What does it mean to reflect on the “political implications of the Internet” today—a most challenging task that I’ve been asked to accomplish in the present essay? One easy answer—all too easy perhaps—is to simply follow the intellectual path beloved by the media, the pundits, and the cultural critics: we can just assume that we all know what the Internet is. Like the proverbial judge asked to define pornography, we might have great difficulty defining it but we know it when we see it.
If that’s the path we want to take, then our inquiry into the political implications of the Internet is likely to be contentious, inconclusive, and most likely infinite. For every invocation of some positive aspect to the Internet—”Look, the Internet was good for the Arab Spring: just look at how many people showed up to topple Mubarak!”—our imaginary interlocutor is likely to bring up some equally negative aspect—”Look, the Internet was bad for the Arab Spring: just look at all the surveillance and the failure to mobilize the digital masses after the first wave of protests in 2011.”
Such intellectual ping-pong—with one side finding a suitable positive example, only to be challenged by the other side finding a suitable negative example—has been going on, in one form or another, for the past 15 years. The impact of the Internet on both authoritarian states and democracies has been analyzed this way—down to very minute details about how specific political regimes operate. So debates about the filter bubble (courtesy of Eli Pariser) or audience polarization (courtesy of Cass Sunstein) or the brain-wrecking/brain-boosting aspects of social media (courtesy of Nicholas Carr and Clay Shirky, respectively) can all be nicely framed as part of this broader conversation about whether—to simplify our initial question even further—”The Internet is good or bad for democracy and politics as such?”
As an active participant in some of these debates over the last five years, I’ve quickly reached a depressing conclusion that, on many of the highly contested issues, it’s not uncommon for both parties to be wrong and right—simultaneously! Often, the opponents are either talking past each other or are focusing on two (or several) quite different aspects of the problem, somehow oblivious to the fact that once we abandon our quest to arrive at some ultimate score—if we stop comparing the negative side of our Internet ledger with the positive one—we might actually accommodate both perspectives—or reject them. The reason why we don’t do that is because there seems to exist some strange gravitational pull within our debates about technology—and that pull drags every single conversation toward teasing out some implications for the Internet at large. Forget about learning about the world: let’s just learn something about the Internet! Now, that’s a trendy subject.
In my two books, I’ve dubbed this gravitational pull Internet-centrism and, for my money, this is by far the most important “political consequence of the Internet”—not least because it erects a barely perceptible set of barriers and intellectual traffic lights of sorts that guides our debates toward certain outcomes or, in the worst case, gets them stranded in all sorts of intellectual traffic jams where they tend to remain for decades. The only way out of this intellectual impasse is to clear away those traffic jams; we shouldn’t be making things worse by continuing to traffic in dubious metaphysical assumptions of our own.
Consider an example that I’ve already mentioned: the debate about the impact of the Internet on the Arab Spring—a high-profile debate that is made all the more complex by the fact that those revolutions are still ongoing. Why is it so hard for us to accept that the proliferation of digital technologies could—given the favorable political, economic, and social conditions—allow a group of highly motivated young people to mobilize their supporters and advertise their protests while at the same time enabling those in power—and, above all, the secret police—to get a better handle on tracking the movements of their opponents? Or why can’t we accept that, in the absence of those favorable political, economic, and social conditions, those in power are likely to exploit the same digital technologies for their own gain, be it to spread propaganda or surveillance or harassment or censorship or espionage? Or that there might be an important role that these digital technologies are playing in creating both favorable political, economic, and social conditions—by allowing access to more information, creating new jobs, weakening the role of dogmatic authority—for enabling democratization while at the same time creating political, economic, and social conditions—the weakening of mainstream political parties, the further marginalization of the disconnected lower classes, the ability to spread religious propaganda—that might further inhibit it? Why can’t we seem to hold all these multiple perspectives on the Internet in mind at the same time?
The broader point I’m making here is that, as virtually every one of our social activities is being digitized, it’s very arrogant of us to expect that, somehow, we would be able to figure out what the role of the Internet in all of this is. Given how ubiquitous and cheap both digitization and connectivity are, what we call the Internet—and I here I don’t just mean computers, laptops, and routers but also smartphones and the Internet of Things and cheap sensors—is invading every single corner of our existence. This is not by any means a bad thing in itself. Properly designed and governed, this can actually be extremely emancipatory and be a healthy development for democracy. But what we need to come to grips with is that, once the Internet is everywhere, a question like “What are the political implications of the Internet?” loses much meaning, in part because it’s like asking “What are the political implications of everything for everything?” A giant supercomputer might answer this question but, alas, we don’t have it yet.
