LAST YEAR, Apple CEO Tim Cook gamely defended the company’s most comically dysfunctional product to Walt Mossberg, the Wall Street Journal’s tech columnist:
“What makes Siri cool is that she has a personality. She becomes many people’s best friend.”
“Isn’t that a little sad?” Mossberg replied.
“Hey, I’m not a judge.”
Cook has it half right: Siri may be indispensable and endlessly amusing to lonely techies, but not because she has a mind of her own. Like other successful Silicon Valley exports, she is an extension of the user. She straddles the two most recent generations of artificial intelligence devices: those that enable users to be more efficient but that only approximate user intent and those that more precisely decipher user queries. The goal is to create a machine that knows users better than they know themselves. Economist Tyler Cowen believes the end result will be a partnership between people and devices, with less control but greater efficacy. “There will be a thriving market for intelligent agents, virtual helpers, expert systems and a host of other services that leverage advancements in machine learning,” according to Alex Williams at TechCrunch. And consumers are not the only use case: the enterprise space is looking to enhance (or replace) human capital with capable digital actors. For now, however, the focus is on users: Siri, Google’s new Hummingbird algorithm, which improves search results for more conversational queries by factoring in context, and Amazon’s recommendation engine are just some of the more visible examples.
This is “the crazy shit VCs used to go for,” as Thomas Pynchon puts it in Bleeding Edge. They’re going for it again, and the chronicler of nuclear domination could not be timelier. His lifelong interest in the entropy of modern science makes him an ideal observer of tech ambitions. It’s a world that should be a godsend for Pynchon: one that believes it is poised to “organize the world’s information,” according to Google. Even more fitting, as Siri demonstrates, it aims to do so through technology that mimics human beings, and provokes human behavior to respond in kind. What Bleeding Edge reveals is that, though he has long been regarded as an opponent of techno-science (a Luddite even), Thomas Pynchon’s literary career has shared more with the dehumanizing ethos of Silicon Valley than has been recognized. Siri is a Pynchon character.
Pynchon’s career-long obsession with entropy began when he read Norbert Weiner’s The Human Use of Human Beings. The book posited that entropy inevitably leads to, as Pynchon described in the introduction to his short story collection, Slow Learner, “universal heat-death and mathematical stillness.” Pynchon, along with much of the restless youth of his generation, reacted with “somber glee at any idea of mass destruction or decline.” Weiner would revise the volume four years later to read more optimistically, with a new humanist focus on the ability of machine learning, modeled after its biological counterpart, to help resist entropy.
In The Crying of Lot 49, Pynchon offered his own solution to the problem of thermodynamic entropy by way of a famous physics thought experiment called “Maxwell’s Demon,” and it suggested a very perceptive view of what Weiner’s vision would eventually become. Through sorting hot and cold molecules at the direction of psychic communication with a human “sensitive,” the machine would decrease entropy. According to John Nefastis, the inventor of a Nefastis Machine — a device containing a Demon — feedback between the sensitive and the machine is what makes the whole experiment work. After a valiant attempt at powering the Demon with her mind with Nefastis’s guidance, Oedipa, Pynchon’s heroine, concludes that, “the true sensitive is the one that can share in the man's hallucinations, that's all.” Pynchon was mocking Weiner’s willingness to give man and his inanimate mimic a common purpose and place them on equal footing to achieve it. For Pynchon, both are too inept to help each other.
He was not alone in his skepticism of Weiner. French philosopher Jean Pierre Dupuy offered the best critique of Weiner’s “humanism”:
“Will it be possible one day to design a machine that thinks?” The cybernetician’s answer, rather in the spirit of Moliere, was: “Madame, you pride yourself so on thinking. And yet, you are only a machine!” The aim of cognitive science always was — and still is today — the mechanization of the mind, not the humanization of the machine.
However, unlike Dupuy — and despite his protestations in Slow Learner and The Crying of Lot 49 — Pynchon instinctively rejects the metaphysical conception of human beings. Gravity’s Rainbow ostensibly argued for the fallacy of human will:
Pavlov believed that the ideal, the end we all struggle toward in science, is the true mechanical explanation … His faith ultimately lay in a pure physiological basis for the life of the psyche. No effect without cause, and a clear train of linkages.
“A screaming comes across the sky” heralded the death of humanism by neurological signals. Unlike his cybernetics friends, Pynchon did not have to assert that humans should act like machines because, in his anti-humanistic view, we were in fact already machines. Pynchon’s corpus largely evaded engaging with objections to this belief by burying it in the existential preoccupations of Cold War era science. He would likely have been able to carry on had he not chosen to tackle a technology industry convinced that such questions were its responsibility to solve.
