1   Introduction

Margaret A. Boden and Ernest Edmonds

This book is about computer art and its relations to art of a more traditional kind. Specifically, the book focuses on generative art, in its various forms, but particularly on art for which the artists use programming, computer code, as a significant element of their work. We also deal with the context of such art, so some chapters consider much of what is often called digital art, but the focus remains on the generative. Much has been written about digital art in general, and we do not repeat what is easily read elsewhere except when understanding the context requires it. At the core of computing is software—computer programming and code. Perhaps rather strangely, the role of code in computer art has received relatively little attention; hence it is our primary focus. We aim to highlight important continuities as well as the many exciting differences.

A wide range of questions is discussed, for the book is a collection of chapters—most newly written for this volume—by authors with different backgrounds. Margaret Boden is a philosopher, with a special interest in creativity and in how concepts drawn from artificial intelligence (AI) can help us understand it. Ernest Edmonds is a pioneering computer artist and a professional computer scientist. Although each chapter can stand alone, they are unified by the authors’ shared interest in the history and philosophy of computer art.

In preparing the book we interviewed a range of artists and curators and had more informal conversations with many people engaged in the art world, most with a specific interest in computing. Perhaps it is not surprising that we have found significant differences among generations. Whereas older people, including many of those who form the art establishment, see computer technology as almost disturbingly new, and not infrequently question the possibility of computers contributing to art at all, young people take it for granted that computers are an integral part of almost all aspects of life, including art. Senior artists may talk about the new opportunities that the computer gives them, whereas young artists were often writing computer programs years before they ever thought about becoming an artist. For them, using a computer to make art is as natural as using paint.

Notwithstanding the comfort that the young have with computers and computer software, we still find that computer art is largely presented and viewed with suspicion in the conventional and established art gallery world. In this respect, it seems that visual art is behind music in adopting the new technologies. A good deal of modern music involves the use of computers in one respect or another. Apart from the major digital art events that are staged around the world, internationally a small number of commercial galleries specialize in computer art and a small number of public galleries or special wings or units in public galleries also present such work. Venues are certainly growing, but even when the growth is strong, more often than not the computer art is not integrated into traditional art exhibitions. So we find that the problem of computer art (is it really art?) is very much a current problem, even though the young may not agree and even though almost every film, much theater, and a great deal of music is today made with the aid of computers. When we come to chapter 14 and hear directly from artists themselves we do not find any doubts at all. However, there is often difficulty in dealing with the established visual art world.

We see, then, that the question of whether computer art should rightly be called art at all is still with us. Are the continuities that link it to older forms of art strong enough to settle that question? Or can it be settled merely by pointing out that several major art galleries have exhibited examples of this work? Is it relevant that other modern technologies (photography and film) have been welcomed by influential art critics, such as Walter Benjamin (1968), as ways of doing—and democratizing—art? Or does the use of computers compromise this field’s aesthetic credentials in ways that the use of cameras does not? If so, how? In particular, is the compromise even greater in cases of generative computer art, in which the program is left to produce the artwork by itself, with little or no input from a human being? And if not, why are so many members of the general public skeptical about the very idea of computer art? In addressing these questions we show that some of the links with older art forms are quite strong even though they may not often be noted. We also give examples of very strong public interest in interactive art, although the naming of the work as art may not always be particularly evident.

Some closely related questions concern its history. Besides reviewing how the various types of computer art developed, we say something about how they have been met—or rejected—by denizens of the mainstream art world.

Mainstream art comes in many forms: painting, sculpture, ceramics, music, dance, poetry, drama, and more. So too does computer art. But these chapters highlight the visual arts. Certainly, the philosophical points made in the more analytical chapters apply to art in general. Moreover, one chapter is largely about music, and several mention examples from other genres, such as literature. Nevertheless, most of the discussion draws on art (whether mainstream or computer based) in the form of painting or graphics.

The six chapters in part II, by Margaret Boden, are broadly philosophical in tone. The first two set the scene for what follows—by identifying distinct types of computer art and by describing how creativity has been studied (or ignored) in AI. The other four focus on links between computer art and traditional views in philosophical aesthetics. Part III—comprising six chapters by Edmonds—considers some historical questions from the point of view of someone who has been a practitioner of this genre right from its start. In addition, his chapters touch on what the experience of creating computer art is like and how it differs from straight painting or musical composition. Part IV contains a single chapter that comments on many of the questions raised in the previous chapters in a set of interviews (conducted by Edmonds) with well-known computer artists, all of whom write code. We present many of the issues discussed in the book in the artist’s voice: what it means to make art today using the media of today.

Chapter 2 provides a taxonomy of computer art. This analysis distinguishes different types within the broad genre of computer-based art and outlines some of its links with more familiar forms.

