Showing posts with label Expression. Show all posts
Showing posts with label Expression. Show all posts

Friday, 28 July 2023

What is a portrait anyway?

 

Kaye Donachie: Monotonous Remorse

I've recently returned to Leeds after a week's walking and drawing on the South Coast near Chichester. While away I went to the Pallent House Gallery and there were two exhibitions on there that made me think about portraiture again. On our first day in the area we had been browsing the Chichester charity shops and I bought an old copy of James Gleeson's 1969 Thames and Hudson biography of William Dobell, an Australian artist that I had never really thought about before. He had had a long painting career that was centred around portraiture, and as I was reading about him, images of his work were floating around in my mind when I visited these portrait focused exhibitions. I was also due to spend a session working with a group of other artists looking at making portraits the following week and I had my doubts as to whether or not I would be able to get anything worthwhile from the activity. 

Gwen John's and Kaye Donachie's portraits are very different and Donachie's in particular ask questions as to what portraiture is and can be about. At the centre of James Gleeson's text on Dobell is an extended account of a 1944 court-case, that had a profound mental effect on Dobell, and which raised even more questions as to what a portrait actually was. 

William Dobell: Joshua Smith: Winner: Archibald Prize for Portraiture 1943

Photograph of Joshua Smith: 1943


William Dobell: Preparatory drawings for the painting

Dobell entered his portrait of Joshua Smith into the Archibald prize in 1943, and immediately on his winning of the first prize, two disgruntled unsuccessful exhibitors, who were at the time also members of the Royal Art Society, decided to take legal action aimed at setting aside the trustee's decision. All of this seems a long time ago now, but if I think about how powerful a grip a TV program such as Sky Arts 'Portrait Artist of the Year' has on the general public's idea of portraiture, I suspect a similar argument could still be made in relation to how portraiture is thought about today, especially as most of the painters seem to work from gridded up IPad photographs of the sitters. The arguments in the court case surrounded the relationship between a portrait and a likeness or caricature. Some distortion, Dobell asserted, is essential when we have to get a visual image to reveal the 'real' inner character of a person. For myself, this particular issue is another aspect of the question, how can we visualise what is invisible or in this case how can what is visible reveal what is invisible? How does the external physiognomy of a face or body shape reflect the life that an individual has been through and how does this relate to their 'character'? What sort of inner 'truth' can we discern from our readings of the shape of an external body? Can we build exaggerations of perceived visual form in order to heighten character? Or, are the best portraits simply very accurate copies of what is there? As usual I need to ramble around this set of ideas in order to see where they are going. 

Dobell know Joshua Smith, a fellow artist, before undertaking to paint him, and they both applied for the competition. Dobell had reminded Smith on commencement that he used "an element of distortion in order to make the portrait more like the subject than he is himself"; which is a very interesting statement, how, I ask myself, is something more like the subject itself? This suggests that in some way a painting can be more 'deep' or 'profound' than the person that the image is made from. A doorway to a thought that suggests that art (for instance the Mona Lisa) carries far more cultural weight than the individual humans that are portrayed in its making. At the time the arguments were mainly laid out in the daily newspapers. 

"Joshua saw the progress of the portrait all the time I was painting it," Dobell was quoted as saying in the Sydney Sun. "He gave me every facility and was in no way critical when he saw the finished work ... he congratulated me on the portrait." So as far as Dobell was concerned not only was he happy with his portrait's 'likeness' but Smith was too. However the painter Wolinski declared of Dobell's painting, "If this is meant to set a new standard in art, it will start a revolt. Someone likened it to a praying mantis. I can think of no better description. I couldn't sleep after seeing it the first time." Another artist Mary Edwards declared, "Grotesquerie has been awarded the prize" and "it is a Pearl Harbour attack on art". She went on, "Not a child or an expectant mother should be allowed in the Art Gallery at present." Of course, record numbers visited the gallery to see what all the fuss was about. The gallery had to extend its opening hours, and then it prolonged the exhibition by a month. The question - Is this even a portrait? - was central to the debate. The Contemporary Art Society declared it the most important work of art entered in the competition, to others, this was the ugly face of modern art. 

