Can computers be considered creative?

“Creativity”, according to Margaret Boden, “can be defined as the ability of generate novel, and valuable, ideas.” (Boden) The topic of computer creativity has become increasingly topical, as computers have begun to show themselves increasingly capable of generating novel outcomes, if not necessarily ideas. There has been a recent spate of books addressing computer creativity. (Miller, du Sautoy) Within architectural circles, too, there has been considerable interest in this topic (Bolojan, Manovich). Is the reason why we often call computers ‘creative’ that we have a tendency to anthropomorphise technology, and project human attributes on to it? What about creativity itself? Indeed the whole notion of creativity and the so-called ‘creative economy’ have been called into question. (Mould, Manovich) Others, such as Rochelle King, Elizabeth Churchill and Caitlin Tan have even claimed that creativity is ‘something of a myth’. (King, Churchill, Tan) Or might creativity be alive and well, and being enhanced and reimagined thanks to developments in computation?

Initial assumptions, for sure, were that computers could never be truly creative. For example, Ada Lovelace, daughter of the poet Lord Byron, and herself no stranger to creativity, predicted that the Analytical Machine, an early proto computer on which she collaborated with Charles Babbage, would never be able to produce anything beyond what it was programmed to do: ‘The Analytical Machine has no pretensions whatsoever to originate anything. It can do whatever we know how to order it to perform. It can follow analysis, but it has no power of anticipating any analytical revelation of truths. Its province is to assist us in making available what we are already acquainted with.’ (Fuegi, Francis). More recently, Japanese computation expert, Makoto Sei Watanabe, reinforced this view, ‘Machines are better than people at solving complex problems with many intertwined conditions. In that realm, people are no match for machines. But people are the only ones who can create an image that does not yet exist. Machines do not have dreams.’ (Watanabe).

It has now become increasingly apparent, however, that computers can indeed ‘synthesise’ or generate novel images using neural networks. The assumption that they can be creative, moreover, is written into the title of the TED talk by Blaise Agüera y Arcas, ‘How Computers are Learning to be Creative’. (Agüera y Arcas) Here Agüera y Arcas outlines early explorations into inverting the network and using neural networks to generate artworks through DeepDream and other techniques.Footnote 1 Since then Generative Adversarial Networks (GANs) have become a popular mechanism for synthesising images in a more controlled fashion. (Leach 2022, 26-31) More recently still, DALL. E 2 and MidJourney have begun to generate some astonishing images that show how the domain of AI continues to develop at an exponential rate.Footnote 2

Arguably, however, the question of creativity might be better addressed through the theoretical issues raised by one particular match of Go, that has proved to be one of the most iconic moments in the history of AI. The match alerted humankind – especially Go playing nations – to the extraordinary potential of AI. The match led to the famous ‘Sputnik’ moment, as Kai-Fu Lee has termed it, when China woke up to the extraordinary potential of AI: ‘Overnight, China plunged into an artificial intelligence fever.’ (Lee, 3)Footnote 3 I would argue that the match also offers some fascinating insights into the nature of creativity itself.

This article, then, seeks to interrogate the nature of creativity through the insights raised by this match. The premise is that AI can potentially offer us a mirror in which to understand what it is to be human. As Hiroshi Ishiguro notes in the context of robots, ‘The robot is a kind of mirror that reflects humanity and by creating intelligent robots we can open up new opportunities to contemplate what it means to be human.’ (Ishiguro, 179) Could the same be said of AI? And might AI even help us to understand the nature of human creativity?

AlphaGo versus Lee Sedol

In March 2016 a match of Go was staged in Seoul, Korea, between AlphaGo, a deep learning computer programme developed by DeepMind of London, and Korean 9-Dan professional Go player, Lee Sedol, one of the greatest Go players of all time. The match consisted of five games, with most Go experts – including Lee himself – predicting that AlphaGo would be easily beaten. However, much to their surprise, AlphaGo won the first game and went on to win the match by four games to one. Yet what was most remarkable about the match was not the fact that AlphaGo beat Lee Sedol, but the manner in which it beat him.

