This Post makes available the Chapter 19 of my book “Painting with Light and Colour”. Its purpose is to confront the issue of “feeling” in the appreciation and use of colour in painting.
Although this is the only chapter in this book which uses the word “feeling” in its title, it is far from being the only place where the fundamental importance of the feelings in artistic creativity is mentioned. Even when it is not mentioned specifically, its importance is assumed.
Indeed, one of the key proposals that underpins the series as a whole is that the brain’s “feel system” is essential for ALL “learning” and ALL “creativity” and, moreover, that this is the case in ALL domains of human activity.
While it is common experience that the feelings play a key role in providing the “motivation” that is necessary for everything we do or avoid doing, few seem to realise just how deeply embedded it is in the way our brains work. Above all, it is the arbiter that allows us to use “feedback”, a characteristic that means it is involved in determining the outcome of ALL decision making. Thus, it is inextricably involved in determining judgements between:
Each individual’s assessment of what is “Good” and what is “bad”,
Degrees of “Similarity” and “difference”.
(For more on this fundamental subject, use the two links at the bottom of this page (below the image of the Patrick Heron painting) to access two chapters from of two of my other books. Both have much to say about the role of feeling, in all its manifestation, including ways in which the brain’s “feel systems” are involved).
This Post provides a link to chapter 18 of my book, “Painting with Light and Colour”. It is the last of the chapters in the part of the book dedicated to “painting with light”. Its title, “All you need to know about painting-2” is almost the same as the title of Chapter 1, except for the number 2 tagged on at the end. The grandiose claim was made by my teacher Professor Marian Bohuz-Szyszko, during a brief encounter on the very first day we met. He asserted that “all you need to know” can be summerised in two simple rules.
The purpose of the chapter is to consider the plausibility of his assertion in the light of the ideas developed in the Chapters 2 to 17. These not only delve into the historical origins of the rules, but also provide scientific evidence of their power as tools for artists.
After the link to Chapter 18, I have added a slightly edited version of its “Introductory”, as a means of better preparing you for its contents.
Introductory to chapter 18 of “Painting with Light and Colour”
We have now come to the last chapter and the question as to how to make the best use of the information and ideas presented.
The first chapter introduces the five propositions of Marian Bohusz-Szyszko, the ones that according to him constitute, “all you need to know about painting”. The chapters that follow provide an account of their historical and scientific origins and explain why they are so powerful. At the same time they point out some limitations. However, although the avoidance of repetition and the use of complex, complementary-containing colours can transform what artists can achieve, they certainly do not represent “all there is to know about painting”, not even with the modifications and extensions suggested in this book. Most notably, the Professor’s rules give short shrift to two subjects that many artists consider to be of the utmost importance Thus they:
Have no relevance to the kind of “colour dynamics” that can be generated between juxtaposed colours (the subject of the following chapters)
Do not address what is perhaps the most important topic of all, namely the role of the feelings.
Although a full discussion of the importance of the feelings as a driving force in all domains of creativity is reserved for “Fresh Perspectives on Creativity”, it would not do at all to neglect them entirely in what follows.
.
The earlier chapters from “Painting with Light and Colour
This ‘Post’, as well as providing a link to Chapter 17 of my book “Painting with Light and Colour”, concentrates on a number of practical matters that need to be taken into account before starting to paint. Most of the suggestions concern all mediums. Where there are differences, these will be pointed out. As a bonus two highly complex images are used to illustrate practical ways of differentiating a multiplicity of colours.
The two images below are used to illustrate different approaches to ensuring that no repetition of colour occurs. The first, a pastel painting of a forest scene, was made by means of unaided visual analystic strategies and the second, a constructivist composition using acrylic paint, required the use of systematic mixing procedures.
Differentiating by eye
Figure 1 is a pastel painting. As explained in Chapter 16, to obtain a large range of subtly different pastel colours, it is necessary to have a sufficiently large number of colours to work with. It is well worth remembering that, just as important as having a sufficiency of pure pigment colours available, is to have each of them represented at a range of lightness levels. Particularly important is to include both the lightest and the darkest possible levels. To make the painting of the forest illustrated above, I probably used something like 16 base colours at 7 lightnesses levels.