The term “computer art” is sometimes defined more narrowly. For instance, Dominic Lopes (2009) considers only the interactive examples. By contrast, the taxonomy offered here highlights the variety of approaches that can reasonably be included under this name. For example, it distinguishes generative art (G-art), computer-generated art (CG-art), interactive art (I-art and CI-art), robotic art (R-art), evolutionary art (Evo-art), computer-assisted art (CA-art), and live coding art (LC-art). As we will see, these subgenres are not mutually exclusive but sometimes overlap.

In addition, chapter 2 indicates what philosophical questions and problems are related to this or that subgenre. In this context, the most important distinction is between CA-art and CG-art.

CA-art, in which the computer is used by the human artist as an aid in the art-making process, raises no philosophical problems that do not also arise given the use of paintbrushes or chisels. That is not true of CG-art, in which a computer program is left to run by itself, with minimal or zero interference from a human being. Here, the relative autonomy of the computer threatens to usurp (aspects of) the role of the human artist. Our main interest, in all the chapters that follow, is in the generative examples. That is partly because Edmonds is himself a CG-artist and partly because CG-art presents many more philosophical puzzles than CA-art does.

Different types of CG-art, moreover, suggest different philosophical worries. So the taxonomy is followed by a sketch of the relevant philosophical landscape. Some of its peaks and valleys are considered more fully in later chapters.

Next, chapter 3, “Explaining the Ineffable,” outlines the history of AI research on creativity.

Much as people often raise philosophical doubts about whether AI systems are really intelligent, so they often doubt whether they can really be creative. Some even question whether AI can offer the appearance of creativity. Taking those questions seriously requires one to consider the research mentioned here.

AI research on creativity has blossomed in the last twenty years. But although creativity was identified very early as a key goal for the field, it was not seriously studied for several decades. Insofar as creativity was modeled at all, it was exemplified as verbal or logical problem-solving—hardly ever as music or graphics. Even scientific creativity had to wait many years before becoming an explicit focus for AI.

As it happened, the two people who were most influential in bringing art into the AI arena were both computer artists: the painter Harold Cohen and the composer David Cope. However, AI work on artistic creativity should not be confused with computer art. Often, there is no specifically artistic intention, still less any attempt to exhibit the results in galleries or concert halls. Rather, the interest lies in trying to model the types of thought processes that go on in the minds of human artists when they create. Indeed, Cohen himself, when already a highly acclaimed painter, turned to AI largely because he thought it would help him understand his own creativity. By the time of his death, in 2016, he had come to believe that human and AI creativity are fundamentally different. Even so, he felt that his computer-based work had helped him appreciate this.

That point leads to the key message of the chapter, which is that—contrary to the still-influential Romantic myths—creativity is a psychological capacity that can be understood by science. Specifically, it involves computational processes comparable to those defined in AI.

Three types of psychological mechanisms can produce new thoughts: combining familiar ideas in unfamiliar ways, exploring accepted styles of thinking or making so as to generate novel structures of the relevant kind, and transforming the style to come up with structures of a novel kind (Boden 2004). Each of these three kinds of creativity can be seen in the work of mainstream human artists and in computer art too. And each, as explained in chapter 3, has been modeled in—and clarified by—AI research.

There is an important caveat here, which applies also to several other chapters (e.g., chapter 5). Whether a computer artwork or AI program can appear to be creative is a factual question—to which the answer is often yes. Whether it is really creative is a philosophical question, not a factual one. According to many philosophers, the answer must be no. Irrespective of its actual performance, they say, no computer can provide real creativity.

The reasons advanced for this view are diverse and involve some of the deepest problems in philosophy—including the nature of meaning, consciousness, and moral community. They will be ignored in this book (unless mentioned by any of the artists interviewed in chapter 14), whose focus is on what computer art can achieve in fact. For the record, however, questions about whether computers can be really creative cannot be given a definitive answer, precisely because the philosophical issues involved are so deeply disputed (Boden 2004, chap. 11; Boden 2014).

Chapter 4 considers the broadly accepted assumption that art appreciation normally involves recognition of the making skills concerned. This assumption dates from the predigital age. Indeed, John Ruskin—whose nineteenth-century views on aesthetics are still hugely influential—defined art largely in terms of the human skills that it employs (see chapter 7). Even Robin Collingwood, who (as explained in chapter 6) drew a sharp distinction between art and skill, allowed that we often do admire the skill of the person who made the work of art and that we typically value the artifact all the more accordingly. In short, it is nowadays broadly accepted as common sense that art appreciation normally involves recognition of the making skills concerned.

This fact, with its implications for the appreciation of computer art, is the theme of chapter 4. The appreciation of computer art is even more dependent on knowledge of the making process than is the case for traditional (or anyway, non-Constructivist) forms of art.

