CHAPTER TWO
Computational logic and
ecological rationality
Luciana Parisi
The computational turn in architectural design has led to a new conception
of nature, for which the idea of man-made structures has been surpassed by
an investment in materially driven ecologies. Computational design is now
concerned with the intelligence of materials, their capacity (or potentiality)
to self-organize by changing over time. This attention to a bottom up order
of becoming aims at “empowering matter in contemporary design”1 and
cannot be understood in isolation from a naturalization of logic, in which
computation constitutes the ground of in-distinction between technology
and matter.
Historically speaking, the development of computational design is
associated with the epistemological paradigms of second-order cybernetics
and interactive computation.2 The last ten years have been characterized
by a radicalization of the principles of biophysical self-organization
involving a design thinking, which brings together evolutionary biology and
non-standard geometry (or topology).3 The use of digital modeling inspired
by the Universal Turing Machine involved the manipulation of symbols
to test results and deduce proofs for possible structures. In contrast, this
neo-materialist approach, I would suggest, relies on inductive methods of
reasoning, where data from the biophysical world is algorithmically reactivated to evolve spatio-temporal structures, which are, as it were, empirically
derived from matter. This chapter argues that this naturalization of computation is an important instance of the ecological view of power.
Following Brian Massumi’s diagnostic analysis of governance in terms
of environmental order, this chapter discusses the advance of an ecological
form of rationality (the naturalized logic of affective power), which
feeds off its media-technological condition. The turn to computation in
design is already part of an ecological rationality of governance defined
by the technocapitalization of the indeterminate behavior of materials.
76
GENERAL ECOLOGY
The increasing investment in biotechnology, nanotechnology, information
technology and cognitive science points to a shift towards a dynamic rather
than mechanical instrumentalization of nature. I use ecological rationality
to describe the modus operandi of a logic no longer relying on deductive
reason. Far from simply imitating the physical properties of matter, this
rationality invests in their indeterminacy to generate conditions of affective
governance. I suggest that computational materialism in design is the
manifest image of a technocapitalist culture turning the mechanization
of deductive reasoning into a dynamic logic of computation whose rules
are established by the indeterminate potentialities of physical, biological,
chemical behaviors and their complex interactions.
However, I propose that this shift implies at least two overlapping
tendencies. On the one hand, environmental governance points to the end of
a deductive model of rationality surpassed by an inductive—or as Massumi
says an “affective” mode of governance (from the model of cognitive
mapping to the activities of pre-emptive power). On the other hand, this
technological form of governance involves the reduction of media to a
meta-computational apparatus of data, algorithms, and programs, defining
media as information systems.4 Beneath these overlapping levels, however,
this chapter argues, there is another, as yet unexplored consequence that
concerns the transformation of computational logic and of a mode of
reasoning involved in algorithmic processing. In what follows, I will draw
on Alfred North Whitehead’s notion of the speculative or metaphysical
function of reason to argue that computational logic could instead pose a
challenge to the totality of ecological rationality.5
This is an attempt to unpack the rupture between computational reason
and ecological rationality. My argument about the semi-autonomy of
computational reason (as part and parcel of a generic function of reason)
derives from a concern with the cogent reality of data architecture and its
algorithmic processing, which I argue can hardly be explained in terms
of what is affectively lived, perceived, and thought. I suggest that the
critique of ecological rationality embedded in the techno-computational
strata cannot only be explained in terms of the affective response reflecting
another naturalization of the artificial. If computational design exposes the
naturalization of both computation and technomediatic governance, it also
allows us to explore the historical configurations of computational logic
within the larger scope of a speculative or metaphysical function of reason
embedded in the actuality of algorithmic thinking.
The tendency towards the digitalization of nature is not new in design
and can be traced back to the use of mathematical formulae and solutions
in planning.6 However, with the computational turn in design, the use of
formulae has been replaced by the processing power of algorithms, their
performative elaboration of data exceeding the a priori of axiomatic
principles. The computational function of algorithms shows us that the
deductive logic of truth and a priori axioms is unable to account for—and
Computational logic and ecological rationality
77
to predict—contingent or external factors. The increasing use of large data
volumes and distributive interactive systems in design has not only pointed
to the limits of deductive logic (the general includes the particular) but also
diffused the use of inductive methods of heuristic thinking (starting from
the particular and proceeding by trial and error to arrive at the general)
in which the realm of physical contingencies and not of mathematical
formulae are said to be central to computation. If we read this shift to
physicalism in computation as a symptom of a new logic of power, then it
becomes evident that, as Massumi clearly argues, the chain of contingencies
becomes the driving force for decision-making actions. Inductive reasoning
is then complicit with the naturalization of computation and the emergence
of an ecological rationality modeled upon the premise of indeterminacy.
In particular, as evidenced in computational design, the indeterminacy of
matter (and materials) to generate spatiotemporal forms has resulted in yet
another idealization of physical structures, patterns, and complex behaviors.
While I suggest that inductive reasoning is central to a notion of computational nature, I also argue that ecological rationality can (and must) be
questioned. The computation of matter’s indeterminacy could be read as
the advance of power’s affective intelligence, whose actions, instead of being
deduced from truths, are induced from the behavioral patterns of matter
directly. This new level of equivalence between affect and reason reveals
the paradoxical condition in which the technocapitalization of matter has
led computational logic to become one with the physical indeterminacy of
nature. This chapter is an attempt at unpacking this seamlessly paradoxical
condition by arguing that the deductive limits of computation can rather be
understood in terms of a transformation of the function of computational
reason. I will discuss the computational mode of reason in terms of what
Whitehead calls “non-sensuous” or “conceptual prehension” in so far as
the algorithmic elaboration of data, I argue, partakes of a speculative,
generic or metaphysical function of reason that moves through but cannot
be contained by the biophysical layers of stratification central to ecological
rationality. This chapter suggests that algorithmic processing is a form
of reason that operates or becomes performative of a data environment
through a prehensive synthesis, which mirrors neither the laws of physical
nature nor the realm of mathematical order.7 In particular, the function
of rule-based processing will be discussed in terms of a speculative reason
that complicates the model of both deductive and inductive processing
of truths, and disentangles naturalized computation from an algorithmic
mode of thought. My attempt at halving the unity of computational reason
and naturalized technocapitalism is also an effort to re-address the notion
of reason in terms of a generic speculative schema—constituted by rules,
axioms, procedures—that are neither simply imparted nor proven by
the world. Instead, as debates about the limits of the deductive model of
computation in information theory suggest, rules can be bent and postulates can be revised, both according to contingencies occurring in data
78
GENERAL ECOLOGY
processing, but also because computational processing stretches beyond
given facts or data. In the history of information science, it is well known
that the question of the incomputable (random or infinite strings of data)
came to challenge the dominance of deductive axiomatic truths defining the
universal function of finite rules according to a mechanistic view of nature.
In the age of the algorithm, however, incomputables are no longer exceptions falling outside the remit of computational logic. On the contrary,
the latter has surpassed its own deductive limits, and, contrary to today’s
claims, it cannot be explained in the biophysical terms of the material
world. Instead, and this is my argument, computational reason needs to be
investigated according to its internal pragmatism, its own generic performativity (or even evolution) of data through which hypotheses are generated,
and initial premises are revised. If computational reason could be defined in
terms of its own dynamics, it would be approached in terms of a productive
instrumentalization of reason not simply espousing the project of capitalist
rationality (both formal and ecological). This productive instrumentalization instead involves an engagement with the historical transformation
of automated logic coinciding with the effort to theorize a generic model
of artificial reason, defined by the formation of non-matching forms of
intelligence—i.e., forms that cannot be naturalized into one univocal being.
