Tuesday, 30 June 2020

FAT is dead, long live FAT

Back in 2015 FAT (Free Art and Technology) died, in effect it put itself to sleep. What had begun as a space for hackers and programmers to interject and critique the online world had become a hyper aware arena that was beginning to crumble from the inside as everything around it was becoming either so corporate that its power was beyond cynicism, or the tropes and memes that FAT was playing with were becoming themselves appropriated to such an extent that in some ways their job was done. 

The FAT site was archived and now when I look back at it I find it very like a museum and a very good one. FAT ideas and projects still seem fresh and alive, just in the same way that a Van Gogh drawing doesn't seem to date. Jimmy Cauty's conversion of police riot shields into smily faces, seems as relevant now as it did then. 

Jimmy Cauty: Riot Shield

From fast art ideas to mini events designed to create a moment of serendipity, such as vacations in Paris, the FAT archive is a store house of irreverent and yet thought provoking ideas. If you are wondering why I am promoting this on a blog about drawing, I would argue that FAT is another example of contemporary 'disegno', I am thinking about drawing as a conceptual base around which other forms of art can emerge. Before painting, sculpture or architecture we had 'disegno', because those more solidified discipline specific concepts had to have a platform out of which to emerge. I can still remember when I was first introduced to the computer as a tool that could deal with drawing, I undertook a six week course on programming in BBC basic, after which I could determine points on a screen by indicating geometrical coordinates and then once I had written the correct the code, could join them. Every letter I am typing, has an associated code, that determines its shape and then another mass of code is used to determine where it goes, a simple word processor is using a deeply embedded set of codes for each typeface as well as code needed to control, spacing, kerning, indents, etc. etc. All of which is at its root linked to a type of drawing based on geometric principles, but we forget what lies beneath and FAT didn't. *

*That comment about forgetting what lies beneath gives me another link or entanglement back to my interest in object orientated ontology. Graham Harman wrote about computer languages in response to his own question about the relationship between OOO as a philosophical system and object orientated computer languages. He pointed out the fact that older computer languages were always holistic entities, but what programmers now do is put together independent programming 'objects' that are made to interact with other objects, whilst the older internal information belonging to each holistic 'object' remains discreet or hidden. In effect the reality that lies behind what programmers do becomes more and more hidden or locked deep down inside the structure. These 'objects' are also in Harman's terms 'opaque' to each other as well as to the end user. He then goes on to suggest that unlike other philosophical systems, OOO adherents believe that in a similar way to object orientated programming all objects are mutually autonomous and enter into relation with other things only via a mediator. (See pages 11 and 12 of Graham Harman's 2018 Object Orientated Ontology: A new theory of everything Pelican edition) 

See also:

Thursday, 25 June 2020

The Trinity Buoy Wharf Drawing Prize 2020

I have posted before about the importance of what was formally the Jerwood and which is now the Trinity Buoy Wharf drawing prize. 2020 is the 25th consecutive annual holding of this prize and accompanying exhibition and to celebrate this, the call for entries is going international, and is now open for artists from the UK and around the world. There is another new addition: the Working Drawing Award, which focuses on the role of working drawings in art, architecture, design, engineering, manufacturing, science and if you are a regular reader of this blog you will be very aware of how I have attempted to open out the conceptual framework around which we can think about why drawing matters to include the work done in other disciplines. In particular it would be interesting to look at how even more disciplines could be included, such as police crime scene drawings or those done for archeological digs or circuit diagrams. 

The Trinity Buoy Wharf drawing prize has been led by its Director, Professor Anita Taylor, since its inception, and she is passionate about drawing in all its guises. She has helped raise the profile of what has sometimes been a forgotten art form, and she continues to work hard to communicate and support an understanding of drawing as a medium that is central to the way that humans communicate with each other. 

The entry date has been extended and it is now going to be the 9th of July. All the details for entry are on the Trinity Buoy Wharf Drawing prize website  and there are regional as well as London drop off and collection points. 

See also:

Wednesday, 24 June 2020

The zig-zag

Entoptic phenomena in relation to art forms

Entoptic images are caused by phenomena within the observer's own visual perceptual system. If you look at the drawings to the left of the image above, zig-zags and wave patterns have clearly been picked out as being one of the basic six sets of entoptic phenomena. 

