Internet activism through social networks like Facebook and Twitter clearly can work. Activists played important roles in bringing down dictators in the Arab world, stopping the Stop Online Piracy Act (SOPA) in Congress and electing Barack Obama—just to name a few examples. But how much did the Internet matter in making these watershed events possible? How effective is it likely to be in the future? And how would we measure whether activism “works” for society—not just the activists?

“Technology,” quipped the British architect Cedric Price, “is the answer, but what was the question?” Evgeny Morozov uses this pregnant observation to attack the “unthinking admiration of technology as panacea” in his iconoclastic 2010 book The Net Delusion: The Dark Side of Internet Freedom. Morozov is, of course, right that too many advocates of “Internet freedom” consider the Internet the solution to every problem. If all you have is a hammer, nails abound!

Morozov rejects “cyber-utopianism,” the assumption that the Internet will necessarily make “modern authoritarian regimes more open, more participatory, more decentralized, and, all along, more conducive to democracy.” Instead, he demands cyber-realism—a “more down-to-earth approach to policymaking in the digital age.” But is Morozov right to be quite so skeptical of digital activism?

He is certainly correct that there is much more at play in these conflicts than the Internet, and that we cannot understand the role of digital activism in each situation without studying the various other historical, cultural, and economic factors at work. (For instance, rising bread prices played a key role in both Tunisia and Egypt—just as in the French Revolution.) We in the West who cheer Internet freedom all too easily overestimate the impact of tools like Twitter in countries such as Iran, where the number of people with effective access to digital media remains small.

Certainly, the Internet isn’t a one-way ticket to democracy. Internet-fueled revolutions succeeded in Egypt, Tunisia, and Yemen, but Iran’s “Green” or “Twitter” Revolution in 2009 failed to unseat President Mahmoud Ahmadinejad after a disputed election. The Internet has certainly made the violence of Syrian President Bashar al-Asad’s crackdown more visible, but it has not (yet) ended the Syrian Ba’ath party’s 46-year monopoly on power—nor shaken China’s Communist Party.

This is, as Morozov notes, partly because governments like Iran’s and China’s have successfully used social networking tools against digital activists. They have monitored dissidents to a degree unimaginable in the pre-Internet era, by closely watching social networks and forcing companies that operate them (or email or search engines) to turn over data. Such countries have also countered the messages of liberalism with messages of nationalism and resistance to outside interference. Illustrating how complex these debates can be, the Syrian government has branded its opponents as enemies of secularism and supporters—or at least dupes—of Islamic terrorism.

That Egypt’s Facebook revolution has, at least in the short term, served to replace a secular (if repressive) regime with a government dominated by the Muslim Brotherhood underscores Morozov’s point: We cannot assume that everything the Internet touches turns to democratic, pluralistic gold. Our assumption that everyone wants to be “liberated” is no less naïve and simplistic today as part of the “Internet Freedom Agenda” than it was when the same idea drove President George W. Bush’s neoconservative advisors to “liberate” Iraq—saying things like “We’re an empire now, and when we act, we create our own reality.”

But is Morozov too pessimistic? In the afterward to the paperback edition of The Net Delusion, written in October 2011, Morozov rejects the “cyber-pessimist” label used by many of his critics and instead re-summarizes his views as cyber-agnosticism: “an unyielding refusal to take any stance on the question of whether the Internet is a tool of liberation or repression.” Few would disagree that we must evaluate digital activism not “solely on the efficacy with which it achieves the goals it sets for itself” but on its “ecological effect on the broader political culture that produces it.”

Indeed, those who share what philosopher Thomas Sowell called the “constrained vision” of the perfectibility of man and society will find much to admire in Morozov’s rejection of “utopian social engineering—ambitious, ambiguous, and often highly abstract attempts to remake the world according to some grand plan.” But Morozov’s work is only the beginning of a self-conscious Internet realist enterprise. As Adam Thierer notes in his review of The Net Delusion, the problem is that Morozov offers little guidance on how to apply Internet realism in practice. Most of all, Morozov’s normative judgments rest on descriptive predictions about how Internet activism is likely to evolve. Having such a conceptual framework, even if not perfectly predictive, is essential to assessing whether digital activism “works” (normatively).

Fortunately, we need not build such a framework from scratch. Friedrich Hayek once said that “The curious”—meaning both unique and relentlessly inquisitive—“task of economics is to demonstrate to men how little they understand about what they imagine they can design.” This rigorous skepticism has already been fused with political science into the “Public Choice” school—“politics without romance,” as the Nobel Prize winner James Buchanan artfully put it.

Public Choice theorist Mancur Olson offers a good starting point. In his seminal 1965 work The Logic of Collective Action, he tested the following seemingly straightforward hypothesis:

[I]f the members of some group have a common interest or objective, and if they would all be better off if that objective were achieved, [then] the individuals in that group would, if they were rational and self-interested, act to achieve that objective.

Yet if this were true, activism would be so easy as to be unnecessary: whatever problems were in our interest to fix, already would have been fixed—or, at least, there would be activist movements with strength proportionate to the popular recognition of the problem. But of course, in the real world, political “markets” are not perfect—as any protester in Syria or political organizer in the United States could tell you. Why not?

Applying the logic of economics, Olson explained the central problem of “collective action”:

[U]nless the number of individuals in a group is quite small, or unless there is coercion or some other special device to make individuals act in their common interest, rational, self-interested individuals will not act to achieve their common or group interests.