CG-artists not only use the computer but celebrate its potential as a medium. Similarly, oil painters may celebrate the colors and textures of their paints and jazz clarinetists the potential of their instrument. Moreover, the different forms of CG-art use the computer in very different ways. It is not merely that individual CG-artworks look (or sound) very different from each other but rather that they can exhibit very different aspects of the indefinitely varied generative potential of the digital computer.

That is all very well, and the multifariousness offered by computers is a genuine cause for celebration. But it raises a problem for the art audience, most of whom have never done any programming and have little or no intuitive sense of what it involves (see chapter 10). Even people who (like most members of industrial societies) use computers very frequently indeed may have no such understanding, because they experience computer programs ready-made. The complex skills that underlie user-friendly applications such as spreadsheets and the commercial search engine are invisible, even unsuspected, by their users. As the computer artist Harold Cohen has put it (personal communication), most of us are consumers of programs, not producers of them. Thus, we are not well placed to appreciate, criticize, or even compare individual works of CG-art with respect to the programming achievements involved in them.

Up to a point, computer artists and curators can arrange to make these technical achievements more accessible. Several methods of doing this are suggested in chapter 4. But even in a culture whose children are introduced not only to computers but also to simple programming, the extent to which this can be done is strictly limited. The unfamiliar, elitist skills of computer artists are even less imaginable to the general public than the specialist skills of traditional artists are. We can all remember using a paintbrush, and we can all imagine forging wrought iron. Whether we can picture how to program computers to generate interesting graphics or music or to operate robots is another matter entirely.

The conclusion of chapter 4, then, is decidedly downbeat. Computer art can be enjoyed by anyone. Some examples can engage virtually everyone. But it can never be as fully appreciated as the more familiar types of art. It will always carry an aura of magic: awesome, perhaps, but largely unintelligible.

Chapter 5 explores the potential for creativity in one particular form of computer art, evolutionary art (Evo-art in the taxonomy of chapter 2). The key question is posed by its title: “Can Evolutionary Art Provide Radical Novelty?” In other words, is the creativity of Evo-art forever limited in a way that human creativity is not?

As remarked earlier (and in chapter 3), one of the three types of creativity is transformational creativity. In this, some initial style is transformed with the result that structures of a novel kind can arise. In general, the more radical the stylistic transformation, the more it is valued. Although if the change is too radical, it may be rejected. If computer art cannot offer us transformational creativity, then it is less worthy of our appreciation than human art is.

The computer artists who practice Evo-art hope, by using genetic algorithms, to effect stylistic mutations that vary the resulting artifacts (e.g., images) in highly unpredictable ways, perhaps even jumping out of one style into another. If the second style is not just new but radically new, so much the better.

But is that, in principle, possible? Many critics argue that it is not, because the computational potential of any program—including one equipped with rule-changing genetic algorithms—is fixed. It may be huge, even infinite (a point that is important in chapter 4). But there are always some conceivable structures that are excluded. Evo-art, on this view, simply cannot provide truly radical transformations.

Such transformations do sometimes happen, but only (so this objection runs) outside cyberspace. For instance, they occur in biological phylogenesis, in which fundamentally new organs (e.g., light sensors) can arise—and once arisen, can then evolve into highly differentiated forms. The explanation given for this is that biological evolution takes advantage of the physical contingencies in the environment. Because cyberspace involves no such contingencies, no opportunity for an evolving computational system to engage with the physical world, this spur to stylistic novelty is forever absent.

Robots, however, do not inhabit cyberspace. They not only take in perceptual data from the real world but are material objects moving around in a material, often cluttered, environment. Chapter 5 describes some work in evolutionary robotics wherein radical (anatomical or perceptual) transformations have happened as a result of previously unconsidered physical conditions, even accidents. It also mentions ongoing research by a leading computer artist who hopes to evolve drawing robots that, as a result of physical interactions, generate fundamental, but aesthetically valuable, stylistic changes.

Biology is not the only noncomputerized forum for radical transformation: history-book artistic (and scientific) creativity in human minds and societies counts too. But here, physical contingencies are only rarely relevant. Usually, the relevant environment is psychological or cultural. These factors can be modeled in computer systems. And they do not have to be specifically provided beforehand by the programmer: some (non-Evo) computer artists already allow their programs to take current data—words or images—at random from the Internet. One such example is The Living Room, an installation built by Christa Sommerer and Laurent Mignonneau.