Zig-zag patterns alongside dots and other abstract shapes, can be 'seen' because of the fact that eyes are embedded into your body. We sometimes forget how much a part of the body the eye is, it cant be separated out from the way blood circulates or the nervous system works, but we tend to think of our eyes as being a sort of window through which we perceive the world, rather than a working organ that like all our other organs is integrated into the rest of the body. The eyes operate in such a way that they are also looking at our inner selves at the same time, but we tend to forget that.  One of the reasons that we see entoptic images is because of the movement of white blood cells in capillaries in front of the retina. Another reason is that floating coagulations of vitreous jelly can drift through the interior of the eyes and you sometimes see them or if not these 'floaters' themselves, traces of their movement, as your eye attempts to focus on them. The zig-zag you see being your brain's attempt to decode your own eye's movements as it tries to 'see' something. One zig-zag form in particular is called the “Purkinje Tree”, this is when you see your eye's blood vessels when light shines into the pupil from an unexpected angle. Another way to get the brain to 'see' a zig-zag is to apply pressure to your closed eyes and this generates a phosphene, perceived as veiny or zig-zag-like lines. These phenomena only come into vision when you are faced with a plain background such as a white wall or a clear sky. Normally there are too many interesting things going on for the brain to be bothered with these insignificant visual events. “Prisoner’s Cinema”, an effect that occurs when you are in darkness for a long time, is another example. A “light show” in the mind gradually emerges out of the blackness, and often begins with abstract zig-zags and dots. There is a further stage to this effect, as time goes on and there is no relief from the dark, the mind begins to 'see' other more recognisable things and abstract amorphous forms gradually take on animal or human like shapes. This effect it has been theorised, is why cave painters produced both abstract forms and representational ones and most importantly mixed them together. 

Pech Merle Horse and spots

The Pech Merle cave in France has drawings on it's walls from 25,000 years ago, (the spots are made by a controlled 'spitting' of the pigment mixed with saliva and that is how the hand stencils are made too). Imagine 'seeing' a cloud of dots, and then gradually a horse form emerges from them and then as the horse form takes shape, you are also breathing out spots of paint. Perhaps drawing a line with one hand, whilst spitting paint over the other hand that you are using to steady yourself against the cave wall. One image grows from another, and then as the horse emerges, you begin to 'decorate' its edge by spitting dots to surround it. In many ways our minds are still in that primeval darkness, every time we close our eyes we slip back into our old skull cave. 

A 430,000 year old zig-zag 

There is an engraving that was excavated from a riverbank in Indonesia that is so old that it couldn't have been made by the humans we call Homo Sapiens. It is thought to have been drawn by an individual that belonged to the species we now call Homo Erectus. That these creatures were human like was at one point doubted and the fact that they have drawn designs on objects such as shells, in many ways changes our view of what we are and how we relate to everything else. The divide between ourselves and other species is perhaps not as sharp as we thought, and we are not alone in feeling the need to scratch a sign into the world. The fact that the oldest drawing we know of is a zig-zag, also suggests that we ought to pay a little more attention to zig-zags than we usually do. 

The video below describes the history and science behind entoptic phenomena, as well as how to produce them yourself. 

Entoptic phenomena

The zig-zag is also something that is seen in our daily lives in a variety of other forms, forms that have been used by artists as powerful metaphors. Artists and cultures have over time responded to various different zig-zag associations in order to create their metaphors, some like branching trees, others like waves or folds, some like cracks and some like wrinkles. 
Doris Salcedo: 'Shibboleth' 

Zig-zags can be like cracks. The crack in the floor of Tate modern by Doris Salcedo, entitled 'Shibboleth' refers to the Biblical tribe the Ephraimites, who when attempting to flee their persecutors across the river Jordan were captured by their enemies, the Gileadites. In order to check they were of the Ephraimites tribe, every person was asked to pronounce the word, 'shibboleth'. Their language did not include a 'sh' sound, so if they couldn't say the word, they were executed. A shibboleth is any custom or tradition, usually a choice of phrasing or even a single word, that distinguishes one group of people from another. It is one of those differences that gives people the power to judge, to reject others or to kill them. 

For Salcedo, the crack represents a history of racism, running parallel to the history of modernity; of the divide between the rich and poor, northern and southern hemispheres. She invites us to look down into this crack, and to confront discomforting truths about our world, truths that are at the moment coming home to roost. 

A crack in the basement floor

Cracks, suggest that something is breaking or has broken, we worry about what might be behind the crack. Cracks in the fabric of space and time have been used both to drive science fiction concepts and by scientists to describe stress lines in the fabric of space. Cracks in the fabric of spacetime, are supposedly many light years long, are extremely thin, have enormous amount of energy and form billions of small loops along their lengths. Sometimes the cracks meet, cross and intertwine and in doing so give the writers of Doctor Who ideas. 

Hypothetical model of space-time cracks

A crack can also be repaired, in Japan broken pottery can undertake a 'golden repair' or Kintsugi technique, the zig-zag line of the crack of repair being fixed with lacquer mixed with powdered gold. In Japanese culture the aesthetic philosophy of wabi-sabi, helps people to cultivate an appreciation of flaws. This is something becoming more and more important to think about in a throwaway society like ours, we could learn a lot from an idea that celebrates repairs and gives honorific value to the lived experience of objects. The third Saturday in October is International Repair Day, perhaps there are other ways to visually celebrate an engagement with something that we all need to think about if we are not to continue using up the world's resources. 

