We cannot agree on an historical point of origin for the Internet. (Bletchley Park? The telegraph? The diorama? The abacus? The Atlantic Cable? Painting? Writing?) Its techniques and tools are still in the process of development, perhaps even in their infancy. Internet culture is heterogeneous and dynamic. Its economy is not stable, seeming sometimes as fantastic and illusory as the Internet itself. Its status as global tool or tool of globalization is still unclear. Most importantly, even the object of study, and so the appropriate methodologies for study, are unclear. Like other nascent forms of representation before it, the Internet in its infancy presents itself as -- and may actually be -- the site of cultural, political, and ideological contestation. Or it may not: the contest may in fact have ended before it began, in which case scholars interested in such things can, like Lawrence Lessig, write only about who won and who lost. The grandest claim one might plausibly make is that the Internet at the present moment is the material actualization of the post-structural indeterminacy that characterizes post-Nixon/Mao/Gandhi representation and cultural theory, from the post-1949 Middle East, to the films of Peter Greenaway, to deconstruction, to White Noise. However, it behooves the critic to find a sector of critical theory through which some of these assertions might be more clearly elaborated.

The cultural-critical "app" I choose for discussing this poststructural indeterminacy is Michel Foucault's notion of panopticism: first, because it is one of the more straightforward poststructuralist notions; second, because it is not just an important and conventional touchstone within the community of poststructural critics, but has also made its way into popular discourse; third, because it already has a familiar application within Internet studies; and fourth, and most importantly, because the Internet and the panopticon make significantly similar assumptions about the creation of the subject within discourse. Both panopticism and the Internet construct space with a special attention to the subject's internalizing a particular model of space, and a particular notion of how people are distributed throughout space in relation to one another, and with a special attention to the defining of the individual through the space she occupies. Further, both are intensely interested in the construction and distribution of authority over and within the subject. So I will forward a limited thesis connecting the Internet to this corner of poststructuralism: in literalizing of Foucault's panopticon, the Internet makes us question the notion of and, perhaps, redefine panopticism. This redefinition in turn allows us to ask questions about the nature of Internet representation. We may ask not only whether the Internet signifies panopticism, but whether it redefines "construction" and "signification" themselves.

Panopticism 1.0

In fact, panopticism seems an appropriate poststructuralist model for this instance because the Internet has been tentatively read through the lens of panopticism before, for example in Communications Studies. It has something of the same status as a reflection on the coming of new technologies as Martin Heidegger's "Question Concerning Technology" and Walter Benjamin's "Work of Art in the Age of Mechanical Reproduction." However, though Foucault and panopticism have a hefty presence as the objects of critical study on the Net (274,000 Google results for "Foucault," and 17,100 for "panopticon," including a trading card site featuring Foucault and several other poststructuralists),[1] his actual utility as a theorist of the Net is limited in scope to studies of visible surveillance on the Net: in other words, to studies of such phenomena as information-gathering about individuals (e.g.: Carnivore software), attempts to evade such cataloguing as data encryption, or, more generally, late capitalism itself.[2] Other special properties of panopticism -- particularly spatialization, totality of experience, coercive discourse, and ambiguous/internalized authority -- are not frequently linked to the Internet. Neo-Foucauldian cultural critics understand surveillance society as a top-down phenomenon in which an otherwise scarcely visible oligarchy utilizes new technology as a tool of social surveillance.[3] They understand panoptic society to be a sort of Orwellian 1984 or Kafkaesque Castle in which power is invested in the powerful if invisible. The aim, then, of the neo-Foucauldian critic is to bring the powerful to light by revealing how she uses the technology for surveillance purposes. Such critics are interested in the way that society constitutes its constituent members as either prisoners or jailers, not the way that society itself is in fact a prison house in which surveillance is distributed in a manner that makes us our own prisoners. For example, though the most exhaustive work on panopticism and new technology -- David Lyon's The Electronic Eye: The Rise of Surveillance Society [4] -- contains a lengthy discussion of the way in which panopticism is defined by "uncertainty as a means of subordination" (in other words by how the authoritarian gaze is unverifiable), his discussion of panopticism per se is largely concerned with the various data-collecting agencies that use the Internet to exert an external coercion on the individual, not with how such authority is internalized: "The prison-like society, where invisible observers track our digital footprints, does indeed seem panoptic." [5] A little less often, scholars are interested in the ways that the Net limits our ability to think outside the Net, in other words in questions about discourse and discipline.[6]

The question of whether corporations and governments exercise power over the individual by collecting data about her are probably far more politically useful than the questions raised in the present essay. Nevertheless, I shall eschew the traditional discussion of actual agencies of control in favor of a discussion of those other conditions that define panopticism. Is the Internet surveillant? Without question. But is the Internet surveillant after the manner of the panopticon? We cannot answer this question by means of sociological accounts that are simply interested in the government and corporate tendency to get to know us better through Internet spying, The panopticon does not use information just to know us; it also deploys information to create us, to constitute us as compliant workers and consumers. Essentially, if it is panoptic, the Internet must serve the same panoptic/enlightenment function of social control through a physical control of the body in space and a rhetorical control of the definition of subjectivity that other panoptic institutions do. This larger question may be parsed: is the Internet an "institution"? For example, does the Net have a discourse that determines its own boundaries of action? Is Net surveillance coercive? Does the subject in front of the screen internalize this coercion? If coercive, does the Net establish a diffused authority? Does the Internet encourage a pro-social uniformity in its citizens? How does the Internet define the relationship between spatialization and discourse? Is the Net a discipline in both senses: a body of knowledge and a form of coercion, the knowledge being the vehicle of coercion? What other ways of thinking about bodies in space does virtuality either interrupt or promote?

Foucault's Panopticon

If we wish to know whether the Internet is panoptic rather than simply surveillant, we must briefly revisit Foucault's own writing in Discipline and Punish.[7] This synopsis of the plot of Foucault's "Panopticism" chapter is not exhaustive; rather, it is blazon-like, meant to itemize and summarize those dynamics I shall examine later.