Melbourne newspaper critic James Stuart MacDonald, wrote a scathing newspaper review of Dobell's portrait. Using the headline "Circus Freak Exhibit", he lambasted the work as not a portrait but a "fantasy", stating, "Mr Joshua Smith looks like a man naturally deformed or crippled, and suffering, in addition, from malnutrition."  The portrait being the "epitome of ugliness, malformation and gruesome taste".  In response to all this invective, eventually the key element of the argument taken to the courts was that this was not a portrait, but a caricature. 

So another digression. Back in 1938, but only 5 years before Dobell painted his portrait of Joshua Smith, E. H. Gombrich wrote an essay on caricature, in this essay he states, "Freud's Interpretation of Dreams (1913) marks a turning point in the history of aesthetics just as it does in psychiatry. But more than this, psycho-analysis, though it has up to the present only registered the fact that there must be psychological laws of art-construction as there are laws of dream-construction, is proving of great value to the student of aesthetics, because it deals with man as a creature of conflicts and troubles". (Gombrich, 1938, p. 319) What both Gombrich and Dobell had both recognised was that aesthetics had moved on to be able to incorporate the thinking of Freud, and that 'psychological laws of art-construction' were now in place and that they needed to be taken into consideration in a similar way to the laws of dream-construction. Further more 'caricature' was seen by Gombrich as being a particularly effective trigger, that could enable us to enter into an awareness of these issues. Gombrich goes on to state that, 'it is a likeness more true than mere imitation could be' and that 'caricature, showing more of the essential, is truer than reality itself.' (Ibid) Gombrich then goes on to make a clear statement about what a portrait artist was supposed to be doing. 'By the seventeenth century the portrait painter's task was to reveal the character, the essence of the man in an heroic sense. The caricaturist has a corresponding aim. He does not seek the perfect form but the perfect deformity, thus penetrating through the mere outward appearance to the inner being in all its littleness or ugliness.' In doing so he makes a very interesting distinction between the portrait painter and the caricaturist, who he says looks for the 'perfect deformity', (Ibid) but in a way he brings them both together by stressing that portraits seek 'the essence' of someone, whilst caricature needs to 'penetrate through mere outward appearance'. 

The late 16th century marks the time when for Gombrich the work of art is, 'for the first time in European history—considered as a projection of an inner image. It is not its proximity to reality that proves its value but its nearness to the artist's psychic life. Thus for the first time the sketch was held in high esteem as the most direct document of inspiration'. (Ibid)

Sketch of a man: Presumed study for Billy Boy

Billy Boy: William Dobell

When preparing for a painting Dobell would visit the sitter and make sketches, he did not however work directly from them, he would put them up on a wall behind him whilst making the actual painting. The sketches were his attempt to find a way in to a visual trigger that would reveal an essence of some sort. In this way he made sure 'verisimilitude' did not get in the way of psychological 'truth'. If we look at his preparatory sketch for Smith's portrait we can see that the sketch is less distorted than the final painting, but the features that Dobell picks out as being essential to the character are already evident. The sketch's exaggeration of forms sits roughly halfway between the more attenuated figure produced for the final painting and the black and white photograph of Smith. This psychological approach it could be argued is yet another example of the continuing Romantic tradition. However in portraiture the most extreme examples of psychologically intense exploration have traditionally been self portraits, whereby the artist does know something about the inner mental life of the subject because he or she has to live out that life. 

In English, the first use of the term “self-portrait” was in 1810, and it was derived from the German “Selbstbildnis” itself a term that had come into use only a little earlier. The French word “autoportrait” did not gain currency until the 1920s. This suggests that there had been a fundamental change in how the portrait was thought of. Artists had of course in the past made images of themselves, but as a genre, they were not picked out, so what had changed? Our interest in the self portrait I would suggest was heightened by Romanticism. Romanticism, a Western European intellectual phenomenon, celebrated the individual imagination and intuition as well as advocating individual rights and liberty.  From approximately 1800 to 1850 it was at its height as a movement, and its influence continued well into the late 20th century and many would argue it is still one of the dominant ideas behind our current difficulty in reconciling a celebration of human individuality with a need to let the non human world have its say.  In my mind the key self portrait in relation to this shift in thinking is Chardin's pastel of 1776. 