Korean tea bowl with golden repair

Several contemporary artists have responded to the Kintsugi idea; Yeesookyun, Tomomi Kamoshita, and Charlotte Bailey are all interesting to look at, however I'm particularly interested in the work of NeSpoon, who's street repair interventions I find open out this type of thinking into the wider public sphere. Her decorative surfaces can be like graffiti found on the sides of buildings and she has pioneered the idea of street ceramics as a way of 'healing' broken urban spaces. 


The zig-zag crack is very like the crack of lightening. Walter de Maria's 'Lightening Field' being probably the most well known artwork to use lightening as its central idea. 

Walter de Maria 'Lightening Field' 1977

Sited in New Mexico, the work consists of 400 stainless steel poles arranged in a 1 mile x 1 kilometre rectangle, and is still operational. The poles range in height from 15 to 26 feet 9 inches because the terrain is uneven and their tips have to be level. De Maria had a long history of working on a huge scale, making a mile long line drawing in the desert back in 1968, a practice that he informed us was based on his interest in the Nazca lines. When asked about his work de Maria stated, 'the Invisible is real', and for me this is where the zig-zag enters the arena of the sublime. Both Burke and Kant refer to lightening as being sublime, and the Abstract Sublime a term used by Robert Rosenblum to describe the feelings suggested by certain paintings, such as those by Barnet Newman, Clifford Still or Rothko, tended to see these paintings as images representing cosmic forces, forces that were sometimes 'cracks in the fabric of space and time'.

Clifford Still

There are of course much more prosaic uses of the zig-zag. The zig-zag suggests movement and when we dance we are often lost in the moment of our moving, it being no accident that many cultures have associated the active visual dynamic of this form with dance. 

Butlins dance floor

From the zig-zag of a Butlins dance floor to the zig-zag of a decorated Australian didjeridu seems a long way, but on reflection perhaps they are not that far apart. 

Anangu dancer

The multi-modal art forms of Australian aboriginal peoples link the zig-zag of dance, with hand gestures, body decoration, painted objects as well as speech patterns. The 1920s Western European concept of Art Deco does the same. 

Flapper zig-zag look

Lobby of Claridges Hotel London c. 1935

Try and picture in your mind jazz playing, the dancers trying to zig-zag their bodies in an accompanying visual rhythm, in a room with a zig-zag floor and with zig-zag art Deco decoration and people's speech patterns being inflected by cool jazz type syncopation. This is art as total culture, not unlike what happens elsewhere and else-when throughout history. In aesthetics the term 'gesamtkunstwerk' is used to refer to a total work of art, something that encompasses a range of approaches rather than just one media.
We are embedded into our world and the concept of an art form that is separate from the culture we inhabit is a very strange one. Kant's aesthetic idea of 'disinterested interest' is an idea that allows us to stand back from the art we make in order to contemplate it, and it stands at the centre of the history of European aesthetic discourse. But there are other aesthetic traditions where art forms engage with the world as part of life. 

In the book 'Everywhen: The Eternal Present in Indigenous Art from Australia', by Skerritt, Perkins, Myers, and Khandekar, we are introduced to the idea of the 'ever-present', time being something that is bent into a loop, old things being interwoven with new ones in an endless cycle. This 'dance' being the one we occasionally find ourselves lost in, in those moments when perhaps lost in the rhythm of an enfolding piece of music or when looking carefully at something as we draw it, we inhabit the moment rather than regard at it as something for disinterested contemplation. In times such as the one we inhabit now, we will need to make decisions as to what art is for and we need to understand how art is changing in the way it operates within our society. Art has never been fixed as to how and why it is used, each culture and time period uses it and understands it in a different way and that is why an awareness of other times and cultures is vital, as it allows us to see that there are other possibilities and that no one way to engage with art is right. 

See also:

Drawing and perception

Saturday, 20 June 2020

Photography as an extension of drawing

Application for a patent for a camera

I had several conversations with first year students this year about the relationship between drawing and photography. Because of the apparent ubiquity of the photographic image, there is a strong belief that drawing is not as relevant as it was. I still beg to differ. 
Before a camera can be made, all its parts will have been drawn, both for the patent office and as technical drawings that can be converted into information for machines to follow so that the individual components can be made. 

When you need to explain how a camera works, diagrams are nearly always used to show how the various components work. 

A pentaprism is often used in SLRs to get light to the eye without inverting the image. An understanding of how this works requires a basic grasp of geometry and geometry is a type of drawing. 

If your camera uses mirrors again it would be geometric drawing that clarified how the angle of incidence equals the angle of reflection. 

When a camera is put together in a factory an assembly line diagrammatic drawing is made to ensure that everything is put together in the right order. 

When you need to know how all the individual components fit together a drawing is needed. 