Pynchon is hardly thrilled about technology designed to intuit human behavior. DeepArcher, the Second Life-like game that serves as Bleeding Edge’s technological backbone, quickly blends into reality such that it’s difficult to tell where New York City begins and the stimulation ends. Troubling under most circumstances, but all the more so once it becomes a part of 9/11 grieving:
Yet it’s as if they want to engage — they get eye contact, smile, angle their heads inquisitively. “Yes, what was it?” or “Problem?” or “Not right now, OK?” If these are not the actual voices of the dead, if, as some believe, the dead can’t speak, then the words are being put there for them by whoever posted their avatars, and what they appear to say is what the living want them to say. Some have started Weblogs. Others are busy writing code and adding it to the program files.
It’s a harmless escape, isn’t it? But the problem with technology that helps us deny reality is that it delivers on a false promise: providing a best friend where one cannot exist.
The team behind DeepArcher is a brilliant but vacuous and immature group from Silicon Valley who moves to New York. Maxine, Bleeding Edge’s heroine, asks them if they understand the deeper philosophical implications of their creepy product, to which she receives the following response: “‘It’s code,’ Justin a little bewildered, maybe, ‘just keep the thought, couple geeks up all night on cold pizza and warm Jolt wrote this, not exactly in VRML but something hypermutated out of it, ’s all it is.’” It falls to Justin’s wife, Vyrva, every bit the California blonde but for the Stanford degree, to sheepishly explain that, “‘They don’t do metaphysical, ’ … flashing Maxine a smile falling noticeably short of fond amusement.”
Though Pynchon appears to disagree with these hackers’ lack of humanism, he once again cannot escape replicating the attitude he criticizes in his own work. Taking a page from the tech playbook, Pynchon atomizes his characters into their most essential characteristics. Critics of Weiner might well identify Pynchon’s signature form of storytelling as fundamentally lacking in its understanding of people. As Dupuy wrote, “the makers or re-creators of the world know nothing of the beings who inhabit it, only lists of characteristics.” Anti-humanists do not realize that, “the most perfect simulation still fails to capture something.” Dupuy does not mean to offer a different approach that would avoid these shortcomings, only to point them out lest we begin to think that technology is already on its way to becoming human. And what is it exactly that machines currently lack? They have not solved the mystery of love, for, “When we love somebody, we do not love a list of characteristics.”
Love is not one of the irresistible driving forces in Pynchon’s universe. In Mason & Dixon, he allows for tenderness between Charles Mason and Jeremiah Dixon and Gravity’s Rainbow teems with tragic love-death sexual orgasm. But the most consistent charge against Pynchon’s work is that his characters are not sympathetic, and usually not very important anyway. His first loyalty is to ideas. If flat archetypes and half-formed illustrations are all that is needed to convey those ideas, then why not? For Pynchon’s defenders, it’s not that simple. Thomas Jones, in his review of Inherent Vice, had the following to say:
As readers we may rely on our liberal humanist ability to ‘empathise’ with immaterial strangers, but we can still tolerate with bland equanimity the manifold suffering of the wretched of the earth when we put down our novels and turn on the evening news. That’s OK: if we couldn’t, we’d all be suicide bombers. Still, in this respect, Pynchon’s alienating novels are altogether more ‘realistic’ than any number of finely wrought explorations of individual consciousness.
Fair point, but the argument against Pynchon was never about realism or sentiment. Without humanity, characters are unnecessary. If their lives nevertheless occupy a good portion of Pynchon’s books, we are left wondering, to put it bluntly, what is the point? It doesn’t help that his characters are in a bit of a no-man’s land: too well delineated to seriously indict the ease with which people dehumanize one another and too caricatured and inconsequential to serve as anything more than functional gaps in a dense and overwhelming narrative. Therefore, despite all of Pynchon’s drama, erudition and prescience, his novels never really have anything at stake, even when they address powerful themes.
In Bleeding Edge, the cast’s mechanical nature comes across as inartful at best. At worst, it renders his characters too obvious to be interesting. Maxine is a living New Yorker cartoon, with reflexive high-low, biting wit, a divorced mother with a thing for her ex and an unusually dedicated fraud investigator. Pynchon discourages any further plumbing by reducing her internal life to, “PAUSE, then STOP, then POWER OFF, smiling without visible effort.” She carries on an affair with a mercenary covert agent, moonlights at a strip club, accompanies her parents’ crazy neighbor to a drug transaction and keeps an emotherapist, all of which are equally inexplicable.
Her supporting cast, though not expected to uphold nearly as much of the action, receives about the same number of brushstrokes. In a former life, Maxine’s spy lover “acquired a portfolio of pain and damage applied to various human body parts that might have added up to hundreds — who knows, maybe thousands — of deaths on his karmic ticket.” He married a local girl in Guatemala before leaving her and leaping into his present life. We know he is trouble because of his job, and “the fucking voice, sonorous, overcoached, phony as a cold call on an answering machine.” March Kelleher evangelizes about conspiracy theories starring the US military-industrial complex. Her “Shirley Temple”-esque daughter is married to Gabriel Ice, the Oliver Peoples bespectacled geek overlord — the great villain of Bleeding Edge — who embezzles funds from his hot security company to help the United States catch terrorists. It’s never in doubt that his true dream is world domination through armies of servers. A Stuyvesant geek with foot fetish is bent on uncovering Ice’s crimes; former Russian spies are called upon to explain the novel’s most advanced weaponry.
California, Pynchon’s former hippie paradise, incurs all the wrath of disappointed nostalgia. Vyrva confides to Maxine:
“We came to New York, we all did, so innocent ... Back in California it was fun, just write the code, go for the cool solution, the elegance, party when you can, but here, more and more it’s like—”
Grown up. But only just: Vyrva wears “what Barbie used to call an Executive Lunch Suit” to meetings with VCs. Maxine’s emotherapist has “that vacant, perhaps only Californian, the-Universe-is-a-joke-but-you-don’t-get-it smile.” Justin and Lucas, the creators of DeepArcher, are primarily concerned about getting hacked, never mind the virtual reality they’ve created in the seedy Deep Web.
Pynchon had been up till now relatively safe placing sketches of people alongside technology, for they seemed to have so little in common. Pairing people with the Internet, however, he flirts with the danger of discovery: that, despite a gallant show of intentions to the contrary, he is in fact a subscriber to human mechanization. It turns out that before Bleeding Edge, Pynchon’s common themes of conspiracy, corporatism, the frightening consequences of science, paranoia and entropy were discussed in minor key, unresolved. All of these settle in the tech industry, signaling that science fiction is no longer stuck in the terrifying but one-dimensional specter of the Cold War apocalypse — it can advance once again to the visions of humanity’s future, and be realized. Today, the Nefastis Machine wouldn’t be a delusional fantasy but a shiny product with its own launch party.
It’s likely that if it were a real tech company, DeepArcher would be worried as much about the stickiness of its features as much as its security. Though they may have already worked that out: Maxine can fluidly click “on everything, faces, litter on the floor, labels on bottles behind the bar, after a while interested not so much in where she might get to than the texture of the search itself.” Her clickstream eventually ends up, as it is supposed to, at one of DeepArcher’s more developed features, which welcomes her with, “Take your time, young lady, we’re holdin her for you.” The program anticipates the needs of the user, before the user even knows she has those needs. That is the tech secret sauce that allows a newborn to correctly navigate an iPhone and Google Instant to predict the rest of your search query, among other tech-user partnerships.
Early this year, Google hired legendary technologist Ray Kurzweil to explore the concept of “the singularity,” the “theoretical moment in time in which artificial intelligence surpasses the human brain.” It has fascinated Larry Page and Sergey Brin for a long time, and they want it to happen first at Google. At a talk at Singularity University (yes, it really exists), Kurzweil predicted that humans will go beyond the "limitations of our biological bodies and brain" with the help of artificial intelligence. In his book, The Singularity Is Near, Kurzweil states that, “Along with the accelerating improvement cycle of nonbiological intelligence, nanotechnology will enable the manipulation of physical reality at the molecular level.” Further, the movement towards “greater complexity, greater elegance, greater knowledge, greater intelligence, greater beauty, greater creativity, and greater levels of subtle attributes such as love” will be facilitated by humans becoming more machine. The ultimate goal is to become immortal, to be “on a path that will get us to the next point” in our physical lives.
At least for one of Pynchon’s machines in Bleeding Edge, that moment has already arrived. With Maxine’s gun pointed at him, Gabriel Ice declares:
“It doesn’t happen,” Ice carefully watching the muzzle.
“How’s that, Gabe.”
“I don’t die. There’s no scenario where I die.”