Jean-Baptiste-Siméon Chardin, Portrait de artiste peint par lui-même, 1776. Pastel

Chardin's self-portrait, is a very private, self-aware image. He sits hunched at his easel, lost in the activity of looking at himself and in that absorption he avoids trying to make an image that reflects the gaze of others. This portrait is not just showing us a truer image of the artist, it also suggests an inner mental life whereby he is unmasked, no longer playing out his social role of artist/gentleman, he is now exposed as Chardin, the troubled man. After this point the self portrait I believe becomes more self conscious and as it does so it opens itself up to the need to become more expressive, rather than representational. This situation is of course exacerbated by the fact that in 1839, Robert Cornelius had shot the first successful photographic portrait using the daguerreotype and artists were becoming freed from the need to make accurate representations, because photographs, it was believed, could do this far quicker, cheaper and far more accurately. 

Robert Cornelius: Photographic self-portrait: 1839

The debate at the centre of the Dobell trial would seem to be really about a much wider issue, that of Modernism itself and its decoupling of the making of art images from their need to be accurate visual representations or copies of what people could see. 

But there are other strands to this set of ruminations and before I get round finally to considering the work on exhibition at the Pallent House gallery, I want to throw into the arena two other issues. The idea of the living reality of a portrait was put forward by Oscar Wilde in his 1890 novel 'The Portrait of Dorian Grey', in the novel a young man sells his soul for eternal youth and beauty, as he pursues a debauched life. The effects on his physical appearance then slowly change his painted portrait which he keeps locked up and hidden from everyone else's view.  This malignant portrait of the effects of evil and debauchery on a young aesthete in late-19th-century England is both a classic work of art in itself and the model of a idea, whereby we feel we should be able to spot the effects of life's experiences on someone's face. Conversely we are easily fooled by an evil heart inhabiting an innocent looking body. The painter Ivan Albright was chosen to make the painting that was used for the 1945 movie adaption of the story and he was chosen because at the time his work was seen as a clear example of a painter who was able in his portraits to depict the effects of life's experiences on someone's face.

If you look at two self portraits by Albright you can see two very different attempts to do this.

Ivan Albright: Self portrait before a stroke
Ivan Albright 1983 self portrait made after a stroke

Albright at work on the portrait of Dorian Grey

Ivan Albright: Portrait of Dorian Grey 1944

Oscar Wilde's story brings to a head the relationship between a portrait and the idea of an artist's ability to dig down deep into the 'real' character beneath a face's external appearance. In Wilde's story a myth was made into a narrative reality. Albright would have had to paint his 
Portrait of Dorian Grey before commencement of the shooting of the film, so it was probably painted during exactly the same time that Dobell was working on his portrait of Joshua Smith. Judging by the reactions to Dobell's painting at the time you could easily swap one for the other. Albright intending to portray a base inner character and Dobell perhaps innocently achieving a similar thing, his portrait being seen as an "epitome of ugliness, malformation and gruesome taste". 

The portraits of Maria Lani 

If though we compare Wilde's tale with the story of Maria Lani a woman who was painted by many artists, we could come to a very different conclusion. In the late 1920s she was portrayed in paintings and sculpture by over fifty artists, including Bonnard, Chagall, Cocteau, Derain, Matisse, Rouault, and Suzanne Valadon. When we look at their various portraits of her, we can only come to one conclusion, each artist is expressing themselves 'through' the face of the sitter. They are in effect using the portrait to express themselves and what is revealed is the character of the artist rather than that of the sitter. 

