War Goes Viral

Like most every­thing today, the campaign was launched with a hashtag. But instead of promoting a new album or a movie release, #AllEyesOnISIS announced the 2014 invasion of northern Iraq—a bloody takeover that still haunts global politics two years later.

Revealing a military operation via Twitter would seem a strange strategy, but it should not be surprising given the source. The self-styled Islamic State owes its existence to what the internet has become with the rise of social media—a vast chamber of online sharing and conversation and argumentation and indoctrination, echoing with billions of voices.

Social media has empowered ISIS recruiting, helping the group draw at least 30,000 foreign fighters, from some 100 countries, to the battlefields of Syria and Iraq. It has aided the seeding of new franchises in places ranging from Libya and Afghanistan to Nigeria and Bangladesh. It was the vehicle ISIS used to declare war on the United States: The execution of the American journalist James Foley was deliberately choreographed for viral distribution. And it is how the group has inspired acts of terror on five continents.

So intertwined are the Islamic State’s online propaganda and real-life operations that one can hardly be separated from the other. As ISIS invaders swept across northern Iraq two years ago, they spammed Twitter with triumphal announcements of freshly conquered towns and horrific images of what had happened to those who fought back. A smartphone app that the group had created allowed fans to follow along easily at home and link their social-media accounts in solidarity, permitting ISIS to post automatically on their behalf. J. M. Berger, a fellow with George Washington University’s Program on Extremism, counted as many as 40,000 tweets originating from the app in a single day as black-clad militants bore down on the city of Mosul.

Media reports from the region were saturated with news of the latest ISIS victory or atrocity, helping to fuel a sense of the Islamic State’s momentum. There was no time to distinguish false stories from real ones. Instead, each new post contributed to the sense that northern Iraq had simply collapsed in the face of the ISIS onslaught.

And then it did. Terror engulfed Mosul, a city of 1.8 million people. The 25,000-strong Iraqi garrison may have been equipped with an arsenal of American-made Abrams tanks and Black Hawk helicopters, but it was disoriented by reports of the enemy’s speed and ferocity. Already beset by low morale and long-festering corruption, it crumpled under the advance of a mere 1,500 ISIS fighters, equipped mostly with small arms. The Islamic State was left to occupy the city virtually uncontested, seizing vast quantities of weapons and supplies, including some 2,300 Humvees.

In the abrupt surrender of Mosul and collapse of defending Iraqi forces, one could find echoes of the similarly shocking fall of France to the 1940 German blitzkrieg. The Germans relied upon the close coordination of tanks and planes, linked together by radio. Radio gave their forces speed—and also the ability to sow fear beyond the front lines.

ISIS spread a similar panic online. Immaculately staged photos, filtered through Instagram, transformed a ragtag force riding in dusty pickup trucks into something larger than life. Armies of Twitter bots twisted small, one-sided skirmishes into significant battlefield victories. Hashtags were created and pushed (and others hijacked) to shape and hype the story. Through this fusion of activities, ISIS stumbled upon something new. It became, in the words of Jared Cohen, a former State Department staffer and now the director of Jigsaw (Google’s internal think tank), “the first terrorist group to hold both physical and digital territory.”

Tweets, photos, and other visuals from ISIS and ISIS-friendly accounts feature a mixture of slick production and attempts at intimacy and personal connection.

It will not be the last. The fate of the self-declared caliphate, now under the assault of nearly two dozen national militaries, is uncertain. Yet the group has already proved something that should concern any observer of war and peace, law and anarchy. While the Islamic State has shown savvy in its use of social media, it is the technology itself—not any unique genius on the part of the jihadists—that lies at the heart of the group’s disruptive power and outsize success. Other groups will follow.

And not just terrorist groups. This is only the beginning of a larger revolution, one that is already starting to reshape the operations of small-time gangs on one end of the spectrum, and the political and military strategies of heavily armed superpowers on the other.

More than a year ago, we set out to understand the use of social media as both a tool in conflict and a shaper of it, tracking how online chatter has begun to intersect with real-life violence in dozens of armed confrontations around the globe. In doing so, we sought to untangle a seeming contradiction. The internet has long been celebrated for its power to bring people together. Yet as it turns out, this same technology is easily weaponized. Smartphones and social apps have clearly altered the nuts and bolts of violent conflict, from recruiting to battlefield reporting. But the greatest effects may be more fundamental, expanding the causes and possibly the incidence of war, and extending its reach. Social-media platforms reinforce “us versus them” narratives, expose vulnerable people to virulent ideologies, and inflame even long-dormant hatreds. They create massive groundswells of popular opinion that are nearly impossible to predict or control.

Social media has already revolutionized everything from dating to business to politics. Now it is reshaping war itself.

“A Bond of Perpetual Peace”

War, as the 19th-century military theorist Carl von Clausewitz famously put it, is simply the continuation of politics by other means. Social media, by democratizing the spread of information and erasing the boundaries of time and distance, has expanded the means, transforming war to an extent not seen since the advent of the telegraph.

