<<Back to Introduction

'Colour School'
Thoughts on the Teaching of Colour Theory

by Roy Osborne

Previous to the 1930s, the models adopted in ‘progressive’ colour teaching tended to derive from Post-Impressionist models, and notably the work of Vincent van Gogh, Paul Cézanne, and the later work of Edgar Degas and Claude Monet.

In the 1920s there had been a return, generally speaking, to ‘objective’ observation (rather than ‘subjective’ expressionism), manifested in America by Georgia O’Keeffe and painters of the Precisionist School (including Charles Sheeler), and in Britain by teachers at the Euston Road School (including William Coldstream and Claude Rogers).

The 1930s saw the emigration of influential teachers and their students from the Bauhaus in Germany (often via England) to the United States. Here, new approaches to art teaching which had been developed at the Bauhaus started to emanate, at first from László Moholy-Nagy’s New Bauhaus in Chicago (1937-39), and has infiltrated art schools throughout the USA, Europe and elsewhere by the 1960s, since when there has been something of a decline in the teaching of the perceptual arts in favour of examining conceptual approaches to art.

Three particularly influential colour teachers developed their artistic and educational theories while teaching at the Bauhaus, an art school which had previously been the Weimar School of Arts and Crafts in Erfurt:

Paul Klee (1879-1940)

Johannes Itten (1888-1967)

Josef Albers (1888-1976)

A factor that impressed Klee about Paul Cézanne was his use of colour not merely to record the obvious visual appearance of objects (as the Impressionists had largely done) but to establish the artist’s right to modify naturalistic colouring in order to harmonise or balance colours within the pictorial composition itself.

Klee may also have been impressed that Cézanne appeared to show no embarrassment that his paintings revealed the manner by which they had been painted and/or had evolved over an extended period of time. For Klee this ‘building-block’ approach to composition may have strengthened his own desire to produce works that placed more emphasis on formative powers than with the finished forms.

A second major influence stemmed from Klee’s contact with members and associates of the Blue Rider group, based in Munich. Not least among these was Wassily Kandinsky, who later became his colleague at the Bauhaus, but who at that time published a short, personal manifesto, On the Spiritual in Art (1911), in which colour is cast in the role of the spiritual component of what Kandinsky believed would be a universal visual and abstract language for the communication of feeling.

Perhaps more significantly for Klee was the inclusion in the Blue Rider exhibition of 1911 of a selection of paintings by the Parisian painter, Robert Delaunay. Klee admitted particular fascination for recent paintings by Delaunay ‘which carry on a perfectly abstract existence without the benefit of motifs drawn from nature’. Delaunay, experimenting with sunlight dispersed through prisms, inspired Klee also to consider the question which confronted many artists of the time: when colour is released from its descriptive role, what might be a basis or structure for in its application in abstract art?

For Delaunay, one possible source was those principles formulated by Chevreul while working at the Gobelins tapestry works some 70 years previously:

‘At this moment, about 1912 to 1913, I had the idea for a kind of painting which would depend technically on colour and its contrast but which would develop in time, simultaneously perceived, at a single moment. I used Chevreul’s scientific words: simultaneous contrast.’

Summary of the colour theories of Paul Klee

Principal reference: Jurg Spiller, ed. (1956), Paul Klee: Das bildernische Denken. Basle: Schwabe. English translation (1961): Paul Klee: The Thinking Eye. London: Lund Humphries.

In 1919, Klee received an invitation to teach at the Bauhaus in 1919. While preparing for his first painting classes, he wrote home to his wife, Lily: ‘Here in the studio I am working on half a dozen paintings, drawing, and thinking about my course, all at once, for everything must go together or it would not work at all.’ (Spiller p.32)

The most coherent collection of his colour teaching notes, entitled ‘order in the realm of colours’, was prepared for classes presented to his Bauhaus students late in 1922. His first statements acknowledge respect for the theories of Johann von Goethe (Theory of Colours, 1810). Klee then charts an orderly analysis of colour in relation to practical aspects of painting (specifically in the watercolour medium) and rapidly established a passage from the theoretical to the practical, so that simple formulas inspire experimental and open-ended possibilities.