In short, the message of chapter 5 is that Evo-art could provide radical transformations. Some of them would doubtless be rejected by us, but that is another matter. It could do this even if its artifacts were situated not in the real world but in cyberspace. The fact (repeatedly cited by skeptics about computer art in general) that a computer program can do only what it is set up to do does not exclude this possibility. Certainly, if an Evo-art installation were to benefit from the cultural contingencies available on the Internet, the web’s contents would have to be accessed by some instruction in the program. But that is just to say that creativity—even radically transformational creativity—does not happen by magic. Evo-art has an open, and intriguing, future.

The remainder of part II turns to key topics in philosophical aesthetics. Many competing accounts of art have been offered by philosophers, including various modernist and postmodernist positions—not to mention the still-influential views of Plato, Aristotle, and Kant. This rich body of literature is touched on in several chapters but is foregrounded in chapters 6 and 7.

Fashions in theoretical aesthetics change. Constructivism, for instance—which led most of the pioneers of computer art to experiment with the new medium (see chapters 9 and 14)—stressed aspects of the art-making process, and of the finished artwork, that had not been valued before the early twentieth century.

The conceptual art of the 1960s went even further, for it challenged most of the then-accepted assumptions about what counts as art and even what sort of thing an artwork can be (Boden 2007). According to that viewpoint, many computer artworks are not art at all—or anyway, not of an interesting kind. Even CG-art, wherein the core of the artwork is often held to be the process of art making (see chapter 15), does not necessarily qualify. It does not normally offer a conceptual outrage—a shocking conceptual combination (see Boden 2007)—to the audience. Or rather, it does not offer any specific shock to the audience—as when the conceptual artist Walter de Maria, for instance, buried a perfectly crafted and very expensive object (a two-inch-thick steel cylinder, one kilometer long) so as to make it invisible.

For some audiences, however, CG-art in general is irredeemably suspect, just because its artworks are made by computers not by human hands. Are those audiences wrong? Perhaps so, but this cannot be taken for granted: the point needs arguing.

In particular, one needs to consider the aesthetic theories—or rather, the largely unexamined philosophical assumptions—that are most widely held by the general public (and within the art world; see chapter 8). These concern the role of emotion in art and the role of the art maker as a human individual—the topics of chapters 6 and 7.

The theme of chapter 6 is the relationship between computer art and Collingwood’s theory of art. Why Collingwood? Well, not because either of us accepts his views on art; we do not. Rather, because it is a common—perhaps the most common—criticism of computer art that there can be no such thing, because of the importance of emotions in art. Because computers cannot have emotions, it is said, they cannot be artists either. And Collingwood (1958) is among the most influential of the many philosophers who have claimed that art necessarily involves emotion.

Prima facie, then, his account challenges those who claim that CG-art is, truly, art. Indeed, his ideas underlie much of the hostility to computer-generated art that is found in our culture—not least its uneasy position vis à vis the art world (see chapter 8). If it could be shown that CG-art is not quite as inimical to Collingwood’s theory as it appears to be at first sight, the widespread resistance to it would be weakened.

Chapter 6 argues that his approach does indeed threaten the status of computer art as art. But that is not because of any reference to emotions in general. There is no reason in principle why computer art could not express or arouse emotion, and some of it—especially computer music—actually does. Rather, it is Collingwood’s extreme particularism that prevents CG-art from being (on his view) art proper.

He defined art in terms of the expression—and the detailed construction—of some highly particular emotion in the artist’s mind, with an equivalent process occurring also in the mind of the audience. He even said that to regard a novel as depicting “the feelings of women, or bus-drivers, or homosexuals” is not to treat it as art at all but rather as psychology (Collingwood 1958). As a work of art, it must construct or express the particular experiences of the individual novelist—who may happen to be a woman, a bus driver, or a homosexual but whose work cannot be regarded (qua art) as representative of the group.

Collingwood’s view goes against the spirit of most computer art. That is because a large part of the point of CG-art—especially when the originating human artist has a relatively hands-off role—is to explore, and to exhibit, a certain range of possibilities. Each CG-artwork is unique, to be sure. But those generated by a given computer program are unified—both perceptually and historically—by it. The audience is expected to notice that unity and to appreciate it. Whether this appreciative response requires knowledge of computing, as well as sensitivity to observable stylistic variation, is discussed in chapter 4.

Given this conflict, the question arises whether we should say “So much the worse for computer art!” or rather “So much the worse for Collingwood!” The alternative chosen in chapter 6 is the latter. For besides undermining the rationale of CG-art, Collingwood’s position also conflicts with common intuitions about traditional art.

Shakespeare’s plays, for example, are not art proper for Collingwood—although his sonnets may be. They do not, except perhaps occasionally, construct and express some highly specific emotional experience of Shakespeare himself. On the contrary, they often aim at depicting the emotions of some general class of people—monarchs, for instance (“Uneasy lies the head that wears a crown”). Similarly, a painting that aims to make a political point—such as Pablo Picasso’s Guernica, or Gustave Courbet’s Peasants of Flagey Returning from the Fair—is seen by Collingwood as propaganda rather than art unless it happens to capture the painter’s particular experience and arouses an equivalent emotion in the audience.

