Anna Bentkowska

Moving Images, Shifting Notions. Who Can Trust the Digital Image?

Keywords: digital image, manipulation, Artificial Life images

This paper is about the manipulation of images, and digital images in particular. This vast subject has been approached from a variety of positions. The focus here is on those techniques of digital imaging that challenge established theories and conventions of image making and interpretation.

The fast pace of the ongoing digital revolution has generally been accepted as inevitable and far reaching. Computing brings disciplines that used to exist in separation closer. There is, however, a tendency of setting limits to what areas of computing might be applied to one's own field. The humanities scholars' approach to computing science is especially selective. Those whose interests are in visual culture, for instance, are unlikely to see the latest techniques in medical or biological computing as immediately applicable to their research. Recent developments in medicine and biology, such as cloning of live organisms, have relied heavily on digital methods. Research, currently undertaken in Australia, into bringing back to life the extinct Tasmanian thylacine tiger, following the successful reconstruction of its genetic blueprint, brings closer the so far unthinkable possibility of resurrecting dead organisms. Frankenstein-like fiction is becoming something of a reality. Although, seemingly, of remote interest to the majority of cultural historians, the concepts behind such developments and the wave of new imaging techniques may have direct implications in our ability to understand the way images can be created and analysed. The DNA code of Sir John Sulston is the subject of his portrait by Marc Quinn in the National Portrait Gallery in London. Created by using a now standard method for DNA cloning, Sulston's portrait contains his DNA replicated in bacteria, which have been stabilized in agar jelly. Without the knowledge of the processes involved in its making, this small, c. 85x120 mm image is likely to be described as showing tiny organic forms that resemble drops of olive oil floating on the surface of water. Such terms as 'portrait', 'sitter' and 'true likeness' are more relevant to this entirely abstract work than to any realistic depiction of man, yet the shift in these notions is unprecedented; so is the challenge of seeing the invisible. In order to recognise the possible link between genetic research and art imaging, one would have to accept the emerging fusion of different realities - lost, imagined and existing ones.

In his book on dinosaurs as icons of American culture in the 20th century, William J. Mitchell elaborates the idea that images might live like a species and can be regarded as an extension to organic life. ¹Mitchell is an iconologist and his idea of living images refers first and foremost to the evolution of images, or rather the evolving iconography of a specific subject. To better understand his point, one may compare the life-size Megalosaurus, made of brick and reinforced concrete, shown at the Crystal Palace in London during the 1854 Great Exhibition, with the recent animatronic and computer graphics used in Tim Haines's BBC television production Walking with Dinosaurs. If one thinks what an enormous step, in terms of knowledge of the subject as well as rendering techniques, has been made between 1854 and 1999, one realises what the evolution of images of dinosaurs can mean. Mitchell was unable to make this particular comparison, because the publication of his book preceded the TV production. But if he watched Walking with Dinosaurs, he must have been pleased to see images transformed into living organisms; the computer models used in this series illustrated his point to a tee.

The use of computer graphics, animation and sound has forced palaeontologists and natural scientists to re-examine existing knowledge of prehistoric animals. Digital visualisation techniques have forced them to go even beyond this knowledge, and venture into hypothetical reconstructions of behaviour and appearances, including such features as the colour and texture of the skin, speed of movement and the sounds that dinosaurs might have made, all of which so very little is known. ² As a result, the beholder has been confronted by a visually convincing construct of a multimedia image of the dinosaur; this new image embraces the re-examined science and the assumed knowledge of the way dinosaurs lived, looked and behaved. Only those equipped with the in-depth knowledge of the evidence are able to tell what in this living imagery is real and what is artificial.

Although Mitchell's other research is concerned to some extent with digital culture, his idea of the analogy between the image and organic life pertains to images in general and not just digital images. It is not his original idea. Mitchell follows here the French art historian Henri Focillon and his once celebrated, but now rather forgotten work La vie des formes, published in 1934, and translated into English in 1948 as The Life of Forms in Art. To sample Focillon's metaphoric style, this is what he says about a drawing: "In a drawing the paper is an element of life; it is the very heart of the design." ³ The paper is not a mere support; it gives the drawing its feel, substance, existence, life.

