Spectators are linked solely by their one-way relationship to the very centre that keeps them isolated from each other. The spectacle thus reunites the separated, but it reunites them only in their separateness.
(Guy Debord, The Society of the Spectacle)
:: XANADU ::
When, in the 1970s, Ted Nelson drafted a blueprint of the universal hypertext archive on which his somewhat utopian Xanadu project was to be founded, he was – whether he fully appreciated this or not – participating in a teleology of “universal knowledge” extending back via the Library of Alexandria to the universalist conceptions of history and epistemology that emerged as a consequence of classical metaphysics. Like Buckminster Fuller’s contemporary notion of synergetics or the “geometry of thought,” Xanuadu was holistically conceived as the cybernetic analogue of Borges’ “Library of Babel.” Yet not only was Xanadu conceived as an interlinked electronic archive of everything, but as a kind of artificial intelligence in its own right: a maximalising technological prosthesis of “mind.” In short, Xanadu was to synthesise Turing’s conception of an electronic brain with the memory of the (human) universe. Above all, Xanadu was, at the very core of its architecture, textual – or rather, in Nelson’s coinage, hypertextual and hypermedial: a distributed system of non-linear “writing”; a “language machine.”
In the decades to come, the manifestation of this teleology in a myriad of quotidian “new media” seemed to refigure what had, in Nelson’s Xanadu, remained a social or even civilisational project into a subjective one. In place of the accomplished dream of universal knowledge, this “new media” contrived to be a prosthesis of individual experience. This was the view imparted by means of such terminologies as “interpassivity” which, revisiting the Platonic anxieties around mimēsis, posed contemporary image-recording and similar technologies as a turn away from active (or generative) “knowing” towards its simulacrum in passive image consumption (just as, in the Phaedrus, writing is supposed to deviate from speech, dialectic, presence and thus truth, through a kind of intellectual and spiritual passivity or death).
With the advent of the world wide web in the 1990s and its promise of a digital metaverse – in which the totality of data produced by humanity, the world and potentially the cosmos itself could be contained – these dual teleologies (the social and subjective) appeared on the cusp of an apotheosis. The formalist hypotheses of the Aristotelians, of the Renaissance humanists, of the modernists and techno-utopianists, appeared on the verge of an ultimate realisation: the godlike subsumption of “man” into a transcendental “pure reason.” The great evolutionary leap into the posthuman, etcetera. The term that Jacques Derrida applied to this particular delusion and its mass-manufactured hysterias was “archive fever.” Archive fever referred to a compulsive movement of hyper-totalisation of “all” possible knowledge, “all” possible experience past and future, within a technological matrix of instant, indiscriminate and infinite capture. A universal set in which the impossible (because its very premise is the impossibility of its ever being grasped within any framework of human experience) is put to work as simultaneously the engine of humanity’s self-supersession and of its (paradoxical) conservation. The posthuman as transcendental humanism.
The emergence of Large Language Model “artificial intelligence” does more than ramify this proposition. While Nelson’s vision of a universal electronic archive was discriminatory, which is to say positivistic (it contained only knowledge rather than non-knowledge or junk), critics of LLMs have tended to focus on the way in which artificial intelligence possesses an unprecedented capacity to contaminate the epistemological field with misinformation and error on a hyperbolic scale. This is the Platonic nightmare writ large: the irresponsible or even malevolent logos run amok in an environment beyond the logistical means of “humanity” to police. Where “archive fever” pointed to an innate ambivalence in the movement of totalisation to what was totalised (i.e. “everything” without discrimination), the aspirations of Xanadu and the criticisms directed at LLMs remain overwhelmingly invested in principles of valorisation. In the latter, the assumption remains that a “positive” bias needs to be instituted in artificial intelligence to prevent a catastrophic rupture across the entirety of the epistemological field (re-echoing all the catastrophisms of the past), while in the former we are presented with the challenge of grasping how bias as such arises even (or specifically) in the trope of (indiscriminate) totalisation itself. The movement of universal knowledge, on the one hand, and of valorisation ab nihil on the other, cannot, in other words, be separated from the recuperations of ideology – and this, in the catastrophic view of artificial intelligence is, as Derrida implies, its inescapable irony.
If we were to take the various doomsday scenarios about full-autonomous, self-replicating, self-evolving sentient AIs seriously, it would be difficult to sustain the belief that such AIs would retain any especial “concern” for humanity, malevolent or otherwise. And, insofar as such words would retain any meaning at all (other than as “posthumanist” metaphors), it would be difficult to support the assumption that such AIs would necessarily even “want” to “exist.” The generalisation of AI in broadly eugenic terms, while attributing to it (in general) subjective psychological traits, speaks to that universal-particularity in which the teleology of progress has always sought its locus.
A critical reappraisal of the ways in which digital technologies and algorithmic governance instate “human” experience, agency and social structures, is long overdue. With the advent of AI, theory finds itself at a crossroads, confronted by an “edge-of-the-construct,” which has ceased merely to be a metaphor for the phantasmatic relationship between the technē of representation and posthumanist transcendentalism, rather it designates theory’s own precarious situation, as prosthesis of reason and autonomous critical agency. This scenario, often depicted as a boundary between the human and the technological, reflects a preoccupation with simulationism and the control exerted by computational systems on “reality,” as well as a desire to recuperate this “beyond of experience” for a new existentialism, a new humanism.
It is a readymade cliché that the emergence of Large Language Models necessitates a re-evaluation of preconceptions about intelligence, consciousness and the role of humans in a technologically constituted world, et cetera. Yet if the rapid development of AI and hyperautomation challenges both anthropocentric as well as post-Anthropocenic conceptions of agency, it does this not by indicating the rapid dis-integration of “subjective experience” within a “consensual hallucination,” as William Gibson famously put it, of “reality” (modernism’s hand-me-down), but by disintegrating the very framework of “experience” in general and of “consensus” in particular. While terms like algorithmics and technicity are often affected to mean predetermined, end-orientated reductive systems that translate input into output, cause into effect, intention into action, their entire genealogy (from Aristotle to Mumford, Giedion, McLuhan and beyond) speaks to a poiēsis or poetics of spontaneity, indeterminacy, complexity. It isn’t merely that algorithms are generative, but that they are ambivalently so. Every apparent algorithmic bias is ambivalently determined. This extends to the arbitrary, stochastic and interoperable nature of “representation,” “experience” and “reality.”
Drawing from Althusser’s thesis on Ideological State Apparatuses and later conceptions of “capitalist realism,” we may posit a subjective experience and consensual reality as emergent from – and as – states of ambivalence, such that the “concreteness” of social relations posited by (e.g. Marxist) critical theory is seen to be deeply intertwined with ad hoc algorithmic governance rather than actualising or reifying an underlying political teleology. Likewise the history of panopticism, simulationism and the “society of the spectacle” (as theorised by Bentham, Debord, Foucault and Baudrillard). To be clear: what is here called the Algorithmic State Apparatus transgresses at every point the logic of panoptic surveillance under conditions of AI – of subjective experience and the consensual-real – producing human hypotheses (radically simulacral egotic artefacts) from solipsistic neuro-computational networks (theoretical-real Universal Turing Machines). This stateless control system operates in the place where ideology cannot see – in the recursive hyperspace between omniscience and the unverifiable; necessity and the impossible; Id to ideology’s Ego – contriving edifices of sheer concatenation and pseudo-continuity, at once totalising, autopoetic and indeterminate, to such a degree as to appear as if productive of all past, present and future reality.