The importance of the artist who visualised how all the elements fit together is reflected in the fact that this drawing is signed. 

The drawing above was made in 1976, not long after I had been working in a draftsman's office myself and my job was to produce drawings like this to enable people to understand how things worked. As in all forms of work that involve skill, some people are better than others at what they do, in this case J. la Guernie was one of those people who had a talent for this sort of drawing at that time in France. 
It doesn't matter what sort of camera or aspect of a camera, such as a bellows extension or lens diaphragm, in order for the idea to be realised it would need to have been visualised and therefore drawn. It is usually at the patent office that all the first drawings are kept.  

Patent drawing for the box camera

Patent drawing for the polaroid camera

Patent drawing for a movie camera

This tradition continues and although the drawings are now done using CAD, drawings are still primary information, you can't feed a photographic image of an idea into a machine and get it to be made. All 3D printers need three co-ordinate wire frame constructions to work from, even if the initial information comes from a camera taking multi-viewpoints, these many viewpoints are stitched together into a framework that is geometrically determined and coordinates plotted on a graph. 

The complexity of coordinates in what is actually a very simple drawing of a staircase is resolved if you work your way through them one at a time. You can see how easily the number of reference points builds up, and then if you wanted to twist the drawing through let's say just 5 degrees and then a further 5 degrees, the density of drawn coordinates would quickly become far too difficult to navigate, which is why at a certain level of complexity its best to let a computer handle the maths. However even at its most complex a computer generated drawing is still basically a drawing. A drawing constructed by plotting points along x, y and z axes and then joining them using lines. 

Using x, y and z axes to determine where a point in space is.

Over the years technical drawings have reflected changes in drawing technology, the advent of the computer being just the latest, but on the street and in places where decisions have to be made without new technology, drawing with a pencil or biro by hand is still vital to an understanding of what is happening. 

People thinking about how to link video cameras up

In the two cases above, it is clear that as technology develops, there is still a need for drawing, if only to clarify how to connect the various elements of a complicated wiring system.

In both still and moving image photography it is not just the technical equipment that relies on drawing. Composition and selection for film is often storyboarded and printed storyboard templates for this are still available.  
Film storyboard template

Storyboard for Psycho shower scene 

Saul Bass draw the storyboard for Alfred Hitchcock's famous 'shower scene'. His visualisation would have helped Hitchcock think through what the scene needed to look like. The drawings being there long before the camera was used. 

Ridley Scott did all his own storyboards for Alien and of course decided to use HR Giger's drawings for the alien creature. 

H R Giger: Sketch for alien 

We can sometimes forget that every fantasy movie begins with drawing, the fact that comic book franchises have become so important, further reflecting the role of drawing as central to realising ideas. You cant take a photograph of an image in your head. 

Long before cameras were invented artists realised that the proportions or aspect ratio of a rectangle could be used to help control the way people looked. Most DSLR camera sensors have a 3:2 aspect ratio which is the same as 35 mm film. This aspect ratio was decided when Leica made the first 35mm film cameras. The 3:2 aspect ratio is double an old movie film frame; because Oscar Barnack, the Leica inventor, rotated the film for cinema cameras through 90 degrees and in doing so doubled the width of the frame to create the ‘full frame’ 35mm still camera format. In the early days of the cinema because of a lack of standardisation, films were made using a variety of different film widths and projection speeds, but eventually the 35-mm wide Edison film stock became the standard format. This format would have been decided upon initially as a drawing, using a ruler and set square, because when it comes to deciding on the ratio of a rectangle we use geometry, and because manufactured things have to be first of all understood as technical drawings.

35mm film used in a movie camera

35mm film used in a still camera

Ratio is a geometric idea and geometry is a drawn construction. So when you are trying to compose an image or make a selection using your 35mm SLR, it is worth thinking about all those artists in the years before the invention of the camera, who also used a frame to help them compose. 

Van Gogh, although living at a time when the camera was coming into everyday use, still used an old fashioned perspective frame. Do read his letters if you ever have the chance to, because they are wonderful examples of how an artist thinks. He refers to the use of a perspective frame several times in them. In one of his letters to his brother Theo, he complained about trying to work in the dunes, where the ground is very uneven, "This is why I’m having a new and, I hope, better perspective frame made, which will stand firmly on two legs in uneven ground like the dunes". Then in another letter he complained about expenses, "I had more expenses in connection with the study of perspective and proportion for an instrument described in a work by Albrecht Dürer and used by the Dutchmen of old. It makes it possible to compare the proportions of objects close at hand with those on a plane further away, in cases where construction according to the rules of perspective isn’t feasible. Which, if you do it by eye, will always come out wrong, unless you’re very experienced and skilled. I didn’t manage to make the thing the first time around, but I succeeded in the end after trying for a long time with the aid of the carpenter and the smith. And I think that with more work I can get much better results still." Later once he has got to grips with how to use the frame he states, "Long and continuous practice with it enables one to draw quick as lightning - and, once the drawing is done firmly, to paint quick as lightning too.."  From: The Letters of Vincent Van Gogh, Penguin Classics 1997.  