I have two complementary approaches to ensuring that no colour is the same as any other colour:
The first depends on the fact that any repeated colours jump out of illusory pictorial space (See Chapter 10). The result is an ambiguity between a real surface and an illusory pictorial space interpretation. These provide the eye/brain with an insoluble interpretive problem (as explained in previous chapters). The result is both disturbing and attention grabbing.
The second depends on the fact that our analytic-looking systems are extremely good at detecting very small differences between any two compared colours.
Thus, the first strategy for ensuring that no colour is the same as any other colour, is first to allow attention to be drawn to the colours that jump out of the illusory pictorial space and, then, to differentiate between them by adding subtle touches of colour. When doing so, it is useful to remember that almost imperceptible differences can have a transformative effect.
The second strategy is to make comparisons between all the regions of colour on the picture surface, however small they may be. In a painting as complex as the one imaged in Figure 1, this means a very large number of comparisons, which is why it took such a very long time to complete.
Differentiating by means of systematic procedures.
Figure 2 is an acrylic painting that posed the problem of creating an array of 289 different oranges that, at first glance, viewers could easily mistake for being the same. This requirement implied, not only that neighbouring colours on the same row would not be perceived as being different but also neighbours from both the row below and the row above. The solution adopted was to premix and store in separate pots, a sequence of oranges, each of which was just noticeably different (JND) from its neighbours. Any mixture between any two in this sequence would produce colours that coud not be distinguished by eye. Accordingly, any progression of mixtures between them, would produce an invisible progression analogous to the progression colours that occurs between any adjacent parts of a uniformly painted flat surface (for example a wall). It is only when you compare colours situated at a distance from one another that the fact that a progression has taken place becomes evident.
This Post provides a link with CHAPTER 16 of my book “Painting with Light and Colour”. It is the first of three chapters that provide different approaches to reviewing the ideas presented so far. To whet your appetite, after an image of a painting made with a view to exploring the rules of Professor Marian Bohusz-Szyszko, there follows a reprise of the “Introductory”:
.
.
Introductory
We have now arrived at the final section of “Painting with Light” . In it are three chapters. The first (this one) reviews ideas that have been presented in previous chapters. As explained there, some of these have a long history, going back both to the Italian Renaissance and to 19th century developments in science. Others are much more recent, arising from advances in the relatively new disciplines of neurophysiology and computer modelling. From the point of view of the subject of “painting with light”, the most important of the new insights concerns ways in which the eye/brain mediates our perception of surface solidity, surface form, illusory pictorial space and ambient illumination (the prevailing quality of light). It tells of two great breakthroughs. The first being the late 18th century realisation that colour and, indeed, all visual experience is made in the head. The second, the discovery, roughly two centuries later, of two functionally independent visual-systems. One of these has much to tell us about surface-reflection (the subject of the previous chapters) and the other about body-colour, the subject of the chapters that follow. This chapter is also in two parts. The first contains a review of the theory. The second moves on to a list of basic questions relating to it. Each of these is accompanied by a short answer.
Below are images of seven paintings, made between 1964 and 2015 that explore the ideas of Professor Marian Bohusz-Szyszko, as presented in previous chapters. It is important to stress that although they give some idea of the range of possibilities, they cannot reproduce the all important “harmony that runs parallel to nature”. For that it is necessary to have real paint on a real flat picture surface, illuminated by an external light source (preferably a natural light one)
Clement Greenberg, Jackson Pollock and a space within the picture surface
Clement Greenberg was a high profile mid twentieth century art critic whose thoughts on Modernism in Painting influenced a generation of artists. He had much to say on the paintings of Jackson Pollock to which he accorded a special importance. One of the reasons why was that he saw in them an unprecedented type of “pictorial space”. This he called “a space within the picture surface”
.
.
A moralistic reaction against photographs
His claim had a history. In the late 1860s. a number of young artists saw a threat in the degree of realism manifested in images taken by the recently invented camera. The approach they chose for countering it was purely conceptual. It was based on a two step rationalization. First, they argued that realism of photographic images was such that it deceives the eyes and, second, that deception is “immoral”.