He begins his classes by discussing the hues of the rainbow (Spiller p.467). In order to obtain a useful teaching tool, he modifies and rationalises its contents by adopting Goethe’s six-section colour circle, with red opposite green, orange-red opposite blue, and yellow opposite violet.

Klee them positions his colour circle to form a disc located horizontally within the colour sphere, similar to an arrangement devised by Philipp Otto Runge in 1810. This three-dimensional arrangement permitted Klee to arrive at a useful organisation of colour in which white is located at the top of the figure and black at the base. Importantly, it allowed Klee to clarify the relationship between hue, colourfulness (chroma) and tone (value).

Having established a theoretical model, Klee moves to the visual experience of colour comparison, with reference to the observations of complementary colour afterimages that had formed the basis of Goethe’s principles of colour harmony:

‘Experiment shows that if we expose our eyes to red for a while and then suddenly look away, am n astonishing after effect is produced: we see not red but green’ (Spiller p.473).

Eager to transform this observation into practical experiment, Klee then turns to his watercolour box to see what happens when, as transparent paints, complementary red and green colorants are glazed repeatedly one over the other. The outcome is that, although no grey paint is used, where the number of red layers equals the number of green layers, a neutral grey is obtained. Hence ‘this median colourlessness decreases ... to the left in favour of increasing red, and to the right in favour of increasing green’ (Spiller p.475). Comparable observations are made with other complementary colour pairs.

Klee then directs his students to experiment for themselves, working methodically, and taking care to allow each new application of paint ‘to dry before the stove’.

For Klee, the simple activity of glazing similar or contrasting colours over one another appears to have given him the foundation he sought in order to link hue, colourfulness and tonality with the practical teaching of colour theory. Such principles were not just exploited in the art-room but significantly in a series of increasingly sophisticated watercolours he painted between 1921 and 1923.

Having rejected traditional methods of pictorial representation, Klee set about seeking alternative strategies for holding his paintings together visually. A benefit of utilising small selections of colours (often as few as two or three) was that the resulting impression appeared remarkably unified and harmonious.

For Klee, unlike the Constructivists proper (such as Moholy-Nagy, or even Albers) it was not enough to combine colour and contour for themselves alone. He rarely considered his works complete until they had been subjected to the unpredictable inner response of their creator: ‘Now a further dimension is added, the dimension occupied by questions of content’ (Spiller p.90).

In other words, Klee’s abstract compositions were not intended as ends in themselves. Once the artist had taken the care to construct a colour-form foundation, he next opened himself to extend the work into the realm of ‘free association’ or artistic fantasy. Klee admits therefore that ‘all the partial operations suggested by my exploration of the colour circle’ were always intended to be integrated ultimately ‘with the more subconscious dimensions of the picture’ (Spiller p.95).

A significant aim of the colour experimentation which Klee encouraged in his students, as well as in his own work, was therefore to produce a colour foundation the principal aim of which was to cross the ‘bridge’ from the simple colour exercise to the complex work of art, that is, to stimulate the student to contemplate and invent for example ‘Figures which may be called constructions in the abstract, but which may be named concretely: a star, vase, plant, animal, head, man, etc, according to the association they have conjured up’ (Spiller p.90).

Summary of the colour theories of Johannes Itten

Principal reference: Johannes Itten (1961), Kunst der Farbe, Ravensburg, English translation by E. van Haagen as The Art of Color, New York: Van Nostrand Reinhold.

A principal influence on the educational theories of Itten was Adolf Hölzel, with whom he had studied at Stuttgart Academy (1913-16). Hölzel’s teaching, unusually for its time, stressed spontaneity of personal expression in the use of colour and drawing. While teaching in Vienna, Itten also encountered the liberal teaching theories of Franz Cizek, and early supporter of child-centred education, intent, according to Dick Field (Field p.53) on encouraging his pupils to ‘grow , develop and mature’. Both approaches became considerations for Itten, following his appointment to direct the preliminary programme at the Weimar Bauhaus between 1920 and 1923 (when he was replaced by Moholy-Nagy).

