Becoming Beside Oneself

Brian Rotman

ucla.doc 2/22/99

introduction: the machinic self

Undoubtedly, something irrevocable is happening to the self in the presence of the robotic-informatic web of media systems, external memory devices, and digital procedures threading through it. As offices, corporations, manufacturing and computing become distributed and virtualized, and an explosion of images, hypertexts, internet connectivity, telepresence and technologies of immersion increases communicational and interactive bandwidths, a new self is being installed, or better, is emerging, at the intersection of these techno-effects and our ancient, enbrained bodies. The electronic dolls of the future, all those proxy selves, avatars, technoid personas and digital puppet we can't wait to play with, are getting more enticing by the moment.

But in relation to the effects and affects of technology, the self -- either as too present subject or elusive object -- becomes ever more difficult to locate and theorize. Even when it escapes the salvationist hope or dystopic fear hovering over technology,[1] it risks being drowned in nostalgia for a `natural', `human', less `machinic' version. Raymond Barglow's opens his book, The Crisis of the Self in the Age of Information, with the question: "But who is this `self' ... and why is it at this particular juncture in the history of Western societies the very identity of the self becomes problematic?"; surely not a bad question.(1994:1) But, despite many interesting observations along the way, the puzzle it raises never gets discharged: one is left where one -- the self, the reading/writing subject -- came in. Perhaps any account of the self framed by a psychoanalytically oriented perspective that understands the psyche as a transhistorical, universal and `natural' phenomenon makes such a failure to engage inevitable. After all, psychoanalysis -- either in Freud's original biologistic form or Lacan's linguistically framed True Return to him -- is constrained to offer a psyche whose genesis and vicissitudes are transcultural and independent of history. Technological change is neither of these things, so that only a psyche that is also conceived as being put together within society and history from outside itself, as an extraneously made thing, could possibly register a relation to technology.[2] Perhaps the question has to be: what and how (and not "who") is this self, since the making of a technological self (technological making of a self) involves machinic processes of construction, impersonal fabrication, and formal/material assemblages, not reducible to familial conflicts, imaginary self-identifications or the perennially tragic crises of an always fallen person.

Which is not to say that `selves' can be separated from persons, identities, psyches, consciousnesses, subjects, and all kinds of intentional agents; or that human self-consciousness -- our experience of interiority, auto-awareness, the self-presence of being here -- is illusory. Nor is it to advocate some mysterian humility which refuses all attempts to explicate such a self within a vocabulary not already compromised by it. Far from it. It's to say rather that such interiority cannot provide an explanatory or conceptual origin, and that the deepest question of the self, certainly in relation to technology, is how it gets to be, its assembly. The self may be as is said a natural kind, but it is also a made thing and my understanding here is the machinic processes of technology are (always have been) part of its making. Such an understanding goes most profitably with a view of corporeality/desire that departs radically from what is usually meant by these terms. There are in the broadest sense two models or concepts of desire: desire as lack, incompletion, shortfall, that is, desire as a hole to be filled; or desire as excess, becoming and production, that is, desire as a source. The first is the normal, entirely orthodox concept enshrined within a philosophical tradition from Plato to Hegel, and operative as an unchallenged axiom in Freudian and Lacanian psychoanalysis for which a human psyche's state is one of permanent lack and necessary separation from its `objects'. The second, heterodox and oppositional, was elaborated by Spinoza, re-created by Nietzsche, and made into an materialist ethico-aesthetics by Foucault and Deleuze/Guattari to counter the psychoanalytic, consciousness-centered notion of the `human'. Corresponding to this division of desire, there is, as Elizabeth Grosz illustrates[3], two concepts or models of the body: the (Platonic) privative or lack-body that subordinates corporeality itself to some transcendently ideal mind or soul with respect to which it is always low and inadequate, and the (Spinozist) productive or excess-body that assigns corporeality ontological primacy and whose behaviour is the means for the bringing into being all so-called mental idealities. Plainly, it will be the model of desire/corporeality as excess that will be used here to theorize a self whose assembly is inseparable from technological change.

For Guattari: "[T]echnological machines of information and communication operate at the heart of human subjectivity, not only within its memory and intelligence, but within its sensibility, affects and unconscious fantasms." He goes on to distinguish three dimensions of machinic subjectification:

"1. Signifying semiological components which appear in the family, education, the environment, religion, art, sport ... 2. Elements constructed by the media industry, the cinema, etc, 3. A-signifying semiological dimensions that trigger informational sign machines, and that function in parallel or independently of the fact that they produce and convey significations and denotations, and thus escape from strictly linguistic axiomatics." [1995:4]

Of course, many besides Guattari have recognized from a variety of perspectives the role of socially constructed material systems, practices, and techno-apparatuses in altering/creating human subjectivity[4], but his emphasis on trans-signifying, non-intentional, impersonal effects, operating outside the arena of the directly communicative or beneath its intentional surface is particularly valuable. And within this, the Deleuze/Guattari concept of an assemblage, which permeates their writings, will prove crucial to the digital re-making of selves of interest here. But its very generality requires us to specify a field of operation and a local grounding for it if it is to do work for the idea of self.

Thus, before we go on, I want to locate what my conception of an assembled self might be to a contemporary movement of thought; one that over the past twenty years has been transforming social anthropology, cognitive science, epistemology, learning theory, artificial intelligence, evolutionary/developmental biology, and robotics, and is currently impinging on linguistics and the social anthropology of logico-mathematical thought. I refer to the so-called `contextual revolution', the vector to situatedness in the human sciences, that insists -- contrary to centuries of analytic method in the physical sciences -- on the primacy of context, situation, embodied groundedness, particular setting, and on the importance of the genesis and umgebung of the practices it studies. Central to this shift is the move from top-down descriptions, theoretical analyses and functional decompositions to bottom-up explanations that favour historical, genetic and evolutionary accounts[5]. The shift is from smooth, rule-bound mathematical abstractions and plan-driven schema to the opportunistically put-together particularities of ethology; from context-free abstractions in pre-given terms to situated activity within a vocabulary of emergent effects; from precise, causal chains to loose assemblages of co-occurrent processes; from linear reasoning with its chains of implications, vertical hierarchies, and sequential logic to various forms of horizontal or concentric thought; from, in the lexicon of Deleuze and Guattari, the arborescent to the rhizomatic.

