DEPLETION DESIGN
7
algorithmic architecture
luciana parisi
In the field of algorithmic architecture and in the works of, amongst many, digital architect Greg
Lynn, computational design takes inspiration from vector fields, used to model, for example, the
speed and direction of a moving fluid throughout space, or the strength and direction of some
force, such as the magnetic or gravitational force, as it changes from point to point. Here the
architectural form is the result of the computational processing of biophysical variables (e.g., the
distribution of weight, gravitational pressures, the circulation of air, the intensity of traffic, the
frequency of movement). Influenced by the second-order cybernetics of evolving feedbacks, during the 90s algorithmic architecture started to adopt biophysical dynamics to input change in the
software program, for instance by retrieving biophysical data as a new parameter to be added to
the sequence of algorithms.
By closing the gap between mathematical models and bio-physical contingencies, algorithmic
architecture entered the topological field of spatio-temporal connectedness, involving a continual
transformation of form without cutting or tearing. From this standpoint, biophysical unpredictability became superior to mathematical axioms and the reality of abstraction slipped behind the
concreteness of matter. Similarly, the digital design of space has now left behind the Euclidean
matrix of extension, and has added temporal evolution to fixed points developing a topological
surface of continual variation enveloping all points.
The figure of the blob or topological surfaces of continuity now dominates digital architecture. In
particular, parametricism, according to Patrick Schumacher, now represents the new global style
for architecture and design. When applied to large-scale urbanism, for instance, parametricism
is said to transform the differential distance between points into an integral surface of continual
variations. From this standpoint, parametricism implies the inclusion of contingent objects (e.g.,
atmospheric, geological, biological, physical elements), which become variable parameters whose
temporal evolution contributes to the overall transformation of the architectural whole. This means
that variables, for instance, are not only added to the program (as if from the ‘outside’), but rather
partake of the software environment of parametric relations. Parametric programming, therefore,
is not just concerned with the computation of existing elements, but also, and significantly, with
how feedback relations between finite parameters can engender the infinite variations of the
architectural form.
From this standpoint, parametricism is, as Hatherley calls it, a manifestation of the “cultural
logic of late neoliberalism”, whose topological operations of continual transformation, structural
coupling and mutual correspondence between the inside and the outside are now defining the
choreographic arrangement of data. This is not a new argument, and it is also one that is, to
a degree, separate from the issues that I aim to pursue here. I am not specifically concerned
with criticizing parametricism or its excessive formalism for its inability to address infrastructural issues and the political implications of lived space. Instead of arguing that parametricism promises a formally open-ended and flexible space that does not physically match real-
8
theory on demand
ized architectures (as in the case of the Evelyn Grace Academy Zaha Hadid Architects) and
instead of contending that parametricism is the direct incarnation of the spirit of the neoliberal market, I would suggest that parametricism is not abstract enough to meet the possibilities offered by a radical algorithmic formalism that exposes the algorithmic governance to its
internal inconsistencies, which, I argue, imply an understanding of algorithms as actual objects.
This means that a critical approach to parametricism does not and cannot exclusively disqualify
the computational production of spatio-temporalities on the grounds that this is always already an
expression of totalizing neoliberal governance. Whilst it is not my intention to deny that parametricism is an instance of the topological aesthetic of governance, I also want to problematize the
tout court rejection of the agential character of computer programming and the actual reality of
algorithmic objects that seem to lie behind the political critique of parametricism or of what is also
called Deleuzian architecture. This is to say that algorithmic objects are necessarily implicated in
the sociality that they invisibly structure. But the stealthy intrusion of computational programming
into everyday culture requires a close engagement with the nuances of the digital apparatus and
of the axiomatic thought that indirectly infects such a culture.
From this standpoint, the topological architecture of relations expressed by parametricism is precisely what needs to be challenged in order to reveal the transformation that algorithmic objects
have brought to digital formalism. A close analysis of this transformation may help us to explain
how structural changes in programming are not negligible, but are in fact ontological expressions of computational culture and power. This analysis may also contribute towards indicating
the incongruence, the asymmetry and not the equivalence between algorithmic architecture as a
totalizing system of governance and as a series of fractal or inconsistent events.
One can argue that an immediate level of transformation that parametricism exerts upon digital
formalism is its attempt to incorporate contingencies into the enterprise/model by including real
time data in software programming. However, I would argue that the introduction of temporality
into programming does not fully challenge the formal nature of algorithmic architecture. Instead,
it affords its formalism the pretension of describing how mathematics can incorporate physics by
creating a totalitarian system of relations, according to which a few mathematical rules ground
the evolution of complex structures by establishing continual feedback with the environment.