Consider an intriguing, even if a bit odd, parallel. Suppose we take the same case study—the Arab Spring—but instead of the Internet, we want to figure out the political implications of money. So everyone—the military, the dictators, the secular opposition, the Islamic opposition, the religious institutions—are given $100 million to spend as they wish. Now, it’s obvious that, if we only rely on theory and speak in the abstract, we won’t be able to predict what the impact of this cash infusion would be. Perhaps the opposition will use it to print more leaflets or establish stronger alliances with the trade unions. Or perhaps they will send some of their leaders for training abroad. Or maybe they will just steal some of the money. Perhaps the government will use it to buy more weapons. Or perhaps they will hire more police. Or maybe they will buy more surveillance equipment. But then maybe the religious institutions will use the money to build a splendid mosque that would somehow relieve the tensions.
To answer a question like “What are the political implications of money?” in this case would require knowing everything about how a given society operates, having an excellent grasp of its social fabric, being able to predict what alliances are likely to emerge and when. Clearly, this question is much harder than it appears at first sight; otherwise, the billions that the American government—and it’s not exactly short on Middle East experts—poured into foreign aid to some of the regimes in the Middle East would have resulted in democracy long time ago. In retrospect, this looks like a silly question—and few of us would ever seriously pose it.
But why don’t we feel the same constraints when it comes to inquiring into the “political implications of the Internet”? And is there a better way to preserve the spirit of this question—and still get some answers—even if we pose it differently? Tackling the first question would give us a clue to the second one. The reason why we keep asking “So, on the whole, is the Internet good or bad?” type of questions has to do with our strong belief that it’s a medium and, as a medium, it has some coherence—a logic of sorts—that, once applied to political and social institutions, can meld them in accordance with what the logic of the Internet demands.
One can counter that, when it comes to money, we are dealing with a medium as well—its logic, some might say, is to create markets. This is trivially true but our set of assumptions about the Internet and its logic runs much deeper and wider. For example, most of us to believe that it’s an either/or type of medium: it’s either a tool of enslavement (i.e., it would favor the governors) or emancipation (i.e., it would favor the governed). That it might do both—and that it might do it differently at different times, depending on the exact historical conditions in a given country—is an insight that is hard to square with how we think of this medium.
For what is this Internet? It’s a set of services, platforms, standards, and user behaviors. It might seem that the platforms, to take just one example, are the same everywhere—but, of course, they aren’t. And it’s not just a matter of digital device. Online platforms that are popular in Russia—LiveJournal or VK—have different modes of governance, different policies with regards to free speech, different functionality—than platforms that are popular in either America or China. Yes, we might call all of them online platforms or blogging platforms but, at the micro-level—the level that shapes user interaction and user behavior—these are profoundly different.
These platforms—whose evolution has been shaped by the peculiarity of political conditions in which they emerged—give rise to different citizens and different politics. This is not to say that they can’t give rise to democratic politics, protests, and manifestations of public anger—as we all know from the news, they do it regularly—but even if they do, they probably do it via different routes and modalities of behavior. All of this is to say that it’s probably not a good idea to take a snapshot of the totality of such platforms, behaviors, and users in one country, call it the Internet, and then compare it with a snapshot of the totality of other platforms, other behaviors, and other users in another country on the false assumption that all of this too is somehow the same Internet. This is not the same Internet, it never was and it never will be.
But even in the context of a single country, it seems impossible to answer our initial question about the “political consequences of the Internet.” If, say, the Russian Internet is made of platforms, standards, user behaviors, and so on—and if we grant that both their individual shape and the form of their mutual entanglement are themselves the product of history, politics, economics, and culture—then we are essentially asking about the “political consequences of politics,” a tautology if there ever was one. The Internet, as this term exists in popular discourse, is not the Internet as it’s experienced by users on the ground. There’s no Platonic idea of the Internet or a stable abstract object around which we can build a philosophy or a social science or on which implications we can reflect. That is, it certainly exists as a ubiquitous presence in our public debate but this is not the Internet as it is experienced by actors on the ground—those who are actually making politics.
What we tend to forget about the history of computing and digital networks is that the modes of behavior that we currently practice on the Internet today—sending e-mails, looking up information, shopping, engaging in debate—predate the idea of the Internet as such.
The myth that most of us have bought into is that, in the middle of the Cold War, a bunch of wise people with funding from the U.S. Defense Department got together, thought through about everything that the Internet could do, and then simply started implementing that agenda item by item, as if they had it all figured out.