A recent book, Catching Ourselves in the Act, by Horst Hendriks-Jansen attempts to make the elements of this revolution cohere into a framework for a science of the human, based on an explanatory mechanism he calls interactive emergence. According to this (and I simplify enormously), human behaviour can only be properly understood as emerging from an interaction between our biology and a constructed socio-cultural matrix of contexts, artefacts and naratives whose properties and very presence cannot be separated from the demands and facilitations of this biology. In particular, this means that characteristically human mental traits -- having intentions and auto-aware consciousness, using language, desiring, experiencing oneself as a speaking/signing being -- have to be seen as emergent behaviours, constructed or assembled bottom-up from simpler neuro-cultural activities. Let me follow Hendriks-Jansen by illustrating the idea of interactive emergence with two simple, highly suggestive examples, one from from situated robotics the other from ethology. Consider the mobile robot constructed by Meya Mataric at MIT which, when placed in an enclosure with walls, moves parallel to a wall as soon as it finds one; a behaviour that emerges from the interaction of its low-level sensors and actuators with an environment of which it `knows' nothing:

There are no explicit instructions inside the robot that tell it to follow walls. No formal definition of walls is required to produce its behavior. The emergent behavior of the system as a whole is the result of various autonomous activities interacting with each other and with the environment, and not of a centralized system making decisions based on internally represented courses of action or goals. The robot therefore is unlikely to follow a particular wall in exactly the same way on different occasions. Its route will depend on its approach in a specific instance, on noise in its sensor readings, and on numerous unspecified contingencies ... But it can be relied on to follow the wall every time. (8-9)

Or, to take a less mechanical example illustrating a similar effect from ethology that goes back to Konrad Lorenz, consider the so-called `parenting behaviour' of a duck towards its ducklings. This simple-seeming behaviour, it turns out, is an assemblage made up of different activities, with different evolutionary histories, carrying out different functions, and triggered by different stimuli. "The only point at which these activities can be said to intersect is on the duckling `out there'"(7-8). In other words, the duck can be relied upon (otherwise there'd be no ducks) to follow parenting proceedure, but nothing inside it corresponds to an instruction `act a parent'.

Likewise, and this is the point of the examples, nothing inside our heads corresponds to instructions to act a `self'. Yet we do so act. What makes such self behaviour possible is the presence of the appropriate environmental items, the occasions and artefacts which do for the self what walls do for Mataric's robot or ducklings for Lorenz' duck. Of course, humans no less (in fact many orders of complexity more ) than robots and ducks have internal representational structure, they can remember, imagine, introject, re-present and conjure to themselves absent items of their environment and so be/have a self in solitude. But this in no way implies that the self could be solely an internal production. Rather, what is being claimed here is that the self is a behaviour of the body that is a dual, inside/outside phenomenon. It cannot be determined externally as the introjection of the socio-cultural (social, economic or technological determinism), though processes of socialization and subjectification necessarily take place all the time and are fundamental to it. And it cannot be determined endogenously (psycho-biological or evolutionary developmentalism), though such are nonetheless always and crucially present, not least via the body's ancient, pre-conscious auto-awareness, its proprioceptive abilities (which long ago facilitated the construction of pre-linguistic forms of `self'-aware behaviour witnessed in the mirror behaviour of chimpanzees and ourang outangs). The self's duality, then, is what makes it possible for it to be an emergent phenomenon, scaffolded and narrativized on the intersection of a neurology already enculturated and a culture inseparable from natural selection.[6]

Of course, the complexity of this phenomenon is astonishing, as becomes clear from any attempt to think the environmental items --the manifold objects, occasions, events, processes, artefacts, entire technologies, narratives, expectations/procedures of personhood, scaffoldings, etc, -- which allow it to be realized in their presence. The self appears thus as the most complex assemblage of embodied behaviours we can imagine. But it is, nonetheless, a co-presence of less complex flows and happenings, an assemblage of heterogenous behaviours with different intensities and speeds, functions, histories, and futures. And it is, I argue, only as such a fabricated coming together of differnces that its relation to technology can be understood.

So much for how one might begin to think a self that is both biologically grounded and culturally created but reducible to neither. I want now to look at two particular arenas within the contemporary scene, each in its own way enabling a dimension of self to be formed in the presence of a newly technologized practice. The arenas, of computing and visualizing, will provide the necessary external object, akin to the wall for Mataric's robot or the duckling for Lorenz' duck, which functions as the occasion of the emergence of a `self'. The resulting selves correspond to two senses of the phrase "beside oneself" of my title. Thus, beside one self: meaning the move from a monadic 1-self to a multiplex n-self, a move which abandons a singular, primary and originating nucleic interior in favour of the coming into being, or better re-recognition, of a plural self that is heterogeneous and collective. And beside oneself: meaning the displacement of a self occupying its own centre through a movement of difference between selves -- as in `she was beside herself with excitement' -- a trans-self carried over, momentarily de-territorealized or ported alongside itself. I claim that though separable -- as phenomena of machine thought (de-singularized cognitive self) and digital imagery (transported visual self) -- these structures of self arrive and are experienced together in the contemporary informatic scene as a post-serial subject, part of the larger formation of subjecthood being midwifed before, as they say, our very eyes.

computational self: serial 1-self to parallel n-self

For the past 20 years computer science has been reconceptualizing its fundamental verb. `To compute' is now dominated by parallel processes, algorithms and computational architectures, as once it was by serial ones: "Parallel computers", according to Stanford's John Koza, "are the future of computing. Period." [7] If, as MIT's Norman Margolus recently observed, computer architectures are ultimately "abstractions of how we think about reality" 8, then the change from serial to parallel computing is significant; not least since the self (as an effect and/or an illusion) is part of the reality; both in its thinking no less than being thought about. Quite independently of this, the difference between serial and parallel is a widely dispersed and deeply embedded semiotic phenomenon, and some indication of this might highlight just how consequential the shift within computing might be.[9]

The poles of the duo are familiar. Serial refers to doing one thing after another (domain of succession and sequence), parallel to doing many things at the same time (co-occurrence and simultaneity). Though able to be separated and assigned positive content, these poles function differentially, as opposed axes which form the armature of conceptual apparatuses or machines organizing and assembling objects, states of affairs, or abstractions, which subsequently transcend, unite or dissolve the very difference between serial and parallel that facilitated their creation. Some examples. Music with harmony and pitch as parallel, and melody and rhythm as serial (so-called "vertical" and "horizontal" music respectively). Arithmetic with ordinal as serial against the parallelism of cardinal number.[10] Quantum physics: interchangeability of an ensemble of single, serially occurring events with a plurality of simultaneous ones. Speech: the linked hierarchies of distinctive features, phonemes, and syllables. And so on, for symbolic communication, film editing, types of memory, electrical circuits, and entire technologies -- telegraphy (serial transmission of electical pulses) against photography (simultaneous imprinting of light).

