider instinctive behaviour, sensory-motor skills, or creative problem solving. It seems that the neglect of this fundamental distinction between invariant rules and variable strategies has led to much confusion in psychology, although the distinction becomes obvious when we turn to concrete examples. Take two creatures at opposite ends of the scale: a common spider spinning its web and a chess champion pondering his next move. The chess-player is controlled by the fixed rules of the game which define the permissible moves, but at the same time he is applying flexible strategies in his search for the most promising move among the permissible ones, guided by past experience and feedbacks from the environment, i.e., the position on the board. Thus the canon determines the rules of the game, strategy decides the actual course of the game. Now, the common spider too employs flexible strategies -- it will suspend its web from three, four, or more points of attachment, according to the lay of the land, but it will always arrive at the well-known symmetrical pattern where the radial threads bisect the laterals at equal angles according to genetically fixed rules of the game.

Let me now turn to the deceptively trivial experience that one and the same activity, such as tying my shoelaces, can be performed automatically, without conscious awareness of my own actions, or accompanied by various degrees of awareness. Driving along a familiar road with little traffic, I can hand control over to the ‘automatic pilot’ in my nervous system and think about eternity. In other words, there is a shift of control of the ongoing activity from higher to lower levels of the hierarchy. Vice versa, overtaking another car requires an upward shift to partial or full alertness and a rapid evaluation of risks which involves higher levels of the cognitive hierarchy.

The acquisition of a new skill, such as touch-typing, requires a high degree of concentration, whereas with increasing mastery and practice my fingers can be left to look after themselves ‘automatically,’ as we say. This condensation of learning transforms mental into mechanical activities -- mind processes into machine processes. It starts with infancy and never stops. To paraphrase Ludwig von Bertalanffy: we are not machines, but most of the day we act as machines. We do so even in such complex activities as adding up a column of figures. Karl Lashley once quoted a colleague of his, a Professor of Psychology, who told him: "When I have to give a lecture I turn my mouth loose and go to sleep."

Thus consciousness may be described in a negative way as that special attribute of an activity which decreases in direct proportion to habit formation. This is less paradoxical than it sounds, if one remembers that Norbert Wiener, following in Schroedinger’s footsteps, defined information as "essentially a negative entropy."

The transformation of learning into routine is accompanied by a dimming of the lights of awareness. We expect therefore that the opposite process will take place when routine is disturbed by running into some unexpected obstacle or problem: that this will cause a sudden switch from ‘mechanical’ to ‘minding’ or ‘mindful’ behaviour. Let a kitten suddenly cross the road on which you have been driving absentmindedly, and your previously absent mind will return in a flash, to take over control, to make a rapid decision whether to run over the kitten or risk the safety of your passengers. What happens in a crisis, or in any less dramatic problem situation involving unexpected, puzzling or discordant experiences, is this sudden shift of control of an ongoing activity to a higher level in the many-leveled hierarchy, from a semi-automatic to a more conscious performance, because the decision to be made or the problem to be solved is beyond the competence of the automatic pilot and must be referred to higher quarters. Needless to say, the implementation of the decision must still rely on automatised sub-routines -- braking or swerving -- which are triggered into activity by the command post.

It would seem that this sudden transfer of control of behaviour from a lower to a higher level of the hierarchy -- analogous to a quantum jump -- is the essence of conscious decision-making and of the subjective experience of free will. Its opposite is the mechanisation of routine, the enslavement to habit. We thus arrive at a dynamic view of a continual two-way traffic up and down in the hierarchy of behaviour.

Habit-formation implies a steady downward traffic, as on a moving escalator. There is a positive and a negative aspect to this phenomenon. On the positive side, it conforms to the principle of parsimony, or least action. If the laws of grammar and syntax did not function unconsciously, in the gaps between the words, we could not attend to meaning.

