Are We Posthuman Yet?
N. Katherine Hayles
How We Became Posthuman: Virtual Bodies in Cybernetics, Literature, and Informatics,
Chicago: University of Chicago Press, 1998
N. Katherine Hayles's How We Became Posthuman is a difficult book to evaluate because of its chief virtue: its narrativity. I find myself falling into the supine position of writing a book report rather than a book review. In support of her contention that unacknowledged everyday narratives construct and maintain assumptions about the nature of the subject, the human, and the (gendered) body, Hayles offers a narrative of her own: a narrative of narratives. Her technique is so effective that I am tempted to retell it, to get the "story" out, particularly to non-twentieth-century specialists like myself - much as I passed around Neuromancer in the 80s to anybody who would take direction. In the case of Hayles's book, the reason for this contagion is not that the story is unproblematic, not that the story is especially radical or new: it is more, I think, that her narrative works as a kind of fixative; it integrates its myriad components into a coherent fabric. Or better yet, to invoke the theme of one chapter, narrative, by enlisting embodied memory through subvocalization, through silent reading, gets under the skin in the way atemporal, conceptual exposition does not. Hayles's book got under my skin, like a satisfying work of speculative fiction.
How We Became Posthuman tells a twentieth-century tale of "how information lost its body." Hayles gestures towards the origins of this tale in the formation of the liberal humanist subject, citing C.B. MacPherson's concept of "possessive individualism," the freedom to dispose of property at will, particularly, the property of one's body. But she presents her book as a response to a contemporary belief, specifically, the proposition that we are close to the technological capability to 'upload' human consciousness into a durable machine, shuffling off at last that mortal coil. Despite the circulation of this belief in a variety of pop-cultural settings (William Gibson wrote an X-Files plot based on it), Hayles finds this belief strange.
In order to explicate the grounds for such a belief, Hayles lays out the history of post-WWII cybernetics and information technology as a narrative of corporeal erasure. She accomplishes the difficult task of following her thread across diverse, historically contentious terrain by means of a carefully crafted and deliberate organizational structure. Hayles divides the development of "cybernetics, literature, and informatics" into three stages, each of which gets a critical exposition, a historical exposition, and a discussion dedicated to thematically related fiction. The book's introductory chapter presents these three stages in an information-rich timeline, reminiscent of texts in general humanities charting the progress of civilization through several interdisciplinary paradigms. In fact, Hayles's book would work very well as a combination course text and teacher's guide to, say, "The Culture of Information: Mind and Body by the End of the Millennium," no doubt a very popular course. Like a textbook, her work indexes a great number of primary and secondary texts in literature, critical theory, history, philosophy, and science, and organizes them along a thematic and developmental scheme. Whatever reservations scholars might have about the generalizing that invariably accompanies such a presentation, its coherence is so emphatically useful that it is well worth the dissent it might inspire concerning particular claims.
After opening with a summary of its argument and a historical overview of the relevant terrain, Hayles elaborates several facets of a paradigm shift that, she claims, facilitates the transition from the notion of the "human" to a multiply articulated and controversial "posthuman," of which the hybrid cyborg is both symbol and example. Underlying this paradigm shift is a transition from the presence/absence of the psychoanalytic dialectic, the Lacanian elaboration of Freud's fort/da, to an informatics dialectic of pattern/randomness. While the presence/absence dialectic stresses embodiment - the body (or its parts) constitute what is present or absent - pattern/randomness clearly moves up the abstraction ladder to an emphasis on configurations of bodies, not bodies themselves. Thus, information is to this degree indifferent to bodies. The governing anxiety in the new configuration becomes whether one is human or not, rather than whether one is male or female, and the posthuman becomes an updated analogue of castration anxiety. The paradigm shift intertwines with an economic shift: no longer is wealth a function of possession of discrete valued things (including one's body), but a function of access, facilitated by codes. Indeed, information, in its literal sense "informing" those who access it, is capital, from which subjects, in their ungrounded mobility, becomes less and less distinct. Hayles cites William Burroughs's description of dope-dealing in Naked Lunch as the practice of new capitalism: the "junk merchant does not sell his product to the consumer, he sells the consumer to his product. He does not improve and simplify his merchandise. He degrades and simplifies the client" (42).
