An Introduction to Computational Architecture
Digital, parametric, architecture has been given new adjectives since the early 1990s. These adjectives refer to an architecture designed, modelled, manufactured and/or implemented by an architect assisted by a computer. To this end, these terms are as general as they are generalizing. They contribute to the construction of the ambiguous and conflictual relationship between the architect and his tool.
The term computational architecture emerged more recently(1), and encompasses architectural practices dealing with computation. Computation is the calculation operated by a computer to process information. Computational architecture thus refers to the procedures that have emerged as a result of the introduction of the computer into the design process. Computational architecture considers the computer no longer as a representation tool, but as a pure calculation instrument.
Since the middle of the 15th century, architecture is represented in two-dimensional geometric projections: plans, sections and elevations(2). This classical design paradigm has been transposed to the computational context since the end of the 1990s. This linear transposition induce the use of the computer as a representation tool. Considering it as a computational tool not only upsets architectural design, but also the status of the architect. The architect is known to be the one who draws floor plans. In his computational design process the architect manipulates algorithms, drawing is secondary. An algorithm is a sequence of operations on a set of data. One can imagine, for example, that an algorithm can process the program elements of a project in excel form, and produce a plan by resolving the levels of proximity between the program elements. It is no longer a question of projecting a finite form, but rather of designing form possibilities. The architect establishes the rules for the emergence of the architectural object.
We can then ask ourselves: should we compute architecture? Architecture is perceived by its protagonists as an art that cannot be calculated, a pure product of human sensibility. The practice of architecture is imbued with the culture of fine arts, making architecture an art and the architect an artist. This vision of architecture closes in on the projection of an idealised reality, and tends to forget that in its realisation, all the elements of a building are calculated. Frame, structure, materials, façade, occupation, circulation, program elements, etc., are all computable and computed variables that directly model spatial phenomena, and their resultant spatiality. Computational architecture can then be seen as the bringing into coherence of these computed variables, through the sensibility of an individual, the architect.
The question that then emerges is: how to compute architecture? Most architects refute the idea of computation, which they associate with the automation of the thinking process. They look at computational architecture from the classical design paradigm, where space is thought from its geometric projection. In the computational paradigm, space is thought with the manipulation of computation models and variables. The complexity inherent in each architectural project is shared between the computer and the architect. This condition allows the architect to free himself from Euclid in his design process(3). His subjectivity, which is his own, is inscribed in the algorithm. We then witness a return to a form of craftsmanship, to a sensitive relationship between the designer and the tool, which he models according to his thinking.
It is from this redefinition of the architect that the friction between the proponents of the classical paradigm and the proponents of the computational paradigm arises. The transposition of the classical design paradigm into the computational context since the end of the 1990s has left the classical architect faced with a simplified interface, hiding the invisible growing complexity of the machines and computation models on which these tools, necessary for the production of architecture today, are based.
Architecture is not a discipline as such. Architecture is the harmonization of the disciplines involved in the construction of a building. What defines an architect is the point of tension between these disciplines. In this position, the architect is particularly dependent on the evolution of technology and the state of shared knowledge in a society. As such, he must take up the computational sciences, which have been the technological engine of the world since the beginning of the 21st century(4).
[1] Andrasek A (2009) Biothing, Hyx, Orléans.
[2] Alberti LB (1452) De Re Aedificatoria.
[3] The euclidean space is the traditional three-dimensional Cartesian geometric space (x, y, z).
[4] Bratton BH (2016) The Stack: On Software and Sovereignty. 1er édition. Cambridge, Massachusetts: The MIT Press.