On the negative side, the downward traffic on the escalator tends to turn men into slaves of habit. The mechanisation of behaviour, like rigor mortis, affects first the extremities -- the lowest, subordinate branches of the hierarchy, comparable to the ethnologist’s fixed action patterns -- one’s characteristic gait, handwriting and accent of speech. However cleverly you try to disguise your handwriting, you cannot fool the expert. But mechanisation also tends to spread into the cognitive hierarchy, and force the stream of consciousness into fixed riverbeds -- Bergson’s homme automate programmed by a computer with built-in prejudices.

We can now turn from this all too familiar phenomenon to the upward traffic in the hierarchy: the abrupt transfer of the controls of an ongoing activity from a mechanical to a mindful level, and the concomitant experience of free will governing choice and decision. A striking description by Wilder Penfield of an experiment during a brain-operation may be relevant in this context (Penfield 1961):

When the neurosurgeon applies an electrode to the motor area of the patient’s cerebral cortex causing the opposite hand to move, and when he asks the patient why he moved the hand, the response is: "I didn’t do it. You made me do it". . . .

Once when I warned such a patient of my intention to stimulate the motor areas of the cortex, and challenged him to keep his hand from moving when the electrode was applied, he seized it with the other hand and struggled to hold it still. Thus one hand, under the control of the right hemisphere driven by an electrode, and the other hand, which he controlled through the left hemisphere, were caused to struggle against each other. Behind the ‘brain action’ of one hemisphere was the patient’s mind. Behind the action of the other hemisphere was the electrode.

It is interesting to compare the reaction of Penfield’s patient with the reaction of subjects who are made to carry out a post-hypnotic suggestion -- like taking off a shoe while having tea at 5 o’clock in the afternoon. In both cases the subject’s actions have been caused by the experimenter; but whereas the subject who does not know that he is obeying a post-hypnotic command automatically finds a more or less plausible rationalisation why he took off his shoe, Penfield’s patients realise that they are obeying a physical compulsion: "I never had a patient say, ‘I just wanted to do that anyway!’"

One must conclude that the hypnotist’s intervention affects a higher level of the mind-brain hierarchy than the surgeon’s needle. And to convince an opponent in a discussion that his theory is mistaken affects an even higher, more complex level. The step down from the level of reasoned persuasion to the level of hypnotic suggestion, or operant conditioning, means a transition from a highly mental to a comparatively mechanical level, and the same applies to the further step down to the neurosurgeon’s needle. But each transition is a relative, not an absolute, affair. To quote Thorpe (1966) at the Eccles Symposium: "The evidence suggests that at the lower levels of the evolutionary scale consciousness, if it exists, must be of a very generalised kind, so to say unstructured; and that with the development of purposive behaviour and a powerful faculty of attention, consciousness associated with expectation will become more and more vivid and precise."

However, these gradations in ‘structuring, vividness and precision’ exist not only in philogeny but also among members of the same species at different stages of development and in different situations. Each upward shift in the hierarchy produces more vivid and structural states of consciousness; each downward step is a transition from the mental to the mechanical. Classical dualism knows only a single mind-body barrier. The hierarchic approach implies a serialistic instead of a dualistic view; the transformation of physical into mental events, and vice versa, is effected not by a single leap over the barrier, but by a series of operational steps up or down the hierarchy.

Consider for a moment how we convert variations of air pressure arriving at the ear-drum, which are physical events, into ideas, which are mental events. It is not done in one go. In order to understand language we must perform a series of quantum jumps, so to speak, from one level of the speech hierarchy to the next higher one: phonemes are meaningless and can only be interpreted on the level of morphemes, words must be referred to context, sentences to a larger frame of reference. The spelling out of a previously unverbalised idea or image involves the reverse process: it converts a mental event into the mechanical motions of the organs of speech. This again is achieved by a series of steps, each of which triggers off pre-set neural mechanisms, activates linguistic holons of a more and more automatised type; the canons of grammar and syntax exert their rule unconsciously, automatically; and the last step in articulating the sounds of speech is performed by entirely mechanised patterns of muscle contractions. Chomsky’s phycholinguistic hierarchy was anticipated in A Midsummer Night’s Dream.