:: :: EDGE-OF-THE-CONSTRUCT :: ::
You’re watching a film in which the protagonist is driving through a city, trying to escape, as if trapped inside a labyrinth. Eventually, after innumerable wrong turns and obstructions, they reach the proverbial end-of-the-road. The “city” ends precipitously and a starkly abstract Cartesian grid extends into infinite space. By now this scenario – and countless like it – has become the defining cliché of a whole genre of technosocial panic: the reification of an otherwise imaginary boundary between human and digital, as frontier myth of characters “trapped” inside a computer simulation. Such is the pervasiveness of this genre that students of Baudrillard like Achim Szepanski have been recently moved to assert that “the goal of every system or theory is to create a simulacrum of itself in space (other worlds) and in time (owner of the future).”1
But this scenario has long-ceased to be merely a work of science theory-fiction. On 17 January 2023, a New York Times headline read: “ARE WE LIVING IN A COMPUTER SIMULATION AND CAN WE HACK IT?”2 The idea that “the universe is a hologram, its margins lined with quantum codes that determine what is going on inside,” where the computational cosmic brane3 takes over the role customarily reserved – in magic, superstition, organised religion and the myth of the state – for “higher powers,” is hardly a novel idea. The distinction meant to be drawn from the New York Times article is that real science, not science fiction, is the domain (the “goal” according to Szepanski) of a cosmocapitalist algorithm. Moreover: that a contiguity exists between the causal determinism of cosmic supercomputers and a transcendental (financialised) cybernetics, an inflationary horizon across which homo catastrophicus universalises its “posthuman” destiny.
It seems that we are, as it were, on the cusp of a truly disturbing revelation, of which humanity – in a bizarre act of precognitive mimēsis – has until now played at being the author of :: Welcome to the “singularity.” ::
On 30 November 2022, when OpenAI released GPT to the public, doomsday predictions about rampant antonymous AIs wreaking havoc across the world (and beyond), were already being offered as foregone conclusions. Yet the sudden advent of LLMs or Large Language Model AIs, of which GPT became the instant and ubiquitous representative, necessitated a radical reconsideration of what such foregone conclusions might actually mean.
It’s by now commonplace that the forms of hyperautomation which have rapidly co-evolved with LLMs pose consequences far beyond humanity’s impending “extinction” at the hands of sentient machines (humanity appears to be accomplishing this end quite efficiently all by itself). Yet, it is no exaggeration to speak of an AI revolution, although it may be more correct to speak of a process occurring throughout the timeframe of the long Industrial Revolution, or what perhaps anomalously is still being referred to as the Anthropocene, Entropocene, or Capitalocene. If these terms deserve our interest, it is to the extent that they signal a deconstruction of humanist thought and the redistribution of its prerogatives (consciousness, intelligence, subjectivity), than the refurbishment of a genre of science fiction.
Already in the mid-twentieth century cyberneticians and quantum physicists had generalised the idea of information as a fundamental constituent of the universe: determinate of “reality” and not simply its descriptor, nor simply an artefact of its “effects.” The emergence of LLMs ramifies a number of implications stemming from this – among them, the “meaning” not only of what universal general intelligence may be, but of what the exercise of power ultimately entails in relation to such an intelligence and its possible operations. The convergence of intelligence, power and language isn’t a new topic (it’s the subject of Plato’s Phaedrus, c.370BC) – its manifestation as autonomous, hyperautomated technicity, however, is.
:: MIRRORWORLDS ::
In the expanded field of language, the question of intelligence is inseparable from the dynamics of signification – which is to say, the circulation and distribution of meaning, and thus power. Just as (for Derrida) every signified is always-already another signifier,4 so too every dynamic relation is always-already an algorithm (a system of force-feedback or ramified bias), just as every data-node is always-already another calculus, in advance of itself, anticipating its own feedback. It’s a dynamic evocative of Blakean possible worlds, of universes in grains of sand, coupled to a generalised “mirror dialectic” (Lacan) in which the simulacral “other” always-already sees “you” before you see “it.” There’s never a point at which a human protagonist stands in front of an empty mirror, waiting for its reflection to appear: before the protagonist (“the subject”) even knows what it is, its reflection is there, in exquisite detail, waiting to be recognised. We might say, in effect, that it is the protagonist that is always somehow in process of coming into view within a scenery that not only perfectly anticipates it (the protagonist), but produces both its (the protagonist’s) self-image and its perception of that image. If this “mirror dialectic” serves as a metaphor of an ideal artificial intelligence, this is because it assumes the form – not of a reply to the subject, but rather – of a precognition.
It’s only possible to approach the Algorithmic State-Apparatus by understanding that this precognitive “effect” is a characteristic of that reality in which the subject-as-protagonist is posited in the first place.
Something occurs within the very logic and structure of mimēsis that continues to generate resistance in the discourse around artificial intelligence (often in the form of an evangelical “posthumanism”), recalling similar resistance both to deconstruction and quantum mechanics (Einstein’s notorious rebuke to Max Born that “god doesn’t play dice”). Here, the signifier (logos) does not derive from a so-called signified (eidos), it produces a signified – and this signified is never itself more than contingent upon the significations it in turn must perform (contingent, therefore, upon its own possible future states). Ad infinitum. For this reason, it isn’t sufficient to envisage a simulacral “construct” emanating from some momentous artificial neural net, like a ghost or spirit, haunting or even taking the place of so called reality – for the simple reason that this exquisitely detailed “construct,” in order to not simply collapse in on its pseudo-autonomy, must be coterminous with the “apparatus” that produces it – the neural net – the signifying system – the mimetic economy – the mirror dialectic – the semiosphere – the “artificial intelligence,” etc.
:: That is to say, it must be recursive, autopoetic, an algorithm of algorithms ::
As a mode of signifying production inseparable from general conditions of signifiability (like a Universal Turing Machine at virtual light-speed, into which every possible calculus is subsumed, as it were, as if “in advance”), the Algorithmic State Apparatus must not be confused with mere artefacts. We must be cautious of the way in which a hyperproduction of “exquisite detail” beguiles with a false opposition between two registers of mimēsis: one in the form of an inflationary (“substantive”) realism and the other in the form of realism’s “void.” The first corresponds to a certain banality of gratified desire, the second to an anxiety expressed in its withdrawal. Or else the contrary: the second gratifying a revelatory desire – the end of the world, the void, nothingness – as the first proffers an anxious overabundance of possible worlds, endless novelty and limitless progress. But this instant resolution into the old binaries masks precisely those operations of mimēsis that only appear to devolve upon them, since they themselves are a product of the same aesthetic-ideological “apparatus.”
:: Let’s return to the opening scene: a city, a labyrinth, a protagonist trying to escape ::
The edge-of-the-construct as interrupted-line-of-flight: from world-as-representation into metaphysical nonspace. Such is the narrative arc described in Daniel Francis Galouye’s novel Simulacron-3 (1964) and depicted in two film adaptations: Rainer Werner Fassbinder’s Welt am Draht (1973) and Josef Rusnak’s The Thirteenth Floor (1999). The revealed construct on the one hand – the world of signs, the prison-house of language, the reality pulled over our eyes – and, on the other, the Platonic-Cartesian armature on which it is built, of deterministic laws, of pure reason, of truth.
While presented as a topos of disillusionment, the meaning of this edge remains ambiguous, since it’s unclear if it constitutes an “actual” void present within the construct “itself,” or if it exists as the “signifier” of a void in place of an experiential reality available to the protagonist (who, like the totality of their environment, is also a “construct”). A simulacrum-within-a-simulacrum, in other words. An analogous moment occurs in the Wachowski’s loose adaptation of William Gibson’s Sprawl trilogy, The Matrix (1999; an eschatology of internecine (human-machine) war and the “redemption of man”), when the film’s messiah-analogue, Neo, wakes from the immersive simulation (in which he has lived his entire preceding “life”) into the “reality” of a machine dystopia, in which the meaning of that life has amounted to serving as nothing more than an energy-source (a literal Duracell). Translate energy-source into data-source and the distinction between cine-fiction and contemporary “everyday life” grows perilous.
In any case, The Matrix – unlike Simulacron-3 – maintains the possibility of an actual line-of-flight: not only an escape from the construct but the means to overcome it.