As the son of a prolific engraver who had a successful printing business, Focillon had direct experience of graphic art. However, his theories are not medium, period or subject specific and relate to processes common to all media. This is why his thought still appeals. Focillon's idea of images being dependent on principles of life goes beyond superficial comparison and the borrowing of terminology. It is not a simplistic metaphor either, but rather a system of complex ontologies that link artistic forms with biological ones. While looking at the life of the digital image, I shall retain Focillon's approach and touch upon select aspects of the life of the digital image: its birth and progenitors, body and habitat, its reproduction, as well as its death and after life. ⁴

The life of the digital image has never been easy. During its short and rebellious existence the evolving formats of computer graphics and their applications have been regarded as insubordinate to the established boundaries and criteria of art. The immaterial nature of the digital image and the ways in which it can be created, challenge the long-held and still popular view that the work of art is constituted by a material object made by an artist, and in any case by a human being. Among those who have predicted that the physical component, 'present in all arts' will be affected by 'profound changes' was Paul Valéry. Early in the 20th century he remarked, 'We must expect great innovations to transform the entire technique of the arts, thereby affecting artistic invention itself and perhaps even bring about an amazing change in our very notion of art". ⁵

The digital image is born

Uniqueness is a feature associated with organic life. Even so-called identical twins are never truly identical and can even be of different gender (brother and sister). Not even a base pair of units in the DNA strands, are identical: variations in any two copies do occur once in every 1000 base pairs. Dolly, the sheep and other recently cloned animals are not truly perfect clones either, but with the advance in genetic engineering exact copies of organisms may eventually become a terrifying reality. Uniqueness has also been traditionally regarded as THE criterion of a work of art. "A work of art - says Focillon - is an attempt to express something that is unique." ⁶ The uniqueness of an artistic idea can only be checked against knowledge; if one is unable to trace the model, influence or inspiration behind an artistic concept, one is likely to take it for original, authentic and unique. The artist Tracey Emin, currently in the British art scene spotlight, can be regarded as an innovative artist only by those who are unaware how heavily she borrows from others. Her tent installation Everyone I have ever slept with, video Interview and short story Exploration of the Soul, for example, are all reworked ideas of Billy Childish.

The uniqueness of a work of art in a physical sense still adds to its appreciation and value. Given the choice, no one wishes to acquire a copy of a Van Gogh, however faithful, instead of his original. Works of art do exist in editions, copies, replicas and variants; but somehow looking at Michelangelo's plaster David in the Cast Court at the Victoria and Albert Museum, is not the same as admiring the original marble. But if one looks at the former without the knowledge of the latter, which judging by the comments made by visitors to the V&A is sometimes the case, one's experience may not be much different from looking at the original. It is the knowledge of the original that does not allow us to put its copy, however artistically fluent and technically accurate on an equal footing with the model. Rubens's paintings after Titian display all the mastery of the copier, some would argue even excel Titian, yet lack in the originality of the idea. The passage of time affects the status of a copy, adding to its appreciation. Hieronymous Cock's engravings after paintings by Peter Bruegel the Elder originated as mere reproductive prints. Their role was to disseminate the art of the great master, but with time became graphic works of art in their own right. Are digital reproductions likely to rise to such a status? Time will tell. What we are already able to do is to date some digital images of art and architecture. We can, for example, differentiate between early computer graphics, such as rather crude computer models of architecture from the 1980s, and recent ones, which have been rendered through more sophisticated photo-realistic techniques.

The issue of uniqueness of a digital image as a work of art is particularly complex. The computer artist makes a copy of his work as soon as he considers it worth preserving. The concept of the original blends, therefore, with that of a replica, copy and reproduction. This is not much different, one may argue, from a bronze sculpture, which exists in a number of editions, each of which may retain the attribute of originality but never that of uniqueness. The comparison with any other solid and tangible medium may seem ill-conceived because physically the digital image does not exist at all. And yet, it is extremely difficult to renounce the traditional criteria of art entirely when dealing with digital media. Attempts, so frequently futile, at protecting the status of the digital image through strict copyright, prove how difficult it can be to redefine the old criteria.

