The role of digital computation in the design of objects and spaces is ever increasing. Digital tools overtook analog ones decades ago for the purpose of documenting designs, and there is more frequently a seamless digital path that leads from the designer’s desktop to the computer-controlled machinery that manufactures components of the project or product (even in architecture, industrial design’s cranky, slow moving and conservative uncle). There is a widespread belief that the digital tools at our disposal should have an effect not just on the efficiency with which we are able to conceive and realize our designs, but at the deepest levels on the utilitarian and aesthetic grammar of the finished work. But should we let our tools tell us what to make? Because they are means and not ends, computational design processes have no aesthetic language of their own- they cannot tell us what or how to design. The tools at hand will leave marks on the work, but they are not the soul of the project: that part comes from the humans that design, and the humans they design for.
There’s been recent buzz about the dawning of a new epoch in design, defined by a radical new conceptual approach derived from computational processes. The first point to make here is that all design is lowercase “p” parametric (design is the process of creating a solution by resolving a given set of parameters), whether we use parametric modeling software suites like Rhino / Grasshopper, traditional manual graphic static techniques, massive inverted chain sculptures, or intuition as our favored generative technique. Second, the rise of capital “P” Parametricism, if we are going to accept Patrick Schumacher’s co-opting of a specific term in computational design to describe the generic trend, is evolutionary, not revolutionary. Of course the tools of any trade affect the aesthetic qualities of the work itself―many of the great Modernists made this relationship explicit by favoring forms that expressed the capabilities and limitations of the machinery available. Precisely sawn timber, cleanly extruded metal sections and large planes of glass are elegantly expressed in the international style of the early twentieth century. Modernism espoused radical honesty about process―there were of course radically honest modes of construction preceding modernism, just as there are now. It’s OK for computer generated work to look computer generated- but computationally assisted design processes (specifically generative ones) will not by themselves magically impose any particular compositional grammar. We have to tell them to do that by limiting variables.
Another frequent assertion made in this debate is that access to digital computational technology has made possible feats of design complexity that were not previously achievable. It’s a bit strange and amusing that as postmodernism ran its course, it deposited adherents to this notion (believers in the wild form making power of the computer to outshine any human-derived logical process) in an essentially Victorian mindset: the bigger, stranger and more baroque the project, the better. I have a sneaking feeling, though, that these folks can’t see the forest for the trees. Let’s dispel the notion that radical scale or formal complexity are the exclusive purview of digital designers:
This is Salisbury Cathedral, built in the thirteenth-century. Completed in under 40 years (record time in a building type that often saw construction times measured in centuries), you’d be hard pressed to argue that there is anything geometrically simple happening here. There is an underlying formal hierarchy derived from the structural logic of stacked bricks, and there is underneath all that complexity a lot of symmetry, but there is more going on here than on person could keep track of with pencil and paper.
Baroque sculpture exhibits certain kinds of mathematical precision (proportion, scalar relationships), but it’s also pretty exuberant. A contemporary designer with the right digital tools might be tempted to turn that tunic into a roof. Many computational tools and techniques were available to a seventeenth-century sculptor, but no obviously not computers.
Again, there’s a compositional logic here that is derived from structural principles. The form, the scale, and the audacity of this public space are such that it could just as easily be an icon of our age, rather than the Modern. Saarinen himself was acutely interested in the potential of computational design techniques (he referred to them as “IBM machine methods” in a 1959 interview in Horizon magazine), and could be seen in part as a progenitor of their coming proliferation-but he ranked intuition as the most important tool at his disposal.
The point of all this luddite rumbling is this: computers by themselves don’t open otherwise unavailable formal or functional doors to designers, and their use merely facilitates, rather than defines, the things we make with them. If we value the puckish formal gesture, the space elevator, or the perfectly functioning intermodal rail station, then the computer can help us solve the design problem more efficiently than could be imagined even a generation ago, which is why every designer on earth should be exploring ways in which cutting edge computational design techniques can improve their process. The trouble is, it’s easy to let the computer’s solution-finding efficiency, and the semi-magical nature of the process of generative designing (did you SEE that?) put stars in your eyes. Even in the eagerly awaited fully generative processes (in which the designer will input project variables and let the software take over, spitting out a finished design) the designer acts as a sort of watchmaker god, ensuring certain outcome parameters by selecting a limited set of input variables. If you’re going to turn St. Theresa’s tunic into a convention center, you probably should think about why rather than how.
Ultimately, what we make is by and about us (until machines achieve sentience, and then it’s a different ballgame), and neither smoke nor mirror can remove the designer’s personal responsibility to (or for) what is created.
_Daniel S. Meyers