The Creative Imagination -- by Michael Polanyi

"The Creative Imagination" by Michael Polanyi [PSYCHOLOGICAL ISSUES (1969) 6: 59-91] explores the role of intuition and imagination in the growth of scientific knowledge.  Due to the richness of the material, we have decided to devote two successive posts to the article.  Polanyi (1891-1976) had first hand knowledge of the scientific process: he was an esteemed physical Chemist turned philosopher of science.  He proposes here in the first part that scientific knowledge like ordinary perception depends on an internal integration of a variety of clues, many we may not be aware of. We focus on the object and only have subsidiary awareness of the clues that determine our perception; because the integration occurs internally, he calls this "personal knowledge".  In ordinary perception, this depends on the development of a coherence among our various senses.  In a similar vein in scientific endeavor, like a ball rolling down an incline, we are guided by the slope of greater coherence leading to an experience of deeper meaning.   The second part of the essay considers how the standards on which this coherence is based can change in the context of scientific work.

Polyani’s own statement in the article provides the best bridge between his view of the nature of scientific knowledge and spirituality. “ Science is based on clues that have a bearing on reality. These clues are not fully specifiable; nor is the process of integration which connects them fully definable; and the future manifestations of the reality indicated by this coherence are inexhaustible. These three indeterminacies defeat any attempt at a strict theory of scientific validity and offer space for the powers of the imagination and intuition.”  

An earlier version of the article was given as a lecture at Wesleyan University in 1965 and is available online. 

…[N]either imagination nor intuition are deemed rational ways of making discoveries. They are excluded from the logic of scientific discovery, which can deal then only with the verification or refutation of ideas after they have turned up as possible contributions to science. (....)

However…[n]o scientific discovery can be strictly verified, or even proved to be probable. (....)

There is in fact no sharp division between science in the making and science in the textbook. The vision which guided the scientist to success lives on in his discovery and is shared by those who recognize it. It is reflected in the confidence they place in the reality of that which has been discovered  and in the way in which they sense the depth and fruitfulness of a discovery.  (....)

...[T]eachers in philosophy are likely to raise their eyebrows at such a vague emotional description of scientific discovery. (....)

[Yet] Copernicus discovered the solar system by signs which convinced him. But these signs convinced  few  others.  For  the  Copernican  system  was  far  more complicated than that of Ptolemy: it was a veritable jungle of ad hoc assumptions. (....) He did not stop to consider how many assumptions he had to make in formulating his system, nor how many difficulties he ignored in doing so. Since his vision showed him an outline of reality, he ignored all its complications and unanswered  questions. 

(....) In spite of its vagueness and its extravagances, his vision was shared by great scientists like Kepler and Galileo. Admittedly, their discoveries bore out the reality of the Copernican system, but they could make these discoveries only because they already believed in the reality of that system.                         

We can see here what is meant by attributing reality to a scientific discovery. It is to believe that it refers to no chance configuration of things, but to a persistent connection of certain features, a connection which, being real, will yet manifest itself in numberless ways, inexhaustibly. It is to believe that it is there, existing independently of us, and that for that reason its consequences can never be fully predicted.

Our knowledge of reality has, then, an essentially indeterminate content: it deserves to be called a vision. (….)  

This vision, the vision of a hidden reality, which guides a scientist in his quest, is a dynamic force. At the end of the quest the vision is becalmed in the contemplation of the reality revealed by a discovery; but the vision is renewed and becomes dynamic again in other scientists and guides them to new discoveries. I shall now try to show how both the dynamic and the static phases of a scientific vision are due to the strength of the imagination guided by intuition. 

(....) I have pursued this problem for many years by considering science as an extension of ordinary perception. When I look at my hand and move it about, it would keep changing its shape, its size, and its color but for my power of seeing the joint meaning of a host of rapidly changing clues, and seeing that this joint meaning remains unchanged.  I recognize a real object before me from my joint awareness of the clues which bear upon it. (....)

We can recognize here two kinds of awareness. We are obviously aware of the object we are looking at, but are aware also--in a much less positive way--of a hundred different clues which we integrate to the sight of the object. When integrating these clues, we are attending fully to the object while we are aware of the clues themselves without attending to them. We are aware of these clues only as pointing to the object we are looking at. I shall say that we have a subsidiary awareness of the clues in their bearing on the object to which we are focally attending.

While an object on which we are focusing our attention is always identifiable, the clues through which we are attending to the object may often be unspecifiable. (....) 

But it is a mistake to identify subsidiary awareness with unconscious or preconscious awareness, or with the Jamesian fringe of awareness. …[I]t can have any degree of consciousness so long as it functions as a clue to the object of our focal attention. (....)