The last chapter in part II, like chapter 6, engages with a writer in traditional aesthetics: Ruskin. His approach is prima facie even more inimical to computer art than Collingwood’s is—which makes it even more interesting as a philosophical challenge to the field. That is primarily because it stressed the personal individuality, and the creative freedom, of the human artist-craftsman.

Much as we do not share Collingwood’s theory of art, we do not share Ruskin’s either. But we do admit to having some sympathy with his view, for it has entered deeply into our culture.

In particular, Ruskin’s chapter on Gothic architecture in The Stones of Venice (1853) was hugely influential. It molded people’s ideas about art in general, remaining “the greatest influence on taste” until the early twentieth-century doctrine of pure form (Clark 1949). For instance, it insisted on the aesthetic importance of the artist’s cultural (e.g., religious) motivation and on his chosen methods of hand-working. By inspiring William Morris, it led indirectly to the Arts and Crafts movement. And, again largely via Morris, it also affected the development of socialism in England.

The ideas expressed in that famous essay are still widely current. Many of us today share, or anyway sympathize with, Ruskin’s stress on the individuality and fallibility of the working artist and on the importance of the cultural interests that imbue the artwork. Indeed, the suspicion of computer art in the mainstream art world today is largely grounded in aesthetic values inherited from him. Ruskin’s essay was an impassioned complaint against the industrial age and especially against the effects of machine technology on the workman’s role in art and craft.

High-tech computer art is very different from the medieval masonry so admired by Ruskin. There is no question but that it fails to satisfy his criteria for superior art. However, chapter 7 argues that this new approach is not quite so beyond Ruskin’s pale as one might think.

For example, the individuality of the originating artist still shines through, even given the mediation of the computer (see also chapter 10). Ideological values, such as religion and politics, are not typically involved in CG-art—although, in principle, they could be. They are often prominent in other forms of what chapter 2 terms electronic art, or Ele-art: video-art, for instance. But if CG-art normally eschews ideology, it allows plenty of room for infelicities and failures on the part of the originating human artist. And computer-assisted art (CA-art), of course, leaves even more room for mistakes by the human individual. It may seem strange to laud mistakes and infelicities as criteria of art: surely they are to be avoided if at all possible? That was not Ruskin’s view. And we need to understand his legacy to understand much of the resistance (especially in England) to this new way of art making.

Ruskin identified six principles of the Gothic, which he used to evaluate all forms of art. More accurately, perhaps, he identified twelve such principles: two parallel sets, concerning the form or content of the artwork itself and the moral climate expressed in it by the artist. We will see that computer art can satisfy his six, or twelve, principles only up to a point. To the extent that it fails to do so, computer art would be regarded by him as an inferior type of art.

Or rather, he would have regarded it as inferior if he had accepted it as being art in the first place. The final section of chapter 7 shows that he would not have done that. And his reasons for not classifying computer art as art are shared by many people today, some of whom even cite Ruskin in making their argument. In brief, his nineteenth-century ideas remain very relevant to our twenty-first-century concerns.

The six chapters in part III touch on many of the philosophical points already mentioned. But because of Edmonds’s long experience as a computer artist, they are more practical and historical in tone.

Chapter 8 explores the sociological relations between computer art and the art world and considers the changes in these relations that have taken place over the last half century. The attitudes of the major art galleries, for instance, are relevant—and, as we will see, deeply ambivalent.

Sociology and history, in these respects, are closely linked to philosophy. One influential account of what it is for something to count as art is that it be accepted as such by the art world (Danto 1964, 1981; Dickie 1974, 1997). On that view, what counts as art depends on the collective response of the socioeconomic network comprising artists, critics, museum curators, gallery owners, dealers, and collectors—and, where relevant, publishers too (Becker 1986, 1995).

Acceptance or rejection by the art world, of course, rests largely in the application of a range of familiar aesthetic criteria (autonomy and authenticity, for instance; see Boden 2010b, 2010c). But it depends also on judgments, and even prejudices, of other types.

For example, one historian of computer art has remarked that, although “artists throughout history have appropriated new technology and manipulated it to serve their own purposes,” the technology used in computer art “has tended to be viewed by the art world as deriving from military applications” and that “perhaps this is [one] reason why the art world has been reluctant to embrace it” (Mason 2008, xii). A related source of skepticism is the role played by computer technology in modern methods of mass production (see chapter 14).

Yet another prejudice against this new art form arises (as remarked previously, in relation to chapter 6) from the widespread assumptions that art and art making must involve emotion and that this can have nothing whatsoever to do with computers.