This chapter suggests that divorcing computational logic from the
technocapitalist naturalization of computation is a fundamental step for a
speculative or metaphysical theorization of reason, that is, a generic architecture of reason that has infinite varieties of data environments and modes
of abstraction. Computation, I would argue, is only one mode and the
transformation of the logic of computation importantly reveals algorithmic
actuality and its automated reason. This also means that computation
needs to be disentangled from a totalizing notion of reason that ignores
the artificiality of abstraction (or computation as a mode of abstraction)
and its concrete structures of thinking. But how to engage with this mode
of abstraction, which is accused of quantitatively reducing thought to a set
of procedures without potentiality, chance and imagination? One way to
do so may be to attempt to articulate a generic or speculative function of
reason through a materialist approach that could explain the relation—and
not the equivalence—between biophysical constraints and the artificiality
of abstraction.
The final section of this chapter (“Speculative reason”) draws on Alfred
North Whitehead’s brief excursus on the centrality of reason in the history
of civilization, which is useful for our reconceptualization of computation
because it explains that the function of reason involves the abstraction
of causes from the physical chain of things. This involves counteracting
the continuous process of causes and effects with a concrete abstraction
of thinking. But why is a notion of speculative reason so important for
counteracting the ecological rationality of technocapitalism today? Does
it help us to move away from a totalizing technocapitalist naturalization
Computational logic and ecological rationality
79
of computation in which information is said to derive from the energetic
(affective) activities of matter? In short, can a speculative notion of computational reason go beyond ecological rationality?
These questions could be answered with concrete examples. However,
more (or less) than offering specific cases to evaluate these points, this
chapter argues that the becoming-environment of computation does not
mean that computation is nature. Instead I will consider computation as
an evolving mode of abstraction that reveals alien, intelligible capacities for
processing incomputable data.
While, as Massumi illustrated, the neoliberal form of technocapitalist
environmentalism has replaced deductive rationality with affective (nomo)
logic, it is here contended that the reservoir of reason left to computation
coincides neither with deductive logic nor directly with affective thought.
Instead, its alienness remains a symptom of a non-mutual relation between
ecological rationality and computational logic.
From this standpoint, this chapter suggests that it may not be sufficient
to ask how and in what ways a notion of speculative reason in computation can help us to think what it may mean to live in an algorithmic
environment. The alienness of automated reason rather involves the more
fundamental problem of confronting the actuality of a non-sensible thought
(algorithmic prehensions), amenable to neither logos (deductive rationality)
nor affective thinking. The analysis of techno-mediatic computation thus
requires a critical effort towards the articulation of automated reason.
What follows is an attempt to account for the function of computational
reason, questioning the ontological equivalence (or mutual co-constitution)
between the natural and the technical. I will discuss how computational design risks a renewed idealization of biophysical causes (and
natural contingency) through the idealist conviction of an immediacy of
computational logic and matter. The critique of deductive reasoning in
computational design may thus risk falling into a crude materialism, in
which the physical potentialities of material elements coincide with the
primacy of aggregate causality (i.e., of indeterminate correlations). This
view works to disqualify rather than explain the materiality of algorithms,
their actual functions of extraction and abstraction of data, which are
performative of a new order of finality leading to the production of rules on
behalf of algorithms. Matter-oriented design seems to overlook the materiality of artificial data environments. Paradoxically, here, the acceleration of
algorithmic automation has led to an anti-speculative approach to computation, in which the order of abstraction (and the intelligible processing of
data) has been reduced to the fluctuating dynamics of matter.
I will now address first the use of deductive approaches in computational
design, then discuss the shift to the dominant inductive designing of spatiotemporal structures.
80
GENERAL ECOLOGY
Digital nature
The Milgo Experiment, also known as the AlgoRhythms Project, devised by
architect morphologist Haresh Lalvani, exemplifies well the use of deductive
logic in computational design.8 Since 1997 Lalvani has been working with
Milgo/Bufkin, a metafabrication company, to realize curved sheet-metal
surfaces designed through digital programming. The development of the
Morphological Genome is described as the search for a “universal code for
mapping and manipulating any form, man-made or natural.”9
The universal structures that form matter, for Lalvani, must be derived
from simple genetic rules: cellular automata that specify a family of
related parameters, with each parameter controlled by a single variable
of form corresponding to a base in the DNA double-helix genome.10 In
other words, and in conformance with formal principles of computation,
Lalvani believes that the infinity of all possible forms can be specified
by a finite number of morph genes.11 Lalvani’s model of a continuously
generating morphological genome embraces the deductive logic of a digital
metaphysics, according to which cellular automata and discrete entities are
universal codes from which it is possible to deduce, just like with DNA,
an infinite variety of processes that enable the generation of new form.
Lalvani’s use of algorithmic architecture is not too far removed from the
fundaments of so-called digital philosophy, according to which digital or
discrete codes are the kernel of physical complexity: code is ontology, that
is, and finite sets of algorithms are the axioms upon which it is possible to
build any complex world.
According to digital philosopher Edward Fredkin, all physics can be
explained through the simple architecture of cellular automata, or discrete
entities that form a regular grid of cells, existing in a finite number of spatiotemporal states.12 According to this digital view of physics, the universe is
a gigantic Turing cellular automaton: a universal machine that can perform
any calculation and program any reality through a finite number of steps.
For Fredkin, cellular automata are the ground on which physics can be
explained. The universe is digital and not continuous. Cellular automata or
discrete units are the ground of nature. This digital conception of computational nature divides the Parmenidean infinitesimal continuum into finite
small particles, or atoms, within which complexity is contained.
This deductive view however has been challenged by a bottom-up
method in digital design. Architect Neil Leach, for instance, pointed out
that the use of swarm intelligence challenged this view resting on the use
of the discrete logic of fractals, L-systems, and cellular automata.13 This
meant that biophysical indeterminacy or the contingency of environmental factors had to be accounted for by computational modeling. By
adapting an inductive form of reasoning in which environmental variations
became a central factor in computation, digital design turned towards a
Computational logic and ecological rationality
81
generative form of modeling, incorporating temporalities and championing
the indeterminacy of biological systems.14
In the field of algorithmic architecture, one example of the adoption of
inductive logic aiming to include contingency in digital modeling can be
found in the works of architect Greg Lynn. His design takes inspiration
from biophysical vector fields that involve the growth of an emerging
algorithmic form.15 Computational modeling here becomes an evolving
system in constant coupling with the environment. Just as computational reason associated with cellular automata was attuned to first-order
cybernetics and deductive logic, so too this view of computation involves
notions of self-organization and interaction aligned to second-order cybernetics and its inductive reasoning. This shift, highlighting the centrality of
interaction among agents, already constituted the germs of an ecological
rationality in which computational systems are attuned to evolutionary
dynamics.
Leach takes the 2008 design by Kokkugia of the Taipei Performing Arts
Center as an example of swarm modeling, whereby interactive self-organizing multiagents define objects in terms of unity and the parts thereof, as
being both one and many.16 Here the parts of an object are conceived as
semiautonomous agents able to evolve their own set of interactions with
other objects without reproducing the same set of instructions. Similarly,
changes are only dictated by the emergence of contingent solutions.
This emphasis on self-organizing agents and partially interacting objects
producing a whole bigger than its parts exposes second-order cybernetics’
emphasis on the behavioral capacities of biophysical properties to coevolve
over time. The inductive premises of computational design are thus defined
by emergent and not preprogrammed properties of interactive algorithms.