For Foucault, history is dynamic, and authority and coercion define the subject in different ways at different times. Before the enlightenment, effective punishment of the criminal was visible punishment: hangings and decapitations were made visible to the masses as a spectacle.[8] But the rise in European population -- and its geographic mobility -- in the seventeenth and eighteenth centuries made feudal and monarchical models of social control outmoded. For the beginning of the enlightenment, the practice of authority is defined as procuring for a small number, or even for a single individual, the instantaneous view of a great multitude" (D&P, 216). Spectacle is replaced by surveillance within the prison-panopticon. Unlike the medieval object of punishment, the body of the prisoner of the panopticon is not tortured; it is simply separated -- and thus alienated -- from other prisoners, and watched. This attention to the body of the prisoner is total: it implies an interest in and effect on all the movements of the prisoner. Not knowing whether or when they are on view, prisoners ultimately internalize the notion of a surveyor. This prison-panopticon serves as Foucault's central metaphor for the rise of a society in which all institutions are disciplinary in both senses of the term. First, they represent a body of knowledge. Second, this disciplinary knowledge is always coercive, enforcing discipline -- particular modes of behavior and belief -- on the individual. Discourse plays an important role in this social coercion. Disciplinary knowledge is articulated through its own proprietary language, generally accessible only to adepts. And discourse is coercive in the sense that it is impossible to define or to think anything without the languages of the limited number of discourses available through a large but finite number of social institutions. Discourses define subjectivity. Finally, because the notion of a central surveyor is a fiction, and so authority is internalized by each individual surveying herself, panoptic society works by employing institutions to distribute power throughout society. This is not to say that power is democratically or equally distributed, or distributed for the purpose of democratizing society. Such distribution is simply the most efficient way to maintain a stably quiescent and productive society, by making certain that everyone can potentially be surveyor or surveyed. In brief, Foucault thinks that the underside of the enlightenment that we inherited from the seventeenth and eighteenth centuries is the desire to control increasingly large numbers of people in a manner that monarchies were not able to accomplish: both by distributing self-regulating bodies regularly through space and, as a consequence, by having people police themselves because they believe they are being surveyed.

Bodies in Space: Comparison to the Gaze-as-Authority in Film and Television

The panoptic gaze is -- at least initially -- unidirectional and fictive: while surveying the prisoners, the (implied) guard is herself invisible, the force of her fictional surveillance enough to keep the bodies of the prisoners evenly distributed and quiescent. Can the Internet be presumed to be phallic in this fashion: simultaneously powerful and nonexistent? (Foucault himself is silent on this point in D&P.) I think that the answer depends on how one views the apparatus that connects one to the Internet: the monitor.

Like those of many film and media scholars interested in questions of ideology, my own gaze is drawn toward the coercive qualities of the screen, now the computer monitor.[9] The monitor is that part of the larger Internet apparatus that most immediately reminds one of now more traditional visual entertainment/information media. Governmental and corporate surveillance aside, I wonder whether, like other media, the monitor through which we view the world is always monitoring us.

American television and film are easily identifiable as panoptic institutions. They are disciplinary in the sense that they provide their own defining discourses, mainly variations on the themes of entertainment, desire, and consumerism. (Genre rules, for example, constitute a discourse.) Second, their promise of desire fulfilled keeps their audiences still and attentive.[10] Both film and television attempt to be total experiences, not only at the moment of spectatorship, but in their peripheral phenomena: their omnipresent paratexts, their connection to other forms of coercion (the "military/industrial/entertainment complex"). However, film seems more spatially panoptic than television. Like students in the classroom, patients in the hospital, and prisoners in the penal colony, film viewers are equally distributed in the space of the theater in a manner that gives each person more or less equal access to the film screen, reciprocally giving the screen equal access to each viewer. Because film takes place in the dark it is most often a monadic experience: each spectator is an island unto herself. However, like the central prison tower, the central object of attention (the screen itself) is well lit. And, as with the prison tower, we are to keep our attention riveted to this central structure; stillness is enforced. Finally, though we believe we have chosen to go to the movies in a way prisoners do not choose prison, we are metaphorically imprisoned both in the sense that our culture still gives us precious few authentically practic options, and in the sense that, like prison, movies are instructive. Films give us imagoes with which we identify: models for culturally acceptable or desirable modes of thought and behavior.

Television does not seem spatially panoptic in the manner of cinema. The experience of television is less centralized; hundreds of people are not equally distributed throughout within a partitioned, grid-like space. However, citizens of television are distributed throughout the same living room throughout at least USA culture. (When the television monitor is in the bedroom or the gymnasium, it simply turns these spaces into living rooms.) More importantly, solitary viewing does reproduce the experience of the monadic prisoner in her cell. And while, like film-as-entertainment, television provides us with imagoes for emulation, television is if anything even more instructive -- more perceptibly coercive -- than the movies.

Ironically, however, because their gaze is bi-directional, film and television constitute a twentieth-century variation on an eighteenth-century theme; these constructions of space and of the gaze are cryptically panoptic. By cryptic I mean that the prisoner's relationship to the tower is ironically slightly less deceptive than our relationship to the screen; the illusion of presence provided by the panoptic tower is doubled by the media illusion that surveillance is nonexistent. Prisoners understand themselves to be under direct surveillance. But, while we believe ourselves to be watching television and film, these media are watching us along those axes by which we are allowed social definition: our viewing habits and so (presumptively) our desires, through Nielsen ratings, advertising sales, bottom lines, pre-emptive censorship, and so on. While, during the experience of watching, we believe the gaze to originate from the spectator and onto the screen, in fact the gaze is relayed from the screen/tower to the spectator in a way that coerces her to internalize consciously and unconsciously the lessons of the screen. This, at least, is the assumption that advertisers take on faith.

Television and film (as well as Foucauldian theory itself), then, give us the following five dimensions along which to think about whether mechanical forms of visual representation like the Internet are or are not panoptic: (1) the gaze, (2) space, (3) authority, (4) totality, and (5) discourse. Is the gaze of the Internet (cryptically) unidirectional; in other words, does the gaze operate consciously, or does it convince us that we are watching it while it is actually watching us? How does the Internet define space? In other words, where is the Internet citizen when she is watching the screen, and where is she in the screen and in relation to others similarly engaged? How does the Internet invest and then deploy authority? To what degree is the Internet a "total" experience? Can the Internet be said to be uttered in a unified discourse and, if so, what is the nature of that discourse? Does this discourse institute desire and, if so, is this desire self-regulating, external, and productive of a tendency toward social uniformity?

The Gaze.

The Internet is both more and less panoptic than television or movies, and for the same reason: its gaze is a special version of those media's bi-directional gaze. While, like the dream screen of film or television, the computer monitor observes us as we observe it, the bi-directionality of the monitor seems genuinely reciprocal in a manner to which neither television nor movies can aspire. Unlike these media, the Internet provides us the opportunity to create choices when we create web pages, or even talk back to the screen through bulletin boards, email, chat rooms, online games, and other such venues. Because we have traded the remote control for mouse and keyboard, the Internet gaze seems overtly rather than cryptically bi-directional. In short, the Internet is at present the best version of what Roland Barthes calls a "writerly" text, a text whose meaning can in some measure be constituted by the reader. This at any rate was the assessment of such foundational 1990s Internet theorists as George P. Landow.[11] Utopic critics still characterize the Internet as the very opposite of the panopticon; it is more consciously "empowering" than other media because, unlike movies or television, it allows a greater range of behaviors, a greater number of choices, and a certain ability for creative self-expression that are not possible in other electronic media.