The digital image and its progenitors

Irina D. Costache, who discussed some of the same issues at one of the earlier CHArt conferences has argued that, "The term 'master' in cyberspace is given to the technician not the artist". ⁷ While this may be true in relation to images 'born' through digitisation, condemning the artist completely to oblivion does not seem to answer Costache's question "Who is the author?" The choice is wider than that between an artist and a technician. There is a relatively little known, though by no means new genre of computer art, which questions the role of the human creator altogether. I am referring here to computer graphics that are products of ArtLife.

Fig. 1 An Artificial Life image can replicate itself ad infinitum. See The Artificial Painter Gallery at http://www.daimi.aau.dk/~hhl/ap_gallery.html for further examples.

ArtLife stands for Artificial Life and, alongside Artificial Intelligence, is a discipline of computer science. ⁸ It brings together people of very different backgrounds: cognitive scientists and evolutionary biologists, mathematicians and ornithologists, philosophers and physicists, scholars of social and adaptive behaviours, as well as artists and those generally interested in aesthetics. Artificial Life scientists view the Darwinian evolution as a discovery of "the power of algorithm" and the mechanical basis of the process of reproduction. The ultimate goal for art-life scientists, believing in 'strong AL', as opposed to 'weak AL' is to surpass the biologists who have already succeeded in creating life in vitro, by creating life in silico. Sceptics wonder whether this is a serious science or a new form of Frankensteinism.

The principles of Artificial Life go back to the 1950s and John von Neumann's abstract cellular automaton, capable of generating a replica of itself over and over again, ad infinitum. In this process the concepts of the original and copy are made redundant, as every copy becomes a model for the next copy. The issue of who should be credited with the act of creation is particularly difficult to resolve. An algorithm starts the creative process, as a result of which very complex structures or models can be constructed. In Artificial Life it is believed that the same mathematical process governs biological systems. The process does not require man's control, hence the term 'automaton'; man is responsible for the formula that initiates the creative process and can stop it at any stage, but his control over its products, 'living' images, is limited. The behaviour of a particular element of the structure, or model is controlled by its neighbouring elements. In 1986 Craig Reynolds introduced new AL software, which he called Boids (see www.red3d.com/cwr/boids). Boids are bird-like shapes whose behaviour can be compared to birds flying in formations, able to avoid an obstacle, regroup, split, and create new groups. The three steering behaviours are separation, alignment and cohesion. Examples of boids are available on the Internet, alongside a multitude of other AL products. Most computer users have no technical facilities to view boids in action and can only see AL images reduced to video or stills. AL has been used to create special effects in a number of feature films, with viewers frequently unaware of the technique employed. Bat swarms in Tim Burton's Batman Returns (1992) and the wild beast stampede in Disney's The Lion King (1994) are both manifestations of boids.

The striking feature of many Artificial Life images are their aesthetic qualities and decorativeness, features which contemporary art has been so keen to renounce. Although my aim here is not to evaluate AL images, I am not afraid to refer to them as art. Some AL images may lack deeper meaning or sophistication other than technical, but in my view art history should not be ignoring their existence. The problem is that art historians and critics may find themselves speechless in front of AL works, unable to fully understand the technicalities of the medium or unwilling to assimilate the terminology mostly alien to the old history of art. In order to describe the behaviour-based events characteristic of AL, cognitive and computer scientists make an over-extensive use of collective terms borrowed from biology and psychology. They refer to the artificial creations as if to real, living animals or humans. They personify their behaviour and endow them with feelings. In AL literature, the life-related terms are seldom used in inverted commas. While art historians are perfectly familiar with personifications and other forms of anthropomorphism in art, they may find venturing beyond such metaphoric constructs difficult. If works created through the use of AL software are to become a subject of art historical studies, we must find ways of reconciling the artifice of art (which has traditionally been the focus of such studies) with Artificial Life's ambitions to play Nature.