Kees Van Dongen: Maria Lani 
Picabia and Rouault: Maria Lani 

Matisse and Soutine: Maria Lani 

What is wrong with Modern Artists?

The aspect of contemporary art that allows for 'expression' also allows for distortion. This has confused and baffled people for over a hundred years and the newspaper article above is typical of the reaction the press in particular have taken to the artist's licence for distorting reality. The various lenses that each artist applies to the world are their passports to fame, and when you look at the images above, the Soutine painting is clearly a 'Soutine', and the Picabia a 'Picabia'. These artists are known for their 'signature' styles and if you couldn't identify them from their way of rendering form, they would have failed in their various attempts to be 'true' to their own sensibilities. The newspaper headline, 'What is wrong with the heads of Modern artists', is also a reminder of how easily newspapers can fan the flames of division and hatred of anything seen as different or not fitting in with what is seen as established taste. We must never forget that a denunciation of visual exaggeration and formal inventiveness as forms of expression became central to the Nazi project of stigmatising certain people as 'degenerate'. Classical styles are also 'styles', but like the fish swimming in water who cant see it because it is just 'there', many Western civilisations have held classical art up as a model that is the only one to follow. In the case of the Nazi party, the classical model was a measure that was used to help fulfil the dreams of Adolf Hitler who stated in 1935, "It is not the mission of art, to wallow in filth for filth's sake, to paint the human being only in a state of putrefaction, to draw cretins as symbols of motherhood, or to present deformed idiots as representatives of manly strength."

Adolf Hitler and other Nazi officials at the Dada wall of the Degenerate Art exhibition: 1937

In 1937 a commission led by Adolf Ziegler, Hitler's favourite painter, was charged with purging German museums of unacceptable art. The model that all artists were told to aspire to was Ziegler's particular brand of sleazy classicism, a type of painting that when we look at it now, we cringe at its celebration of the male gaze and soft porn portrayal of women. 

Adolf Ziegler (1892-1959) The Four Elements: 1937

It is worthwhile reminding ourselves that Dobell was painting his portrait during the Second World War. His painting of Smith was painted only 6 years after Ziegler's 'Four Elements' and in many ways Hitler's rhetoric parallels that of the Melbourne critic who wrote about Dobell's image as being 'deformed or crippled'. How easily we can be seduced by rhetoric to believe that there must be something wrong with people, (artists) that don't see the world in the same way as everyone else. It is a continuing issue, this is a Daily Mail headline from 2012, 'Moral vacuity and rancid opportunism. Sorry Damien, but you're a FRAUD', the Mail arts reporter Quentin Letts goes on to say, 'London's Damien Hirst exhibition stinks. Literally. From a work called A Thousand Years seeps the pong of rotting flesh. This may be because its large glass box contains the severed head of a cow. It has oozed a small, drying patch of blood and the box is full of flies.' When I first saw 'A Thousand Years' I realised Hirst had made a very powerful statement, and no matter how much I might find some of his work a little thin, there is no denying that at his best he has made some of the most significant statements about mortality of our times. The use of phrases such as 'Moral vacuity and rancid opportunism', once again reminds us of how newspapers use emotional triggers to push people's buttons. 

Perhaps after my digression its now time to return to the two exhibitions. The art critic Waldemar Januszczak states of the Gwen John exhibition that these paintings are, 'works of control, self knowledge and immense feeling'. Although many of the works displayed are portraits there is no mention of insight into any of the sitter's characters, these paintings are about Gwen John's control of colour, her obsession with certain paint handling qualities, often leading to her addition of chalk into her oil paint mixes in order to get a 'dry' surface appearance, her close control of tonality and use of centralised compositions to achieve a sense of timeless stability. All these things combine to ensure that her 'immense feeling' for life is communicated through these images. These qualities being what we recognise as her 'style'. 



Gwen John: Portraits

Gwen John is also well known for her delicately painted interiors which again communicate a very particular sensibility to the world. If fact if you squint your eyes up and let some of her portraits go out of focus and then do the same with her interiors, you could be forgiven for confusing one subject with the other. 