What is activism but an attempt to “make individuals act in their common interest?” Yet simply rallying large numbers of people to a cause often fails. Ever wonder why American food manufacturers use high fructose corn syrup rather than sugar? It’s because the tiny group of sugar cane growers in Florida, and the even less efficient sugar beet farmers in the Midwest, have captured agricultural policy—using import tariffs to raise the domestic price of sugar above the price of corn syrup. Since nearly all Americans bore part of an estimated $3.86 billion in higher sugar prices last year, why didn’t they—we—rally long ago to stop this naked transfer of wealth? Because large groups (like all consumers) have great difficulty organizing, while tiny groups can often out-lobby them. Activism therefore often fails to materialize at all.

Can the Internet help? In a 2003 article, Arthur Lupia and Gisela Sin, political scientists at the University of Michigan, began from the premise that “technology changes longstanding expectations about what people can learn about each other”—thus challenging some of Olson’s key assumptions. Most importantly, they find that “evolving technologies can transform groups once crippled by large numbers into groups capable of collective success.” Yet they also conclude that digital technologies can actually make activism harder for some groups. This ambiguous result led them to a call for further research that sounds remarkably like Morozov’s call for cyber-realism:

In sum, our work shows the importance of being more specific about the role of communication in theories of collective action. Without such specificity, it is difficult to understand whether and how technological advances that change communicative incentives and opportunities alter who joins with whom.

So, let’s take the most obvious specific example: SOPA. Copyright critics succeeded spectacularly in using social media to organize the general public against special interests—as this graphic makes clear.

How? And why did they succeed in stopping SOPA after failing to stop so many earlier expansions of copyright?

Obviously, social media lower organizational costs, especially of recruiting members. Less obvious is that, as Lupia and Sin explain, social media increase noticeability: “members’ ability to notice each other’s actions.” Even in 2003, there was little way to tell whether your friends actually followed through when you asked them to help join a cause. But today, it’s easy to encourage them to re-share material on Facebook or Twitter—and to “notice” whether they’ve done so. As political activist Patrick Ruffini recently told NPR:

We’ve long known that the most powerful thing in determining how you’ll be influenced to vote is a recommendation from a friend… And the ability to see in your Facebook news stream somebody taking action on behalf of a campaign who’s a trusted connection of yours is something that everybody who’s going to be active this year is going to want to look to harness.

Similarly, Lupia and Sin concluded that “evolving technologies convert venues where individual actions are effectively anonymous into settings where people can hold each other accountable for their actions.” While this has an obvious dark side (easier government surveillance), it is also critical to understanding why social media facilitate activism.

Social media allows members of large groups—think Twitter followers—to be “continuously bombarded with propaganda about the worthiness of the [cause],” creating “social pressures not entirely unlike those that can be generated in a face-to face group.” What better way to replicate face-to-face pressures than by replacing your Facebook profile photo with a “STOP SOPA” logo, as many did?

Jerry Brito, an Internet policy scholar at the Mercatus Center, celebrates the defeat of SOPA as an example of the Internet facilitating effective collective action while also addressing the related public choice problem of “rational ignorance” (it’s not worth the time for most voters to learn about the sugar subsidies, copyright over-reach, etc.). Yet he argues that it’s

unlikely we’ll see [such outpouring of public concern] very often. The reason is that the SOPA issue has a unique set of characteristics that allowed it to take advantage of the Internet’s latent power to overcome rational ignorance and to facilitate collective action by large groups.

Specifically, SOPA (a) was a simple issue (at least in its superficial, but effective, rhetoric: “copyright as censorship”); (b) rallied diverse free speech interests; and (c) appealed to corporate interests, who helped amplify grassroots activism. These are all sound points. But as with Morozov’s predictions about the dark side of Internet activism, it’s hard to say how the balance will tip.

The Internet clearly empowers large, dispersed groups (like dedicated Internet users) to organize against small but concentrated interests. As anyone who works in technology policy in Washington can attest, SOPA’s implosion made Congress more cautious—at least about Internet regulation, where fear of a digital activist backlash is greatest. Yet the House of Representatives just passed cybersecurity legislation (CISPA) despite efforts by some activists to brand it “Son of SOPA.” Does this mean Internet activism “failed?” Have politicians already recalibrated their approach to do essentially what they would have done anyway? To some extent, yes. But much more clear is that the victory of digital activists against SOPA led to more careful deliberation of CISPA, informed by independent experts, than with SOPA. The progress is even greater when compared to the attempt last summer by SOPA’s architect, Representative Lamar Smith (R-TX), to ram through a sweeping mandate that broadband providers track Internet users—and turn that information over to government. Only time will tell how the balance of power in these debates will tip.

And even then, it’s not clear we’ll learn the right lessons from whatever happens. The core point of this essay is that we have a limited analytical tool kit for predicting whether we will see more protests like that against SOPA—and whether they will work—just as we poorly understand how to support long-term democratization. Answering Morozov’s normative questions about digital activism requires better descriptive models—such as those offered by Public Choice, integrated with the historical, cultural, and social analysis Morozov uses.

Four points bear making to better understand digital activism—and to help steer its use for good. First, the same tools and methods of activism used to influence, change, or overthrow governments are used every day to rally support for philanthropic, philosophical, or cultural causes, and to discipline companies—from our favorite coffee shop to giant multinationals. These examples may teach us more about activism because there are more frequent and less complex than revolutions.

