Terzidis’ Algorithmic Architecture

Algorithmic Architecture, 2006
Kostas Terzidis
Architectural Press, Oxford

So let’s start with an (admittedly simplistic) question: Are algorithms radically new ways of generating form or just another toolkit for design? Though this presents a false-fork, it does caricature two sides of an imbalanced debate…a debate that Algorithmic Architecture enters into in an odd way.

Algorithmic Architecture is a strange book.  I want to be clear: I am not opposed to strange books.   Flatland is a strange book.  Through the Looking-Glass is a strange book…but Kostas Terzidis’ work is strange for different reasons.  All three books confront the “otherness” of mathematical, mechanistic, computation and frame them in humanistic terms.  All three books are riddled with paradox and contradiction.  However, whereas the later two classic works of fiction approach abstraction in a lively and poetic manner – sometimes morphing contradiction into fleeting insight – Algorithmic Architecture contradicts itself in much more confounding fashion…

Let me first say that the structure of Algorithmic Architecture is truly eclectic.  In the first chapter – “The strive to capture the elusive” – Terzidis traces the linguistic and philosophical roots of design, computation and the algorithm.  At the outset, he presents many structural supports on which to frame his argument, but little actual architecture.  In the second chapter – “The intricacy of otherness” – the author sets up a theoretical stage on which to choreograph the roles of designer and computer.  The third chapter – “A brief history of algotecture” – surveys the intersection between algorithms and architecture.  Chapter four is where  the book becomes fascinatingly and overtly strange.  In this chapter, the author explains the basics of MEL (Maya Embedded Language) scripting, and gives numerous code examples (pages and pages worth).  This chapter is almost completely different in tone and pitch: with a few minor exceptions, it is a clear and concise introduction to many abstract computational constructions, giving sample code to illustrate each.  For some readers, this might be new territory; to anyone who has already done any programming whatsoever, having variables, arrays and for loops explained (again) is unnecessary and a little comic wedged in the middle of a theoretical treatise on the role of computation in architecture.  Subsequently, chapters five and six jump back to the abstract theorizing, engaging the notions of  1) “Amphiboly,” the intentional ambiguity of form in design; and 2) “Periplocus,” the indirect yet intentional activity of fabrication.  These are the chapters that posit the notion that algorithmic architectural practice could lead to “novel” designs.  The book ends with an epi(multi)logue discussion of the concepts presented in the book and a number of interesting departures, contradictions and refutations. So now that I’ve spoiled the ending, let’s take a look at the arguments…

The work begins by distinguishing between “computation” and “computerization.”  According to Terzidis, what many architects see as digital design – the automation, mechanization, conversion, or digitization of preconceived or predetermined design artifacts – is only mere “computerization.”  Tactfully put.  What is of main concern in Algorithmic Architecture is “computation” proper: determining or calculating using logical or mathematical methods.  Terzidis’ main philosophical contention is that “algorithmic computation” is a fundamentally intellectual and computational process that complements and is the “Other” (Greek: allo) of human cognition.  The implication of this position for design is that the computer (or the algorithm) ought not to be seen as an extension of man, but as a completely different intentional entity with which one collaborates.  Related, though obliquely, is Terzidis’ view that design is a process of rediscovery, not of inventing the new.   Terzidis clearly has a love for Pre-Socratic philosophy and traces the words and terms of design back to their Greek roots, and with this tracing comes an early adoption (in the book) of the Parmenidian, Xenophanes-esque dictum that “nothing comes out of nothing and nothing disappears into nothing” (pg. 5).  The implication for design (at least early in the book) is that seeking the radically new or novel is futile in the face of a deterministic and causal universe…to caricature: “every design comes from another design, or another thought, or another idea, or another …etc.”  With this contention, Terzidis takes up a difficult philosophical and theoretical position with relation to design and intentionality…ok, this will take some explaining…

In the opening chapter, Terzidis claims that a computer should not be seen as a tool or an extension of man.  However, when he seeks to define the algorithm, he states:

“an algorithm is not about perception or interpretation but rather about exploration, codification, and extension of the human mind” (pg. 27).

One possible reading is that what he means here by “human mind” is equated with “human intellect.”   But this conception of “intellect” as inclusive of human intellect seems at odds with the idea that the “computational intellect” is somehow fundamentally different the human intellect.  While you may agree that this might be the case, accepting this radical difference has direct implications for the theoretical “comprehensibility” of this “Other” (allo) computational intellect.  Terzidis is not clear on this point.   If, as Terzidis holds, humans today have become capable of exceeding their own intellect (pg. 15, 17, 53, 117), he is forced to take a difficult position relative to the intentionality of the human designer and the computational complexity of the “algotecturally” designed artifact.  Terzidis’ position is ambiguous.  On the one hand, if you accept his claim that the computer can handle a mode of thought completely foreign and inaccessible to humans, you must dispense (to a certain degree) with the notion of purely human agency within the design process.  However, Terzidis also seems to claim (and disclaim?) the notion that human intention is key to the design process when he posits that design is simply about “rediscovery.”  Terzidis seems caught between the desire to take a radical position (e.g.: computational intellect is radically separate from human intellect, and algorithmic design is a collaboration or hybridization of two distinct intentional entities) and a lingering linguisitic/philosophical doubt concerning the novelty of computation as simply an extension of human activity.

The author’s love for the etymology and philology of design-related words and terminology is fun and contagious.  Most of this philosophical and linguistic exploration is contained in the first two chapters of the book.  In my reading, these were the best parts of Terzidis’ work.  However, having said this, it might be quicker to just read his paper “The Etymology of Design: a pre-Socratic Perspective”, (Design Issues: MIT Press, 2006).  With each passing chapter, I craved a sentence that would have been formulated thus: “What makes interesting for architects and designers is…”  Terzidis seems convinced that there is a fundamental break between human cognition and computation, and if a more collaborative and accepting culture of design embraces this, it will result in a sweeping paradigm shift within architecture.  Though I may sympathize and share affinity with many of Terzidis’ positions – I agree that knowledge of data structures and programming can enrich architectural education and practice – I find myself wondering just why this should be the case.  Though there are some hints that algorithmic design may (someday) be the dominant paradigm of dealing with spatial and formal complexity, I wanted a more plausible picture of such a future.  Early algorithmic buildings will probably be spectacular failures, and it is up to Terzidis to convince me otherwise.  Perhaps it is appropriate that this book reads like a series of fragments, rather than a cohesive whole…for fragments are all that remains of the works of the pre-Socratic philosophy Kostas Terzidis takes as his point of departure.