Lesson Eleven – Special Effects of Colours

Zeuxis and Arte are again at Arte’s home. It is raining, and they are sitting in the kitchen. Arte is sipping from a Coca-Cola while she browses through the newspaper. Zeuxis is sitting in front of her. He plays impatiently with his fingers on the kitchen table.

Zeuxis: Arte, since it is raining, we might maybe go into some heavy stuff of theory.

Arte, looking up and putting her newspaper aside: OK, Zeuxis. I am in a studying mood anyhow, and very alert. I am all ears.

Zeuxis: Well then let’s start. I have to talk about superimposed colours and about juxtaposed colours.
In painting on a canvas, a large area consisting of one particular colour absorbs all wavelengths of light but the wavelength of the colour of that area. It reflects the wavelength of that colour only. Painters have to take care with the processes of colours that combine in the viewers’ eyes.
Painters use emulsions of pigments in oils or, as in the earliest times, in egg yoke (tempera process). That meant they use small powders, small grains surrounded by oil. Painters bring this film on the canvas. When they do this in light brushstrokes and when using much oil and few pigment, special effects happen. If they use for instance blue, then all wavelengths but blue are absorbed by the pigments and blue colour is reflected. But since the emulsion is brought lightly on the canvas, some light can get through to the underlying layer. If this layer is also of an emulsion with for instance green pigments, some of the green may be reflected back and passes through the blue emulsion on top. The result will be a combination of green and blue to our eyes, and that is the colour cyan. The painter may be astonished to have blue on his brush but cyan on the canvas. If he brings the blue over green and over a juxtaposed white area he would see the same brushstroke be blue over the white and cyan over the green.

Arte: I would avoid doing that if I were a painter.

Zeuxis: Painters can either avoid such superimposed colours, or exploit these effects, which is a very difficult feat indeed. In order to avoid the effect they can use very dense emulsions, as most modern paints are, or put thick brushstrokes. All paints used by painters are to some degree relatively transparent or opaque.
Layers of superimposed transparent paints mix colours in the eye of he viewer. Moreover, the viewer receives then also a sense of depth. This technique is called "glazing". The best and more sensible genius painters exploit the effects to obtain a very broad palette of colours.
This was for instance the case of some of the Flemish Primitive painters of the fifteenth century, who experimented with the new medium of oil painting, and discovered its various effects, including the effects of various granularities of the emulsions. But only the painters who had a very sensible eye for such effects and wanted to experiment, only these used this very difficult process. Generally it was too difficult to imagine the thousands of various possible combinations of two or more superposed layers.

Zeuxis: One effect that was frequently applied however was to paint on ground layers of uniform white or black paint.
The English Victorian Pre-Raphaelite painters like William Holman Hunt coated their canvas first with white very bright paint, and then they worked on that substrate in pure colours without repents (over-painting) to obtain very vibrant colours. The white underlying layer reflected light instead of absorbing it.
Some painters preferred dark undertones. Lodovico Carracci used dark grounds and so did Tiziano. Boschini published a treatise in Venice in 1674 with the title "Riche Minere della Pittura Veneziano", in which he noted that Tiziano used to say that a painter should be acquainted with three colours only; white, black and red ^G76. And indeed, especially in his later years Tiziano’s paintings received the lower, darker and subdued tones of great sadness for the tragedies he painted like his "Pietà", the "Flagellation of Christ" or the "Death of Actaeon". It may be that these paintings are merely unfinished ones, as Titian painted at one picture over long periods. His Pietà in the Galleria dell’Accademia of Venice was indeed finished, only slightly, by Palma Giovane. But as we know them now they are powerful and sufficient. Why Titian used black backgrounds, is a story that we can only fully explain in the next chapters.

Arte: Johann Wolfgang von Goethe spoke in his "Theory of Colours" of the importance of the grounding of pigments and of the effects due to the fineness of layers of paint. He explicitly spoke of Jan van Eyck’s light brushstrokes and the associated effects. Leonardo da Vinci recognised already the effects of several layers of paint. He noted that white surfaces as a ground for colours have the effect of making the colour pigments appear more vibrant.

Zeuxis: Good, Arte. You have not forgotten Goethe at all. Now on to what happens when we put colours next to each other.

Arte: Zeuxis, you are strange now. Whatever would happen when we put colours next to each other?

Zeuxis: Some patience, Arte! Very strange effects occur simply by putting colours next to each other!