Second, activism works largely by imposing reputational costs on its targets—be they autocrats whose thuggery is broadcast on YouTube, Congressmen who fear being labeled sponsors of censorship, or local businesses who want to avoid negative blog posts or Yelp reviews. But reputation also disciplines activist groups—addressing one of the Morozov’s key concerns. While economists have long studied reputation markets, we still do not fully understand how the Internet has changed them. Santa Clara Law Professor Eric Goldman offers a conceptual framework in an his essay, the “Regulation of Reputational Information,” in the free collection of essays TechFreedom published last year: The Next Digital Decade: Essays on the Future of the Internet (including an essay by Morozov summarizing his arguments). Goldman’s key insight was that reputation systems (like customer review sites or eBay’s ratings) on the Internet serve as a “secondary invisible hand” helping to guide Adam Smith’s “primary invisible hand” of self-interest by “helping consumers make better decisions about vendors.” This was true even before the Internet, but online reputation markets deliver information much faster and more cheaply than ever before. Such rating systems, in turn, are themselves subject to reputation markets—a third invisible hand.

Ultimately, the efficiency of information markets depends on the availability of useable data. Thus, through “Smart Disclosure,” government can actually help provide raw material for activism. As Brito argues, lowering the cost of information helps to overcome the rational ignorance of voters. The example of Google’s Transparency Report (showing how often governments demand that Google turn over user data or take down content) illustrates both how private companies can bring reputational pressure to bear on governments and how they can enhance their own reputations by being more transparent about dealings with government.

Third, activists, working largely through reputation markets, are already changing the way Internet companies relate to their users. Most notably, in 2006 Facebook launched News Feed, a rolling ticker of everything their friends shared. Ironically, users who considered this “Facebook stalking” used Facebook to rally against the site—taking advantage of the increased noticeability of their message made possible by News Feed itself. Facebook has responded to this and other missteps by developing a process for soliciting user comment and votes on changes to its rules. With over two million Likes, the “Facebook Site Governance” page may be the beginning, however modest, of the kind of participatory self-government of social networks Rebecca MacKinnon calls for in her new book Consent of the Networked: the Worldwide Struggle for Internet Freedom.

Fourth, perhaps the most important policy debate flowing out of this issue is that over deputizing online intermediaries. As MacKinnon rightly notes, making “intermediaries… liable for their users’ and customers’ behavior… is precisely the legal mechanism that enables an unaccountable government to delegate the bulk of censorship and surveillance to the private sector.” However noble the goals—protecting children, defending copyrights, promoting cybersecurity or punishing defamation—intermediary liability bolsters the power of oppressive governments and encourages companies to censor or not to provide open fora in the first place. Either way, governments can indirectly hamstring digital activism. Avoiding such indirect “architectural censorship” requires a better understanding of digital media.

In the end, understanding digital activism may also help governments manipulate the Internet in precisely the ways Morozov worries about—just as understanding public choice can make it easier to craft electoral choice mechanisms to favor incumbents. But the same is true of medicine and many other branches of science. The Internet may not necessarily make the world a better place in every way, but the more we understand how it changes our relationships with each other, the better equipped we will be to steer its evolution in more humane directions.

As a lifelong political activist I would like to believe that “digital activism” had tremendous impact and leverage for change. However, as someone who has built his career upon communicating the “magic” of technology to the public on behalf of the leading companies in Silicon Valley, I remain skeptical concerning the democratizing impact of the Net through its newest expression, the social network. It’s my opinion that social networking, as an activist tool, is being vastly oversold. However, this is not without precedent or purpose: Great IPO fortunes depend on this popular misconception.

Given my background, I consider myself inoculated from charges of Luddism or “cyber-pessimism” – a pejorative that I also reject for Mr. Morozov and others who have liberated themselves from what I call “the cult of tech.” Simply defined, the cult of tech is the nexus of technology companies, telecom service providers, tech think-tankers and assorted digerati that derive their livelihood from promoting a digital panacea. These combined interests exert undue influence over an often befuddled popular media struggling to keep up with the “magic” of new tech offerings. For example, the cult of tech jumped at the marketing opportunity to brand an indigenous anti-authoritarian uprising in Iran as the “Twitter Revolution” with scant evidence of the application’s actual impact, negative or positive.

Technology always cuts two ways. Although the personal computer provided empowerment and creative liberation for individuals, and the Internet gave us access to information, they came at a cost. Experiences over the Net require a service provider to mediate connections amongst us. The early 90s freewheeling Internet with hundreds of independent ISPs has devolved into less than a handful of huge players. This new concentration of power, whether as a public or private entity, is cause for concern. Since centralized power is inherently non-democratic, these monolithic network entities are not inclined to liberate humanity. Therefore utopians better think twice if they are depending on the Net to promulgate democracy and freedom. For example:

Governments of the Industrial World, you weary giants of flesh and steel, I come from Cyberspace, the new home of Mind. On behalf of the future, I ask you of the past to leave us alone. You are not welcome among us. You have no sovereignty where we gather.

John Perry Barlow’s quote is Utopian indeed, a poetic touch, but unfortunately the last sentence is magical thinking. In reality, cyberspace, and the au courant “Cloud,” are not ethereal things – they are physical assets that depend upon tangible presence and resources. Just as we don’t have minds without brains in which consciousness can reside, cyberspace exists in data centers and network switching racks in real locations, owned by huge corporations and governments. Those in control of these physical assets rule over the network and the flow of information. Although they allow for variable amounts of chaos depending upon the cultural context, they are fundamentally authoritarian in structure.

