The Net Advance of Physics RETRO:

2015 November 30

Ouroboros: Benzène by Haltopub, 2013.


The source of scientific creativity has always been controversial. In the earlier Nineteenth Century, under the influence of Romanticism, scientific "genius" was seen as a quasi-divine gift, like its counterpart in poetry. (As may be seen elsewhere on this site, Romantic poetry was often very pro-science, and a few scientists, notably Sir William Rowan Hamilton, were themselves Romantic poets.)

At the same time, there was a push in some circles (friendly as well as hostile) to view science as "cold" and data-driven. This attitude, enthusiastically adopted by biologists in particular, became the universally dominant one by 1900 or so, although many mathematicians and theoretical physicists continued to resist it.

In 1922, when the distinguished American historian of science Walter Libby gave the talk below, a neo-Romantic and philosophical approach to truth characterised the physics avant garde in Göttingen and Copenhagen, but most scientists in the English-speaking world viewed such metaphysical concerns with distaste. As we shall see, Libby considered the subjectivism of mainstream Victorian physicists like Tyndall nearly incomprehensible. From such a viewpoint, the irrational inspiration of much scientific progress was either an embarassment or, as Libby thought, a subject deserving the attention of psychological researchers.

It should be mentioned that Kekulé's dreams are more problematic than the account to follow suggests; the chemist gave several inconsistent accounts of his experiences over the years.


First of a series of lectures on the "Psychology and Logic of Research, " given before the Industrial Fellows of the Mellon Institute of Industrial Research of the University of Pittsburgh, February 14 - May 2, 1922
By Dr. Walter Libby, University of Pittsburgh.

[ Scientific Monthly 15, 263 (1922)].
Libby's words are in bold

IN books and articles touching on the psychology and logic of research, a certain confusion has frequently arisen from the use of terms like intuition, illumination, and inspiration, which seem almost to defy definition, as well as from an unwarranted use of terms like imagination and conception, regarding the denotation of which there is some approach to harmony among the recognized exponents of mental science.

Among the philosophers, Wundt, Bergson and James, for example, acknowledge --- each in his own way, to be sure --- a close relationship between the imagination and the memory. Both of these mental processes admit of analysis into simple sensory elements. Reproductive imagination differs indeed from memory only in so far as it is unaccompanied by a sense of repetition. The productive, or creative, imagination, though it differs from the reproductive in the freedom with which it manipulates and rehandles sensory data, is nevertheless as dependent as it on the materials furnished by the eye, ear and other sense organs. We may rearrange and recombine the data supplied by sensation and retained in consciousness ; we can create nothing absolutely new.

Nearly all of the chapter on imagination in James's Principles of Psychology would be equally relevant in a discussion of the memory. The point of view of this eminent philosopher and psychologist is so opposed to the views of writers like Tyndall and Pearson, who are inclined to identify the scientific imagination with creative thought in general (which it is our purpose to analyse), that it seems worth while to examine in some detail the phenomena of retention and recall, and, by differentiating one type of memory from another, obtain a clue to the various types of imagination in the strictest sense of that term.

Student in his Study by Jan de Heem, 1628. [Ashmolean Museum]

Cases of remarkable powers of visual recall have been put on record by James. One of these he quotes :

The more I learn by heart the more clearly do I see images of my pages. Even before I can recite the lines I see them so that I could give them very slowly word for word, but my mind is so occupied in looking at my printed image that I have no idea of what I am saying, of the sense of it, etc. When I first found myself doing this I used to think it was merely because I knew the lines imperfectly; but I have quite convinced myself that I really do see an image. The strongest proof that such is really the fact is, I think, the following:

I can look down the mentally seen page and see the words that commence all the lines, and from any one of these words I can continue the line. I find this much easier to do if the words begin in a straight line than if there are breaks. Example (La Fontaine, 8. iv) :

Étant fait ............
Tous .............
A des .............
Que fit .............
Céres .............
Avec .............
Un fleur .............
Comme ............

In an experimental study undertaken by the writer a group of ten college students were asked to memorize words and sentences in Italian, a language which none of them had studied before, and of which the experimenter was also fairly ignorant. Each of the eight exercises, employed within the space of two months, consisted of ten detached words and of about fifty words connected in sentences. The procedure was to place typewritten sheets of the words and sentences, with their translation, before the students for twenty minutes. The sheets were then collected and all copies and notes made during the study period were destroyed. Forty-eight hours later the members of the group were asked to write down all the words which could still be recalled.

In the third exercise of this sort one student succeeded in reproducing correctly nine out of the ten detached words and all of the words in the sentences. The spelling, the punctuation, and even the use of accents were almost perfect. This student was a well-marked type of motor memory. She found it impossible to memorize anything effectively without writing it down. To hold a pencil in the writing position aided her to some extent to fix in memory an ordered statement of ideas. When she had tried to learn the Italian words and sentences by visualization, they seemed quite strange to her after the lapse of forty-eight hours; but, when she had copied them down, they were as old friends.