Art works that explore the Artificial Life environments are shown at SIGGRAPH - Special Interest Group on Computer Graphics conferences and workshops, held annually since the 1978. 'The Bush Soul', shown in 1998, was an interactive installation created by a team of computer scientists from the University of California in Los Angeles, lead by the computer-artist Rebecca Allen (see http://emergence.design.ucla.edu). This installation used new Emergence software to display, in real time, computer-generated three-dimensional AL environments. As Allen describes her project, the virtual characters are programmed to communicate and interact "through gestures, sound and the emission of energy in the form of light particles. As in real-life, feelings and resulting behaviours can change based on certain interactions or due to time-based events. Complex social environments can emerge from the interaction of simple behaviours."

Fig. 2. Rebecca Allen, The Bush Soul. Reproduced courtesy of the artist

If one resorts to old conventions of representation, one could describe Allen's environments as metaphysical landscapes of serene beauty. The credit for this work goes first and foremost to computer science but reflects artistic, psychological, social and biological interests of its creators. They invented the software and conceptual framework for this experiment, but had little influence on the behaviour and interaction with the participating humans and the way the events have developed. So, where is the author? One may, of course, draw on the 20th century critique of the author that attempted to renounce the role of biological individuals, and conveniently adopt Rolland Barthes's view that the author is dead and all there is, is Text ('text' being, of course, a generic term for any carrier of information, whether textual or visual). This view has been tested successfully against various unconventional manifestations of art, but in the context of digital images it is not totally satisfying. This is because, in Barthes's own words, 'the Text is not to be thought of as an object that can be computed', nor should it be separated out materially from a work. ⁹ The digital image, being an abstract construct that can be computed, escapes Barthes's definition of Text.

As far as the quest for the author is concerned, my suggestion is to resurrect Focillon's view that a progenitor of every image is another image. Form bears form. It could be argued that this theory frees the artist from seeking originality. Although formulated long before the advent of digital culture, it can be aptly applied to the creative processes of Artificial Life, as well as some of the animated forms of the Image-Text specific to a multimedia format. An example of the latter, which I would like to refer to appeared briefly in online lecture notes on manipulation of digital images on a website of an UK college. Worryingly powerful in its propagandist content, the combined Image-Text consisted of the well-known logo of Microsoft Windows, transformed into a swastika. It represented an extremist view in the debate on Microsoft dominating the world computer market. The wording 'Microsoft Office 2000' was changed into 'Final Solution 2000'. The design of the logo was not altered at all, just animated to show its different parts, in black and red, in timed sequence. This construct is exactly Focillon's image-ex-image. So are ArtLife images.

Taxonomy of the digital image

When a new species is discovered, one that does not share its characteristics with an existing one, such a new species is given a new name. When introduced, digital images have not been given a new name but the old 'image' continued to be used. Rolland Barthes reminds us how deeply this term is rooted in the concept of imitation and points out the problems arising therefrom, for the semiology of images.¹⁰ (It is of note, however, that this etymological issue is limited to Latin-based languages only.) When introduced, the digital image was regarded as a natural extension to existing imaging and reproduction techniques, photography in particular. Photographic theories of representation cannot, however, be always directly applied to digital images. Let us take, for example, Barthes's view of photography as "the mechanical analogue of the real", or his assumption that photography "can lie as to the nature of the thing", but "never as to its existence".¹¹ While Barthes views can be true of digital images that are created through the translation from the real, the Artificial Life works will never comply.

Benjamin, a contemporary of Focillon, is frequently quoted in discussions of the relationship between photography and digital imaging; a number of CHArt papers have been constructed around his article 'The Work of Art in the Age of Mechanical Reproduction" (or 'Technical Reproducibility', as Stephen D. Ross believes reflects closer Benjamin's thought).¹² Surprisingly, perhaps, I should like to refer here to an earlier text by Benjamin, one in which he is concerned with "The Task of the Translator". In this text, published in 1923, Benjamin says, "Fidelity and freedom in translation have traditionally been regarded as conflicting tendencies, the two are difficult to reconcile, in fact [one] seems to deny the other all justification." ¹³ The digital image is able to reconcile both tendencies, offering a faithful translation of the real into virtual (analogue into digital) as well as freedom of manipulation.

