Is there a future for figurative art?

How do pictorial representations work? What enables us to read a depiction as a representation of an object, a situation or an event? Until the early 20th century, such questions would have seemed, to most people, pointless and hardly worth asking. If pressed for an answer, they might have explained that representations work because they resemble the things they depict. However, with the passage of the century this increasingly appeared to be an unsatisfactory or false answer.

Scepticism about the “resemblance theory” of depiction became widespread amongst philosophers, and something of a polemical battleground. Resemblance theorists tried to shore up what looked like a crumbling theoretical position, while sceptics presented seemingly unanswerable challenges to what they regarded as little more than a folk myth.

Probably the most articulate and persuasive of the sceptics was the American philosopher Nelson Goodman. His book Languages of Art, first published in 1976, is, arguably, one of the most important books of the 20th century. Goodman’s position is both complex and subtle, and difficult to encapsulate in a simple statement. However, it would be reasonable to suggest that Goodman regarded pictorial representations as symbol systems which denoted the thing represented, rather than depending upon some inherent process of resemblance. Goodman related pictorial representations to the many other codes of representation, such as natural language (for example, the words I am using to write this article), musical notation or Morse code.

Parallel to this philosophical scepticism about the traditional resemblance theory of representation, art practice itself increasingly abandoned many long-held assumptions about the purpose and nature of pictures. Starting with Cubism, artists progressively moved away from conventional representation of the material world into the fields of abstraction and non-objective art. There was a widespread feeling in art practice and also in art theory that the idea of depicting the material world was exhausted; that the great epochs of Western art, such as the Renaissance, the Baroque and Impressionism, had said everything that was worth saying; and that artists must seek elsewhere for new inspiration and new ways of picture-making. This was given additional force by the invention of new media, such as photography, film and video, which appeared to usurp the traditional functions of representational art, rendering the graphic creation of representations pointless.

Figurative art continued to be respected and admired as, increasingly, a museum art, a pictorial dead language which we could still interpret and understand, but which few living practitioners wished to use. Anyone who frequented the private galleries and salerooms of a city like London over the last half of the 20th century will recall how new figurative art became marginalised. The dominant ideology of non-figurative art itself fragmented into a series of recognisable sub-categories, such as constructivism, abstract expressionism and colour field painting. Many of the foundational practices of figurative art, such as drawing and painting from observation, disappeared from the curricula of art colleges.

Theoretical attacks on the validity of figurative art often identify linear perspective as their chosen battleground. Pictorial perspective was perfected in the early years of the Renaissance, giving artists a systematic method for coherent representation of the material world; as such, it has often been cited in defence of the resemblance theory of depiction. But it can easily be demonstrated that conventional linear perspective is deeply flawed and unreliable as a record of visual experience. It does not, for example, take account of phenomena like parallax or binocular vision. The existence of different and rival systems of perspective, for example in Indian and Chinese art, suggests that perspective is in fact no more than yet another visual code, a conventional symbol system which has to be learned in order to make sense.

My thesis in this article is that the debate over the nature and value of pictorial representation has been led astray. The question most frequently asked is: “Does this representation look like the thing represented?” I would re-frame the question as follows: “Does this representation resemble our perception of the thing represented?”

Of course, perception itself is a complex, mysterious and to some extent inaccessible phenomenon involving physical, neurological and psychological processes. The study of perception ranges widely from the philosophical domain of metaphysics to the laboratory work of behavioural psychologists. Perception is to some extent a reconstructive process, in which the percipient receives, collects and organises sense data in order to make coherent sense of what is being perceived. However, there are some things one can identify in visual perception which have clear analogues in our interpretation of representational art, and in this respect representational art is more than a code or symbol system, as we can demonstrate by systematic comparison with a true symbol system such as natural language.

I would suggest that there are at least four features of our perceptual processes which relate them clearly to our reception of figurative art: these are orientation, configuration, occlusion and scale.

First, orientation. When we perceive a visual array in the material world, such as the room which surrounds you while you read this article, we perceive it in a clear relation to its orientation to our station point – the position from which it is viewed. We see things as, for example, to the left, to the right, above or below. This distribution of the contents of the perceived world is reflected in many physiological and psychological behaviours, such as eye movement and tactile response.

The Still Life with Bottle, Glass and Loaf in the National Gallery, attributed to a follower of Jean-Siméon Chardin, represents a glass of wine, a bottle, a loaf and a knife resting on a newspaper. The glass is to the left, the loaf is to the right and the bottle is between them. The knife lies diagonally on the table, between the loaf and the wine glass. It would be reasonable to assume that the painting was an attempt to create a faithful record of an observed visual array. Given suitable objects, we could easily reconstruct a simulacrum of the subject of the picture in material space.