Zeuxis: Objects have areas of a certain colour. We only perceive these colour hues due to the constitution of our eyes and mind. Gradually, scientists and artists became aware of ever more complicated phenomena of vision. They learned that our perception of colours altered when one colour was set next to another.
Any awareness and understanding of such phenomena would only start to be accepted when it dawned in the common consciousness that not merely physical but also physiological processes participated in vision. The effects of adjacent colour areas then could be comprehended as effects that were not totally absurd, but in the realm of possibilities. In the following chapters we will try to explain what happens when different colours are seen in adjacent areas, and what these effects could mean for painters and for viewers of paintings.
Leonardo da Vinci had already remarked that adjacent colours influence each other. And again, Johan Wolfgang von Goethe was the first to draw attention by his experiments on the contrasts of colours. Take a sheet of red paper and place it on a yellow background. Take that same red sheet and place it upon a magenta background. The colour of the sheet will be perceived as having changed. Yet the object, the sheet of red paper has not physically changed. Our mind has changed the perception.
The honour of having fully noted, then accepted and passionately studied the effect of juxtaposed colours fell to a French chemist who had been appointed to a position of director of colouring at a tapestry manufactory.
Michel-Eugène Chevreul (1786 – 1889) had become interested in colours by necessity. He was a French chemist, a Professor of Chemistry working at the Museum of Natural History of Paris, but Louis XVIII appointed him as Director of Dyeing at the Gobelin Manufactories of Paris. He wrote several treaties on the chemical characteristics of animal fats, and these were also at the origin of some paints. A chemist before had held the directorship of dyeing at the Royal Manufactories, so the choice of Chevreul was not so strange.
Chevreul was called in at the Gobelin Manufactories in 1824 because customers complained about the colours of the tapestries. The Manufactories were a national French pride, so a famous French chemist was necessary to rectify the errors.
Chevreul noticed that some of the colours were indeed not stable. But in other cases, the products used in colouring were of good quality. He was puzzled, so he started to experiment with colours. He noticed that the perception of colours changed when two different colours were set next to each other. He was a scientist, so such an effect must have been hard to accept, but Chevreul had confidence in his sight, and yielded to the evidence. He used a sound scientific method of experimentation, and he was thorough and very meticulous.
Chevreul studied, and then published a paper in 1839 on colour theory that became quite influential. He called his treatise "On the Law of the Simultaneous Contrast of Colours". In this he explained the effects of juxtaposing various colour areas.
In a first principle, Michel Eugène Chevreul wrote, “When the eye perceives at the same time two colours that are similar in hue, these hues appear as dissimilar as possible, both from a view of their optical composition, as from their tonal value”.
Chevreul had found that there were two ways by which simultaneous contrast can operate: in intensity and in optical composition. We know now that three parameters can modify, and these are the three constituting features of a colour: its intensity, its hue and tone. Thus, a red colour will be perceived as being brighter against a dark background and less bright against a bright background. A red will appear more powerfully red on a yellow background and more orange-like on a reddish background. The red will look more coloured in hue on a grey background than on a background of bright pure colours.
In this way, scientists and artists discovered that one does not just have to see assemblies of small dots, the impression of which is added in the eye, to see changing colour tones in an additive physical process that works by distance alone. When larger zones of colour are juxtaposed, changes in colour impressions also are perceived. A large area of blue completely surrounded by a large white area will give a more brilliant hue to the blue. When we see that same middle area of blue surrounded by a large dark area of brown, the blue will have an entirely other hue. In the latter case the blue may take on some of the red and darker tones of brown, and the whole will give a more subdued, sad, solemn impression.

Arte: That is indeed strange, Zeuxis. Now my mouth drops open! Colours change just by setting one next to the other. Your magic is nothing compared to that!

Zeuxis: Let me repeat, Arte, and explain somewhat differently, and emphasise the three effects at play when juxtaposing large colour areas.
First, an achromatic colour (hues reduced in intensity of primary hue by adding white or black) will look more intense (have more of the primary hue) when surrounded by its complementary colour. By complementary colours we mean here the colour on opposite sides of the colour wheel.
Secondly, complementary pairs that are juxtaposed intensify each other. For instance, when red and green are painted next to each other, the red will become more fiercely red and the green more fiercely green.
Goethe remarked in his "Theory of Colours" that “to satisfy itself the eye seeks a colourless space next to every hue in order to produce the complementary hue upon it”. He said that in this resided the fundamental law of all harmony of colours. In other words, next to a colour should be expected its complementary colour, for that is the natural physiological effect.
Chevreul wrote, “In the harmony of contrasts, the complementary combination is superior to any other”.
Thirdly, when we view for some time an intense hue, then a complementary colour halo appears to surround the hue. This complementary glow will combine with the surrounding colours, especially when they are weak. Thus an intense red colour surrounded by a grey will have the effect that the grey becomes greenish since the green halo that is complementary to the red will interfere with the grey.

Arte: I would like to know more about this guy, Chevreul. This is extraordinary. How could he see so clearly what centuries before nobody had remarked?

Zeuxis: All right, Arte.