The body of the image, its parts and transplants

Manipulation brings first and foremost negative connotations. The ease with which images can be edited or doctored digitally, in a way that is impossible to detect, provokes a far more serious criticism than any other imaging technique. According to Joan Fontenberta, photography, for example, has managed to "accommodate the myth of objectivity" and yet, "not only permitted deceit but [even] facilitated it". ¹⁴ The same can be said about earlier visual art genres that have managed to retain the image of truthful representation of the real. Seemingly realistic views often lack topographical exactness, history and genre scenes interpret events freely. Portraits by old masters may seem less prone to fabrication, and yet it is not difficult to give examples of 'doctored', even cloned effigies from the past. Peter Lombart's 'Headless Horseman' is a well-known example in British art of an engraved portrait, which depending on current political situation served either as the effigy of Charles I or Oliver Cromwell. The transplantation of the same head onto a different body, or the use of the same torso for different sitters is a practice known from a number of 17th century workshops. When finished, such a cloned portrait functioned as a true likeness. Modern art is also 'guilty' of deception of various kinds. Some of Edouard Degas's post-impressionist landscape studies of fleeting light effects were not created in the open, as one would have expected, but in the studio, by looking at models made of stones, small branches and clay figurines. This old method, first recorded by Cennino Cennini around 1400, was used by many later artists. Their constructed works are frequently taken for what they are not. Images have always been manipulated for artistic, as well as political, ideological and commercial reasons, sometimes out of pure vanity, to fulfil one's expectations of beauty or ambition.

Image processing software makes the cloning of the digital image and transplantation of its parts extremely easy. This means that editing techniques can be readily used and abused. It is not always easy to recognise a doctored image. Our judgement is put to test daily, every time we look at photographs in newspapers. When reading and interpreting manipulated images one can easily became a victim of one's own overconfidence, lack of judgement or observation.

An image can deceive through its content as well as technique. This figure shows what appears as childish, hand drawing. It is only when we look at a series of similar images that we realise the repetition of pictorial motifs. All these elements have been originally drawn by hand, then scanned and put on a transparent background ready to be reused in a variety of designs (Fig. 3). Such a technique affects the creative process, which cannot start with a general idea for the whole composition and progress by filling it in with particular details. The creative process imitates here the jigsaw principles and progresses from the detail to the whole, and not the other way round as it was usually the case before the advent of computers.

Fig. 3. Postcards by 'Fabryka & Forma', 1999.
Reproduced courtesy of Zofia Kumpera.

The next examples also refer to the idea of transplantation of parts of one image into the body of another image. A variety of iconographic and stylistic investigations in pictorial arts can benefit from this technique, but it is rarely applied. (This is the subject of my current research which I intend to present elsewhere in more detail.) The technique of a digital collage of various images seems particularly well suited to support the art historical debate on the role of quotation in pictorial arts. One of the subjects of such a debate has been Caravaggio. Mieke Bal recent study examines the ways in which this Baroque master have influenced artists in the 20th century, but also how the latter have changed our reading of Caravaggio's works.¹⁵ Another famous example of the use of the quotation in painting is Eduard Manet's Le déjeuner sur l'herbe. The idea for this painting came to Manet after he had seen Giorgione's Concert champêtre in the Louvre (now attributed to Titian). "Je veux refaire cela et le faire dans la transparence de l'atmosphere", he is known to have said. As we know, Manet modelled the figures in the foreground of his Déjeuner on a lost drawing by Raphael depicting The Judgement of Paris. Raphael's drawing was known to Manet from an engraving by Marcantonio Raimondi. A number of other artists used three seated figures from Raimondi's print in new works. The relationship of these works has been the subject of a debate involving art historians and semiologists. ¹⁶ Some scholars see the engraving as a direct model for the painting, some talk about a collage or trompe l'oeil, an image within an image; others wonder whether this borrowing should be seen as a manifestation of inter-contextuality, stressing the influences in both directions: the earlier works influencing the later ones, but also the more recent ones changing the reading of the sources from which they derive. The issue of the originality of the quotation is central to this debate, which evolves around comparative analysis of pictorial motifs, their relationship and role as sources for another composition. Even a simple superimposition of images and quickly made collages of relevant motifs, are a helpful visualisation of the debated issues and serve especially well the analysis of the alterations made to the model image.