Does social media make any kind of impact in molding opinion? Yes. As with all media types it serves both for good and evil, truth and lies. However, Mr. Morozov and I are on the same page in the belief that cultural and physical realities are the determining factors far more than “friending” a cause. Whether we like it or not, bullets and batons are more potent than bytes. Reality generally trumps virtuality.

In the opening paragraph of Mr. Szoka’s essay, he headlines three “successes” attributable to social network activism – the Obama election, the North African uprisings, and SOPA’s defeat. Below I will argue that all three are actually perfect examples of the medium’s failure to deliver change.

The efficacy of the network as a tool of activism is best examined in three different contexts:

1. Democratic states

2. Authoritarian states

3. Commercial “states”

1. SOCIAL MEDIA IN DEMOCRATIC STATES

THE 2008 ELECTION

Although the Obama campaign was adept at fundraising, messaging their base, and organizing volunteers, once elected, the power center for Obama dramatically shifted. It was not longer the network of activists and “the professional left” who mattered; it was those within earshot of the White House, those who lunched and dined with the people who had the president’s ear. Pre-election virtual lost out to post-election proximity and presence. Polling indicated that the “public option” was the leading choice for a new healthcare model. Bloggers blogged, but to little effect as a few well-funded senators derailed a social network of millions of vocal activists. In elections, even a well-organized social media effort can be undone by one individual tampering with digital voting machines.

ONLINE PETITIONS

The popularity of online petitions is astonishing. Yet they are served to no one and have no force of law. The purpose of their existence is simply data collection for issue-oriented fundraisers and spammers. Similarly, millions can “like” a cause on Facebook and have no impact on reality. These devices are akin to “close door” buttons on elevators and “walk” buttons at pedestrian crossings. They have the effect of making people feel like they are taking action, but in reality, the effect is nothing beyond self-satisfaction. It might actually deter real political action since people “already gave” at the web browser.

OCCUPY

Although there had been progressive blogging and activism of note since the advent of MoveOn.org and Daily Kos, it wasn’t until activists stopped merely talking to themselves and made their continuing presence felt on Wall Street that they became hard to ignore. Reality again trumped virtuality.

SOPA

Many “netizen” activists seem to be taking a victory lap on the deferment of SOPA legislation. However, when we look at the players, it wasn’t little the guy versus the omnipotent. It was two powerful entities at odds over the rulebook for revenues from intellectual property. It was Silicon Valley billionaires versus Hollywood billionaires and Hollywood overplayed its hand. Next time the studios and companies such as Google, Facebook, Twitter, and Microsoft will make a deal with content companies and exclude the rest of us. And make no mistake – they will be back.

CISPA

Vitally more important than SOPA is CISPA. We see organizations as politically divergent as the ACLU and Cato showing up to fight this heinous violation of the Constitution – but neither the technology giants nor their considerable lobbying/political-action surrogates are working for freedom this time.

2. SOCIAL MEDIA IN AUTOCRATIC STATES

If you ask yourself what’s new and different about the recent events in North Africa and the Arab world, the answer would be “not much.” The pattern of autocracy repeats over time; as Mark Twain said, “history rhymes.” A corrupt leader is replaced by a reformist in a revolution. After a few decades, the newer leader succumbs to corruption and begins to siphon off the national wealth at the expense of everyone else, just like the old leader, and starts positioning his heirs to succeed him. The populace grows poorer and more enraged and then overthrows the government once again. The spark is always different, but the rage is the same. This time the spark was literally set by a self-immolation in Tunisia, not by Facebook.

In Libya, a nation had divided in two and then NATO intervened on the side it favored to win. It was fierce town-by-town fighting determining the course, not social media. Despite blood being spilled, not much democracy has emerged from Libya. Egypt remains in a state of chaos, and most likely the Egyptian Army is being restrained by its multibillion dollar benefactor, the U.S. government. It was also made clear during the Egyptian uprising that the Internet Kill Switch was a playable card for the government.

At other times, the technology winds up being a net negative for organizers. Iran’s 2009 uprising is a frightening example of what can go wrong. Hailed as the Twitter Revolution, social media gave a false sense of empowerment to the protestors. Yes, the world was watching, but so were the secret police. Worse, the intermediators who controlled the wireless and the Net were government entities and enabled by western technology companies. The government used these networks to track down, jail, torture, and kill those striking for human rights. The world mistook watching for taking action, which was not possible.

In confronting authoritarian situations there can be technical workarounds to protect identity and bypass filtering, yet there is never a technical measure that doesn’t eventually have a counter measure. We can also be certain that well-financed state authorities will have early access to the most expensive and capable tools to monitor, filter, and eliminate opposing opinions – and these are supplied by the same Silicon Valley companies whose leaders who glibly and regularly espouse “freedom.”

3. SOCIAL MEDIA IN COMMERCIAL “STATES”

Finally we arrive at the purpose of social networks…and some success in activism. However, it is only what could be called “consumer activism.” Social media applications were designed as advertising platforms and consumer intelligence-gathering channels. Using increasingly powerful software tools, marketers scan for opinions and even complaints about their offerings. In the commercial context, the feedback of social networks is an asset – it is data collection for “business intelligence.” Consumer activism can have an effect here because the criticism is valuable to marketers in their efforts to sell more stuff, even if at times criticism is uncomfortable.

Now consider this sales proposition from a social networking company: If our application can overthrow a government, certainly it can help you sell more fizzy sugar water. How many businesses decided they needed a Twitter strategy after seeing the “Twitter Revolution” taking place in Iran? Why would companies such as Facebook and Twitter not want to capitalize on the misconception that their tools powered social change when it increases their perceived value, especially with potential retail share purchasers at IPO time?