Even social snobbery has played a part. Computer art in Britain was pioneered predominantly in the newly formed polytechnics (although the Slade School of Fine Art is an important exception), which did not endear it to artists and scholars identified with the major art schools and the universities (Mason 2008, chaps. 7 and 8).

Not least, of course, is the always-present arts-science divide. Despite having been included in the first international exhibition of computer art in 1967, and despite having been nominated for an Arts Council of Great Britain award by the chairman of the Institute of Contemporary Arts, an early kinetic sculpture by Edward Ihnatowicz was rejected by the relevant committee because “the Committee took the view that … there was not really enough evidence of its prospects of success as a work of art as distinct from a piece of electronic ingenuity” (Mason 2008, 96).

So in chapter 8, Edmonds recalls some of his experiences vis à vis the art world over the last fifty years and reports on interviews with individuals situated in different parts of this variegated curatorial and commercial world. Some very recent and—to both of us—inappropriate responses to computer art are discussed. Encouragingly, at least a few of the experts had positive and constructive visions to share.

The next two chapters are not only historical but also partly autobiographical. In chapter 9, “Formal Ways of Making Art: Code as an Answer to a Dream,” Edmonds provides a historical context for computer art and describes how, as a young man and already a committed painter, he welcomed the new technology of computing as a way of advancing certain artistic concerns that had been informing his work.

The new concepts and capabilities that the computer introduced were very appropriate in a particular art-making tradition that preceded the invention of computers. It can be seen in the discussions of the General Working Group of Objective Analysis in 1921 in Moscow. This was the beginning of the art movement generally known as Constructivism.

The group drew a distinction between composition and construction in making their art. Briefly, composition was seen to be about arranging perceptible forms according to relationship rules, and construction was about making a work according to a plan for its production. The key point was the introduction of the notion of making a visual artwork according to a plan, or underlying structure, that took creative precedence over the final object. Satisfying the goals or constraints guiding the art-making process was considered to be more important than concentrating on the visual (or audible) composition of an art object.

In 1921, of course, the plans were executed by the artists themselves, but in CG-art they are executed by computers. Constructivist work continued through Systems group art and other groupings, leading on to generative art. What art historians term “generative art” includes any approach wherein the realization of an underlying plan, or abstract structure, is highlighted. Full appreciation of the work therefore depends on knowledge of the creative processes involved (see chapter 4). Early examples, although rarely discussed in this context, include the dice music of Haydn and Mozart. With the arrival of computers, the generative artists found the answers to the prayers that their constructivist predecessors had been making.

Chapter 9 goes on to give the particular example of the developments in Systems art in the United Kingdom, influenced by Max Bill, for example. It explores how making art with software grew out of that movement. The UK’s Systems artists both directly and indirectly informed recent software-based art in the UK.

If programming can aid art, as declared in chapter 9, it can also count as art—or so Edmonds argues in chapter 10. Here, he draws on his long-standing personal experience of computer programming, for both artistic and nonartistic purposes, to emphasize the significance of software in art. The concerns of software-based art include a significant conceptual element. The computer enhances the artist’s ability to shape the underlying structures of artworks and art systems. In some artists’ hands the software goes well beyond the role of a manufacturing tool. It is also a thinking aid or inspiration and increasingly must be seen as a new medium in itself.

The art in software is increasingly recognized. Software in art has been underrated, however. Even many of the books on digital or computer art pay little attention to the computer code that drives and largely defines such art. That is largely because most people, including most art critics, have little or no experience of programming; to use a computer is not necessarily to program one. The artist’s challenge is software, not because it is difficult but because it is the conceptual representation of the new art.

Chapter 11, “Diversities of Interaction,” discusses the increasing concern with the active role of the audience that has been encouraged by computer art. The audience can interact with an artwork in many ways. We have long understood that perception is an active process so that even contemplating a Mark Rothko painting involves interacting with it in a certain sense. In that example, of course, there is no material change in the painting when it is looked at. But many artworks do, in fact, materially interact with their environment—for example, with shadows changing as the sun moves round a sculpture or with a breeze changing the configuration of a mobile. With the advent of computers in art, however, there has been a significant growth of interest in observable interactions with artworks.

In these cases, we can see members of the audience move in some way and can also see the artwork change as a result. Such observable interaction can be considered in the light of action-response models. In other words, the causal effects of the gallery visitor’s behavior on the artwork are relatively direct. They may even be open to intentional direction by the audience members, in which case the audience actively participates in the creation of the final artwork (see also Boden 2010a; Edmonds 2011).