The way these artists, later known as the Impressionists, sought to avoid analogous immorality in their own figurative paintings was to emphasise the reality of the actual picture surface. Their main strategy for doing this was accentuate “surface texture” and “personalised mark-making”. Luckily for the history of painting, one outcome was that they discovered the exciting potential of exploiting the dynamic relationships between the reality of the picture surface and the illusion created by the figurative aspects of their work.
.
.
Two new developments
.
.
In the course of time, the successors of these early Modernist painters discovered that figuration is not necessary for creating illusory pictorial space. They found that perceptions of it could be evoked in non figurative paintings, by means of “cognitive cues”, the most important of these being “overlap” and depth-indicating “diagonals”.
.
.
However, despite the exciting possibilities offered by real surface/illusory pictorial space dynamics, Piet Mondrian and other purists could not help concluding that any kind of illusory pictorial space was a “deception” and, accordingly, both “immoral” and to be avoided at all costs. Nor did they consider this to be a trivial matter. Mondrian permanently broke off his close, working relationship with Bart van de Leck because he had included a diagonal in one of his paintings.
Poor Mondrian had to soldier on alone. At times he must have felt despairing. However hard he tried, he could not find a way of altogether eliminating perceptions of illusory pictorial space. Eventually, his deeply religious worldview rescued him. It suggested a way of rationalising himself out of the morality problem, namely, to rename the “pictorial space” that he was unable to eradicate as “spiritual space”.
An illusory space within the picture surface
Seemingly unaware of Mondrian’s discoveries, Greenberg discerned what he described as “a space withing the picture surface” in the paintings of Jackson Pollock. This he argued was fundamental different from the “illusory pictorial space” found in the work of his Modernist Painter predecessors. However, although influential for a number of years, his claim did not convince number of later artists and theorists. Although these had no difficulty in seeing the space that Greenberg has identified, they insisted that it was illusory and they were right to do so. We now know that they were seeing was the same kind of illusory pictorial space found in the the“vase/face illusion”, in which the vase can be perceived as being either in front of or behind the inward looking faces .
The problem that now arose was that this kind of illusory space provoked an optical disturbance of a kind that some saw as an undesirable (described in the chapter on “negative spaces” in “Drawing on Both Sides of the Brain”). To remove this in the interests of creating a pure, undisturbed colour experience, artists of the 1950s and 1960s, felt obliged to eliminate all traces of Greenberg’s “space within the picture surface”.
.
.
However, when they tried to do so, they found the task they has set themselves to be more difficult than they had bargained for. Indeed many decided that they were faced with an insoluble problem. They had come to the conclusion that, as long as there is more than one region colour on the picture surface, there is no way of avoiding some degree of visual tension. The reason was no longer necessarily the repeated colours of Mondrian and Pollock, it could also be the figure/field separation that occurs at the earliest stage of perceptual processing, in which the object (figure) is picked out at the expense of its context/background (field). In the view of artists like Ellsworth Kelly, the only remaining way of producing pure colour experience would be by means of single colour paintings. As these could be described as “coloured objects hanging on a wall”, the question that then arose was whether they should be classified as “paintings” or as “sculptures”. Many seem to have plumped for the latter.
.
In 1968, the Museum of Modern Art in New York provided the art world with what turned out to be a watershed exhibition. Its title, “The Art of the Real” ,was deeply significant. From the conceptual point of view, it signed the death knell of one of the most influential strands of the broad tapestry of “Modernism in Painting”. For those that believed this to be the only strand, the future of painting had to be in some manifestation of “Post Modernism”.
None of this necessarily deterred artists from carrying on regardless with all the different possibilities. They will always be free to embark on new explorations of:
All are to be found in work produced up to the present day.
Footnote 1
As a footnote I think it worth mentioning that the issues discussed above were central to the teaching of both my main mentors.
Professor Bohusz-Szyszko felt the main importance of his rules was that they ensured what he described as “the integrity of the picture surface”.
Michael Kidner followed Greenberg and Pollock in believing in the importance of eliminating illusory pictorial space, as defined by them. He also made good use of what they described as “the space within the picture surface”, in other words what I have been likening to the illusory pictorial space exemplified in the vase/face illusion.