I have found the same technique most useful in iconographic and stylistic analyses and interpretations of 16th-century landscape painting. In stylistic analyses high-resolution images must be used; this is particularly important in the analysis leading to resolving the attribution of a painting; in such a case every brushstroke counts. The technique in question is one of the possible responses to Chris Bailey's paper "Compare and Contrast" which he presented in 1999 at the DRH (Digital Resources in the Humanities) and CHArt conferences on the ways in which art historians use digital images. ¹⁷ Elements can be taken from different pictures and implanted into another painting in order to prove specific points about stylistic and iconographic similarities; the same method may help in attributing the core painting to the artist of the works from which the details were taken. When these elements blend with the body of the core image and cannot be recognised as implants, the case is made. There is, however, a great risk of such a method of comparative analysis. The resulting image is always a false, misleading construct, especially when taken out of the context of specific research and given an independent life. One would have to have an in-depth knowledge of all paintings used in the test and an absolute photographic memory to physically deconstruct the new image.

Morphing techniques are even more dangerous, for in order to morph one image into another it is usually necessary to alter both images beforehand to anchor common points of the forms to be metamorphosed. This technique is used in popular culture photography and much enjoyed for its frequently unexpected, if not spectacular results. In the history of art, morphing can be applied to visualize various stages in the creative processes, for example to show the relationships between preparatory drawings and a finished painting, or between a series of interdependent works (e.g. a model and its repetitions). The images used in such visualizations are inevitably scaled and their composition altered.

The computer artist, Lillian Schwartz has morphed Leonardo's self-portrait into The Mona Lisa. Schwartz has published her work and commented on the processes involved in the making of the double image. ¹⁸

Fig. 4. An animation simulating the effect of Leonardo's self-portrait morphed into the Mona Lisa. Intermediate images reproduced courtesy of Lillian Schwartz.

A series of intermediate images was created to demonstrate a gradual transition from one face into the other. Many artists, and Marcel Duchamps most famously perhaps, manipulated The Mona Lisa, exploring new techniques and searching for this image's new meanings. We have no problems with accepting Schwartz's digital artwork as a significant contribution of this kind. The judgement becomes more problematic when the Leonardo/Mona Lisa image is seen not as art but as evidence, and used to support a hypothetical identification of the female sitter of the famous painting as Leonardo himself, first suggested by Maurice Vieille in 1913. With no records of Leonardo's true intentions, Schwartz's manipulated work presents a challenge to art historians if they wish to argue against this identification through solely historical arguments. Leonardo believed that an artist should always lend his mind and body to the image he is creating. Schwartz proposes a literal interpretation of Leonardo's view and her visual argument has much power.

The judgement of doctored images depends on the objectives of the manipulation as if the end could justify the means. There are instances when manipulation of visual records brings them closer to the truth. Some of the recent TV documentaries on World War II add colour to the old black and white material. Colour is an important element of historical memory and bears deep meanings: the survivors of concentration camps recall the drama of looking at the blue sky and sunshine through barbed-wire fences, seen as symbols of freedom and captivity. Colour is an essential element of any visual record, and yet conventional art reproductions tend to interpret the colour freely. If we take a number of reproductions of the same painting, we will be amazed how much they can differ in colour. Digital reproductions make colour correction easy, but to use this kind of manipulation the image must be checked against the original and this is rarely the case.

