Breed (1995-2007) is a computer program that uses artificial evolution to grow very detailed sculptures. The purpose of each growth is to generate by cell division from a single cell a detailed form that can be materialised. On the basis of selection and mutation a code is gradually developed that best fulfils this “fitness” criterion and thus yields a workable form. The designs were initially made in plywood. Currently the objects can be made in nylon and in stainless steel by using 3D printing techniques. This automates the whole process from design to execution: the industrial production of unique artefacts.
Computers are powerful machines to harness artificial evolution to create visual images. To achieve this we need to design genetic algorithms and evolutionary programs. Evolutionary programs allow artefacts to be “bred”, rather than designing them by hand. Through a process of mutation and selection, each new generation is increasingly well adapted to the desired “fitness” criteria. Breed is an example of such software that uses Artificial Evolution to generate detailed sculptures. The algorithm that we designed is based on two different processes: cell-division and genetic evolution.
The underlying principle of the Breed morphogenesis is division of cells. One initial cell, a cube, engenders throughout successive stages of cell division a complex, multi-cellular “body”. Morphogenetic rules determine how the division of a cell occurs, dependent on its situation between the cells surrounding it. Every potential situation has a separate rule, so a cell surrounded on all sides by other cells may divide differently from a cell that only has a neighbour on the left and underneath, or a cell with nothing at all in the vicinity, etc. Each rule is coded in a gene, and the complete set of rules forms the genotype of the growth.
A parent cell divides by halving the cubic space that it takes up according to length, width and depth. This creates eight smaller spaces, each of which either contain a massive cell or remain empty, according to the rule applicable to the parent cell. The new cells again function as parent cells during the following division stage. Each division refines the cells and differentiates the form further, until growth ceases after a number of stages. The ultimate form, the phenotype, is not specified at a high level in its genotype, but is the result of the recursive application of simple rules on the lowest organisation level, that of the cell.
The appearance of such a virtual Breed object may take many forms. The majority of these are comprised of many parts floating separately from each other in space. This is no problem while it remains a computer model on the screen where gravity does not count, but turned into real material, subject to gravity, such an incoherent structure would collapse. To automate the search for constructable results, it is necessary to establish objective and measurable preconditions for constructability. One crucial condition is that the phenotype should be completely coherent, i.e. consisting of a single part. The computer can compare two models with each other by counting the number of separate parts the model contains. The model with fewer parts satisfies the criterion of “coherence” more, or rather is “fitter” than the model with the higher number of parts. We are now able to implement a computerised process of trial and error that incrementally evolves in the direction of potential solutions. The simplest method (two-membered evolution strategy) is already effective: take a randomly composed genotype as base, generate the phenotype and test it for fitness; mutate the base genotype, generate the phenotype and test it for fitness; compare both results with each other, and take the result with the highest fitness as the new base. Repeat the mutations until the result satisfies the stated requirements.
An initial series of six sculptures was executed manually in plywood. The tracing, sawing and gluing was a time consuming technique and in any case, we wanted an industrial procedure. It was not possible to produce such complex forms via computer controlled Rapid Manufacturing techniques until the end of the nineties. Only with the arrival of the SLS (Selected Laser Sintering) technology did it become possible to computerise the whole line from design to execution. A second series of nine samples was realised with this technique in DuraForm nylon. In 2007, a series of six sculptures is produced with the Printing Metal technique. These objects are made out of stainless steel, infiltrated with bronze. The results became smaller and more detailed so that the separate cells of the objects began to be absorbed into the total form.
“We do not wish to copy nature,” Arp once said. “We do not want to reproduce, we want to produce. We want to produce as a plant produces a fruit and does not itself reproduce. We want to produce directly and without mediation. As there is not the least trace of abstraction in this art, we will call it concrete art.” — Jean Arp (1886-1966).
Young Projects is pleased to present “Deep Dive”, the first solo show of the Dutch Artists Driessens & Verstappen on the West Coast. The exhibition is made possible by the generous support of the Mondriaan Fund, which promotes relevant Dutch visual art and heritage in the Netherlands and abroad.
Based in Amsterdam, Erwin Driessens and Maria Verstappen have been working together since 1990. They graduated from the Maastricht Academy of Fine Arts and the Rijksacademy in Amsterdam and shortly thereafter began a multifaceted practice that touches on sculptural practices, installation works, bio-art, new media and more.
Central to their practice is the exploration of natural systems and how they operate as self-organized environments. Many projects involve never-ending systems of their own; systems that either replicate themselves repeatedly ad infinitum or evolve—almost imperceptibly—over long periods. “We like to create systems that themselves create other things,” say the artists. “It’s meta creation.”
They’ve applied the process of cell division to video imagery, for instance, where a viewer can continually zoom deeper and deeper into the core of the image’s pixels (or bit map) without ever reaching an end point. That idea, as seen in works E-volved Cultures (2005-2011) and Formulae E-volver (2015) is to breed animations by applying artificial genetics and evolutionary techniques. In a continuous feedback loop between man and machine, infinite variations of computer code can evolve from a ‘primordial soup’ of elementary building blocks. “We create an art that is not entirely determined by the subjective choices of a human being,” says the artists, “but instead is generated by processes that operate more or less autonomously… [for us] few things are more satisfying than seeing a process unfold, seeing something that has not been seen before, no longer under our control.”
The artists have used a similar process of artificial genetics and evolutionary techniques to create sculptures as well. Breed (1995-2007) for example, uses a computer program to generate physical objects by incremental multiplication, starting with little more than a single cell. (On the basis of selection and mutation, a code is gradually developed that best fulfills a “fitness” criterion). Similarly, Accretor (2012) is based on accretion of small particles (cubical grains or voxels) to ‘grow’ sculptures that are so highly detailed they surpass any human capability. By executing these works with 3D printing techniques, the process from design to execution is fully automated. Thus, the unique nature of the sculptures, which results from industrial production methods, seems to contradict classical notions that uniqueness can only be achieved by personal style and manual labor. For the artists the process also comments on the ever-increasing consumption of mass produced goods, as opposed to the increasing desire for sustainability and a hyper individual lifestyle.
Given that their work hones so close to nature it uses similar ideas of emergence, complexity, form theory, dynamical systems and chaos. What’s more, it also touches on the subject of the sublime, in particular the sublime’s relationship to the infinite. in many ways, Sam Taylor Coleridge’s remark about the sublime beauty of the circle could easily be used to describe Dreissens & Verstappen’s generative works. “The circle is a beautiful figure in itself,” he wrote. “But it becomes sublime when I contemplate the eternity under that figure.”
Driessens & Verstappen’s work has been exhibited at museums and galleries worldwide, including the Central Museum Utrecht, Netherlands; Cordoaria Nacional, Lisbon, Portugal; IVAM Centre, Valencia, Spain; George Eastman House, Rochester, NY; Aperture Gallery, New York; and many others.