Let's return to computing and the serial architecture which dominated it until recently. I'll cite two explanatory proposals for this domination. Hendriks-Jansen makes the suggestion that in formalizing a mechanical mathematical procedure to create his model of thought, Turing (acknowledged father of the sequential model) drew on the idea of an assembly line, "a mechanism that was familiar and well understood by most of his contemporaries" (1996:94), transferring the belt into the endless tape of his machine and enforcing linearity by allowing the objects so assembled to change the ongoing operations of the line itself. The idea is interesting in relation to the construction of a self in that, by tracing sequential computation back into Taylorism and its roots in time-and-motion study, it makes a connection between the sequential behaviour of machines and of the individuals constructed/necessitated by them. From another direction, James Bailey, also interested in this connection, asks why human `computers' (as they were called) were so easily replaced by electronic counterparts when first introduced, as if we'd prepared for them. His answer is that in a sense we had: science's adherence to linearity, top-down analysis and cause-effect explanations ensured that we'd build machines that "inherited all the sequential ways of expressing and formulating science that had developed over twenty-five hundred years". (Bailey 1992:67) An interesting suggestion of why parallelism was avoided, but one that says little about the seriality embodied in the universal Turing machine dominating computational architecture, and nothing about the self accompanying it.

It's said that Turing arrived at his celebrated idea of a computing machine through introspection; observing his inner self carrying out the individual steps of a proof or calculation. Without question, the conception Turing arrived at -- a linear, one-thing-after-another, self-contained, sequence of separated computational moves -- matched what most at the time and many still consider as the workings of our minds performing a `chain' of reasoning; a circumstance that has allowed it to serve, in Edward Hutchins' description, as "an origin myth of [a] cognitive science" committed to a computational, symbol-processing understanding of human thinking. Such a science rides on a confusion concerning the locus of computation:

[W]hen the symbols are in the environment of the human, and the human is manipulating the symbols, the cognitive properties of the human are not the same as the properties of the system that is made up of the human in interaction with these symbols. The properties of the human in interaction with the symbols produces some kind of computation. But that does not mean that that computation is happening inside the person's head. (1995: 361)

The computational architecture fostered by this misapprehension "is not a model of individual cognition. It is a model of the operation of a sociocultural system from which the human actor has been removed." (363) What this entails is that both the materiality of the signifying apparatus and Turing's own corporeal activity in relation to it are absent. Elsewhere I've pursued the transcendentalizing consequences of this supressed corporeality.[11] Of interest here is that remedying it, inserting a properly corporealized agent back into the picture of cognition, as Hutchins' critique demands, would escape from the model of a single, linearly proceeding invisible, inner self and open the way to linking the cognizing self to a plurality of other -- explicitly acknowledged -- agents. A path that leads in other words to parallel cognition and computation.

Parallel or distributed computation divides tasks, data, algorithms, resources, instructions and memory between separate but interconnected agents who/which perform their operations simultaneously. The idea is as obvious, familiar and fundamental as social action and the division of labour, as computer scientist B. Chandrasekaran noted at the beginning of parallel computation's rise to prominence: "Social organizations from honeybee colonies to a modern corporation, from bureaucracies to medical communities, from committees to representative democracies are living examples of distributed information processing embodying a variety of strategies of decomposition and coordination." (1981: 1) To which could be added crowds, workgroups, packs, networks, couples, families, and audiences; not to mention non-human collectivities from slimemoulds, reefs, and colonies to all manner of flocks and swarms. Computer science's conversion to parallelism amounts, then, to a belated recognition of co-present or co-active subjects or agents at a site long, deeply and mistakenly held to be the sole creative province of an individual, serially conscious agency.

Long familiar in the biological sciences (multicellularity, social organization, symbiosis, the structure of organs like the eye, the simultaneity of the body's systems, and so on), distributed and parallel architectures are now starting to diffuse through the technosciences. An example from the study of fluid phenomena: in twenty years, investigation of the equations governing fluid flow has gone through three stages: a) imagination (attempting to solve the equations of flow via sequential reasoning of mathematical proof with a theoretically infinite accuracy), b) computation (solving the equations to any specified degree of accuracy via sequential calculations), c) simulation (replacing the equations by parallel micro-computations of a cellular model of the reality they idealise). We shall meet another version of the transformation from theoretical imagination to material simulation in connection to vision, where it will manifest as the replacement of the transcendental Euclidean point by the discrete, material pixel.

In other words, parallel machine thought is the facilitating artefact, the wall in our robot example, in whose presence an n-self in place of the 1-self of linear computation emerges. But for this to happen, there must be present an appropriately collective internal arrangement. Just such a collectivization of the self from below, an interior multiplex self, invisible to the conscious, introspecting monadic self, has long been thought within philosophical psychology, but only recently taken seriously by mainstream neuro-cognitive science when it recognized the end of the homuncular mind, in which a tiny agent, an internal I/Me, is invoked to run the psyche. The disappearence of such an internal director leaves what Fransisco Varela calls a "selfless self" (1991), a brain with no center, no controlling entity standing in for a unitary self, but a distributed collection of independent, linked systems or subminds working together and against each other.

