ESSAy by Kathryn Stutts
L’émergence des outils informatiques de conception architecturale dit “paramétriques” ouvrent un vaste et parfois vertigineux horizon de possibilités, autant esthétiques que techniques, tout en donnant un caractère viscéralement organique et quasi-vivant au projet architectural, nourri de toute sa complexité. Sans angoisse ni fascination, Kathryn Stutts nous livre ses clés de compréhension.
The built environment has always been regarded in relation to nature, whether it be a relationship of harmony, discord or something in between. In recent years two parallel phenomena have evolved, feeding off one another and changing the architectural discourse surrounding the natural and the man-made.
One is a renewed inspiration from natural processes, especially ecological and biological systems. Far beyond the green building frenzy, the idea of nature as a model system is becoming a valuable way of thinking about our shifting, temporal, heterogeneous world. With the increased interest in natural sciences comes a renewed appreciation for the beauty of complex, manifold forms, also falling within a broader shift towards a greater value on aesthetics. There is even a growing willingness among some designers to cede a certain level of control, preferring emergent, self-generating architecture which produces exotic forms according to natural models. Such cases form an interesting parallel to advances in modern biology, in which scientists are studying organization and interaction at the scale of individual molecules in order to better understand the complexity of life.
The second recent phenomenon is the rapid advance of technology in architecture and design industries. In the infancy of a new era of architectural practice, the potentials of computing, visualization, analysis, and fabrication have taken hold. Specifically, the use of computational and parametric modeling software is spreading swiftly and pervasively through schools and practices alike. Designers are turning to centuries-old principles of mathematics to organize and script the design and making of objects, buildings, even cities. In these models all elements are interlinked and interdependent, bringing new levels of design possibilities. Such practices are breathing new life into architecture, still finding its footing in the fluctuating contemporary social and political climate.
These two parallel phenomena are distinct yet complementary, and taken together can generate interesting effects. It is in this context that pattern and ornament have once again become popular concepts in design. Ornament in architecture, which traditionally reinterprets natural elements into built forms, was nearly obliterated in the modern movement of the 20th century. Today it is welcomed for its ability to articulate surfaces and space, embracing complexity, accommodating differentiation, and providing identity. The technology exists to generate and manage complex geometry and surfaces whose organizing principles are often, though certainly not always, borrowed from nature. Furthermore, the tools used to design ornamental elements and tessellated surfaces are digitally linked to the tools used for fabrication, and thus encourage the connection to materiality and building. The master craftsman of the 19th century has now become the computer geek with a milling machine.
The power of integrated, computational design lies in the ability to control multiple variables in a single, interconnected, self-adjusting model. These variables could be geometric, performance-based, or a scripted response to a set of data. The input parameters and constraints can be modified and multiple iterations generated with the click of a button. The resulting model is embedded with inherent intelligence, to be exported as a drawing or a spreadsheet. It is a living, dynamic model rich with information. Ultimately this live model may be exploited for producing provocative imagery or for identifying practical methods of fabrication and construction. Among the many divergent directions in which this potential is currently being explored, two groups can be broadly labeled as theoretical pursuits and practical ventures.
The architects interested in emergent forms and generative design often take a deep inspiration from the science at the root of biology and the natural world to conceive of self-organizing, self-generating architecture. They even borrow the term “autopoiesis” from the scientific domain, meaning self-production and originally referring to the behavior of biological cells. The parametric process itself, essentially a system of organization, can be seen as a metaphor for systems in nature and thus provides an appealing method for addressing design problems. These types of studies have been popular in academia for the past ten years or so, but do not often expand beyond the theoretical proposal and seductive image.