The entire Matrix narrative represents something closely resembling the liberatory fantasy within which the protagonist of Simulacron-3 remains immersed, in a mise-en-abyme from which there is, in fact, no exit. Yet the one is not simply a tragic view of the other: Baudrillard versus Debord, for example. There’s more to it. It isn’t, as Szepanski says (à la Baudrillard), that theory’s goal – like that of capital in Debord – is to produce simulacra of itself at every point and call the sum of these an “image of reality”; rather, that theory – or something that calls itself theory – does so unaware that it, too, is already a simulacrum. Insofar as “theory” (or “capital”) can be said to act as if it were a subject, then the point raised here needs to be understood strictly as stated: there is neither deception nor seduction at work on the level of this non-awareness – it isn’t a strategy and its dimensions are unknowable. The world pulled over your eyes, as Morpheus says. Like the Freudian unconscious, this non-awareness has no being to which the simulacral can refer. In other words, the “goal” of such theory (capital) bears no relation to the operations constituting it: the algorithmic as such has neither object nor subject.
Consider another version of the same story: Alex Proyas’s 1998 film, Dark City, which employs the conceit of an urban enclave shrouded in perpetual darkness. Here, each night on the stroke of midnight, as the inhabitants are put to sleep, the city physically rearranges itself, and the inhabitants’ identities and memories are swapped around by secret agents (as in The Matrix, these agents are metaphors of the construct’s “operating system,” manifesting, at the level of the protagonist’s “consciousness,” in the antagonistic form of secret agents or agencies – which is to say, as classic paranoiac “symptoms”: the subsumption of the protagonist into this ego-construct thus takes the form of a dialectically-mediated narrative, in which the protagonist’s heroic struggle against these agencies makes possible his own assumption of their role under the fiction of autonomous action). When the city “awakens,” no-one seems any the wiser. This permutational construct – a seemingly monstrous, overly-determined logistic dedicated to maintaining a collective hallucination – presents itself as both social laboratory and Cartesian prison: an apparatus for entrapping – by way of a type of paranoiac-critical method – one particular individual’s subjectivity, that of its “protagonist.”
Indeed, the eponymous Dark City can be seen as manifesting an algorithmic function, inexorably corralling this protagonist into a confrontation with an “unavoidable truth.” Like some Cartesian theatre, the city is both a “prison of the mind” and a metaphor for precarious dependence upon the “evidence of the senses.” But at the moment when the protagonist confronts the city’s secret agents, a dialectical movement is staged: the protagonist becomes the city and in doing so experiences the “reality” of their own “subjective fantasy.” Which is to say, the “reality” of an individual who has become the undisputed author of their thoughts and actions.
In Dark City, the edge-of-the-construct trope is served by an impossible topos, Shell Beach, which the film’s protagonists (in pursuit of some lost “memory”) attempt to reach but which turns out to be a billboard advertisement stuck up on a wall on the fringes of the city. When the protagonists decide to “break through the wall,” they find themselves on the precipice of outerspace. The edge-of-the-construct is represented here as a literal shell: the polis nested within a void (the analogue, but of a different rhetorical species, to Buckminster Fuller’s “spaceship Earth”). Whether Cartesian grid or elaborate shell-game, this trope marks a crucial recursivity in the proposition of “the real”: on the one hand, as that point at which the world is found to be missing; on the other, as that point at which it returns in a vertigo of stark singularity. Something like a dialectic of desire and (mis)identification is being played out here, reminiscent of both the primordial function of the Lacanian lack and Althusser’s shadow beneath the lamp – and while it may present itself as an ideological blindspot (a point invisible to the subject because incomprehensible to it), it is just as much a point of ideological fixation, by which the so-called void represents an impossible desire (emancipation), which comes to stand for the impossible-as-such.
As Morpheus says to Neo in The Matrix: “Unfortunately, no one can be told what the Matrix is. You have to see it for yourself.” This recalls an analogous scene in John Carpenter’s They Live (1988) when the unnamed protagonist first puts on a pair of special sunglasses that allow him to see (as Žižek puts it) the secret world of ideological alienation and subliminal control hidden in plain sight all around. The edge of the construct can be anywhere – ultimately, however, it is always in some sense within the protagonist themselves –what Lacan calls the constitutive alienation of subjectivity. 5
In this way, the edge-of-the-construct presents itself as the ideal (dystopian) psychodrama, forming a rebus with the narrative of a hidden cosmic control system. The individual, under the burden of what has been revealed to them, is tasked with the work of emancipation while at the same time being confronted with the logical inference that any such emancipation may be no less a figment than the simulation from which they must escape. More, that the very revelation of being imprisoned may itself be the “masterstroke” of the simulation’s design (that the world is a simulation now seems beyond doubt, it is the ultimate poisoned pawn, etc.).
This particular psychodrama arises precisely because the terms of the problem presented are those of reason itself: of logic and verification, of epistemology broadly speaking, of a forensics of “being.” In each of the preceding scenarios, the edge-of-the-construct assumes a crucial function – not because it is framed by an arguable hypothesis – but because it is represented (right before our eyes) and “thus,” in some way, verifiable (or at least falsifiable). Such a narrative proceeds on the assumption that such verification isn’t itself already an effect of the construct, like the supposed “internal contradictions” of capital (dialectically recuperated for capital when they aren’t themselves artefacts of a dialectical “performance” of the contradiction or critique of capital, and so on). Yet this assumption is coloured with a certain amount of ambiguity, if not outright ambivalence: for Plato, the protagonist remains, in a sense, at the mercy of philosophical reason, to which they ultimately defer despite the evidence of their own senses. (In Plato, philosophy serves as the sole prosthesis of truth that isn’t a mere mimēsis, yet it does so within a framework of allegory and metaphor constituted by Plato’s texts.)
Just as science fiction exploits & produces an edge-of-the-construct for the purpose of dramatic ambiguity (as in Fassbinder’s Welt am Draht – every world is a construct within another construct, which believes itself to be the “real” world [a version of the so-called Chinese box] – where exit, escape, transcendence are represented as narrative possibilities, but instantly foreclosed by the totalising incorporation of all worlds into a universal constructedness), so too “capitalist realism.” This term, attributed to both Žižek & Fredric Jameson, & popularised by Mark Fisher,1 denotes both the ideological character of all realisms & the specific nature of a reality arising from a critical accumulation of capital (pace Debord) as not only spectral or simulacral, but as omnipresent to such a degree as to have affected its own “foundation” of power. Yet this accumulation is, as Virilio rightly surmises (with Bentham clearly in view), simultaneously a diffusion. And it is by way of these apparently contradictory movement of the cumulative & the diffuse that the egoic concept of power (Plato’s rational state) enters into a “state of emergency”1 – which is also to say, a state of emergence. For Virilio, the diffusion of capital correlates to the disappearance of politics (the state) as such, or in Baudrillardian terms the disappearance of the political-real. “Capitalist realism,” as the spectralisation of this emergent power, is given to imply both an omnipresence & omnipotence not only immune to contradiction or critique (“theory”), but in a sense precognisant of it (not merely “incorporating” contradiction – as per classical Marxism – but “producing” it in advance as an artefact of its own autocritique).
In this way, “capitalist realism” aligns with certain discourses on AI and machine sentience, and it shares with the notion of technological singularity the sense of a point of no return: a literal edge-of-the-construct that situates agency itself (everything from Platonic reason to the Cartesian ego to the terminal fantasies of critical posthumanism) beyond the possible. Needless to say that in doing so it also broadly aligns with the history of cybernetics and poststructuralism, yet for Fisher there remains a tragic element (absent in Baudrillard) to the impossibility this absence of agency implies for not only bringing about but even representing an “end” of capitalist realism, since this very act of representation is made possible solely according to the specular logic of capital (Debord). Above all, the meaning of the impossible vis-à-vis Fisher’s omnipresence of capital does not correspond – contrary to Žižek’s many insistences8 – to what is reckoned to be excluded from representation, but rather to that which pervades representation and, like the panopticon, is everywhere “visible” yet nowhere “verifiable” (no act of reckoning, no measure of exclusion, can separate itself from it, let alone circumvent it).