Computer scientist Marvin Minsky's phrase "Society of Mind" (1987), perhaps the most cited contemporary metaphor for this inside-the-head parallelism, is a rediscovery of past rediscoveries. A century before him, the idea was widespread. William James wrote of an inner mental multiplicity; Robert Louis Stevenson asserted: "Man is ... truly two. I say two, because the state of my own knowledge does not pass beyond that point. Others will ... outstrip me on the same lines; I hazard the guess that man" (and he means individual humans not mankind) "will be ultimately known for a mere polity of multifarious, incongruous and independent denizens."[1979: 82]. Whilst Nietzsche, ahead of the game as usual, suggested: "The assumption of a single subject is perhaps unnecessary: perhaps it is just as permissable to assume a multiplicity of subjects whose interaction and struggle is the basis of our thought and our consciousness ... My hypothesis: The subject as multiplicity." (1968: 260) More than a century before this, David Hume likened the mind to "a kind of theatre" and argued that "The true idea of the human mind is to consider it as a system of different perceptions or existences, which are linked together ... and mutually produce, destroy, influence, and modify each other. ... I cannot compare the soul more properly to anything than to a republic or commonwealth, in which several members are united by the reciprocal ties of government and subordination." (17xx)

These (persistently socialized) figures for the mind -- society, polity, system of existences, commonwealth, theatre, republic -- testify to a recurrent refusal to see it as a private, undifferentiated singularity. The historian Hillel Schwartz commenting on multiple personas and doppelgangers that emerged at the end of the 19th and 18th centuries respectively, sees these as a response to social and economic dislocation, to "societal fears of the loss of generational continuity, most acute at the ends of centuries". (1996: 81)[12] Maybe. Though Hume wrote mid-century, and at this century's end the effect is inseparable from a converging of ideas in neurology, robotics, and post-cognitivist theories of mind onto the notion that even the simplest mental activities, for example counting, are the result of a co-ordinated multiplicity of actions.[13] Moreover, Schwartz's explanation conflates two sorts of multiplicity: a derivative multiplicity, a many which comes from a prior one by splitting, copying, and so on; and a primary or aboriginal multiplicity of differences from which units are formed by way of bundles, assemblages, and so on. There is a gulf between the two. Thus, for example, n seen under the former represents a split whole, in the case of 2 a multiplicity captured by the binary, familiar from Freud's model, Jung's Shadow persona, and the popular picture of schizophrenia as split self. Against this, n conceived under the latter is an assemblage, in the case of 2 a binary captured by multiplicity, familiar from dual or so-called multiple personality disorder. Thus, bearers of multiple identities, who may well have been, as is now claimed[14], narrativized into existence by the therapists who treated them, are (by virtue of this very possibility) a handy emblem/symptom of a postmodernity in which an n-self -- multiple or collective, dispersed, distributed -- appears inescapable.[15]

visual selves

There is a moment in Notre Dame de Paris when Victor Hugo dramatizes the demise of images and architecture as the means of educating an illiterate population by having a priest hold up a book against the cathedral and declaim "this will overcome that". There is more than one moment in Imagology when media philosophers Mark Taylor and Esa Saarinen declaim the arrival of the "age of post-literacy" in which the book becomes Hugo's cathedral, and held up, poised to overcome it, like a pre-literate pictogram returning on an Hegelian spiral, is the digital image.

In the thirty years since McLuhan, much has been written on the effects of writing (as technology or medium not literature) on the self, from Walter Ong's "writing restructures consciousness" (1982: 78) to an appreciation of the medium's role in creating the self's temporality, as in Roy Harris' observation that the writing of diaries allowed "[i]ntegrating knowledge of an earlier self with knowledge of a later self [to become] a semiological process, subject to conscious control and evaluation."(1995: 44) Are we about to witness a replay of this in which a post-literate self is emerging, patterned less on the word -- stable, integral, discrete, enclosing a unique, interior meaning -- than the fluid multiplicities of the image? Artist Helen Chadwick noting that "The solitary repressive ego harnessed in language is sovereign", goes on to ask, "What if dangerous fluids were to spill out, displacing logic, refuting a coherent narrative, into a landscape on the brink of I." The consequences for the self's containment, as Chadwick avers -- "Before I was bounded, now I've begun to leak ..." -- are likely to be dramatic and intense in their subversion of a self-contained self. (1989: 29).

A becoming fluid already announced by Jean Baudrillard who proclaimed the "liquifaction of all referentials" (1983:4) at preciely the point when data started to flow in streams and alphanumeric solidity appeared on the surface of the liquid crystal screen. Since then "liquid television", "liquid architecture", "liquid software", "liquid typography", and many other meltings. The large-scale vehicle for the liquifaction of the I conjectured by Chadwick is likely to be not art (notwithstanding its representation in her own paintings) but the informatic or semiotic image, by which I mean information-bearing, instructional, explicatory, and otherwise pragmatically oriented images -- maps, diagrams, tables, charts, schematics, figures, wire-frame renderings, simulations, arrays, scans, and so on, rather than artistic images. We are in the midst of an explosion of such visual artefacts, whose sheer scope, richness of content, heterogeneous means, stylization, subtlety, and communicational depth prompted James Elkins to goose the art historical community with the thesis that such non-art images "engage the central issues ... and are [as] fully expressive and capable of great and nuanced a range of meaning as any work of fine art." (1995: 553) Surely Elkins' provocation is worth pursuing. And surely one can go further along the route it opens up: just as writing midwifed a certain self, so semiotic images, as much as those of art, are now capturing the sites where psyches are narrativized, analogized and reproduced; operating -- in terms of the robot metaphor -- as the wall for the emergence of a new visual self.

There are precursors: the semiotic image has risen to power twice before, riding each time on the inadequacy of the written word, whose linear logic, sharp boundaries, and sensitivity to syntax seem too rigid, precise, foregone in meaning, to cope with a wave of novel, rapidly changing knowledge. First, in the mid-fifteenth century, as a result of perspective drawing which produced, according to historian Samuel Edgerton, "a pictorial language that ... could communicate more information, more quickly, and to a potentially wider audience than any verbal language in human history." (1980: 189); a phenomenon which (subsquently consolidated by the printed book) is inseparable from the emergence of the autonomous Renaissance self. Second, in the latter half of the nineteenth century, in the wake of the extraordinary image fest unleashed by photography and numerous image producing and recording instruments, which led Jean-Louis Comolli to speak of a "frenzy of the visible" (1980: 122); a phenomenon inseparable from the emergence of a self characterized by Ian Hacking as a "consensus of vital properties" (Quoted Schwartz 1996: 81).