In the seventh exercise a second student succeeded in recalling the ten detached words with one mistake in spelling and, with remarkable fidelity, a song of eleven lines from an Italian opera. He relied not on motor or visual, but on auditory imagery. He insisted at the beginning of the experiment on having the words and sentences read aloud. Later he was able, so he said, to surmise the sound of them. His mistakes corroborate his introspection concerning his type of memory. His spelling, punctuation and use of accents were less accurate than the first student's. He was able to give only the first syllable of a five-syllable word which was indistinctly pronounced by his rather incompetent instructor. The spelling luto for lutto was probably also due to the experimenter's defective pronunciation of Italian. In one line of the song the student elided two words by doubling the initial letter of the second word, but without deteriment to the rhythm. In this same exercise a third member of the group was able to recall only fourteen words out of sixty-four. His impression that his mind is of the visual type is supported by the fact that all of the words definitely remembered occur in conspicuous positions in the exercise --- the first line of a stanza, the end of a line, the beginning and the end of the list of words, etc.

If the claim is put forward that great scientific discoverers have been gifted with particularly vivid imagery, we must bear in mind the actual achievements of young people of college age submitted to definite tests. In this chance group of ten students, one, as we have seen, relying on auditory imagery, was able to recall sixty-four words out of sixty-four, while another, by means of kinesthetic imagery recalled sixty-three words out of sixty-four. Remarkable as their performances were, these students were surpassed in the total experiment by a student who was conscious (as many of us must be even in such a simple experience as holding a telephone number in consciousness for a few moments) of relying on both auditory and visual imagery.

Inspiration by William-Adolphe Bouguereau, 1898.

The address of the Irish physicist Tyndall on the "Scientific Use of the Imagination," delivered before the British Association in 1870, gives evidence of the functioning of his own imagination, and raises a number of questions in reference to the use of the imagination in scientific research. He is of the opinion that in explaining sensible phenomena we habitually form mental images of the ultra-sensible. He holds that the action of the investigator is periodic, and that the emotions play no inconsiderable part in the intellectual life. Tyndall quotes Sir Benjamin Brodie as stating that the imagination is both the source of poetic genius and the instrument of discovery in science.

When, however, Tyndall says that, with experiment and accurate observation to work upon, imagination becomes the architect of the theories of physical science, he seems to pass from the consideration of imagination in the strict sense of the term to the consideration of the speculative process involved in the setting up of hypotheses; and, when he claims that without the exercise of the imagination the conception of force would vanish from our universe and that causal relations would disappear, his enthusiasm for his theme has apparently rendered him oblivious of the distinctions between mental processes.

"There is," he proceeds, "in the human intellect a power of expansion --- I might almost call it a power of creation --- which is brought into play by the simple brooding upon facts." After this sample of amateur psychology one is not surprised to hear the physicist speaking of a composite and creative power in which reason and imagination are united. Having confused the imagination with the reason, he invents a tertium quid that includes them both. The example of scientific thought which he gives, concerning the process of developing analogies between the wavelets on the surface of a pond, sound waves in water or air, and light waves in the ether, does not further the differentiation of the imagination and the reason. He regards as a product of the imagination the inference that the people by whom we are surrounded are possessed of reason because they behave as if they were reasonable. For him the world of sense itself, the phenomenal world of the physicist, is largely the outcome of things intellectually discerned, and is, therefore, dependent on the imagination. In short Tyndall imparts to the term imagination the maximum extension and the minimum intension.

Francis Galton's essay on Mental Imagery, 1881, provides an antidote for the extreme views of Tyndall. In this essay Galton expresses the conviction that scientists as a class are not good visualizers. When he questioned his friends of the scientific world, including Fellows of the Royal Society and members of the French Institute, he was amazed to find that few of them could picture to themselves things recently seen, such as the breakfast-table at which each had sat a few hours previously. When, however, Galton addressed himself to persons whom he met in general society, he obtained results altogether different. Girls, boys, women, many men, could recall sights like the morning's breakfast-table with photographic vividness and in their appropriate coloring and illumination. The power to visualize is more marked in the female sex than in the male and is somewhat more active in adolescent boys than in men. A study of the drawings of the Bushmen of South Africa and of the remains of prehistoric art indicates that the visual imagery of primitive man may be of an almost hallucinatory vividness.