Hayles begins the historical tale of disembodiment with the Macy Conferences on Cybernetics, a series of post-WWII interdisciplinary meetings joined by leading mathematicians, engineers, psychologists, and others associated with the emerging and yet to be defined field. Hayles's purpose is not a detailed historical account of the conference, as Steve Heims has already given us, so much as a selective examination of how assumptions, often rooted in disciplinary power, and more subtly in gender, disembody the concept of information, and how it came to center a "constellation" of concepts and research trajectories around homeostasis: those environmental interactions that keep an entity stable, drawing a skin around a kind of artificial body. Conceptually, homeostasis permits a body-erasing comparison between human and machine because it focuses on identity conservation rather than identity itself. Human and machine are "black-boxed," their specific structure erased, excluded from the equation, in order to measure and compare input/output relations across material contexts. On the model of the famous Turing Test, if a machine responds to input the same way that a human does, other differences, specifically differences in embodiment, fade into irrelevancies.
The conception of "information" itself, Hayles explains, constitutes a hugely important building block of the "black-boxing" of embodiment. The engineering thrust of Claude Shannon's and Warren Weaver's approach to the problem of defining information favored maximizing its independence from context: information is that which is conserved in its journey from a sender to a receiver, given the exclusion of noise. Since the quantity of noise depends on the channel in which this journey takes place (air, water, copper wire), as well as on the complexity of the language of the message (that is, the number of sign options), the concept of information had to possess a degree of abstraction that embraced both medium-as-language and medium-as-channel. Rather than a particularized, embodied dialectic of presence and absence, the conception of information builds on a pattern/randomness distinction, an abstraction and serialization of an assortment of bodies into disembodied ratios. It essentially black-boxes distinctions between code and natural language, permitting the convergence of input/output and pattern/randomness conceptions of behavior and information into a less and less refutable model of intelligence as discrete from embodiment. If entities behave the same, then they are the same. If they don't behave the same, then one is "noisier" than the other. As these abstractions converge, it becomes less and less possible to discuss other approaches to language, information, humans, and machines - although alternatives did exist. Hayles sets the advocates for decontextualization and disembodiment against a minority of dissenters, including Lawrence Kubie and Donald MacKay, who held out for the significance of "reflexivity" against homeostasis, that is, for the factor of the observer/actor as, to use a Deleuze and Guattarian anachronism, "desiring machines" that both create and subvert the input/output behavioristic modelling of information and its effects.
These assumptions raise their own frightening issues, which Hayles discusses in a separate chapter on Norbert Wiener, probably the most widely known, if not the most powerful, of the Macy participants. Wiener supported the basic assumptions of those who dominated the conference, but Hayles puts particular stress on Wiener's ambivalence towards the human/machine analogy. As a liberal humanist, locating humanity's defining characteristic in enlightened self-interest, in an agency and autonomy that rest on rational response to social and natural exigencies, Wiener found himself confronted as if in a mirror with the implications of chess-playing, military strategizing computers. In these contexts, where machines, like the liberal model of self-interested humans, calculated response based on probabilistic knowledge of the environment, Wiener nervously asserted what the driving machine/human analogy could not itself assert: the superior sophistication and flexibility of human response, and its consequent unpredictability to potential machine opponents. Wiener thus truncates the human/machine hybrid in the face of the possibility that the machine might develop plans for the human. This boundary enforcement, though, fails to interrogate the assumptions - the hygienic, body-observing assumptions - that create humans and machines as competitors in the first place.
The literary selection concluding this first cybernetic wave is probably unknown to many readers, and out of print: Bernard Wolfe's post-nuclear apocalypse dystopia, Limbo, written in 1952. Hayles points out its enactment of many of the issues raised by the Macy conferences, and particularly in Wiener's work, The Human Use of Human Beings. Wolfe sets the novel in a period of reaction to the supermachine that engineered WWIII by dominating human (male) agents - a phallic mother of a machine. Surviving humanity tries to amputate war-bent aggression by amputating human limbs; it places a high status value on voluntary quadriplegia. But just as psychoanalysts provided the most powerful theoretical source of dissent at the Macy conferences, psychoanalysis is a strong theme in Wolfe's work: amputation brings about a return of the repressed, reconstituting a class system, gender hierarchy, and physical competition. But even Wolfe, in Hayles's view, cannot control his material (and its crude sexist satire makes this striking): the Freudian principles the book seems to validate fall apart. Gender seems determined much more powerfully by projection than by observation, and language becomes the more reflexively trenchant prosthetic of all. Although complementarity - in Wolfe's terms, "the hyphen" - provides the principle of balance and deviation in the novel, hybridity - the "splice," the "circuit," according to Hayles - remains its repressed but seismically active double, unamputatable and posthuman.