As imagination bodies forth

The forms of things unknown, the poet’s pen

Turns them to shapes, and gives to airy nothing

A local habitation and a name.

Each step downward in the conversion of airy nothings into spatial motions of the vocal chords entails a handing over of control to more automatised automatisms: each step upward in the reverse conversion leads to more mentalistic processes of mentation. All this seems to indicate that the mind-machine dichotomy is not localised along a single boundary, but is present on every level of the hierarchy.

On this view, the categorical distinction between mind and body fades away and instead of it ‘mental’ and ‘mechanical’ become complementary attributes of behavioural holons on every level. The dominance of one aspect or the other -- whether the holons’ activities are preformed mindfully or mechanically -- depends of course partly on its position or rank within the hierarchy, but also, and perhaps even more, on the momentary flow of traffic in the hierarchy, whether the shifts of control proceed in an upward or downward direction. Thus even the lower reaches of the hierarchy regulated by the autonomic nervous system can apparently be brought under mental control by bio-feedback or meditation; and vice versa, I can perform the supposedly mental activity of reading a printed page -- without taking in a single word.

We are in the habit of talking of mind as if it were a thing, which it is not -- nor is matter, for that matter. Minding, thinking, mentating are processes in a complementary or reciprocal relation to mechanical processes. Let me hasten to say that I am using here the physicist’s term ‘complementary’ as a very vague analogy, and not as an explanation. But as an analogy it has its uses. An electron will behave as a wave or a corpuscle according to circumstances. Andrew Cochran (1972, pp. 235-249) has suggested that the quantum-mechanical wave properties of electrons reflect a rudimentary consciousness of matter, and that the "dual aspects of man are a direct result of the dual aspects of matter." Here Cochran seems to be guilty of the sin which Whitehead called ‘misplaced concreteness,’ but at least the physicist’s perplexities are a comfort to the biologist; and though the complementarity principle leaves a host of problems unanswered, at least it poses a few new questions.

We know that the manifestations of the electron’s wave aspect or corpuscular aspect depends on the experimental set-up. In the cloud chamber it behaves as a corpuscle; in other experiments it is refracted as a wave. With animals and men it also depends to a large extent on the experimental situation -- the contingencies of the environment -- whether their mentalistic or mechanical aspect will dominate. Thus a skill practised and repeated in a monotonous environment in the absence of novelty tends to degenerate into mechanical routine. And vice versa, we have seen that a challenging environment which contains unexpected features or obstacles, and poses new problems which must be referred to higher echelons in the hierarchy, will cause a shift from mechanical to mindful behaviour. Such sudden upward shifts from mechanical routine towards originality and improvisation have been observed by ethologists throughout the animal kingdom from insects upward to birds, rats and chimpanzees. They may be regarded as precursors of human creativity, and point to the existence of unsuspected potentials in the organism which are dormant in the normal routines of existence but emerge in response to new challenges offered by the environment -- a zoological analogy to Toynbee’s paradigm of Challenge and Response.

To return to man: the ‘spelling out’ of an intention -- whether it is the expression of an idea or just the lighting of a cigarette -- is a process of triggering patterns of sub-routines into action, behavioural holons on subordinate levels; it is a process of particularisation of a general intent. On the other hand, the referring of decisions to higher levels is an integrative process which tends to establish a higher degree of coordination and wholeness of experience. Thus every upward shift would represent a quasi-holistic move, every downward shift a particularising move, the former characterised by heightened awareness and mentalistic attributes, the latter by diminishing awareness and mechanistic attributes.