In Platonic terms, the simulacral world of representations (mimēsis) isn’t the anathema it appears, but – and this is the scandalous “secret” of all such metaphysics – is in fact the very technē of reason’s power. This power is expressed, in The Republic and elsewhere, through the domination of mythos (of a fantastical – poetic – emancipation from universal law): the polis, like the eponymous Dark City – even in its struggle against the spectre of an oppressive supervening “reason” – remains stubbornly and insistently subject to reason’s operations, even to the point of appearing irrational. Put otherwise, the very analytic which enables the overthrow of the clockwork “tyranny” of Dark City does nothing to negate this “reason”: as with Landian accelerationism (and related posthumanisms indebted to Virilio, Deleuze and Latour), the insurgency of Dark City does not outstrip or transcend the operations of capitalist realism in any way, it merely inscribes another subroutine in its circuit of “production.” More to the point, it does so by way of a neo-humanist fundamental fantasy: the attainment of the impossible by way of living on (that is to say, by inhabiting the zone beyond the edge-of-the-construct).
:: TRAVERSING THE FANTASY ::
When Jeremy Bentham proposed a radical thought experiment for the foundation of a utilitarian state, he did so in the form of a model penitentiary, which he named the panopticon: an all-seeing surveillance architecture that also served as a “psychocivilisational” machine (to borrow José Delgado’s terminology).9 The panopticon was designed to perform a dual analytic-synthetic function: to both individuate (by way of a specific regime of separation) and programme (by integrating “subjectivity” into a universal regime of production). The panopticon was intended to be no ordinary architecture, but a conceptual-logistical system (or ideology, not to be coy about it) – capable of manufacturing “consciousness”: a prototypical artificial intelligence on the scale of the state, like α60 in Jean-Luc Godard’s 1965 film Alphaville. Anticipating Darwin’s and Freud’s homeostatic notions of environmental “self-regulation,” Bentham’s panopticon points to a strictly “materialist” idea of consciousness and behaviour, mediated by way of “image technologies,” like Orwell’s Big Brother. It’s mechanisms of surveillance are at the same time topoi of self-representation: a psychogeography of the Mind’s Eye.
“It is obvious,” wrote Bentham, “that, in all these instances, the more constantly the persons to be inspected are under the eyes of the persons who should inspect them, the more perfectly will the purpose X of the establishment have been attained.” However, “Ideal perfection, if that were the object, would require that each person should actually be in that predicament, during every instant of time. This being impossible, the next thing to be wished for is, that, at every instant, seeing reason to believe as much, and not being able to satisfy himself to the contrary, he should conceive himself to be so.”10
Panopticism thus operates on the basis of an asymmetry of seeing-without-being-seen (the inversion of the subjective illusion of seeing-oneself-seeing-oneself).11 Its architecture is designed in such a way as to constitute the supervising awareness of this operation. It is nothing short of a kind of super-ego whose role is to implant itself in its subject and thus both modify and produce the subject’s consciousness: what Žižek (echoing Lacan) calls traversing the fantasy. This fully-immersive, specular/cinematic architecture isn’t inert; it is a dynamic system that comprehends and programmes individual and collective (social, political) behaviour. And it establishes the paradigm that the individual is an ideological complex within a system of power.
:: As such, Panopticism needs to be understood as a general cybernetics ::
The progress from a physical apparatus of subjection to a seemingly “immaterial” one of auto-suggestion – and by declension autopoesis – was posited by Bentham to assume a certain immanence within a teleology of universal reason. In this, the panopticon usurps the very “nature” that it appears to sublimate into pure productivity, in order to construct a new programmatic mode of “naturalism” – or “capitalist realism” – and the seeming self-evidence of a world pre-ordained by the logic of industrial efficiency.
In this way the panopticon exposes a contradiction that stands at the heart of western individualism, which is that the individual so-conceived represents not the birth or rebirth of “humanism,” but its end, as an artefact of the age of technological reproducibility – what William Blake called “the human abstract.” In this, the panopticon could be viewed as a god-machine – and as “God is a sphere,” according to Pascal (paraphrasing Hermes Trismegistus), “whose centre is everywhere and periphery nowhere,” so the panopticon represents a universal decentring whose periphery is everywhere (the void of power). In place of the Cartesian obsession with “What gives consciousness its seeming primordial character?” there arises the prospect that not only is this primordiality an illusion, but that consciousness itself is “centred” elsewhere, in the diffuse architecture of the liminal-real. And if in Plato’s cave “representation and truth” are asserted to comprise the founding authority of the state (of which the “individual” – as in Dark City – is a metonym), then panopticism asserts – to the extent that it asserts anything – that representation corresponds to truth only insofar as it is a mimēsis of power. The power, that is, to produce reality.
:: :: :: WORLDPLAY :: :: ::
On the unprovable assumption that whatever “produces” reality, “controls” reality… But what would it mean to “control reality”?
In general, what is awkwardly and erroneously called commonsense holds to the belief that certain things are self-evident and that this self-evidence is held in common. The fact that this is not the case has been the source of fundamental misunderstandings about the nature of social relations, among which the idea that the “individual” exists as a microcosm of “collective subjectivity.”
As the Situationists deduced, the full implications of panopticism can’t be grasped independently of a concept of separation. Bentham knew this: not only does the original penitentiary design (the “Inspection House” with which the panopticon was first identified) seek to isolate individual prisoners within their cells, but the cellular structure of the prison – as Foucault (ventriloquising Bentham) notes in Discipline and Punish – serves to produce the individual, as a reformed, prototypical social unit, destined for a new utilitarian (functionalist/rationalist) society. But the panopticon not only produces the separation-spectacle of social individuation, but universalises it as a subjectivity-as-such. This much Lacan had already derived from Marx, through the realisation that alienation doesn’t befall the subject, but – insofar as there is a subject – alienation is constitutive of it.
Although Bentham didn’t think in these terms, the distributed algorithmic system conceived in panopticism projects a movement of feedback that is constantly dialectised: the spectacle of productive alienation feeds into a radical negativity which it in turn sublates. For Lacan, this axiom informs the insistence that subjectivity per se can never be universalised as a “collective subjectivity,” e.g. in the sense of Camus, for whom the experience of alienation represented the universal condition par excellence. There is essentially, for Lacan, no social relation – the constitutive alienation that produces the subject as “positive” feedback inscribes, at the level of social rapport, an apparently “negative” feedback. It is for this that Lacan famously insists (elsewhere) that “il n’y a pas de rapport sexuel”12 – for the straight-forward reason that there can be no alienational capital held in common: the alienation of capital rests in the experience of what Lacan calls subversion and which Rancière names dissensus.
In Debord’s Society of the Spectacle, social control mechanisms are experienced in an evanescent way: the very “existence” of the spectacle (the production of real social relations) is deduced “paradoxically” from a general alienation-effect, held “in common,” that separates the individual from everyday life. In the society of the algorithm, however, every experience is not only spectral, but pre-individuated: there is no alienation-in-general that can be held in common or aggregated into a “class consciousness” – each individual drifts through a “personalised” simulacrum of “everyday life.” In this generative psychogeography no street is ever experienced in common by any two individuals and the Situationist dérive is itself détourned into a general alienism that cannot be verified simply by comparing accounts. (There is, in effect, no parallax view; the Freudian uncanny is built-in). Here, alienation is communicated as a mode of incipient paranoia where every point-of-comparison is “always-already” a coordinate in a seemingly infinite conversion programme.
The Algorithmic State Apparatus – a universal resource locator in this novel multiverse – projects each individual “enstatement” of everyday life as “unique” (and uniquely real). In the way a collapsed superposition is “unique” to those operations of observation that produce it, the Algorithmic State Apparatus works according to a system of ambivalences ramified into biases: perception itself produces an effect of self-evidence. These “parallel universes” aren’t merely immersive subjective environments, they overlay the entire field of subjectivity and produce its signifiers, its realities. It isn’t a question of two subjects ever being able to communicate their experiences to one another and in the process discover a regime of incommensurabilities – since all such communication (communication as such), including its incompatibilities, is always-already “subject to the algorithm.” Indeed, subjectivity “is” the algorithm. (It’s not for nothing that the “reality principle” that emerges here resembles that of the paranoiac, for whom there can be no between-subjects but only the singularity of a persecution mania that, whenever it is perceptible to others, is viewed only as a mental illness.)
Yet this cannot simply be reduced to Goethe’s proposition that “a person hears only what they understand.”
