[7]
Instrumental Reason,
Algorithmic Capitalism,
and the Incomputable
Luciana Parisi
Algorithmic cognition is central to today’s capitalism.
From the rationalization of labor and social relations
to the financial sector, algorithms are grounding a
new mode of thought and control. Within the context
of this all-machine phase transition of digital capitalism, it is no longer sufficient to side with the critical theory that accuses computation to be reducing
human thought to mere mechanical operations. As
information theorist Gregory Chaitin has demonstrated, incomputability and randomness are to be
conceived as very condition of computation. If technocapitalism is infected by computational randomness
and chaos, the traditional critique of instrumental
rationality therefore also has to be put into question:
the incomputable cannot be simply understood as
being opposed to reason.
In Alleys of Your Mind: Augmented Intellligence and Its Traumas, edited by Matteo Pasquinelli,
125–37. Lüneburg: meson press, 2015.
DOI: 10.14619/014
126
Alleys of Your Mind
In the September 2013 issue of the journal Nature, a group of physicists from
the University of Miami published the article “Abrupt rise of new machine
ecology beyond human response time.” In the article, they identified a transition to “a new all-machine phase” ( Johnson et al. 2013) of financial markets,
which coincided with the introduction of high frequency stock trading after
2006. They argued that the sub-millisecond speed and massive quantity of
algorithm-to-algorithm interactions exceeds the capacity of human interactions. Analyzing the millisecond-scale data at the core of financial markets in
detail, they discovered a large number of sub-second extreme events caused
by those algorithms, whose proliferation they correlated with the financial
collapse of 2008.
In this new digital environment of trading, algorithmic agents make decisions
faster than humans can comprehend. While it takes a human at least one full
second to both recognize and react to potential danger, algorithms or bots
can make a decision on the order of milliseconds. These algorithms form “a
complex ecology of highly specialized, highly diverse, and strongly interacting agents” (Farmer and Skouras, 2011), operating at the limit of equilibrium,
outside of human control and comprehension.
The argument I develop here takes this digital ecology of high-frequency
trading algorithms as a point of departure. Thus, my text is not specifically
concerned with the analysis of the complex financial ecology itself, but aims
more directly to discuss a critique of automated cognition in the age of algorithmic capitalism. For if financial trading is an example of a digital automation
that is increasingly autonomous from human understanding, this system has
become a second nature. Therefore it seems to be urgent today to ask: What is
the relation between critical thought vis-à-vis those digital ecologies?
My question is: Can the critique of instrumental rationality—as addressed
by Critical Theory—still be based on the distinction between critical thinking
and automation? Can one truly argue that algorithmic automation is always
already a static reduction of critical thinking? By answering these questions,
we cannot overlook an apparent dilemma: Both, philosophical thought and
digitality, rely on principles of indetermination and uncertainty while featuring
these principles in their core complexity theories. As such, both challenge and
define the neoliberal order at the same time—a paradox.
To question this paradox, I will turn to the notion of incomputability as theorized by computer scientist Gregory Chaitin, who contributed to the field of
algorithmic information theory in his discovery of the incomputable number
Omega. This number has a specific quality: it is definable but not computable. In other words, Omega defines at once a discrete and an infinite state of
computation occupying the space between zero and one. From a philosophical perspective, the discovery of Omega points to a process of determination
Instrumental Reason, Algorithmic Capitalism and the Incomputable
of indeterminacy involving not an a priori structure of reasoning but more
importantly a dynamic processing of infinities in which results are not contained in the logical premises of the system.
This centrality of the incomputable in information theory, I suggest, brings
not only the philosophical critique of technical rationalization into question,
but also the instrumentalization of reason. Thus, in the following text I argue
that it is no longer sufficient to side with the critical view of technoscience
on the basis that computation reduces human thought to mere mechanical
operations. Instead, the paradox between realist philosophy and the realism
of technocapital can be read as a symptom of an irreversible transformation
in the history of critical thought in which the incomputable function of reason
has entered the automated infrastructure of cognition.