But does this bi-directionality really imply consciousness and mass empowerment, or is the keyboard simply a glorified remote control? Certainly, it is the remote, at the level of simple consumerism. As with television and cinema, the Internet gives us false, essentially consumerist options: ABC or CNN becomes eBay or Amazon.com, AOL or MSN. We are encouraged to think that using Netscape strikes a blow at the insidiously secretive Microsoft because it is an "open-source" application, though the equally monolithic AOL/Time Warner now owns Netscape. Two Internet devices important to most conversations about freedom and creativity are perhaps even more illustrative of the ambiguously determinative quality of the Internet: hypertext (or hypermedia), and the avatar. In the early and mid-1990s, hypertext was touted as the cure for the inactivity enforced by other forms of representation, as if hypertext could save other forms of representation from ennui and entropy.[12] Hypertext gives the web surfer a sense of choice, a sense of the Internet as an exciting experiment in bricolage. But, though it is still an interesting and innovative technique within many websites, hypertext seems across the Net to have a limited number of uses, almost all of them (with the exception of hyperlinks that take the user from one site to sites of similar interest) again simply commercial: banner ads connecting to retail, auction, and pornography sites. Further, early theorists of hypertext were unaware that the "pop-up" advertisement (this "anti-choice" hyperlink did not appear until 1997) would be so omnipresent.[13] Even the apparently non-commercial use of hypertext to get from search engines to the information they offer scarcely conceals the fact that many search engine companies are paid to find commercial hits first. Many scholars shy away from the Internet when they discover that most search engine algorithms favor commercial over non-commercial sites. The notion of freedom to "surf" is at least in part undercut by the gentle corporate appropriation of hypertext.[14]

With the exponential growth of sophistication and interest in visual representation and online gaming, the second, newer route to apparent self-determination has been the avatar, the icon the Internet citizen can make -- or more usually buy and download -- that may represent the citizen in online games or in more sophisticated visually oriented chat sites like The Palace and ActiveWorlds. The avatar seemed for awhile the very apotheosis of the American dream, and the antithesis of the panopticon: it embodied the fantasy that we can refashion our selves in any manner we wish. We can't be under surveillance if no one knows who we are: what we look like, what color we are, what religion and ethnicity, what sex. The actual deployment of the avatar on the Net, however, has been mixed. While the characters played in online games are thriving, the visual chat rooms in which participants could engage in some serious personality changes have not fared so well. The Palace is now defunct, and ActiveWorlds now charges for what was originally a free universe. The price of freedom is not eternal vigilance, but, at the time of this writing, about $9.99/month. (Online gaming is even more expensive; game avatars have sold on eBay for four figures.) Further compromising the notion that the avatar means freedom is -- finally -- the fact that the Internet is surveillant, not in the sense of the government's collecting data on its citizens, but in the much more permeative sense that corporations need to verify our existence and status as stable, desiring consumers. The avatar construct is met at every turn by credit card verification. Online gaming is a sort of metaphor for the antagonisms between identity and commerce: I cannot be Palomar the Invisible without swearing fealty to PayPal the Verifiable.

Further, while both utopic and panoptic critics of the Net describe the Net as visible spectacle, neither takes into account the Net's hardware -- its architecture -- as a determinant of the gaze. Remember that, for Foucault, Bentham's panopticon ultimately determines the design of schools, factories, hospitals, and other kinds of public architecture. While utopic accounts of the Internet, for example, oppose the notion of the Internet as significantly surveillant because it is not associated with a visible architecture, the simplified diagram for relationship between surfer and Internet or intranet looks suspiciously like the schematics for Bentham's prison.

Whatever the direction of the gaze, it is located in a hardware space defined as center and periphery. The Figure (1) illustration even contains a central tower radiating light. For those interested in surveillance phenomena like Carnivore, the server is the locus of authority; the server master and webmaster can and do censor our access and input, encouraging Net-citizen uniformity. In short, the flowchart for Internet hardware is panoptic. Even insofar as we are not describing a physical architecture but rather the flow of information, we are still simply indexing the move from a capitalism invested in things to a late capitalism for which ideas are the principal trope for the buyable.[15]

However, even if the Internet is not really utopic but rather a tool for late capitalist appropriation, we must still account for the fact that, unlike film and television as entertainment, the Internet cannot yet constitute all spectators as one spectator, or even a small set of spectators. Entertainment seeks to create unconscious and monolithic audiences, but, because the Internet is a modality as much as it is a medium in the sense that it provides different kinds of venues for expression -- from web sites to MUDs to chat rooms to BBSs to newsgroups to client servers -- entertainment and consumerism, though a large presence on the Internet, constitute a small percentage of its representational capabilities. As a consequence, the Internet enables different and simultaneous possibilities for awareness, including but not limited to the consciousness/unconsciousness of media representation. Expanding on the habitually untheorized notion of the surfer, the following list is not exhaustive; rather, it suggests the heterogeneity of surfers' stances in the same way that the educable sons of the Passover Seder suggest all children without defining them:

1. The surveyor. This gaze is external, overt, and directed toward the surfing subject, who is conscious of being watched. This is the surveillant gaze of Carnivore, employer surveillance of employee email, and so on. The object of desire is constituted as the satisfaction of the regulations of that gaze. In short, the Internet is what critics have designated as Panopticon Classic.
2. The moviegoer. This gaze is external but cryptic toward the surfing spectator, whose desire is constituted externally and so is unsatisfiable except through conformist/consumerist behavior. In short, as consumer entertainment, the Net is cryptically panoptic. The Internet is a movie or a shopping mall, and the subject is the traditional spectator/shopper.
3. The gamer. Originating from the surfing self, this gaze is outward, into the Net. The object of desire is external. In making us active participants, the Internet makes us critics: we are aware of the conditions of the Net in a way we are not of a movie. We achieve total immersion, but not total unconsciousness. Our desire is directed toward non-consumer sites, and much of our time is spent trying to circumvent capitalism. The same subject who is unconsciously influenced by a film's product endorsements will still avoid banner ads, spam, and pornography.
4. The writer. The gaze is inward; the desire is for self. Perhaps, in co-constructing the Internet, the subject recreates desire differently, for example reconstituting desire less as a question of subject and object, and more as a function of the subject refashioning her own subjectivity.
5. The lurker. Originating in the surfing spectator, this cryptic gaze is specifically voyeuristic; it is the desire to see without being seen, the individual's desire to appropriate some of the authority of social surveillance through imitation of it. As such, this gaze shares with Laura Mulvey's notion of the masculine gaze the belief that the gaze can be sadistic and exploitative, resides in the spectator, and reflects the deployment of surveillance-as-power in society at large.[16]
6. The spy. A "proactive" lurker. The user of applets, spyware, etc.