Inevitably, the mix of commercial and political brings us to China. Defying all the pre-conceived nonsense that commercial liberalization and political freedoms are tied together, China blesses engagement with the web and social media to market products, but actively blocks the Net as a means of expressing dissent or diversity of opinion. In these efforts they are being enabled by best of Silicon Valley. The cult of tech preaches freedom, but happily sells the tools of oppression.

Conclusion

Overall, I am by nature an optimist. However, the social network as it is presently constituted is not a serious tool for substantive social change. It is concentrated, centralized and controlled. At best, in the autocratic context, it is akin to China’s “Democracy Wall.” In the democratic context, it is similarly a way to vent, and perhaps organize, but as of yet not much more. However, if you are selling widgets, the social network looks more promising.

Berin Szoka is right to argue that while the Internet brings new dimensions and power to activism, we must not be naïve about the power of networked technologies. It is important to unpack the factors behind successful—and unsuccessful—online activism. Examples of both abound. Internet connectivity and widespread social media adoption do not on their own guarantee activism’s success. The Internet is not some sort of automatic “freedom juice.”

Success or failure of digital activism depends on a plethora of variables—economic, cultural, religious, commercial, political, personal, and accidents of history. In his seminal book The Digital Origins of Dictatorship and Democracy: Information Technology and Political Islam, Philip N. Howard, a professor at the University of Washington and expert on technology and political change in the Islamic world, concludes that while the Internet and mobile technologies do not cause change, change is unlikely to happen without sufficient mobile and Internet penetration.[1] Indeed, the two Arab countries in which dictators were deposed without civil war in 2011 were Tunisia and Egypt—both of which have relatively high rates of Internet penetration and social media use compared to many other parts of the Middle East and North Africa. However, as I discuss at some length in my book Consent of the Networked: The Worldwide Struggle for Internet Freedom, the revolutions of Tunisia and Egypt did not spring immaculately from Twitter and Facebook. Movements for political change in these countries developed and matured over the course of a decade; then when the right moment came activists were in a position to take advantage of them. Activists experimented with networked technologies, honed their messages over time, built support networks, and generally worked to use Internet and mobile platforms to their maximum advantage. They also spent a decade building offline relationships both nationally and regionally and honing offline protest skills. The revolutions’ successes in Tunisia and Egypt, as Szoka rightly points out, had much to do with widespread economic grievances and anger over state corruption.

Another factor was the relative lack of sectarian divisions in Egypt and Tunisia as compared to other countries in the region. This contrasts sharply with Bahrain which also boasts deep Internet penetration and widespread social media usage, but whose society is torn asunder by a deep sectarian divide between majority Shiites and Sunni political elites. This divide has enabled the ruling Al Khalifa family to suppress dissent violently and with impunity—aided by other geopolitical factors including support from neighboring Saudi Arabia, which considers Sunni activism on its doorstep to be a dangerous sign of Iranian political meddling. Then there is the presence in Bahrain of the U.S. Seventh Fleet, a geopolitical rather than a technological reality that makes rapid political change in Bahrain all the less likely. In Syria, Internet penetration was much more shallow and online communities much weaker to begin with. This combined with a sharp sectarian divide has meant that while activists have been able to use the Internet to get information out to the world about the Assad regime’s atrocities against its own people, conventional geopolitics—not new media—will be the decisive factor in deciding when and how Assad will fall from power.

Success or failure of digital activism in authoritarian states also depends on the regime’s technological capacity, skill, foresight, and planning. As I describe in detail in the third chapter of my book, the Chinese government took the Internet seriously as both a political threat and economic opportunity from the moment it began to allow commercial Internet services in the mid-1990s. The Chinese government built the world’s most sophisticated system of filtering and blocking for overseas websites, including most famously most Google-owned services, Facebook, and Twitter. At the same time, the government encouraged the development of a robust domestic Internet and telecommunications industry so that Chinese technology users can enjoy an abundant variety of domestically run social media platforms, online information services, Internet and mobile platforms, and devices produced by Chinese companies. By imposing strong political and legal liability on Internet intermediaries, the government forced companies—many financed by Western capital—not only to foot the bill for much of the regime’s censorship and surveillance needs, but to do much of the actual work.

Online activism still does occur in China, but due to multiple layers of censorship and surveillance, activism’s successes have for the most part been local, presenting minimal threat to the power of the central government and Communist Party. Users of the Chinese Twitter-like social networking platform Weibo have ruined the careers of local and provincial officials by exposing their corruption. Chinese “netizens,” as they like to call themselves, have also called attention to specific errors or incompetencies of specific parts of the bureaucracy, which the central government has then moved to fix—which in many ways boosts the central government’s power and credibility as compared to local governments or specific ministers seeking to develop independent power bases. To date, activists who have tried to use social media to build national movements for systemic political change have consistently gone to jail or been placed under house arrest, their supporters and friends often harassed and threatened with loss of jobs and educational opportunities even if they have not technically committed any crime by Chinese law. The case of the blind activist Chen Guangcheng may or may not serve as a watershed moment for Chinese activism—it remains too early to tell. But if it does, the reasons for digital activism’s success in China will have as much to do with offline domestic and international factors as with anything technological: a leadership crisis at the top of the Communist Party precipitated by the downfall of the power-hungry Chongqing Party Secretary Bo Xilai; plus specific developments not only in the U.S.–China diplomatic relationship but also U.S. domestic partisan politics, which Chen’s supporters have taken skillful advantage of, using social media of course.