Footnote 2
Incidentally, Professor Bohusz-Szyszko told me that the fact that his rules ensured the “integrity of the picture surface” was one of the main virtues of the dogmas he shared with his students is. According to him repetitions of the same colour in different parts of a panting broke the “integrity of the picture surface“, either by jumping out in front of it or by creating the illusion of holes within it. We now know that this theorising was wrong. As explained in many places in my books, what they jump out of is illusory pictorial space. What actually happens is that they are perceived as jumping from surfaces in an illusory world, and integrating with the actual picture surface. From the practical point of view this theoretical distinction does not matter for in both cases the outcome is that the eye and the brain are confronted with incompatible and therefore disturbing perceptual cues.
During the twelve years I worked among scientists at the University of Stirling in Scotland, a transformation took place in my understanding of just about everything to do with the role of the eye and the brain in the organisation of the the main perceptual and motor skills used in the making of drawings and paintings. PART 2 of my book “What Scientists can Learn from Artists” tells of experiments done by myself, colleagues and other scientists that made especially significant contributions to this exciting development.
Chapter 7, (accessed by clicking on link below) offers an autobiographical introduction the contents of PART 2 that gives a flavour of what I was up to in those years. A theme that runs through its pages is that the transformative learning was a two way process, offering benefits to all concerned. Time revealed many unexpected advantages in my being a combination of an experienced artist/teacher and a naive beginner in all the scientific disciplines in which I was to participate. My new colleagues found themselves faced with a drip feed of questions coming from unfamiliar perspectives that were to prove their value as catalysts capable of stimulating new ideas for a surprising number of highly expert scientists, working in a variety of disciplines. In return, their often participatory responses enabled me to put together the body of ideas that underpin the originality of my books, my teaching and, to an important extent, my work as an artist.
Over the centuries, at least since the Italian Renaissance, artists have sought to represent three dimensional objects and scenes on two dimensional surfaces. By implication, this required them to create an eye-deceiving third dimension (‘trompe-l’œil‘).
To achieve their objective, they and their successors: (a) mastered the laws of ‘linear perspective’, (b) delved deeply into the subject of‘anatomy’, (c) explored the form-making properties of ‘gradation’, (d) recognised the importance of‘overlap’, (e) provided explanations for the phenomenon of ‘aerial perspective’, (f) explored whole-field lightness relations (‘chiaroscuro’) and (g) demonstrated the value of existing knowledge of the form of objects and the layout of scenes in influencing how viewers would perceive them (‘cognitive cues‘). In other words, over the centuries the artists have pioneered our understanding of just about everything that psychologists of perception needs to know about illusory pictorial space.
However, there was one big absence and it is this that dominates the discussion of illusory pictorial space in two of my books. In“Painting with Light and Colour” the subject is approached from the perspective of artistic practice. In “What Scientists can Learn from Artists”, which contains the chapter that can be obtained by clicking on the link below, its treatment has both scientists and artists in mind.
The chapter also touches briefly on the issue of what they saw as the immorality of deceiving the eye, which was to have both a decisive and long lasting effect on the evolution of painting from the Impressionists until the late 1960s at least. More on this in my other books.
Elizabeth Cavé and her great friend Eugene Delacroix
.
In an earlier Post I told of the teaching of Horace Lecoq Boisbaudranand its widespread influence. In it I did not mention another important figure who also developed a method for training the memory. Her name was Elizabeth Cavé. Like Lecoq Boisbaudran her method eventually found favour with the establishment and was to some extent introduced into the national curriculum. She was also, over some 30 years, a personal friend and confidant, often described as “mistress”, of Eugène Delacroix, who was something of a Father figure to the young Impressionists, including:
Edgar Degas who, as a young man, went, with his friend Gustave Moreau (teacher of Matisse), to visit Delacroix in his studio. When they arrived, despite having been warned to expect a testy old cumudgeon, they were given a warm and generous welcome. In contrast they described his intellect “icy”, maybe a reaction to the scientific bent that led him to be an early champion of the ideas of Michel-Eugène Chevreul, the chemist who first enunciated the law of ‘simultaneous colour contrast’
Probably all the other young ‘Impressionists’ for whom it is said that a ‘must see’ experience was the application of Chevreul’s law to be found in the frescoes painted by Delacroix towards his life (between 1857 and 1861), in the L’Eglise St. Sulpice, Paris.