And sets the stage for the next several chapters, investigating cybernetics's second wave. Hayles mentions Gregory and Mary Catherine Bateson as transitional figures from the Macy era, who continued to stress reflexivity and the situation of the observer against the dominant emphasis on cybernetics. One of Gregory Bateson's inspirations was a 1959 article on the operations of the frog's visual cortex, written with Humberto Maturana. Maturana, later collaborating with his student Francisco Varela, was to set up the narrative that would bring about a significant paradigm shift from the earlier homeostasis phase of cybernetics to a focus on reflexivity. Maturana's basic insight followed from the axiom that "everything said is said by an observer" - it is observers who constitute themselves and their objects, and reconstitute them with every sentient pulse, with every perception, every judgment. The theme of reflexivity, the emphasis on the observer, accompanies a complementary and complete epistemological constructionism: cognition is not the reception of information on the environment so much as the active constitution of both self and environment, termed autopoiesis. "Systems" - temporal, complex - are what carry out these operations, and now systems, as opposed to input/output boxes, constitute the model for the human, but with less anxiety about the posthuman per se. Nonetheless, the threat of subordination still looms large. Where in Wiener's world, the chief threat to the autonomy of the human arose from domination by an extraordinary chess-playing machine, that is, a competition in probability calculation and prediction, the context of Maturana and Varela re-accents this struggle as a conflict between autopoiesis and "allopoiesis" - where allopoiesis refers to the making of others in service of one's self-making. And given that all other selves are, from the point of view of the autopoietic organism, "environment," the threat is a significant one, arising from the basic nature of autopoietic system construction. Maturana and Varela's systems are "open" with respect to structure - they physically grow, change matter, form, and location, and so forth - but "closed" with respect to organization; hence the continuity of "self" across various structural states. Organization consists of the reinstantiation and reiteration of self-environment distinction. But this condition of the subject amounts to autism, as Hayles relates; "intersubjectivity" at best refers to a finely-tuned degree of structural coupling with the environment, accompanied by pleasure, perhaps. In any case, it does not refer to any actual diminution of the barricade between self and other. It also offers no special privilege to human association per se, other than the fact that the delicate calibrations of language form a significant part of the human environment, rendering it more complex than self-other species relations. The upshot of this theorization of reflexivity, in short, is what Hayles sees as the end of the disembodied liberal subject, and its replacement with embodied processes and the constructedness of self and world. Its upside is its return to the body; its downside is the extremity of subjective isolation.
Hayles's discussion of the novels of Philip K. Dick, particularly of his repeated concern with the trope of the "schizoid android," provides a literary reflection and deflection of the autistic solipsism corresponding to the dark side of autopoietic reflexivity. Schizophrenia and androidism connect in Dick as internal and external conditions of alienation, of the absence of empathy. While for the android this condition is located under a deceptive surface appearance, its reverse, the schizoid, undermines the android's difference from the human by betraying the possibility that androidism is in the eye of the beholder - the human beholder. An irrefutable realization that one's world is one's projection-in-rebound turns others into phantoms, or into stony objects of use or abuse, just as surely as if a barricade were erected by biological design. So schizoid and android mix and double each other.