Michel-Eugène Chevreul’s theory of juxtaposed colours

Zeuxis: Before I explain Chevreul’s findings in further depth, we have to look at some of his definitions.
Chevreul argued that red, yellow and blue were the primary colours. Thus he used as a basis the traditional primaries of the subtractive process, but interpreted like painters did. He recognised also the orange, green and violet secondary or compound colours.
chevreul used the word "hues" – or nuances – of a colour the modifications that this colour receives by the addition of a small quantity of another colour.
Chevreul called "tone" of a colour the modification, which that colour takes when black or white is added. We called this "saturation" in our theory.
He used the word "scale" for the collection of all tones of the same colour. So, scale indicates the various gradations that one hue undergoes when adding white or black.
When white was added to a colour he used the word "weakening" of tone. Adding black "deepened" a tone. He also used the words pure, broken, reduced, grey and dull. Pure colours were the primary colours red, yellow and blue, as well as those colours, which result from their binary combinations in equal quantities, such as the pure secondary colours orange, green and violet. The broken colours were these pure colours mixed with black.

Arte: When painters mix paint, colours are subtracted. So some blue added to red and yellow produces some black and thus "reduces" or "breaks" an orange colour.

Zeuxis: Yes. Chevreul called "luminous" the colours yellow, orange, red, and light green. Sombre colours were blue and violet.
Chevreul explained in his book many phenomena of optical effects of adjacent colour areas. He formulated a law stating that in looking at two contiguous colours a viewer perceives what he called "simultaneous contrast of colour and simultaneous contrast of tone". He wrote that when the eye saw two adjacent colours, these would appear as dissimilar as possible in optical composition and in the height of the tone.
Chevreul did not discover why his law was valid, but he thought he also had discovered the process by which his law worked. He wrote that for two adjacent colours to be as dissimilar as possible, the complementary of one colour had to be added to the other. He knew that this was a physiological reaction of our eyes and brain, because the physical nature of the contiguous surfaces did not alter by placing them next to each other.

Zeuxis suddenly throws pieces of paper on the table and startles Arte at the brusqueness of the movement, but she is all ears again.

Zeuxis: Here is a green piece of paper and a yellow piece of paper.
By Chevreul’s process if we put for example this green and yellow area contiguously, red, the complementary of green, added to the yellow, will make it incline to orange. The complementary of yellow is indigo. This colour added to the green will make it bluer. As a result we will see orange next to a green blue. The greater the difference between a colour and the complementary of the other added to it, the more striking would be the resulting modification in hue.
Colours can also be strengthened in intensity by this effect. Take red and green surfaces juxtaposed. The complementary of green, that is red, will be added to the red and the complementary of red, green, will be added to the green. In this example both colours are strengthened. The effect worked best when strips of coloured paper of about ten centimetres wide were placed next to each other.

Arte: Indeed! Wonderful! But how do painters then know what colour they put on a canvas, when colours change immediately?

Zeuxis: The effects described by Chevreul are of course extremely important for painters!
On the one side, painters are annoyed by the effect, because when they juxtapose colours on the canvas – which is all what painting is about – they will perceive other colours than the hues of their preparations of paint on the palette. Preparing paint on the palette while the palette is already full of colours from earlier patches of dry paint, may give an impression that is different from the hue of the pigment in other circumstances. Therefore, painters have to think carefully before laying down paint on the canvas, taking into account the law of Chevreul and the surprising resulting effects of colour that will be obtained. Often they have to correct their colouring to obtain their desired effect.
On the other hand, if painters work with complementary colours next to each other the colours will not only remain as prepared but will also be strengthened. This last effect occurs of course only for exactly complementary colours. It is not easy to correct colours in a painting. It is possible with oil painting, but may deepen the colours when successive layers are brought on the canvas. Corrections are impossible in fresco painting; with this technique the colours have to be right immediately.

Zeuxis: The pioneering Impressionist painter Camille Pisarro (1831-1903) wrote in 1887 in a letter to his son Lucien, "It is clear that we could not pursue our studies of light with much assurance if we did not have as a guide the discoveries of Chevreul and other scientists. I would not have distinguished between local colour and light if science had not given us the hint; the same holds true for complementary colours, contrasting colours, etc." ^G87.

Arte: Is this law really always valid, in all circumstances, even with white and black?

Zeuxis: Chevreul investigated also the effects of white and black in combination with other colours. He found his law was always valid.
White next to any colour will make that colour brighter and deeper, whereas the white will take on the complementary colour of the other.
Black or grey next to any colour will make that colour more brilliant, and usually lighter. The black takes on a delicate hue, as the complementary of the other colour is added to it. But it will not only remain dark, but also even deepen by the law of contrast of tone.
Especially with these last observations, Chevreul explained a concept of painting that artists had discovered since many centuries in an intuitive way, and had applied as a painter’s technique. Painters have used black next to colours often, because, as Chevreul remarked, this improved the hue and brilliance of the colours. Black between colours also much annihilated the combinatory effects of the simultaneous contrast between the colours so that these colours kept their own hue. This method is sometimes called "outlining". Such effects are not reached by separating the colours with white. The white subdues the colours by its brightness, which is enforced by the law of simultaneous contract of tone, and this breaks the hues. Thus, black or dark tones in the background are the hallmarks of the greatest painters like Rembrandt, Titian and van Dyck. And so we understand also why ladies prefer their husbands to wear a dark suit to accompany them when they themselves wear brightly coloured robes!
Painters became gradually aware of these effects. It is remarkable to see for instance how the evolution worked from Robert Campin to Rogier van der Weyden. Robert Campin, a Walloon painter of the fifteenth century, made a portrait of Robert de Mammines in which the face of the model was painted against a light background. But Rogier van der Weyden, his student, painted his portraits against a dark, even black background, to enliven the colours of the face.