Itten encouraged his students to explore their personal responses to colour, and to deduce what selected combinations of colour might offer about the personality or ‘inner life’ of the artist. He also held that appropriate theoretical grounding was useful if not indispensable in selected circumstances:

While acknowledging that, ‘In moments of strength, problems are solved instinctively, as if of themselves’. He also acknowledged that 'Doctrines and theories are best for weaker moments .... We can be released from subjective bondage only through knowledge and awareness of objective principles’ (Itten p.12).

Itten does not propose an identifiable colour theory as such (any more than Kandinsky did), but rather offers a framework within which the student may relate essential areas of colour investigation. His contribution, in his own words, was to furnish ‘a serviceable conveyance in which the colour artist may travel a longish distance upon his way (Itten p.153). Central to this approach was to divide colour study into three initial categories:

Impression (the objective observation of colour)

Expression (the emotional response to colour)

Construction (the symbolic or abstract use of colour)

In each of these the student is directed to explore seven categories of colour comparison, as follows:

1. Contrast of hue, including contrast of hue with black and white

2. Contrast of light-dark (tonality), to include lightening and darkening of colorants with black and white

3. The cold-warm hue contrast

4. Complementary colour contrast, in which vivid colours are progressively neutralised when mixed with others opposite on the colour circle

5. simultaneous colour contrast

6. Contrast of saturation or chroma

7. Contrast of extension (comparison of varying size)

Itten’s framework can be considered thorough, but for its omission of the temporal experience of colour perception (included by Klee and Albers) such as afterimage and visual search patterns). In other respects, it is not dissimilar to the seven categories of ‘colour contrast’ proposed by Hilaire Hiler (Hiler p.9) and the six variations of colour contrast proposed by Paul Renner (Renner p.54).

Parallel with such explorations, Itten urges awareness of the expressive potential of such formal factors as transparency and opacity, near and far, lightweight and heavyweight, and rare and dense.

Itten’s significant impact on the first three years of the Bauhaus’ existance was curtailed in 1923 when he was dismissed and replaced by Moholy-Nagy. As a consequence, Itten’s emphasis on the expression of subjectivity by graphic and colour means was extinguished in favour of the relatively objective language of Constructivism. Moholy-Nagy’s arrival also seemed to simulate a self consciousness in the romantic tendencies of Kandinsky and Klee, and Klee’s magnificent series of imaginative watercolours halts in favour of much more minimalist compositions.

Summary of the colour theories of Josef Albers

Principal reference: Josef Albers, Interaction of Color, New Haven & London: Yale University Press, and subsequent paperback editions.

Though born in the same year at Itten, Albers was a student of Itten’s for a number of years before being promoted to assist in actual teaching of the compulsory Basic Course. From then on, there was a much greater emphasis on the mechanical and the functional, the ‘mathematical-geometrical’ and the exploration of the intrinsic and performance qualities of materials.

A major advance in contemporary colour teaching undoubtedly occurred with the publication of Albers’ Interaction of Colour in 1963.

The original publication consisted of a boxed portfolio of some 150 small-scale silkscreen prints, illustrating remarkable examples of so-called optical colour illusions accompanied by a sparse (and not entirely accurate) text, for example, with respect to optical and subtractive colour mixing The images include some of the finest examples of Albers’ students’ colour work undertaken at Yale University (USA) between 1950 and 1960.

Though Albers colour teaching was influential on so called Op Art of the 1960s, all of the illusory effects he refers to would have been familiar to Michel-Eugène Chevreul, whose publication of 1839 represents perhaps the still the most exhaustive survey of optical colour contrasts. Albers significant achievement was to enthuse in a new generation of students theories of colour established a century before, by such writers as James Clerk Maxwell and Ogden Rood.

An obvious omission from the Interaction of Colour, as Albers acknowledges, is mention of the physics of light and the physiology of vision. In Albers’ opinion ‘the scientific analysis of the physical quantities [of light] is not the problem of the colourist’. This might be true if the colourist was restricted to making collages of ready-coloured papers (as Albers recommends for his teaching), but not where an artist may also be working with projected light.