Van Gogh's perspective frame and above that a drawing he did of him using it

A film on Van Gogh's use of perspective frames

The gridded frame for Van Gogh was something he took from Dürer, an idea that stretches back through hundreds of years of painting and drawing. 
There is a certain back and forth with artists moving between the idea of using camera obscuras and tracing frames but the fact that images had to be made by hand and often transferred from one surface to another and that they often also had to be re-sized, meant that a squared up image became central to many artists' thinking when it came to composition. The squaring up also allowed artists to link the internal divisions within an image with the external dimensions of the frame.  

Study for a portrait of the Artist's wife, Robert Polhill Bevan 1915

When you look through the view finder of a camera, you are using a selection frame, one that you can if you want to, add a grid to. This is again something that as an idea began with drawing. 

Using the grid on a phone camera

When you enable the grid on your mobile phone you are using a Renaissance idea that has been used in drawing studios over and over again for hundreds of years.

The Renaissance grid method

The thing about drawing is that it can be used in so many ways. From technical drawings for engineering, via geometry and measured drawing to rough diagrams on scraps of paper and expressive sketches explaining how someone could frame up a scene in a horror movie. It is this versatility that explains why it is still around and why it is still something to consider when using a camera, after all photography is in its own words, 'drawing with light'.

See also:

Mathematics and rightness (Includes information on the Golden Section) 
Drawing and film
The grid as a cage
The weaving of grids

Friday, 12 June 2020

The frame and the screen

I have posted before on the frame and the screen but have yet to bring together the various ideas associated with them both.
Framing is both a physical and an intellectual idea. The physical act of putting a frame around something operates as if it cuts whatever is framed away from the rest of the world. But it is not quite as simple as that.
Each type of physical framing changes the way we think about what is framed. 

How to window mount

The window mount

The window mount is probably the most common form of presentation and you will see images framed in this way in most art shops. It has various concepts behind it. One is the window of course, a window we look through in order to view the world like a picture, 'the picturesque'. It relates to the 'view finder' in a camera, a frame of selection, whereby the artist presents the best view, or a special selected image for contemplation. But there are other issues at work. Framing is often about degrees of separation, the cutting away or separating of the artwork out from everything else. If we look at the image of the landscape above it is framed by a wooden edge, as well as by a window mount. The physical edge of the wooden frame, creates a certain 'objectness' for the image, it brings it into the realm of 'other furniture', so that although the artwork is not necessarily 'furniture' it can be pulled into an association that allows it to be included within a 'domestic' frame of reference. For instance I'm sure you will have been in a hotel room with a framed abstract image installed within it that is obviously chosen to chime with the curtain design rather than to make you think deeply about existence. This doesn't always work of course, I often find myself thinking a lot about my existence after looking at the framed images on my hotel walls but I doubt these images are designed with me in mind. 

Image of an Edward Hopper painting taken from a framing service website

This is why when you go to a framing service website they will often have images showing you how the artwork will look framed alongside other items of furniture. However, sometimes within the frame, we have a second frame, or a third or even a fourth. 

This frame within a frame, is a device to make it appear as if you are entering into the image. It is meant to make you feel like you are walking down a corridor towards a window that is directly in front of you. You are in effect walking into the image and in doing so you are stepping from one world into another, rather like stepping into the mirror in Alice through the looking glass. This is a very powerful effect, but one that takes you away from reality, so only use this framing method if that is the effect you want.  

A frame within a frame

The card cutout we often call a window mount is called in the framing business a passe-partout, and is put between the picture and actual frame, which is the physical surround that holds everything together.

A collection of passe-partout window mounts

'French matting' uses gold lines to separate the image even further, to enhance the 'aura' of the image. The frame in many ways 'celebrates' the image as a possession. In cutting it out from the world, it can also be a way of making it easier to possess. If you look back at the framed image of a bird in a landscape above, you will also notice that the framing cuts into the edges of the image, (in this case the printing process has cut the edges away to leave a white border, but it is often the window mount that hides the edges of images) a small amount of the image will therefore never be seen and this is another issue artists have to think about when deciding whether or not to have work framed and window mounted. Compare the French matted landscape above with the image below by the Connor Brothers. 

Image 'floated' within a window mount.

The Connor Brothers' presentation fits two ideas together. On the one hand they are trying to reassert the 'objectness' of the image by showing us the edges of the paper by 'floating' the image within a window mount, which itself operates to separate the image out from the world and enhance its 'aura'. The wooden frame, operates to establish both an overall physicality or 'furniture', as well as establishing a degree of separateness from the world, which is furthered by the large expanse of white card between the image and the surrounding wooden frame. We are again looking at frames within frames and the corridor effect is still in play. 