Zeuxis projects a painting made by Titian.

Zeuxis: Let me show you an example now, Arte. We have seen this picture before.

-> Tiziano Vecellio (1488/1490-1576). Pietà. Galleria dell’ Academia. Venice. 1576.

Zeuxis: One of the most magnificent examples of this knowledge of colour and tone is Titian’s Pietà.
In Titian’s last years his palette grew more sombre, and he used spare colours. But in doing, this his pictures gained in force, and his few colours like the red cloak of Nicodemus, obtained a quality of their own which are admirable in their isolation. It has been argued that this effect was merely a physiological effect of ageing, that the discrimination of colours is less acute with age, so that the palette of the very old painter becomes more limited as with Titian, or more strident as with Claude Monet. We find that hard to accept for a great painter like Titian, especially as there are other painters, like Michelangelo, who reached a very old age and yet whose palette did not darken or become more extreme. It has been argued also that these pictures are unfinished ones, as Titian worked over long periods. Titian would have painted only the rough composition and just placed a colour here and there. But these patches of colour are so diligently placed in the picture, and the other tones already so balanced, full and complete, that also this conjecture is difficult to hold. In Nicodemus, Titian probably painted a self-portrait.

Arte: I did not know that painters had such a hard time with colours.

Zeuxis: You see, learning to paint is not that easy! The effects of Chevreul’s law of simultaneous contrast make it also difficult for painters to predict what the coloured result would be when they used very many colours juxtaposed. Most painters have therefore restricted their palette to a few colours only for a particular painting. Many of the greatest painters stated in their later years that in order to make "beautiful" paintings, only a few colours were necessary.

Arte: Did Michel-Eugène Chevreul discovered other things on colours?

Zeuxis: Besides the law of simultaneous contrast, Chevreul described other colour effects.
He called successive contrast the effect when the eye, having looked for some time at one coloured surface, turns away and then sees that same surface still for some time in its complementary colour.
Chevreul called mixed contrasts when one looks with this complementary colour in the eye and mind to another coloured surface. The two colours will combine. Thus, when we look first at a red surface of a painting for a certain time, turn away and have the successive contrast of green virtually in our eyes (red’s complementary), this green will combine with a yellow surface at which we might look in another part of the painting to a mixture of yellow and green. And this is not such a nice effect on the yellow.
Painters cannot do much to control these last effects, except making their coloured surfaces not too large and separate them by neutral hues such as white or preferably black. Therefore, viewers will find on many canvases really pure colours separated by darker zones, often the chiaroscuro dark shadows between the pure hues. The darker shadows diminish and isolate the colours, so that Chevreul’s effects are less remarked.
The juxtaposition of colours does not just change the colour hue itself. When the contiguous colours are not of the same intensity, the colour that is deep in tone appears deeper and that which is light will appear lighter. This happens always in order for the colours to become as dissimilar as possible. This explains also all the more why the most sensitive painters preferred darker backgrounds for their pictures. With for instance a clear, very bright blue background, colour surfaces of red, yellow and green will look duller, and the bright blue even lighter. On a dark background of black, grey or dark brown, the background will look darker and the colour surfaces lighter.

Zeuxis brings his magic, transparent screen out again and he shows a painting by Rembrandt.

-> Harmensz Rembrandt van Rijn (1606-1669). The Man with the Golden Helmet. Staatliche Museen Preussischer Kulturbesitz. Gemäldegalerie Berlin. Berlin. Around 1650-1655.

Zeuxis: Rembrandt, as no other master, had these marvellous effects of hues contrasting with black in his mind. This was how Rembrandt saw colours, and the only way he could represent them. He was apparently so sensitive to colours that he was obliged to surround his beautiful hues by black so that they were resplendent.
A good example is the painting "The Man with the Golden Helmet". When one arrives before this picture, the impression of the gold in the helmet and the bright face of the soldier strikes like lightning. In any hall where paintings are hung and accompany Rembrandt’s picture, no such marvellous hues are seen.

Zeuxis makes the man with the golden helmet disappear.