To complicate matters further, the Internet itself—its technical architecture as well as the regulatory constraints shaping what people in different places can and cannot do with it—is a variable. We cannot treat the Internet as constant—either across geographical space or across time—in our calculations about the success of online activism. In Thailand, for example, the relative weakness of online activism is the result in no small part of heavy liability placed by national laws on Internet intermediaries. Thailand’s Computer Crimes Act holds Internet service providers and website operators legally responsible for the activities of their users. This, combined with an antiquated lèse majesté law banning insulting comments about the king, has resulted in the arrest of people involved with running activist and opposition websites and made it difficult for online activism to achieve critical mass.[2] In Azerbaijan, a blanket surveillance system imposed by the government on Internet and mobile network operators is combined in a politically insidious manner with media manipulation, arrests, and intimidation of online activists. This has, in the words of Internet scholars Katy Pearce and Sarah Kendzior, who recently concluded a multi-year study of the Azerbaijani Internet, “successfully dissuaded frequent Internet users from supporting protest and average Internet users from using social media for political purposes.”[3]

Thus while the Internet often empowers activism, it is also used in many parts of the world as an insidious extension of state power—sometimes with the direct collaboration of companies seeking market access; sometimes much more indirectly due to the fact that Internet and mobile companies are conduits and repositories of vast amounts of citizens’ personal data, and also make commercial decisions that have a profound impact on people’s digital lives and identities.

Back home in the United States, this is precisely why recent political movements against legislation like SOPA and CISPA that Szoka describes are so important. SOPA would not only have built a “great firewall of America” but would have imposed liability on Internet intermediaries in a manner that was ripe for political as well as commercial abuse. CISPA, in the name of securing America’s networks, will also legalize and institutionalize mechanisms of government access to citizens’ private communications that lack accountability, inviting abuse against which citizens will have no meaningful recourse as the legislation is currently written.

Whether the Internet remains conducive to political activism, or with liberal democracy for that matter, is by no means guaranteed. We face a virtuous or vicious cycle depending on whether you are an optimist or a pessimist: Activism is urgently required—nationally and globally—to ensure that the Internet remains compatible with activism.

Notes

[1] Philip N. Howard, The Digital Origins of Dictatorship and Democracy: Information Technology and Political Islam (New York: Oxford University Press, 2010).

[2] Danny O’Brien, “Verdict Postponed in Landmark Thai Internet Freedom Case,” Committee to Protect Journalists, April 30, 2012.

[3] Katy E. Pearce & Sarah Kendzior, “Networked Authoritarianism and Social Media in Azerbaijan,” Journal of Communication (2012) ISSN 0021-9916.

Berin Szoka’s essay asks the right question – Can Internet activism work? – and wisely reaches out to political theory for an answer. It is the lack of such a realistic theory that leads to ungrounded optimism or pessimism about the new medium. Mr. Szoka’s analysis turns on Mancur Olson’s insight that concentrated interest groups have lower information and organization costs than diffuse ones. As a result, concentrated groups pursuing narrow and selfish interests can gain benefits from the state at the expense of more diffuse groups pursuing broader goals, like economic growth or good education for children.

Mr. Szoka is importantly correct that the Internet can help redress the balance between special and more encompassing interests by reducing the cost of accessing information. Such reduction redounds to the advantage of diffuse groups more than concentrated groups because reduced costs can temper the former groups’ larger problems of coordination. The potential of better information technologies to empower the relatively powerless can be best gauged by the reaction of the powerful, like authoritarian leaders. Authoritarian leaders together with their band of supporters themselves constitute a potent special interest group that lives off the state. Fearing that more information may enable citizens to better organize to attack their privileges, they have tried to restrict emerging technologies of free communication as long as these technologies have been around. Today, Iran and Syria shut down the internet when they are threatened. Years ago, many nations refused to license the printing press.

In a democratic state like ours, the primary interest groups are not authoritarian cliques, but private actors, like public sector unions and trade associations, which have the leverage to pressure politicians to use public power on their behalf. And, like authoritarian leaders, such groups are desperate to avoid transparency to retain their benefits. A case in point is the opposition of teachers’ unions to publishing evaluations of schools and teachers on the Internet. And many interest groups have tried to prevent laws requiring Internet disclosure of campaign contributions.

But the Internet does more to help diffuse groups than merely to improve access to information. Along with other new technologies of the digital age, it improves the quality of information—a development that also benefits the pursuit of encompassing interests. We know already that many individuals act not only in their narrow self-interest, but also for common goals, like faster growth and better education. Thus, political scientists have consistently found that most people vote based on mixture of what they think is good for the country and what they think is good for themselves.

Yet the results of policies are contestable. And it is often hard for citizens who are distracted by many enterprises more interesting than politics to find good information about policies’ likely outcomes. Most people also have a better intuitive sense of how policies will affect their short-term interests than the long-term interests of society, even if the long-term effects may be of great personal as well as social benefit. Thus, citizens are less likely to organize and vote for encompassing interests when they are unsure of the policy instruments to achieve them. But the more easily citizens can discover policies that likely work, the more likely are they to organize and vote for such common goals.

The Internet provides an important mechanism of such social discovery. Because of the greater space and interconnections that the Internet makes available, web-based media, like blogs, can be dispersed and specialized and yet connected with the wider world. As a result of this more decentralized and competitive media, the web generates both more innovative policy ideas and better explanations of policy than were available when mainstream media dominated the flow of political discussion. Specialized blogs can address policy issues with a level of sophistication and depth that the mainstream media never could achieve. In my own field of law, the web hosts scores of widely read blogs, often with a quite particular focus, from tax law to contracts, from empirical studies to law and economics.