In 1838, Sidney Morse wrote to his brother Samuel to congratulate him on the recent unveiling of the telegraph, which Sidney called “not only the greatest invention of this age, but the greatest invention of any age.” He prophesied, “The surface of the earth will be networked with wire, and every wire will be a nerve. The earth will become a huge animal with ten million hands, and in every hand a pen to record whatever the directing soul may dictate!”

In his 1998 book, The Victorian Internet, Tom Standage describes how starkly and suddenly the telegraph altered many aspects of life. A culture of tinkerers and hackers arose around the device, with its own lingo and even its own courtships and romances, conducted in Morse code. Businesses could track their supplies with a level of accuracy hitherto unimaginable, and coordinate far-flung operations more closely. Newspapers, which had barely contained any international coverage before, were suddenly stuffed with reports of recent events taking place thousands of miles away. Overnight, these distant occurrences assumed great weight in political discourse, even though their actual effect on people’s lives had not changed at all.

As telegraph cables crisscrossed the globe, many observers felt that history had turned a page. According to the historian Johanna Neuman, great thinkers of the day believed that “the knowledge relayed by the telegraph would make nations so conversant with the national interests of their one-time enemies that war would come no more.” The first transatlantic cable was laid between North America and Europe in 1858. In an exchange of congratulations, President James Buchanan expressed to Queen Victoria his belief that the telegraph would “prove to be a bond of perpetual peace and friendship between the kindred nations, and an instrument designed … to diffuse religion, civilization, liberty, and law throughout the world.”

Within a few days, Britain would use the same cable to send orders to its military.

The telegraph swiftly became an important new tool of war. Beginning in the Crimean War (1853–56), what might once have been broad instructions, traveling weeks by sea, became, to the lament of officers in the field, micromanaged battle orders sent by cables from London to Russia. A new kind of generalship emerged during the Prussian Wars of Unification (1864–71), as the movements of whole armies were coordinated in real time. In the American Civil War (1861–65), Confederate and Union soldiers, each seeking an edge over the other, laid some 15,000 miles of telegraph wire.

The telegraph also reshaped the public experience of war. One journalist marveled, “A battle is fought three thousand miles away, and we have the particulars while they are taking the wounded to the hospital.” This immediacy, in turn, introduced new opportunities for ideologues and media entrepreneurs to stoke public outrage and even enthusiasm for war: The competitive “yellow journalism” that preceded the Spanish-American War (1898) is the classic example. As news reporting increasingly became a contest of speed, accuracy became a secondary concern. Members of the Associated Press were so intent on keeping readers informed of every lurid detail of the conflict with Spain that they chartered boats that sailed frantically through naval battles to reach the nearest telegraph station.

Citizens around the world were suddenly privy to “news”—whether true or not—that had once been the exclusive domain of monarchs and ministers. Meanwhile, information obtained by newspapers could drive government action. The world had shrunk. The pace of international events increased.

Similarly seismic changes are now being wrought by social media. Today, there are 3.4 billion internet users, rendering Sidney Morse’s bold prediction of “ten million hands” rather modest by comparison. Roughly 500 million tweets are sent each day. Nearly seven hours of footage is uploaded to YouTube each second, in up to 76 different languages. With 1.7 billion active accounts, Facebook is the largest “country” in the world. According to Pew, clear majorities of American Twitter and Facebook users now get their news from these platforms. Fifty-nine percent of American Twitter users rely on the service to follow news events as they happen in real time.

Yet we are not at the crest of the wave. Nearly half of the world’s adult population is still not online. Many of the new connections will be concentrated in regions most susceptible to violence and conflict. According to the International Telecommunication Union, internet use in the developing world grew by an average of 16 percent each year from 2005 to 2015. The U.S. National Intelligence Council has estimated that more people in sub-Saharan Africa, South Asia, and the Middle East have internet access than have electricity.

Such global connectivity has long stood as Silicon Valley’s holy grail, in the pursuit not just of profits but also of peace. It is why Google seeks to release giant balloons into the stratosphere, beaming internet access down to people who lack it, and why Facebook is building solar-powered drones to do the same.

In 2005, when “The Facebook” was still a Palo Alto start-up, a college-age Mark Zuckerberg was interviewed by camcorder in the office lounge, red Solo cup in hand. “The goal wasn’t to make an online community,” he explained of his new platform, but “a mirror of what existed in real life.”

Social media is indeed a mirror, one that reflects all manner of human interests and ideas, invariably extending into the realm of politics and violence. Last year, the most-talked-about event on Twitter was not a silly meme or a feel-good story: It was the November 2015 terrorist attacks in Paris, in which 130 people were killed by a coordinated team of ISIS gunmen. Millions watched as images and snippets of video captured the chaotic scenes. The most-powerful updates came from the victims trapped in the Bataclan theater, who naturally turned to social media to plead for help, even as jihadist murderers stalked the halls.

The duality of human nature is readily apparent when social media fixates on conflict. Thanks to the internet, war crimes have been laid bare by citizen reporters examining evidence from thousands of miles away, and a voice has been given to suffering civilians who previously had none. Strangers can be moved to tears by the image of a drowned Syrian toddler washing up on the shores of Turkey, and the world has never seemed so small. But social media has also opened new avenues for extraordinary cruelty. In January, Syrian-regime loyalists, learning of a rebel-held town that was starving under siege, taunted the residents by posting pictures of what they were eating for dinner.