These premises seem to anticipate a holistic view of computation in which
the multiagent swarming intelligence points to computational modeling as a
variable whole. This holistic view, however, cannot help but reify the notion
of a computational nature in which it is impossible to discern the continuity
of biophysical complexity from the discrete character of computational
abstraction. Ultimately, swarm models reveal that the temporal dynamics
intrinsic in the biophysical environment of continuous interactions is the
motor of computation. The next section will discuss the radicalization of
this inductive form of reason in computational design.
Computational nature
The most powerful and challenging use of the computer … is in the
learning how to make a simple organization (the computer) model what
is intrinsic about a more complex, infinitely entailed organization (the
natural or real system).17
82
GENERAL ECOLOGY
Moving away from computation as a form of symbolic representation of
physical elements, the definition of so called “material computation” has
radicalized the inductive method of reasoning, arguing that the biophysical
world of matter already provides us with a model of computational
processing.18 From this standpoint, the elemental properties of materials
and their generative rules constitute spatiotemporal structures in nature.
Instead of following geometrical and mathematical patterns, this (hyperinductive) vision of material computation aims to directly follow material
processes of self-assembling that result from the interactive relations of
loose elements. Physical computation corresponds to pattern formations
in both living and nonliving nature driven by the analogical process of
local interactions, giving rise to material self-organizing structures and
behaviors.19 Central to material computation is the question of design
in nature. Against the Darwinian selection mechanisms, design is here
explained in terms of physiological process and energy systems.20 The transcendent model of natural selection based on rule-based design is replaced by
an emergentist conception, whereby design is led by the inherent morphogenetic potential of material to grow and evolve into new structures.
Material computation is concerned with immanent processing in which
information has acquired an energetic pulse and has become itself a process
in-formation. The scope is not simply to induce algorithmic processing by
establishing a continuous feedback between programmed instructions and
the biophysical environment. More radically, it involves an ontological
merging of computational processing and physical process. This radicalization of inductive reasoning problematically implies a naturalization of
computation, claiming that the potentialities of biophysical substrates are
now central to what can be constructed, thus ultimately dissolving any
binarism between thought and things, concepts and objects.
This approach in design offers us an entry point to this ontologization of
computational nature. For instance, architect Achim Menges’ project ICD/
ITKE Research Pavillion realized at the University of Stuttgart is conceived
as a “bending-active structure” in which the feedback between computational design, advanced simulation and robotic fabrication is set to explain
how the material behavior of wood coincides with a complex performative
structure.21 Instead of relying on high-tech equipment that could simulate
or activate material, Menges points out that the responsiveness of material
is embedded in the computational capacities of the material itself.22 This
approach involves a direct manipulation of the material and “the physical
programming of humidity-reactive behavior of these material systems,”
which leads to “a strikingly simple yet truly ecologically embedded
architecture in constant feedback and interaction with its surrounding
environment.”23
These challenging bio-technical structures are understood in terms of
intensive aggregation (not partes extra partes but an intrinsic self-differentiation of parts), explaining how the overall system behavior results from
Computational logic and ecological rationality
83
the interaction of loose elements. Aggregates are thought to be able to
“materially compute their overall constructional configuration and shape
as spatiotemporal behavioural pattern.”24 Computation coincides here
with the continuous integration between material, form and performance,
in which material systems’ performative capacity is, as it were, enacted
again. From this standpoint, the computational qualities of nature are
here doubled, expanded and regenerated. While material computation is a
radical step away from formal logic and symbolic language, it involves a
radical attempt at integrating computation with matter, not only because it
aims to re-enact computation in nature, but because it reveals the computation of materiality itself. “Inspired by nature’s strategies where form
generation is driven by maximal performance with minimum resources
through local material property variation,” material computation proposes
to “analyse model and fabricate objects, with non-binary continuously
heterogeneous properties designed to correspond to multiple and continuously varied functional constraints.”25 This new level of designing matter
is already instantiated by “rapid prototyping and manufacturing”: one of
the many attempts to use material computation to establish an economical
/ ecological way of minimizing energy expenditure, resources, and environmental impact.26 The tendency of material computation therefore is not
simply motivated by a reproduction of the computational processing
of information found in nature. Instead, the radicalization of inductive
reasoning here implies that the computation already found in biological,
physical, and chemical systems explains that aggregate material properties
constitute design. Differently from the crude empiricism of biomimetics,
for which the natural order of design is reproduced and optimized in the
development of nature-like structures, this new convergence of computation and materiality instead conceives of what is given in nature in terms
of potentialities or indeterminacy. In other words, it is indeterminacy and
not the already measured value of physical, biological, chemical processing
of data that explain the tendency of nature to become more than what it is.
This form of computation aims to explain an eco-logical order of nature.
Ecology here involves not an (associationist) interaction of parts, but the
capacities of the environment, defined in terms of a multiplicity of interlayered milieus or localities, to become generative of emergent forms and
patterns.
Materialist computation specifically draws on the work of biologist Jakob
von Uexküll, offering an ecological conception of space defined in terms of
an immanent condition of subjective experience.27 In particular, his theory
of the Umwelt (“surrounding world” or “environment”) explains natural
design as intrinsic capacities of all human and nonhuman elements to feel
and sense. Here the signaling processing (the transmission of information)
or the interaction of perceptual and sensual signs is the condition for the
activity of a selective mechanism continuously shaping and unleashing
the capacities of organisms to transform and be transformed by the
84
GENERAL ECOLOGY
environment.28 The philosophy of Von Uexküll rejected Darwin’s theory
of natural selection because it excluded the inner worlds of animals and
their capacity to act upon and become co-constituted with their physical
environments. Here, environments uniquely afford the internal capacities
to generate new functions and behaviors.29 Von Uexküll’s ethological study
of the environment as a fundamental factor in the constitution of beings
explains not the optimization of species but the emergence of new patterns,
orders, configurations. No longer deduced from finite sets of rules, but
rather hyperinduced by the potentials of local interactions, allowing for a
constant transformation of energy into information, this ecological view
of computation defines the primacy of ever-evolving relations over a static
order of matter. While this approach is now central to computational
design, I argue that it also instantiates the complex constitution of an
ecological rationality in contemporary power.
Ecological rationality
Brian Massumi argues that the contemporary regime of power could be
understood in terms of an environment autonomous activity operating
through the regulation of effects rather than of causes.30 In particular,
he develops Michel Foucault’s insights about the environmental qualities
of power defined not by formal rationality (transcendent law), but by
inductive or local responses to governability, involving the performativity
(evaluation, selection, ordering) of external variations and the establishment
of general codes of conduct. Following Foucault, Massumi suggests that
Environmentality works through the “regulation of effects” rather than
the re-establishment of causes, and must remain operationally “open to
unknowns.”31 Massumi investigates the form of rationality involved in the
calculation of risks and suggests that within a global mode of operative
power, rationality is surpassed and replaced by an affective field.32 The
questions “What order is this? Does it still have the rationality of a
system?”33 point to the affective reconfiguration of the order of biopower.
For Massumi, this new order involves naturalization or naturing
nature34—a concept that needs to be redefined away from any categorical
opposition to the artificial because it operates in a zone of logical and
ontological indistinction between nature and culture. Here the environment
is not One un-adulterated given, but “indeterminacy” driven by interconnected levels of complexity, comparable to the intricate unpredictability
of weather systems, for instance.35 Massumi explains that nature now
coincides with the primacy and the immanent reality of the accident—with
indeterminate indeterminacy. But to understand this nuanced naturalization of power, Massumi suggests that it is necessary to develop a concept
of “naturing nature coming to cultivation.”36 Here nature is at once
Computational logic and ecological rationality
85
produced and presupposed, induced and deduced, active and passive. This
concept describes the constant correlation that constitutes the environmental processing of interlayering strata from which an ongoing emergence
of patterns is—as it were—excarnated from the “biosphere or noosphere.”