The Algorithms of Cognitive and Affective Capital
Capital has been said to have entered all aspects of personal and social life.
Before explaining the question of the incomputable in algorithmic automation, it is important to point out that with the so-called technocapitalist phase
of real subsumption, digital automation has come to correspond to cognitive
and affective capital. With this, the logic of digital automation has entered the
spheres of affects and feelings, linguistic competences, modes of cooperation,
forms of knowledge, as well as manifestations of desire. Even more, human
thought itself is said to have become a function of capital. Our contemporary understanding of this new condition in terms of “social capital,” “cultural
capital,” and “human capital” explains that knowledge, human intelligence,
beliefs, and desires have only instrumental value and are indeed a source of
surplus value. In this automated regime of affection and cognition, capacities
are measured and quantified through a general field defined by either money
or information. By gathering data and quantifying behaviors, attitudes, and
beliefs, the neoliberal world of financial derivatives and big data also provides
a calculus for judging human actions, and a mechanism for inciting and directing those actions.
Paradoxically, in the time when “immaterial labor” is privileged over material production (Hardt and Negri 2000), and when marketing is increasingly
concerned with affective commodities such as moods, lifestyles, and “atmospheres” (Biehl-Missal 2012), capitalist realism seems to be fully expressed
(Fisher 2009), guided by the findings of cognitive psychology and philosophy of
mind. Central to these findings is the plasticity of the neural structure as well
as the extension of cognitive functions—from perception to the capacity to
choose and to judge—through algorithm-based machines. It is not difficult to
see that nowadays the social brain is nothing else than a machine ecology of
algorithmic agents.
127
128
Alleys of Your Mind
A different aspect is discussed by Stiegler’s view of technocapital. He sees
thinking and feeling as the new motors of profit, which are repressed or captured by capital and transformed into mere cognitive and sensory functions
(2014). In other words, technocapital is what denies desire and knowledge,
reason and sensation. Instead, it reduces these potentialities to mere probabilities determined by the binary language of yes and no, zero and one.
Exploring this further, Lazzarato (2012) has argued that a critique of technocapital can focus neither on the capitalization of cognition nor its automation.
In The Making of the Indebted Man, Lazzarato (2012) maintains that knowledge
exercises no hegemony over the cycle of value, because knowledge (and thus
thought) is primarily subject to the command of financial capital. Here, the
neoliberal form of capital in its current phase of real subsumption corresponds to the production of a new condition: the general indebtedness. This
form of neoliberalism governance has entered all classes, even those that do
not own anything. Hence, the most universal power relationship today is that
of debtor and creditor. Debt is a technology of government sustained by the
automated apparatus of measuring and evaluation (credit reports, assessments, databases, etc.). Lazzarato understands this axiomatic regime in terms
of a semiotic logic, whose core scientific paradigm and technological applications are always already functioning to capture (by quantifying in values)
primary aesthetic potentials.
From this perspective, automation is the semiotic logic par excellence, which
does not simply invest labor and its cognitive and affective capacities, but
more specifically becomes a form of governmentality, which operates algorithmically to reduce all existence to a general form of indebtedness. This algorithmic form of governability is also what has given way to a diffused financialization of potentialities through which aesthetic life is constantly quantified and
turned into predictable scenarios.
Not only Lazzarato, also Massumi (2007) has noted the diffused ecological
qualities of this new form of algorithmic governmentality, which he describes
in terms of pre-emption, a mode of calculation of potential tendencies instead
of existing possibilities. The calculation of potentialities that describe this
dynamism is no longer based on existing or past data. Instead it aims at
calculating the unknown as a relational space by measuring the interval
between one existing data and another. This form of pre-emptive calculus
indeed transforms the limit point of this calculation—infinities—into a source
of capitalization.