How does free will fit into this tentative model? To repeat what has been said before: the subroutines on which our thoughts and actions are based are governed by fixed rules and more or less flexible strategies. The rules of chess define the permissible moves; strategy determines the choice of the actual move. The question is, how are these strategic choices made? Down on the visceral level, they are guided by the closed feedback loops of homeostatic regulations. Even such a complex sensory-motor skill as riding a bicycle is self-regulating in the sense that the cyclist’s strategy is governed by kinesthetic and visual feedbacks without the necessity of referring decisions to superior levels -- except if the road is barred. But when we enter the higher levels of the hierarchy, the relative importance of rules and strategies undergoes a subtle change. Compare playing noughts and crosses (tick-tack-toe) with playing chess. In both cases my strategic choice of the next move is ‘free’ in the sense of not being determined by the rules of the game. But noughts and crosses offers only a few alternative choices, guided by relatively simple strategies, which can even be codified to form a secondary set of non-statutory constraints, as it were. The chess-player, on the other hand, is guided by considerations on much higher levels of complexity with an incomparably larger variety of alternative choices, that is, more degrees of freedom. Moreover, the strategic precepts which guide his choice again form an ascending hierarchy. On the lowest level are tactical precepts such as occupying the centre squares, avoiding loss of material through forks and pins, protecting the king -- precepts which every duffer can master, but which the master is free to overrule by shifting his attention to the next higher levels of strategy, where material may be sacrificed and the king exposed in an apparently crazy move which, however, is more promising from the viewpoint of the game as a whole. Thus in the course of a game decisions have to be constantly referred to higher echelons with more degrees of freedom, and each shift upward is accompanied by a heightening of awareness and the experience of making a free choice. Generally speaking, in these sophisticated domains the constraining code of rules (whether of chess or of the grammar of speech) operate more or less automatically, on unconscious or preconscious levels, whereas the strategic choices are aided by the beam of focal awareness. In writing down a sentence, I do not have to worry about the rules of spelling, but ah, for the choice of the fitting adjective, the mot juste!

To repeat: the degrees of freedom in the hierarchy increase with ascending order, and each upward shift of attention to higher levels, each handing over of decision to higher echelons, is accompanied by the experience of free choice. But is it merely a subjective experience? I think not. After all, freedom cannot be defined in absolute, only in relative, terms, as freedom from some specific constraint. The ordinary prisoner has more freedom than one in solitary confinement; democracy allows more freedom than tyranny; and so on. Similar gradations are found in the multi-leveled hierarchy, where with each step upwards the relative importance of the constraints decreases and the number of choices increases.

However, this model will only be found useful if we assume that the hierarchy is open-ended toward infinite regress, both in the upward and downward direction. And there seems to be some justification for this assumption. Matter is no longer an ultimate concept; the hierarchy of macroscopic, molecular, atomic, subatomic levels trails away without hitting rock-bottom until matter dissolves into patterns of energy-concentration, and then perhaps into tensions in space. In the opposite direction we are faced with the same situation: there is an ascending series of levels, leading from automatic and semi-automatic reactions, through awareness and self-awareness, to the self’s awareness of its awareness of itself, and so on, without hitting a ceiling.

To put it in a different way: the higher the level to which the decision is referred, the less predictable the choice. We tend to assume that the ultimate decision rests with the apex of the hierarchy -- but the apex itself is not at rest. It keeps receding. The self, which has the ultimate responsibility for a man’s actions, eludes the grasp of its own awareness. Looking downward in the hierarchy of behaviour, a man is only aware of the task in hand, an awareness that fades with each step down into the dimness of routine, the darkness of visceral processes, the various degrees of unawareness of the growing cabbage and the falling stone, and finally dissolves in the indeterminacy of the Janus-faced electron.

But in the upward direction the hierarchy is also open-ended and leads to the infinite regress of the self. Looking upward, or inwards, man has a feeling of wholeness, of a core to his personality from which his decisions emanate, and which in Penfield’s phrase "controls his thinking and directs the searchlight of his attention" (Penfield 1961). But the metaphor is deceptive. When a priest chides a penitent for indulging in sinful thoughts, they both assume that behind the agency which controls his actual sinful thinking there is another agency which decides what subject he should think about; and so on ad infinitum. The ultimate culprit, the self which directs the searchlight of my attention, can never be caught in its focal beam. The experiencing subject can never fully become the object of his experience; at best he can achieve successive approximations. If learning and knowing consist in making oneself a private model of the universe, it follows that the model can never include a complete model of itself, because it must always lag one step behind the process which it is supposed to represent. With each upward-shift of awareness toward the apex of the hierarchy -- the self as an integrated whole -- it recedes like a mirage. "Know thyself" is the most venerable and the most tantalizing command. Total awareness of the self, the identity of the knower and the known, though always in sight is never achieved. It could only be achieved by reaching the peak of the hierarchy which is always one step removed from the climber.