If we now compare the painted image with the words that I have used to describe it, we see immediately that the code we describe as natural language behaves quite differently. In the sentence, “The loaf is to the right of the glass”, our language sign for “the loaf” is in fact to the left of our language sign for “the glass”. This is because the rules of construction, in English at least, dictate a left-to-right sequence. The rules for Hebrew, Arabic or Japanese would demand a different ordering of units, from right to left or vertically as appropriate. Unlike the graphic depiction of reality, orientation in natural language, and indeed in all coded systems of communication, is arbitrary and determined by agreed conventions.

Secondly, configuration. In the picture, the part of the image representing the glass is wider at the top than at the bottom. The neck of the bottle, as represented, is narrower than its shoulders. We assume that in this respect, the image reflects and records the visual perceptions of the artist, who saw the objects as having these configurations. Here again, as with orientation, there is no analogy with any constructed code of representation, such as natural language, Morse code or naval semaphore. Although these could communicate a descriptive account of what is conveyed in the pictorial image, it would be a rather tedious and inadequate version of what is so rapidly and easily accessible in the picture.

Thirdly, occlusion. In the picture, the loaf obscures, or occludes, a part of the bottle. We do not assume that it is a rather strangely shaped bottle, with the piece at the bottom right missing, but that part of the bottle is temporarily obscured because the loaf intervenes between the bottle and the percipient. This is a familiar and important part of our perception of the physical world. Things are constantly disappearing, partly or wholly, behind one another. The influential Swiss child psychologist Jean Piaget (1896–1980) suggested that infants have to acquire or develop the sense of the persistence of objects – that objects, which are temporarily removed from direct sense perception, nevertheless continue to exist.

As with orientation and configuration, occlusion can only be represented in natural language if it is converted into the alphabetic code from which language is constructed. In the sentence “The cat disappeared behind the sofa”, “the cat” remains as fully visible as if it read, “The cat sat on the sofa”. Occlusion in natural language can only be represented by the utilisation of semantic and syntactic codes.

Finally, scale. The dominant principle of scale is that objects and their characteristics appear to diminish as they recede from the percipient. This applies not only to the physical dimensions of objects, but to factors such as colour, tone and texture. Meindert Hobbema’s The Avenue at Middelharnis is a classic example of this phenomenon in pictorial representation. We assume that the diminution in the size of the depicted trees indicates differences in distance from our station point as an observer, rather than concluding that the avenue consists of trees of radically different sizes. This is precisely the kind of conclusion which we would reach were we presented with the material reality of the subject itself.

Perspective systems, of which there are many, attempt to reduce this principle to a graphic code, and with varying degrees of success. However, the demonstrable flaws in perspectival representation should not lead us to conclude that they are all based upon a flawed premise; it is just that in order to produce a consistent system which can be practised, taught and learned, many concessions must be made.

Similarly, colour, tone and texture can all be modified by the increasing distance between the percipient and the contents of a visual array. When inversions of this principle arise, for example when we see an unexpected splash of colour or brightness some distance from us in a landscape, we assume that it is the result of an unusual occurrence, such as a shaft of sunlight breaking through a cloud.

What I have argued so far is that pictorial representation is a unique phenomenon, which utilises aspects of resemblance to the perceived world, and that in this respect it is quite different from all other forms of encoded communication such as natural language. However, one might agree with this assertion but at the same time believe that it no longer has a function or purpose in modern culture, either because it has been superseded by other media, such as photography and film, which are much better at doing what figurative art once did, or that the language of visual representation has been exhausted by prolonged use, and it is no longer possible to say anything worthwhile or original in it.

I do not believe this to be the case. Figurative depiction of the world around us always carries the potential for surprising novelty. This is, first, because the world itself is in an endless flux of change, and, secondly, because the people who inhabit it see it in different, and often new, ways. One has only to examine the work of a handful of contemporary figurative artists of significance, such as Freud, Rego and Hockney, to see that it could not have been created at any earlier historical moment. This is not a judgement of quality, but an index of the inexhaustible fertility of figurative art. To suggest that figurative art has exhausted its potential makes no more sense that declaring that after an illustrious past theatre is now finished and there is no longer any point in writing or producing new plays.

I began by suggesting that during the second half of the 20th century there was a profound loss of faith in the validity and purpose of figurative art, and that this was reflected in art practice and its institutional support systems, such as art criticism, art galleries and art education. There are signs that the tide has now turned. Galleries and sale rooms suggest a growing appetite for good figurative art of all kinds. Art colleges which had more or less abandoned practices such as drawing and painting from observation are becoming more aware of the source of pictorial wealth which they offer. Let us hope that these are signs of a continuing revival which will enrich the art experience of future generations.