Indeed, the more we’ve learned about behavior on social media, the more apparent it has become that the mirror is distorted—or rather, that it distorts us. For all the hope that comes from connecting with new people and new ideas, researchers have found that online behavior is dominated by “homophily”: a tendency to listen to and associate with people like yourself, and to exclude outsiders. Social networks are bad at helping you empathize with people unlike you, but good at surrounding you with those who share your outlook. The new information ecosystem does not challenge biases; it reinforces them.

A review by the analytics firm Gnip (since acquired by Twitter) of 11.5 million tweets during and about the November 2012 Israeli-Palestinian clash, for instance, found that only 10 percent of this conversation occurred between supporters of the opposing sides. A similar examination of online activity during the 2014 race-related protests in Ferguson, Missouri, found that liberals and conservatives in the U.S. cited or put forth completely different facts and arguments and seemed hardly to acknowledge each other’s existence. Since May of this year, The Wall Street Journal has run a project called “Blue Feed, Red Feed,” showing side-by-side Facebook streams of news sources popular with, respectively, liberal and conservative audiences. The resulting social-media feeds look like they’re from two parallel universes.

Within a circle of friends or like-minded acquaintances, social media certainly fosters connection. But the further one zooms out—to whole societies or the course of global affairs—the more this connection is marred by tribalism and mutual mistrust.

This problem is particularly disturbing because of another feature of social media: Its users are not passive consumers, like TV viewers or radio listeners or even early internet users. Via platforms that range from Facebook and Instagram to Twitter and Weibo, we are all now information creators, collectors, and distributors. Civilians in conflict areas can take and publish inflammatory photos of collateral damage; suburban teens in Marseille or Seattle can follow the lives and losses of individual combatants and interact with them directly. And of course, messages that resonate can be endorsed, adapted, and instantly amplified.

Both ends of the communications process have been democratized in a way that no prior technology has accomplished. Social media has made a great many of us participants in, as well as observers of, conflict. The implications of this wide-scale participation extend far beyond the virtual realm.

War: The Viral-Marketing Campaign

How can a group use social media to involve people deeply in a distant conflict—and even persuade them to join it? As a case study, consider the Islamic State.

The ISIS propaganda machine is equal parts frightening and surreal: Prisoners who are about to be beheaded are fitted with lavalier microphones; synchronized murders are set to booming chorales; brutal clips of death and martyrdom are stitched together with Final Cut Pro. Just how did a throwback death cult with a seventh-century worldview come to dominate 21st-century social media so swiftly and completely?

While ISIS may represent something new in its targeting of both physical and digital domains, it hasn’t, in fact, invented anything new. Its members, in the words of the Australian counter-terrorism researcher Haroro Ingram, are “more strategic plagiarists than geniuses.” ISIS has simply adapted the time-tested tactics of terror to the new rules of the social-media age.

Terrorism has always been theatrical. Some 2,000 years ago, Jewish zealots known as the sicarii, or “dagger men,” stalked Roman-occupied Jerusalem. Rather than killing quietly in alleyways, they made sure to slay Roman sympathizers before a crowd. The aims of these town-square assassinations were the same as those of the Islamic State’s YouTube beheadings: to send a signal to as large an audience as possible.

It was inevitable that terrorists, eager to spread their message, would be among the first to recognize the promise of social media. What we know as the Islamic State emerged from a mix of former lieutenants of Saddam Hussein and vicious jihadists of al-Qaeda in Iraq. They found common cause in Syria, broke with al-Qaeda, and were joined by a fresh wave of Millennial-generation recruits who had come of age during the 2011 Arab Spring—and who had seen the attention-grabbing power of Facebook and Twitter firsthand.

William McCants, a scholar of militant Islam at the Brookings Institution, has tracked the evolution of terrorist propaganda, from audiotapes passed around by hand to hour-long sermons on VHS snuck out of Afghanistan to digital videos that look like movie trailers, tailored for sharing. ISIS mastered the latter, and this mastery, McCants says, helped it supplant al-Qaeda as the brand in favor among a new generation of jihadists. “Al‑Qaeda videos look like something you’d see on Charlie Rose or PBS NewsHour,” he says. “ISIS videos have more of a Vice feel about them: They’re very visceral, very immediate. They’re from the battlefield.” But McCants downplays the suggestion that this formula makes ISIS some kind of social-media innovator. The technologies to create these types of videos are now cheap and readily available. “It’s not mind-blowing—it’s what a normal PR firm might do.”

Indeed, strip away the religious claims and the on-camera killings, and the ISIS online playbook looks much like any of the dozens of social-media-marketing “how‑to”s circulated by consultants. The principles that have guided the Islamic State’s viral success are the same ones used to publicize a new Taylor Swift album or the latest Star Wars movie. They are out there for anyone to copy.

Two media specialists at the University of North Carolina at Chapel Hill, Cori Dauber and Mark Robinson, have studied how the Islamic State builds its message, and discovered a consistent and conscious effort to mimic the “Hollywood visual style.” Colors are saturated, contrasted, and crisp; subjects are kept in clear and tight focus. A former ISIS cameraman, now in a Moroccan jail, described to The Washington Post how he worked with nine other crew members to document the massacre of 160 captured Syrian soldiers in the desert south of Raqqa. Like the camera operators who film The Bachelor and other reality shows, they wove among “participants,” recording from a host of different angles, seeking the perfect shot.