This is an old conundrum, but it seems to blossom into new life in the context of the open-ended hierarchy. Determinism fades away not only on the subatomic quantum level, but also in the upward direction, where on successively higher levels the constraints diminish, and the degrees of freedom increase, ad infinitum. At the same time the nightmarish concept of predictability and predestination is swallowed up in the infinite regress. Man is neither a plaything of the gods, nor a marionette suspended on his chromosomes. More soberly, similar conclusions are implied in Karl Popper’s (1950) proposition that no information-processing system can embody within itself an up-to-date representation of itself, including that representation. Somewhat similar arguments have been advanced by Michael Polanyi (1966) and Donald McKay (1966).

Some philosophers dislike the concept of infinite regress because it reminds them of the little man inside the little man inside the little man. But we cannot get away from the infinite. What would mathematics, what would physics be, without the infinitesimal calculus? Self-consciousness has been compared to a mirror in which the individual contemplates his own activities. It would perhaps be more appropriate to compare it to a Hall of Mirrors where one mirror reflects one’s reflection in another mirror, and so on. Infinity stares us in the face, whether we look at the stars or search for our own identities. Reductionism has no use for it, but a true science of life must let infinity in and never lose sight of it.


Cochran, Andrew. 1972. "Relationships Between Quantum Physics and Biology." Foundations of Physics 1:235-249

Heisenberg, W. 1969. Der Teil und das Ganze. Munich: R. Piper.

Koestler, A. 1967. The Ghost in the Machine. New York: Macmillan.

Koestler, A. 1970. "Beyond Atomism and Holism -- The Concept of the Holon." In Koestler, A.. and Smythies, J. R. (eds.), Beyond Reductionism. New York: Macmillan, pp. 192-216. London: Hutchinson.

MacKay, D. M. 1966. "Conscious Control of Action." In Eccles, J. C. (ed.), Brain and Conscious Experience. New York: Springer-Verlag, pp. 422-445.

Pattee, H. H. 1970. "The Problem of Biological Hierarchy." In Waddington, C. H. (ed.), Towards a Theoretical Biology, Vol. III, pp. 117-136. Edinburgh: University Press; Chicago: Aldine Publishing Co.

Penfield, W. 1961. "The Physiological Basis of the Mind." In Farber, S. M., and Wilson, R. H. L. (eds.), Man and Civilization: Control of the Mind. New York: McGraw-Hill, pp. 3-17.

Polanyi, M. 1966. The Tacit Dimension. New York: Doubleday.

Popper, K. R. 1950. "Indeterminism in Quantum Physics and in Classical Physics." British Journal for the Philosophy of Science 1:117-133,173-195.

Skinner, B. F. 1953. Science and Human Behavior. New York: Macmillan.

Thorpe, W. H. 1966. "Ethology and Consciousness." In Eccles, J. C. (ed.), Brain and Conscious Experience. New York: Springer-Verlag, pp. 470-505.

Watson, J. B. 1925. Behaviourism. London: Kegan Paul & Co.; New York: People’s Institute Publishing Co.

Weiss, P. 1969. "The Living System -- Determinism Stratified." In Koestler, A., and Smythies, J. R. (eds.), Beyond Reductionism. New York: Macmillan, pp. 3-42; London: Hutchinson.



By Charles Hartshorne

Charles Hartshorne has taught philosophy primarily at the University of Chicago, Emory University, and the University of Texas; at the latter he is now Emeritus Ashbel Smith Professor of Philosophy.