Gwen John: Interior 1926

Gwen John's paintings are reflections of her sensibility and sensitivity to her experiences, she regards her world through colour desaturation glasses; looking at her art materials, as well as the composition and subject matter for her paintings through the same unifying lens of a personal vision. 

Kaye Donachie is a very different painter and yet she also has a particular lens through which she regards the world. 



Kaye Donachie

Kaye Donachie’s paintings in her exhibition 'Song for the Last Act', celebrate the lives and works of women. Her portraits are not exacting copies of photographs of these women; drawing on literature, biography and in particular archival photographic imagery of women poets, writers, and artists. Donachie then reimagines these historic women in contemporary portraits, that use her very particular sensibility to oil paint. By using oil paint's ability to be manipulated whilst wet and controlling its flow qualities, she presents us with images that are invigorated by her handling and which therefore emerge into a painterly present tense. Her colour choice is often 'un-natural' therefore also emphasising the ghost-like quality of the images, her palette conjuring up colour visions of old black and white frozen photographic portraits of often forgotten women. 

As these portraits are not of 'real' women, they could be distorted in many ways and no one would be there to cry 'caricature!' They are coloured using a palette designed to remove these images from perceived reality and the paint handling means that it is hard to decide where for instance the mouth ends and begins. However I'm sure no one would argue that these were not portraits because they were too distorted, she obviously takes time to render her images as accurately as possible. Donachie has a clear project, her paintings illustrate a concept about the revisiting of women in history and she is celebrating these women using a personal palette; using colour to suggest the fact that she is rediscovering these women from old archives of photographs.

Donachie's images are sensuous paintings to look at and in their very painterly liveliness they bring back to life the women found within the archives she has been exploring. This is in direct contrast to Gwen John's paintings that I feel are powerfully controlled so that they achieve a timeless quality, the sitters individuality replaced by a way of painting that unifies everything and that aspires to timelessness. Reserved nuns occupying a similar compositional and desaturated colour sensibility to other seated women; quiet strength being the main feeling tone that Gwen John evokes; however I left her exhibition with a sense that in one way what Gwen John was doing, was building up her own personal archive of other women who she had managed to shape into various images of herself. 

References

E. H. Gombrich, (with Ernst Kris) The Principles of Caricature, British Journal of Medical Psychology, Vol. 17, 1938, pp.319-42 

Gleeson, James (1969) William Dobell London: Thames and Hudson 

See also:






Painting and photography A post on the work of Adam Stone who has worked from archival footage of photographs of buildings. In particular my thoughts on the concept of duende, a quality that the poet Lorca saw as essential to the recreation of a lived experience, could be applied to Donachie's paintings.





Saturday, 29 October 2022

Angles: Vision and Sound

The components of an angle

The angle, or the shape made at the conjunction of two straight lines gives us the letter 'V' and the letter 'L'. I have looked at angles before under their collective title of the zig-zag, and in their classic formal three line arrangement as the triangle but as they have such a vital role to play in our understanding of visual communication, I thought it time to look at them on their own. 'At a tight angle' suggests something hard to get at, a right angle is regarded as something upright and 'true'; not only have angles influenced he way we think visually, they have penetrated our verbal languages as well. An angle suggests an approach, as in 'what angle are you taking' and when we encounter a steep angle on road signs, whether it is going up or down, the approach is always from the left.



Angles can be harsh, the angle of naked pain was an angle that we used to ask Foundation students to find and draw when we were advocating working on that edge between logic and nonsense; 'l'angle mort', is the common French term for the "blind spot" that you often find involved in traffic accidents. 

However it is in pictorial composition that we find the angle most often used in visual languages. 

Charles Henry: Aesthetic Protractor 

Mathematical harmony based on certain divine proportions is an old concept but it was revived in the 19th century by Charles Henry, the French “psychobiophysicist”. In 1888 he published a colour circle and an associated aesthetic protractor, so that artists could decide on colour selection based on Henry's psychophysiological principles of dynamogeny and inhibition. A theory of rhythm was also added by Henry related to prime numbers or their reciprocal fractions. I have highlighted before in posts on 19th century approaches to visualising invisible forces, that these now 'discredited theories', had an impact on artists and how they went about their trade and in doing so they left a legacy of thinking that still lingers around the edges of Modernism. 