But these people had no idea what the Internet was for, what it would be, or that it would soon be imagined as global village or cyberspace. For much of the 1970s and early 1980s, this Internet coexisted with many other similar networks. Even when the World Wide Web came into existence in the early 1990s, it coexisted with several other approaches—Gopher and WAIS were the most prominent ones—that, under somewhat different conditions, could have given us a digital environment that is very different from the one we have today. There’s simply no teleological logic leading to the World Wide Web; much of it wasn’t built according to a grand master plan. Different practices give rise to different technological infrastructures to enable them and it just happens that the network that links these infrastructures—the Internet—is now thoroughly confused with the plurality of both infrastructures and practices.
So, if we really want to be very specific about our language—a prerequisite, as I would argue, to talking about politics—we should state the following:
The practice of social networking in Egypt is different from the practice of social networking in China, even though both have some functional similarities.
Users in Egypt do and expect different things from social networks than do people in China—which makes perfect sense given that they live in different cultures, with different political, social, and cultural concerns.
Now, the actual social-technological infrastructures that enable social networking in Egypt are almost certainly different from those in China; in the former case, much of this social networking probably happens on Facebook—an American site that might have a complicated outlook toward its Egyptian users—whereas in China, much of social networking activity happens on local sites that are tightly controlled by the government. Their servers are probably located inside the country, not outside. They probably have a team of native speakers to do censorship—not necessarily the case in Egypt/Facebook. Such differences in socio-technical infrastructures that enable the practice of social networking have profound implications on how much freedom users have in each case; how they relate to each other; how subversive they have to get to express their discontent; how easy it is for state authorities to monitor their actions, and so forth.
Finally, there’s one final network—the Internet—which is actually of rather trivial importance in this comparison, as Egyptian users in Facebook and Chinese users on a local Chinese social network probably do not have much to say to each other. Yes, it’s true that they are all wired by the same network—and that network does have the same standards and protocols—but this insight is of little consequence here. Once we switch to a practice-based view of the world, we discover that, even though the Egyptian users and the Chinese users browse the same Internet for the purposes of social networking, their experience on it is profoundly different. Moreover, as already noted, even within each country, we are likely to see lots of other variations that depend on where and when we look: in times of upheaval, social networking could be more or less useful for protesters depending on what their goals are and how much surveillance and censorship power the authorities have.
To believe that we can collapse all these differences into just one Internet and then study its political implications seems naïve and actually irresponsible. As enjoyable as it has been, this debate—about whether “the Internet is good or bad for dictators”—must end, in part because there’s simply nothing interesting to be said about this abstract Internet thing. This doesn’t just work for dictators, by the way; it applies for studying political changes in democratic regimes as well. Anyone who has enough knowledge and patience to map out the political culture of a given democratic regime—and then do a similar mapping of its media-technological-knowledge infrastructures—would discover the impossibility of predicting—and then aggregating—the totality of changes in political culture that are triggered by shifts, even tiny ones, in how the media-technological-knowledge infrastructures operate.
A few examples might suffice. A country with strong freedom of information laws might suddenly discover that, thanks to search engines, the documents that were previously public but stored in a library are now widely accessibly online, at no cost and extra effort. Is it good or bad for democracy? This is not a question we can answer in the abstract. Or we might discover that, suddenly, search engines and their autocomplete function allow us to see which politicians are believed—or speculated—to be taking bribes as the word bribe follows their name in the search query. Is it good or bad for democracy? This again is hard to say in the abstract. And this is just the search engines—but think about social networking, databases, Wikipedia, smartphones, sensors, Big Data, algorithms—all of this is part of the Internet too. The idea that, somehow, all of these technologies will have similar effects—and those effects will hold regardless of the political culture where these technologies are put to use—seems delusional.
The only way forward for responsible researchers who are actually interested in figuring out the connections between media-technological-knowledge infrastructures and politics is to proceed slowly and carefully and without operating with such ambiguous concepts like the Internet. Yes to the study of individual practices, yes to the study of particular segments of the media-technological-knowledge infrastructures—no to the totalizing language of the Internet debate, with its assumption that it’s a single and coherent medium—”just like the printing press,” as the pundits like to say—which is manifesting the same effects everywhere it goes.
How and why we have settled on this language and set of metaphors—i.e., the idea that the Internet is an agent of change that is similar to the printing press—is itself a profoundly important question that our digital intellectuals shouldn’t shy away from tackling in full force.