The death and afterlife of the digital image

The digital manipulation of images is a powerful tool. So powerful that it has been regarded as instrumental in the death of the image as we used to know it. In the late 1980s and early 1990s when the digital imaging revolution fed on the factors of the novel and unfamiliar, alarmed voices pointed out to a host of serious legal and ethical implications of image manipulation. A list of titles of the articles published at that time confirm the fear of the digital image, as for example, "Computer as Accessory to Photo Fakery"¹⁹ , "Ask It No Questions: The Camera Can Lie"²⁰ , or the truly catastrophic, "Digital Retouching: The End of Photography as Evidence of Anything"²¹ .

It is not the computer that should be blamed for the panic. The reason for the understandable fear lies, in my view, in the realisation that we are no longer in control of images, far too often not knowing what has and what has not been altered within the image; sometimes we do not even know how the image came into being. The medium is capable of seamless alterations.

It is not the death of images which image processing is likely to cause. As far as art images are concerned, it is rather the death of strict conventions and mystifications intrinsic to the traditional history of art and its methods. Digital images require a new flexibility of approach. "Once upon a time, pictures were simple" - reflects, not without certain nostalgia, James Elkins in his book Why are our Pictures Puzzles? In which he considers strategies for containing (or unfolding) the ambiguities of modern images. ²²

Images are always simple when we feel in control of them. What we do not like is when images question our knowledge, expertise and professionalism. And manipulated digital images are very good at fooling the beholder. So who can trust the image? Only one who has the ability of reconstructing all the subsequent stages in the life of the image, right from its conception, as well as all the possible treatments the image has undergone. This knowledge is certainly available to the image-maker and manipulator, but sometimes also to a knowledgeable and observant viewer.

Pedro Meyer is a photographer with special interest in digital imaging. ²³ His book Truths and Fictions. A Journey from Documentary to Digital Photography tries to teach how to look at images. ²⁴ His 'Freeing the Film' (1987-93) could be taken at first for a photograph taken at an exhibition of Richard Avedon's photographs.

Fig. 5 Pedro Meyer. Freeing the Film, 1987-93 and photographs used to create the final image. Reproduced courtesy of the artist.

A second look gives away the tricks with some of the characters represented. The man in the black hat has a strange double-presence: in one of the pictures on the wall and as a visitor to the gallery. The 'Freeing the Film' is a collage of a number of photographs registering events in one day in the life of Pedro Meyer when he went to see the exhibition of Avedon's photographs on show in New York in 1987. Walking down the street he saw the old man with a billboard. He took a couple of photographs of him. At the exhibition he took further photographs of the photographs on display. He took photographs of visitors as well. Years later Meyer assembled them all digitally. "The image of the man standing there, like a Magritte figure, is in the process of freeing film from its traditional moorings of representation - says Meyer, showing the evolution of meanings engaged in the dialogue between artefacts, their media and mediators bringing all into the picture".

Roger Malina, the spiritus movens behind the "Leonardo" journal, also active at the SIGGRAPH Artificial Life workshops, stated - in response to Walter Benjamin's seminal article on reproducibility of works of art - that, "The unique computer tools available to the artist such as those of image processing, visualization, simulation and network communication are tools for changing, moving and transforming, not fixing, digital information."²⁵ A similar shift from the fixed to the flexible may become characteristic of today's art history, forced upon the discipline by the ambiguous nature of its objects, subjects and methods. The challenge seems hard to resist.

Acknowledgements: The author wishes to thank the artists Rebecca Allen, Zofia Kumpara, Pedro Meyer and Lillian Schwartz for their permission to use their images and Hazel Gardiner for help with preparation of this paper.

September 2000

Notes

NB. All internet links active at the time of publication.

1. Mitchell, W. J. (1998), The Last Dinosaur Book. The Life and Times of a Cultural Icon, Chicago and London: The University of Chicago Press.

2.See Martill, D. and Nash, D. (2000), Walking with Dinosaurs: The Evidence. London: BBC.

3. H. Focillon (1948), The Life of Forms in Art, p. 15. New York: Wittenborn.

4. I am unable to assess the relevance to my paper of Peter Mason's forthcoming book The Life of Images, to be publish in November 2001 by Reaktion Books. Despite its title, Mason's study, being concerned with the evolving iconography and migration of ethnographic images of natif Americans from the late 16th century to the present, seems to focus on different issues.