A further development is based on a systems view, in which the audience and the artwork are seen as collections of systems that exchange information. Stroud Cornock and Edmonds (1973) presented the concept of seeing the artwork, the object itself, and members of the participating audience as interacting systems and suggested that the art system might be seen as the collection of all these things. Going one stage further, Sean Clark and Edmonds showed ColourNet at the CHI (Computer-Human Interaction) conference in Paris in 2013 (Clark and Edmonds 2013). This art system allowed separate, possibly mobile, artworks by others to be included within the distributed system. The version demonstrated at CHI 2013 consists of two digital artworks—the core Shaping Form component by Edmonds and the Transformations component by Sean Clark. Shaping Form is projected on a large screen. Participants access Transformations on their smartphones. Both components are able to interact with multiple users simultaneously and can work independently. However, they are also able to interact with each other by exchanging color information via the Internet and using that information to influence the colors used.

In works such as Shaping Form, an action does not necessarily produce an immediate response. It may change the internal state of the other system and so influence its behavior later in time. The causal influence from audience to artwork is no less real but is much less direct—and often unfathomable.

As for chapter 12, “Correspondences: Uniting Image and Sound,” this is the only one that focuses on or beyond nonvisual art. An important form of interaction that has developed considerably as the result of computer programming is interaction between different media. For instance, artworks that integrate visual displays with sound are common today. Such work is often linked with an interest in synesthesia. However, the link is more one of artistic inspiration than science, because the condition (color-and-sound mapping, for example) does not seem to be consistent across individuals.

The key point explored in the chapter is that the computer enables real-time dynamic linking between the generative production of sound and of image. For instance, software technology can analyze live music as it is played, detecting pitch and amplitude (for example) and feeding that information into a computer-based art system that responds to or transforms those sounds—producing music, images, or forms in many other media. A tight integration of media is thus made possible in ways that were imagined but not realized in earlier times.

Chapter 13 “Diversities of Engagement,” describes how computer-based interactive art has encouraged a concern for better understanding of audience engagement. At the simplest level, this is a measure of how long someone looks at something: maybe a painting or a sculpture or a video. In the museum world, such simple measures are used to judge how attractive an exhibit is—or perhaps more accurately, how attracted to an exhibit the members of the public are.

The concept of engagement can be elaborated, however. For instance, at least three kinds of engagement can be identified. These are attraction (the work drawing attention to itself), sustainability (the length of time that someone finds it engaging), and repeatability (the extent to which someone wants to experience the work again and again).

Artists who make digital works that take engagement into account become involved in strategies for drawing the audience in and retaining their interest in quite explicit ways. This process can be one that includes an element of audience education. The artwork can be designed so as to attract the public and interact with them in ways that develop interest and engagement. Specific aspects encourage that result. Examples of this can be seen in the curated program of interactive art exhibits and their evaluation in Beta_Space, a joint venture of the Powerhouse Museum in Sydney and the Creativity and Cognition Studios at the University of Technology, Sydney (Candy and Edmonds 2011).

The artworks that are made for the new audiences for whom these artists are aiming can be ambient. That is, they may exist in the environment without any specific signals indicating that they should be seen as art. Of course, the context itself can be seen as a strong indicator in this sense. So we might want to distinguish between such art in a museum and art in the street, for instance. Engagement and ambient art are concepts that are important in the digital art world, but they have good continuity with the nondigital world. Dynamic and interactive artwork that is an integral part of an ordinary environment, such as a meeting space, bar, or city square, and that is made by use of computer systems is growing.

Finally, part IV, consisting of only the substantial chapter 14, provides a series of conversations with eleven computer artists. Most focus primarily on the visual arts, although a few of them sometimes integrate or combine vision and music in their work. All the artists are established, even long-established, professionals in the field, although some are known only in very specialized communities. This section of the book presents the artist’s voice: the concerns of the book as seen from the point of view of the practicing artist.

In leading these conversations, Edmonds presents the artists with the key questions raised in the previous chapters. As one might expect, there was little or no comfortable consensus on many of these. On some, however, there was a significant measure of agreement. One obvious difference was closely related to age. The older, pioneer artists discovered computers and programming when already into their career and, for various reasons, became engaged in ways that extended or changed their artistic practice. At the other extreme, younger artists report having used computers, including programming them, before they dreamed of becoming artists. For them, the computer did not change their art, because it was always there.

The artists interviewed here work in the visual domain, in performance, and in sound and some work across these media. A strong theme is the importance of programming. The computer code is significant in several ways. For at least one of them, the code is the art. For several, writing it is a vital part of the thinking process. They would never make the kind of art that they do without going through the thinking process required by programming. For them, software is a way of both making art and thinking about art. One of the artists puts a significant amount of effort into constructing the technology that is used in making the art. At least two others even find the need to make their own software tools. A performance artist finds that the specification of dance movements can be inspired by and even determined by the code.