For if we do want to understand the “political implications of the Internet,” we can see them right here, in the way in which most of the debates about the Internet are set up: in framing questions in a particular manner—”Tell us how the Internet affects X…”—we make certain answers and certain modes of thinking impossible. We take them off the table, so to say, and instead prefer to continue with the exciting game of intellectual ping-pong where we are constantly asked to update the score. Twitter enabling new protests in Russia? Great: score one for “the Internet is good for democracy.” American firms selling surveillance equipment to dictators in the Middle East? Too bad: score one for “the Internet is bad for democracy.”
We must learn to register such developments—Twitter-based protests, after all, are as important as the murky sales of powerful surveillance software—without feeling any need to update the score in that ping-pong game. For if we are truly concerned with the future of democracy in the world, we must make sure that (a) Twitter is most useful to protesters worldwide and that its commercial ethos doesn’t undermine its utility for activists, and (b) Western governments have enough regulations in place that would prohibit their own firms from shipping dangerous tools of surveillance to dictatorial regimes; in most cases, this would be a very difficult problem to solve, in part because these tools are built to satisfy the surveillance needs in democracies—we got a painful reminder of this thanks to Edward Snowden’s revelations.
Both of these questions—the utility of Twitter for protest and the challenges involved in containing the sprawling surveillance apparatus built by democracies—would require a lot of soul-searching and force us to task lots of uncomfortable questions: about the future of capitalism, privacy, personal data, responsibility of companies and governments, the Western obsession with the war on terror, and so forth. None of these questions will be easy to answer on their own but they would get maddeningly difficult to answer if we also confuse ourselves with an unnecessary urge to somehow make sure that our answers cohere to some vision of the Internetas a singular network, a single medium with coherent logics and demands. No, this vision won’t serve us any good and we might as well abandon it from the very start; the questions we need to answer already look complicated enough.
It would be naïve to think that, as we move forward, our intellectual predicaments will become lighter and our challenges easier. Of course not: we’ll face even more practices, more infrastructures, more techniques of creating, manipulating, and disseminating knowledge. All of them will change the political culture of each and every state in ways that no one can really predict. Yes, there might be similarities and the ongoing inter-networking and the intercommunication enabled by the English language would yield certain homogenization in practices. But it would be incorrect to expect that such occasional homogenizations would trigger more differentiations or result in completely new actors, practices, or techniques. That Islamist groups use Twitter to publicize their terrorist acts does tell us something about globalization but it doesn’t tell us much about the direction in which it would be moving, let alone about what it has in store for democracy or cosmopolitanism. For all we know, the global exposure enabled by the internetworking might spawn more local copycats who would pursue their own highly localized projects of terror.
The great intellectual mistake that we could make in this regard is to assume that, somehow, if only we think hard enough about the Internet, we would arrive at the right answer as to what would happen to the world once everything is interconnected and digital.
To reiterate: this is a false hope. Such intellectual mastery would never happen—in part because digitization or connectivity are not like physical or chemical processes whose consequences we can predict. And this has nothing to do with the protean nature of the Internet or it being the most complex force in history; no, it simply has to do with the fact that what is being digitized and connected are various parts of our society—and it’s those parts that defy any logic of prediction.
Think about it this way: the Arab Spring has proved as impossible to predict in advance as the Cold War—all of this despite the fact that almost everyone carries a mobile phone, there’s plenty of Big Data on social media sites, and the computing power available for churning out predictions is much more impressive than it was in the 1980s. And yet, with all this data and with all this computing power, even the CIA, with its impressive models and its penchant for game theory and data collection, failed to even remotely predict it in advance. Actually, given these immense technological resources, the failure to predict the Arab Spring looks far more remarkable than the failure to predict the fall of the Soviet Union and the eventual end of the Cold War. So don’t hold your breath for anyone being able to work out “the political implications of the Internet” any time soon.
Does it all mean that we should just abandon all hope and do nothing, hoping that, somehow, now that everyone has access to a smartphone and Google, things will work themselves out and democracy will eventually prevail? Well, no: this would be too irresponsible. The best we can do is to develop a better set of optical tools—the ones that would allow us to zoom in on particular practices and notice the actual bits and pieces of the many infrastructures hiding behind the Internet label—and embrace a form of epistemological modesty, where, every time we are asked to opine on “What does the Internet do to Subject X?” we politely decline and stay silent. Or, if we are of a more dissenting breed, we point out the explicit danger of asking such questions.
Comments on this publication