.
My letter to LRB
With all this information in my head, you can imagine how my interest perked up when I came across a quotation from Delacroix in an article by T.J.Clark, published in the London Review of Books in October 2019. In this Delacroix tells us that he experienced a paradigm shift in his approach to painting, from being “hounded by a love of exactitude” to employing his memory to sift out “what is striking and poetic”. He also states that this transformation occurred as a spin-off from his “African voyage” in 1832.
On reading this endorsement of the virtues of channelling experience through memory, I was immediately reminded of the philosophy of Lecoq Boisbaudran. From there my mind jumped to Elizabeth Cavé and to wondering whether Delacroix’s change of direction had any link to her teaching method. When I discovered that their liaison had started in earnest in 1832, I could not resist the thought that either she had influenced Delacroix or, perhaps more likely, vice versa. If so, there seemed to be quite a lot to add to what T.J.Clark had to say. Below is what I wrote.
The letter
T.J.Clark (LRB 10-10-2019) quotes Eugene Delacroix as dating a change from being hounded by a love of exactitude to making work based on “recalling” what is striking and poetic. He asserted that it came after his “African voyage”, which mean after his return from Morocco in 1832. When I read this I immediately realised that this date roughly coincided with the beginning of his relationship with Elizabeth Cavé in 1833. Whether or not her ideas were influenced by Delacroix or visa versa , she published ‘Le dessin sans maître’, which received a laudatory review from her, by now long standing, friend (in the ‘Revue de deux Mondes’ of September 1850). In it, she explained her method of teaching drawing which, according to her, she had been practising since 1847. Key to this was training of the memory. Two years earlier, in 1848, Horace Lecoq Boisbaudran published a compilation of two texts, ‘L’Éducation de la mémoire pittoresque’ and ‘la formation de l’artiste’, in which he explained his method, also based on training the memory. His connection with Delacroix can be inferred from the personages in the 1864 painting ‘Homage à Delacroix’ by his pupil Henri Fantin-Latour, in which we see others two students of Lecoq Boisbaudran, Alphonse Legros and Felix Bracquemond. Also in the painting is James MacNeil Whistler who is know to have learnt Lecoq Boisbaudran’s method from Alphonse Legros and who famously demonstrated it to a doubter. He did this, first, by looking at an unfamiliar landscape and, then, turning his back on it and painting it from memory (for more about the influence of Lecoq Boisbaudran and its plausible ramifications see http://www.painting-school.com/horace-lecoq-boisbaudran-influence
So how does all this relate to the quotation from Delacroix? The clue lies in his youthful “love of exactitude” being replaced by a more mature approach based on “recalling what was striking and poetic.” What Lecoq Boisbaudran would surely have argued is that the great man’s earlier obsession with ‘accuracy’ prepared him for his later personalised use of memory with all its benefits, for this was exactly what his teaching method (and presumably that of Elizabeth Cave) aimed at achieving. The main differences, he could argue, lay in the shortness of the time in which his students were expected to make their transition and the methodical progression from simple to complicated that characterised the learning exercises that made it possible. Surely, both Delacroix and Lecoq Boisbaudran would have concurred with Edgar Degas, significantly a great friend of Alphonse Legros, when he said, “It is always very well to copy what you see, but much better to draw what only the memory sees. Then you get a transformation, in which imagination works hand in hand with the memory and you reproduce only what has particularly struck you.”
.
As well as the personalisation of artistic output, the method had huge advantages in terms of rapidity of information pick up. The famous late watercolours (‘Cambodian dancers’, etc) of Rodin, another student and a lifelong admirer of Lecoq Boisbaudran and his teaching, illustrate both these advantages. Likewise the post-African paintings and drawings of Delacroix. Also, I find it hard to believe that there is not some connection here with Delacroix’s famous assertion that “any artists worth his salt should be able to draw a man that has been thrown out of a sixth floor window before he hits the ground.”