Convinced, by his systematic investigation of the comparative dearth of visual imagery among men of science, Galton arrives at the conclusion that habits of highly generalized and abstract thought, the pursuit of language and book learning, are antagonistic to the faculty of perceiving mental pictures. He admits, however, that there are instances in which persons see mentally in print every word uttered in a conversation or an address, and that the highest minds are probably those in which visualization is not lost but is held as a rule in abeyance, ready for use on suitable occasions. In fact, in later studies he records the remarkable visualizing powers of men like Professor Schuster, Flinders Petrie, and the Rev. George Henslow, botanist. Mr. Henslow recognized that visual images, which he could summon at will, differed from the original perceptions, and that they were dynamic, undergoing changes, in many cases due to a suggestiveness, in the images, of something else. At times the images oscillated or rotated in a perplexing manner.

Karl Pearson in The Grammar of Science, 1911, resembles Tyndall rather than Galton and James as regards the scope and range he ascribes to the activity of the imagination. According to Pearson the discovery of law is the peculiar function of the creative imagination. He declares that the man with no imagination may collect facts, but that he cannot make great discoveries. After an elaborate classification of such facts has been made and their relations and sequences carefully traced, the next stage in the process of scientific investigation is the exercise of the imagination. Pearson, however, insists that it is the disciplined imagination (comparable, no doubt, with Tyndall's composite and creative power in which reason and imagination are united) that has been at the bottom of all great scientific discoveries. He also admits that the classification of facts is often largely guided by the imagination as well as by the reason. At the same time he maintains that all great scientists have, in a certain sense, been great artists, and by describing a work of art as concentrating into a simple formula a wide range of human emotions and feelings, he attempts to bring the products of artistic creation into line with scientific laws. It is evident that Pearson, while emphasizing the importance in research of the creative imagination, has not contributed substantially to its analysis and differentiation.

Reserving for later consideration the complex mental processes so boldly broached by Tyndall and Pearson, let us glance, in the spirit of Galton and James, at some of the evidence concerning the employment of imagery by the scientific discoverer.

Some of the chemical symbols used by John Dalton.

Dalton seems to have relied on visual imagery; as has been remarked by others, his mind was of a corpuscular turn. In the early stages of his meteorological work he thought of aqueous vapor as made up of minute droplets diffused among the gases of the atmosphere. To the particles of these gases he ascribed definite form, and represented by diagram his idea of the constitution of the air. About 1803 Dalton began to picture atoms as of different sizes. He formed visual images of molecules of nitric oxide and nitrous oxide, of carbon monoxide and carbon dioxide, of ethylene and ethane. In his laboratory note-book during the autumn of 1803 he made entry of his symbols for hydrogen, oxygen, nitrogen, carbon, sulphur, and several of their compounds. Dalton could not accept with equanimity the less graphic method of representing chemical elements and compounds. As late as 1837 he wrote: "Berzelius's symbols are horrifying: a young student in chemistry might as soon learn Hebrew as make himself acquainted with them. They appear like a chaos of atoms .... and to equally perplex the adepts of science, to discourage the learner, as well as to cloud the beauty and simplicity of the Atomic Theory." Would the development of modern chemistry have proceeded more rapidly if the symbols of Dalton, which appeal to the imaginative thinker, had triumphed over the symbols of Berzelius, which appeal to the conceptual thinker ? [Berzelius originated the set of one- or two-letter abbreviations, such as H for hydrogen, used in modern chemistry.]

Come unto These Yellow Sands by Richard Dadd, 1842.

Kekulé, Berichte der deutschen chemischen Gesellschaft, 1890, pages 1305-1307, has left an intimate record, worth reproducing in extenso, of his own experience as a scientific discoverer:

Genius has been spoken of, and the Benzene Theory has been designated a work of genius. I have often asked myself what, exactly, is genius, in what does it consist? It is said that genius recognizes the truth without knowing the proof of it. I do not doubt that from the most remote times this idea has been entertained. "Would Pythagoras have sacrificed a hecatomb if he had not known his famous proposition till he found proof?"

It is also said that genius thinks by leaps and bounds. Gentlemen, the waking mind does not so think. That is not granted to it. Perhaps it would be of interest to you if I should place before you some highly indiscreet statements as to how I arrived at certain ideas of mine.

During my stay in London, I lived for a long time in Clapham Road in the vicinity of the Common. My evenings, however, I spent with my friend Hugo Müller at Islington at the opposite end of the metropolis. We used to talk of all sorts of things, mostly, however, of our beloved chemistry.