Several artists, particularly those involved in live coding art (LC-art, in the taxonomy of chapter 2), are very interested in to what extent the audience should see the code and to what extent it is desirable that they understand it. One of these is designing a programming language with the very purpose of facilitating audience understanding of the software that drives the work.

It is clear from the interviews in the final chapter that computer artists’ aims, methods, and beliefs differ significantly and also that they have often changed over the years. We hope that this book will inspire its readers to consider computer art from both historical and philosophical perspectives and, perhaps, also move their thinking on in valuable ways. We must acknowledge that our own thinking has been partly influenced by many conversations with the late Harold Cohen, who sadly died during the later stages of this book’s preparation. Stroud Cornock, who also sadly died during these late stages, inspired much of Edmonds’s work. Ernest Edmonds also thanks Linda Candy for endless valuable conversations and tireless support.

Perhaps, while considering computer art, some readers might also get involved in doing it. Or at least in hanging a CG-piece on their living room wall. Impractical? Not necessarily: some small wall-mounted interactive displays show dynamically changing images and sounds, triggered by passersby. Let us help bring computer art into the wider art world!

References

Becker, H. S. 1986. Doing Things Together: Selected Papers. Evanston, IL: Northwestern University Press.

Becker, H. S. 1995. “A New Art Form: Hypertext Fiction.” In How to Read a Hypertext, edited by M. Bernstein. Cambridge, MA: Eastgate Systems.

Benjamin, W. 1968. “The Work of Art in the Age of Mechanical Reproduction.” Publ. in German in 1936, trans. 1968. Reprinted in R. Kearney and D. Rasmussen (eds.), Continental Aesthetics: A Reader (Oxford: Blackwell), pp. 166–181.

Boden, M. A. 2004. The Creative Mind: Myths and Mechanisms. 2nd ed. London: Routledge.

Boden, M. A. 2007. “Creativity and Conceptual Art.” In The Philosophy of Conceptual Art, edited by P. Goldie and E. Schellekens, 216–237. Oxford: Oxford University Press. Reprinted in M. A. Boden, Creativity and Art: Three Roads to Surprise, chap. 5. Oxford: Oxford University Press, 2010.

Boden, M. A. 2010a. “Aesthetics and Interactive Art.” In Boden, Creativity and Art, chap. 11.

Boden, M. A. 2010b. “Authenticity and Computer Art.” In Boden, Creativity and Art, chap. 10.

Boden, M. A. 2010c. “Autonomy, Integrity, and Computer Art.” In Boden, Creativity and Art, chap. 9.

Boden, M. A. 2014. “Artificial Intelligence and Creativity: A Contradiction in Terms?” In The Philosophy of Creativity, edited by E. S. Paul and S. B. Kaufman, 224–246. Oxford: Oxford University Press.

Candy, L., and E. A. Edmonds. 2011. Interacting: Art, Research and the Creative Practitioner. Faringdon, Oxfordshire: Libri Press.

Clark, K. 1949. “Roger Fry.” In The Dictionary of National Biography 1931–1940, edited by L. G. Wickham Legg, 298–301. London: Oxford University Press.

Clark, S., and E. A. Edmonds. 2013. “ColourNet: A System of Interactive and Interacting Digital Artworks.” In CHI ’13 Extended Abstracts on Human Factors in Computing Systems. New York: ACM Press: 2911–2914.

Collingwood, R. G. 1958. The Principles of Art. New York: Oxford University Press.

Cornock, S., and E. A. Edmonds. 1973. “The Creative Process Where the Artist Is Amplified or Superseded by the Computer.” Leonardo 6:11–16.

Danto, A. 1964. “The Artworld.” Journal of Philosophy 61 (19): 571–584.

Danto, A. 1981. The Transfiguration of the Commonplace: A Philosophy of Art. Cambridge, MA: Harvard University Press.

Dickie, G. 1974. Art and the Aesthetic: An Institutional Analysis. Ithaca, NY: Cornell University Press.

Dickie, G. 1997. Art Circle: A Theory of Art. Chicago: Spectrum Press.

Edmonds, E. A. 2011. “The Art of Interaction.” Digital Creativity 21 (4): 257–264.

Lopes, D. M. 2009. A Philosophy of Computer Art. London: Routledge.

Mason, C. 2008. A Computer in the Art Room: The Origins of British Computer Arts 1950–80. Hindringham: JJG Publishing.

Ruskin, J. 1853. “On the Nature of Gothic Architecture: And Herein of the True Functions of the Workman in Art.” In The Stones of Venice, chap. 6. Vol. 2 of The Sea-Stories. London: Smith Elder. (Published separately, with an introduction by William Morris, in 1854, by Smith Elder.)