These gazes cannot all be said to be subsumed to an overarching capitalist gaze simply because they exist in a medium created by and permeated by consumerism whose hardware relegates some to the center and others to the periphery. The consumerist tendency is panoptic, but the working out of the gaze is as yet too heterogeneous to be considered appropriated.

Space.

Perhaps because both gaze and space are genetically linked to spectacle -- and so to the heterogeneity of Internet spectacle -- in a way that, say, the notion of discourse is not, Internet spatiality, like the Internet gaze, is and is not panoptic, and for the same reasons: spatial heterogeneity militates against panopticism, while the need for verifiability tends to result in a spatially uniform Internet citizenry. On the one hand, the experience of surfing the Net seems spatially heterogeneous. The Internet contains an enormous variety of spaces, limited only by the web designer's ability to re-envision two- and three-dimensionality. Seeing a film in a theater, people sit evenly spaced in semi-circular fashion around a single image, panopticon-style; as Jean-Louis Baudry notes, the scene of filmgoing reproduces Plato's allegory of the enslaving cave which is itself, we might note, panoptic (if anachronistically so) in the sense that it provides an illusory world in order to enforce immobility in its viewers.[17] Internet citizens, in contrast, log on in different places at different times; the Internet often seems mood-dependent. If she has an Internet connection at home and work, or also has access to a cyber-café, or has wi-fi, the surfer is not even limited to surfing from a single space.[18]

However, like television, the Internet also renders all spaces as one space -- the monitor-user space to which most users (at least those without an Ethernet connection) have spatial access in the same way: through the almost cabalistic invocation of that ethereal nomen which must never be spoken: username and password. For the simple surfer, the Net is also like television when, as a consumer, she is watching the same dozen sites as everyone else. The Net seems a little more homogeneous when one considers that the expense of logging on limits participation to the members of just a few classes, and significant development to just a few nations. And if we consider Internet technology as a determinant of meaning rather than simply its vehicle, such technology -- the interlaced order of pixels by which a monitor renders objects visible to the orderly and hierarchicalized distribution of surfers, servers and modes along a fiber optic grid -- suggests the even distribution through space reminiscent of panopticism. If the critic is looking over the shoulder of the user, she may decide that his surfing is a liberating experience. Observing, however, from the server site or the fiber optic engineering room at Qwest, she may see him as a point on the grid.

And, while the Internet is a more temporally heterogeneous experience than the experiences of going to the movies and watching television, it is still panoptic to a much greater degree in the sense that it is genuinely monadic in a way that these media may only approximate. Whatever actual communication the Internet fosters, the physical experience of surfing the Net is almost always absolutely monadic: we are alone, in a physical space differentiated from all others. For Foucault this partitioning is crucial for defusing and diffusing social and political opposition: "Avoid distribution in groups; break up collective dispositions; analyze confused, massive, or transient pluralities. Disciplinary space tends to be divided into as many sections as there are bodies or elements to be distributed" (D&P, 143). The Internet is not panoptic in the sense of evenly distributing bodies on a grid · la the assembly line or the military formation (two of Foucault's favorite examples). We don't all surf at once, or for the same number of hours a day. And Internet use is still not as universal as television spectatorship. But, while the movie and even the television come to an end for us, we are "structured to feel" (to paraphrase Raymond Williams) that the Internet is temporally endless, that the conversations, games, negotiations, and exchanges happen whether we are awake or not.[19] Television and movies are packaged in a manner that suggests peaks and troughs in viewers' interest. At least this is the assumption behind "prime time" and "bargain matinees." But the Internet has no prime time and no bargain matinees, both of which take note of the fact that movies and television have peak hours of popularity. Or, to say it correctly, though the Internet has peak bandwidth-use hours, it has no peak hours of representation and few hours of participation that are commercially and socially recognized as such. Internet traffic actually does conform to measurable patterns of peak use, but chat rooms, MUDs, role-playing games, music-exchange networks, commercial sales sites, auction sites, and pornography sites take almost no notice of these spikes in bandwidth use. While games like Everquest hold special events during peak hours of play, it is still the case that online gamers can and often do remain in the game for days at a time. Like panoptic society, the Internet gives the impression of being eternal and/or eternal: all times are one uniform time.

Authority.

While the Internet construed as gaze and space is at best only ambiguously panoptic, the notions of authority and totality suggest that the Internet is capable of a perfect panopticism in a way that perhaps no other social institution or form of representation is. Surveillance on the Internet is panoptically (if mischievously) democratized, and the Internet is capable of being the total experience to which other media can only aspire.

To begin with an earlier point: in opposition to the actual top-down surveillance of corporations and the government, the gaze of authority in film and media is not visibly surveillant after the manner of the panoptic tower. Its coercion is cryptic and unconscious: as consumers of entertainment we are not usually aware of being under surveillance. Bentham's originary panopticon is characterized through visible surveillance by a guard, who, all the same, is not necessarily there; he is a fiction of authority displaced by self-regulation and regulation by one's neighbors. (Slavoj Zizek refers to this fiction of authority as "radical uncertainty."[20] ) This fictionality has three important consequences: our relationship to authority and our own identities, based on a fiction, are also fictional; the citizen is uniformly self-policing; and everyone has a little power to police the state.

Does the Internet similarly institute an internal agency that ensures a vision of authority and society in which each person is her own and her neighbor's monitor? Certainly software exists that encourages the tendency toward self- and peer-surveillance, and among other blandishments, such software is advertised as security against corporate and governmental surveillance.[21] Anti-virus software, spyware, anti-spyware, anti-pornography software, firewalls, Trojans, anti-Trojan programs, worms, data-erasure programs, and other forms of self-surveillance -- software more or less readily available to all Internet citizens -- can infiltrate other computers or monitor the penetrations into one's own computer; it is possible to locate the source of the attack, thus monitoring the activities and strategies of individuals and corporations. While some of this software seems to constitute a resistance to panopticism by maintaining individual privacy, it reflects the degree of self-consciousness we have developed about our Internet behavior. Derived from the belief that we may be under surveillance at any given time (and so suggesting the present/absent guard in the tower), such software keeps everyone in the game, distributing a little bit of power and authority to every player who owns a modem. This distribution of power -- not the actual top-down intimidation of Carnivore -- is panoptic. As Foucault asserts, rather typically describing the panopticon itself as a machine,

Any individual, taken almost at random, can operate the machine... The more numerous those anonymous and temporary observers are, the greater the risk for the inmate of being surprised and the greater [his] anxious awareness of being observed. The Panopticon is a marvelous machine which, whatever use one may wish to put it to, produces homogeneous effects of power. (D&P, 202)