Game 2 proved to be the turning point. After Game 1, Lee was surprised, but after Game 2 he was lost for words: ‘Yesterday, I was surprised. But today I am speechless. If you look at the way the game was played, I admit, it was a very clear loss on my part. From the very beginning of the game, there was not a moment in time when I felt that I was leading.’ (Metz) The biggest talking point of the whole match was a remarkable move played by AlphaGo in this game: ‘In Game 2, Lee exhibits a different style, attempting to play more cautiously. He waits for any opening that he can exploit, but AlphaGo continues to surprise. At move 37, AlphaGo plays an unexpected move, what’s called a ‘shoulder hit’ on the upper right side of the board. This move in this position is unseen in professional games, but its cleverness is immediately apparent.’ (Moyer) Hassabis goes even further: ‘Anyone can play an original move on a Go board by simply playing randomly. Yet a move can only be considered truly creative if it’s also effective. In that sense, Move 37’s decisive role in game two represents a move of exquisite computational ingenuity that not only changed the game of Go forever, but also came to symbolize the enormous creative potential of AI.’ (Hassabis, Hui, 84).

European Go champion, Fan Hui, also recognised the ‘creativity’ of the move – Move 37, as it has become known: ‘When AlphaGo chose that move, I assumed that it had made a mistake. I immediately looked to see Lee’s reaction. At first, he seemed to smile – as though he too thought it had made a mistake – but as the minutes rolled by it was clear that he was starting to realise its brilliance. In fact, after the match, he said that when he saw this move he finally realised that AlphaGo was creative.’ (Hassabis, Hui, 89) Interestingly, Hui describes the move as also being ‘beautiful’, ‘I’ve never seen a human play this move. So beautiful.’ (Moyer) Lee himself also describes the move as being both creative and beautiful: ‘I thought that AlphaGo was based on probability calculation and that it was merely a machine. But when I saw this move, I changed my mind. Surely AlphaGo is creative. This move was really creative and beautiful.’ (Hassabis).

Leading Go players commenting on the match initially thought that Move 37 was a mistake. (Hassabis) Move 37, however, was no mistake, and it paved the way for victory, as 100 moves later that stone became connected with a series of other stones on the other side of the board. It was as though AlphaGo could operate not just one move at a time, but could calculate 100 moves ahead – a feat that no human being could ever hope to match. In fact it was only one of a series of moves that have been referred to as ‘slack moves’, because their strategic brilliance did not become apparent until several moves later. Not only did these ‘slack moves’ effectively change the game of Go forever, challenging accepted approaches to playing the game, but they also showed that AI could demonstrate a level of ‘creativity’ well beyond human creativity. As Lee comments: ‘AlphaGo showed us that moves humans may have thought are creative, were actually conventional.’ (Kohs) Indeed, we could even say that it demonstrated the limits of the human imagination, as no one was even capable of even recognising the strategic brilliance of the move until many moves later. We could also say, however, that the AlphaGo match illustrates the limits of what we call ‘creative’. Clearly ‘creativity’ exists only in so far as we can recognise it. This, surely, is the crucial point.

Architectural implications

What, then, are we to make of all this, from an architectural perspective? First of all, and most obviously, we can see parallels presenting themselves with the domain of architecture. Similar incidents have been noticed when Spacemaker AI software has been used in architectural design. It would seem that AI has the capacity to outperform architects in certain design challenges – especially those that require a strong strategic input, not unlike AlphaGo. Just as AlphaGo was able to come up with moves that no humans would have thought about, so too Spacemaker AI software used within an architectural context has been able to make design proposals that no architects would have thought about. As Håvard Haukeland observes:

‘The places where the architects thought that it would be smart to build tall buildings, and the places where they thought it would be smart to build a dense wall, all the things that they intuitively thought would be smart – because they had hundreds of projects of experience – were flipped around. Because when you get the complexity of thinking of a multi-objective organizational problem… you are really not able to see the patterns that a computer can find. So what happened was that the computer was able to find a pattern as to how to solve that site that you would never come up with yourself.’ (Haukeland).

We should be careful, however, to distinguish between the different modalities of creativity found in architecture, and the kind of creativity displayed in a game of Go. There is a danger in promoting a monolithic definition of creativity. In this sense, creativity is no different to design. There are not only many approaches to design, but also many different understandings of design. Design for an engineer or a computer scientist, for example, takes on a primarily functional character. Performance is the primary concern. Appearances are not so important. Architectural design, however, as Vitruvius and so many other commentators have pointed out, would appear to involve a combination of performance considerations and aesthetic concerns. Not only that, but in the design process itself, we often consider background strategic decisions – in the form of traffic flow calculations, environmental analysis and other issues – before we begin to really address the appearance of a building. At an urban scale we can see this articulated in the distinction between ‘urban planning’ and ‘urban design’. Urban planning is a primarily strategic operation, involving policy makers, politicians and so on. If there is any design deployed in this stage of a project, it might be limited to the design of infrastructure – roads etc. Urban design, by contrast, takes place at a later date, and involves primarily aesthetic concerns. We should therefore draw a distinction between creativity at a strategic level and creativity within the aesthetic domain.