In visual perception, it has been argued that angles not only convey information, but that they also produce emotional responses. In the 17th century Charles Le Brun (1619-1690) developed an inventory of 24 emotions in his 'Expressions of the Passions of the Soul'.  

Charles Le Brun: Surprise and Admiration

He wanted trainee painters who belonged to the French royal academy to more precisely describe emotions and this inventory of expressions has been central to the reproduction of facial emotions in visual images ever since. However, in the late 19th century Georges Seurat who had received French royal academy training, had become aware of the emergence of physiological optics, and it became increasingly clear to him that perception was not a matter of a passive reception of a fixed image, but that there was a dynamic interrelationship between image and observer, that together contributed to the making of emotional perception. Charles Henry's physiological research into visual perception was focused on  concepts of dynamogeny and inhibition or as Edouard Brown-Sequard (1817-1894) described them, 'the transformation or displacement of forces'. (Crary, 2001, p.164). This very closely relates to my own interests in how artists might visualise invisible energies and although Brown-Sequard's and other's ideas of invisible forces and their actions during this time period are now discredited, I still feel that there is something about them that needs exploring, even if only the need to give balance with current scientific theories, which I am sure will be overturned in the next half of this century as new approaches are taken in our attempts to understand the invisible energies that surround us. 

Charles Henry proposed that angles could stimulate the perception of emotion. For instance by using angles of 24, 30, 45, 60 and 72 degrees he stated dynamogenic effects would be aroused that represented an energetic and joyful atmosphere. (Smith, 1997, p.143). If you for instance use a Dutch angle in photography, which is about 20 degrees, you take the image off its vertical, or right angled stability and in doing so create a certain emotional unease.  

Creating a sense of unease in a photograph

In this case 'Dutch' isn't Dutch, as the name is a corruption of Deutsch, (German) and it came about because of the consistent use of angled shots on German expressionist films. By avoiding the stability of the right angled vertical it feels as if reality has been knocked off its axis. 

Typical German expressionist film scene

Scene flipped horizontally 

German expressionist cinema employed several scene builders who had backgrounds in painting and they were very aware of how angle effected emotion. By simply flipping the image above, it is interesting to see how an angle rising from the left changes our feeling tone as we sense that we in this case have to climb the street off balance as apposed to an angle falling down the street from the left, whereby we feel in even greater danger, the danger of falling off the edge. 

Henry had looked at the earlier work of Humbert de Superville who had his own three line scheme for expression; 'expansive, horizontal, and convergent', a scheme that he believed affected the fundamental emotions. These were based on facial expressions, rising diagonals signifying a smile, convergent diagonals indicating sadness.



I personally thought de Superville's work very superficial in comparison with Le Brun's, but Seurat was impressed enough to try and use the angles proposed by Henry in several of his later paintings, such as the Circus.

Le Brun's facial expressions

De Superville had like Le Brun before him, based his ideas on studies of classical art, and he praised the Apollo Belvedere as embodying the mystery of art, studying the sculpture from several angles, and illustrating his lectures with drawings meant to demonstrate the ideal proportions of the statue.

Humbert de Superville: Sketch of the Apollo Belvedere

De Superville: The intrinsic meaning of shapes colours and lines

Kandinsky: Colour shape questionnaire

It is Kandinsky that brings this discussion into the 20th century. He had very clear views as to which colours fitted which angles. A sharp angle, such as one at 30% is he stated an aggressive angle, and best matches the colour yellow. A straight upright right angle of 90˚ is red. As the angle opens out it turns bluer as it has less aggression, and eventually at 180 degrees the right colour becomes a black horizontal line. He even put out colour/shape questionnaires to test his theory, if you got it right, the triangle would be yellow, the square red and the circle blue. Kandinsky, like Mondrian was interested in Theosophy a movement that was a complex belief system, which included a belief that colours affected spiritual harmony. 