A study of 1,300 ISIS propaganda videos by Javier Lesaca, a visiting scholar at George Washington University’s School of Media and Public Affairs, found that 20 percent were directly inspired by Western entertainment: Call of Duty, Grand Theft Auto, American Sniper. The irony is as rich as it is gruesome—a group that sprang from al-Qaeda in Iraq copies shots from a Clint Eastwood film about an American serviceman who won glory while fighting al-Qaeda in Iraq.

The Islamic State’s careful audiovisual engineering hints at a future of war propaganda that will lean almost entirely on evocative and shareable images—everything from doctored photographs to video screenshots to infographics. ISIS militants have discovered, as marketing experts have long known, that compelling imagery matters far more than any accompanying text in determining whether or not something goes viral. Indeed, when the Turkish military launched an August offensive into Syria to sweep ISIS militants from its border, it cribbed many of the very same online tactics, creating a Twitter account for the operation that pushed out everything from soldier selfies to dramatic, staged videos of commando raids.

During a recent campaign in Syria, the Turkish Armed Forces borrowed a page from ISIS’s playbook, using social media to instill a sense of unopposable force.

But the Islamic State also understands the importance of intimacy and authenticity to social-media outreach. Professionally choreographed videos are complemented by rougher, first-person shots of chaotic gun battles. And both are posted by actual fighters, who also opine on everything from religion to potato-peeling duty. For years, a Dutch jihadist fighting in the ISIS ranks maintained a personal Tumblr bursting with arresting images: his fellow fighters at rest; his newborn baby; even his cat, stretched alongside a suicide belt.

These qualities have lain at the heart of the Islamic State’s success in online recruiting. Contact with sympathizers has often been made in an open forum, and then moved to private message exchanges. Plenty of radicalized Westerners, pulled back from the brink of recruitment, have described online relationships that unspooled over weeks or months. In time, the jihadists living on the other side of the world (or in some cases, pretending to) ceased to be seen as recruiters. They became friends—or at least the social-media version of friends.

While choreography might seem to be in opposition to authenticity and intimacy, their clever combination is actually how the pop singer Katy Perry has accumulated more than 90 million Twitter followers, more than any head of state. Her tweets are usually casual and abbreviated, as if dashed out to a small group of friends. They intermix promotion with mundane, real-life moments. Likewise, the ISIS fighters who talk up the glory of the caliphate also muse online about, say, the death of the actor Robin Williams and their childhood love of his movie Jumanji. This sense of authenticity wins and inspires followers in a way that official government press releases cannot.

The scale of the Islamic State’s online efforts has been striking, reflecting the group’s recognition of social media’s importance to its ends. In an October 2015 study for the Quilliam Foundation, the terrorism analyst Charlie Winter found that in a one-month period, the group released nearly 1,150 “propaganda events”—batches of related videos, articles, photos, and essays—originating from 35 different media-production units. This cascade splashed through tens of thousands of accounts associated with ISIS, strewn across more than a dozen social-media platforms.

Most of these releases never go viral, but then again, neither do most of the more than 200 articles a day posted by the online publishing giant BuzzFeed. Like BuzzFeed, ISIS appears to realize that while the internet never forgets, it is also true that people have never so quickly forgotten the things they see on the internet. (Remember Kony 2012 or Cecil the Lion?) Peter Bray, a social-media analyst, has found that the average tweet reaches the zenith of its popularity just 18 minutes after it’s sent. ISIS keeps its content fresh and in front of viewers by making many small bets, knowing some of them will pay off big.

The group also tailors some of its propaganda to be picked up directly by the mainstream media, baiting them into amplifying the message further. There was initial puzzlement, in the hours following the August 2014 release of the horrific video showing the beheading of James Foley, as to why these brutal ISIS militants had not made the footage more gruesome, in the style of al‑Qaeda’s past executions. Why had ISIS instead cut to black right as the murder began? Some news outlets unwittingly provided the answer by posting graphic stills on their websites and linking to the full video: The event had been filmed in such a way as to make it shareable by conventional media outlets. Why do the hard work of spreading propaganda when others can be relied upon to do it instead?

This same thinking has informed the Islamic State’s stratagem of hijacking breaking news. ISIS supporters have appropriated hashtags for global events like the World Cup, regional news like an earthquake in Napa, California, and even events as inconsequential as an interview with a minor YouTube celebrity. Each of these tiny invasions has generated its own echoes in the press—including this article.

The Islamic State’s online information war has unquestionably been effective. The past three years have seen a marked increase in local acts of terrorism “inspired”—but not directed—by the Islamic State and other entities. (In the U.S., the number so far in the 2010s is already more than twice that of the previous decade.) While a vast majority of people—Muslim and non-Muslim—reject the group’s toxic ideology, social media has nonetheless enabled ISIS to find sympathizers and converts all over the world.