From this standpoint, one can suggest the following: Contrary to the logic of
formal subsumption, which corresponds to the application of unchanging sets
of rules, whose linearity aimed to format the social according to pre-ordained
ideas, the logic of real subsumption coincides with the interactive computational paradigm. This paradigm is based on the responsive capacities of
Instrumental Reason, Algorithmic Capitalism and the Incomputable
learning, openness, and adaptation defining human-machine interaction as
well as distributed interactive systems. With the extension of quantification
to the indetermination of the environments—and thus to contingency—an
intrinsic transformation of the logic of calculation has happened. In fact, the
development of this interactive approach has been crucial to the now dominant form of real subsumption.
Historically, interactive algorithms were invented to circumvent the algorithmic constraints of the Turing Machine. The concept of this machine was
insufficient or unable to cope with the complexity of the empirical world—a
complexity that one could say, philosophically speaking, has its own nonrepresentational logic. Here, the advance of real subsumption cannot be isolated
from the emergence of a dynamic form of automation, which constitutes a
historical development in computer science from Turing’s algorithmic modeling. Back then, Turing’s conceptualization of a mechanism, which is based on
a priori instructions, strongly resonated with a mechanism as defined by first
order cybernetics (a closed system of feedback). Today, the combination of
environmental inputs and a posteriori instructions proposed by the interactive paradigm embrace second order cybernetics and its open feedback
mechanisms. The goal of this new dynamic interaction is to include variation
and novelty in automation to enlarge the horizon of calculation, and to include
qualitative factors as external variables within its computational mechanism.
Contrary to Lazzarato’s critique, it seems important not to generalize automation as being always already a technocapitalist reduction of existential
qualities. The task is rather to address the intrinsic transformation of the automated form of neoliberal governability and to engage closely with the question of the technical. However, rather than arguing that the technical is always
already a static formal frame, delimited by its binary logic, I suggest that
there is a dynamic internal to the system of calculation. If so, it is necessary
to engage with the real possibility of a speculative question that according
to Isabelle Stengers (2010 and 2011) is central to the scientific method: What
if automation already shows that there is a dynamic relation intrinsic to
computational processing between input data and algorithmic instructions,
involving a non-linear elaboration of data? What if this dynamic is not simply
explainable in terms of its a posteriori use, i.e., once it is either socially used or
mentally processed?
The interactive paradigm concerns the capacity of algorithms to respond and
to adapt to its external inputs. As Deleuze (1992) already foresaw, an interactive system of learning and continuous adaptation is at the core of the logic
of governance driven by the variable mesh of continuous variability. Here, the
centrality of capitalism in society forces axiomatics to open up to external outputs, constituting an environment of agents through which capital’s logic of
governance increasingly corresponds to the minute investment in the socius
129
130
Alleys of Your Mind
and ultimately life variations. The question of the undecidable proposition
is important, because it defines an immanent and not transcendent view of
capital, as Deleuze and Guattari (1987) remind us. This is the case in so far as
the extension of capital to life requires its apparatus of capture to be open to
contingencies, variations and unpredictable change.
It is here that the organizational power of computation needs to be more
closely investigated to clarify the transformation that automation itself has
undergone with the re-organization of capital from formal to real subsumption. Interactive automation of cognition and affection should be examined
anew. Whether we are faced with the critical conception of cognitive capital,
or with the critical view of an automated governance based on a general
indebtedness, we risk overlooking what can be considered the most radical
process of artificialization of intelligence that human history has ever seen;
this involves the conversion of organic ends into technical means, whose consequences are yet to become unpacked.