Hayles's narrative here is particularly affecting in a way that draws the larger narrative of the book into question. Narrative acquires an undecidable relation to truth - stories deform their truths in the telling, multiplying possibility in excess of the probabilities that structure it. Hayles's exposition deftly interweaves biographical incident with novelistic plot, breaking down the border between criticism and its object and subject. As she recounts, the central and governing trauma of Dick's infancy consisted of the death of his twin sister Jane: Dick garnered the lions's share of his mother's inadequate milk supply, and his sister failed to thrive. Hayles sketches this incident without the detail required of formal biography, but with the effect of rendering Dick as much a character of her plot - and of his many novelistic plots - as Jane's is in his life story. Remembering a doubtful tale from his mother's lips, pursued by a female ghost of himself, saddled with guilt but deprived of both agency and access to certainty, Dick is left equivocally at the crossroads of feeling and fact, psyche and soma. In his novels, as Hayles relates, the result of a progressive overburdening of the self with responsibility for the world - exemplified in one instance by the sense that one is psychokinetically omnipotent, a solipsistic world with no outside - precipitates collapse into tomb-world, a world of eons of immobility, the end result of entropic decay brought on by the impossibility of novelty that accompanies perfect control. Gradually, tomb world picks up, slowly evolves back into an outside that is no longer the pure product of an inside: the healing of a schizophrenic episode.
And one begins to suspect that How We Became Posthuman operates like a Dick novel, climbing out of tomb-world into Disney World. The immense burden of reflexivity in Dick gives way, in Hayles's own narrative, to life from the outside once again, fake life. She begins the last phase of the book with a brief critical-theoretical review that sets the body back into place, setting the stage for its simulation. Hayles sees Foucault's subordination of the body to discursive effects of power as dangerously complicit with the general trend of the discourse of disembodiment encountered in cybernetics. In contrast to Foucault, she praises Elaine Scarry and Elizabeth Grosz, who attend specifically to the body's instantiated and individuated friction with normative or coercive forces. Hayles sets up the interplay between ideology and material embodiment on two axes - on the one hand, as an exchange between normative concepts of the body and actual embodied states, and on the other hand, the portable, easily transmissible technologies of inscription relating that body to its environment in friction with less portable, less conscious forces of body habituation to the environment. As the seat of articulation of each of these axes, "embodiment" always alters the abstract trajectories projected onto the body and into inscription: "Formed by technology at the same time that it creates technology," Hayles writes, "embodiment mediates between technology and discourse by creating new experiential frameworks that serve as boundary markers for the creation of corresponding discursive systems" (205). With respect to language in particular, Hayles refers to Garrett Stewart's work on subvocalization - the bodily involvement with silent reading, a site of simultaneous and intertwined self-and-other narration. In counterpoint, Hayles brings in a discussion of William Burroughs's The Ticket That Exploded, a protest against parasitization by "the word" as a kind of social virus attacking the embodied individual simultaneously from without and within. Both Burroughs's work and the theoretical work on the gap between embodiment and inscription insist on the existence of a kind of friction, "noise," randomness, that makes the word's viral penetration and complicity with subjectivity incomplete and unpredictable. This noise, Hayles stresses, is the force of mutation: random, unprogrammed, and unaccountable alterations in transmission.
And mutation, together with the possibility of its complicity in an evolutionary response, provides the basis for the new science of "artificial life." In artificial life, the uncompromising quality of autopoietic organizational closure is overcome - at least partly as the result of a theoretical shift in point of view. Reflexivity now gives way to a concept of emergent behavior. Accompanying this shift is the resurrection of analogy, with the same sleight-of-hand effect of the Turing Test's human and machine: artificial life sets the terms for defining life-in-general, in this instance as a capacity to respond robustly to environmental change - that is, to evolve. But while no window exists from which to view most organic evolution, computer simulation easily frames the scene for artificial life. With greater intensity than first-wave cybernetics's rebound on the human via models like the neural net, artificial life plots the progress of an entire "species" of cyber-organisms - and calling them "organisms" or "creatures" is indeed part of the defining "constellation" of artificial life, given that they are, in material actuality, code. Researchers at the Santa Fe Institute assembled a program of codes to replicate themselves in the environment of a computer, codes with a built-in mutation element. The codes both altered and replicated at a spookily unanticipated rate, producing in the lab what cyberpunk produces in novels: a sense of life "out there" in cyberspace, beyond the control of its human inventors.