If science is a manner of perceiving things in nature, we might find the prototype of scientific discovery in the way we solve a difficult perceptual problem. Take for example the way we learn to find our way about while wearing inverting spectacles. …[Y]ou feel completely lost and remain helpless for days on end. But if you persist...eventually [you] can even drive a car or climb rocks with the spectacles on. (....)

The inverted image has been reconnected to other sensory clues, to touch and sound and weight. These all hang together with the image once more, and hence, though the image remains inverted, the subject can again find his way by it safely. A new way of seeing things rightly has been established. (….)

We see how the wearer of inverting spectacles reorganizes scrambled clues into a new coherence. He again sees objects, instead of meaningless impressions. (....) He has made sense out of chaos.

In science, I find the closest parallel to this perceptual achievement in the discovery of relativity. Einstein (Schilpp, 1949, p. 53) has told the story of how from the age of 16 he was obsessed by the following kind of speculations. Experiments with falling bodies were known to give the same results on board a ship in motion as on solid ground. But what would happen to the light which a lamp would emit on board a moving ship? Supposing the ship moved fast enough, would it overtake the beams of its own light, as a bullet overtakes its own sound by crossing the sonic barrier? Einstein thought that this was inconceivable, and, persisting in this assumption, he eventually succeeded in renewing the conceptions of space and time in a way which would make it inconceivable for the ship to overtake, however slightly, its own light rays. (....)

Relativity alone involves conceptual innovations as strange and paradoxical as those we make in righting an inverted vision. (....)

We generally see things as we do, because this establishes coherence within the context of our experience. So when Einstein extended his vision to the universe and included the case of a light source emitting a beam, he could make sense of what he then faced only by seeing it in such a way that the beam was never overtaken, however slightly, by its source. This is what he meant by saying that he knew intuitively that this was in fact the case. (....)

Science is based on clues that have a bearing on reality. These clues are not fully specifiable; nor is the process of integration which connects them fully definable; and the future manifestations of the reality indicated by this coherence are inexhaustible. These three indeterminacies defeat any attempt at a strict theory of scientific validity and offer space for the powers of the imagination and intuition.

This gives us a general idea of the way scientific knowledge is established at the end of an inquiry; it tells us how we judge that our result is coherent and real. But it does not show us where to start an inquiry, nor how we know, once we have started, which way to turn for a solution. (….) 

This·quest is guided throughout by feelings of a deepening coherence and these feelings have a fair chance of proving right. We may recognize here the powers of a dynamic intuition. (….)

Physics speaks of potential energy that is released when a weight slides down a slope. Our search for deeper coherence is likewise guided by a potentiality. We feel the slope toward deeper insight as we feel the direction in which a heavy weight is pulled along a steep incline. It is this dynamic intuition which guides the pursuit of discovery.(....)

But we must yet acknowledge further powers of intuition, without which inventors and scientists could neither rationally decide to choose a particular problem nor pursue any chosen problem successfully. …[T]hink of Einstein, when as a boy he came across the speculative dilemma of a light source pursuing its own ray. (....) His intuition told him that there must exist a principle which would assure the impossibility of observing absolute motion in any circumstances. Through years of sometimes despairing inquiry, he kept up his conviction that the discovery he was seeking was within his ultimate reach. (....)

The power by which such long-range assessments are made may be called a strategic intuition. (....) Without this kind of strategic intuition, he would waste his opportunities on wild goose chases and soon be out of a job. (....)  

It is a skill for guessing with a reasonable chance of guessing right…. The fact that this faculty often fails does not discredit it; a method for guessing 10% above average chance on roulette would be worth millions.

But to know what to look for does not lend us the power to find it. That power lies in the imagination.

I call all thoughts of things that are not present, or not yet present--or perhaps never to be present---acts of the imagination. (….) 

Take the example of learning to ride a bicycle. The imagination is fixed on this aim, but, our present capabilities being  insufficient,  its  execution  falls  behind.  By  straining  every nerve to close this gap, we gradually learn to keep our balance on a bicycle. (....)

This is the mechanism to which I ascribe the evocation of helpful clues by the scientist's imagination in the pursuit of an inquiry. But we have to remember here that scientific problems are not definite tasks. The scientist knows his aim only in broad terms and must rely on his sense of deepening coherence to guide him to discovery. He must keep his imagination fixed on these growing points and force his way to what lies hidden beyond them. We must see how this is done. (....)

No  quest  could  have  been  more indeterminate  in its aim than Einstein's inquiry which led to the discovery of relativity. Yet he has told how during all the years of his inquiry, "there was a feeling of direction, of going straight towards something definite. Of course," he said, "it is very hard to express that feeling in words; but it was definitely so, and  clearly to be distinguished from later thoughts about the rational form of the solution." We meet here the integration of still largely unspecifiable elements into a gradually narrowing context, the coherence of  which  has not yet become  explicit.(....)