AlphaGo is particularly strong at a strategic level. It employs a Monte Carlo tree search – along with a policy network and a value network – and is therefore exceptionally good at searching. However, we can also recognise a further potential of AI within the aesthetic domain. If algorithms can track our preferences in terms of news, books, movies and music, should they not be able to track our preferences also in terms of architectural aesthetics? Likewise if Gmail is capable of suggesting how we might complete a sentence based on the way that we have expressed ourselves in the past, might AI not be capable eventually of suggesting how we might how might complete an architectural drawing? In other words, the writing would appear to be on the wall. AI is clearly capable of outperforming human beings at strategic decisions, and will soon also be able to match human beings in terms of aesthetic choices.

AI, creativity and consciousness

Can we consider computers, however, to be genuinely creative? Was AlphaGo actually being creative, or was it simply conducting a search for the best possible solution? Or is the question of computer creativity an impossible one to answer, as Boden argues: ‘Whether a computer could ever be “really” creative is not a scientific question but a philosophical one. And it’s currently unanswerable, because it involves several highly contentious—and highly unclear—philosophical questions.’ (Boden 2009).

Memo Akten, at any rate, seems to think that computers can be creative: ‘By saying that a machine can be creative, you are not anthropomorphising the machine, but liberating it by expanding the term ‘creativity’ to go beyond humans. Creativity is not limited to people. I’m a biological machine. Humans can create art. Why not machines?’ (Akten).

Others are not so sure. Indeed, if we are to follow the thinking of Melanie Mitchell, AI cannot be said to be creative, in that it does not possess consciousness: ‘I also believe that being creative entails being able to understand and judge what one has created. In this sense of creativity, no existing computer can be said to be creative.’ (Mitchell, 272) Of course, we could argue that we human beings might not be fully conscious of an idea, when it suddenly pops up in our mind in a flash of inspiration. Indeed, Max Tegmark notes. ‘Neuroscience experiments suggest that many behaviors and brain regions are unconscious, with much of our conscious experience representing an after-the-fact summary of vastly larger amounts of unconscious information.’ (Tegmark, 315) However, even if this might be the case initially, very soon afterwards – the ‘aha’ moment – the brilliance of the idea dawns on us, as we use our consciousness to appraise and appreciate the idea. From this perspective, AI cannot be said to be creative, since AI is not capable of appreciating its ‘creativity’.

Perhaps the most famous argument about the importance of consciousness has been made by philosopher, John Searle, using his famous ‘Chinese Room’ thought experiment. (Searle) Imagine that a computer program has been designed that seems as though it is capable of understanding Chinese. The program is able to take English text and translate it into Chinese text so convincingly that it passes the ‘Turing Test’, in that no one realises that it is just a computer program. Now imagine if Searle were to find himself in a closed room equipped with an instruction manual in English that described the translation process. He would be able, surely, to process the material in a similar way. The only problem is that he does not understand Chinese. This is analogous to the question of consciousness. A computer might appear to be conscious, but it does not understand what it is doing. Searle might appear to be able to translate Chinese, but he has no knowledge of Chinese. The Chinese Room experiment can therefore be understood as a critique of the Turing Test. There is a significant difference between something appearing to do something, and actually doing it.

AlphaGo, according to this argument, might have ‘won’ the match of Go, but it cannot have been aware of the fact that it was even playing Go, because it does not possess consciousness. As such, AlphaGo might be good at processing ‘symbols’, but it has no idea what they mean. Indeed it has no more capacity to ‘think’ than a pocket calculator.