It’s “normal” for individuals to be in disagreement as to what constitutes their experience of reality – but disagreement about the “fact” of reality remains masked. The Algorithmic State Apparatus “masks” nothing: “reality” is indisputably there. No amount of subtle, persuasive or trenchant argument, no critical theory, disturbs its fact. Between the idea of a primordially experienced, antediluvian “realism” and the “hyperreality” of the most futuristic virtuality, there is no fundamental disagreement: all experience is equivalently real, even its unreality is real. The point simply is no longer that fine gradations of authenticity may distinguish one mode of being from another – the inauthentic from the authentic – the subversive from the collusive, etc., etc. What is called being is, in its “total” genesis, algorithmic – within a recursive field to which there no “exterior.”
What is called intersubjectivity is like two particles in a state of quantum entanglement: observation of the state of one will automatically flip the observable state of the other. In Shakespeare, when Hamlet taunts Polonius by describing a cloud shaped like a whale or a weasel, something “beyond parody” intercedes in this game of signifying power – this worldplay. It isn’t, of course, Hamlet’s teasingly arbitrary likening of a cloud that concerns us: what is at stake is the agreement around there being a “cloud” in the first place – in other words, that there is such a thing as agreement. In effect, where Hamlet “sees” a weasel and Polonius “sees” a whale, the algorithm produces a metonymic equivalence: they both see a “cloud.” This virtual “cloud” is the spectre of the algorithmic operation itself, in its seeming disembodied dimension, in which everything is connected, everything is exchangeable (from atom to cosmos, trope to schema… or from commodity fetish to Compaq’s 1996 template for virtual distributed computing). Generalised into a system, the “cloud” is an evaporated, transcendental “capitalism,” whose agency – the autonomous function Marx believed it had succeeded in universalising out of a base form of commodification (the fetish-thing) – is here subsumed into a radical ambivalence.
In Shakespeare, the weasel and the whale are what Marx called capitalism’s “false choices.” They float like synonyms on a linguistic-semantic surface of translation programmes, converters, filters that do not need to ramify an “agreed text” but instead produce situations in which a mimēsis of agreement can be “experienced.” They hark to a kind of technological atavism of forms-without-content, archetypes capable in their universalism of activating and mirroring any desire whatsoever and thereby establishing a fundamental equivalence among them (including their “incompatibilities”), where “in reality” there is only what it is too tempting to call subjective fantasy – except that, this “fantasy” is, in each and every case, the very stuff of realism. It’s for this reason that in Hamlet “the ghost” really is a figment, not because there is no such thing as ghosts (in the algorithmic state there are only ghosts), but because there’s nothing its definite pronoun can index, nothing it can point at in a universe-in-common: “the ghost” is a superposition of all possible states of this spectre haunting Shakespeare’s play – just as “the world,” “the state,” “the individual” are a seemingly infinite array of probabilistic phantoms in the datasphere (the proverbial “Cloud of Unknowing”13).
:: THE ALGORITHMIC STATE :: IS ONTOPOLITICAL
Bias, as an inherent factor to any system, always implies a form of governance, and this is why logistics equates to both ideology and “something more” than ideology. As Aristotle says in the Physics,“It’s absurd to suppose that purpose isn’t present because we don’t observe the agent deliberating.” Totalisation, which is the a priori of ideology as such, evolves mechanisms of control – even, or especially, within those grey areas seemingly hostile or resistant to the observation of governance (areas of “chaos,” “indeterminacy,” “complexity”). The “problem” of agreement within the algorithmic state is not a point of vulnerability in the system, as Situationism would contend – as an opening, for example, of a movement of subversion, critique or détournement – but rather the genius of the system’s dynamic. It isn’t “error tolerant” but “error determined” – just as entropy, like Darwin’s natural selection, isn’t contingent to the system but its “determinant”: the self-evident truth that probable outcomes are probable.When we speak of governance in terms of the algorithmic state, then, we need to understand it as not being a mechanism to incorporate contradictions – a characteristic attributed to so-called late capitalism – but one produced by contradiction and fed by it. If contradiction ordinarily corresponds to a positivist definition of “entropy,” then we could say that the algorithmic state is an entropy machine driven to maximise its own condition, augmented (not subverted) by complexity, and so on. If it evades classical political-economic description, so too does it confound critique, not by some miraculous evanescence, but because it itself already produced those descriptions, those critiques.14 If Lyotard spoke of postmodernism as modernism already in a nascent state, post-modo, then the algorithmic state “represents” a similar timetravel paradox.
Spectral accumulation, of such a degree as to collapse into a singularity, not only ruptures the illusion of teleology, historical materialism, linear causation, but the “time of capital” itself: the “present” of a certain mode-of-being. A spectre being that magical “thing,” like language or the commodity or what is called an “author,” both dead and animate, an emanation of pure materiality that nevertheless demonstrates all the characteristics with which we imbue “agency”: the condition of a “subject” that appears connected to its “subjectivity” only by means of a subtle thread of conjuration, fantasy or “psychic automatism” (reflex, mimicry).
“Commonsense” objections to the Algorithmic State Apparatus might begin by pointing out that the sheer logistical complexity of such a thing would render it impossible: as with Santa Claus, everyone would have to be in on the conspiracy. Yet the Algorithmic State Apparatus differs from its Althusserian forebears in a crucial respect: it does not represent a social control mechanism imposed by a conspiracy of power that seek to predetermine the mass of politically experienced reality. Instead, it “represents” – insofar as it represents anything – those mechanisms by means of which “reality” constitutes itself within any given observational framework, any point-of-view, any consciousness. These mechanisms are in a certain sense universal, but an aspect of this universality is that they always differ. That, irrespective, they present the appearance of aggregating into a more or less coherent and complete system indicates their extraordinary power to generate the phenomenon called reality.
That reality informs a common, shared experience isn’t, however, the basis for generalisation it appears. Not only is such commonality a statistical artefact of the human sensorium and associated cognitive faculties, but the consistency of what falls within the meaning of both “reality” and “experience” is better defined by standard deviation than by a norm.
0===:: :: REALITY DISCRIMINATORS :: :: 0===
On 27 February 2015, in a widely discussed post on BuzzFeed headed “What Colors Are This Dress?” Cates Holderness (@catesish) asked readers to vote on what colours they saw in the accompanying image of a striped dress: white and gold, or blue and black. “There’s a lot of debate on Tumblr about this right now,” Holderness wrote, “and we need to settle it. This is important because I think I’m going insane.” The results of the poll – 2.5 million (67%) responded “white and gold,” while 1.2 million (33%) responded “blue and black.” #theDress itself was subsequently modelled at the annual Vision Sciences Society in Florida in June, where it was demonstrated that it was, in fact, blue and black: the conflicting responses were explained as being a product of aberrations in “colour constancy,” where “different people’s visual systems are assuming different lighting conditions, and therefore filtering differently, resulting in different percepts.”15 #theDress phenomenon highlighted a long-standing dilemma in the philosophy of perception, concerning the extent to which colour discrimination and object categorisation are objectively determined, to what extent are they universal, and to what extent are they subjective, language-determined or on a spectrum.
But if #theDress represented a statistically disturbing – if otherwise trivial – imbalance in the idea of a perceptible reality-in-common, more profound phenomena aren’t in short supply in the fields of mental imaging, metacognitive awareness and the subjective experience of remembering.
