If you've spent a lot of time lately riding on elevators frequented by executives from Deloitte and Touche, Archer Midland, Dean Witter, or Clark and McCullough, then you will doubtless be familiar with the term "disintermediation." For the rest of you the indulgence of an introduction is in order. Disintermediation means the elimination of the "middleman," especially as made possible by the Internet and electronic commerce. Among the functions that, through information technology, are now being widely disintermediated are those of wholesalers, retailers, distributors, bank tellers, insurance brokers, travel agents, and gas station attendants. As long as it's not one's own livelihood that's up for disintermediation, one can be happy to some extent with these developments. I've been to more than one fast-food place and chain store where I won't go back until the place is staffed entirely by robots. When they figure out how to automate some of the customers, they'll really have something.
But it's clear that the only people who ultimately benefit from disintermediation are the ones who can sell a product or service that disintermediates someone else, make a pile of money, and then get out before they in turn become disintermediated. Which means we are headed for mass unemployment. Today's disintermediate world is rapidly becoming one in which everyone unemploys everyone else, because there is no end to the human functions that can be disintermediated.
This goes not only for manufacturing and service jobs, but also for intellectual functions. Claims have been made that the Internet, with its potential to link together unlimited stocks of digitally represented knowledge, will disintermediate the library and the university, both subject to the fatal disadvantages of spatial location and world-time. A hot area in computer science is computational linguistics: the generation of computer models for understanding and using language. The advances underway in this field open up a potential universe in which reading and writing take place largely or entirely without humans. (Some of Norman Mailer's recent books have made great strides forward in this direction, I suspect).
Current text-processing software has already thoroughly changed the way many people make and use texts. Electronic libraries with millions of books and articles can be searched in seconds for words, phrases, and their logical combinations. It's no exaggeration to say that in the information age, texts aren't read, they're searched. But, as Heraclitus said, "If you do not expect the unexpected, you will not find it." To search a text instead of reading it is to renounce its capacity to surprise us, to make of the text more than ever before a tool, and to restrict its range of implication and suggestion to the ends we assign it.
With the right software we can train a machine to do our reading for us, combing digital newspapers and magazines for a few phrases that seem to match our interests, and to send us e-mail to tell us about what it found. Although many celebrate the time-saving potential of such technology, all it really does is push back the threshold between "productive" and "wasted" time. We still have to tell the machine what to look for and where to look, and we still have to decide whether or not to read what it finds, and if we decide to read it we still have to do so. For most of us the predictable result is more backlog of stuff that we know about but will probably never read.
This category is frighteningly, endlessly extensible. It has an infinite capacity and I have the queasy feeling that more and more texts are being produced on purpose just to fill it. It can easily accommodate stuff we might have read had we come across it in some other way than having it dredged up for us by an electronic agent that mostly feeds us junk; if someone had told us about it, for example, or if someone had written it to us and was waiting for our reply.
If being readily able to find texts that match a particular query affects our attitude toward texts, so does the expectation that search software will be used on them. On the Web, it's standard procedure for a writer to pad a document with "META" terms more or less describing its content, increasing its chances of turning up in response to a search. Writers and editors are under increased pressure to put proper nouns and subject descriptions as early as possible in an article and repeat them often, tactics that ensure a higher rank when the document is indexed for those terms. Most brazen of Web scribes are the pornographers who fill pages with random combinations of all the words that somebody somewhere might associate with sex: Try searching for "strap" and "Tina Louise" some time. (Here's a tip: If this article ever ends up on the Web and you want to find it, just remember to search for "Tina Louise" and "Heraclitus." Make sure it's Heraclitus and not Archimedes, though, because Tina Louise was in a movie about Archimedes.)
These texts aren't written for people; they're written for search engines. Which is to say they're not written at all. Writers and readers, in my definition, are people who expect the unexpected. Without the expectation of a human reader amplifying, agreeing, objecting, or frowning—an unpredictable range of responses, in other words—there's no getting off the treadmill of the writer's own mind, and there's certainly no escaping the feedback loop of the word-vomiting software program.