Foucault speaks of a web of power relations in which absolute authority is invested, not in an oligarchy made up of -- depending on your conspiracy of choice -- media owners, the NSA/CIA/FBI, owners of multinational corporations, heads of state, or Jews, but in everyone who has internalized that prison guard. Which is to say everyone on the Net. Anyone can install Net Nanny, AdAware, Norton Firewall, or other "user-friendly" applications. Everyone has a little power, though power is shared unequally oppressively.[22]

Totality

The total distribution of power along the Net grid reminds us of the notion of panoptic totality; not only is the distribution of power among Net citizenry absolute, the experience of being subject to authority is complete as well.[23] Television and film aspire and fail to be total experiences in and of themselves; the impossibility of avoiding the paratexts that complete their cultural significance indicates both the desire and the failure. Much more than film and television, the Internet has been endlessly discussed since before its existence as a portal to one kind of total experience that still eludes us, but that is not the subject of this paper: virtual reality.[24] However, we may look closer to home -- or the workplace -- to find another kind of total experience in the making, for which the virtual worlds of Neuromancer or The Matrix (1999) are at present simply metaphors. Conceived both as more than simply surveillance but still less than a total experience, the Internet as a technology is itself gradually being assimilated into at least two bits of new-technology hardware: the personal digital assistant (PDA), and the cell phone. Even the name of the latter device suggests Foucault's notion that each citizen inhabits her/his own prison cell. Both devices are quickly assimilating other communications and representational devices as well: email, the pager, the digital camera, the video camera, the tape recorder, global positioning systems, and the video game. When the problem of screen size is solved (perhaps by the addition of video glasses, making all individual movement an extension of surfing the Net as driving has become an extension of the cell phone), William Gibson's cyberspace as the experience of total immersion will be more than realized. The PDA can now download all forms of representation, from poetry to pornography. The PDA operative -- probably affluent, certainly computer literate -- is now immersed, not just in the web of the Internet, but in every sector of communications. No paratext is necessary -- or indeed possible -- to explain the experience of the cell device, because the device has access to all texts that refer to it.

The notion of total immersion echoes an often-cited, much discredited "humanist" bit of film theory, André Bazin's "The Myth of Total Cinema," the most important assertion of which is that cinema always tries to convey an ever more realistic sense of the real.[25] Total cinema equals the total immersion in the experience unfolding onscreen. Writing in the 1950s, Bazin implicitly assumes that cinema's ever-increasing ability to present us with a window on the actual is nevertheless asymptotic: representation may go thus far and no farther. Hence, the notion that cinema can be "total" is a "myth." But, in combination with other new technologies, the Internet promises to make that asymptotic line tumesce just enough to touch its own limit.

This tumescence determines the erasure of another asymptotic separation, between the "virtual" and the "real." That is to say the ideological construed as the representation of reality becomes -- or was always already (as poststructuralists have proclaimed for thirty-five years now against both New Critics and traditional Marxists, who insisted on either a formal or a material real) -- the real.[26] This erasure is especially true in the notion of the panopticon, where the only reality possible is that presented us by social constructions of disciplines, discourses, and institutions. Panoptic society is always already virtual. We can always know only what we need to know in order to ensure the continued operation of the panopticon. In this sense new technologies-as-panopticon is the apotheosis of the notion of panopticism. If what we really mean by the panopticon is a vehicle for the conveyance of ideology, new technology is meta-panoptic: it disseminates itself as ideology.[27] Total immersion as ideology is the vision of The Matrix, in which, for most citizens, no reality is visible but that created by the technologies that keep them unconscious, still and drained in a hidden post-apocalyptic landscape, while convincing them that their lives are rich and full.

But new technologies (if not the Internet alone) have an even more immediately total effect on the body. For Foucault, the panopticon is the social model for which the Enlightenment strove; it is the social raison d'être for the rise of the sciences and other disciplines. In a counter-psychoanalytic manner, the panopticon acts on the mind through the disposition of the body, rather than the other way round. Like ideology, it seeks to be the sum of all experience for the citizen. On its own, but more especially in conjunction with other new technologies, New Technology is the closest thing to an apotheosis of that desire for totality; it is along this dimension, then, panoptic.

The Internet as part of a matrix of new technologies affects the body panoptically, which is to say in an unconsciously oppressive manner. Paradoxically, but in a manner that confirms the increasingly diffused quality of power through new technologies, the affluent sector that can afford PDAs -- which is to say the presumable wielders of power -- is most in thrall to the technology: at the beck and call of anyone who pages, phones, or faxes them. The PDA and cell phone owner is not the mistress of the manor but its chatelaine, which is to say medieval middle management: Herbert Marcuse's One-Dimensional Man or Sloan Wilson's Man in the Gray Flannel Suit.[28] However, the Internet citizen may be differentiated from corporate middle management in one important respect: the corporate citizen exists in a vertical hierarchy at the top of which is a CEO, and at the bottom of which is, depending on your point of view, the worker and/or the consumer. In contrast, in the new-technology panopticon, everyone is middle management: again, everyone has a little power: all are oppressive. Everyone is answerable to a higher authority, and everyone is a higher authority. In cyberspace immersion, everyone has approximately the same status, despite minor differences between, say, the surfer and the web master. We are assaulted by banner ads, but we can also flame anyone with an email address, a website, or a chatroom or Listserv presence. It is the One-Dimensional Man made to look three dimensional in two dimensions. Put another way, for the new middle management, new technologies are the apotheosis of Foucault's notion that panoptic society recreates the body as mechanical: "the machine required can be constructed" (D&P, 135). New Technology is the literal embodiment of the somewhat figurative language Foucault uses in describing a discipline as "a type of power, a modality for its exercise, comprising a whole set of instruments, techniques, procedures, levels of application, targets; it is a 'physics' ... of power, a technology" (italics mine).[29] While other forms of representation (movies, novels) are only metaphorically technologies (or, at best, the products of an invisible technology), the Internet is visibly technological. While poetry captures the imagination, the integrated cell phone/PDA envelops the body.

So, while the direction of the gaze or the constitution of space does not define the Internet as panoptic, the Internet as a total experience can be a literalization of Foucault's panoptic society in a way that no other institution can. As a consequence, the Internet tends to leave us aware of our minds but not of our bodies. One might compare the Net with masturbation fantasy, an experience that, while implicating the body, takes place mostly in the mind's eye, which is to say virtually.[30] Finally, as a sector of new technologies, the Internet represents a continuation of the twentieth-century media impulse to refashion the panoptic "by other means." If the twentieth-century media panopticon is cryptically and incompletely panoptic, the twenty-first-century matrix of Internet/new technologies is perhaps volitionally but completely panoptic, which is to say that -- for now -- we may still choose not to enter this particular prison, but also that, once in, our movements are determined by the technology.