Albers begins his ‘experimental way of studying colour’ by encouraging the student to utilise simple abstract designs in order to make the same ‘physical’ colour look different psychologically within differently contexts. This is most easily accomplished by placing two small rectangles of the colour on differently coloured backgrounds. Following repeated experiments, ‘it is discovered that certain colours are hard to change, and that others are more susceptible to change’ (Albers p.9).

Somewhat like Klee’s colour teaching, it is from such simple (but challenging) beginnings that the student is invited to explore more open-ended possibilities, through, for Albers, only within the abstract composition. The benefit of repeating such exercises results not only in greater awareness of effects which a casual or untrained observer might not notice, but also in sensitising the student to the subtle modifications which occur when colours are placed in close juxtaposition. Further exercises involve the student in the accurate assessment of transparency or translucency illusions

A principal objection to the Albers approach to colour teaching might be the danger that such studies do not feed directly into the personal practice of the student as a pictorial artist, but remain only as somewhat impersonal exercises. In other words, practice may stimulate theory, but remain stillborn if it is not then integrated within the greater, overall artistic development of the student.

In The Hidden Order of Art, first published in 1967, the Anglo-German educator Anton Ehrenzweig commends Albers’ compilation of colour studies, but notes that ‘the all-important relationship between form and colour’ is conspicuous by its absence. Ehrenzweig argues that, comprehensive though the Interaction of Colour may be, Albers has omitted at least one principle able potentially to provide a rich basis for encouraging ‘by trial and error – an eye for colour’:

‘Broadly speaking, a strong composition inhibits the mutual enhancement of colour surfaces (simultaneous colour contrast, colour interaction); conversely the mutual enhancement of colours tends to weaken form and tonal contrasts, [and] the relationship between figure and ground and illusions of depth produced by perspective’ (Ehrenzweig p.155).

In other words, Ehrenzweig seems to argue that, while Albers ‘systematic exercises’ cannot but succeed in increasing the student’s awareness of colour interaction (or the relationship between colour and colour), they do so at the expense of tediousness of form. (Albers, it may be remembered has his roots in the Bauhaus at a time when the ‘functionalist’ ethic was at its height.)

The probable reasoning behind Albers’ approach was that simplicity of form will throw greater emphasis on to colour. The suppression of one or more parameter in the methodical study of any subject will tend to sharpen the focus on another, as one may find with the early Minimalist colour-field paintings by William Turnbull, Kenneth Noland, Gene Davis, and even Albers himself.

Paul Renner similarly recommended simple grids of horizontal lines be used for his colour exercises at least as early as 1947.

Ehrenzweig admits a disappointment that Albers’ ‘magnificent book ... hardly mentions how much active colour interaction depends on comparative weakness of form,’ especially as ‘his own painting is perhaps the best example of this law.’ Much of the last 25 years of Albers’ life was dedicated to painting variations of the Homage to the Square series, in which a large square encloses squares of decreasing size. As Ehrenzweig observed, ‘the weakest shape one can put into a square is another smaller square’. As a consequence, the interaction of colour within such a formally unchallenging design is ‘greatly enhanced’.

There is little evidence that Albers’ studied colour methodically until the late 1940s; and I propose that it was not until he had distilled his formal compositions to the ultimate simplicity of the Homage to the Square format, that he found himself facing something of a ‘formal dead-end’.

In looking for new avenues of exploration, it may have seemed an obvious possibility, at least for the time being, to hold the format constant, and begin a methodical exploration of colour. This exploration was to hold so many possibilities (in his art and in his teaching) that they remained far from exhausted even at his death in 1976. It was perhaps the stability of the Homage to the Square format that offered such an ideal foundation from which to examine so extensively what Albers acknowledged to be the ‘relativity and instability’ of colour relationships and illusions.

Ehrenzweig observes that ‘Equally weak is a circle within a circle in the manner of fashionable ‘target pictures’ (by Robert Delaunay, Jasper Johns, Peter Sedgley, etc). While this strategy of ‘non-distracting form’ may have satisfied Albers’ own artistic aims, from an educational standpoint there is a danger of it presenting something of a dead-end for students wishing to work within much more complex (and especially pictorial) formal compositions.