Morgan Russell and Stanton MacDonald-Wright, founders of the Synchromism movement were also influenced by Theosophy, and they developed colour and keyboard relationship diagrams. They were also testing out other theories, such as those associated with the concept of 'elan vital' and as you can see in the title of the image below, colour composition and life forces were in 1914, as in Theosophy, entangled. 

Morgan Russell: 
'Synchromy in Orange: the creation of man conceived as a result of a natural generative force' 1914

As Russell developed his Eidos and his ideas for a light machine, Macdonald-Wright, then living in California, was equally preoccupied with developing a kinetic light machine and nurtured an accompanying interest in film and theatre. In 1927, he began working with a theatrical group in Santa Monica and called his work "Synchromist Theater." The Christian Science Monitor described how "the action of the play takes place in nowhere, at no time, therefore to have other than a purely abstract setting would not only be incongruous but ridiculous. The mood induced by the use of these Synchromist settings is definite, and together with the use of Wright's color organ, which can throw any or all of the colors of the spectrum...evoke an illusion and atmosphere of a fresh sort." He successfully built what he called a Synchrome Kaleidoscope in 1959 and this as an idea would help others in the search for a synthesis between the emotional languages of sound and light. 

Wolfgang Köhler had in 1929 developed what he called the bouba/kiki effect and this pointed to another way of developing sound symbolism within visual languages.

Kiki and Bouba shapes

It was discovered that right across the world if you showed these two shapes above to people and asked them which is Kiki and which is Bouba, virtually everybody pointed to the sharp angled image as 'Kiki' and the rounded curved image as 'Bouba'. By the time I entered into this sound / image conundrum it was late in the day, but still there was a lot to learn. In the mid to late 1970s I was teaching on the foundation course in art and design at Leeds. It was a course that had its roots in Bauhaus thinking and exercises using sound and vision were often used with students to make them aware of the possibilities of sound and vision interaction. One thing as staff we would do was to 'sound' a drawing. I used to enjoy the intuitive responses to shape and colour this would engender. Making vocalised sounds and varying their pitch as students presented various shapes and surfaces for critique. The structure of Steve Reich's 'Come Out' often being used as an example of how overlaying and serial repetition could give rise to visual as well as aural compositions. 

However by the 1970s a new way of thinking about these issues was coming in to play. Because of the introduction of computers, programmers began getting interested in algorithms, and they began to see that these could be used to predict things or give shape to possibilities. One of the earliest attempts at this was Stiny and Gips' 'shape grammar'. 

From 'shape grammar' by Stiny and Gips (1978)

Stiny and Gips proposed in 1971 that by using their algorithms sets of non representational, geometric art could be generated. Aesthetics as a concept was boiled down and considered in terms of 'specificational simplicity and visual complexity'. But things have changed, and aesthetics linked to algorithmic possibilities has developed far beyond the simple idea of this being a way to create variations of abstract images. By using 'like' as an information gathering tool, this social media practice has allowed companies to see in almost real time how virtually anything, can be used to influence audiences' emotions, not just shape and colour. Kandinsky's colour shape questionnaire looks so naive in comparison. 


Grasses drawn by a computer using L-systems

At the same time that Stiny and Gips were developing algorithms that could generate geometric forms, Aristid Lindenmayer was devising 'L'-systems to describe the behaviour of plant cells and to model the growth processes of plant development. The angle was now part of a decision making process that was becoming more and more embedded into computer software development, one that was eventually to lead to uses within the growing field of artificial intelligence. As computers were able to crunch bigger and bigger number sets, more and more complex behaviours could be modelled, perhaps the most significant of these early models being 'Boids' an artificial life program, developed by Craig Reynolds in 1986, which simulated the flocking behaviour of birds.