If the Internet is a response to the Enlightenment interest in the effect of the mechanical on the body, we are still left wondering about the valence of that relationship: do new technologies subvert or apotheosize the Enlightenment? Is the web an enlightenment project or, as Ravi Sundaram, citing Donna Haraway and other utopic theorists, suggests, "a crucial opposition from the old Enlightenment oppositions of nature and culture from which flowed the representations of human praxis and subject formation"?[31] Is the Internet, as Haraway and others suggest, a reaction against the Enlightenment's careful divisions between the organic and the inorganic, a reaction against disciplinary thinking? Is it, in other words, Haraway's cyborg culture, in which distinctions between feminine and masculine, first world and third world, collapse with the collapse of the categories organic/inorganic? Or is the truth, as Foucault suggests before Haraway, that the Enlightenment already conflated the organic and the mechanical, so that new technologies are in fact the apotheosis of the Enlightenment project: the defining of the organic in terms of the inorganic; the exercise of the machinery of power on the body in such a way that the body itself becomes a machine? If it is an Enlightenment project, is it the traditional face of the Enlightenment -- the Encyclopedia, tentative experiments with democracy, and Isaac Newton -- or is it the panoptic face: discreet oppression, disciplinary disciplines, Jeremy Bentham, and Napoleon Bonaparte?

Discourse at the Level of Code: Writing and Reading the Internet

The appropriate and most frequent object of Internet studies is the individual Internet subculture: adult chat rooms; game sites; art forums; pornography sites; nationalist, racial, ethnic, and religious websites; and so on.[32] Nevertheless, I have eschewed such subjects for the sake of a quixotic tilt at the Internet as a whole. The sole advantage of such an approach is that it allows me to ask at this point: does the Internet itself constitute a discourse? Is there a discourse of the Internet? In other words, is the Internet already defining and defined in the manner of all disciplinary discourses?

Because of its diversity, the Internet as spectacle (what is actually visible on the screen) cannot easily be considered to have a unitary discourse. At best one might assert the assumption of most cultural-studies research: that each Internet subculture has its own discursive practice. But we may be in error if we identify the diverse languages on the Internet as the language of the Internet. Though the Internet is international -- and so necessarily of many languages (though not all: no Masai) -- such globalism does not imply a freedom from the constraints of discourse. Other languages than those of the spectacular Internet determine the limits of the Internet and its way of knowing and expressing the world. Strictly speaking, the codes and markup language in which internet content is written -- java, html, xhtml, and so on -- constitute the discourse of the Internet in the very specific sense that they determine the absolute limits within which the discipline operates. We can create only what these languages allow us to express. Or, as Lev Manovich asserts,

As we work with software and use the operations embedded in it, these operations become part of how we understand ourselves, others, and the world. Strategies of working with computer data become our general cognitive strategies. At the same time, the design of software and the human-computer interface reflects a larger social logic, ideology, and imaginary of the contemporary society. So if we find particular operations dominating software programs, we may also expect to find them at work in the culture at large.[33]

If, as Manovich implies, the Internet and then the culture within which it is imbedded belong to whoever writes the code and codec, we may regard as significant the fact that so much software and hardware have originated in Japan, and that for economic reasons, the United States now imports much of its software development from other nations (India, for example). It becomes further significant that much of America's science elite -- including computer developers -- is (and traditionally has been) imported from elsewhere. The importation of Indian software implies an interface with post-colonial Indian culture. The Internet becomes transnational and multi-ethnic as much through its invention and development as through its use by various countries. Claiming America as the originator of the computer and the Internet constitutes the same order of historical dis-information as asserting that Thomas Edison invented the movies.

If it matters where the code comes from, we must assume that the writing of code reflects the conditions of the creator of the code. Put another way: the writing of code is not a completely scientific endeavor; rather, it is also cultural/sociological -- it is an affect of the writer and her culturo-socio-economic background. The knowledge that Alan Turing was persecuted for being homosexual might allow us to queer computer science.[34] We must still ask, however, whether computer code is a discourse; whether it creates texts that constitute subjectivity after the manner of other disciplinary discourses; and whether, as a result, the Internet code and markup language are interpretable in the same fashion as other texts.

If the base languages of the Internet are not interpretable then this conversation is at an end.

However, some kinds of interpretation already pre-exist the asking of the question. An aesthetics of code has existed among programmers almost as long as programming itself. Programmers refer to "beautiful" programs, not because of the spectacle it enables on the monitor or in any other instrument, but because of the elegance, clarity, and brevity of the code-text itself. In another arena, writers argue about the poetics of code, taking computer code as the starting point for the production of poetry and narrative.[35] Perhaps more significantly for this essay, a kind of semiotics of programming has explored code's potential ability to enable the computer to originate meaning; as Jay David Bolter suggests, this notion is part and parcel, for example, of investigations into artificial intelligence:

The computer is a machine for creating and manipulating signs; the signs may be mathematical, verbal, or pictorial. Computer programming and indeed all kinds of writing and reading by computer are exercises in applied semiotics. The first lesson any sophisticated computer user must learn is the difference between a sign and its reference, between the address of a location in the computer's memory and the value stored at that address. This dichotomy characterizes the machine at all levels: it is at the essence of hypertext and of programs for artificial intelligence, in all of which text is simply a texture of signs pointing to other signs.[36]

Though I will assert in a few moments that hypertext or any other originary Internet language is more than simply the usual and presumably endless chain of signifiers, I am intensely interested in the almost paradoxically hermeneutic/cultural-studies endeavor implied in this kind of study. The Internaut as hermeneut, or the Interneut. Let us concern ourselves, then, with the semiotics of the Internet code (Semenet), or, strictly speaking, of Internet code as discourse (Intercourse, Disinter, Disintery, or perhaps just DisCo).[37]

We might find a model that will help us, if not read the Internet, at least imagine how the Internet might be read, in Ferdinand de Saussure's oppositions between langue and parole and between signifier and signified. (These models also inform Foucault's notion of discourse.) Most critics of the web implicitly assume that the spectacular web is a signifier after the manner of television and media: it has something to do with representing the real world.[38] However, at a much more obvious -- if counterintuitive -- level, considered as the signifier, programming code is deployed on precisely the same assumption that Saussure makes about language: it is binary opposition. If one then not zero; if zero then not one. Platform and programming languages may be seen as langue, while the actual programs written are its parole. Read this way, we have as the object of study a sort of global language, the origin of whose rules matter less than the actual deployment of those rules by particular persons and cultures.