Art Museum framing of a Vermeer

It's interesting to look at how as an object's worth increases, both as financial and cultural capital, its framing becomes more elaborate. In this case it is a complex double framing because the older frame that sits around the painting, is itself now an object of cultural capital, the black surrounding frame having the simplicity of a more 'Modernist' aesthetic. We still in effect walk down that invisible corridor into the image, but it is a corridor decorated in both modern and classical styles. 

Allan McCollum: Forty Plaster Surrogates

Back in the late 1970s Allan McCollum began making his plaster surrogates. What was interesting to me was the fact that they made you very aware of the framing process of art and of how this process was making new objects that were themselves part of the art idea. Take away the artist's work from inside the frame and you still have a powerful signifier, which is the frame itself. In McCollum's case his casts also make you very aware of the objectness of the frame. The paradox being that what began as surrogates were now very definitely art objects in their own right. 

Artists, myself included, sometimes want to reinforce the 'objectness' of their drawing or painting by floating the work within a larger box frame, in order to avoid the entrapment or suggestion of a window on the world which comes with a window mount. But in floating an image within a frame, there is an in-built contradiction. Looking back at Allan McCollum's Plaster Surrogates, you can see that the frame never goes away, it still operates as a separator between to put it very crudely, 'art and life'. 

So you may say, "Why not simply hang the painting or drawing directly on the wall?" The issue here is though that walls in the art world are also frames. 

If you look at the 'salon hang' from the late 19th century above, the first thing you become aware of is that there is no wall space. The heavy frames do all the work separating each image one from each other. The owner has bought the lot, and is showing off his or her purchase power. 

In the spotlit image above it is the lighting that frames the work, in fact it is the lighting that reinforces the idea of the 'aura'  of the work of art. The gallery has literally given each art-piece its own individual auratic glow. Turn the lights on and we are left with a white cube space.

The white cube, as written about by O’Doherty is not as a gallery space a neutral container. It is a historical construct that operates to 'frame' the artwork within it in a particular way.  It is an aesthetic object itself, with a particular aesthetic history, which we can trace back to sanitarium design and the need for hygiene. O’Doherty would go on to argue that this aesthetic actually overpowers the work that resides within it. Those white walls are not context free, the apparent neutrality of the gallery walls is just that, apparent, and the gallery's attempts to attain an aura of timelessness by either spotlighting the art or giving it lots of surrounding white space is a sort of salesperson's trick or slight of hand. 

O’Doherty looks at the gallery as a sacred space that is like an archeological tomb, undisturbed by time and containing cultural riches. The gallery is thus constructed to give the artworks lasting value; it is a space designed to immortalise the cultural values of our elite i. e. very rich people. Reminding us that galleries are also shops and like most shops they are designed to get you forget the worries of the outside world, the white cube establishing a crucial distance between that which is to be kept outside (the social and the political) and that which is inside (the everlasting value of art). In this case the artwork is framed to ensure that it is separate from the rest of the world, so that the buyer can clearly purchase its aura. In the case of this sort of art the buyer is also in buying into and establishing a certain set of credentials. 

Of course once you understand the game you can enter into it. You can play with the conventions and subvert them to your own ends. Framing and presenting can be a political decision as well as an aesthetic one. But it is important to remember that as the number of paintings are removed from the old collection that we see in the 19th century stereoscopic image above, more and more wall space is revealed, and this wall space operates as a frame. It might not be the golden, florid, decorative thing of the late 19th century, but it is just as much a frame, but this time it is masquerading as a neutral space. 

As you can see from this argument, the frame operates in the real world. It operates as a way of giving the observer a cut off or separate space within which to contemplate the artwork, but this operation is a complex one. The space of the frame oscillating between the 'real' world and an ideal space, a space within which art's spiritual or aesthetic values may be cultivated. As the world outside meets the world inside there is a complicated arena within which a certain duality comes into play. Deridda is fascinated by this and he calls it the parergon. 

It was Kant that linked the idea of the parergon to that of a painting's frame. On the one hand the frame is seen as an addition or ornament. It embellishes the artwork, but suggests Kant, it doesn't add anything to it. On the other hand Kant sees the frame as an example of something that is neither one thing or another, it is something detached or separate; detached not only from the thing it enframes but also from its surroundings, (the wall where a painting is hung). According to Kant, the parergon is like the gold leafed frame for a painting, a mere attachment added to gain superficial charm or grace, and which could in reality detract from the genuine beauty of the art. The frame belongs neither to the artwork nor does it operate as an article of useful furniture, you can't sit on it, eat from it or keep yourself warm, but in sitting between the two it operates as a permeable boundary, a space between the domain of the artwork and the environment of the room the work is hanging in. 