Although my thoughts are still in an early phase, I want to consider the possibility of theorizing that algorithmic automation heralds the realization of a
second nature, in which a purposeless and impersonal mode of thought tends
to supplant the teleological finality of reason, echoed by Kant’s conception
of reason in terms of motive—i.e., the reason behind the action—that substantiates the difference between understanding and reason. This is also a
proposition, which more importantly works to challenge the theory that there
is a mutual relation or undecidable proposition between philosophy and technology as well as between thought and capital. Instead of the idea that the
refuge of thought and of philosophy from an increasingly dynamic technocapitalism lies in the ultimate appeal to intellectual intuition and affective thought
as the safe enclaves of pure uncertainty and singularity, I want to pursue the
possibility that algorithmic automation—as rule-based thought—may rather
be indifferent to these all too human qualities, whilst actively encompassing
them all without representing philosophical and or critical thought. This is a
proposition for the emergence of an algorithmic mode of thought that cannot
be contained by a teleological finality of reason, which characterizes both
capitalism and the critique of technocapitalism.
The Turing Experiment and the Omega Number
As we know, algorithmic automation involves the breaking down of continuous
processes into discrete components, whose functions can be constantly reiterated without error. In short, automation means that initial conditions can
be reproduced ad infinitum. The form of automation that concerns us here
was born with the Turing Machine: an absolute mechanism of iteration based
on step-by-step procedures. Nothing is more opposed to pure thought—or
Instrumental Reason, Algorithmic Capitalism and the Incomputable
“the being of the sensible” as Deleuze (1994: 68) called it—than this discretebased machine of universal calculation. The Turing architecture of prearranged units that could be interchangeably exchanged along a sequence is
effectively the opposite of an ontogenetic thought moving through a differential continuum, through intensive encounters and affect.
Nevertheless, since the 1960s the nature of automation has undergone
dramatic changes as a result of the development of computational capacities
of storing and processing data. Previous automated machines were limited
by the amount of feedback data. Now algorithmic automation is designed to
analyze and compare options, to run possible scenarios or outcomes, and
to perform basic reasoning through problem-solving steps that were not
contained within the machine’s programmed memory. For instance, expert
systems draw conclusions through search techniques, pattern matching, and
web data extraction, and those complex automated systems have come to
dominate our everyday culture, from global networks of mobile telephony to
smart banking and air traffic control.
Despite this development, much debate about algorithmic automation is still
based on Turing’s discrete computational machine. It suggests that algorithmic automation is yet another example of the Laplacian view of the universe,
defined by determinist causality (see Longo 2000 and 2007). But in computational theory, the calculation of randomness or infinities has now turned
what was defined as incomputables into a new form of probabilities, which
are at once discrete and infinite. In other words, whereas algorithmic automation has been understood as being fundamentally Turing’s discrete universal
machine, the increasing volume of incomputable data (or randomness) within
online, distributive, and interactive computation is now revealing that infinite,
patternless data are rather central to computational processing. In order
to appreciate the new role of incomputable algorithms in computation, it is
necessary to make a reference to the logician Kurt Gödel, who challenged the
axiomatic method of pure reason by proving the existence of undecidable
propositions within logic.
In 1931, Gödel took issue with Hilbert’s meta-mathematical program. He
demonstrated that there could not be a complete axiomatic method, not a
pure mathematical formula, according to which the reality of things could
be proven to be true or false (see Feferman 1995). Gödel’s incompleteness
theorems explain that propositions are true, even though they cannot be verified by a complete axiomatic method. Propositions are therefore deemed to
be ultimately undecidable: They cannot be proven by the axiomatic method
upon which they were hypothesized. In Gödel’s view, the problem of incompleteness, born from the attempt to demonstrate the absolute validity of pure
reason and its deductive method, instead affirms the following: No a priori
131
132
Alleys of Your Mind
decision, and thus no finite sets of rule, can be used to determine the state of
things before things can run their course.
Turing encountered Gödel’s incompleteness problem while attempting to
formalize the concepts of algorithm and computation through his famous
thought experiment, now known as the Turing Machine. In particular, the
Turing Machine demonstrates that problems are computable, if they can
be decided according to the axiomatic method.1 Conversely, those propositions, which cannot be decided through the axiomatic method, will remain
incomputable.