However, the same problem arises with this argument as with most structuralist arguments: it is at least partially ahistorical in assuming that, while the writing of particular programs (parole) is affected by the culture within which the codes are written, the context and structure within which programming language exists (langue) is irrelevant. To make programming code(s) the Master Code is simply to reify the notion that the white culture within which it is assumed to have been produced is substructure to every subsequent digital representation's superstructure. The analogy to Saussure's model is further flawed because that model is itself a partial explanation for the difficulties in translating from language to language, or from langue to langue. It assumes a multitude of different cultures speaking in different tongues and thus producing different kinds of meaning. Most of the Net, however, is uttered in html, xhtml, or one of a handful of other web-creation languages. In this sense, nothing really distinguishes the production of meaning by New Delhi or Silicon Valley software developers. So even if the web is not monolithic by virtue of its parentage, it is to some degree structurally monolithic. As such it is discourse, a language spoken by adepts in a particular discipline (programming) that determines that the text it creates will reproduce the same pro-social meaning over and over again.

However, complicating our vision of Net discourse as monolithic is the realization that we have examined it, so to speak, only horizontally: we understand that the use of html cuts across all national borders. We have not, however, rightly understood that programming has a vertical dimension: the abstract "assembler language" makes possible more linguistically recognizable programming languages, which in turn enable the creation of browsing software. Browsers read another code, html, which in turn makes xhtml, Flash, or Dreamweaver possible. Various plug-ins and filters are constructed on top of these already precarious edifices, creating, not one, but several Towers of Babel in which each kind of code must learn to live in often-uneasy détente with other kinds of code. Finally, proprietary programming codes -- often variations on the major programming codes -- are deployed by individuals and corporations. In this scenario, the initial programming languages are not themselves monolithic edifices so much as the building blocks for an olio of edifices. Not substructure but bricks and mortar. Perhaps a clearer analogy is not with French or English, but with Proto-Indo-European, Altaic, Uralic, and Austro-Asiatic, themselves probably not monolithic, and finally all but erased and displaced by the flood of premodern and then modern languages to which they give birth.

Another, more contemporary analogy may make the possibility of code plurality and heterogeneity clearer. From 1895 until about 1950 there was no significant mass technology for viewing the moving image other than celluloid film projection. However, despite the realization that media has an undeniably homogenizing effect globally, most film theory and criticism since World War II has assumed that this basic commonality in technology has little effect on the medium's ability to express different cultures and ideologies in a heterogeneous fashion. (In fact, film historians tend to be nostalgic for the unifying internationalism of silent film.) If one imagines for a moment that programming language is of the same order of artifact as film -- in essence an "apparatus" rather than langue in the traditional sense -- then the contributing languages of the Internet become at least as flexible in representing or resisting ideology and ideological difference as film is thought to be.[39]

While the structuralist Saussurean linguistic model seems inadequate for describing programming code or HTML, at a different level of signification, code embodies a certain post-structural idea about language: like language, code not only signifies the Net, it actually creates what it defines. But again, while this notion as observed about writing or speech is conceptual -- a truth that must be inferred -- it is self-evident in the case of the Internet. While we assign the inventive function of language to some vague and invisible ideology that requires our seeing the world in a trickily, repressively, and coercively pro-social fashion, we understand more directly that computer code is the abstract prime mover of effects we can see: Internet effects. In a certain sense the Internet embodies the inventive quality of language because it is representationally the opposite of language. While writing is palpable and the things it represents are abstract and arbitrary, code is (for all but the programmers who originate, improve, or steal it) invisible, while the things it represents -- for example the Internet -- is palpable, if virtual. In spoken/written language, the signifier is audible/visible while the signified -- ultimately ideology -- is invisible if ever-present; conversely, the visible effects represented on the Internet paradoxically signify both the existence of an invisible code/markup language, and also the signified of the code that has uttered them into existence. In a dynamic reminiscent of the now rather self-evident if principal declaration of deconstruction, the Internet is the signifier of its own system of signification, or the signifier of its signifier. It is both signifier and signified.[40]

However, the very materiality and palpability of the Internet -- in contrast to the invisibility and evanescence that characterize the ideological ambiguity and arbitrariness to which deconstruction always refers -- suggests that something other than simple deconstruction is at work here. The Internet is like language-as-ideology in being a construct, but unlike ideology in being a material construct. Again, ideology must be inferred while the Internet is palpable. While, like other critical strategies, deconstruction depends on subverting the Saussurean naturalization of signification by observing that the relationship between signifier and signified is only conventionally binary and so, consequently, ideological, the relationship between the signifier and the signified for the Internet is (as Christian Metz asserted twenty-five years ago of the film image)[41] of a different order, requiring a different hermeneutics. It is as if the Internet -- and software more generally -- is the palimpsest of its code-as-signifier, which covers its own existence with the artifact it creates in a way that the written word never does. Code is not so much a self-consuming as a self-effacing artifact.[42]

What does this reciprocity of signifier and signified mean about the meaning of the Internet, and its status as panopticon-in-training? The answer is, I think, fourfold: the Internet's panoptic status depends, first, on whether its reciprocal model for the sign becomes the dominant trope for the way in which language works in culture[43]; second, on which social movements are conceived during the time of the invention of the Internet, third; on how the Internet inflects those movements; and fourth, on the awareness of the surfer of the encoding that underlies her Internet travels. To elaborate briefly: for Foucault, the panopticon has survived for over three hundred years as a defining if evolving technology that reflects changes in the way language is conceived: both as a series of discourses meant to define and categorize, and as a method of ordering people during the rise of the modern nation-state. In an analogous manner, the Internet arises at a time of industrial globalization and cultural border crossings, in other words during the dissolution of the nation-state. The new linguistics of the Internet will arise when the language has been foregrounded once more as an important political/ideological issue. It is not so much a question, as Manovich asserts, of the Internet's becoming the dominant cognitive model so much as a question of how Internet language may internalize and then re-inflect contemporaneous social phenomena. If, for example, the rise of the novel is significantly co-incident with European colonialism, then it may be important that the Internet's growth thus far is roughly coextensive with both the establishment of the World Trade Organization and the publishing career of Homi Bhabha.[44] The Internet in this respect is problematically transnational; it enables both access and exploitation. While we understand that it is the product of late capitalism, of an economy interested in the ownership of ideas more than of things, we still do not know whether the Internet is Habermas's new modernism or Jameson's postmodernism. Is it the cultural arena for solving the problems of being modern, or simply another tool of late-capitalist exploitation? At the moment it seems to be both.[45] The self-consciousness of the surfer of the code is approximately analogous to a consciousness of ideology. It is the awareness of Internet as an artifact whose production of meaning is up for grabs by anyone who understands that it does produce meaning in very material -- which is to say manipulable -- ways.