A puzzling question to me is the topic currently discussed under such captions as ‘systems analysis’ (Bertalanffy, Laszlo), or ‘holograms’ (Bohm), or as ‘holons’ by Koestler. The mathematics of the discussion is, alas, beyond me. Informally I have some difficulty in grasping the distinction between:

1. Any member of a set of interacting individuals (each a sequence or society of momentary actualities) is influenced by all others;


2. Any member is influenced by the whole formed by the entire set.

Of course, an existing whole is no mere ‘sum of the parts,’ since each part or member interacts, has definite relations of influencing and being influenced by the others. Each member has its own perspective on the others. This is built into the psychicalist idea of ‘prehension,’ the relational aspect common to memory and perception, the aspect Leibniz failed to appreciate. But what is added by taking the totality of individual agents as itself an agent?

There is one very important qualification to the above. Given a system or composite of individuals, all on a comparable level, for instance molecular or cellular, then in some cases, given integration of activities (thus molecules in a cell, or cells in a vertebrate animal), there may be what Leibniz called a ‘dominant monad’ or what Whitehead calls a ‘society of presiding occasions.’ The sequence of experience in a waking vertebrate animal is the paradigm case. In one sense such a presiding individual is only another of the interacting agents, but in another sense it stands for the totality. It does this because its dominance consists in an exceptional adequacy of prehending and being prehended. Thus a cell in a nervous system interacts directly and strongly only with a few neighboring cells, whereas a human experience directly prehends and is prehended by a multitude of cells. It sums up to some extent the bodily totality. But it is not literally the totality. Also, in dreamless sleep there are no presiding occasions above the cellular level; the cells are on their own. And this seems to be true at all times of the constituents of the blood, such as the phagocytes. It is also probably true of the cells of a tree. Each cell, in such cases, responds to neighboring cells and to lower level singulars -- molecules, atoms, particles.


By Bernhard Rensch

Bernhard Rensch is at the Zoologisches Institut of the Westfaelischen Wilhelms-Unversitaet in Muenster, West Germany.

I fully agree with Dr. Koestler’s description of mental functions as a hierarchical order of different levels between ‘mechanical’ and mindful processes. But in my opinion it is possible to assume that even voluntary thinking can be regarded as a determined process. This means that we have not a ‘free will.’ When, for instance, a chess player has a choice between different moves, he will not only be guided by the rules of the play, but he will evaluate several possibilities. His decision is guided by the dominant idea to choose the most advantageous move. All our voluntary thinking is guided by such dominant ideas, when we want to solve a problem or plan an action. All voluntary thinking is motivated and this means ultimately that it is determined. However, it is often difficult to analyze the course of our thinking afterwards, because we have to do with an interaction of many different factors. Our thinking is guided by the abilities of our brain, by personal mental abilities or weaknesses, by the feeling tone of relevant associations, by memory traces which we have stored, by actual sensations, mainly by exciting ones, and by special moods. As this interplay is extremely complicated, we can not pretend to have certainty that our voluntary thinking and acting is not free, but we can say that this is most probable.

In simpler cases it is possible to judge the course of the brain functions. When I train a rat to discriminate a cross from a circle and to prefer the former one (because only this is rewarded during the training period), then I can predict that the rat will choose the cross later on, although now both patterns are rewarded. Or when I know a child very well and know what he has learned before in arithmetic, then I can predict with a high degree of probability the course of his considerations and their result when I offer him a new arithmetrical task which does not exceed the realm of his knowledge. The voluntary thinking processes of adult men are only too complicated to become sufficiently analyzed. And we can understand when Max Planck once said that even Kant’s deepest thoughts and Beethoven’s best musical productions were ultimately necessitated. However, later on Planck made some reservations because he became aware of the moral and juridical difficulties when we deny free will (cf. B. Rensch, Hippokrates 24: 1019-1032, 1963; B. Rensch, Biophilosophy, ch. 7A, New York: Columbia University Press, 1971).