By proving that some particular functions cannot be computed by such a
hypothetical machine, Turing demonstrated that there is not an ultimate decision method of the guise that Hilbert had wished for. The strength of Turing’s
proposition is that his Turing Machine offered a viable formalization of a
mechanical procedure. Instead of just crunching numbers, Turing’s computing
machines—and indeed contemporary digital machines that have developed
from them—can solve problems, make decisions, and fulfill tasks; the only
provision is that these problems, decisions, and tasks are formalized through
symbols and a set of discrete and finite sequential steps. In this respect,
Turing’s effort can be seen as a crucial step in the long series of attempts in
the history of thought geared towards the mechanization of reason.
However, what is more important is how the limit of computation and thus of
the teleological finality of reason—automated in the Turing machine—have
been transformed in computer science and information theory. Here, the work
of mathematician Gregory Chaitin (2004, 2006, and 2007) is particularly symptomatic of this transformation as it explains what is at stake with the limits of
computation and the development of a dynamic form of automation. Distinguishing this transformation from the centrality of the interactive paradigm in
technocapitalism is crucial. This paradigm, born from the necessity to include
environmental contingencies in computation, mainly works to anticipate or
pre-empt response (as Massumi 2007 has clearly illustrated). Instead, and
more importantly for me and my proposition of algorithmic automation as a
mode of thought, it is a serious engagement with the function that incomputable data play within computation. To make this point clearer, I will have to
explain Chaitin’s theory in greater detail.
Chaitin’s algorithmic information theory combines Turing’s question of the
limit of computability with Shannon’s information theory demonstrating the
productive capacity of noise and randomness in communication systems,
to discuss computation in terms of maximally unknowable probabilities. In
every computational process, he explains, the output is always greater than
1
See Turing 1936. For further discussion of the intersections of the works between Hilbert,
Gödel and Turing, see Davis 2000.
Instrumental Reason, Algorithmic Capitalism and the Incomputable
the input. For Chaitin, something happens in the computational processing of data, something that challenges the equivalence between input and
output, and thus the very idea that processing always leads to an already
pre-programmed result. This something is, according to Chaitin, algorithmic
randomness. The notion of algorithmic randomness implies that information
cannot be compressed into a smaller program, insofar as between input and
output an entropic transformation of data occurs, which results in a tendency
of these data to increase in size. From this standpoint, the output of the
processing does not correspond to the inputted instructions, and its volume
tends in fact to become bigger than it was at the start of the computation. The
discovery of algorithmic randomness in computational processing has been
explained by Chaitin in terms of the incomputable: increasing yet unknown
quantities of data that characterize rule-based processing.
Chaitin calls this algorithmic randomness Omega (the last letter of the Greek
alphabet refers to the probability that this number is infinite). Chaitin’s investigation of the incomputable reveals in fact that the linear order of sequential procedures (namely, what constitutes the computational processing of
zeros and ones) shows an entropic tendency to add more data to the existing
aggregation of instructions established at the input. Since this processing
inevitably includes not only a transformation of existing data into new inputs,
but also the addition of new data on top of what already was pre-established
in the computational procedure, it is possible to speak of an internal dynamic
to computation.
From this point of view, computational processing does not mainly guarantee the return to initial conditions, nor does it simply include change derived
from an interactive paradigm based on responsive outputs. This is because
Chaitin’s conception of incomputability no longer perfectly matches the notion
of the limit in computation (i.e., limit for what is calculable). Instead, this limit
as the incomputable is transformed: It becomes the addition of new and maximally unknowable algorithmic parts to the present course of computational
processing; these parts are algorithmic sequences that tend to become bigger
in volume than programmed instruction and to take over, hereby irreversibly
transforming the pre-set finality of rules. Chaitin’s re-articulation of the incomputable is at once striking and speculatively productive. What was conceived
to be the external limit of computation (i.e., the incomputable) in Turing, has
now become internalized in the sequential arrangement of algorithms (randomness works within algorithmic procedures).