The notion of a useable indeterminacy in meaning seems even more sensible as one understands that the fact of the Internet's signifying its own code is different from the relationship between other media and their material bases. Cinema does not signify the actual celluloid and emulsion that constitute it. And it can only inferentially signify the process of creation; even shot-for-shot remakes of films (Psycho, 1960, 1998) can't reproduce with any exactness the original. And for other reasons -- expense, access to means of distribution and exhibition -- film and television are difficult for everyone to produce. However, the Internet's code is instantly visible (if not immediately coherent) to anyone who clicks the appropriate browser button. And so the codes are revisable in a way that no other medium is.[46] In this respect, computer code allows at least the second coming of Walter Benjamin's notion of the reproducibility of texts as a kind of resistance: more people can appropriate the code of the Internet because it is more or less readily available for reinterpreting and rewriting in a way that film and television are not. At most, computer code suggests a re-thinking of how sign and discourse create the subject because, since the relationship between signifier and signified is no longer completely arbitrary (though it is still completely conventional), neither then is the real simply a construct. Virtuality is, ontogenetically, neither completely material nor completely constructed. One would have to invent a new descriptor for its status, perhaps "materucted," or "consterial," or even "convircted."

The Internet, then, stands in relation to discourse as it stands in relation to other elements of panopticism: it is and it is not. It is discourse in the sense that it is language; it is not traditional discourse because the organization of its sign system departs radically from that of traditional systems of signification. Because the nature of Internet discourse is -- at least for now -- ambiguous in its newness, it is also difficult to assert anything about Internet subjectivity. If discourse constitutes the subject, what are we to say about a subject whose very materiality is in question, not because of the traditionally poststructural sense that it does not exist as such, but rather because the sign system of the Internet is impossible to define as simply material? At most, we may assert that the very ambiguity of Internet subjectivity suggests something different from the pronounced panoptic subjectivity created by the integrated PDA/web browser/cell phone.

Conclusion: Panopticism 2.0

If we have decided that the Internet is readable as code, even if that code is not clearly a disciplinary discourse, we still have not enabled critical activity. Though beautiful to the programmer, code and markup language are of course anathema to the critic and historian who, whatever flourishes her rhetoric may contain in the direction of corporate professionalism, has probably entered the discipline of literary or cultural studies because of the more or less neurotic love of the object of study -- the novel and the poem -- that the discipline demands. The experience of reading code can in no way be thought of as the substitutively erotic experience of reading literature. It is not, in short, jouissance.

I can imagine as a consequence two responses to the reading of code. In the first, we may change the mode of processing pleasure away from the immediate gratification of the novel reader, and toward the delayed gratification of the archaeologist, most of whose pleasure probably does not derive from the initial reading of the text but from interpretation: translation, reconstruction of cultural context, and so on. One model for this kind of pleasure may be found closer to home in Walter Benjamin's projected Arcades Project, which was in some large measure supposed to be the reconstruction of all-but-disappeared texts.[47] This is the pleasure that critics in any case claim for themselves: reading is always interpretive. Reading code would simply be a test of the truth of that claim.

In the second response (as I have tried to hint throughout this essay), the interpretation of code might indicate a truly "professional" turn in literature studies because it would not evoke the traditional text's moiety of jouissance. Approaching the Internet through its codes will reflect the direction of literature and cultural studies as discursive practices toward an ever-increasing utilitarian function in the disciplines, a function closer to linguistics than to the reading of literary texts. In short, whether or not we discover the Internet to be panoptic, the study of code threatens to reveal literary scholars to be contained in an increasingly rigid panopticon that reduces the "pleasure of the text" to degree zero. However, though hemmed in on one side by the constraints of the interpretive disciplines, politicized scholars might be involved in debates about representation that will have important political/ideological consequences over the next few years: dialogues about open source codes, Internet censorship, intellectual property, and copyright are always already questions about representation and interpretation.

Even more liberating is the possibility that understanding the signifying system of the Internet would change our notion of how signification works. For Foucault, this kind of altered understanding implies an alteration in the way that we experience the world. In The Order of Things Foucault discusses how our attitude toward language changed and was itself changed by our culture's changing relation to epistemology. For example, he discusses the sixteenth- and seventeenth-century transition from the "classical" to the "modern" in linguistic terms:

[I]n the sixteenth century, one asked oneself how it was possible to know that a sign did in fact designate what it signified; from the seventeenth century, one began to ask how a sign could be linked to what it signified... The profound kinship of language with the world was thus dissolved... Things and words were to be separated from one another. Discourse was still to have the task of speaking that which is, but it was no longer to be anything more than what it said.[48]

Foucault historicizes the "fact" of our realization that sign and signifier are related only arbitrarily. He finds a moment that predates Saussure by some two and a half centuries; the very beginning of the "long Enlightenment" is the moment of the emergence of the kind of discourse that produces modern disciplines and institutions out of the realization that, though words mean, that meaning is manipulable. In an analogous fashion, we may wonder whether the possible emergence of the Internet as a sort of ambiguously constituted Panopticon 3.0 does not bring with it a fundamental shift in the way we understand the world. It is perhaps the case that as a consequence of an accident in technology, we return to a revised or neo-classical version of pre-modern signification, in which words are once more genetically -- or at least mechanically -- linked to the things they signify.

One can imagine this realization as a liberation at least from the constraints placed on us by contemporary notions of signification. Under the current linguistic regime, cultural criticism often seems a sort of game in which we continually have to try to imagine a way out of our almost inevitable realization that definition is absolutely limited by language. We have different strategies for this game that, depending on the daring of the player, are conceived as either critique or subversion: "structures of feeling," "signifyin'," "the carnivalesque," "heteroglossia," and so on. What if we could go on at least to a different game that questioned in some more fundamental way the limitations placed on us by signifying-as-ideology? In short, it is conceivable that the Internet as ambiguous panopticon might be re-envisioned as a vehicle, not for oppression, but for a kind of liberation, not perhaps the human/machine dichotomy collapse imagined by Haraway and others (alas, a collapse already envisioned not only in the panopticon but in the aesthetic of Marinetti's fascist Futurism), but a kind of liminal space in which identity formation can be questioned. The problem with this liberation is of course that it is, for the most part, ersatz: it exists only virtually. But virtuality is its virtue as well. Virtuality is the ground of the Internet's ambiguity. Perhaps the linguistic oddity that characterizes the Internet will provide a way out of the otherwise inevitable binary opposition in discovering meaning. Perhaps there is another way of framing the future of the Internet than as one of two opposing stances, in other words as either the site of, if not liberation, at least of significant mass resistance (as opposed to, say, the bulk of academic Marxism or American independent films), or the helpless reproduction of the ideology of the culture that produced it, as in Benjamin's oft-quoted Kafka quotation: "Oh, plenty of hope, an infinite amount of hope -- but not for us."[49] A slightly more optimistic vision of the Internet might be found in a slightly different Benjamin:

Only when in technology body and image so interpenetrate that all revolutionary tension becomes bodily collective innervation, and all the bodily innervations of the collective become revolutionary discharge, has reality transcended itself to the extent demanded by the Communist Manifesto.[50]