This ecological interdependency among layers involves, for Massumi, a
notion of “auto-conditioning” of naturing nature moving across scales and
milieus. Naturing nature can also be understood in terms of an involutive
(non-linear) process defined not by gradual steps but by intensive temporalities.37 Instead of the laws of nature deduced to explain the emergence of
patterns, the machinic operations of a naturing nature involve a radicalization of inductive forces in which unknown effects are driving forces, a
quasi-causal motor of power. Borrowing from Gilles Deleuze, one could
understand this radicalization in terms of “transcendental empiricism.”38
The continuity of affects involves the temporal anticipation of potentiality,
which Massumi understands in terms of preemption.39 This implies not
a rational logic aiming at repressing the future through the pressures of
the past, but more precisely a future-oriented logic entering the achronological fullness of time, absorbed by the affective experience of duration.
This temporal performativity can no longer be defined according to
Jameson’s critique of cognitive mapping—a critique of a deductive rationality grounded in representation or conceptual framing of matter. Instead,
ecological power involves an inductive logic of effects that are generative of
infinite aggregations without primary cause.
If material computation in design is an instance of this form of ecological
rationality, where power does not simply instrumentalize nature, but cultivates nature and anticipates its becomings, how can a critical conception
of computation—or automated mode of reason—avoid the trap of technocapital naturalization? In order to develop a notion of reason away from
the technocapitalist naturalization of matter’s behavior, it seems urgent
to explore the computational and philosophical limits of deductive and
inductive reasoning. If the deductive logic of technocapitalist naturalization
involved the rational application of truth to the world through transcendent
laws and rules, the technocapitalist adaptation of inductive logic rather
implies that the truth becomes one with the lived world. Here reason
becomes immersed in the capacities of a body to feel and make intuitive
decisions rather than follow sequential, logical steps. While this ecological
form of rationality explains what is at stake with technocapitalist naturalization today, it discards the historical transformation of automated logic.
As discussed in the next section, the transformation of theories of computational logic highlights the existence of blind spots within a seemingly
totalizing ecological rationality. I will attempt to develop a theory of
algorithmic thought in terms of a speculative notion of reason by arguing
that the intelligible capacities of algorithms to transform randomness into
information patterns exceed the ecological rationality of capital. Borrowing
from Alfred North Whitehead’s discussion of the speculative function of
86
GENERAL ECOLOGY
reason, the next section will also address the limits of both deductive and
inductive logic to account for how thinking operates. It will be suggested
that beyond the tension between affective (or inductive-empirical) and
rational (deductive-representational) logic, it is possible to discuss computation in terms of physical (efficient cause) and conceptual prehensions
(final cause).40 If the ecological rationality of capital exposes the univocity
of nature and culture through the centrality of affective thought, then
the turn to computation may offer us a way to discuss the emergence of
intelligible activities that challenge the ecological self-generation of being
and thought. It will be suggested that Whitehead’s notion of speculative
reason may be a productive starting point for a materialist approach to
computation concerned not with re-enacting the computational qualities of
nature, but with engaging with the intelligible tendencies of algorithms to
process infinite varieties of infinite data.
Speculative reason
This final section will help clarify how and to what extent it is possible to
approach computational logic without reducing its operations to technocapitalist governance. My attempt at specifically theorizing the computational
function of reason wants to suggest that this is not naturalizable insofar
as it has fundamentally developed through the artificial construction of
data-environment, and because it involves an algorithmic order of intelligibility—an alien reason—intrinsic to the actual processing of data.
Importantly, the articulation of a computational function of reason requires
a theoretical engagement with the problem of the limit of computability,
which has also been discussed in terms of randomness, incomputability or
the famous “halting problem.”41
While this classic question of the limit of computability has been
exhaustively addressed in the history of computational theory, Gregory
Chaitin’s quest for Omega, or for an algorithmic pattern of randomness,
seems to offer us one of the most promising views about the hypothesis
that algorithmic procedures are not merely instruments of elaboration of
primary data.42 Instead, one could argue that they can also be understood
in terms of their prehensive activities of recording, storing, selecting, and
elaborating patternless data. Chaitin’s views importantly contribute to the
development of a theory of computation concerned with the transformation
of automated logic. In particular, his renewed engagement with the mathematical theory of information (especially the emphasis on the ratio between
meaningful patterns and noise) and the problem of entropy (the measure
of chaos) in a communication system offers us the opportunity to consider
the issue of the limit of computation in terms of a historical realization of
the limit of logic in the first place. Chaitin’s long-term study of the problem
Computational logic and ecological rationality
87
of randomness in information theory is based on a specific notion of
entropy, which he understands in terms of irreversibly increasing volumes
of information generated at the input-output levels of computation.43
Bringing together Alan Turing’s question of the limit of computability
with Claude Shannon’s information theory, Chaitin tackles the question of
indeterminacy in computation by demonstrating how randomness (noise
or incompressible quantities of data) is rather central to computation.44
For Chaitin, computation corresponds to the algorithmic processing of
maximally unknowable probabilities. In every computational process, he
explains, the output is always bigger than the input: something happens
in the processing of data that breaks the equilibrium between input and
output. Chaitin calls this phenomenon algorithmic randomness.45
The notion of algorithmic randomness implies that information cannot
be compressed into a smaller program, insofar as between the input and the
output there emerges an entropic tendency of data to increase in size (i.e.,
involving an increase in patternless information within the system). Chaitin
explained the discovery of algorithmic randomness in terms of a rule-based
processing that no longer follows the approach of deductive logic for
which results are already contained in their premises. During the 1990s and
2000s, Chaitin identified this problem in terms of the limits of deductive
reason. He claimed that the problem of the incomputable defining results
that could not be predicted in advance by the program are to be explained
in terms of “experimental axiomatics,” a postulate or decision immanent
to the patterns evolving in the algorithmic processing of primary data.46
The increasing quantity of patternless information, emerging from within
computational processing, points to a dynamics internal to algorithmic
operations, whereby patterns are consequent to the synthetic relation
between algorithms and data.
However, this dynamic is not derived from or induced by the biophysical
activities of the environment, but operates within the data environment itself,
according to which automated logic involves neither deductive (a priori
rules) nor inductive (biophysically driven) reasoning. This is also to argue
that to reject ecological rationality and its technonaturalized governance,
we need to develop a new critical view of computation. One step towards
the articulation of this view involves the enlargement of the question of
automated logic through the theorization of a speculative function of reason,
which, as will become clearer later, could account for the elaboration of
generic rules from the prehensive elaboration of materially embedded
data. Here the analysis of the historical transformation of computational
logic as demonstrated by Chaitin’s method of “experimental axiomatics”
is paramount. This means that computational logic can no longer be
critically rejected because of its limited formal, symbolic order of reason,
syntactically working to achieve arbitrary connections between units / bits
of data. Similarly, automated logic and rule-based reasoning cannot simply
be jettisoned in favor of locally induced inputs. I have discussed earlier the
88
GENERAL ECOLOGY
predominance of these two models in computational design, particularly
marking the shift from the model of cellular automata to aggregate causality
in material computation. This historical shift however has to be accompanied by a theoretical reconceptualization of computational logic, and, I
propose, it needs to be re-examined through Whitehead’s argument for a
speculative function of reason (as discussed later in this section).