These 'liminal' or 'threshold' moments are essential to an understanding of animism and other forms of web-of-life spirituality that encompass both human and non human understandings of the world and I personally have thought of them as being similar to what Lewis-Davis in his book 'The Mind in the Cave' calls the membrane, or how neolithic peoples perhaps regarded cave walls that were used to support paintings. On the one hand they were just that, walls of caves that you could paint or draw on, but they also served as permeable barriers that allowed people to imagine a space beyond the depictions, one that included the world of spirits and a space in which negotiations between different animals and environments could be performed, often with a spirit guide or shaman. 

My understanding of Derrida in relation to this comes via Hegel. Hegel wrote extensively about the master/slave dialectic. The idea being that in a relationship where there is an imbalance of power what can happen is that gradually the one begins to depend on the other and as this dependency deepens the power begins to move from one to the other. I'm not sure how much this happens in reality, but it is a very interesting idea. 

Derrida is interested in dualities, he is always looking for openings that allow him to point out that what you think is happening is one thing but in reality it is something else. So in the case of the frame he suggests that what was thought of as an addition is in fact more important than the actual artwork. In doing this he spends a lot of time discussing the position of the frame as both in and out of the world and this is where I would like to bring the screen into the discussion. Like the frame, the screen also sits between one thing and another. It frames the images that appear within it and operates as a physical object in the 'real' world, you can hold it and touch it and treat it like a piece of furniture, but you can also forget it exists and fall into the world it contains. 

The selective frame and the circular lens

The formal issue that has always intrigued me is the relationship between the circle and the rectangle that happens between the lens and the sensor. The cartoon above epitomises the problem, the lens has a circular focus that cuts away the rest of the world, thus whatever that part of the world is linked to and shaped by is cut away, but then this cutaway is re-framed within a rectangle. 

The Edinburgh camera obscura

If you go to see a working camera obscura, such as the one in Edinburgh, the first thing you become aware of is the nature of the curved image. This is what it is like inside the camera, but imagine a frame that is then placed over this circular image, and this is where the film or other photosensitive area is positioned. 

The reframing is for a technical reason, the quality of the circular image worsens as we move farther from the lens' point of convergence. As we move towards the edge of the circle, images are dimmer, blurred and smudged. This is due to the lens, not the sensor. The lens converges light towards its centre, which means as you approach its edges you get less and more diffused light, hence the image edge's fuzzy-ness.  This can be compensated for by the camera's sensor. A camera sensor compensates for a circular lens that distorts towards its edges in various ways, and because of a range of distortions, including aspherical elements, chromatic aberration, coma, low dispersion, and a high refractive index, has a lot of work to do. Sometimes it is worth looking at technical issues just to highlight how much the nature of a specific medium is shaping communication, so in this case because we are looking at how framing in photography effects meaning and the fact that framing is also a way of minimising but not eliminating lens distortion, I'm going to try and non-scientifically pass on some information. 
At the centre of the problem with a lens in relation to focus is field curvature. Curvature of field, is a natural aberration of all lenses, due to their curved structure and how light moves through them and onto a flat plane. The edges of an image can therefore appear soft or distorted compared to the sharper central area. One of the most difficult things to resolve is chromatic aberration, which is when a lens can't focus the different colour wavelengths all at the same point. Wavelengths of light enter a lens and disperse as they pass through it, in order to get all the different wavelengths to come back together at the point of the sensor, these wavelengths need reorganising in order to become focused. Some very high quality lenses can do that, but there is always some difference in diffraction, the problem is technically called colour fringing. A drawing will as always clarify the issue.
Colour fringing due to the different diffraction rates of light wavelengths
Lateral Chromatic Aberration is what occurs when different wavelengths of light are focused on the same plane, but at different positions, this being due to the angle of light entering the lens and is visible at the edges of the frame, rather than near the centre, which is why post-production or in-camera solutions are needed to alleviate this.
Spherical Aberration  is caused by light rays entering the lens and not converging at the same point. This impacts on image clarity, sharpness, and resolution and is more likely to be seen further away from the centre of the image.  
Light entering at different parts of the lens is not refracted accurately 
Coma or comatic aberration occurs when light rays pass through a lens at an angle, as opposed to straight on. When a lens design cannot focus these angular light rays at the same point, the point light source will be depicted as a teardrop highlight, rather than a circular highlight. This can be minimised by stopping down your lens.
Light coming at the lens from an angle
Photographers are used to compensating for these things, in particular stopping down helps enormously. Stopping down means using a higher f-stop number which decreases the size or diameter of the lens aperture. Reducing the aperture size increases the depth of field, i.e. the focus is sharper over a greater distance. This is why pin-hole cameras really do need a tiny 'pin hole' for light to pass through. 