Over time, this online propaganda—and the heightened visibility of terrorism itself—has burrowed deep into the psyches of people far beyond the Islamic State’s physical control. According to public-opinion research by Gallup, over the past two years, American fears of terrorism have risen to a height not seen since the aftermath of 9/11. Even when violence is isolated and sporadic, social media ensures that it is never far from people’s minds. That in turn encourages ugly stereotyping and harmful overreactions by citizens, media, and politicians. The result is a widening of divisions and the spread of anger and fear—an ecosystem in which ISIS thrives.

Other extremist groups are already using elements of the Islamic State’s playbook to try to win converts and attention—and seem to be succeeding. A recent paper by J. M. Berger, the expert on extremism at George Washington University, tracked the coordinated use of social media by American white-nationalist groups, whose ranks on Twitter have increased by 600 percent since 2012. Members, Berger wrote, push out hashtags and messages “in concert at high volumes” in order to build enough momentum to grab the attention of mainstream media outlets. They use platforms like Reddit for broad messaging—but also to draw individual users further into their web. Controversial and vibrant images are engineered to go viral, while videos and songs build on the success of existing memes: everything from spoofs of the Matrix movies to racist reworkings of “Let It Go” from Disney’s Frozen.

Netwar Is Here

More than a quarter century ago, two defense analysts with the Rand Corporation began to think seriously about how conflict might be shaped by the nascent internet. In their groundbreaking article “Cyberwar Is Coming!,” published in 1993, scarcely two years after the first website had been created, John Arquilla and David Ronfeldt predicted a future of military operations in which software code would be used as a means of attack.

They also went a step further. Just as militaries might clash in cyberspace, they argued, entire societies would collide in a phenomenon they called “netwar.” In this sort of conflict, reality itself would be up for grabs. Netwar, they wrote, “means trying to disrupt, damage, or modify what a target population ‘knows’ or thinks it knows.” Information could be fashioned into a dangerous weapon.

Today, netwar is a daily reality. After lingering in the shadows of Russian military planning for decades, Soviet-style “information warfare” entered a period of renaissance in the past decade. Russian officials felt increasing pressure from the forces of Western liberalization and internet technology as they watched “color revolutions” engulf many nations of the former Soviet bloc. So they set out to harness the power of the internet to their own ends, controlling it at home and using it to divide foes abroad. An association of nearly 75 education and research institutions was devoted to studying the finer details of how the internet works, coordinated by the Russian Federal Security Service—the successor to the KGB.

The flagship of the Russian propaganda machine is Russia Today—or just RT, as it is emblazoned on New York City buses and street signs lining Fifth Avenue—which promises the time-honored service (and perceived truth) of “the second opinion.” A glitzy and contrarian news service that received roughly $250 million in government subsidies for 2016, RT injects Russian state opinion into international reporting; it broadcasts in English, Arabic, and Spanish, and posts additional items online in Russian, French, and German. It has become the most popular television news network on YouTube.

Yet Russian information operations are like icebergs: RT and other branded propaganda outlets are just the small part that is visible. Beneath the surface, Russia maintains a vast digital network of bloggers and paid social-media commenters, many of whom do not advertise themselves as Russians at all. It is surprisingly easy to draw their ire. Just post something unfriendly toward the Russian position on Crimea or the 2014 shooting-down of Malaysia Airlines Flight 17 over Ukraine and you will soon find yourself receiving insulting messages from people you’ve never met and friend requests from mysterious lingerie models eager to change your mind (and keep a closer watch on you besides).

Many of the real people behind these fake accounts are young and chic—aspiring writers who show up each day to work in “troll factories,” darkened office buildings nestled in the suburbs of Moscow and St. Petersburg. They manufacture dozens of online personae, working 12-hour shifts. From cramped cubicles, they vent fog into discussions about geopolitics, NATO, Ukraine, American elections, and everything in between. As a European Union official who studies Russia’s propaganda put it, “The aim is not to make you love Putin. The aim is to make you disbelieve anything. A disbelieving, fragile, unconscious audience is much easier to manipulate.”

In the past, information-warfare campaigns have typically come at great cost and had little prospect of success. Even if the propaganda reached its intended audience and found a sympathetic ear, what then? How could dissidents locate one another, much less coordinate enough to have a meaningful political effect?

Not so today. Thanks to social media, this same sort of propaganda effort can be conducted cheaply and almost invisibly. Even the most trivial sign of a political fissure—a few hundred angry users in an internet forum—represents a potential opportunity to sow discord and chaos in a rival nation.

Sometimes, the goal is simply to stack tinder, throw matches, and see what happens. Far-right political parties (nationalist and isolationist) in countries such as Hungary, Greece, and France have been bolstered by Russian cash, accorded disproportionate coverage by Russian media, and then spun up with social-media support. In the United Kingdom, the unsuccessful 2014 Scottish-independence referendum was loudly condemned as “rigged” by Russian observers seeking to delegitimize democratic processes and stir the pot of resentment. The 2016 “Brexit” campaign calling for Great Britain to leave the European Union was similarly lavished with attention by the Russian press and backed by an army of trolls and Twitter bots.