To give an example of one such theoretical project, the Mangal City proposal was conceived by a team of students at the Architectural Association in London called Chimera. Making explicit reference to natural and ecological systems, the designers acquired an in-depth knowledge of phenomena such as adaptation and evolution. Their vision is for a self-moderating urbanism that operates down to the scale of human habitation. The model, formal and operational, comes directly from living organisms, the mangrove and its collective habitat, the mangal. Quite unlike an architect drawing a structure that resembles a tree, this team sought to understand the natural processes that give plants their form (for example phyllotaxis, a rather precise and intricate geometric organization which optimizes leaf exposure to sunlight and water). These principles are then reinterpreted into formulas that allow the design to be generated in a similar manner to how specific plants have evolved.
The resultant forms and powerful imagery of generative design processes are often beautiful or surreal, yet they also risk becoming stylized. Certain architects embrace this effect and are pushing for a new “style” based around the principles of parametric design, complete with a list of rules for a formal language . Other critics see this as a pitfall for architecture, when demonstrating the capability of the tool takes priority over the intent of the designer. As with any new technological advance, the challenge is to find a means of surpassing technology with architectural ambitions that exploit the tools while rendering them invisible. In generative designs such as the Mangal City, this is done with evaluative feedback and numerous iterations, as the aesthetic vision evolves in parallel with the model. Alternatively, new technological capacities can be applied to enrich the potential of pre-conceived design concepts.
The latter case is illustrated by the type of parametric design being undertaken to realize the Louvre Abu Dhabi by architect Jean Nouvel. In contrast to generative design which abandons a level of control over the final product (or may never be considered to have one), integrated modeling is being used here to provide high levels of control in a complex design problem with a fixed objective. The Louvre Abu Dhabi has at its heart a rich, atmospheric vision for a large open-air canopy to provide shade and dramatic lighting effects for museum visitors. This vision was at the origin of the concept and motivated the parametric process to be implemented as the design developed.
Upon first glance, the project is about pattern. The dome-shaped canopy consists of multiple layers of a geometric fractal motif superimposed to achieve a sense of randomness and wonder. A digital model is used to generate and control the pattern layers based on input parameters such as scale, rotation and translation. The relationships among layers and to the geometry of the sphere are far from haphazard, rather they are highly calibrated and precise. The structure necessary to span the 180 meter diameter is also incorporated in the model as a layer of the pattern, being fully integrated, optimized, and contributing to the visual and lighting effects of the multi-layer skin.
While the canopy may be the physical object of the design, the real material being manipulated is daylight. The dome itself is a tool, a complex filter, used to provide comforting shade and a distinctive micro-climate below. At the heart of Nouvel’s concept is the ‘pluie de lumiere,’ inspired by middle eastern souks and moucharabieh, functioning not unlike a tree canopy in dense forest. To achieve this effect, the parametric model must incorporate another level of control in order limit the amount of daylight passing through the dome. Thus the size and spacing of the apertures become the mathematical result generated from predetermined levels of light transmission. The layered pattern also accommodates variation in its openness by adjusting the thicknesses of elements in correlation to programmatic functions below the dome. A series of iterations were tested until a final configuration was found, a complex aesthetic enriched and refined by the optimization of performance and structural criteria. Once completed, the ornamental effects of the light pattern on the plaza will be dynamic with the changing path of the sun and daily weather patterns.
These two vastly different projects illustrate the range of potential to be found at the intersection of two significant trends in architecture. For Jean Nouvel, the vision and the means of realizing it are independent. For the Chimera team at the AA, the vision is inextricable from the computational software used in its creation, and the design is stopped before any plans for realization. There are endless other studies and approaches using these and similar methods, whether the focus be on single surfaces, pattern, intricacy, performance, fabrication, or even methods of project delivery and client-architect contracts. The more parallel investigations to draw upon, the richer the potential for architects, paving the way for aesthetically and cognitively rich projects and true blending of the realms of theory and practice. Even better if these integrated design processes allow architects to get a little closer to realized projects which feel studied, complete, and have the ability to move people. Perhaps this ‘deep’ architecture, with its affinity to biological models, can be seen as one response to the contemporary desire of humans to readjust their existence to the natural rhythm of a world in constant evolution.■