What if we were to judge AI, however, not in terms of human intelligence, but on its own terms? As such, the question of consciousness becomes irrelevant. The issue becomes what tasks AI can perform, not whether AI can display human-level consciousness. As Yuval Noah Harari notes, ‘There might be several alternative ways leading to superintelligence, only some of which pass through the straits of consciousness.’ (Harari, 314) This leads Harari to pose the important question, ‘Which of the two is really important, intelligence or consciousness?’ (Harari, 314) As far as corporations and armies are concerned, the answer is clear: ‘intelligence is mandatory, but consciousness is optional.’ (Harari, 314) Harari goes on to give the example of a taxi driver: ‘The conscious experiences of a flesh-and-blood taxi driver are infinitely richer than a self-driving car, which feels absolutely nothing.’ (Harari, 314) But as far as the passenger is concerned, the consciousness of the taxi-driver is irrelevant. Soon, Harrari notes, we will have autonomous taxis that are far more reliable than taxis driven by conscious drivers. ‘Taxis’, he comments, ‘are highly likely to go the way of horses.’ (Harari, 315) In fact Harrari notes that a Google self-driving car was once involved in an accident, when hit from behind by a sedan ‘whose careless human driver was perhaps contemplating the mysteries of the universe, instead of concentrating on the road.’ (Harari, 314) Consciousness, in other words, is not only optional; it might also become a liability.

Could we not say the same about AI generated buildings? Many architects, for example, might claim that a computer cannot display the same emotional intensity that they invest in the design process. But does it really matter whether AI is conscious or not when it generates a design? Surely, the important question is whether or not AI can produce a good design. ‘Is our goal,’ asks Daniel Bolojan, ‘to create machines that mimic human intelligence and creativity, or are we aiming to create machines that are capable of being intelligent and creative in their own right?” (Bolojan).

Architects and creativity

Are we architects as original as we think we are?

We human beings tend to be very predictable. Indeed, we need to recognise that we have a tendency to develop certain ‘signatures’. By this I mean not simply our actual hand written signatures, which remain remarkably consistent over time, but other forms of self-expression. How easy is it to recognise someone’s voice without seeing who it is, or equally to recognize them from a distance by their mode of walking, running or other forms of behavior. It is as though human beings develop habits, routines and familiar modes of operation that constrain their forms of self-expression. The same, of course, would apply to architectural design. How often can we guess the architect, when we see a building?

We might even go so far as to comment that architecture itself relies upon a ‘canon’ of permissible forms of expression. If a student were to produce a highly unusual architectural design that resembled a pineapple, for example, s/he might risk failure. However, a student producing a design that loosely resembled a building by Zaha Hadid or Frank Gehry – but importantly that is not a direct replica – would be deemed to have passed. Indeed, Ahmed Elgammal has pursued a very similar approach in producing Creative Adversarial Networks (CANS), and generating art that appears to be close enough to the ‘canons’ of artworks, but at the same time slightly different. (Elgammal) By extension, we could also imagine the possibility of AI generating novel variations, and thereby serving as some kind of muse, as Michael Hansmeyer has called it: ‘We’ve been using computers to increase our efficiency and precision. Let us view the computer as our muse, as a partner in design, and as a tool to expand our imagination.’ (Hannsmeyer) Meanwhile, even though Gehry at some stage has produced ground-breaking designs, such as his design for the Guggenheim Museum in Bilbao, nonetheless he has ended up repeating a similar version of this design in several subsequent projects, including for the Walt Disney Concert Hall in Los Angeles.

It could even be argued that copying is in fact intrinsic to human nature, and no more so than in our contemporary condition. In effect, reproduction, replication and copying have come to define our present age in what Hillel Schwartz has called the ‘Culture of the Copy’. (Schwartz) It is not just a question of following trends or fashions in a time where role modelling and theming have become endemic. Rather we need to understand that culture continually replicates itself through a form of copying. Indeed Benjamin himself notes: ‘Nature creates similarities. One need only think of mimicry. The highest capacity for producing similarities, however, is man’s.’ (Benjamin, 332) German philosopher and a close colleague of Benjamin, Theodor Adorno, also comments: ‘The human is indissolubly linked with imitation: a human being becomes human at all by imitating other human beings.’ (Adorno, 154) Moreover, as Kendall Walton has argued, it is in the childhood games of playing roles as ‘doctors and nurses’ or ‘cops and robbers’ that human beings learn how to play the roles of managing directors, captains of industry and even architects in later life. Imitation lies at the heart of the human condition. (Walton, 11).

A series of more contemporary theoretical insights have served only to corroborate this general view. A perfect example is the notion of the ‘meme’, a relatively recent neologism. Coined by Richard Dawkins in 1976, the meme refers to the way in which ideas spread: ‘Examples of memes are tunes, ideas, catch-phrases, clothes fashions, ways of making pots or building arches.’ (Dawkins 1982) In effect, culture is propagated through imitation: ‘Just as genes propagate themselves in the gene pool by leaping from body to body via sperms or eggs, so memes propagate themselves in the meme pool by leaping from brain to brain via a process which, in the broad sense, can be called imitation.’ (Dawkins 1982) Susan Blackmore subsequently took this further: ‘Everything that is passed from person to person in this way is a meme. This includes the songs you sing and the rules you obey.’ (Blackmore, 7) The result is a form of contagion based entirely on the principle of copying. Indeed, the very way in which the use of the term ‘meme’ has spread is a perfect example of its power.