At Chaitin’s own admission, it is necessary to see algorithmic randomness as
a continuation of Turing’s attempt to account for indeterminacy in computation. Whereas for Turing there are cases in which finality cannot be achieved,
and thus computation—qua automation of the finality of reason—stops when
the incomputable begins, for Chaitin computation itself has an internal margin
133
134
Alleys of Your Mind
of incomputability insofar as rules are always accompanied and infected by
randomness. Hence, incomputability is not simply a break from reason, but
rather reason has been expanded beyond its limits to involve the processing
of maximally unknown parts that have no teleological finality. To put it in other
terms, automation is now demarcated by the incomputable, the unconditional
of computation. Importantly, however, this challenges the view that computational processing corresponds to calculations leading to pre-programmed and
already known outputs. Instead, the limits of automation—that is the incomputable—have become the starting point of a dynamism internal to computation, which exceeds the plan for technocapital’s instrumentalization of reason.
From this standpoint, relating Chaitin’s findings to the positioning of critical
thought and technocapitalism reveals a new aspect: the incomputable cannot
be simply understood as being opposed to reason. In other words, it is not
an expression of the end of reason and cannot be explained according to the
critical view that argues for the primacy of affective thought.
According to Chaitin, the incomputable demonstrates the shortcomings of the
mechanical view of computation, according to which chaos or randomness
is an error within the formal logic of calculation. But incomputables do not
describe the failure of intelligibility versus the triumph of the incalculable—on
the contrary. These limits more subtly suggest the possibility of a dynamic
realm of intelligibility, defined by the capacities of incomputable infinities
or randomness, to infect any computable or discrete set. In other words,
randomness (or the infinite varieties of infinities) is not simply outside the
realm of computation, but has more radically become its absolute condition.
And when becoming partially intelligible in the algorithmic cipher that Chaitin
calls Omega, randomness also enters computational order and provokes an
irreversible revision of algorithmic rules and of their teleological finality. It is
precisely this new possibility for an indeterminate revision of rules, driven by
the inclusion of randomness within computation, that reveals dynamics within
automated system and automated thought. This means the following: While
Chaitin’s discovery of Omega demonstrates that randomness has become
intelligible within computation, incomputables cannot, however, be synthesized by an a priori program or set of procedures that are in size smaller than
them. According to Chaitin, Omega corresponds to discrete states that are
themselves composed of infinite real numbers that cannot be contained by
finite axioms.
What is interesting here is that Chaitin’s Omega is at once intelligible yet nonsynthesizable by universals, or by a subject. I take it to suggest that computation—qua mechanization of thought—is intrinsically populated by incomputable data, or that discrete rules are open to a form of contingency internal
to algorithmic processing. This is not simply to be understood as an error
within the system, or a glitch within the coding structure, but rather as a part
Instrumental Reason, Algorithmic Capitalism and the Incomputable
of computation. Far from dismissing computation as the evil incarnation of
technocapitalist instrumentalization of reason, one realizes that incomputable
algorithms emerge to defy the superiority of the teleological finality of reason,
but also of sensible and affective thought.
Speculative Computation
It would be wrong to view this proposition that incomputables define the
dynamic form of automation with naïve enthusiasm. Instead, it is important
to address algorithmic automation without overlooking the fact that the computation of infinity is nonetheless central to the capitalization of intelligible
capacities—even in their automated form. My insistence that incomputables
are not exclusively those non-representable infinities, which belong to the
being of the sensible, is indeed a concern, with the ontological and epistemological transformation of thought in view of the algorithmic function of
reason. Incomputables are expressed by the affective capacities to produce
new thought, but more importantly reveal the dynamic nature of the intelligible. Here, my concern is not an appeal to an ultimate computational being
determining the truth of thought. On the contrary, I have turned to Chaitin’s
discovery of Omega, because it radically undoes the axiomatic ground of truth
by revealing that computation is an incomplete affair, open to the revision
of its initial conditions, and thus to the transformation of truths and finality.