Writing for computers is at the opposite pole from writing as "a striving to depart from one's own words, with which nothing essential can be said." Mikhail Bakhtin, who wrote this phrase, may have been the thinker who most deeply thought through the relationship between creativity, responsiveness, and the unfamiliar, and he may have much to tell us about the role technology should play in this nexus. Bakhtin (1895-1975) spent most of his life in exile and enforced obscurity, effectively forbidden to publish (for long periods) by the Soviet authorities, and only posthumously emerged as perhaps the greatest literary theorist of the 20th century. His reputation in English-speaking countries rests mainly on his concept of the "carnivalesque" (Rabelais and His World), which has become a convenient knapsack for carrying stuff to cultural studies class, from David Bowie and the Wu-Tang Clan to Buffy the Vampire Slayer and Jerry Springer. Let's just say that the concept of the carnivalesque (which, for his part, Bakhtin always used as a historical category, and which, in his writing, always sounds as if resonating from his own decidedly noncarnivalesque historical situation), has turned out to be all too easily assimilable to late capitalism's self-image, and has thereby become overfamiliar and lost its force. But there is another, perhaps more radical and universal Bakhtin, the philosopher of polyphony and dialogue.
In the later writings and notebook entries collected in English under the title Speech Genres & Other Late Essays, Bakhtin develops most suggestively his concept of "understanding as dialogue": "An utterance is never just a reflection or an expression of something already existing outside it that is given and final. It always creates something that never existed before, something absolutely new and unrepeatable" ("The Problem of the Text"). This uniqueness derives from, and takes its moral importance from, the unique viewpoint each speaker embodies: "The whole utterance... has not mere formal definition, but contextual meaning." This context includes not just the speaker's space-time position, linguistic range, and psychology, but also the fact that no utterance is ever the first: It always responds to, takes up, amplifies, criticizes, ridicules, questions, interprets an utterance that came before it. "Any understanding of live speech, a live utterance, is inherently responsive," Bakhtin says in "The Problem of Speech Genres." "The listener becomes the speaker."
Dialogue for Bakhtin isn't just the actual, specific situation of a conversation or colloquy; the category includes also, and more importantly, the dialogue conducted across centuries, over "great time," about ultimate and irresolvable matters. In view of this broader dialogue, Bakhtin conceives (in "The Problem of the Text") of a "superaddressee." "Any utterance always has an addressee (of various sorts, with varying degrees of proximity, concreteness, awareness, and so forth), whose responsive understanding the author of the speech work seeks and surpasses... But in addition to this addressee (the second party), the author of the utterance, with a greater or lesser awareness, presupposes a higher superaddressee (third), whose absolutely just responsive understanding is presumed, either in some metaphysical distance or in distant historical time (the loophole addressee)." This "third party" understands me better than anyone at hand can, even better than I can myself. It reflects the participation of my speech in a dialogue that no individual will ever experience as finished, but which will conclude nevertheless: "Sooner or later, what is heard and actively understood will find its response" ("The Problem of Speech Genres"). This concept, in the context of Bakhtin's own enforced silence and his words' inability to find the response they craved, becomes tragic. In what may have been the second-to-last sentence he ever wrote, in "Methodology for the Human Sciences," Bakhtin returns to this theme: "Nothing is absolutely dead: Every meaning will have its homecoming festival." Mallarmé said that everything in life exists to end up in a book; Bakhtin agrees but adds the assurance that the book will actually be read.
To try and consider the utterance apart from the dialogue to which it belongs, from the meaning that comes only from this belonging, and from responsiveness whether direct (that of the immediate addressee) or delayed (that of the superaddressee), is to render it inert and valueless. "If we anticipate nothing from the word, if we know ahead of time everything that it can say, it departs from the dialogue and is reified," says Bakhtin in "The Problem of the Text"—an insight that needs to be read in conjunction with Heraclitus's about failing to expect the unexpected.