Other times, the misinformation campaign works toward narrower policy purposes. This summer, a small, peaceful anti-U.S. protest outside Turkey’s Incirlik Air Base was transformed, in characterizations by Russian media and internet trolls, into a much larger mob riot—portrayals that filtered into U.S. media and online discussion. Soon after, patently false rumors spread via social media that American nuclear weapons kept in Incirlik would be relocated to a military base in Romania—the same base where, in fact, a U.S. antimissile system had just been activated, over angry Russian objection. The aim of these falsehoods was to exaggerate the “disintegration” of U.S.-Turkish relations and to incite Romanian resentment against the NATO missile shield, in order to weaken acceptance of the U.S. military presence in Europe.

Information warfare can also serve more-chilling ends. Russia’s infiltration and invasion of Crimea and eastern Ukraine was preceded by a relentless online campaign to stoke pro-Russian protests and cast the new (Western-friendly) Ukrainian government as, quite literally, a bunch of Nazis. What appear to be Kremlin planning documents, later leaked online, describe the campaign as playing on the “centrifugal aspirations” of Ukrainian minorities in order to initiate a “pro-Russian drift.” Similar smoke-and-mirrors efforts appear to be under way against Estonia, Latvia, and Lithuania, each of which has a large ethnic-Russian population—and each of which is a member of NATO.

Few targets loom larger than the United States. This election cycle, Russian hackers targeted the U.S. political system, digging up embarrassing information and spreading it as widely as possible. Russian trolls posed as angry U.S. supporters of one or another political campaign while outlets like RT leapt to enlarge the divisions that other parts of the propaganda machine had helped create. What felt new and strange to many Americans followed a familiar script: provoking restive minorities, strengthening the hand of potentially friendly politicians and political movements, undermining trust in democratic processes, and generally raising the volume of anger and dissent.

The ultimate intent is not so much victory for a certain side, but a loss for everybody: sapping the credibility of U.S. institutions and tearing open as many wounds as possible. After Election Day, we should not be surprised to find a vocal group of internet users with mysterious IP addresses decrying the result as a fraud and driving talk of conspiracy—and even of resistance or secession. In time, we may see a multiplying number of homegrown violent extremists (along the lines of the infamous Oregon militiamen), encouraged by the subtle manipulation of a certain rival government.

Although Russia has pioneered this modern version of information warfare, it is hardly alone. Following a series of anticorruption protests in Turkey and a spate of critical international media coverage, for instance, the Turkish government hired thousands of professional trolls in a bid to build a social-media army. In Venezuela, authorities have used pro-government Twitter bots to manipulate one of the few news sources not already controlled by the state; the fake Twitter followers of Venezuelan President Nicolás Maduro were so loyal that he became the third-most-retweeted public figure in the world, behind only the king of Saudi Arabia and the pope.

All of these efforts share the same two broad objectives. The first is to overwhelm the state’s adversaries, be they foreign or domestic, with misinformation: to challenge the very basis of their reality. But the second is just as important: to mobilize their own citizens and supporters and bind them to the state. The power of social media is used to intensify nationalism and demonize the enemy. In this strategy, homophily is not something to be feared or avoided. It is the goal.

The combination of untruth and homophily—set against a global battle of competing narratives—hints at a dark future. A world without facts, cleanly segregated by ideology and national allegiance, will be a more dangerous one. Such cynical use of the internet not only threatens to keep people in a perpetual state of mistrust; it may also increase the likelihood of conflict itself.

When it comes to social-media mobilization, China stands in a league all its own. The Chinese Communist Party has long stoked the fires of nationalism among the 700 million Chinese internet users in order to bolster the state against the perceived threats of outside information. The strategy is equal parts censorship and manipulation. China employs as many as 2 million internet censors and trolls, who, far from operating in the shadows, have their own system of professional certification. In the words of Chinese President Xi Jinping, the ultimate goal of online expression must be “condensing public opinion into consensus.”

This apparatus will hit its next stage with the planned implementation of a national “social credit” system, in which the government will score citizens for trustworthiness and civic “goodness.” Akin to an Orwellian Klout score, the measure will draw from a combination of factors ranging from an individual’s employment history to her online behavior to even that of her friends and family, creating a self-policing system. In turn, the score will be used to determine real-world benefits and punishments.

Such programs offer the lure of control, which is growing ever more attractive as China enters a period of economic and political uncertainty. Their danger is that the regime will instead find itself in a position of “騎虎難下,” a proverb dating back to the Jin dynasty (a.d. 265–420) meaning, literally, “Riding a tiger and it being hard to get off.” China’s cybernationalists have been shaped into a potent force, but they are also a hive that erupts angrily at the slightest perceived provocation from the United States, Taiwan, or Japan—and not always at their masters’ bidding.

During this year’s Taiwanese elections, one of the most popular phrases on the Chinese social-media service Weibo translated as “Use force to unify Taiwan.” And while China was in discussions with its neighbors over disputed islands, Chinese networks commonly featured messages such as “Even if China is a graveyard, still need to kill all Japanese. Even if no grass grows in China, still need to recover Diaoyu Islands.” Following a July ruling by the International Court of Justice, which rejected many of China’s sweeping territorial claims to the South China Sea, Chinese social media exploded with hundreds of thousands of furious comments, many calling for war. The anger spooked senior party officials; censors and state media worked overtime to restrain the very forces they had once helped unleash.