Since Omega is at once a discrete and infinite probability, it testifies to the
fact that the initial condition of a simulation—based on discrete steps—is and
can be infinite. In short, the incomputable algorithms discovered by Chaitin
suggest that the complexity of real numbers defies the grounding of reason in
finite axiomatics and teleological finality.
From this standpoint, several thoughts unfold. I agree that the interactive paradigm of technocapitalism already points to a semi-dynamic form of automation, which has enslaved the cognitive and affective capacities and established
a financial governmentality based on debt. But beyond this, there still remain
further questions regarding the significance of algorithms.
If we risk confusing the clear-cut opposition between digitality and philosophy (Galloway 2013), what and how are algorithms? For now, I want to point
out that algorithms, this dynamic form of reason, rule-based and yet open
to be revised, are not defined by teleological finality, as impersonal functions transform such finality each time. This is not to be conceived as a mere
replacement or extension of human cognitive functions. Instead, my point is
that we are witnessing the configuration of an incomputable mode of thought
that cannot be synthesized into a totalizing theory or program. Nonetheless,
this thought exposes the fallacy of a philosophy and critical thought, which
135
136
Alleys of Your Mind
reduces computation to an inferior mechanization of reason, destined to mere
iteration and unable to change its final directions.
Here, my argument was mainly concerned with the critique of computation as
the incarnation of the technocapitalist instrumentalization of reason. It was
an attempt at suggesting the possibility that algorithmic automation coincides
with a mode of thought, in which incomputable or randomness have become
intelligible, calculable but not necessarily totalizable by technocapitalism.
Despite all instrumentalization of reason on behalf of capitalism, and despite
the repression of knowledge and desire into quantities, such as tasks, functions, aims, there certainly remains an inconsistency within computation. This
is the case insofar as the more it calculates, the more randomness (patternless information) it creates, which exposes the transformative capacities of
rule-based functions. In the algorithm-to-algorithm phase transition that most
famously characterizes the financial trading mentioned at the beginning of
this essay, it is hard to dismiss the possibility that the automation of thought
has exceeded representation and has instead revealed that computation itself
has become dynamic.
To conclude I want to add this: dynamic automation cannot be mainly
explained in terms of a necessary pharmacological relation between philosophy and technology, knowledge, and capital, or the conditional poison allowing for a mutual reversibility defined by a common ground as Stiegler (2014)
does. Similarly, one has to admit that the dynamic tendencies at the core of
algorithmic automation are not simply reducible to the technocapitalist logic
of semiotic organization declared by Lazzarato (2012) or to the exploitation/
repression of the cognitive-creative functions of thought.
The challenge that automated cognition poses to the post-human vision—that
thought and technology have become one, because of technocapitalism—
points to the emergence of a new alien mode of thought, able to change its
initial conditions and to express ends that do not match the finality of organic
thought. This also means that the algorithm-to-algorithm phase transition
does not simply remain another example of the technocapitalist instrumentalization of reason, but more subtly reveals a realization of a second nature
in the form of a purposeless and automated intelligence. If algorithmic
automation no longer corresponds to the execution of instructions, but to the
constitution of a machine ecology infected with randomness, then one can
suggest that neither technocapitalism nor the critique of technocapitalism can
contain the tendency of the automated processing of randomness to overcome axiomatic truths.
Instrumental Reason, Algorithmic Capitalism and the Incomputable
References
Biehl-Missal, Brigitte. 2012. “Atmospheres of Seduction: A Critique of Aesthetic Marketing
Practices.” Journal of Macromarketing 32 (2): 168–80.
Chaitin, Gregory. 2004. “Leibniz, Randomness & the Halting Probability.” Mathematics Today 40
(4): 138–39.
Chaitin, Gregory. 2006. “The Limits of Reason.” Scientific American 294 (3): 74–81.
