The new field of ‘computational creativity’(CC) (also known as ‘artificial’ or ‘algorithmic creativity’, ‘creative computation’ or ‘creative AI’) attracts growing popularity with the recent advances of deep learning and other cutting-edge forms of machine learning algorithms. During the past twenty years CC has grown into a discipline of its own, part of the domain of artificial intelligence, which explores ‘the capacity of machines to both generate and evaluate novel outputs that would, if produced by a human, be considered creative’ (Veale and Cardoso, 2019: 2). The focus of computer scientists is to/systematise/,/formalise/, and thus/automatise/what has once been deemed a rationally inexplicable and almost supernatural capacity of some human individuals to generate new ideas and artistic masterpieces. In their attempts to mathematicize creative behaviour CC researchers reduce creativity to specific processes, algorithms and knowledge structures. At stake here is to program the least programmable, which is the freedom to deviate from the program and from predetermined rules. The goal is to design machines that produce outputs that are novel, surprising and have value – all these judged by human users. Novelty, surprise and value are the features of creativity advanced by Margaret Boden (2004) along with her classification of the three types of creativity: exploratory, combinational, and transformational.
The underlying assumptions about the concept of creativity, implied in the technical formulations above have evoked the critical responses of various humanities researchers, who have pointed out the reductiveness and intrinsic contradictions in these definitions from aesthetic, philosophical, and historical perspectives (see Still and d’Inverno 2016, Du Sautoy 2019, Fazi 2019, Zylinska 2020, Stephensen 2021). Some of the researchers emphasise the historical richness and ambiguity of the concept of creativity itself (Still and d’Inverno 2016) and, hence, its ideological bias and political consequences (Stephensen 2021), others insist on the intrinsic potential of algorithms for creative expression according to their own nature (Fazi 2018), thus denouncing the flaws of the ‘imitation game’ (Turing) and summon us to recognise machine and other forms of intelligence and perception as co-creative agencies (Zylinska 2020).
In light of these critiques we would like to pose the question of automated creativity from at least three different perspectives: the human-centred, the machine-centred, and the systemic.
Creativity is the last resort believed to be impenetrable to automation. When fortress after fortress of ‘uniquely human’ cognitive capacities – memory, reasoning, learning, intentionality, complex pattern recognition, symbol manipulation, prognosis/prediction – fall one after another to the computational logic of algorithmic automation, creative imagination is still believed to symbolise every bit of what means to be an embodied human being. Creativity as it was advanced by Kant’s theory of the genius is radically opposed to imitation and pre-programmed rule following. Illuminated by this understanding, computer scientists’ definition of creativity formulated in the footsteps of Turing’s ‘imitation game’ seems a contradiction in terms. On the other hand, the proposed ‘mechanisation’ of the creative process calls for more nuanced rethinking of the dynamics between imitation and originality, and has to be considered as a potential challenge of the very notion of individual authorship and human exceptionalism.
The second direction questions the notion of automation itself. Thinking about automated creativity we imagine the mechanical automatons of Descartes and nineteenth century industrial machines, their operation fully controlled by predetermined procedures. However, as Yuk Hui (2019) and other researchers have shown, machines after mid-twentieth century cybernetics are becoming less and less mechanical and more and more lifelike. While deterministically driven mechanisms break if they accidentally deviate from their predetermined program, cybernetic systems are open to contingency and incorporate it as part of their operative power. Thus, instead of mindlessly repeating the same outcome over and over, cybernetic machines utilise feedback, thereby generating novelty at every recursive turn.
We could return to the question of creativity again, going this time beyond the dichotomy between human and machine. If we consider creativity as the force that brings forth the new into the world, a force, which is universal, how could we think of computation as creative? How can we reimagine the concept of computational creativity as a shifting/epistēmē/, an underlying sensibility in the sense advanced by Hui (2019, 2021), which sets the conditions of our existence in relation to the cosmos and which, consequently, demands from us to generate new syntheses of thought? From this perspective, the question of computational creativity needs to be articulated in light of algorithmic logos creating systemic conditions for structural transformation of the world, reconfiguring the bonds between space-time, objects and actions. It is a new algorithmic governmentality operating beneath the surface, which Benjamin Bratton (2015) termed ‘the stack’ – a megastructure of multiple layers of systemic organisation stretching at a planetary scale, which follows the contingencies of its own internal logic. If the current creative power of computation is rather systematizing and totalising, tunnelling the future into mere probability of the past, what forms of resistance to this closure could we think of? How can we mobilise this creative power in the opposite direction so that it generates diversification and radical openness of the future?
The discussions in this special issue are invited to go beyond the mere opposition between automation and creativity and mobilise cosmotechnical (Hui) interpretations of computational creativity from non-anthropocentric, Indigenous, Afrofuturist, post-communist, East Asian, techno- feminist and other embodied localities, cultural intuitions and technogeneses.
Possible topics could include (but are not limited to) the following:
- Creativity and automation
- Creativity and cosmotechnics (Hui)
- Automation and de-automation
- Computational aesthesis (Fazi)
- Computational creativity and deindividuation of authorship
- Computational creativity and systematization
- Planetary scale (Bratton) creativity, the noosphere (de Chardin)
- Computational logic and aesthetic sensitivity
- Techno-logos, instrumentality, creativity
- Technological innovation, scientific discovery and artistic
creativity/Tertiary protention/(Hui) and creativity - Creativity and recursivity
- Creativity and the ‘imitation game’ (Turing)
Confirmed contributions by Yuk Hui and Anna Longo.
Submitted papers should not exceed 8,000 words (including references, an abstract of about 150 words, and a short list of keywords). Papers should be sent to the journal’s email address at:(balkanjournalofphilosophy /at/ gmail.com).
SUBMISSION DEADLINE: December 30, 2022.
This special issue will appear in2023.