Notably, the hive no longer roils at foreigners alone, but also at any Chinese-government actions that fall short of the most stridently patriotic standards. Following the October 2015 transit of a U.S. destroyer through contested waters, the fury of Chinese social-media users was directed not merely toward the United States, but also toward their country’s own military—once an unassailable institution. “Stop boasting and fight!” became a common refrain.

Such loud and abrasive internet users will not cause a war on their own, but they will complicate diplomats’ future efforts to avoid one. For the Chinese government, dependent above all else upon the illusion of consensus, the spontaneous political movements enabled by the internet represent a potentially existential threat. When the crowd cries for violence, its desires cannot be satisfied—but neither can they be wholly ignored. “Domestic voices calling for a more muscular Chinese foreign policy have created a heated political environment,” Thomas Christensen, a Princeton professor and former State Department official for China policy, has written in Foreign Affairs. “Gone are the days when Chinese elites could ignore these voices.”

It has become a cliché among international-relations scholars to draw parallels to 1914 Europe, but the potential challenges posed by social media make the comparison apt. Then, as now, regimes toyed with the power of nationalism, amplified by new communications mediums, in order to maintain stability at home. They discovered too late that the popular forces they sought to manipulate were beyond their control.

When Archduke Franz Ferdinand was assassinated in Sarajevo in June 1914, few thought that global conflict was at hand. But over the next several weeks, diplomats and monarchs were left feeling helpless as their nations barreled toward World War I. For some, the prospect of disappointing their own nationalist citizens scared them more than the war itself. German Chancellor Theobald von Bethmann-Hollweg lamented how the public clamor for blood constrained his choices, while Russian Tsar Nicholas II feared the very loss of his throne if he chose any other option than a march to war. He chose war, and the people eventually toppled him all the same.

Reading the frantic diplomatic missives that traversed the telegraph lines in the final days before hostilities commenced, one is struck by how the threat of conflict quickly adopted its own terrible logic and momentum. Impassioned populations and real-time reports of mobilizations and countermobilizations helped fuel a sense that, far from a conscious choice, war had become inevitable.

Notably, this was the prevailing mood in an age when all the European royal families were related, when diplomats hailed from the same genial institutions, when governments exercised vastly more power over the popular press than they do now. Lines of communication were largely controlled by the state, and formal correspondence usually unfolded over days, not hours or minutes.

Today, national leaders engage in Twitter spats, and rapid-fire hashtags draw international attention. Public sentiment can be readily manipulated or even manufactured. And events, filtered through social media, can quickly go viral—the very definition of spinning out of control.

Perhaps the greatest danger in this dynamic is that, although information that goes viral holds unquestionable power, it bears no special claim to truth or accuracy. Homophily all but ensures that. A multi-university study of five years of Facebook activity, titled “The Spreading of Misinformation Online,” was recently published in Proceedings of the National Academy of Sciences. Its authors found that the likelihood of someone believing and sharing a story was determined by its coherence with their prior beliefs and the number of their friends who had already shared it—not any inherent quality of the story itself. Stories didn’t start new conversations so much as echo preexisting beliefs.

This extreme ideological segregation, the authors concluded, “comes at the expense of the quality of the information and leads to proliferation of biased narratives fomented by unsubstantiated rumors, mistrust, and paranoia.” As smartphone cameras and streaming video turn every bystander into a reporter (and everyone with an internet connection into an analyst), “truth” becomes a matter of emotional resonance.

Every Wire a Nerve

News of the meeting between the U.S. Army colonel and the unpopular governor of Kirsham province had seeped into social media long before the colonel’s convoy arrived in the muggy town of Dara Lam. Angry citizens began organizing a demonstration in front of the consulate building. Their #justice4all hashtag, which had begun trending around the world, did not escape the notice of the local Faqih insurgency. Rebel proxies infiltrated the online discussions, seeking to whip the protesters into a frenzy. Their plan was to ambush the U.S. troops as they exited the building, using the protesters as human shields. Cameramen stood ready to record the attack and post it quickly to a network of rebel social-media accounts: The massacre would be live-streamed, and it would likely go viral.

But others had also noticed this flurry of social-media activity. At the U.S. Army brigade’s tactical-operations center, news of the brewing online storm was passed swiftly up the chain of command. The colonel and his escort cut the meeting in Dara Lam short and discreetly left through a back entrance of the governor’s mansion. The attack was averted. The Faqih propaganda machine was out of luck.

You will not find a record of this event in the news, just as you will not find the town of Dara Lam on a map. It is a fake settlement in a fake province in a fake country, all part of a fake war that breaks out every few months in Louisiana, at Fort Polk’s 72,000-acre Joint Readiness Training Center.

Fort Polk has played an outsize role in U.S. military history. It was the site of the “Louisiana Maneuvers” in the early 1940s, when the Army made its transition from horses to armored tanks and trucks as it prepared to plunge into World War II. The fort has since become the military’s training ground for new kinds of conflict—first the simulation of Cold War– and Desert Storm–style mechanized maneuvers, and then, after 9/11, the complex counterinsurgency operations of Afghanistan and Iraq. It is replete with fake villages, “opposing forces” that use the tactics of the Taliban and ISIS, and even actors playing local civilians and hard-nosed journalists. And it is at Polk where the U.S. military is now learning how to fight a social-networked war.