And now this third wave in which, according to historian Anne-Marie Willis, "The means of production of ... visual imagery is undergoing a mutation as significant as the invention of photography" (1990: 197). The mutation is digitization, the enabling technology of the post-photographic practices[16] behind the vast upsurge of contemporary semiotic images that Richard Friedhoff glosses as the "second computer revolution" (1989). And, as with its predecessors, there is the question of the transformation of the self accompanying it.

The visual image has proliferated as the control provided by digital software has closed the gap between the imaginable and the imagable: any item designed, visually specified, made up from images can be presented as a single visual object. Once, one would have called these objects montages or collages differentiating them from the standard unitarily conceived image presented at the centre of itself, but now such "Techniques of assemblage", as Tony Fry calls them, are "the general condition of production." (1990: 170) The image, as assembled co-presence of images, constitutes the contemporary visual paradigm encountered across the board from what is virtually a default format of magazine pages through multiple Picture In Picture use on TV to overlapped image techniques of music videos. The revolution appears as the invention of a form of pictorial polyphony with visual samples as voices; as if the image, or rather the optic which delivers it to us and by means of which we, as visual subjects read it, has gone parallel, confronting us with an imagized image -- an image made of constituent images which variously overlap, juxtapose, underlap, penetrate, displace, obscure, truncate, and embed each other.

En route to expositing the emergence of a corresponding imagized self, there are two revolutionary dimensions of the digital image -- one pragmatic, the other existential -- I want to note. The pragmatic dimension can be observed by considering, from the above range of relations, superimposing or layering one image on another as this is deployed for example in contemporary map making. The digitized image's ability to be copied without degradation allows cartographers to repeat it within any number of independently controllable, accessible and updateable layers so that they can be superimposed and interact with each other. The result -- the GIS (Geographical Information Systems) cartographic revolution -- is a class of images emergently different from the pre-digital map.[17] As soon as a map is created in this way, it ceases to be an inert, static picture of a world exterior to it, but becomes dynamic, a proactive tool capable of impinging epistemically on the world it represents. Pragmatically, such images deliver a parallel image, a co-present assemblage of images that had previously to be viewed serially or side by side. In this way, they internalize the relation of one image, and one cartographic observer, operating beside or inside another. The result, by no means confined to cartography, is an image and viewing subject that is decentered or distanced from itself, in that it is assembled externally from an open-ended design rather than from a singular, endogenously constrained and mimetically conceived source.

The existential dimension can be observed within mathematical diagrammatics, where visualization software is creating an experimental form of a subject that has long and proudly understood itself as purely theoretical. Digitization enables mathematicians to plot graphs instead of laboriously using the calculus; thus pushing toward obsolescence an apparatus that took centuries of development to predict/calculate geometrical atributes because it was unable to draw them. It enables recursive iterations of mathematical models of simple physical systems to be visualized via cellular automata. It is also adding content to the subject via fractals and chaos maps; allowing the visualization of unsuspected topological surfaces; making thinkable mathematical ideas, concepts and processes not previously imagined; and so on. The cumulative effect is that, like GIS maps, digitized images in mathematics are a medium for construction, for creation and active exploration; in this, they exceed the existing category of diagrams insofar as these latter have hitherto been conceptualized as passive descriptions of an external and invisible (mathematical) world.

There is an issue here of mathematical ontology that bears on the self. The very use of digital imaging tools foregrounds the existential status of mathematical entities, since mathematicians cannot but be confronted by the juxtaposition of two opposed understandings of the ideal objects they study. The classical, orthodox viewpoint: mathematical objects as transcendental, invisible and imagined; the digital, experimental viewpoint: mathematical objects as materializable, variously idealizable and imagable. The difference is fundamental: the first describes, for example, a Euclidean point, contentless, infinitistic and zero dimensional; the second, a material pixel with real, specifiable dimensions and variable informational content. Thus, by displacing the immaterial, zero-sized point, pixelation achieves performatively, in a stroke as it were, the dismantling of an entire metaphysical apparatus of classical ontology that philosophical critique of mathematics has yet to accomplish. In this, it is a philosophically potent version of the shift from theoretically proof to material simulation of fluid flow mentioned earlier.

Such materiality is an important link between the digital, post-photographic subject and a pre-photographical self. Thus, according to Jonathan Crary, the notion of self -- he says "visual observer" -- inculcated by perspective and the camera obscura, a "privatized isolated subject", a "unitary consciousness detached from ... the exterior" with its "metaphysic of interiority", (1988: 33) collapsed in the early nineteenth century under the onslaught of a physiological investigation of sight and consequent understanding of vision, as physical, material, embodied. This collapse was masked, however, by the realistic claims made for photography and film as `objective' and truth-producing representational apparatuses -- the camera doesn't lie -- leading to the re-institution of the camera obscura's model of the subject as what Crary calls a "mirage" of this overthrown self. What the post-photographic digital image does is dissolve the mirage and put in its place a dispersed, material visuality which finally jettisons the Renaissance perspectivalism still clinging to the photographically real self.

It can achieve this because the installation of such a post-perspectival self results from a process that reaches down to the level of the image's atomic structure. As an optical regime, perspectivalism constructs, and is constructed upon, a self, i.e., an ensouled individual, with a point of view, a transcendentally specified location mirroring the vanishing point. But digitization, substituting pixels for points, replaces the psychic architecture and "metaphysic of interiority" of the Renaissance individual by an architecture that, because it must be specified in relation to the physiologically meaningful substrate of the pixelated image, cannot transcend the space it physically occupies, and so cannot enact a metaphysical drama of viewing the world from a position outside it. In Crary's terms, the invisible seeing soul has been finally replaced by the physiologically sighted body.

To Refuse perspective in this way is to refacilitate embodied motion: the visual self, free to leave the rigid stasis imposed by the requirements of Alberti's window, can move, both in imagination and literally. Within non-art this leads outside the stand-alone semiotic image I've been considering to simulated fly-through and, via haptic and kinesthetic feedback loops, to telepresence and immersive extensions of imaging; an extension already at work, as Tim Lenoir (Lenoir in press) has demonstrated, to the re-making of the surgical subject.