Zeuxis: Michel-Eugène Chevreul’s theory is merely a process of combining colours to obtain the observed effects. Chevreul formulated a law, but he never tried to explain the phenomenon, and science in his times was not advanced enough to offer an exact explanation. Moreover, Chevreul was not a physiologist. He was therefore primarily interested in the reality of the effect and since he worked in a factory of tapestries, in the practical implications of the combinations for the colouring department of which he was the director.
Chevreul designed a colour circle with the three basic colours red, yellow and blue plus their complementary colours opposite them and the compound colours in between. He explained how colourists at the tapestry factory could use this circle but also of course painters.
Take any two colours A and B and place them contiguously. Look where the colours are on the circumference of the colour circle. Take the complementary of A on the opposite side of the centre and let that be AA. Then look where AA lies between two primary colours. Thus one finds the primary colours of which AA is a compound. While looking at colour B on the circle, one knows also where its complementary BB can be found, that is on the other side of the centre point. One finds then also the primary colours that combine to make BB. Consequently one has to mix A with the two primary colours that make up BB and the result will be the hue of the new A. One does the same with B, adding the two primary colours that make out AA. Furthermore, one knows from Chevreul’s law that the deeper colour of A and B is deepened and the brighter colour brightened.
This process is not so easy for many colours, simple almost only for combinations of primary colours. Painters will have it easier to experiment and put the colours next to each other to see for themselves what the effect on the canvas could be. Thus painting a small example, experimenting by trial and error is more rapid and surer than reasoning with the colour circle.

Arte: If I understand well, Zeuxis, Chevreul did not really know why colours changed when he set them next to each other, just as we here put these pieces of paper together and notice the changes. But with your wisdom, you must at least have an idea? And since the nineteenth century many scientists must have studied the effect.

Zeuxis: Not really so many, Arte. But there is a theory indeed, that I believe is right and that could give a plausible explanation. We go to photography now!

The Retinex Theory of Colour Vision of Edwin H. Land

Zeuxis: Michel-Eugène Chevreul’s theory was formulated as a law, but Chevreul gave no explanation for the effects. An answer was only found, and that partly by accident, in the 1960s. A new theory was then formulated, the results of which are still being examined currently. This theory may not be so relevant to the art of painting, since Chevreul’s rules offer a construction of perceived results of juxtaposing colour surfaces and a good understanding of that is all what painters actually need. Yet, this modern theory does explain conclusively the reasons for the effects that Chevreul studied, and thus this newer theory is more than relevant for a book on painting like this.

Arte: Who found it then?

Zeuxis: A guy called Edwin Land.
Edwin H. Land (1909-1991) was the inventor of the Polaroid –Land photographic camera process of dry-fixing colours on photographic film. I saw you use such a camera too, Arte.
Land made many experiments with photographs taken of coloured pictures of rectangular colour patterns, which he called "Mondrians". During his experiments he was particularly intrigued by an arrangement of superimposed pictures made with filtered light.
Land had made two black-and-white photographs of a Mondrian. Then he projected these on a screen, but one projected through a red filter and the other through a green filter. The two superimposed images showed astonishingly all the colours of the original Mondrian! Land tried to understand how the information of the colour blue for instance could have remained in black-and-white photographs. This was completely in contradiction with all previous traditional theories of vision.
The traditional colour theory would have predicted merely a superimposition of red and green images with zones in various shades of black. The result would have been images in shades of black and yellow. But Edwin Land saw a true colour image with blue colours and red and green.
Land proposed that it is not the wavelength of the reflected light itself and only that, that is processed in our brain. He mused that our colour perception is somehow decoded using the ratio of the longer and shorter wavelengths reflected by the object. Land proved that there was not a one-to-one correspondence between colour perception and wavelength, and thus he proved Newton wrong and Goethe right.
Edwin Land devised a theory from his experiments that he called the "Retinex" theory, a name derived from retina and cortex. This theory proposed that colour perception depended only on the neural structure of the human visual system. According to the Retinex theory, the perception of colour depends more in particular upon the relative reflectance of a coloured area. This relative reflectance is determined by the mean reflectance of random paths leading to that area. So, Land proposed that a parameter "lightness" be associated with every colour surface. This lightness can be a measure of the brightness of a surface compared to the surrounding surfaces. It has been possible to describe this effect in mathematical formulas and thus to model our subjective colour perception.

Arte: So, Land’s theory proposed an explanation for the phenomena of colour illusions as described by Michel-Eugène Chevreul and by Johann Wolfgang von Goethe?