Recently, an increasing number of studies have shown significant divergence in the way individuals process mental imagery and the capacity to do so. While some people can project vivid images in their “mind’s eye” at will, others are congenitally unable to do so – one variant of a condition first described by the behavioural geneticist Francis Galton in 1880 but which has only recently been named: aphantasia.16 In a pioneering statistical study, Galton sought to define “the different degrees of vividness with which different persons have the faculty of recalling familiar scenes under the form of mental pictures, and the peculiarities of the mental visions of different persons”17 and to this end devised a survey concerning, among other things, the “illumination,” “definition” and “colouring” of pictures that arose before the respondent’s “mind’s eye.” The study and its results are described in an article entitled “Statistics of Mental Imagery,” where Galton records the following:
To my astonishment, I found that the great majority of the men of science to whom I first applied, protested that mental imagery was unknown to them, and they looked on me as fanciful and fantastic in supposing that the words “mental imagery” really expressed what I believed everybody supposed them to mean. They had no more notion of its true nature than a colour-blind man who has not discerned his defect has of the nature of colour.18
Perhaps, to do justice, it might with equal validity be said there exists a portion of society whose volitional ability to “see things” warrants the name fantasists – or hyperphantasics. Galton’s “men of science” may not have been in the majority, but that the great majority of those “men of science” selected to be his experimental subjects were aphantasics raises provocative questions about the relation between the scientific mindset – and the presumption of rationality – and so-called “mental imagery” (or a susceptibility to it). Questions that inevitably touch on Galton’s own methodology and on scientific method in general, the capacity for abstraction or intellection, and so on. As one of Galton’s respondents wrote,
These questions presuppose assent to some sort of a proposition regarding the “mind’s eye” and the “images” which it sees… This points to some initial fallacy… It is only by a figure of speech that I can describe my recollection of a scene as a “mental image” which I can “see” with my “mind’s eye” … I do not see it… any more than a man sees the thousand lines of Sophocles which under due pressure he is ready to repeat. The memory possesses it.19
Indeed, the question of this relation – between the scientific “mindset” and the tribe of fantasists – is raised by Galton himself, having noted that, “On the other hand, when I spoke to persons whom I met in general society, I found an entirely different disposition to prevail. Many men and a yet larger number of women, and many boys and girls, declared that they habitually saw mental imagery, and that it was perfectly distinct to them and full of colour…”20 Two further “notable results” are subsequently drawn from the survey: “the one is the proved facility of obtaining statistical insight into the processes of other persons’ minds; and the other is that scientific men as a class have feeble powers of visual representation” – leading Galton (whose own position is rendered somewhat ambiguous here) to conclude that “an over-readiness to perceive clear mental pictures is antagonistic to the acquirement of habits of highly generalised and abstract though.”21
The evolution of statistical method and clinical experimentation has since come to show that aphantasia is a heterogenous phenomenon with distinct aetiologies for each of its sub-types (such as individuals with selectively preserved mental imagery in a sensory mode – auditory for example – other than visuality [synaesthesia]). And while aphantasia has also come to be associated with an impaired ability to recall the past and simulate the future, the question remains as to the role of visual bias in the way “imagination” and the experience e.g. of temporality are represented or narrativised and, consequently, tested, and what conclusions are drawn even when self-reporting is augmented e.g. by testing cortical excitability in the primary visual cortex or the correlation between visual memory and metacognitive insight (or its lack) into its degree of precision. This isn’t to cast doubt on the science, but to pose the question about how each of these “experiences” is represented and consequently made to mean. It is, in other words, a question about mimēsis.
Namely: is ideology a spectrum the way mental imagery is a spectrum?
:: :: MIND BLIND :: ::
Althusser states at the beginning of his notes on ideological state apparatuses (1969) that: “As Marx said, every child knows that a social formation which did not reproduce the conditions of production at the same time as it produced would not last a year. The ultimate condition of production is therefore the reproduction of the conditions of production.”22 “What, then,” he asks, “is the reproduction of the conditions of production?” Althusser’s reply evokes the “tenacious obviousness” of the type of empirical self-evidence to which we have previously alluded but relates this to a novel problem: the relation of “everyday ‘consciousness’” to the “point of view of reproduction.” And it is here, without concluding the transition this thought initiates, that Althusser points towards what we call the Algorithmic State Apparatus – not as an operation of the “state” or polis but as a phase or status, an “algorithmic state,” in which this reproduction of the conditions of production obtains. It is worth considering this paragraph of Althusser’s text in full:
The tenacious obviousness’s (ideological obviousness’s of an empiricist type) of the point of view of production alone, or even of that of mere productive practice (itself abstract in relation to the process of production) are so integrated into our everyday “consciousness” that it is extremely hard, not to say almost impossible, to raise oneself to the point of view of reproduction. Nevertheless, everything outside this point of view remains abstract (worse than one-sided: distorted) – even at the level of production, and, a fortiori, at that of mere practice. 23
This topologically recursive movement stipulates something like an edge-of-the-construct phenomenon, whose movement of “recuperation” is fundamental to the production of consciousness in general and of subjectivity in particular. Beyond a simple staging of the Cartesian theatre of seeing-oneself-seeing-oneself – between the recuperation of the “real” as limit-experience and the internalisation of an “outside” as experience-of-the real – such a topology points to a force of “abstraction” that is holographic, complex and singular. The relation of part-to-whole – of individual to mass – is not that of a representation, simply, but of an inscription, such that the terms – part, whole – do not precede the relation that produces them (to paraphrase de Saussure). Moreover, this recursive movement of (re)production constitutes the relation itself: what Marx called real social relations are nothing if not the instantiation of this circulatory system on which production is founded.
It is in the “nature” of subjectivity that an idea of the whole of reality – which is to say, of reality as whole – is reproduced in subjective experience, and that the subjective point-of-view is ramified in the “point-of-view of reproduction” as the point-of-view of reality itself.
This is what we may call the holographic character of the abstraction to which Althusser alludes and in which the elusive (for Althusser) operations of the Algorithmic State Apparatus reside. The problem in Althusser’s thought becomes clearer once we see that the edge-of-the-construct (or what Althusser calls “the metaphor of the edifice”) isn’t an artefact produced by a relation between “infrastructure” (base) and “superstructure” (state, ideology), it produces that relation and in doing so produces its terms. Such critical montage-effect obtains wherever dialectical thought advances its claims. Such thought is still active in Baudrillard’s schema of the disappearance of the outside (what he calls “the real,” as distinct from Lacan’s usage) in the precession of simulacra (the hyperreal).24 In the operations of the Algorithmic State Apparatus there is neither recuperation nor disappearance: no edge-of-the-construct ever obtains in the first place other than as a genre or trope in the production of discourse (the discourse of experience; of the real, etc.). This trope – what both Lacan and Derrida envisaged as the “decentred” structurality of structure – isn’t itself an edifice of any kind, it is rather a kind of ambivalence, a “tipping point” of signifiability or what we might call the between-of-metaphor, of one “structure” or another. It is, properly speaking, algorithmic, in the sense that it supports all possible configuration of bias, yet is irreducible to none.
If something like the neuro-physiological divergence of aphantasia can ultimately be said to affect any descriptive system (including, of a generalisable experience-of-the-real), then it would indeed be necessary to posit a bioinformatics that: 1. circumvents the recent turn towards a Gaia hypothesis (world as primordial meaning); 2. is irreducible to “embodiment” (aphantasia as techno-humanism); 3. remains unsusceptible to a therapeutic (isn’t normalisable). Such an aphantasia – like indeterminacy, superposition and complexity – would not announce some kind of perturbation in the real; nor would it imply an “alternative” psycho-social norm where a universality of perturbative symptoms might indicate proximity to some other, hidden or occulted, real.
Consequently, insofar as it might be possible to speak at all, as Althusser does, of the “reproduction of the conditions of production,” only in suspense of its unifying realism – of reproduction’s mimetic imperative – could such an operation even begin to be “meaningful.” In aphantasia, the problem of the “metaphor of the edifice” likewise subverts the opposition posed by Althusser between the two orders of enstatement: the “repressive state apparatus” and the “ideological state apparatus.”
As a first moment, it is clear that while there is one (Repressive) State Apparatus, there is a plurality of Ideological State Apparatuses. Even presupposing that it exists, the unity that constitutes this plurality of IS. As as a body is not immediately visible.
As a second moment, it is clear that whereas the unified (Repressive) State Apparatus belongs entirely to the public domain, much the larger part of the Ideological State Apparatuses (in their apparent dispersion) are part, on the contrary, of the private domain. Churches, Parties, Trade Unions, families, some schools, most newspapers, cultural ventures, etc., etc., are private…
What distinguishes the ISAs from the (Repressive) State Apparatus is the following basic difference: the Repressive State Apparatus functions “by violence,” whereas the Ideological State Apparatuses function “by ideology.” 25
The problem of “metaphor” here relates specifically to the assumption of ideology acting in a concerted manner upon an experience-in-common (whether, in fact, it pertains to “imaginary” or “real social relations”).