Such specialized media improve our knowledge of the policy world by energizing and disciplining those who believe they have knowledge. First, specialized media deepen substantive knowledge. Experts participating in specialized media that are closely followed by their peers gain greater incentives to explain policy effects carefully and with nuance. It is not that experts will all agree, but they will try to avoid obvious mistakes and respond to counterarguments. In this world of intense scrutiny, scholars will suffer blows to their reputation if they fail to engage in careful self-monitoring.

Second, as Glenn Reynolds has noted, more mainstream media then takes up the best of the blogs. For instance, Economix, the economics blog of the New York Times, regularly reports on economic ideas from academics and other experts—ideas often at odds with its editorial page. The reporter’s beat is as much the online world as the physical world. Thus, the Internet has become a vast funnel of information that allows specialized but politically salient ideas to course through the wider world in a form in which they can be more readily understood.

The very nature of the Internet’s interconnections encourages a grounding of argument in facts. The reporter links to the academic article he cites, just as the academic may link to the data set on which she relies. More generally, when one side makes factual claims that are crucial to its argument, the other side has incentives to show that those claims are not true. Hyperlinks nest policy in the facts that support them.

Importantly, the same computational revolution that has created the Internet is creating other information technologies that are force multipliers for the Internet’s capacity to increase political knowledge. Exponentially increasing computer power is rapidly improving empiricism—the careful statistical analysis of policies consequences. The Internet also makes possible prediction markets—markets where people can trade on the outcomes of policy decisions. Prediction markets incentivize those who know to bet on the contestable outcomes of policies like tax cuts or stimulus programs. The results pool the information of the dispersed citizenry and helps America know what Americans know.

Along with the Internet, these other information technologies can greatly improve our assessment of policy instruments to achieve common goals. But for the new digital information revolution to achieve its potential, government has to act to let the new technologies wash through our politics. For instance, Congress needs to remove the laws that make domestic prediction markets largely illegal. The President should encourage agencies to conduct randomized trials to allow empiricists to test what policies work. The Supreme Court should make sure that laws protecting journalists do not discriminate against bloggers and should relax campaign finance restrictions to help information technologies deliver political information at the time citizens most pay attention—during election campaigns.

The synergy of information technologies made possible by the ongoing computational revolution can transform politics today, just as the printing press did in its day. Empiricism helps us understand the consequences of past policies. Prediction markets then combine this empirical expertise with information dispersed throughout society, translating that compound into numbers that can command attention when publicized on the Internet. The combination of such technologies can also slowly shift social and political culture. Empiricists become more valuable than theorists at universities. Politicians gain more support for being driven by data rather than grandiloquent claims.

In short, over time the Internet and allied aspects of the computational revolution can create more focused and more accurate knowledge about the consequences of social policies. This knowledge in turn can help more citizens focus more on what they have in common—their shared goals and policies that may achieve them—rather than on the unsupported intuitions or personal circumstances that may divide them. Of course, some citizens will remain ideologues, impervious to updating on the facts. But democracy moves by changing the middle, not the extremes. Like other mechanisms that increase common knowledge, the Internet can give wing to the better angels of our nature.

I agree with Rebecca MacKinnon that institutions matter to how effective the Internet will be at improving governance at any particular time. But institutions themselves are endogenous to the cost of accessing information and creating social knowledge. Lowering that cost makes it easier over time for individuals to coordinate and act on encompassing interests. The encompassing interests include creating good institutions, like democracy in nondemocratic nations, and a more experimental politics in democratic ones.

To be sure, the occasions for actual institutional progress turn on the vicissitudes of politics and the preexisting institutions in a nation, but lowering information costs is very good both for allowing individuals to coordinate and for improving the quality of social knowledge. Thus I cannot agree with Ms. MacKinnon’s bottom line that it is unclear whether the Internet is conducive to liberal democracy in particular and better governance in general. My question to her is whether she thinks it remains unclear whether the invention of the printing press was ultimately conducive to liberal democracy and better governance? If it has become clear now that the printing press was beneficial, what is the salient difference between the press and the new medium of the Internet that further reduces information costs?

I disagree with the main thrust of Jason Benlevi’s piece, namely that the Internet is “concentrated, centralized, and controlled.” To the contrary, at least in free nations, it is the most open avenue for mass communication ever devised. In authoritarian nations, like China, it is a force that the authorities struggle to control, with substantial but not complete success. But such attempts at repression show that the Internet is a force for freedom that strikes fear into the hearts of oppressive regimes.

It is true that private companies, like Verizon, provide access to the Internet in democratic nations like ours. But these providers offer individuals access at a fraction of the price it cost to access an audience in previous years, let alone previous centuries. Anyone can start a blog and some who have done so now have audiences in the hundreds of thousands. Innovative policy ideas have never been more numerous. It is certainly true that advocacy on the Internet has not displaced other forms of political activism, but online information exchange helps refine the goals of such activism and coordinate its direction. In short, the Internet permits more minds to reflect on more common ideas and such widespread recombination of thought is indispensable to social progress.

Disruptive technological change is likely to lower information costs still further. Facebook and other networks now facilitate coordination. But if other networks become more effective at coordination, the costs of scaling up to create a better alternative have also never been lower. In short, the Internet makes is easier for anyone to communicate anything that others want to pay attention to. These developments bring us closer to a regime of spontaneous order than centralized control.