Of course, within artistic practices, embodied movement can be imagined in the face of an unmoving image. Thus, David Hockney, speaking of how, in contrast to the requirement of standing in the middle of a circular perspective of Versailles -- "my body seemed to have been taken away", "reduced to a tiny immobile point" -- he was able to revel in the hours spent "wandering up one street, down another, up another" in such a way that "I still had my body", as he viewed the stylized non-perspective of a Chinese scroll depiction of a city. (1993: 128). Or it can be literal movement, the viewing subject moving in relation to what is being viewed, as occurs in so much contemporary installed art, particularly heavily technologized art such as video installations. Here, the image is de-framed and becomes mobile, dispersed, distributed. Videoart installations routinely surround the viewer/participant in space and require -- like the perambulatory form of medieval theatre -- to be navigated around, physically negotiated and experienced in place. Viewing becomes subordinated to an active engagement with the installed objects and scenes surrounding a moving eye. "Engulfed by the assemblage of temporal parts, the process of looking", becomes, in Barbara London's words, "as much about the physical experience as the composite memories that live on in the mind." And so such installations "illustrate the dissolution of the seriality of time that characterizes the late twentieth century." (1996: 14) In terms of the subject, this can be read as the falling away from a one-dimensional, singular consciousness into parallel, distributed co-presence. What is being analogized and narrativized by such installations is another version of the prepositional self already charted as emergent in the face of the imagized image. Only here, instead of the non-art informatic image as the external screen with respect to which such a self emerges we have art practices in the form of dynamically viewed installations.

Enough, then, about the visually emergent self. I want to close by returning to the nexus of technology and the subject. We've encountered two abstract machines, computing and imaging, operating as the external environmental sources for the interactive emergence of two forms of subjectivity, two meanings of a self beside itself. But though distinct, these selves arrive and are experienced together within the contemporary informatic scene. One reason for this is undoubtedly their common antagonism to (certain effects and usages of) the serial and conversely their immersion in the parallel. Though I've not pursued it here, one can, as just indicated, see this as part of a large-scale contemporary tropism to co-presence, parallelism and modes of distributivity that reject linearity.[18] As such, it is surely related to the construction of a new mode of simultaneity or worldwide co-presence -- globalization -- being put in place by transnational capital able to digitize and hence telecommunicate all that it needs to be present everywhere; a construction of newly universal `now' that appears as an iteration of the creation of World Standard Time by industrial capital at the end of the last century.[19]

And what of becoming? Becoming beside oneself, is not to be identified with imitating, reproducing, splitting or doubling oneself; or identifying with, assimilating, or incorporating something into oneself; or being reborn; though it can couple with or be traversed by all these. It is rather a form of atemporal change, becoming party to a condition other than one's own, a question of self difference, of standing to the side of oneself. At the same time it engenders a condition of de-singularity in which, by acceding to an aboriginal multiplicity, one becomes not one. It's easier perhaps to talk in attributes and processes: of an interactively emergent psyche, a self assembled on the neuro-cultural interface that is internally and externally collective, a self distributed and in excess of unity. But one can still ask: what does it mean, where does it go, how will it feel, to be so? Can I, you, those to come, really not be what we have been for so long in Western culture, an I that is before everything, as a condition for all else, an enclosed, individual, indivisible, opaque, private, singularly rooted Me? Can the self function in ways other than this organized arborescence, can it manifest the qualities of what Deleuze/Guattari, in their all too generous vocabulary, call a rhizome? Do we want it to? Do we desire to become multiply co-present, heterogeneously connected to ourselves, pulled in all directions from the future, always in the process of becoming multiple, beside our selves with glee and dissolution, intermittently present to ourselves, each of us a psyche on the edge of its seat? A self beside itself. Why not?

In directly practical terms, a fake question: do we really have much choice? Aren't we being dragged, like it or not, into a world of ever more numerous co-happenings? Isn't parallel computation and the relentless co-presencing and distribution it facilitates/demands already starting to control numerous social and cultural sites where subjects are produced? Isn't the multiple optic and its post-collage image already deeply embedded in various semiotic and art practices constituting appropriately multiple visual subjects? No doubt. But in aesthitico-ethical terms, the question could not be more real. It asks for an embrace of the post human, an embrace of desire conceived not as lack, shortfall and an always unfillable emptiness -- the humaness of our obeisance to an impossibly perfect deity -- but desire as openness to the other of what we were and always thought we had to be. At the beginning of this essay, as a matter of methodology I rejected the psychoanalytic concept of desire on the grounds that it was incapable of articulating the nexus of technology and self. This has enabled me to get to the point where the question -- why not become beside oneself? -- could be meaningfully posed. So again: a time-honoured, self-contained eternal psyche understood (in the guise of a psychoanalytic axiom) as a familiar hard-nosed Realpolitik of the tragic, fallen-from-god soul; or a distributed, multiple psyche of never-dischargeable excess: a heterodox seeming madness of universal becoming without foundation or end. Which?


[1] For techno-euphoria, in born-again, market-blessed format, see any issue of Wired magazine. For techno-estrangement, see Timothy Druckrey (1995) in which the computer's unexpungeable origin in the military's pursuance of command, control and communication surfaces in a shift from "agency to behavior" whereby "perception becomes scanning, retention becomes retrieval and thinking becomes processing", leading to a "self ... understood as an operating system" [2]. For techno-fatalism as perverse, aphoristic hype and brilliantly turned celebration, see Jean Baudrillard's many writings.

[2] The lack of such externality is the problem with the account of technologically inflected -- specifically internet created -- identity given by Sherry Turkle. She sees multi-user domains (MUDs) as the means by which individual users can "cycle through" a set of virtual identities simultaneously accessible via a multi-windowed screen environment; identities which are "multiple, fluid, and constituted in interaction with machine connections." (1995: 15). But, like Barglow, Turkle is committed to a psychoanalytic concept of the self and with it a voluntarist concept of person that blocks embedding her fieldwork into a larger, theoretically more productive view of the nature of these machine connections.