Similarly, one cannot overlook the increasing concreteness of data
environments embedding the abstraction of social, economic, cultural,
as well as physical data and constituting artificial socio-cultural environments whereby the relation among algorithms and between algorithms
and data leads to the formation of socio-cultural generic patterns, rules
and laws. This level of environmental artificiality cannot be explained in
terms of the affective mechanisms of technocapital communication. These
automated relations produce a surplus of information in which algorithms
have acquired intelligible—or conceptual—functions revealing an order of
decision incompatible with the effective avalanche of affective response.
Instead this order reveals the establishment of algorithmic patterns of
patterns elaborated through the physical and conceptual prehensions of
computational data environments.
Beyond ecological rationality and the technocapitalist imaginary of a
holistic enviromentality, it may be possible to argue for the “actuality”
of data environments that explains the constitution of an artificial world
equipped with its own notations, functions, physicality and conceptuality.
This also implies the formation of a post- or neo-cybernetic phase of epistemological production in which data environments do not simply represent
socio-cultural codes of conduct, but more importantly through a process of
physical prehension they acquire an algorithmic order of intelligibility out
of which socio-cultural rules are re-established (and re-visioned) because
they are embedded in the use of techno-computational language.
As opposed to ecological rationality betting on the technocapitalization
of affective thinking, data environments do not only execute instructions
but also are physically and conceptually prehended to elaborate patterns
at the limit of computational processing. This computational processing
of data importantly points to the intelligible prehension of physical data
that could create concepts or rules from a vast amount of patternless data.
This process of elaboration of data into rules is what may characterize
a data environment beyond the mathematical logic of deduction—establishing a formal or representational schema—and the empirical method of
induction—establishing an experiential relativity of localized responses. A
theory of automated reason may therefore show us the inconsistency of
ecological rationality and its media-technological situation, which seems
to offer us an opportunity to re-invent a critique of computation beyond
the holistic history of technocapitalism. From this standpoint, computation
is not equivalent to naturing nature, but involves a form of intelligibility
able to use algorithmic processing to add a generic order of axioms, codes,
Computational logic and ecological rationality
89
and instructions to what was initially programmed. Programming here
corresponds to the calculation of complexity-by-complexity, exceeding
programming itself.
If axioms are becoming experimental truths, able to postulate unknowns,
computation too may need to be conceived in terms of experimental determinations or intelligibility prehending unknowns and contributing to a
process of revising initial conditions.47 I suggest that this experimental (i.e.,
a priorily improvable) processing of data can be understood in terms of
Whitehead’s prehension because it involves a process of elaborating data,
implying a consequent finality added to what was already programmed.
According to Whitehead, prehension involves the physical and conceptual
modes of selecting and evaluating data and thus the registering, storing, and
processing of existing data. This notion of prehension importantly implies
that the function of reason is not to mentally map physical data, but to
transform—in counter-intuitive manners—physically prehended data, by
adding a level of finality to the physical order. Prehension, in other words,
corresponds not only to the physical mechanics of registering data and
using their existing functions, but more importantly it implies a process of
abstraction, or a conceptual elaboration of data, unfolding another level of
function able to establish or not generic rules and articulating logic. This
prehensive process explains that the function of reason is not mainly to
represent what is physically sensed (or even to re-potentialize sense data).
More importantly, it concerns the capacity to reset and redirect the scopes
of inputted data according to what can be conceptually achieved through
the conjunction and disjunction of patterns, involving the construction of
hypotheses that agree or not with given conditions.
From this standpoint, the centrality of randomness or the entropic
tendency of information to increase in size, resulting in an output that
is bigger than the input, implies that algorithmic prehension involves the
activation of reasoning leading to the experimental (non-a priori) establishment of rules. The generation of a bigger output can be understood
in terms of an experimental intelligibility internal to computation able
to surpass the limits of its deductive premises. Alfred North Whitehead’s
theory of speculative reason could explain computational processing as a
capacity to both surpass and bring forward deduction and induction, truth
and fact as parts of its experimental axiomatics.48 This requires addressing
algorithmic intelligibility in terms of hypothesis generation in which the
data environments constitute the material (non-discursive) level, which is
physically and conceptually prehended by algorithms, in turn establishing
an automated function of reason defined by the propensity towards the
generation of rules. Algorithmic intelligibility can be explained in terms of
the speculative function of reason insofar as it involves the emergence of a
generic form of process or algorithmic abstraction, which is embedded in
the data environment that it retrieves and through which it elaborates an
order of rules beyond deductive schema.
90
GENERAL ECOLOGY
From this standpoint, I suggest that the development of material
computation in design shall not be mainly concerned with the notion of
aggregate causality (efficient causality) coinciding with the complexity of
physical causes, leading to the elimination of rule-based processing. More
importantly, in order to disentangle the computational environment from
the technocapital naturalization of computation, we may need to unpack
the order of conceptual elaboration and of experimental logic within
computation.
This is to say that algorithms are not simply the computational version
of mathematical axioms, but are to be conceived as actualities, selfconstituting composites of data, which is at once recorded and elaborated
beyond its primary condition. As Whitehead explains, at a primary level of
reality there are only actualities and nexuses of actuals.49 These composite
actualities are comprised of physical and conceptual data. From a chemical
element to an idea, actualities are constituted by the very activity of registering, recording, selecting and evaluating data. Actualities are neither
subjects nor objects but the process of nesting data is explained by the
manner in which objective data acquires a subjective form involving the
hypothetical tendency towards a level of finality in which actualities reach
completion through what Whitehead calls “concrescence,” the growing
together of many levels of actualities.50 Central to actualities therefore is
not simply their material aggregation, or biophysical co-causality, but also
the introduction of a conceptual level of causality, defining the aim or the
subjective formation of an actuality. Actualities are thus constituted by the
physical and conceptual prehension of data.
From this standpoint, instead of being merely a set of instructions to be
executed in an environment, the increasing concretization of data environments rather shows that algorithms are composites of data and could be
understood as actualities equipped with their own procedure for prehending
data. Their actuality therefore is not defined by substance but involves data
processing (sequencing, execution, elaboration), which, within information
theory, has been theorized in terms of an experimental logic in which results
cannot be prescribed by inputs. At the same time however, algorithms
are also a form of abstraction or conceptual schema, which, following
Whitehead, is inevitably embedded in actualities. From this standpoint, an
aggregation of actual entities can become an abstraction in which certain
actualities become dominant over others. This process of abstraction is also
to be understood in terms of a speculative function and not simply in terms
of representation—i.e., in terms of a generic function of reason including
abstraction in every actuality and not according to pre-constituted symbols
framing actualities. In particular, Whitehead explains that the speculative
function of reason entails a process of abstraction aiming not at reducing
but mainly at counter-articulating or repurposing registered data towards
generic ends.51 From the standpoint of computation, this process of
abstraction involves the emergence of a generic order of finality or final
Computational logic and ecological rationality
91
causality carried out by algorithms through the conceptual elaboration
of physical data, a sort of algorithmic purpose that is immanent and yet
not reducible to the locality of data environments. Instead, it is worth
noticing that the dominance of the incomputable in interactive parallel and
distributive systems today has turned these environments into uncountable
quantities of search spaces, expanding rather than mainly containing the
possibilities for algorithms to form generic information patterns.