What would happen, however, if we were to take the opposite approach to Akten? Instead of saying that our notion of ‘creativity’ needs to be extended to include machines, what if we were to say that human beings are perhaps not so different to machines? In other words, what if all we were doing in the design process could be understood as little more than a search?

Indeed we could argue that contemporary life is dominated increasingly by the ‘search’. Take the process of taking a photograph, for example. How do we take photos in the age of the digital photography? In the old days of manual photography we would carefully set up the camera for one ideal shot. In the age of digital photography, however, we take a burst of sample shots. We then review them, select our favourite shot and enhance it using editing techniques available on every smart phone. It is no longer a question of the expert photographer setting up the ideal shot, but rather of the informed user taking an array of sample photos and then selecting and enhancing the best one.

Indeed these days it is more than likely that the first step in accessing information is the ‘search’. Ask a student to write an essay, and the chances are that s/he will start by conducting a search based on a few key words associated with the subject of that essay. The astonishing development in the speed and efficiency of search engines over the past few years has meant that within a fraction of a second a search can be conducted of all accessible online information. Furthermore, many analog publications have now been scanned so that access to physical books is not as important, while most journals are now accessible online through digital libraries of academic journals, such as JSTOR.Footnote 4 Once the Google Books project to scan all known volumes has been completed, we will have an even more extensive repository of information. As such, the traditional method of ‘browsing’ through bookshelves will have become if not totally redundant, then at least downgraded to an academic recreational pastime. While it might be possible to conduct a physical search of literature by commissioning a group of scholars to check through all books in a given collection of books that – theoretically – would provide the same information as an online ‘search’, the slow speed of such an operation would make such an exercise prohibitive. And it is precisely the speed of an online search that has begun to afford techniques of knowledge retrieval that are effectively changing the epistemological structure of operations within the knowledge industries.

As such, the ‘search’ becomes the first step in most disciplinary fields. Set within an architectural perspective this is having a major impact on logic of design itself. It shifts the emphasis from the ‘creativity’ of the design process towards the rigor of the search itself. The assumption here, as Kostas Terzidis has noted, is that all possible forms already exist, and it is simply a question of finding them.Footnote 5 (Terzidis 2006, 11; Terzidis 2014, 85) Indeed – if all the potential solutions are already out there, and it is simply a question of searching for them – this potentially undermines the whole notion of the ‘design genius’. Further, if there is any creativity in that process it must surely lie in the creativity of the search itself. As such, the designer emerges less as the demiurgic top down controller, and more as someone who harnesses the productive capacity of informational processes. As I have noted previously: ‘Of the many potentialities afforded by the computer, one of the most significant is its capacity to operate as a search engine. If, then, we think through the logic of the search in the context of ‘design’, what such an approach suggests is that if all possible solutions already exist, it is simply a question of defining a set of constraints and conducting a search, and then selecting one of the many outcomes. The potential implications of this are far reaching. Not only does it challenge the traditional notion of the ‘genius’ of the architect/designer and the originality of the work of art, but it also suggests that if there is still any creativity in the ‘design’ process, it should lie, firstly, in defining the constraints that generate the range of possible solutions to a problem, and, secondly, in developing an effective method of filtering or evaluating them.’ (Leach 2014).

The myth of creativity

As Margaret Boden has noted, ‘Thanks in part to AI, we have already begun to understand what sort of phenomenon creativity is.’ (Boden) The key question raised by the match between AlphaGo and Lee Sedol, however, is whether human beings would even be able to recognise the creativity of AI.

Alan Turing seemed to acknowledge this, when he commented on the potential creativity of computers: ‘We have to have some experience with the machine before we really know its capabilities. It may take years before we settle down to the new possibilities, but I do not see why it should not enter any one of the fields normally covered by the human intellect, and eventually compete on equal terms. I do not think that you can even draw the line about sonnets, though the comparison is perhaps a little bit unfair, because a sonnet written by a machine will be better appreciated by another machine.’ (Turing) His final comment – that a machine could best appreciate the output of another machine – is an interesting observation. Of course, we could question whether a machine could ever actually ‘appreciate’ a sonnet, since it does not possess consciousness, but the point still stands.