Zeuxis: I believe so, Arte. Illusions indeed arise from the contrast between the object and its surroundings, as observed by humans. The contrast is the relative reflectance of an area, relative as compared to the surrounding areas. Land and his collaborator McCann proved that Mondrians produced a perception of colour that does not depend upon the spectral flux, the colour spectrum of old, reflected by the Mondrian patterns. Human perception is independent of the spectral distribution of the reflectivity of the light energy perceived by the eyes. It depends on the relative reflectance of adjacent areas. We do not determine the colour of a surface in isolation. We derive the colour from a comparison of the wavelengths reflected from the surface and from all the areas around it. When we change the setting of juxtaposed colours, our colour perception changes.
The Retinex model also explains the effect of colour constancy. Colour constancy is the fact that our perception of colour is usually, within certain boundaries, independent of the characteristics of the light source. This is also sometimes called chromatic adaptation, the ability of the human visual system to adjust to varying lighting conditions. Our visual system adapts to the intensity of light to preserve the colour of objects. This colour constancy is the tendency for an object to be perceived in the same colour under different lighting conditions. Experiments of Edwin Land proved that familiarity, whereby our mind would substitute colours because we have memories of an object and its colours, were not at play in colour constancy effects.

Arte: Could you explain some more about this colour constancy issue, Zeuxis? It seems we are on to something important here.

Zeuxis gets his magic screen out and he shows Arte a bird’s view of mountains covered with snow. A marvellous sun shines over the mountains.

Zeuxis: To illustrate the colour constancy phenomenon, we take the effects of sunlight on snow. A late sun makes snow yellow on a photographic film. Yet we automatically account for the orange-tinted light and see the snow as white!
This was probably an evolution to perceive the colours of nature the same way in varying lighting conditions, through morning, day and evening. Otherwise, the environment might have been too confusing for humans. Therefore, a tree looks green in the various periods of the day. Yet, our vision system corrects not because we know from memory that snow is white. Land’s Retinex theory clearly explained that the relative brightness of a surface, on which depends according to Land’s theory our colour vision, is unaffected by the conditions of lighting. This explains colour constancy. Memory and familiarity do not need to intervene. We decide on the colour of an area by comparing its ability to reflect short, medium and long wavelengths against that of adjacent surfaces. That relative reflectance does not change much when lighting is modified over all the surfaces together. When lighting conditions change, the relative comparisons between adjacent colour areas remain the same, and so the colour of the area remains the same. But colour constancy disappears when the comparison with adjacent surfaces cannot be made.

Arte: That is it then Zeuxis, the final theory!

Zeuxis: Land’s theory decisively proved that colours are not really attributes of nature, characteristics of a surface, but merely a feature of human vision. Colours are the result of an interaction between the world and humans, the result of neurological processes in the retina and the subsequent handling of the information of relative reflectance in the brain. Colours exist only because there are living beings, and they exist only in the eye and mind of living beings. The human vision complex changes perception, corrects for effects of lighting to some extent. Land’s theory proved Goethe right, as opposed to Newton’s concepts that colour was a physical quality. There is a lot more to colour than a purely physical phenomenon!

Zeuxis stays in thoughts for a while, then he continues: The great Victorian critic John Ruskin (1819-1900) must have felt some of what Edwin Land proved when he wrote what Ernst H. Gombrich quoted in "Art and Illusion". "While form is absolute, so that you can say at the moment you draw any line that it is either right or wrong, colour is wholly relative. Every hue throughout your work is altered by every touch that you add in other places; so that what was warm a minute ago, become cold when you put a hotter colour in another place, and what was in harmony when you left it, becomes discordant as you set other colours beside it; so that every touch must be laid, not with a views to its effect at the time, but with a view to its effect in futurity, the result upon it of all that is afterwards to be done being previously considered. You may easily understand that, this being so, nothing but the devotion of life, and great genius besides, can make a colourist ^G106."

The effect of simultaneous contrast

Arte: Tell me more about the effect of simultaneous contrast, Zeuxis. What would Land’s theory mean for Chevreul’s findings?

Zeuxis: I have already explained somewhat what Chevreul’s effect meant to painters. Land’s Retinex theory complicates matters more than a simple application of Chevreul’s law.
Chevreul made experiments with two colours, three at most. He described these combinations at length. He explained less combinations of more colours. He worked with rectangular pieces of paper or of cloth, with pieces of certain dimensions and he described the effects when he juxtaposed these after having seen them in isolation. He also worked in sunlight, which is of high intensity of illumination, but we see paintings in museums often with harsh white light thrown focused on the pictures.
These experiments, however sophisticated and admirable, are far from situations of reality. In reality zones of many colours and of various sizes and forms are seen adjacently and in combination. Chevreul experimented with bands of paper of particular dimensions and he wrote that with these the effect was most visible. He did not state that he had experimented with other dimensions of coloured areas. We know that the effect diminishes from the border between the two colours and the effect diminishes with higher intensities of incident light. Michel-Eugène Chevreul did not experiment with bands of paper under various lighting conditions, under differing intensities of incident light.
The Retinex theory encompasses effects of all the areas surrounding a particular colour area. It states that the colour of that area is determined by our mind by comparing the reflections of the short, medium and large wavelengths as captured by the S, M and L cones, relative to those of all adjacent zones. We now know that there are neurones in our mind that react to the differences in the wavelengths of light reflected from adjacent areas. The result of such a process is difficult if not impossible to predict by painters, as by any human, and they are more complex than Chevreul could have imagined. And it seems that the Retinex theory does not explain al effects. Further theories and mathematical models are being built.