One beautiful summer evening I was riding on the last omnibus through the deserted streets usually so filled with life. I rode as usual on the outside of the omnibus. I fell into a revery. Atoms flitted before my eyes. I had always seen them in movement, these little beings, but I had never before succeeded in perceiving their manner of moving. That evening, however, I saw that frequently two smaller atoms were coupled together, that larger ones seized the two smaller ones, that still larger ones held fast three and even four of the smaller ones and that all whirled around in a bewildering dance. I saw how the larger atoms formed a row and one dragged along still smaller ones at the ends of the chain. I saw what Kopp, my revered teacher and friend, describes so charmingly in his Aus der Molecular-welt; but I saw it long before him. The cry of the guard, "Clapham Road," waked me from my revery; but I spent a part of the night writing down sketches of these dream pictures. Thus arose the structural theory.

Twelfth Century carving in the Church of Saint Mary and Saint David in Kilpeck, Herefordshire. Photo by Simon Garbutt, 1989.

It was very much the same with the Benzene Theory. During my stay in Ghent, Belgium, I occupied pleasant bachelor quarters in the main street. My study, however, was in a narrow alleyway and had during the day time no light. For a chemist who spends the hours of daylight in the laboratory this was no disadvantage. I was sitting there engaged in writing my text-book; but it wasn't going very well; my mind was on other things. I turned my chair toward the fireplace and sank into a doze.

Again the atoms were flitting before my eyes. Smaller groups now kept modestly in the background. My mind's eye, sharpened by repeated visions of a similar sort, now distinguished larger structures of varying forms. Long rows frequently close together, all, in movement, winding and turning like serpents. And see! What was that? One of the serpents seized its own tail and the form whirled mockingly before my eyes. I came awake like a flash of lightning. This time also I spent the remainder of the night working out the consequences of the hypothesis.

If we learn to dream, gentlemen, then we shall perhaps find truth ---

To him who forgoes thought,
Truth seems to come unsought;
He gets it without labor. ---

We must take care, however, not to publish our dreams before submitting them to proof by the waking mind. "Countless germs of mental life fill the realm of space but only in a few rare minds do they find soil for their development; in them the idea, of which no one knows whence it came, lives as an active process." As I have told you before, at certain times certain ideas are in the air. We hear now from Liebig that the germs of ideas are like the spores of bacilli which fill the atmosphere. Why did the germs of the Structural and Benzene ideas, which have been in the air for a period of twenty-five years, find a soil particularly favorable to their development in my head?

Kekulé thought that the answer to his own question lay partly in the effect of his early study of architecture, which had imparted to his mind an irresistible need of sensory presentation. He could not rest satisfied with an explanation of chemical phenomena unless he could support it by means of definite visual imagery.

Dickens' Dream by Robert William Buss, 1875. [Dickens Museum, London]

Kekulé 's account of the functioning of his imagination seems to stand as a unique confession in the records of scientific discovery. The history of literary composition affords us, however, numerous parallels. Professor Dilthey of Berlin has gathered some of these together under the suggestive title Poetic Imagination and Insanity. Some literary men, like Scribe, are gifted with vivid visual imagery, others, like Legouve, are dependent for their success on auditory images. Scott, Victor Hugo, and Browning seem to belong to the motor type. There is evidence in the case of Flaubert, as well as in that of Zola, that literary imagination may derive its data from the chemical senses. An analysis of the writings of poets like Marston and Helen Keller, defective in sight, in hearing, or in both, is of particular value in the study of literary imagination. Artistic creation in general employs imagery in order to preserve or enhance sensory experiences and to convey to others the moods of the artist.

Is there any class of human being in whom the imagination is more held in control, more disciplined, more subordinated to the reason, than it is in the adult scientist? All the psychic processes, including instinct and inspiration (which has been described as a sort of unconscious imagination), are means of establishing useful relationships with men and things, and it is by no means surprising that the scientific discoverer, who grapples with difficult problems of adjustment, should bring the finest powers of the mind into play. The history of science assures us that the creative imagination is not the monopoly of the painter, sculptor, poet, philosopher, or theologian.

Special investigations of the mental characteristics of Kepler, Newton, Davy, Faraday, Claude Bernard, Ehrlich, Weismann and others must be undertaken before an adequate psychology of scientific discovery can be formulated. The nature of the data of each science, as well as the mental make-up of the individual discoverers must be made the subject of rigid investigation.

Kekulé's pupil Van 't Hoff, who at the age of twenty-two wrote the essentials of La Chimie dans l'Espace, seems to have shared the visual imagination of his master. For Kolbe the idea that the arrangement of atoms in molecules could be determined appeared almost as fantastic as a belief in witchcraft or spiritualism. Berthelot was not less disdainful concerning Wurtz, the teacher of Van 't Hoff and Le Bel. When some friend told Berthelot not to take the atomic theory too seriously, atoms having no objective reality, Berthelot growled:

"Wurtz has seen them!"

The imagination, predominant in one type of scientific discoverer and restrained or suppressed in other types, is at best only one phase of creative thought.

"Wurtz has seen them!". IBM's A Boy and his Atom, 2013.