This means that the topological qualities of parametricism do not in fact challenge but instead
appear as the reification of its formalism, insofar as the computation of biophysical contingencies
only serves to reinforce formalism using the biophysical environment to construct itself as being
open to change. Rather than challenging computational formalism by claiming that computation
can never account for the whole of continual relations, it is necessary instead to unpack the internal limits of computation, and to thereby search for its internal inconsistencies.
Interactive and responsive computation work not to reveal but to occlude the what and the how
of algorithmic objects, which are deemed to remain passive in the face of an ever-changing
governance of continual feedback. We need therefore to turn to information theory to address
its rather complicated and subtle notion of algorithms, which, far from being equivalent to the
continual surface of connecting data, are instead sequential spatio-temporal data structures that
DEPLETION DESIGN
9
are conditioned by infinite amounts of information. I believe that these data structures are actual
spatio-temporalities and are precisely the objects of algorithmic architecture. This means that
an algorithmic object is more than a temporal form or the result of interactive inputs, and rather
defines spatio-temporal structures as the increasing amount of automated data in our computational culture.
Contrary to the view that computation is a reductive form of rationalism, I want to suggest that
the ingression of the incomputable in axiomatics can help us to re-think computation in terms of,
to borrow from A. N. Whitehead, speculative reason. Computation as an instance of speculative
reason does not correspond to the topological order of potential connectedness between points.
On the contrary, Whitehead’s understanding of speculative reason explains that the function of
reason is to add new data to the continual chain of cause and effect – the continual processing
of data for instance.
Similarly, a speculative view of computation implies that this is not equivalent to a mere compression of data or to a structure of relation defined by sets and subsets. On the contrary, a speculative understanding of computation implies that each set and subset of instructions is conditioned
by what cannot be calculated, the incomputable probabilities that disclose the holes, gaps within
and not outside the formal order of sequences. This means that a notion of speculative computing is not concerned with quantifying probabilities to predict the future, but with a concrete system of algorithmic objects defined by randomness or incompressible quantities of data detaching
from the relational order of topological continuity.
This is why speculative computing is not to be confused with the capacity of algorithmic architecture to create temporary forms that simulate what spatio-temporal structures and infrastructures
could become. Similarly, it does not mean that the governing enterprise of a relational database
can always already incorporate into its own structure the amounts of data processed. On the
contrary, the notion of speculative computing advanced here is used to suggest that random
data – which in information theory mean not arbitrary but non-fully compressible into a smaller
program - are the contagious architectures of the present. These architectures, far from withdrawing from the present (and thus remain temporal forms that appear and disappear), rather are
actual objects, which even when they cease to be here, nonetheless remain objective data to be
inherited, evaluated and selected by subsequent algorithms.
It is possible to suggest therefore that algorithmic architecture reveals the immanent reality of
patternless data exposing the inconsistent unity of algorithmic objects, thus determining a fractal
and not topological arrangement of spatio-temporalities in our computational culture. What is of
most importance is that this reality cannot be encompassed by the totalizing and invariant function of a topological model. From this standpoint, algorithmic architecture does not just reveal the
operation of soft-control and neoliberal aesthetics of topological relationality, but rather I want to
suggest that this aesthetics is broken open by the reality of algorithmic events or the automated
selection and production of incomputable data: random information.
Algorithmic architecture does not correspond to a topological whole bigger than its parts, but is
defined by parts that are irreducible inconsistencies divorced from the whole that can be built
10
theory on demand
through them. Algorithmic architecture works not against but with the chaotic parts of information, comprised neither within finite axioms nor within open interactions. Algorithmic architecture
may offer us the opportunity to discuss the nature of algorithmic objects beyond formal mathematical and physical models. It may contribute to unravel the speculative reason of algorithms
that may well overturn what is meant by the digital governance of space-time.
Recommended Readings
Hatherley, Owen. ‘Zaha Hadid Architects and the Neoliberal Avant-Garde’, Mute. Culture and
politics after the net, 2010, http://www.metamute.org/editorial/articles/zaha-hadid-architectsand-neoliberal-avant-garde (last accessed November 2011).
Schumacher, Patrick. ‘Parametricism: A New Global Style for Architecture and Urban Design’,
Architectural Design, Vol. 79, N. 4, (July/August) 2009, pp. 14-24.
Whitehead, Alfred North. The Function of Reason, Boston: Beacon Press, 1929.