Given this general scenario, speculation on the relation of organic life to putatively silicon-based life splits in two directions. One, represented by Hans Moravec and his book Mind Children, sees the potential of computers as much more oriented to artificial intelligence - to AI, rather than to the more general AL. Hayles does little with Moravec's work other than use it as the most extreme expression of belief in the possible obsolescence of the body: Moravec is the main adherent to the belief that consciousness can be uploaded into a more durable, silicon-based housing. AL research, on the other hand, builds and observes "creatures" that have no direct relation to the human per se, but evolve as their own robotic species. Rodney Brooks at MIT is the representative of this group; Brooks's cockroach 'bots have a decentralized bottom-up design that maximizes openness to the environment, and minimizes the initial computational/organizational overhead - allowing the latter, essentially, to emerge from the former. In short, while for Moravec consciousness is information is the human, for Brooks, consciousness is epiphenomenal. However, since computational capacity does seem to emerge from the bottom-up, since something that seems like consciousness does seem to be a product of self-organizing, it has the potential for fueling narratives as misleading and risky as Moravec's. AL as well as AI is full of assumptions, analogies, and nomenclature that sets up "life" in the image of a particular kind of machine. And it is here, Hayles believes, that we now stand.
It is fitting that the last substantive chapter of a powerfully organized work present a constellation of novels by means of a structuralist device. Hayles finishes her exposition with a discussion of Greg Bear's Blood Music, Cole Perriman's Terminal Games, Richard Powers's Galatea 2.2, and Neal Stephenson's Snow Crash along the axes of a semiotic square. The payoff of this section consists primarily in its dramatization of the square's capacity for revealing how rigid oppositions give way to accommodations, gradations, alliances, and mutations - in sum, the movement of the posthuman exterior across the human hermeneutic interior. In Hayles's framework, each novel treats an aspect of and a response to borders between human and non-human - borders potentially permeable, with various implications of threat and opportunity. Yet the chapter concludes with "surprise" at each novel's continuing commitment to the liberal subject. But without more information on her criteria for selection, such surprise is difficult to understand. All of her selections are masculine, straightforward, and unexperimental narratives in a general field full of experimentation. Hayles invites the reader "to construct significance out of ruptures, juxtapositions, and implied links," in analogy with "hypertext lexias" - so why not run the experiment using hypertext narrative? (Much less, poetry?)
A hypothesis emerges: perhaps narrative itself, perhaps particularly novels, those narratives that yield surprises to the dissective ardor of the semiotic square, are a humanist form. Perhaps hypertext narrative isn't narrative at all, but a computer game - implying, in turn, that these games are in some way antithetical to narrative. I would relate this begged question in the final section, moreover, to another slippage that appears at various points in the text: the vulnerability of the concept of evolution to point of view - and thus its tendency to slip towards synonymity with progress, and, in turn, narrative. Evolution shares with hypertext "literature," literature whose form depends in a physical way on reader choices, an element of randomness that resists the linear satisfactions of the story. Hayles discusses evolution at greatest length (still, a brief discussion) in her account of autopoiesis. For reasons that are still not quite clear to me from her argument, she connects autopoietic organizational closure to an inadequacy in Maturana's attempt to account for evolutionary change, an inadequacy that seems based on its geometry: "The tension between evolutionary lines of descent and autopoietic circularity becomes apparent in the authors' claim that autopoiesis is conserved at every point as organisms evolve." But this seems to me an issue in which phenomenological epistemology and ontology diverge utterly. The linearity of evolutionary "descent" is a form given to us post hoc, after selection has quieted the circular pulse in hordes of autopoietic competitors. Am I evolving? Am I evolved with respect to other species? I cannot refer to experience for the answer to such questions. But I can deduce that there are things beyond my experience - and necessarily beyond my experience (like the fact that time goes by when I sleep). Change seems to be one of these things. I believe that I change: maybe this equals I believe that I. (Period) It's not a matter of fact, this "I." The liberal subject seems to be this subject, goaded into continuity by a variety of unphenomenological convictions. I experience myself as a "liberal subject" because only liberal subjectivity turns "experiencing myself" into a line, a narratable notion. When the phrase "experiencing myself" sounds archaic, misguided, irrelevant, when I am less the subject of a story than the singer of a song, a singer who only knows the next note when the train of uttered notes reaches its proximity, when time's arrow slips by like paper tape - THEN - I find that my linear and mislabelled "experience" is better regarded as run-off, possibly an ecohazard. What is narrative in the age of hybrid, networked, informatically penetrated authorship?
But still, I like a good story.