[3] See Grosz 1994, especially chapters 5 and 7, for an illuminating examination of the body in Nietzsche, in Foucault, as well as the uptake of their ideas in Deleuze/Guattari.

[4] Thus, Foucault's elaboration of how self-scrutiny gets assembled via the internalization of the confessional and of panoptic surveillance; Marshall McLuhan's identification of media machines which act "directly toward shaping our most intimate self-consciousness." (Quoted Krokers 1996:); Jean-Louis Comolli's work on "machines of the visible" (1980) linking filmic materiality to audience perception; Friedrich Kittler's tracking of the triple usurpation of the book as chief imagination machine of Western culture by the separate media of film, typewriting and grammophone (1998); Walter Ong (1982), Jack Goody (1977), Roy Harris (1996), David Olson (1994), and Elizabeth Eisenstein (1983) on writing and printing's restructuring of speech and consciousness; Bruno Latour's tracing of non-human agency and sign production (19xx); Timothy Lenoir's articulation of bio-medical theorizing and medical subjectification (in press); Jean-Francois Lyotard's self as "a person [who] is located at `nodal points' of specific communication circuits ... through which various kinds of messages pass." (1984: 15)

[5] A fuller treatment of the metaphorics of bottom-up/top-down and situatedness in relation to social anthropology, robotics, cognitive science and developmental biology is given in Rotman (1994). For an attempt at a situated linguistics, under the label "integrative semiology", see Harris (1995), and for a start on the task of constructing a situated account of the "enculturation" of logico-mathematical practice, see Raven (1996).

[6] One consequence of this way of understanding consciousness -- as emergent on the neuro-cultural interface and not as a purely cortical phenomenon generated from within or as created from without by an internalization of the social -- is to render inoperable the coming into being of a world mind or soul or collective (un)consciousness insofar as this is based on the analogy between the complexity of neuronal connections producing consciousness in the brain and the complexity of individual connectivity, via the internet say, producing the required world entity.

[7] Quoted in Kevin Kelly Out of Control 397.

8 Quoted in David Cary, Cellular Automata Machine,

[9] For more on the serial/parallel duo, see Rotman 1997.

[10] Many more examples from mathematics could be given from such ur-concepts as ordered against unordered pair in set theory to the opposition of dependent/independent events in probability theory. Tobias Dantzig is perfectly correct when he observes that "Correspondence [parallel] and succession [serial]" are "two principles which permeate all mathematics" (1985: 9). The machines associated with numbers are however perhaps the most fundamental. In the finite case, the machine produces abstract numbers which are neutral between cardinal and ordinal, controlled by an arithmetic that runs on this neutrality. Thus, ordinals, which are counted into being, give rise to cardinal magnitudes, plural co-presences, by virtue of the steps in the counting process being recorded (rather than forgotten, which would reduce counting to bare or naked repetition). Conversely, any cardinal, as a collection, gives rise to an ordinal by virtue of its elements being (able to be) listed or well-ordered into a sequence. In the infinite case, the capacity to well-order an arbitrarily presented set ceases to be unproblematic and has to be posited in the form of the axiom of choice. The result is a much more complex, irresolvable and intrinsically separated relation between cardinal and ordinal numbers than in the finite case. Interestingly, some of this complexity and undecidability re-emerges within `finite' arithmetic if one critiques the idea of finite/infinite within classical arithmetic and re-writes `finite' accordingly: the resulting arithmetic exhibits a fundamental separation between ordinals countable from zero and cardinals nameable by notations but not so countable. See Rotman 1993, especially chapters 4-6, for an extended treatment of this question.

[11] The result of this supression is what is meant by the "ghost in Turing's machine" in Rotman 1993. Exorcizing it is not a simple exercise, since without it there is no way Turing could even have formulated the requirement -- crucial to the claim of universality made for his machine (and the proof of that claim) -- that it have an endless tape.

[12] See, van den Berg 1974 cited by Schwartz for a detailed treatment of the links between fin de siecle anxieties and multiple personas.

[13] See Deheane (1997) for an elaboration of the neurology behind this and its implications for a new understanding of arithmetical thinking.

[14] See Acocella 1998 for a recent expose of the role of therapists in the MPD phenomenon, and Hacking 1995 for a history of the iatrogenic origins of multiple personas in the context of the theorization of memory.

[15] But not without risks and humour, as the following posting to a Multitasking Victim Support Group attests: "Maybe we should all just ... yeah, I'm back. No -- Could you fax this? Thanks. Who are you? Okay. ... get used to technology-induced schizophrenia. Delete. Hello?" Jeffrey Obser 11/5/97. Thanks to Rich Doyle for this.

[16] According to Walter Benjamin, the emanation from a unique art work, its aura, is destroyed by the mechanically reproduced photograph. Digital imaging repeats this on the photograph, replacing the immediacy inherent in the chemical fixing of light, the direct trace of the real which is its aura, by unlimited mediation and mutability. As a result, the chemical photograph becomes available to post-photography as a visual effect or style, a citation of the pre-digital real. The camera's guarantee of truth is displaced onto a choice of representation between more or less digitized `fact', instituting what William Mitchell has called "an ontological aneurism ... in the barrier separating fact and fancy." (1994: 189)

[17] GIS maps now dominate cartography on every level from architectural (building interiors) and urban (distribution of roads, paths, sewer pipes and telegraph poles) to geophysical (satellite images of land use, coastline change, troop movements) to galactic (interstellar dust and star charts), from "outer space", as one booster has it, "to slam in your face."

[18] See Rotman 1997 for more on this.

[19] See Kern 1983 for a richly textured account of the latter, as well as a survey of different concepts of simultaneity from the one central to the Einsteinean break with classical, Newtonian physics to various kinds realized in early twentieth century experimental poetry.