It would seem that machines can reach a level of ‘creativity’ that far exceeds ‘human creativity’, much as a dog can sense a range of sounds and smells far beyond the range that a human can. This raises the interest issue – not so dissimilar to the ‘Dunning-Kruger Effect’ – that the dumb do not know how dumb they are. We could even extend this to infer that the intelligent do not know how unintelligent they are, or – perhaps more appropriate in this context – the creative do not know how uncreative they are. Put simply, there are levels of ‘creativity’ that we simply cannot grasp. ‘Creativity’ is nothing, unless we can recognise it.

One way to think about this significant issue is to reconsider Lee’s comment about AlphaGo cited above: ‘AlphaGo showed us that moves humans may have thought are creative, were actually conventional.’ (Kohs) The important word here is ‘thought’. In other words, Lee is implying that creativity is simply a question of perceived creativity.

What becomes clear is that our understanding of creativity is somewhat subjective. Margaret Boden has attempted to define creativity in terms of three types of outcomes list, ‘combinatorial’, ‘exploratory’ and ‘transformational’, as though creativity could be grasped in such objective terms. (Boden) Meanwhile, Demis Hassabis has attempted to outline three ‘levels’ of creativity – ‘intensive’, ‘extensive’ and true ‘invention’. (Hassabis) But can creativity be judged in such as objective way? Who is in a position to even judge creativity?

François Pachet, believes that it makes no sense for creativity to be defined in objective terms, in that it can only be understood subjectively. (Pachet, 26) And what if a creative act is not recognised immediately? And how are we to appraise the creativity of Move 37, if the creativity of that move was not even appreciated until much later? Certainly, as Arthur Miller notes, it is difficult to come up with a clear, objective description of creativity, when the creativity of some artists, such as Vincent Van Gogh, was not recognised in their own lifetime. (Miller, 26) Indeed, the art world is dependent on curators who validate artworks through a form of discourse, so that art itself could be understood as a form of discourse in which artworks are inscribed. As such, artistic creativity is dependent on the discourse that validates it. Presumably, what is true for the art world should also apply to the domain of architecture.

Could we even compare creativity to beauty, as various commentators have done in relation to the AlphaGo match? Indeed, David Hume believed that beauty is highly subjective, “Beauty is no quality in things themselves: It exists merely in the mind which contemplates them; and each mind perceives a different beauty.” (Hume) Could not creativity – like beauty – lie in the eye of the beholder? Or might it also exist in the mind of the creative individual?

Let us return to the example of Searle’s Chinese Room experiment. What Searle was arguing was that there is a difference between perceived consciousness and actual consciousness. For Searle, the Turing Test is inadequate, because it only offers the potential of a computer appearing to be human. It does not test whether the computer is actually human. This surely is the crucial point. Could we not say the same about creativity? A computer might appear to be creative, but can it be truly creative? Certainly not, if we follow Mitchell’s logic. But could we not use the same experiment to argue that human beings might also only appear to be creative? What would happen, however, if we discovered that in reality we are just following a straightforward search and synthesis process – much like AI?

Could ‘creativity’, then, not be viewed in a similar way to magic? Let us be clear – there is no such thing as magic. For magicians do not perform magic. They simply conceal the processes that actual happen, so that the audience comes to attribute them to magic: ‘Like the conjurer’s trick, where the magician conceals the true devices at work, so as to fool the audience into attributing them to magic, so technology, in effacing itself, invites us to believe in its magical potential.’ (Leach 1999) Is ‘creativity’ simply a convenient term used to describe a process which – like magic – we do not fully understand? And is there not also a risk, when dealing with advanced technology (whose operations we might not fully comprehend), of mistaking it for magic? This brings us to the famous quote of Arthur C Clarke, ‘Any sufficiently advanced technology is indistinguishable from magic.’ (Clarke, 14, 21, 36).

As Margaret Boden has commented, there is nothing magical about creativity. (Boden) But could we not push the connection even further? If there is nothing ‘magical’ about magic, is there anything ‘creative’ about creativity?

Is creativity, then, no more than the appearance of creativity, just as magic is merely the appearance of magic? Is creativity, then, really no more than perceived creativity?

In short, do we not therefore need to question our understanding of human creativity? Might there even be no such thing as creativity?

Is creativity, then, no more than a myth?

Neil Leach.