PS. For your interest, I was teaching on much the same principles as Lecoq Boisbaudran for at leat 25 years before I learnt of his existence. These I derived from research done at the University of Stirling in the early 1980s <http://www.painting-school.com/the-course/the-course-director/>.
Back in November 2019 I started posting chapters from “What Scientists can Learn from Artists”, the book which presents the research and the science based ideas that that lie behind much of the contents of my three other books: “Drawing on Both Sides of the Brain”, “Painting with Light and Colour” and ” Fresh Perspectives on Creativity”. I set the ball rolling with with six of the chapters that describe research findings which were in large part responsible for:
Overturning almost all the preconceptions I had about the nature of visual perception.
Providing the building blocks required for replacing them with the coherent picture presented in these books.
When I first came across the material I have summarised in these chapters, their cumulative effect on me was more than just fascinating. It amounted to a paradigm shift. My hope is that reading them will perform the same service for others, particularly when buttressed by the contents of earlier and later chapters.
Below is an extract from the “Preface” to “What Scientists can Learn from Artists”, which summarises its structure. The chapters so far published in my Posts come from PART 2. In the next weeks I will be posting chapters from PART 1 and in the coming months chapters from PART 3. I will wait to see the level of interest before I go on to PART 4, which I have reason to believe will be is considerably more demanding on non scientists.
Also below are links to already published Posts.
The structure of the book
Because the context of the knowledge of scientists and artists is so different, it seems prudent to provide a certain amount of background material which, while likely to be familiar to readers from one side of the arts/science divide, may well not be to those from the other. Thus PART 1 contains a number of general ideas both artistic and scientific many of which may well be familiar to one community and not the other, and PART 3 provides a basic introduction for non scientists to the nature of visual perception that emphasises the variety of visual systems involved in different aspects of visual processing. The function of PART 2 is to describe the main experiments used to underpin the theoretical speculations which lead to the general model of perceptual and cognitive processes that provides the subject matter for PART 4. Throughout the attempt has been made to present ideas in such a way that they will be understood by both groups.
Chapters from my book “What Scientists can Learn from Artists”
These deal with subjects that feature in the other volumes in greater depth.
There are still places vacant for the 2020 sessions of the Painting School of Montmiral. Here a three photos to remind you of our idyllic setting and the seriousness of the teaching
The purpose this Post is to provide a link to “Constraint in artistic aids and practices “, Chapter 9 in my book “What Scientists can Learn from Artists”. As in several other Posts that publish book chapters, I include a slightly edited reprise of its“introductory”, in the hope it will whet your appetite and encourage you to click on the link below. I am hoping that when you have read all the chapters of all my books, you will realise that the answer to the question posed in the heading to this section is “Yes”. The images below illustrate two methods of constraint favoured by artists in former centuries that foreshadow ones that are widely used today: For example, photographs, slide projections, and computer controlled images. All of these, whether consciously or not, make use of constraints, the possibilities of which have been developed by evolution over the millennia, such as standing still, choosing a viewing distance or closing an eye al of which constrain input to our visual systems and, thereby, enable learning and creativity, its corollary.
.
.
.
Introductory
If we want to be creative, we will have to free ourselves from the constraints of old ways of doing things in order to go beyond them into new territory.
In this chapter, we take a step towards the goal of a practical understanding of how this might be done. It starts with my telling how I stumbled on the intuition that constraint may be a necessary condition for exploring the unknown, and provides examples of how the community of artists, whether consciously or not, have made much use of this possibility. Eventually I found myself coming to the seemingly paradoxical conclusion that constraint is necessary if we are to achieve either meaningful freedom or creative self expression. I also came to realise that the use of constraint is one of the guiding principles of our evolution as a species.
My approach to going deeper into the creative powers of constraint, starts with account of how I came to realise their central importance. I use the particularities of my own story because of the insights it furnishes relating to the creative process in general: long periods of gathering data, struggles with the confusion that they seem to engender, a sudden intuition that provides a lead on how order might be found and, finally, doing the work necessary to test its validity.
The inspiration for my breakthrough came when reading a book by J.J. Gibson, one of the most controversial yet influential perceptual psychologists of the day.