I find Mr. McGinnis’s statements that democracy is enabled by the reduced cost structures of digital communications to be, at best, a sidebar discussion, and factually not established – particularly in nondemocratic nations. The reduction in cost for carriage of packets of bits has virtually nothing to do with the argument concerning digital democracy. The principal issue is access and control, not the economy of scale of digital media. From both the information poster’s and the receiving user’s perspective, those in control of the infrastructure of the Net will have a more profound effect on digital democracy than costs. Analogies to the printing press as a dissemination tool fall apart because the printing press did not have controlling authorities; it was inherently decentralized and when necessary operated with stealth and anonymity.

Nontechnical users and policy professionals tend to look at the Net as if it were wild open space, but that perception is not based on an understanding of the true nature of IP networks. The situation is analogous to when you are a passenger on a cross-country flight. As you gaze out the window upon the horizons and vistas there is the feeling that the skies are a wide-open territory – that the pilot has free rein to pick his path – a lone eagle carrying his passengers through the open skies. If you are a pilot, the reality is far less romantic. Every move of that airplane is along well-established routing; the aircraft is constantly monitored by radio and radar as it makes its way from coast to coast. Never once is that craft beyond the control of the hubs and network of flight controllers. All along the way flight data is recorded. Yet, as passengers, we might have the feeling that we are crossing the wild blue yonder with Chuck Yeager at the stick.

On the Net, every bit and every packet of information is identifiable, mapped, routed, tracked, and controlled. In every case, the enabling entity is a commercial firm or a government agency that provides access. In every case that service provider entity retains control of that information stream. They may monitor and filter both the poster of information and the user-receiver. It is not an open and free space. The freedom of network associations is largely an illusion, a lovely tip to an ugly iceberg. Controlling entities have the ability to know exactly who you are, and exactly where you are, and by associating data, they can know exactly what you might be thinking.

On the Net, you may feel anonymous, however only the most sophisticated users can attain anonymity, and only until some cybersleuth counters their counter-measures. It is for a good reason that the arch-hacker non-group calls itself “anonymous.” A lack of anonymity chills dissent, even in democracies.

Although countermeasures such as TOR (The Onion Router) can establish some degree of anonymity for groups struggling for liberation in authoritarian states, given the investment available by interested government agencies (democratic and authoritarian) to defeat such workarounds, there is an irresistible commercial attractiveness to Silicon Valley investors to create tools that unmask dissident users.

Contrary to Mr. McGinnis’s statement, the Chinese government has been quite successful at filtering the Net, and equally successful at using the Net as a disinformation tool. Truth and the Internet are not joined at the hip. If the Net is a less costly way to disseminate information, it is at same time a less costly mechanism for spreading disinformation.

The larger point is that we can trust neither commercial nor government entities with our freedoms because the Net is inherently subject to controlling authorities. The more dependent and deluded we become by the fiction that it is propelling democracy the more vulnerable to authoritarianism we will become.

Rebecca McKinnon has done an excellent job at illuminating cases around the world where the vulnerabilities of digital democracy are demonstrable. Unlike Ms. McKinnon, I don’t believe that laws will be particularly effective at curbing the abuses.

Already in the world’s greatest democracy, the U.S. government and service providers such as AT&T and Verizon have violated FISA and paid no penalty. As for AT&T’s tapping into the Internet backbone on Folsom Street in San Francisco, we have no way of knowing whether that activity has ever ceased.

Although the network has democratizing capabilities, it is a false hope, and marketing hype, to believe it will be the engine of human liberation.

I do agree that the printing press was ultimately conducive to democracy and better governance. But I do not believe that the advent of printing press technology or any technology makes it inevitable that democracy and better governance will result. Democracy and better governance came about because enough people chose to use the printing press in a manner that led to these outcomes, and because people also fought successfully in enough places against laws and modes of governance that sought to constrain the printing press’s potential.

The Internet in its current form – that is, in a still relatively decentralized open form on which it is still possible to be anonymous if you make a concerted effort and if the NSA has not trained all its resources against you specifically – is indeed conducive to democracy and better governance. But the technical setup of the Internet can change in any number of directions. Its technical architecture and capabilities can be changed in ways that will make it less democracy-conducive than it currently is. Many members of the U.S. defense establishment, not to mention the governments of numerous other countries, are making the case very aggressively for eliminating anonymity, for example. Commercial services that collect inordinate amounts of data about people, making government’s surveillance job easier, aren’t helping either. The promise of technology does in my view not absolve human beings of moral and ethical responsibility.

Near the end of my book I discuss the danger of technological determinism, the assumption that technology will inevitably lead human history in a particular direction:

Technological determinism is as dangerous as historical determinism, the worldview underpinning the philosophies of Marx and Hegel, who believed that history was inevitably and inexorably moving the human race toward a certain endpoint. Marxist revolutionaries believed they were in the vanguard of the historically inevitable. Karl Popper, in Volume Two of The Open Society and Its Enemies, his seminal 1945 defense of liberal democracy, warned that historicism “is in conflict with any religion that teaches the importance of conscience.” Real human progress, he argued, can be achieved only “by defending and strengthening those democratic institutions upon which freedom, and with it progress, depends. And we shall do it much better as we become more fully aware of the fact that progress rests with us, with our watchfulness, with our efforts, with the clarity of our conception of our ends, and with the realism of their choice.”

Certainly we can and should use technology to do precisely that.