Chaitin, Gregory. 2007. “The Halting Probability Omega: Irreducible Complexity in Pure Mathematics.” Milan Journal of Mathematics 75 (1): 291–304.
Davis, Martin. 2000. The Universal Computer. The Road from Leibniz to Turing. London: Norton and
Company, 83–176.
Deleuze, Gilles, and Félix Guattari. 1987. A Thousand Plateaus: Capitalism and Schizophrenia.
Minneapolis: University of Minnesota Press.
Deleuze, Gilles. 1994. Difference and Repetition. London: Athlone.
Deleuze, Gilles. 1992: “Postscript on the Societies of Control.” October 59: 3–7.
Farmer, Doyne, and Spyros Skouras. 2011. An Ecological Perspective on the Future of Computer
Trading. The Future of Computer Trading in Financial Markets, UK Foresight Driver Review 6.
London: Government Office for Science. http://www.gov.uk/government/uploads/system/
uploads/attachment_data/file/289018/11-1225-dr6-ecological-perspective-on-future-ofcomputer-trading.pdf.
Fisher, Mark. 2009. Capitalist Realism. Is There No Alternative?. London: Zero Books.
Galloway, Alexander. 2013. “The Poverty of Philosophy: Realism and Post-Fordism.” Critical
Inquiry 39 (2): 347–66.
Goldstein, Rebecca. 2005. Incompleteness: The Proof and Paradox of Kurt Gödel. Norton &
Company.
Gödel, Kurt. 1995. “Some basic theorems on the foundations of mathematics and their implications.” In Collected Works of Kurt Gödel. Vol. 3, edited by Solomon Feferman et al., 304–23.
Oxford: Oxford University Press.
Hardt, Michael and Negri, Antonio. 2000. Empire. Cambridge, MA: Harvard University Press.
Hayles, N. Katherine. 2014. “Cognition Everywhere: The Rise of the Cognitive Nonconscious and
the Costs of Consciousness.” New Literary History 45 (2): 199–320.
Johnson, Neil, Guannan Zhao, Eric Hunsader, Hong Qi, Nicholas Johnson, Jing Meng, and Brian
Tivnan. 2013. “Abrupt Rise of New Machine Ecology beyond Human Response Time.” Scientific
Reports 3 (September 11). doi:10.1038/srep02627.
Lazzarato, Maurizio. 2012. The Making of the Indebted Man. Los Angeles: Semiotext(e).
Longo, Giuseppe. 2000. “The Difference between Clocks and Turing Machines.” In Functional
Models of Cognition. Self-Organizing Dynamics and Semantic Structures in Cognitive Systems,
edited by Arturo Carsetti, 211–232. Dordrecht: Springer.
Longo, Giuseppe. 2007. “Laplace, Turing and the ‘Imitation Game’ Impossible Geometry: Randomness, Determinism and Programs in Turing’s Test.” In The Turing Test Sourcebook, edited
by Robert Epstein, Gary Roberts and Grace Beber, 377–413. Dordrecht: Kluwer.
Massumi, Brian. 2007. “Potential Politics and the Primacy of Preemption.” Theory & Event 10 (2).
Steiner, Christopher. 2012. Automate This: How Algorithms Came to Dominate the World. New York:
Portfolio/Penguin.
Stengers, Isabelle. 2010. Cosmopolitics 1. Minneapolis: University of Minnesota Press.
Stengers, Isabelle. 2011. Cosmopolitics 2. Minneapolis: University of Minnesota Press.
Stiegler, Bernard. 2014. States of Shock: Stupidity and Knowledge in the 21st Century. Cambridge:
Polity Press.
Stiegler, Bernard. 2014. The Lost Spirit of Capitalism: Disbelief and Discredit, 3. Cambridge: Polity
Press.
Turing, Alan. 1936. “On Computable Numbers, with an Application to the Entscheidungsproblem.” Proceedings of the London Mathematical Society, series 2, vol. 42, 230–65.
137