The function of reason in computation importantly points to a historical
transformation of mechanized logic, which in contrast to what is advocated
by material computation cannot be explained in terms of the biophysical
behavior of material substrates. What is needed therefore is not another
technonaturalization of computation (as the approach to material computation in design risks doing) but a materialist approach to abstraction that
is able to explain the evolution from physical to conceptual prehension in
the formation of propositions in which an intelligible registering of data is
followed by an elaboration of generic patterns or rules. Insofar as there is
no actual process that is not accompanied by a conceptual prehension—or
abstraction—of it, the speculative function of reason explains that the
dynamics of abstraction involves a material process of elaboration and
revision of rules, which emerges from and yet extends beyond the local
circumstances of matter’s configuration. Conceptual prehensions define a
final cause that pushes the initial conditions of given facts towards newly
planned actions achieved through the abstraction of prehended data
entering the realm of the generic so that it can yet again enable another
level of actual processing. Final cause, therefore, coincides not with pre-set
aims containing their results, but with the speculative tendency of reason to
become generic and re-determine truths.
Conceptual prehensions define the level of finality of any actuality
because they allow the intelligible elaboration of physical data, and the
capacity of the latter to transform existing concepts beyond their initial
premises. In other words, speculative reason clarifies the purpose of actualities in terms of hypothesis generation or renewed determination of truths:
experimental abstraction.52
From this standpoint, the speculative function of reason serves to explain
the move from indetermination to determination defining computation as
experimental axiomatic in which initial conditions—set ideas or facts—can
change in the processing of data. For Whitehead, the purpose of reason is
to revise its premises rather than being determined by the essence of who
or what does the reasoning. In other words, and contrary to the universal
principle of sufficient reason, any actuality has its own immediate finality
driven by its own mode of reason determined by its own discretization
of data, making infinities partially intelligible and thus extending the
limits of reason towards incomputables. This is an instance of immanent
finality, which rejects both vitalist (empirical induction) and mechanicist
(or idealistic deduction) purposes of reason. It explains the autonomy of
92
GENERAL ECOLOGY
any actuality that above all serves itself, rather than being an instrument
for (and of) something else. It suggests that the computational function of
reason is not a totalizing, sufficient model of reasoning.
A denaturalized conception of computational reason instead starts from
the premise that prehensions involve non-conscious but nonetheless intelligible operations of gathering and selecting data. Algorithmic prehensions
imply no direct bodily sensing and no sense of self-awareness. Nevertheless,
they are not simply the reproduction of existing data seamlessly reprocessing
over and over again. Instead of a mechanical unconscious, which could be
understood in terms of the affective qualities of feeling (and thinking) before
cognition (and rational decision), the notion of algorithmic prehension here
aspires to define an intelligible function of automation able to re-finalize
gathered, selected, and evaluated data. Unlike the consciousness attributed
to rational choice and to the analytic operations subtending decision,
algorithmic prehensions do not achieve sophisticated levels of self-reflection
(or critical view) and are thus instances neither of cognitive functions nor
of the unconscious power of affective thought. How then to articulate
these unfelt—unintuitive—and yet non-conscious automatisms? This is
a challenging question and one that can be addressed if we entertain the
possibility that automation also performs an intelligible function, and,
to some extent, achieves a conceptual determination of incomputables.
From this standpoint, the computational environment cannot simply be
an instance of ecological rationality. Instead, the automatic functions of
algorithmic prehension involve non-conscious53 yet intelligible elaborations
of physically prehended data, showing a contradiction internal to technocapital, which is unable to mend its own schizophrenic constitution.
The computational function of reason thus coincides with the discretization, selection, evaluation of increasingly random data (both external
and internal to the computational environment itself), which importantly
points to the generation of alien inferences advancing in the processing
of data and algorithms (data, metadata, big data). Here a materialist
approach to computation cannot be mainly concerned with the potentialities of computation already existing in nature, but needs to address
the artificiality of an automated elaboration of data followed by an alien
epistemological production. Against the technocapitalist naturalization of
computation which appears to be an extension (or smooth continuation) of
the potentialities of nature, the computational environment of algorithms
defines the development of a semi-autonomous mode of reason involving
a level of abstraction of socio-cultural, economic, and political data: an
artificial environment generating its own rules through conceptual prehensions. The adaptation of this speculative conception of reason to explain
the transformation of automated logic however is here primarily intended
to argue against the holistic view of capital and technology, questioning the
dominant critique of instrumentalization of matter. Whether or not capital
is deemed to be one with computation, the articulation of the speculative
Computational logic and ecological rationality
93
function of reason could enable us both to interrogate this equivalence and
to theorize algorithmic actions as partaking of (and not representing) the
historical transformation of the generic function of reason today.
Notes
1
See Achim Menges and S. Ahlquist, eds, Computational Design Thinking
(London: John Wiley, 2011); Achim Menges, “Material Computation—
Higher Integration in Morphogenetic Design,” Architectural Design 82
(2) (2012). Lisa Iwamoto, Digital Fabrications: Architectural and Material
Techniques (Architecture Briefs) (New York: Princeton Architectural Press,
2009). Michael Weinstock, The Architecture of Emergence: The Evolution
of Form in Nature and Civilisation (London: John Wiley, 2010); Michael
Hensel, Achim Menges, Michael Weinstock, eds, Emergence: Morphogenetic
Design Strategies (London: John Wiley, 2004); Neill Spiller and Rachel
Armstrong, “Protocell Architecture,” Architectural Design 81 (2011).
2
On the paradigmatic constitution of second-order cybernetics, see Katherine
N. Hayles, How We Became Posthuman (Chicago: University of Chicago
Press, 1999).
3
See Michael Hensel and Achim Menges, Morpho-Ecologies: Towards
Heterogeneous Space In Architecture Design (London: AA Publications,
2007); Menges, “Material Computation.”
4
See Lev Manovich, Software Takes Command (London: Bloomsbury 2013);
Friedrich Kittler, Literature, Media, Information Systems, trans. John
Johnston (New York: Routledge, 2013).
5
See Alfred North Whitehead, The Function of Reason (Boston: Beacon
Press, 1929).
6
See Mario Carpo, The Alphabet and the Algorithm (Cambridge, MA: MIT
Press, 2011).
7
It has been argued that “algorithms are mathematical and thus abstract
structures, but should not be mistaken for algebraic formulae, since
assignments or instructions operated by algorithms are non-reversible.”
See Shintaro Miyazaki, “Algorithmics: understanding micro-temporalities
in computational cultures,” Computational Culture. A Journal of Software
Studies 2 (2012). Available online: http://computationalculture.net/article/
algorhythmics-understanding-micro-temporality-in-computational-cultures
(accessed April 2015).
8
See Kostas Terzidis, Algorithmic Architecture (Oxford: Architectural Press,
2006). Michael Meredith, Tomoto Sakamoto, and Albert Ferre, eds, From
Control to Design: Parametric/Algorithmic Architecture (Barcelona: Actar,
2008).
9
See John Lobell, “The Milgo Experiment: An Interview with Haresh
Lalvani,” Architectural Design 76 (4) (2006): 57.
10
Ibid., 58.
GENERAL ECOLOGY
94
11
Ibid.
12
See Edward Fredkin, “Finite Nature,” Proceedings of the XXVIIth
Rencontre de Moriond (1992). Available at the website “Digital
Philosophy” http://www.digitalphilosophy.org/Home/Papers/FiniteNature/
tabid/106/Default.aspx (accessed April 2015).
13
See Neil Leach, “Swarm Urbanism,” Architectural Design 79 (4) (2009).