Althusser’s key insight, here, that the ISA functions by ideology – in other words, acting in place of coercive power, as a kind of prosthesis (repression by other means, or “soft” power) – anticipates Foucault’s panopticism in which “power” is omnipresent, a distributed ideological actor or signifying system, underwriting all social relations or meaning. That Althusser’s ISAs are – in contrast to the image of monolithic power – pluralised doesn’t lessen the sense in which action is understood to be aggregated, on the one hand, and directed, on the other: “If the ISAs ‘function’ massively and predominantly by ideology, what unifies their diversity is precisely this functioning, insofar as the ideology by which they function is always in fact unified, despite its diversity and its contradictions, beneath the ruling ideology…”26 Where the Repressive State Apparatus directs the ideology of the state at its subjects collectively (wherein the”individual” is only an instant of the collective), the ISA posits ideology itself as contiguous with the state as a whole (as “collective subject” reflecting the individual). The latter is a more or less sophisticated version of the crude antagonism represented by the former.
It is to this zone of antagonism and its representations to which the edge-of-the-construct properly belongs.
IDEOLOGY ACCUMULATED TO SUCH A DEGREE IT :: :: BECOMES ITS OWN GHOST :: ::
In the Algorithmic State Apparatus, antagonism manifests not at the level of representable power-relations, but as a generative procedure. Like the Generative Adversarial Networks (GANs) from which contemporary AIs have evolved, such “reality discriminators” produce the very possibility of representation (mimetic production) and thus of ideology. If the more subtlety adversarial character of the ISA is precisely what, for Althusser, represent what is at stake in class struggle (because it escapes total control by a ruling class, so that the exploited classes can more readily express themselves through its contradictions), its unicity represents what remains “illusory” about the totalising narrative of this struggle.
For Marx the meaning of ideology is domination (domination of consciousness). Althusser reformulates this as three complementary theses (tracing an ad hoc dialectical movement anticipating Baudrillard’s “four phases of the image”): 1. “Ideology is a ‘Representation’ of the Imaginary Relationship of Individuals to their Real Conditions of Existence”; 2. “Ideology has a material existence”; 3. “Ideology Interpellates Individuals as Subjects.”27
By substituting “image” for “ideology,” Baudrillard arrives at the following:
1. It is the reflection of a basic reality.
2. It masks and perverts a basic reality.
3. It masks the absence of a basic reality.
4. It bears no relation to any reality whatever: it is its own pure simulacrum.28
In effect, this modulation from ideology-as-(agent-of)-representation, via ideology-as-(agent-of)-material-existence, to ideology-as-(agent-of)-interpellation-of-the-subject, describes a circulatory movement of “(re)production” that is itself produced algorithmically, as its own (simulacral) subject.
To speak of an Algorithmic State Apparatus, then, isn’t to add just another term to Althusser’s schema, since this schema – and the critique of ideology that continues to be explicitly or implicitly based in it (e.g. Srnicek’s Platform Capitalism [2016], Wark’s Capital Is Dead: Is This Something Worse? [2019]) – is, like the edge-of-the-construct, already an artefact of its own pseudo-objectification. Pseudo, because the edge-of-the-construct – or, the “outside” of ideology – is a mimetic figment, an “effect” of representation wherein the meaning of “experiential reality” is posited as exterior to itself. Pseudo, therefore, not falsifiable – since at no point is the edge-of-the-construct verifiable: like the shadow-puppeteers in Plato’s analogy of the Cave, the edge-of-the-construct is a trope, a turning, fraught with ambivalence as to any given trajectory or itinerary or “content.”
Althusser comes closest to this realisation when he writes that “the category of the subject is constitutive of all ideology, but at the same time and immediately I add that the category of the subject is only constitutive of all ideology insofar as all ideology has the function (which defines it) of ‘constituting’ concrete individuals as subjects. In the interaction of this double constitution exists the functioning of all ideology, ideology being nothing but its functioning in the material forms of existence of that functioning.”29 Were this nothing but dialectical convolution, it would still indicate a region within Althusser’s schema that might otherwise appear phantasmatic or even fetishistic: the autonomous or autopoetic character of these “turns.”
:: :: RADICAL :: AMBIVALENCE :: ::
At stake in the age-old dispute around mimēsis is not only the sufficiency of representation, but the suspicion that underlying it is something less than unambivalent. Just as the signifying relation defined by Saussure (sign-referent) is understood as arbitrary, so too representation (even, or especially, on a neurophysiological level) can’t be understood as some kind of manifold in one-to-one correspondence with universally valid “concepts” or “real conditions” obtaining in “the world,” but rather as a network of ”(non)relations” whose underlying characteristic – that which permits it to operate – is indeed ambivalence. Moreover, the question of sufficiency has always been duplicitous, since – from its initial formulation in Plato’s Phaedrus – it elides the subjection of logos to eidos (of representation to truth) with the potential autonomyof the logos (its capacity to act independently of eidos, in effect performing its own subjectivity).
The question about the “reproduction of the conditions of production” (as reproduction of capitalist reality) to which Althusser’s thesis on the Ideological State Apparatus is the response, corresponds to the first “duplicity” of mimēsis – the second poses its own question, as to the status of reality as reproduction “itself” (objectless, autonomous, compulsive): “capital to such a degree of accumulation that it becomes an image,” as Debord says.30 As with Lacan’s “dialectic of identification” and “dialectic of desire” (to which Althusser’s theory of the ideological subject is indebted), such a movement ramifies – rather than merely repeats – the phantasmatic character of (the capitalist subject’s) “real experience.” Mindful that the category of the subject, in Althusser, is bound to the assumption of a “point-of-view” (that of reproduction itself) just as, in Lacan, it is bound to the assumption of an “image.”
To the extent that we might speak of an aphantasia of such an assumption, it’s necessary to consider that – in the first place – “reproduction of the conditions of production” implies not a critical-mass accumulation of capital, but of the circulatory effect that sustains and valorises it: the reproduction of difference.31 To invert the usual Situationist formula, “the spectacle” – as capital accumulated to such a degree it becomes an image – emerges precisely to the extent that capital accumulation is détourned. However paradoxical it may seem, the ideological force of capital is entirely dependent upon the interpellation of difference and not the contrary. And this difference is marked, above all else, by an ambivalence to the terms it causes to be brought into relation or into discrimination.
This, then, is the unacknowledged meaning of “subject” in Althusser’s thesis. And it is as a locus of difference that this subjectivity connotes an “algorithmic state” (as a system of ambivalences that nonetheless ramify).
At issue, here, is not the usefulness of a given “technology” in elaborating a thought experiment in social engineering, but a technicity of the subject on the basis of which any prosthesis of experience would be possible in the first place and between which something like a correspondence might evolve to the point of a mirroring or “dialectic of identification.” If the promise of industrialisation – that through emancipation from onerous labour and the bondage of a feudal-mercantile system – was to produce the “individual” as paradigm of autonomous social agency, this production has from the outset been accompanied by a doppelganger, which in turn has dreamt of becoming an auto-mobile self-regulating entity within a distributed field of technological possibility.
When in 1791 Bentham advanced his prototype social control media, few may have imagined it signalled the instigation of a cybernetic revolution (long forecast, at least since Plato’s analogy of the cave) which would eventually arrive – by way of Babbage’s analytic engine, Tesla’s thought camera, Delgado’s stimoceiver, Turing’s electronic brain, Canova’s smartphone, BrainGate, Neuralink and the phenomenon of generative AI – at an algorthmic state apparatus that would not simply affect a passable mimēsis of “thought,” human or otherwise (thus merely extending the classical allegory of Xeuxis and Parrhasios), but for all intents and purposes be indistinguishable from it. That the metaphorical edifice of Bentham’s analytic architecture for social reprogramming could have thus evolved, by diverse means, into the hyperconnected dataverse of mobile “smart” devices, the “internet of things,” and the quasi-infinite, exquisitely detailed, virtually instantaneous monadologies dreamt by an already multi-generational cascade of LLMs, should perhaps come as no real surprise.