Works Cited

Acocella, Joan "The Politics of Hysteria", New Yorker, April 6, 1998

Bailey, James "First We Reshape our Computers, Then our Computers Reshape Us: the Broader Intellectual Impact of Parallelism", Daedalus, Winter 1992

Barglow, Raymond "The Crisis of the Self in the Age of Information" (London, Routledge: 1994)

Baudrillard, Jean "Simulations" (New York, Semiotext(e), Inc: 1983)

Chadwick, Helen "Enfleshings" (London, Secker and Warburg: 1989)

Chandrasekaran, B "Natural and Social System Metaphors for Distributed Problem Solving: Introduction to the Issue", IEEE Transactions on Systems, Man, and Cybernetics, 11, 1-5, 1981

Comolli, Jean-Louis "Machines of the Visible" in Teresa de Lauretis & Stephen Heath (eds) "The Cinematic Apparatus" (New York, St. Martin's Press: 1980)

Crary, Jonathan "Modernizing Vision" in Hal Foster (ed) "Vision and Visuality" (Seattle, Bay Press: 1988)

Dantzig, Tobias "Number: The Language of Science" (New York, The Free Press: 1985)

Dehaene, Stanislas "The Number Sense" (Oxford, Oxford University Press: 1997)

Deleuze, Gilles and Guattari, Felix "A Thousand Plateaus" , translated Brian Massimi, (London, The Athlone Press: 1988)

Druckrey, Timothy "Information, Interactivity, Neuro-Technology", TightRope, issue 3, 1995

Druckrey, Timothy (ed) "Iterations: The New Image" (Cambridge, MIT Press: 1994)

Edgerton, Samuel Y. "The Renaissance Artist as Quantifier" in Margaret Hagen (ed) "The Perception of Pictures", volume 1 (New York, Academic Press: 1980)

Eisenstein, Elizabeth "The Printing Press as an Agent of Change" (Cambridge, Cambridge University Press: 1983)

Elkins, James "Art History and Images That Are Not Art", Art Bulletin, LXXVII, (4), December 1995

Emmer, Michele (ed) "The Visual Mind" (Cambridge, MIT Press: 1993)

Friedhoff, Richard "Visualization: the Second Computer Revolution" (New York, H.N. Abrams Inc: 1989)

Fry, Tony Art "Byting the Dust - Some Considerations on Time, Economy and Cultural Practices of Postmodernity" in Hayward 1990

Goody, Jack "The Domestication of the Savage Mind" (Cambridge, Cambridge U. Press: 1977)

Grosz, Elizabeth "Volatile Bodies: Toward a Corporeal Feminism" (Bloomington: Indiana U. Press: 1994)

Guattari, Felix "Chaosmos: An Ethico-Aethetic Paradigm" (Bloomington, Indiana U. Press: 1995)

Hacking, Ian Rewriting the Soul: Multiple Personality and the Science of Memory (Princeton, Princeton U. Press: 1995)

Harris, Roy "Signs of Writing" (London, Routledge: 1995)

Hayward, Philip "Industrial Light and Magic" in Hayward 1990

Hayward, Philip (ed) "Culture Technology and Creativity in the Late Twentieth Century" (London, John Libby: 1990)

Hendriks-Jansen, Horst "Catching Ourselves in the Act" (Cambridge, MIT Press: 1996)

Hockney, David "That's the Way I See it" (San Fransisco, Chronicle Books: 1993)

Hume, David "A Treatise of Human Nature" (...)

Hutchins, Edwin "Cognition in the Wild" (Cambridge, MIT Press: 1995)

Jenks, Chris(ed) "Visual Culture" (London, Routledge: 1995)

Kelly, Kevin "Out Of Control: The New Biology of Machines" (London, The Fourth Estate: 1994)

Kern, Stephen "The Culture of Time and Space: 1880-1918" (Cambridge, Harvard Press: 1983)

Kittler, Friedrich "Grammophone, Film, Typewriter" (Stanford, Stanford University Press: 1998)

Kroker, Arthur and Louise "Hacking the Future" (New York, St. Martin's Press: 1996)

Latour, Bruno

Lenoir, Tim + Lecuyer, Christophe "Visions of Theory: Fashioning Molecular Biology as an Information Science" in Wise, Norton (ed.) "Growing Explanations" (Stanford, Stanford U. Press: 2000)

Lenoir, Tim "Authorship and Surgery: Pixellated Presence" in Galison, Peter and Biagiaoli, Mario "Authorship and Science", (Chicago, University of Chicago Press: in press)

London, Barbara "Video Spaces" Performing Arts Journal XVIII (3), 14-19, 1996

Lyotard, Jean-Francois "The Postmodern Condition: A Report on Knowledge" (Manchester, Manchester University Press: 1984)

Minsky, M "The Society of Mind" (London, Picador: 1987)

Mitchell, William J. "The Reconfigured Eye: Visual Truth in the Post-Photographic Era" (Cambridge, MIT Press: 1994)

Nietzsche, Friedrich "The Will to Power", trans. Walter Kaufmann, (New York, Vintage Books: 1968)

Oakley, David and Eames, Lesley "The Plurality of Consciousness" in David Oakley (ed) "Brain and Mind" (London, Methuen: 1985)

Olson, David "The World on Paper" (Cambridge, Cambridge U. Press: 1994)

Ong, Walter "Orality and Literacy: The Technologizing of the Word" (London, Methuen: 1982)

Raven, Diedrick "The Enculturation of Logical Practice" Configurations 4, 381-425, 1996

Romanyshyn, Robert "Technology as Symptom and Dream" (London, Routledge: 1989)

Rotman, Brian "Ad Infinitum ... the Ghost in Turing's Machine: Taking God Out of Mathematics and Putting the Body Back In" (Stanford, Stanford University Press: 1993)

--- "Exuberant Materiality -- De-Minding the Store" Configurations 2, 257-74, 1994

--- "Counting Information: A Note on Physicalized Numbers" Minds and Machines 6, 229-238, 1996

--- "Going Parallel" 204i/Rotman/Beside/top.html, 1997 (to appear in SubStance)

Schwartz, Hillel "The Culture of the Copy" (New York, Zone Books: 1996)

Stevenson, Robert Louis "Dr Jeckyll and Mr Hyde" (London, Penguin Books: 1979)

Tauber, A (Ed) "Organism and the Origin of Self" (Dordrecht, Kluwer: 1991)

Turkle, Sherry "Life on the Screen: Identity in the Age of the Internet" (New York, Simon and Shuster: 1995)

Varela, Fransisco "Organism: a Meshwork of Selfless Selves" in Tauber 1991

Willis, Anne-Marie "Digitisation and the Living Death of Photography" in Hayward 1990