As Dick Field warned (1970 p.64), in the context of overt Bauhaus influences in the art-school teaching of his time: ‘An inevitable problem comes to the front at this point: the difficulty students experience in using the Basic Course ... unless they followed its forms and continued to work in an abstract way. The problem was obscured by the fact that most students [at that time] wanted to work abstractly;... but there were always students unable to move on.

I have attempted to examine methodically the relationship between colour and ‘form’ in the text of The Colour Coursebook (to be published). By this I mean the relationship between colour and various abstract and pictorial ‘formal’ devices, to include figure-ground division and ambiguity, contour, highlight-and-shadow, texture, opacity and transparency, pattern and composition. Integrating an exploration of colour-form relationships (in addition to colour-colour relationships) may be considered an enhancement of the creative practices of the artist, augmenting whatever else may be the content or intention of a visual artwork or design.

Ehrenzweig comes close to recommending such an approach to teaching colour theory in relation to figure-ground perception of ‘form’ when he states, with reference to Josef Albers and Hermann von Helmholtz (the nineteenth-century scientist):

‘The incisiveness of form, such as the comparative sharpness of outline ... can be summed up as qualities of ‘good’ gestalt. We can summarize therefore that colour interaction between figure and ground stands in inverse proportion to the good gestalt of the figure’ (Ehrenzweig p.159).

A principle implied here is that distinct figure-ground divisions will tend to discourage active colour interaction, for the simple reason that colours associated with an ‘advancing’ figure will not so easily interact with colours associated with a ‘receding’ ground. However, Ehrenzweig’s own ‘experimental course for art teachers’ (as described in The Hidden Order of Art), confined colours within simple, geometrical (chessboard) grids, and appears (like Albers) somewhat to have missed the opportunity of exploring the relationship between colour and more-complex (especially pictorial) formats.

In theory, a practice of ‘dissolving’ figure-ground divisions (or encouraging figure-ground ambiguity) will also increase the interaction (or ease of comparison) of colours throughout a composition or design as a whole. Such a strategy would of course destroy legibility in the instance of typographical design, and remove most of the clues for the perception of solid objects in space within an interior design. But within a painting, where such factors need not be of consideration, dissolution of the division between figure and ground can provide a challenging strategy in investigating the relationship between colour and form.

There are numerous ways of undermining clear figure-ground divisions, such as rejecting contrasts of highlight-and-shadow (Pierre Bonnard, etc), fragmenting linear perspective (Picasso and Braque, etc), introducing transparency of both colour and form (Paul Klee in his watercolours, etc), fragmenting contours (Alberto Giacometti, etc), avoiding size clues by adopting regular patterning (Bridget Riley, etc), and so on. Many of these methods were exploited, intuitively or otherwise, by artists throughout the twentieth century.

In their own way Klee, Itten and Albers each offer valuable contributions to the teaching of colour theory. Itten’s text includes information on Newton’s scientific principles which those of Klee and Albers do not. Itten offered a useful framework for the methodical study of colour-contrast effects and proceeded to discuss them as best as he could within pictorial artistic contexts. Itten also appears alone in introducing the student to connotations of colour in the context of cultural traditions.

Albers chooses to concentrate on a narrower aspect of colour study – its relativity in the context of simple abstract formats – though there are rich possibilities in experimenting within this area. Albers’ teaching lacks most conspicuously an obvious ‘bridge’ from simple theory to complex practise, that is, from the simple colour exercise to the integration of colour study within complex pictorial or representational compositions.

In my opinion Klee succeeds most admirably of all in this since, on his own admission, his strategy for teaching colour is not only integrated with drawing and painting but also inseparably with considerations not only of form and composition but of content and pictorial imagination.

In the absence of such considerations, there is a risk of perpetuating art-school colour theory as a specialist subject in itself and not obviously related, by the student or teacher to subsequent diverse artistic practice.


Back to top

Copyright © 2005 Roy Osborne. All rights reserved.
Related Subjects
Areas of Knowledge
What colours are in this season? Visit to find out!