14
For a historical overview, see Hayles, How We Became Posthuman, Ch. 6;
N. Katherine Hayles, My Mother Was a Computer: Digital Subjects and
Literary Texts (Chicago: University of Chicago Press, 2005); Andrew
Pickering, The Cybernetic Brain: Sketches of Another Future (Chicago:
University of Chicago Press, 2010).
15
Greg Lynn, Folding in Architecture (London: John Wiley-Academy,
2004); Greg Lynn, Animate Form (New York: Princeton Architectural Press,
2011).
16
Leach, “Swarm Urbanism,” 58.
17
Sanford Kwinter, quoted in Menges, “Material Computation,” 16.
18
Menges, “Material Computation.”
19
Menges specifically refers to Philip Ball, Nature’s Patterns: A Tapestry in
Three Parts (2009). Menges, “Material Computation,” 17.
20
Menges refers to J. Scott Turner, The Extended Organism (2000) and
The Thinker Accomplice (2007) in which Claude Bernard’s homeostatic
machine, centered on physiological and energy-oriented evolution of
material (bones for instance), is opposed to Darwin’s selective mechanisms.
Menges, “Material Computation,” 17.
21
Ibid.
22
Ibid.
23
Ibid.
24
Ibid., 20.
25
Ibid., 17.
26
Ibid., 20.
27
From this standpoint, it is symptomatic that Jakob von Uexküll’s work
has a special place in the neo-materialist approach to computational space.
In particular, his concept of the Umwelt (environment), also central to
the development of system theory and cybernetics, explains that while
environments are shared, the experienced Umwelt is unique to the organism’s
sensory and affective configuration. This means not that there is one
environment differently adapted to specific organisms, but that the singular
arrangement of affects and percepts of any organism is directly generated
by environmental configurations. See Jakob von Uexküll, “An Introduction
to Umwelt”, in Space Reader. Heterogeneous Space in Architecture, ed. M.
Hensel, M. Hight, and A. Menges (London: Wiley, 2009), 145–8.
28
See Dorion Sagan, “Umwelt After Uexküll,” in Jakob von Uexküll, A Foray
Into the Worlds of Animals and Humans: With a Theory of Meaning, trans.
Joseph D. O’Neil (Minneapolis: University of Minnesota Press, 2010).
Computational logic and ecological rationality
95
29
It would be interesting to explore this theory of Umwelt in relation to
J. J. Gibson’s theory of embedded perception based on environmental
affordances, due to certain strong intellectual affinities between these milieucentered perspectives on the evolution of human or animal perceptual
cognition. James J. Gibson, The Ecological Approach to Visual Perception
(Boston: Houghton Mifflin, 1979).
30
According to Massumi, power can be now defined in terms of the operative
functions of both neoconservatism and neoliberalism. He points to a
symbiotic relationship between neoliberalism and neoconservatism, which
remains however a counterpoint between an investment in life productivity
and the technical determination of any incipient productive force of life. For
Massumi, they are both part of the “proto-territory of life”: a primordial
yet advanced plane that has no bounds. See Brian Massumi, “National
Enterprise Emergency: Steps Toward an Ecology of Powers,” Theory,
Culture and Society 26 (6) (2009): 175.
31
Massumi is referring specifically to Michel Foucault, The Birth of
Biopolitics: Lectures at the Collège de France, 1978–1979, trans. Graham
Burchell (New York: Palgrave Macmillan, 2008), 261; in Massumi,
“National Enterprise,” 155.
32
Ibid., 4.
33
Ibid., 6.
34
Massumi’s notion of naturing nature is inspired by Baruch Spinoza’s
conception of nature defined in terms of a substance in a process of
changing itself by extending within modes, which are modalities,
modifications of substance itself. There is no a priori axiomatics that
guarantees a return to unchanging truths. Instead, as Deleuze’s reading of
Spinoza radically claims, the manners or modalities of nature coincide with
the operations of change, the nurturing that constitutes nature in terms
of becoming. Similarly, Massumi explains that naturing nature involves
a “reiterative playing out its formative forcing” which itself generates
“conditions for a plurality of extensive distinctions and their iterative
regeneration. See Massumi, “National Enterprise,” 34.
35
This nature–culture continuum in which the rules of natural systems
correspond to the rules of war is explored at length by Manuel DeLanda,
War in the Age of Intelligent Machines (New York: Zone Books, 1992).
36
As Massumi clarifies “The emphasis on natured natures’ operative reality
and effective givenness distinguishes this concept from social constructivism’s
notion of naturalization … as presupposed as it is produced, given for the
making and made a given.” Massumi, “National Enterprise,” 34.
37
Massumi is directly referring to Deleuze and Guattari’s concept of
Mechanosphere. Massumi, “National Enterprise,” 34. See also Gilles
Deleuze and Félix Guattari, A Thousand Plateaus, Capitalism and
Schizophrenia (London: The Athlone Press, 1987), 69.
38
The notion of “transcendental empiricism” was first developed by Gilles
Deleuze, Empiricism and Subjectivity, trans. Constantin Boundas (New
York: Columbia University Press, 1991).
96
GENERAL ECOLOGY
39
Massumi developed this concept in several writings. See Brian Massumi,
“Potential Politics and the Primacy of Preemption,” Theory and Event 10
(2) (2007); Brian Massumi, “The Future Birth of the Affective Fact,” in The
Affect Reader, ed. Greg Seigworth and Melissa Gregg (Durham, NC: Duke
University Press, 2010). See also “National Enterprise,” 27–9.
40
My understanding of algorithmic reason in terms of prehension is indebted
to Whitehead’s argument against perception understood in terms of mental
representation and image impression, but also away from being pure
sensori-motor receptor. Prehensions are between the physical registering of
data and conceptual elaborations. Prehensions are also constitutive of the
concrescence of actual occasions insofar as to prehend involves not only
continuity but also discontinuity of being. To prehend implies stability of
organization but also necessary transformation. Prehension is an activity of
selection and evaluation of data that operates through a physical and mental
pole, and ultimately constitutes the subjective form of what is prehended.
41
Alan M. Turing, “On Computable Numbers, with an Application to the
Entscheidungsproblem,” Proceedings of the London Mathematical Society,
42 (2nd series: 1936): 230–65.
42
See Gregory Chaitin, Meta Math! The Quest for Omega (New York:
Pantheon, 2005).
43
Gregory Chaitin’s notion of an incomputable limit is influenced by the
nineteenth-century physicist Ludwig Boltzmann, who defined entropy as
a measurement of the degree of disorder, chaos, and randomness in any
physical system. See Chaitin, Meta Math!, 169–70.
44
Andrey Kolmogorov’s complexity theory is considered to have pioneered
the field of algorithmic information theory. See Ming Li and Paul Vitanyi,
eds, An Introduction to Kolmogorov Complexity and Its Applications,
Third Edition (New York: Springer Verlag, 2008). Gregory J. Chaitin,
“Randomness and Mathematical Proof,” Scientific American 232 (5) (1975).
45
See Christian S. Calude and Gregory Chaitin, “Randomness Everywhere,”
Nature 400 (1999). Gregory J. Chaitin, Exploring Randomness (London:
Springer Verlag, 2001), 22; Chaitin, Meta Math!
46
Gregory J. Chaitin, “The Limits of Reason,” Scientific American 294 (3)
(2006).
47
As opposed to cognitive theories of computation, according to which
to compute is to cognize and thus to produce a mental map of the data
gathered by the senses, and computational theories of cognition, for which
to think is a binary affair determined by pre-set sequences of logical steps,
I draw on Whitehead’s notion of prehension. According to Whitehead,
prehensions are first of all mental and physical modalities of relations by
which objects take up and respond to one another. Alfred North Whitehead,
Process and Reality: An Essay in Cosmology (New York: Free Press, 1978),
23–6.