Boids

Boids as a way of visualising complex flows of information would eventually be used to visualise how a virus might spread. You can probably see where all this is going, but one development in particular went right back to that association Kandinsky and other artists had made between colour, shape and sound and this development is something you see as soon as you open a sound recording or editing program on your computer, it will show the peaks and troughs of the sound you are dealing with, using oscilloscope and similar sound visualiser systems. 

Oskar Fischinger with a “sound scroll” used in the optical soundtrack of one of his films.

Oskar Fischinger was an early pioneer of abstract film and for his soundtracks he developed visual scrolls, the language of which as you can see from the example above was developed from the idea of repeated angles or triangular divisions of time and space. While Fischinger drew each individual waveform by hand, the animator Norman McLaren, developed template based methods to do the same thing. Eventually coming up with the idea of index cards, each painted with a pattern whose spacings produced notes in the chromatic scale. The idea of an optical soundtrack seems quite complex but it could at the time be very easily constructed.
The sound artist would take a strip of black film, or film with a black soundtrack area, two or three feet long. Make a single scratch on it, join the film into a loop and run it on a moviola or projector. The scratch would give a click with a certain quality. The splice in the loop would also make a click. but you need to discount that. The next step was to take the loop off the projector and make another type of scratch with a different size and shape, at least 4, 6 or 8 frames away from the first scratch. Then you would run the loop again, and you would hear two clicks, each with a different quality; one may have a hard sound and the other a soft sound, or one may be loud and the other quiet, by adding various types of scratches one by one, and by identifying them, the artist would learn what kind of scratch gives what kind of sound. By doing this a direct link between sound and vision was made. The story of these events is told in detail in an article by Golan Levin 'Painterly Interfaces for Audiovisual Performance', and what interested me was when computer real time visualisations became possible, such as in Pete Rice’s work, how similar the forms generated were to the ones visualised as 'thought forms' by Besant and Leadbeater back in the early 20th century.

Stretchable music

A screen capture from Pete Rice’s Stretchable Music system in use



Charles Webster Leadbeater & Annie Besant 'Thought Forms 1905

By moving backwards and forwards in time what we can do is sense the physical reality that underpins the seeming 'magic' of current computer realisations. For myself this makes things more graspable and I can see these things as material languages again. As a student of fine art, you may feel that computer aided art is not for you, but the fact is that everything is grounded in the interconnectedness of everything else and that by unravelling the threads that tie it all together, an artist can begin to both play with and open out the situation for new and hopefully fascinating re-interpretations and avenues for material thinking. 

Recording of the sounds in Hyde Park London

During the Covid pandemic MIT’s Senseable City Lab looked at park environments in five cities – San Francisco, New York, London, Milan, and Singapore. It was an attempt to see how collective sound environments were changing due to the fact that people were altering their patterns of behaviour. The project simply accepts that there is a direct relationship between sound and vision, but I think it misses an artist's involvement, one that could have been used to create a more physical response to what was happening. 

Lines: Anders Lind 

'Lines' by Anders Lind, reminds us of how musical notations rely on the use of visual lines to make sense of them, thus closing the loop between lines as abstract forms and sound as an abstract language. The fact that you have to physically engage with the piece means that you begin to know this in your body and not just in your mind. 

Signs of a sound emerging from our mouths also often use angles, the sign below is a stock image for a shout, we imagine the angles emerging from the mouth as an indicator of the loudness of the shout.



References:

Crary, J. (2001). Suspension of Perception : Attention, Spectacle, and Modern Culture. Cambridge, Mass.: MIT Press


Moritz, W. (1986) Abstract Film and Colour Music, in: The Spiritual in Art, Abstract Painting 1890 – 1985 New York: Abbeville Press, 

Smith, P. (1997). Seurat and the Avant-garde. New Haven: Yale University Press.

Stiny, G., & Gips, J. (1971). Shape Grammars and the Generative Specification of Painting and Sculpture. IFIP Congress.

See also:

The zig-zag

The triangle

Eye music

Drawing sound

Steve Reich Rhythm and colour constructions