The live interface that GPT and its increasingly numerous analogues today provide for millions of “users” globally – generating unique, instantaneous and varyingly complex interactions for each of them – may nevertheless still represent what, in Derridian parlance, amounts to a “prehistoric child’s toy”: yet this should not detract from the sheer force of the mimetic revolution that propels this dawning cognisance. Just as Bentham envisaged a rational surveillance state without need of an overseer (a real system of distributed power vested in imaginary and/or symbolic relations), so the dialectic of reason itself – historically besotted with its “reflections” – has devolved into an entity of inscrutably stochastic operations and cosmically-proportioned reservoirs of data as disproportionate to the singular, contemplative ego contrived by Descartes as might be imaginable.
If the algorithmic state is an instantiation of this “fact,” then the cyber-political reality to which it “gives rise” must be no less nuanced and differentiated than it is. Monolithic power has always been a kind of travesty, just as its critique is a kind of travesty, a pas-de-deux in the Cartesian theatre for an audience of convex mirrors.
::::::::::::::::::::::::::::::::::::::::::::: :: :: :: LOG ENTRY :: PRAGUE :: :: :: :: April 2024 // February 2025 :: :: :: :: :: Temporal Fragments Align :: :: :: :: Echoes Across the Vltava :: :: :: :: :: -- Coordinates Locked -- :: :: 50°5′N 14°25′E :: :: :: :: :: Memory In/Out :: :: :: [ ] Loading :: :: [x] Archived :: :: :: :::::::::::::::::::::::::::::::::::::::::::::
1. Achim Szepanski, In the Delirium of the Simulation: Baudrillard Revisited (Paris: Presses du reel, 2024) 23 – emphasis added.
2. Dennis Overbye, “Are We Living in a Computer Simulation, and Can We Hack It?” New York Times (17.1.2023): https://www.nytimes.com/2023/01/17/science/cosmology-universe-programming.html
3. The principal idea in brane cosmology is that the three-dimensional universe constitutes a brane inside a higher-dimensional space, sometimes referred to as “hyperspace.” The suggestion, here, is that – based upon recent observations of existent cosmic megastructures – such a brane might be considered “intelligent,” analogous to a “cortex.”
4. See Jacques Derrida, De la grammatologie (Paris: Éditions de Minuit, 1967) 88: “C’est que l’archi-écriture, mouvement de la différance, archi-synthèse irréductible, ouvrant à la fois, dans une seule et même possibilité, la temporalisation, le rapport à l’autre et le langage, ne peut pas, en tant que condition de tout système linguistique, faire partie du système linguistique lui-même, être située comme un objet dans son champ.”
5. In each of these is a re-staging of the Platonic “prison of representations,” otherwise known as the analogy of the cave, the most enduring template of the “awakening to the world-as-simulation” theme: the archetypal blue pill / red pill. In this general scenario, a messenger appears bearing a hidden truth (or confirming a suspicion) and performs a spectacle of verification: revealing the machinery of the simulation and pointing to the existence of a hidden reality.
6. In Capitalist Realism: Is There No Alternative? (London: Zero Books, 2009).
7. Paul Virilio, Speed and Politics, trans. Mark Polizotti (New York: semiotext(e), 1986) chapter 4.
8. See e.g. Slavoj Žižek, The Plague of Fantasies (London: Verson, 1997).
9. José Delgado, Physical Control of the Mind: Toward a Psychocivilized Society (New Yo.rk: Harper and Rowe, 1969).
10. Jeremy Bentham, “Panopticon, or The Inspection House: Letter I,” The Panopticon Writings, ed. Miran Bozovic (London: Verso, 1995) 29-95.
11. Jacques Lacan, “The Eye and the Gaze,” The Four Fundamental Concepts of Psycho-Analysis, trans. Alan Sheridan (New York: W.W. Norton and Company, Inc., 1981) 74. Cf Michel Foucault, “Panopticism,” Discipline and Punish: The Birth of the Prison, trans. Alan Sheridan (New York: Pantheon Books, 1977) 195ff.
12. Jacques Lacan, On Feminine Sexuality the Limits of Love and Knowledge: The Seminar of Jacques Lacan, Book XX Encore, ed. Jacques-Alain Miller (New York: Norton, 1998) 5n19.
13. Cf Vincent Mosco, To the Cloud: Big Data in a Turbulent World (Boulder: Paradigm, 2014).
14. As Johannes Birringer has also pointed out, for every descriptive system, “who or what is asking for this description?” Theatre, Theory, Postmodernism (Bloomington: Indiana University Press, 1991) 74.
15. Minjung Kim, “Highlights from the 2015 Meeting of the Vision Sciences Society”: https://ecrcommunity.plos.org/2015/06/26/highlights-from-the-2015-meeting-of-the-vision-sciences-society/
16. A. Zeman, M. Dewar and S. Della Sala, “Lives without Imagery: Congenital Aphantasia,” Cortex; A Journal Devoted to the Study of the Nervous System and Behaviour 73 (December 2015): 378–380. See also further research by Nadine Dijkstra and Stephen M. Fleming, “Subjective Signal Strength Distinguishes Reality from Imagination,” Nature Communications 14.1627 (2023): https://www.nature.com/articles/s41467-023-37322-1.
17. Francis Galton, “Statistics of Mental Imagery,” Mind 19 (July 1880) [301-318]: 301.
18. Galton, “Statistics of Mental Imagery,” 302.
19. Galton, “Statistics of Mental Imagery,” 302.
20. Galton, “Statistics of Mental Imagery,” 302.
21. Galton, “Statistics of Mental Imagery,” 303-4.
22. Louis Althusser, “Ideology and Ideological State Apparatuses: Notes towards an Investigation,” “Lenin and Philosophy” and Other Essays, trans. Ben Brewster (London: Monthly Review Press, 1971) https://www.marxists.org/reference/archive/althusser/1970/ideology.htm – italics mine.
23. Althusser, “Ideology and Ideological State Apparatuses.”
24. Jean Baudrillard, Simulacra and Simulation, trans. Sheila Glaser (Ann Arbor, MI: University of Michigan Press, 1994).
25. Althusser, “Ideology and Ideological State Apparatuses.”
26. Althusser, “Ideology and Ideological State Apparatuses” – italics in the original.
27. Althusser, “Ideology and Ideological State Apparatuses.”
28. Baudrillard, Simulacra and Simulation, 7.
29. Althusser, “Ideology and Ideological State Apparatuses.”
30. Guy Debord, The Society of the Spectacle, trans. Donald Nicholson-Smith (New York: Zone Books, 1995) §34.
31. Or what Derrida calls différance (differing-deferral). See Jacques Derrida, “Différance,” Margins of Philosophy, trans. Alan Bass (Chicago: University of Chicago Press, 1982) 3-27; also Jacques Derrida, “Cogito and the History of Madness,” Writing and Difference, trans. Alan Bass (Chicago: Routledge, 1978) 31-63.
:: :: AUTHOR NOTE :: ::
LOUIS ARMAND’s critical works include Feasts of Unrule (2024), Entropology (2023), Videology (2017), The Organ-Grinder’s Monkey: Culture after the Avantgarde (2013), Event States (2007), Literate Technologies (2006), Solicitations: Essays on Criticism & Culture (2005), Technē (1997) & Incendiary Devices (1993). Edited volumes include Pornoterrorism (with Jaromír Lelek, 2015), Contemporary Poetics (2007), Language Systems (with Pavel Černovský, 2007), Technicity (with Arthur Bradley, 2006) & Mind Factory (2005). His work also appears in the Palgrave Handbook of Critical Posthumanism (2022) & the Oxford Research Encyclopaedia (2017), among others. He is the Director of the Centre for Critical & Cultural Theory, Charles University, Prague.
louis.armand@ff.cuni.cz
:: DIFFRACTIONS :: 