48
Alfred North Whitehead, The Function of Reason (Boston: Beacon Press,
1929).
49
Whitehead, Process and Reality, 230.
Computational logic and ecological rationality
97
50
The process of concrescence of an actual entity is therefore defined by a
subjective aim driving the entity to become a unity, to reach satisfaction and
then to perish (i.e., the actual entity then becomes objective data that can be
prehended by another entity). Whitehead, Process and Reality, 22, 104.
51
See Whitehead, Process and Reality; on efficient cause, 237–8; on final
cause, 241; on the transition from efficient to final cause, 210.
52
It may be possible to explain this process in more detail by borrowing
Charles Sanders Peirce’s notion of abduction, referring to a particular kind
of non-deductive inference that involves the generation and evaluation
of explanatory hypotheses. For a recent elaboration of Peirce’s notion
of abduction in the context of computation see Lorenzo Magnani,
Abductive Cognition. The Epistemological and Eco-cognitive Dimensions
of Hypothetical Reasoning (Berlin: Springer Verlag, 2009), 1–41.
Material computation, mainly relying on the inductive logic of physical
interconnections, problematically omits the abstractions carried out by
conceptual prehension for which there can be no direct observation,
intuition or immediate experience. Whitehead, The Function of Reason, 25.
53
On non-conscious cognition in digital media see, Katherine N. Hayles
“Cognition Everywhere: The Rise of the Cognitive Nonconscious and the
Costs of Consciousness,” New Literary History 45 (2) (2014).
Bibliography
Calude, Christian S. and Gregory Chaitin. “Randomness Everywhere.” Nature
400 (1999): 319–20.
Carpo, Mario. The Alphabet and the Algorithm. Cambridge, MA: MIT Press,
2011.
Chaitin, Gregory. “The Limits of Reason.” Scientific American 294 (3) (2006):
74–81.
Chaitin, Gregory. “Randomness and Mathematical Proof.” Scientific American,
232 (5) (1975): 47–52.
Chaitin, Gregory. Exploring Randomness. London: Springer Verlag, 2001.
Chaitin, Gregory. Meta Math! The Quest for Omega. New York: Pantheon,
2005.
DeLanda, Manuel. War in the Age of Intelligent Machines. New York: Zone
Books, 1992.
Deleuze, Gilles and Félix Guattari. A Thousand Plateaus, Capitalism and
Schizophrenia. London: The Athlone Press, 1987.
Deleuze, Gilles. Empiricism and Subjectivity, trans. Constantin Boundas. New
York: Columbia University Press, 1991.
Foucault, Michel. The Birth of Biopolitics: Lectures at the Collège de France,
1978–1979, trans. Graham Burchell. New York: Palgrave Macmillan, 2008.
Fredkin, Edward. “Finite Nature.” Proceedings of the XXVIIth Rencontre de
Moriond (1992).
Gibson, James J. The Ecological Approach to Visual Perception. Boston:
Houghton Mifflin, 1979.
98
GENERAL ECOLOGY
Hayles, N. Katherine. “Cognition Everywhere: The Rise of the Cognitive
Nonconscious and the Costs of Consciousness.” New Literary History 45 (2)
(2014).
Hayles, N. Katherine. How We Became Posthuman. Chicago: University of
Chicago Press, 1999.
Hayles, N. Katherine. My Mother Was a Computer: Digital Subjects and Literary
Texts. Chicago: University of Chicago Press, 2005.
Hensel, Michael and Achim Menges. Morpho-Ecologies: Towards Heterogeneous
Space In Architecture Design. London: AA Publications, 2007.
Hensel, Michael, Achim Menges, and Michael Weinstock, eds. Emergence:
Morphogenetic Design Strategies. London: John Wiley & Sons, 2004.
Iwamoto, Lisa. Digital Fabrications: Architectural and Material Techniques
(Architecture Briefs). New York: Princeton Architectural Press, 2009.
Kittler, Friedrich. Literature, Media, Information Systems, trans. John Johnston.
New York: Routledge, 2013.
Leach, Neil. “Swarm Urbanism.” Architectural Design, 79 (4) (2009): 50–6.
Li, Ming and Paul Vitanyi, eds. An Introduction to Kolmogorov Complexity and
Its Applications. Third Edition. New York: Springer Verlag, 2008.
Lobell, John. “The Milgo Experiment: An Interview with Haresh Lalvani.”
Architectural Design 76 (4) (2006): 52–61.
Lynn, Greg. Animate Form. Princeton: Princeton Architectural Press, 2011.
Lynn, Greg. Folding in Architecture. London: John Wiley-Academy, 2004.
Magnani, Lorenzo. Abductive Cognition. The Epistemological and Eco-cognitive
Dimensions of Hypothetical Reasoning. Berlin: Springer Verlag, 2009.
Manovich, Lev. Software Takes Command. London: Bloomsbury, 2013.
Massumi, Brian. “National Enterprise Emergency: Steps Toward an Ecology of
Powers.” Theory, Culture and Society 26 (6) (2009): 153–85.
Massumi, Brian. “Potential Politics and the Primacy of Preemption.” Theory and
Event 10 (2) (2007).
Massumi, Brian. “The Future Birth of the Affective Fact.” In The Affect Reader,
ed. Greg Seigworth and Melissa Gregg. Durham, NC: Duke University Press,
2010.
Menges, Achim and S. Ahlquist, eds. Computational Design Thinking. London:
John Wiley, 2011.
Menges, Achim. “Material Computation—Higher Integration in Morphogenetic
Design.” Architectural Design 82 (2) (2012): 14–21.
Meredith, Michael, Tomoto Sakamoto, and Albert Ferre, eds. From Control to
Design: Parametric/Algorithmic Architecture. Barcelona: Actar, 2008.
Miyazaki, Shintaro. “Algorithmics: understanding micro-temporalities in
computational cultures.” Computational Culture. A Journal of Software
Studies, no. 2 (2012). Available online: http://computationalculture.net/article/
algorhythmics-understanding-micro-temporality-in-computational-cultures
(accessed April 2015).
Pickering, Andrew. The Cybernetic Brain: Sketches of Another Future. Chicago:
University of Chicago Press, 2010.
Sagan, Dorion. “Umwelt After Uexküll.” In A Foray Into the Worlds of Animals
and Humans: With a Theory of Meaning, Jakob von Uexküll, trans. Joseph D.
O’Neil, 1–34. Minnesota: University of Minnesota Press, 2010.
Computational logic and ecological rationality
99
Spiller, Neill and Rachel Armstrong. “Protocell Architecture,” Architectural
Design 81 (2011).
Terzidis, Kostas. Algorithmic Architecture. Oxford: Architectural Press, 2006.
Turing, Alan M. “On Computable Numbers, with an Application to the
Entscheidungsproblem.” Proceedings of the London Mathematical Society 42
(2nd series 1936): 230–65.
Uexküll, Jakob von. “An Introduction to Umwelt.” In Space Reader:
Heterogeneous Space in Architecture, ed. M. Hensel, M. Hight, and A.
Menges, 145–8. London: Wiley, 2009.
Weinstock, Michael. The Architecture of Emergence: The Evolution of Form in
Nature and Civilisation. London: John Wiley, 2010.
Whitehead, Alfred N. Process and Reality: An Essay in Cosmology. New York:
Free Press, 1978.
Whitehead, Alfred N. The Function of Reason. Boston: Beacon Press, 1929.