Arte: But, Zeuxis, let’s get back to earth now! When I look at a painting, I see what I see. I couldn’t care less for the difficulties a painter had while putting his colours on the canvas. Well … sort of. Poor painters! Should a viewer be concerned at all with these theories?

Zeuxis: The answer on this very pertinent question, Arte, is yes and no. Chevreul proved that adjacent colours altered simply by being viewed together. But he worked on particular settings. Reality is more complicated and less predictable. For painters, the effects of simultaneous contrast are very real and quite a challenge. They can never really be sure what colour they will obtain in a very coloured painting when they add a paint. The colour of their paint on a white palette will look quite differently between other colours on the canvas. O crouse, they can step back, look, and correct. At least, painters can do that; not acquarellists and not freso painters! Luckily, and maybe this is a paradox, for viewers the situation is in fact very simple.

Arte: Like I said. Viewers see what they see. What they see is what they get.

Zeuxis: They see the result of what painters have realised in colour. Painters have corrected for adverse effects. Painters have at their disposal many techniques for their corrections. They can work directly on the hues. They can separate the hues by grey or darker zones. Hues seem to remain individual and unmodified by nearby colours when separated by black, white or grey thick lines or zones. This is the technique of outlining. They can use complementary colours that enhance instead of modify each other. Broken hues, secondary, tertiary and even more quaternary hues will create less perception of afterimages, and thus of modification of hues when juxtaposed, because they contain in them many hues already.

Zeuxis: Great painters use all these techniques so that a viewer sees the same colours as the painters did in their finished work. A viewer might see adverse effects, of dulled and modified colours, but how can a viewer judge whether this was not a desired effect? We know moreover that the processes at play in a real picture that includes many adjacent coloured areas are more complex, with various mutual influences. Sometimes, some observers but not all seem to perceive vibration effects at the boundary between two colours. When many colours are present in a picture two juxtaposed colours change only very locally, at their boundaries alone, but these colour changes will then interfere with colours somewhat further from the boundary and combine to yet other hues. So the effects are complex. More influences work than the simple explanations of Michel Eugène Chevreul. Various influences cancel out each other in the relative reflectances. The Chevreul effect remain often limited to small areas along the edges of coloured areas, where we find often colour transitions and corrections brought in by the painters, and often also naturally the chiaroscuro shadowing to separate colours. So, the issues noticed by Chevreul are more of a problem for the painters than for the viewers.
Chevreul’s work was important, and so is the report I gave you, Arte, because it enhanced much our respect for colouring. Chevreul taught generations to look at colours with deeper interest, whereas before, drawing had received more attention. Scientists and artists now studied effects on humans of combinations of colour. Chevreul’s writing on harmonies of colours, which we will consider further in our lessons, is interesting and this was based on his understanding of the law of simultaneous contrast.

Arte: These theories which explain real effects of colours are interesting and the effects surprising, Zeuxis. Have they been applied in painting? I mean, have painters, once the theories were known tried to use them, and made paintings that used these theories as their basis?

Zeuxis: Yes and no. The painters of Chevreul’s time knew his theories, but remember that the various processes of producing hues were not well understood then. Of course: we are still far from understanding everything, Arte, but knowledge was beginning only to grow by then. I can however talk to you about two styles in this domain: divisionism and camaieu.

Arte: Then please do so, Zeuxis.

Divisionism

Zeuxis: A French art critic Charles Blanc (1813-1882) used Chevreul’s work to propose the optical mixing of colours in paintings, in which dots of juxtaposed pure colours constitute at a distance in our eyes more luminous hues than those that can be obtained by mixing pigments on the painter’s palette.
Artists also eagerly read the works of Ogden Rood (1859-1926), a Professor of Physics at Columbia University. Rood published a popularising book in 1881 called "Modern Chromatics", which was read by the Impressionist painters whose whole art was based on colours. He confirmed the previous findings that the optical mixing of colours was superior to the material mixing of the paints on the palette. Rood also insisted on using small contrasting patches of colour to obtain effects of high luminosity of the hues. Under the influence of these theories, and of course in an evolution to further experiments in colour, Impressionism evolved to Divisionism and Pointillism.