5. Valéry, P., "La conquète de l'ubiquité", Pièces sur l'art; quoted after Benjamin, W. (1999), "The Work of Art in the Age of Mechanical Reproduction", Illuminations, p. 211. London: Pimlico.

6. Focillon, H. (1948), The Life of Forms in Art, p. 1. New York: Wittenborn.

7. Costache, I. D. (1999), "The Work of Art (Historians) in the Age of Electronic (Re)Production", p. 69, Computing and Visual Culture. Representation and Interpretation, Proceedings of the Fourteenth Annual CHArt Conference, Victoria and Albert Museum, London, 24-25 September 1998, London: CHArt.

8. The literature on Artificial Life is extensive, see bibliography compiled by E. A. Di Paolo available at http://www.cogs.susx.ac.uk/users/ezequiel/alife-page/alife.html. A good introduction to AL is Levy, S. (1992), Artificial Life: A Report from the Frontier Where Computers Meet Biology, New York: Random House. A useful review of this book together with further bibliography and a list of internet sites dedicated to AL can be found at http://www.emcp.com/intro_pc/reading13.htm.

9. Barthes, R. (1977), "From work to Text", p. 155, Image-Music-Text, New York: Noonday. 1st French ed., 1971.

10. Barthes, R. (1977), "The Rhetoric of the Image", p. 32, Image-Music-Text, New York: Noonday. 1st French ed. 1971.

11. Barthes, R. (1984), Camera Lucida: Reflections on Photography, p. 87, London: Flamingo (1st French ed. La chambre claire: note sur la photographie, Paris 1980)

12. Benjamin, W. (1999), "The Work of Art in the Age of Mechanical Reproduction", Illuminations. London: Pimlico. 1st German ed., 1936; Ross, S. D. (1994) (Ed.), Art and its Significance. An Anthology of Aesthetic Theory, 3rd ed., see note on p. 526. New York: State University Press.

13. Benjamin, W. (1999), "The Task of the Translator", p. 79. Illuminations. London: Pimlico.

14. Fontenberta, J. (1995), Introduction to Pedro Meyer's Truths and Fictions. A Journey from Documentary to Digital Photography, p. 7, New York: Aperture.

15. Bal, M. (1999), Quoting Caravaggio. The University of Chicago Press.

16. Sczepinska-Tramer, J. (1999), "O Sniadaniu na trawie Eduarda Maneta i o cytacie w malarstwie" [Manet's Le déjeuner sur l'herbe and the role of quotation in painting; in Polish with abstract in English], pp. 411-420, Ars Longa. Warsaw: Arx Regia, is the most recent contribution to the debate on the role of quotation in Manet's paiting and lists bibliography.

17. Bailey, C. and Graham, M. (1999), "Compare and Contrast: the impact of digital image technology on art history", Digital Environments: Design, Heritage and Architecture, Proceedings of the Fifteenth Annual CHArt Conference, University of Glasgow, 24-25 September 1999, available by subscription at http://www.chart.ac.uk/chart1999/index.html

18. Schwartz, L. (1995), "The Art Historian's Computer", Scientific American, April 1995, pp. 106-111. Schwartz's website is at www.lillian.com.

19. Browne, M. W. (1991), "Computer as Accessory to Photo Fakery", The New York Times, July 24, p. A6.

20. Grundberg, A. (1990), "Ask It No Questions: The Camera Can Lie", The New York Times, August 12, Art and Leisure Section, p. I, 29.

21. Brand, S., Kelly, K. and Kinney, J. (1985), "Digital Retouching: The End of Photography as Evidence of Anything", Whole Earth Review, July, pp. 42-49.

22. Elkins, J. (1999), Why are our Pictures Puzzles?, New York and London: Routledge.

23. Pedro Meyer's website can be found at www.zonezero.com

24. Meyer, P. (1995), Truths and Fictions. A Journey from Documentary to Digital Photography, New York: Aperture.

25. Malina, R. (1990), "Digital Image - Digital Cinema: The Work of Art in the Age of Post-Mechanical Reproduction