We still haven't quite squared the circle, which is why we rectangularly frame photographs, when the lens is circular and the resultant images circular too. The corners of a rectangle are further away from the centre than the middle of edges to the left, right, top or bottom, therefore if there is going to be distortion it will be still be revealed in those corners. In fact early cameras often had circular plates such as Thompson’s Revolver Camera from 1862 and 'button' cameras, designed to make small images button sized and shape that could be actually worn like a button. So it wasn't as if the circular format hadn't been considered. The frame as a rectangle, is a powerful concept and somehow it feels right to slice out recorded segments from life with a hard straight edge rather than a circular one. The telescope and the microscope both retain the circular form that reflects the shape of a lens, but as soon as we wanted to record directly what was seen through these optical devices, it was the picture within a rectangular frame that was the right format. 

As soon as an image is made it has to fit a 'rectangular' world. Walls are rectangles and so are tables and shelves. Its easier to make right angled frames and film on a roll or as a plate is easier to use with rectangular formats. So there are a lot of simple practical reasons for retaining a rectangle for the photograph, but they are all linked to our overall shaping of the world with geometry. Cave paintings were not in rectangles, it is only when geometry begins to impose itself on construction methods that the circular or more organic form becomes relegated to history. 

Buildings have been built using geometric principles for thousands of years, but the idea of the frame in relation to a moveable image is quite recent. It is believed that Egyptian Fayum mummy portraits from around 2,000 years ago were made before the person died. They were then hung on a wall until they died and then the portrait was fixed onto the coffin. Whether or not this is true, and there are arguments about how they were used, these images were portable and they would have been placed around the painter's workshop and probably taken into various residences in order for the artists to catch a likeness of the person. There are traces of a surrounding frame on many of them, there frames may have been to fix them to the coffin or to display them on a wall or for both reasons, whichever reason is right, these were portable images that were at one point framed. 

Fayum mummy portrait

Before the portable frame, there was the border and the border operated as a frame. I have been to see the mosaic of the stag hunt at Pella near Athens and you can look down on the floor from a balcony above, as well as walk up to the edges of it on the ground floor. The surrounds operate as a powerful set of symbolic forms, setting the scene of human /animal inter-action within an environment that signifies the wider 'cosmic' arena of nature. The frame in this case is a floor and the image an incident or moment to be picked out within a surrounding world of land and sea life. 

Stag hunt: Pella

We can see the link between architecture and the frame in Classical Greek ceramics. Columns are often used to divide spaces up, the artists obviously linking the idea of an important three dimensional architectural space, often signified by tall columns outlining the entrance, with two dimensional scenes operating around the continuous surface of a ceramic vessel. 

Although borders, or columns acting as frames in ancient art were used to divide scenes as well as provide space for ornamentation in both pottery and wallpaintings, the first wooden frames surrounding images as we know them today appeared on small panel paintings in twelfth and thirteenth century Europe. Often painted onto one solid piece of wood, the area to be painted was carved out, leaving a raised border around its edge. The outer edges were usually gessoed and gilded, before any painting was done, which was often the last part to be completed. Much of the meaning was embedded in the cost of the materials and records from this time emphasise the cost of various materials in craftsmen's contacts, the idea of worth being given to an image by artistic invention was something that would have to wait for the Renaissance.  

The use of mitred moulding strips for making the edges of panels came later, and gradually replaced the simple wooden moulding strips that were attached to the outside edges, especially as larger pieces of wood became harder to get hold of, because more and more of Europe's forests had been cut down for fuel as well as for ships, furniture and housing. 

Paolo di Giovanni Fei: 14th Century painting on wooden panel using an engaged frame

The image above has a frame built around it, (an engaged frame) the frame is deeply carved out, the golden space that the figures exist in being a religious space rather than an actual space, the frame being in effect a spiritual building or architecture to place the iconic image of Mary and Jesus within. The frame is used to give the effect that you are stepping from one world into another, an idea that artists in western Europe will return to many times. 

As you can see a frame is a complex idea and one I shall probably return to again and again. It links newer forms of image presentation with older ones and has always been related to cutting something out of reality and making it special. 

The TV when looked at historically has had a wide range of framing concepts engaged with it, and the particular time periods within which these surrounds were developed also have stylistic impacts on the situation. Because the TV was always associated with the idea of 'modern communications' it was also often placed in a containing surround that was designed to state that modernity.

The state of the art TV screen immediately above has a very thin black border and we are often sold an idea that suggests this border is so thin, it creates no separation between you and the reality the TV depicts. The ultra real 'high definition' screen, allowing you into a world that is as real as the one you are in. Although this frame is wafer thin, in some ways the belief in a new 'realistic' technology is not that dissimilar to the belief in a religion. 

In both cases, the argument is that you can escape your messy, difficult reality by passing through the frame into another world. One is a world of sophisticated technology and imagined universes and the other is of sophisticated technology and imagined universes, we just think they are very different, but the panel painting was at the cutting edge of image technology during its time and the high-definition TV screen is too, but for how long?

See also: