DISPUTATIONS FEBRUARY 6, 2013
In the current issue of The New Republic, Evgeny Morozov offers a critical take on Steven Johnson's Future Perfect: The Case for Progress in a Networked Age, lamenting the “quasi-religion” of “Internet-centrism.” In his response below, Johnson says his book "actually goes out of its way to avoid that kind of naive techno-determinism." And Morozov, in a rebuttal, concludes that "Johnson doesn't understand the substance of my critique."
Anyone worried that Chris Hughes’ ownership of The New Republic would turn the venerable publication into a vehicle for Internet boosterism will be delighted to read Evgeny Morozov’s new essay, “Not By Memes Alone,” running this week in the first official issue of the Hughes reign. Morozov’s essay is ostensibly a ten-page dismantling of my argument for “peer progressive” politics in Future Perfect, and like almost everything Morozov writes, it’s a smart and entertaining piece. He has a very astute riff on the dangers of what he calls “solutionism” in my work, and rightly observes that Future Perfect contains very little discussion of struggle or conflict—both of which strike me as being important critiques of the book.
Unfortunately, the bulk of the essay is a screed against what Morozov calls the “quasi-religion” of “Internet-centrism,” a movement that won’t be content until every institution is reinvented as a decentralized network fashioned after the Web or Wikipedia. This is not a new theme for Morozov, but it’s the first time he has targeted my work as a proponent of this dangerous new faith. I have to admit, everything that Morozov says about the dangers and limitations of the Internet centrists seems utterly convincing to me, and if I ever get a chance to meet some of these cultists, I will be sure to persuade them of the error of their ways. But using Future Perfect as a launchpad to renounce Internet centrism is a strange choice, since the book actually goes out of its way to avoid that kind of naive techno-determinism. This forces Morozov to do a number of awkward and misleading moves to make the book sound more doctrinaire than it actually is.
Some of those moves border on factual errors. Start with Morozov’s treatment of New York’s 311 system, which I endorse in the book as a successful example of decentralized peer networks being used to solve complex social problems. Morozov observes:
But Johnson is completely blind to the virtues of centralization. In discussing 311, he lauds the fact that tipsters calling the hotline to help create a better macro-level view of city problems. But this is a trivial insight compared with the main reason why 311 works: Mayor Bloombergs decision to centralize—not decentralize—previous models of reporting tips... Johnson’s Internet-centric worldview is so biased toward all things decentralized... that he completely misses the highly centralized nature of 311.
Here’s me from the chapter on 311 in Future Perfect:
It should be said that 311 is not a purely decentralized system. There are both literal and figurative headquarters, where the call center is located. In this sense, it is a hybrid form, somewhere between the pure peer network and the older state model. The 311 service vastly increases the number of participants in the system, and gives them the opportunity to set priorities for the city’s interventions. But those interventions are still triggered via a top-down mechanism. To a certain extent, that top-down element may be inevitable.
I think Morozov may be confused about the meaning of the word “completely.” Or perhaps this is just some kind of auto-correct mistake: where he typed “completely blind to,” he meant to type “is perfectly aware of and openly acknowledges.”
In another section, Morozov writes: “For all his talk of political philosophy, Johnson makes no effort to ask even basic philosophical questions. What if some limits to democratic participation in the pre-Wikipedia era were not just a consequence of high communication costs but stemmed from a deliberate effort to root out populism, prevent cooptation, or protect expert decision-making?” And yet half of one chapter is devoted to the problems with direct democracy, including an extended discussion of the way the founders framed those problems in the Federalist Papers. Morozov is free to disagree with my answers, but it is simply incorrect that I “make no effort” to ask the questions.
But enough about Morozov ignoring my words. The most revealing omission in the review revolves around his words. Future Perfect has a chapter called “What does the Internet want?” which Morozov predictably enough invokes as a telltale sign of Internet centrism:
The totalizers would happily follow Johnson in seeking answers to questions such as “So what does the Internet want?”—as if the Internet were a living thing with its own agenda and its own rights.
The problem with this diagnosis is that the chapter is explicitly about the difficulty of imagining the Internet as a unified positive force. It points out that decentralized architectures can be used to build terrorist networks as readily as crowdfunded charity initiatives. Consider this crucial passage from the chapter:
Perhaps it was a mistake to treat the Internet as a deterministic one-directional force for either global liberation or oppression, for cosmopolitanism or xenophobia. The reality is that the Internet will enable all of these forces—as well as many others—simultaneously. But as far as laws of the Internet go, this is all we know. Which of the numerous forces unleashed by the Web will prevail in a particular social and political context is impossible to tell without first getting a thorough theoretical understanding of that context.
You’d think that Morozov would want to mention that passage from “What Does The Internet Want?"—if only because the words were written by Morozov himself, in his earlier book Net Delusion. I quoted them at a very prominent early place in the chapter, precisely to make it clear that easy generalizations about the “logic” of the Internet were prone to failure. The whole chapter is a meditation on avoiding the pitfalls of naive tech essentialism; its answer to the question “what does the Internet want” is: “a lot of contradictory things.” But Morozov is so keen to denounce Internet-centrism that he doesn’t even seem to notice when his own words are being invoked enthusiastically as a critique of Internet-centrism. Now, it would be perfectly reasonable to argue that my critique doesn’t go far enough, or that I’ve misinterpreted Morozov’s position, or invoked it in bad faith. But instead, Morozov just charges ahead as if I haven’t engaged with his argument at all.
The argument that Morozov wants to make here is that we Internet-centrists (a group that apparently also includes Clay Shirky and Yochai Benkler) begin with our one true devotion to TCP/IP, and then conveniently backfill a history of lower-tech antecedents in order to justify our love, as Madonna might say. You wouldn’t suspect it from Morozov’s review, but the discussion of the Internet makes up only a fraction of Future Perfect’s content. He does manage to allude to my section on participatory budgeting in Brazil, but the book also includes long riffs on the prize-backed challenges offered by the Royal Society of the Arts in the mid-1700s; the “democracy vouchers” solution for campaign finance; the extraordinary rise in aviation safety over the past thirty years; the internal organization of corporations; childhood malnutrition in Vietnam, and so on.
These stories hail from very different historical and conceptual frames, but they share two important qualities: they are all directly related to the peer progressive worldview, and they have nothing to do with the Internet, or computers in general.
I can understand why Morozov wants to see Internet-centrism in my work: He’s built his career around debunking that belief system, after all. And yes, I’m glad the Internet and the Web were invented; I think that the world is, on the whole, better off for their existence. I would be surprised if Morozov doesn’t feel that way himself. But Future Perfect goes to great lengths to separate the promise of peer networks from some naive faith in Internet liberation. The main lines of its argument arose in part out of two book-length studies of peer collaboration in the 18th and 19th centuries: The Ghost Map and The Invention Of Air. My last book, Where Good Ideas Come From, ended with a survey of hundreds of peer-produced innovations from the Renaissance to today. The deep roots of the idea date back to reading Jane Jacobs on the “organized complexity” of the city in my twenties, which ultimately led to my arguments for decentralization in my 2001 book Emergence. I’m giving Morozov the benefit of the doubt that he just hasn’t bothered to read any of those books, since he doesn’t mention them anywhere in the review. But if you added up all the words I’ve published on peer network architecture, I wager somewhere around 90 percent of them are devoted to pre-digital forms of collaboration: in the commonplace book or the 18th-century coffeehouse, or urban neighborhood formation, or the traditions of academic peer review, or in the guild systems of Renaissance Florence. If Morozov were only a little less obsessed with the Internet himself, he might have some very interesting things to say about that history. Instead, he has decided to reduce that diverse web of influences into a story of single-minded zealotry. He’s like a vampire slayer that has to keep planting capes and plastic fangs on his victims to stay in business.
The point I tried to make explicit in Future Perfect is one that I’ve been implicitly making for more than a decade now: that peer collaboration is an ancient tradition, with a history as rich and illustrious as the more commonly celebrated histories of states or markets. The Internet happens to be the most visible recent achievement in that tradition, but it is hardly the basis of my worldview. And there is nothing in Future Perfect (or any of these other works) that claims that decentralized, peer-network approaches will always outperform top-down approaches. It’s simply a question of emphasis. Liberals can still believe in the power and utility of markets, even if they tend to emphasize big government solutions; all but the most radical libertarians think that there are some important roles for government in our lives. Peer progressives are no different. We don’t think that everything in modern life should be re-engineered to follow the “logic of the Internet.” We just think that society has long benefited from non-market forms of open collaboration, and that they’re aren’t enough voices in the current political conversation reminding us of those benefits. For peer progressives, the Internet is a case-study and a role model, yes, but hardly a deity. We would be making the same argument had the Internet never been invented.
In his response, Steven Johnson raises four main objections to my review:
Objection I: Johnson claims that he is not an Internet-centrist because 1) Future Perfect went "out of its way to avoid ...naive techno-determinism" and 2) one of the book's chapters is about "the difficulty of imagining the Internet as a unified positive force."
Objection II: Johnson claims that his book is not really about "the Internet," as it also discusses "Royal Society of the Arts in the mid-1700s; the 'democracy vouchers' solution for campaign finance; the extraordinary rise in aviation safety over the past thirty years; the internal organization of corporations; childhood malnutrition in Vietnam"; these stories "have nothing to do with the Internet."
Objection III: Johnson claims that he is, in fact, making an effort to engage with political philosophy—as evidence by his discussion of the limitations of direct democracy.
Objection IV: Johnson claims that I mischaracterize his position on New York's 311 service.
All four objections lead me to conclude that Johnson doesn't understand the substance of my critique. Let's begin with his sly conflation of Internet-centrism with what he dubs "naive techno-determinism.” As I state in the review's second paragraph, Internet-centrists have no problem acknowledging that the "Internet" can be deployed to do bad, evil things. The kind of naive determinism that views the “Internet” as a "positive force" and that Johnson seeks to distance himself from has nothing to do with Internet-centrism; it's a feature more commonly attributed to cyber-utopianism, as I clearly state at the very beginning of the review. That Johnson is not a starry-eyed techno-determinist doesn't make him less of an Internet-centrist.
What should we make of Johnson's question—on Page 120—of “What does the Internet want?” It's a question that he derives from Kevin Kelly's question - “What does technology want?”; both Kelly and Johnson assume that there is some kind of neat intellectual and practical coherence to these two ideas—a view that I vehemently oppose. This question does allow us to make the utopian/centrist distinction even sharper: An Internet-centrist asks the question: “What does the Internet want?” as if that question made sense. An Internet-utopian doesn't even ask that question, assuming that the Internet wants democracy and freedom. I don't know if Johnson is an Internet utopian but he is certainly an Internet-centrist. So while it's clear Objection I doesn't stand, let us still examine Johnson's answer:
“So what does the Internet want? It wants to lower the costs for creating and sharing information. The notion sounds unimpeachable when you phrase it like that, until you realize all the strange places that kind of affordance ultimately leads to. The Internet wants to breed algorithms that can execute thousands of financial transactions per minute, and it wants to disseminate the #occupywallstreet meme across the planet. The Internet “wants” both the Wall Street tycoons and the popular insurrection at its feet.”
I leave it to the reader to decide if this passage implies that there's a certain logic to “the Internet”; my reading is that Johnson does believe this while also arguing that the exact manifestations of that logic would be different in each and every context—which, if one closely looks at the Shirky quote about the “logic of the Internet” that I mention in the review, is very much in line with Internet-centric thinking. But it might be useful to step back and ask whether the very fact of bringing “the Internet” in our explanatory accounts is enhancing or impoverishing our understanding of the technological world that we inhabit. Are we gaining anything by lumping the algorithms used in high-frequency trading with a very different set of algorithms that Twitter uses to decide on its “popular trends” while using the sexy but highly elusive label of “the Internet” to do all that lumping? I don't think so—which is why I've been calling for a highly particularized approach to studying digital technologies—one that would treat each of them on their own terms without having to smuggle in some abstract, macro-level concept such as “the Internet” to smooth over the rough empirical and theoretical edges (As I point out in one of the footnotes, game theorist Ian Bogost has a related but clunky term for this method—he calls it “media microecology”).
Second, I'm well aware that Johnson sees the same spirit of “peer progressivism” that he believes to be at work in “the Internet” also sweeping through modern-day Vietnam, through Britain of the mid-1700s and through half a dozen other non-Internet contexts. But this is exactly what I have criticized him for! This tendency to travel back in time and rummage through other contexts and eras in search of some imaginary proto-Internet—which can then be repackaged in a sexy ideology like “peer progressivism”—is precisely what I find so problematic about Johnson's work in particular and Internet-centrism in general. (As I put it in my review: “Once the elusive logic of the Internet has been located, it is not uncommon to see Internet-centrists move to deflate its actual novelty.”). One cannot refute an accusation of Internet-centrism by proclaiming one's adherence to one of its key principles!
To understand the role that the notion of “the Internet” plays in Johnson's argument, consider a simple thought experiment. Remove the “Internet” and all its affiliated projects, from Kickstarter to Wikipedia, from a long list of examples of Johnson's “peer progressivism”– from childhood malnutrition in modern-day Vietnam to the hurdles of navigation in 18th century Britain—and see how far you'll go in convincing your readers that these examples amount to an original political philosophy—so original, in fact, that it deserves the fancy label of “peer progressivism.” If a book does come out of this, my bet is that you'll have to self-publish. Virtually every political idea that Johnson articulates in Future Perfect has been with us for decades—and it's precisely the vague, lazy and innovation-obsessed culture of our “Internet debate” that lets Johnson get away with inventing an original theory without doing his homework.
What allows Johnson to cut so many intellectual corners is his ability to capitalize on everyone's excitement about “the Internet.” He does so by selecting those pieces from its rich and ambiguous history that fit his overall “peer progressive” narrative while turning non-Internet history into a fishing expedition that supplies intellectual gravitas to the carefully selected Internet anecdotes that define what “peer progressivism” is all about. No wonder all his historical connections make sense: his argument is designed that way! On these grounds, I suspect that Objection II doesn't stand either.
As for Johnson's peeve that I unfairly attack him for not engaging with political philosophy, look no further than “peer progressivism.” What, one might ask, is new about this political ideology? According to Johnson, at least two things. First, its adherents believe that there are some areas of expertise where the public—or the crowd—are more knowledgeable than the experts. Second, “peer progressives”—unlike all those pre-peer progressives—don't have to choose between the state and the market; the two can co-exist, tapping into networks of crowd expertise along the way.
My problem lies not so much with the thrust of these two propositions; both are quite sensible. Rather, my problem is with the manner in which Johnson arrives at them, the fuzzy language that he deploys in the process, the revolutionary novelty that he ascribes to his own insights, and the carelessness with which he treats decades of serious thinking on this subject. Johnson, comfortably ensconced in his Internet-centric bubble, seems to sincerely believe that no one had ever thought about ways to make democratic politics more participatory before the onset of blogs, chats, and social networks. This, of course, is nonsense. The most unfortunate consequence of Johnson's project might be that, in his half-baked efforts to make a case for “peer progressivism,” he might undermine public support for more serious government reforms that are not as excited about “the Internet” but have developed sophisticated theories about involving crowds and networks in both deliberative and participatory processes.
So what does Johnson omit? Quite a few things, in fact. The idea that progressive politics can be combined with market-oriented and decentralized solutions was already in circulation—in the writings of Samuel Bowles and Herbert Gintis but also of Joshua Cohen and Charles Sabel—by the end of the last century (and, more recently, in the work of scholars like Archon Fung). Here are, for example, Cohen and Sabel—writing in 1997 -- on how governments can become more participatory and profit from more decentralized ways of knowledge aggregation: “Instead of seeking to solve problems, the agencies [would] see their task as reducing the costs of information faced by different problem-solvers: helping them to determine which deliberative bodies are similarly situated, what projects those bodies are pursuing, and what modifications of those projects might be needed under local conditions...”Here is Cohen in another essay written at the time:“The availability of alternative methods of problem-solving imposes on legislatures a greater burden in justifying their own direct efforts: They must explicitly make the case that the benefits of those efforts suffice to overcome the advantages of direct-deliberative solutions.”
These two quotes—written a good decade before buzzwords like “open government” hijacked the public conversation—pack more reform wisdom than Johnson's entire book. But Johnson prefers to ignore virtually everything written on this subject—the faux novelty of the Internet licenses him to such frivolity. So he completely ignores Josiah Ober, who has made a fascinating use of Hayek, game theory and political philosophy to argue that the democracy of classical Athens was so effective because it deployed highly innovative and decentralized schemes of aggregating the knowledge of its citizens. Nor does he mention a new strand of scholarship on the political implications of “cognitive diversity”—best exemplified by the work of Jon Elster and Helene Landemore—which has advanced sophisticated, context-sensitive arguments about ways to bring more diverse voices into democratic policy-making. All these efforts start from where reform proposals ought to start: they acknowledge the complexity of the problem that they are trying to tackle and only then do they work their way to their preferred solution. This is not the case with Johnson, who starts with his preferred solution—the “Internet”—and then searches for problems that it can help him solve. Yes, it's nice to see him quote the Federalist Papers but I hope he's at least aware that this is hardly the latest word on innovations in participatory governance. So Proposition III doesn't stand either.
Finally, did I misread Johnson's treatment of 311? I don't think so. My claim is that the reason why Johnson is fascinated by 311 is because he views it through the Internet-centric lens of Wikipedia and other seminal Internet projects. That lens assumes an argument that goes something like this: encyclopedias used to be centralized and now they are decentralized. Likewise, tip-reporting systems used to be centralized and now they are decentralized, with hierarchies being replaced by peer networks. But is it actually true? Were tip-reporting systems ever centralized? Was there ever a competent Big Brother, perhaps in the form of some omniscient inspector—that proverbial “expert” hated by Internet-centrists—who was tasked with tracking all of New York's problems and who now, thankfully, has been replaced by the crowds? Perhaps, there was such an expert– long time ago— but involving crowds in the process of reporting incidents and crimes has a very long history that is not very relevant to the 311 project. The 311 project was not about replacing centralized experts with decentralized crowds; it was about turning a slew of previously decentralized tip-reporting systems and hotlines—systems that already relied on crowds—into one highly centralized system.
Is this what Johnson means when he writes that “311 is not a purely decentralized system” and that “top-down element may be inevitable”? No, it isn't: what he means here is that while under 311, the inputs—the tips—might still be coming from decentralized sources, it still takes a centralized system—some city agency—to deal with the reported problem. But it would be silly to think otherwise—not unless we expect the New Yorkers to bypass various city agencies and fix problems on their own. Johnson completely misses what's novel about the story he is discussing– that the 311 hotline works because it centralized many different hotlines under one roof—and focuses on that part of the story which fits his Internet-centric view of “peer progressivism”—namely that 311 works because many people report tips to it in much the same way that many people edit Wikipedia. But to believe this is to miss the fact many people were already reporting tips to New York's various hotline systems even before 311! Thus, I don't think that Objection IV should be allowed to stand either.
Now, there's one point I must concede to Johnson. I fully agree with him when he writes that "the point I tried to make explicit in Future Perfect is one that I’ve been implicitly making for more than a decade now." This hasn't escaped my attention; the original version of my review ran at 8000 words and contained a long section situating Future Perfect in Johnson's entire oeuvre— a section that had to be cut for space reasons. (Given that I managed to keep a reference to one of his little-known essays from 2005, it is a bit unfair to accuse me of not being familiar with his work.) But I do agree with the thrust of Johnson's remarks: he has, in fact, managed to write yet another book about the same subject—let's just call it “buzz”– this time invoking the notion of the "Internet" to justify the publication. In fact, a close analysis of the source material for Future Perfect reveals that it's based on many essays and blog posts that Johnson had penned before the idea of “peer progressivism” took hold of his imagination.
This is the same trick Johnson pulled with his turn to "innovation" in his previous book, Where Good Ideas Come From, and with neuroscience and sociobiology in his earlier works. Fortunately, we know how Johnson goes about deciding what specific intellectual form to give to this “buzz” in his future projects. Emphasizing the useful feedback that Johnson got from speaking to the clients booked through his agency, Bill Leigh, his speaking agent, recently told New York magazine that Johnson “wanted to take his book sales to the next level...Out of those conversations [with clients] came his decision to slant his material with a particular innovation feel to it.” Where good ideas come from still remains a mystery; where lucrative ideas come from everybody knows. It's surprising that it has taken Johnson so long to discover one such lucrative idea in “the Internet.”