Zeuxis: Painters can exploit the combinatorial effects of small, closely placed dots or of small juxtaposed strokes of paint on the canvas to form in the viewers’ mind a very broad scale of colour tones. This technique is called Divisionism, and it was based on the described scientific investigations into colour of the second half of the nineteenth century. True Divisionists stated that their art of colouring had nothing to do with the stipples and small brushstrokes of Impressionism. They wrote that Impressionists had also worked with dots, but Divisionism really wanted to divide colours and areas as well as other elements of painting. Paul Signac wrote that placing small dots did not guarantee luminosity, as small dots of complementary colours created more a shade of grey than of luminous white. At the same time they exploited the ‘optical mixing’ of colours by dividing the colours on canvas in very small adjacent areas, either dots or small strokes. By "optical mixing" they referred to the additive process of colour formation.
Seurat and Signac, the two best-known Divisionists, had read the treatises of Michel-Eugène Chevreul and of the art critic Charles Blanc. Seurat knew Chevreul, and he also introduced Signac to the now very old Chevreul. Chevreul died indeed at an age of over a hundred years old. He had become famous for his work on colours. He has a monument in the Jardin des Plantes in Paris, and even a street in Paris was named after him. The names of streets called after Goethe in Germany are of course uncountable.
The Divisionist way of painting is very difficult, the effects not easy to predict. Experimenting with the effects beforehand was necessary. Even the best Divisionist painters like Seurat and Signac used in a certain area at most only two or three combinations of colours in their small dots or small brushstrokes. They did not limit them to red, blue and green. They used dots of other colours, but at least Paul Signac proposed to paint only in dots of the pure rainbow’s spectrum colours.

Zeuxis places his magic screen on the white wall of Arte’s kitchen and shows a painting of Georges Seurat.

-> Georges Seurat (1859-1891). Bathers at Asnières. National Gallery. London 1883-1884.

Zeuxis: As an illustration of the theories of the effects of adjacent coloured surfaces I propose to look at Georges Seurat’s painting "Baignade à Asnières". In this picture, the flesh of the bathers is entirely painted in white and so is the shirt of a man lying in the grass. Seurat obtained for the white patches to contrast sharply with the green of the grass, the blue of the water and the sky and with the orange of the bathing suits. In the man lying in the grass, Seurat contrasted black and white with the orange colour of a dog. He was applying colour theories, which as we stated above, ascertain that the colours, here green and orange, are seen even more bright when contrasted with bright surfaces. Furthermore he used green and orange, and also orange and blue, each time two subdued but almost complementary colours, to obtain agreeable impressions and to impressions whereby the hues of these colours are enhanced. Green and blue are together the "fool’s colours", but Seurat used a more magenta hue for the water.

Arte: I like Seurat’s picture, Zeuxis, but it looks so static and solemn despite its wealth of colours. You said something about camaieu?

Camaieu

Zeuxis: Yes, Arte, finally, there exist painting techniques by which pictures are entirely made in the shades of one single colour. The best-known process in which shades of grey are used is called grisaille.
Grisaille was often used to simulate bas-relief sculptures in pictures. When only the colour yellow is used, one speaks of cirage. These painting techniques are all denoted by the word camaieu. The word comes from old ways of imitating cameo colours in miniature painting.
Grisaille paintings have been used to imitate sculpture. Grisaille creates so-called "trompe - l’oeil" pictures, which try at first glance to give an illusion of sculpture. Grisaille paintings use dramatic shadowing and very detailed drawing, that is emphasis on lines, to obtain their illusion. Grisaille paintings were particularly popular with fifteenth century Flemish painters of altarpieces.
These altarpieces originally consisted of wooden boxes that contained many polychromed sculptures. Paintings of pictures were made on the panels of the box that closed upon the sculptures, and pictures were also made on the predella beneath the box. Originally, the paintings were not the most desired representations of the Bible scenes, but the sculptures were. So the painters had the habit not only to paint scenes in colour on the panels, but also to imitate marble statues on the panels of the altarpieces, especially when statues were no longer placed in the boxes of the altarpieces.

Arte: I once went with my mother to Flanders. We went to the town of Gent there and saw Hubert and Jan van Eyck’s famous altarpiece of the "Holy Lamb" in the Saint Bavo Cathedral. This polyptych contains such grisaille paintings.

Zeuxis: Italian painters took up the grisaille technique as an independent means of imitating sculpture in the art of painting. One of the painters who excelled in grisaille was the Renaissance artist Andrea Mantegna (1435 – 1506). Mantegna always was in search for static, tactile monumentality and volume in his figures. It was almost natural for this great artist to grope to grisaille.

Arte, getting from out of her chair: We have to stop here, Zeuxis. My mother will be here soon and I have to prepare dinner … See you a next time.

Zeuxis: That is a pity, Arte, because I was just getting to the crux of all our colour teaching: the subject of harmony. But it will have to wait until a next time.

Back to the navigation screen (if that screen has been closed)

Last updated: July 2010

Book

Copyright: René Dewil - All rights reserved. The electronic form of this document is copyright. Permission is granted for electronic copying, distribution in print form for educational purposes and personal use. If you do reduplicate the document, indicate the source as 'René Dewil - The Art of Painting - Copyright'. No permission is granted for commercial use and if you would like to reproduce this work for commercial purposes in all or in part, in any form, as in selling it as a book or published compilation, then you must ask for my permission formally and separately.