Is there a future for figurative art?

How do pictorial representations work? What enables us to read a depiction as a representation of an object, a situation or an event? Until the early 20th century, such questions would have seemed, to most people, pointless and hardly worth asking. If pressed for an answer, they might have explained that representations work because they resemble the things they depict. However, with the passage of the century this increasingly appeared to be an unsatisfactory or false answer.

Scepticism about the “resemblance theory” of depiction became widespread amongst philosophers, and something of a polemical battleground. Resemblance theorists tried to shore up what looked like a crumbling theoretical position, while sceptics presented seemingly unanswerable challenges to what they regarded as little more than a folk myth.

Probably the most articulate and persuasive of the sceptics was the American philosopher Nelson Goodman. His book Languages of Art, first published in 1976, is, arguably, one of the most important books of the 20th century. Goodman’s position is both complex and subtle, and difficult to encapsulate in a simple statement. However, it would be reasonable to suggest that Goodman regarded pictorial representations as symbol systems which denoted the thing represented, rather than depending upon some inherent process of resemblance. Goodman related pictorial representations to the many other codes of representation, such as natural language (for example, the words I am using to write this article), musical notation or Morse code.

Parallel to this philosophical scepticism about the traditional resemblance theory of representation, art practice itself increasingly abandoned many long-held assumptions about the purpose and nature of pictures. Starting with Cubism, artists progressively moved away from conventional representation of the material world into the fields of abstraction and non-objective art. There was a widespread feeling in art practice and also in art theory that the idea of depicting the material world was exhausted; that the great epochs of Western art, such as the Renaissance, the Baroque and Impressionism, had said everything that was worth saying; and that artists must seek elsewhere for new inspiration and new ways of picture-making. This was given additional force by the invention of new media, such as photography, film and video, which appeared to usurp the traditional functions of representational art, rendering the graphic creation of representations pointless.

Figurative art continued to be respected and admired as, increasingly, a museum art, a pictorial dead language which we could still interpret and understand, but which few living practitioners wished to use. Anyone who frequented the private galleries and salerooms of a city like London over the last half of the 20th century will recall how new figurative art became marginalised. The dominant ideology of non-figurative art itself fragmented into a series of recognisable sub-categories, such as constructivism, abstract expressionism and colour field painting. Many of the foundational practices of figurative art, such as drawing and painting from observation, disappeared from the curricula of art colleges.

Theoretical attacks on the validity of figurative art often identify linear perspective as their chosen battleground. Pictorial perspective was perfected in the early years of the Renaissance, giving artists a systematic method for coherent representation of the material world; as such, it has often been cited in defence of the resemblance theory of depiction. But it can easily be demonstrated that conventional linear perspective is deeply flawed and unreliable as a record of visual experience. It does not, for example, take account of phenomena like parallax or binocular vision. The existence of different and rival systems of perspective, for example in Indian and Chinese art, suggests that perspective is in fact no more than yet another visual code, a conventional symbol system which has to be learned in order to make sense.

My thesis in this article is that the debate over the nature and value of pictorial representation has been led astray. The question most frequently asked is: “Does this representation look like the thing represented?” I would re-frame the question as follows: “Does this representation resemble our perception of the thing represented?”

Of course, perception itself is a complex, mysterious and to some extent inaccessible phenomenon involving physical, neurological and psychological processes. The study of perception ranges widely from the philosophical domain of metaphysics to the laboratory work of behavioural psychologists. Perception is to some extent a reconstructive process, in which the percipient receives, collects and organises sense data in order to make coherent sense of what is being perceived. However, there are some things one can identify in visual perception which have clear analogues in our interpretation of representational art, and in this respect representational art is more than a code or symbol system, as we can demonstrate by systematic comparison with a true symbol system such as natural language.

I would suggest that there are at least four features of our perceptual processes which relate them clearly to our reception of figurative art: these are orientation, configuration, occlusion and scale.

First, orientation. When we perceive a visual array in the material world, such as the room which surrounds you while you read this article, we perceive it in a clear relation to its orientation to our station point – the position from which it is viewed. We see things as, for example, to the left, to the right, above or below. This distribution of the contents of the perceived world is reflected in many physiological and psychological behaviours, such as eye movement and tactile response.

The Still Life with Bottle, Glass and Loaf in the National Gallery, attributed to a follower of Jean-Siméon Chardin, represents a glass of wine, a bottle, a loaf and a knife resting on a newspaper. The glass is to the left, the loaf is to the right and the bottle is between them. The knife lies diagonally on the table, between the loaf and the wine glass. It would be reasonable to assume that the painting was an attempt to create a faithful record of an observed visual array. Given suitable objects, we could easily reconstruct a simulacrum of the subject of the picture in material space.