In terms that are directly applicable to today's technologies of the text, Bakhtin criticizes standard methodologies in the so-called human sciences (he seems to have had in mind structuralism in particular): "The exclusive orientation toward recognizing, searching only for the familiar (that which has already been), does not allow the new to reveal itself (i.e., the fundamental, unrepeatable totality). Quite frequently, methods of explanation and interpretation are reduced to this kind of disclosure of the repeatable, to a recognition of the already familiar" ("From Notes Made in 1970-71"). This reductionism is the same thing as a limitation to one's own consciousness and absence of the other: disintermediation.
Furthermore, the "third party" of the superaddressee is to be sharply distinguished from the neutral, depersonifying observer of the sciences, also called a "third party" (in "From Notes Made in 1970-71"), in whose world "everything is replaceable."
There exists an abstract position of a third party that is identified with the "objective position" as such, with the position of some "scientific cognition." The position of the third party is quite justified when one person can assume another position, when a person is completely replaceable. But it is only justified in those situations... where the integral and unrepeatable individuality of the person is not required.
This occurs when we consider a person only as fulfilling a function, as doing the same thing that someone else, or a machine, could do. This disintermediated man is an abstraction, and his world is removed from "the most vital, experienced life," the only life in which can take place "the complex event of encountering and interacting with another's world." To do justice to this event in its richness, Bakhtin calls for "sciences of the spirit" to take over from the so-called human sciences, which have failed to recognize the primordiality of dialogue.
Each of us, and especially as we're concerned with language, faces the danger of the replaceability Bakthin describes. In light of Bakhtin's warnings, we may wonder: How long before the permanent unemployment now creeping up on us will have also overtaken our ability to produce and interpret texts? We've secured longevity for our words—magnetic and optical storage being indefinitely renewable and much higher—density and considerably cheaper than wax, papyrus, or paper. But our words' ultimate destination is no longer a human reader or a human court of judgment, much less a metaphysical one such as Bakhtin's superaddressee. Instead, judgment will be passed on them by search-and-retrieval technologies that pulverize texts into concatenations of "keywords," by electronic "gatekeepers," accounting systems, and virtual bugging devices that log "hits" and track the "behavior" of the accessing agent.
I think Bakhtin, ever attentive to the vocal textures of language, would have been suspicious of the word "disintermediation." What's so positive about getting rid of an intermediary? Why is intermediation a bad thing? "Sometimes the shortest way through is the longest way around," says a character in Nicholas Ray's Wind across the Everglades. Stages, delays, and detours are indispensable not just to our temporal condition (if we have everything we want at the moment we want it, it's a sure sign we've fallen out of time) but to whatever charm life has.
Theorist Paul Virilio (Open Sky, The Lost Dimension) and film director Riccardo Freda (Theodora—Slave Empress, Sins of Rome) have both recently observed that life is becoming more and more like a Shakespearian tragedy. I'd say that's optimistic: I think it's becoming more and more like a Shakespearian tragicomedy. All's Well That Ends Well, for instance, in which people explicitly no longer believe in the codes they serve, in which the world becomes more manifestly unreal as it fills up with all the things said and not believed in.
Life becomes a tragicomedy when the other is no longer expected or sought, no longer believed in or able to believe in us, and we're driven to more and more freakish actions and gestures in the hopeless attempt to rekindle the other's belief. And finally—this is the tragicomedy—we become happy and successful anyway if we can be experts at manipulating symbols for machines to read in our place: proof that life's compensations are more absurd than its punishments: All's well that ends well.
At least let's try not to end up being like the people of Planet X in Godzilla vs. Monster Zero, whose emotions are "controlled by electronics." That would be awful!
Labor Day, 1998