If we now compare the painted image with the words that I have used to describe it, we see immediately that the code we describe as natural language behaves quite differently. In the sentence, “The loaf is to the right of the glass”, our language sign for “the loaf” is in fact to the left of our language sign for “the glass”. This is because the rules of construction, in English at least, dictate a left-to-right sequence. The rules for Hebrew, Arabic or Japanese would demand a different ordering of units, from right to left or vertically as appropriate. Unlike the graphic depiction of reality, orientation in natural language, and indeed in all coded systems of communication, is arbitrary and determined by agreed conventions.

Secondly, configuration. In the picture, the part of the image representing the glass is wider at the top than at the bottom. The neck of the bottle, as represented, is narrower than its shoulders. We assume that in this respect, the image reflects and records the visual perceptions of the artist, who saw the objects as having these configurations. Here again, as with orientation, there is no analogy with any constructed code of representation, such as natural language, Morse code or naval semaphore. Although these could communicate a descriptive account of what is conveyed in the pictorial image, it would be a rather tedious and inadequate version of what is so rapidly and easily accessible in the picture.

Thirdly, occlusion. In the picture, the loaf obscures, or occludes, a part of the bottle. We do not assume that it is a rather strangely shaped bottle, with the piece at the bottom right missing, but that part of the bottle is temporarily obscured because the loaf intervenes between the bottle and the percipient. This is a familiar and important part of our perception of the physical world. Things are constantly disappearing, partly or wholly, behind one another. The influential Swiss child psychologist Jean Piaget (1896–1980) suggested that infants have to acquire or develop the sense of the persistence of objects – that objects, which are temporarily removed from direct sense perception, nevertheless continue to exist.

As with orientation and configuration, occlusion can only be represented in natural language if it is converted into the alphabetic code from which language is constructed. In the sentence “The cat disappeared behind the sofa”, “the cat” remains as fully visible as if it read, “The cat sat on the sofa”. Occlusion in natural language can only be represented by the utilisation of semantic and syntactic codes.

Finally, scale. The dominant principle of scale is that objects and their characteristics appear to diminish as they recede from the percipient. This applies not only to the physical dimensions of objects, but to factors such as colour, tone and texture. Meindert Hobbema’s The Avenue at Middelharnis is a classic example of this phenomenon in pictorial representation. We assume that the diminution in the size of the depicted trees indicates differences in distance from our station point as an observer, rather than concluding that the avenue consists of trees of radically different sizes. This is precisely the kind of conclusion which we would reach were we presented with the material reality of the subject itself.

Perspective systems, of which there are many, attempt to reduce this principle to a graphic code, and with varying degrees of success. However, the demonstrable flaws in perspectival representation should not lead us to conclude that they are all based upon a flawed premise; it is just that in order to produce a consistent system which can be practised, taught and learned, many concessions must be made.

Similarly, colour, tone and texture can all be modified by the increasing distance between the percipient and the contents of a visual array. When inversions of this principle arise, for example when we see an unexpected splash of colour or brightness some distance from us in a landscape, we assume that it is the result of an unusual occurrence, such as a shaft of sunlight breaking through a cloud.

What I have argued so far is that pictorial representation is a unique phenomenon, which utilises aspects of resemblance to the perceived world, and that in this respect it is quite different from all other forms of encoded communication such as natural language. However, one might agree with this assertion but at the same time believe that it no longer has a function or purpose in modern culture, either because it has been superseded by other media, such as photography and film, which are much better at doing what figurative art once did, or that the language of visual representation has been exhausted by prolonged use, and it is no longer possible to say anything worthwhile or original in it.

I do not believe this to be the case. Figurative depiction of the world around us always carries the potential for surprising novelty. This is, first, because the world itself is in an endless flux of change, and, secondly, because the people who inhabit it see it in different, and often new, ways. One has only to examine the work of a handful of contemporary figurative artists of significance, such as Freud, Rego and Hockney, to see that it could not have been created at any earlier historical moment. This is not a judgement of quality, but an index of the inexhaustible fertility of figurative art. To suggest that figurative art has exhausted its potential makes no more sense that declaring that after an illustrious past theatre is now finished and there is no longer any point in writing or producing new plays.

I began by suggesting that during the second half of the 20th century there was a profound loss of faith in the validity and purpose of figurative art, and that this was reflected in art practice and its institutional support systems, such as art criticism, art galleries and art education. There are signs that the tide has now turned. Galleries and sale rooms suggest a growing appetite for good figurative art of all kinds. Art colleges which had more or less abandoned practices such as drawing and painting from observation are becoming more aware of the source of pictorial wealth which they offer. Let us hope that these are signs of a continuing revival which will enrich the art experience of future generations.