When units deploy into the simulated battles of Kirsham, they must navigate a new addition that mimics what is happening in the real wars beyond: the smeir. Short for “Social Media Environment and Internet Replication,” smeir is a fake internet of blogs, international media outlets, and social-media accounts, all woven together to form a virtual battlefield atop the physical one. Units go out on patrol, villagers tweet about their movements, and the insurgents reshape the story to aid their recruiting—just as in real life.

Major Marc Meyle, an Army intelligence officer who helped create these scenarios, told us the idea is for the soldiers to be “challenged in a full environment: good guys, bad guys, people that can be swayed either way, multiple means of communication being thrown around.” He said that’s just “the way the world is today.”

There is yet no consensus on how the U.S. military should operate in an environment saturated by smartphones and near-universal internet access. Each brigade rotating through Polk handles the experience differently: Some shift their operations as a result of online chatter; others ignore it.

This experiment at Polk represents a first, small step in tackling a vast new operational challenge. Soldiers are learning that social media is an effective way to keep track of the enemy—but also that they are likewise being tracked. Yet even this challenge pales in comparison to the questions that will soon confront the nation that sends these soldiers into battle. When information carries so much power, who, exactly, is a combatant? Will it ever be an American duty to defriend enemy citizens on Facebook? To enlist in “information brigades” to help push back against hostile propaganda?

These questions are no longer so fanciful. In recent conflicts, Israel has established Hasbara (the Hebrew word for “explanation”) war rooms, filled with university students and soldiers who tangle with Hamas and Palestinian sympathizers over what, exactly, is going on in their perpetual war. The scale of this online jousting is astounding. During the 2014 flare-up in Gaza, for example, the two sides’ competing hashtags, #GazaUnderAttack and #IsraelUnderFire, racked up some 5 million uses. The Wikipedia page about the conflict has been edited and reedited more than 7,000 times.

The Israel Defense Forces are very active on social media. During conflicts, “war rooms” of Israeli soldiers and students vie with Palestinians to shape global perceptions.

Even the ranking and targeting of enemies has begun to change. Take the case of Junaid Hussain. A British Muslim former “hacktivist” (who went by the handle “Trick”) and wannabe rapper, Hussain went to jail in 2012, at age 18, for hacking the personal information of former Prime Minister Tony Blair. In prison, he was radicalized and, once out, was seduced by the Islamic State’s online appeals. He went to Syria and set to work spreading ISIS propaganda across social-media platforms. “You can sit at home and play Call of Duty or you can come here and respond to the real call of duty,” his new “Abu Hussain al Britani” persona tweeted to his followers. “The choice is yours.”

By August 2015, Hussain had reportedly become the third-most-important name on the anti-ISIS coalition’s “kill list”—behind only the Islamic State’s self-declared caliph and its top battlefield commander. Rather than any battlefield skills or strategic brilliance, however, it was his social-media-marketing skills that led military planners to prioritize his killing. Ironically, it was also his nonstop internet use that enabled his execution. Hussain was reportedly tricked into clicking a link in a messaging app that had been compromised by British intelligence, allowing him to be geolocated and killed by a Hellfire missile. (A year later, ISIS posted new footage of a young, blue-eyed British boy executing a prisoner with a bullet to the back of the head. British media reported him to be the son of Sally Jones, a former punk rocker who had fallen in love with Hussain online, then joined him in Syria, her 10-year-old son in tow.)

The convergence of online speech and physical violence creates new dilemmas not only for nations, but also for the companies that created and are now responsible for the digital landscape. The small start-ups that have blossomed into tech giants must navigate the limits of neutrality in a digital battlefield of their own design.

This is not just a matter of policing social networks for violent, extremist content—a tough enough task already. If terrorists are to be banned from many popular platforms, for instance, who constitutes a terrorist? An ISIS fighter? An abortion-clinic bomber? An alt-right neo-Nazi? An activist from China’s long-restive Uighur minority? A Black Lives Matter protester? These decisions can carry extraordinary political consequences.

Some experts argue that the focus should be not on the group, but on the content. Yet where should these lines be drawn—and by whom? Banning videos of killings might seem reasonable as a means of curtailing ISIS horror shows, but consider the July Facebook live-stream that showed Philando Castile, bloodied and dying, after being shot by a Minnesota police officer. That video was as brutal as those produced by some terrorist groups (and was briefly removed by Facebook moderators), yet it also prompted renewed discussion in American society about issues of race and policing.

These questions already seem intractable, but what of something we have not seen since the internet came into existence: a war between great powers? Most of us did not associate Twitter with terrorism until the Islamic State stormed into Mosul. We have given similarly scant thought to what might happen if the wondrous tools of the 21st century are ever paired with the scale and intensity of the conflicts that defined the 20th.

What might the responsibility of social-media giants be if a naval skirmish in the Pacific escalated into confrontation between the U.S. and China, or if the conflicts in the gray zone of Ukraine or the Baltics spiraled toward hostilities between NATO and Russia? With its user base of 1.7 billion, Facebook, for instance, can affect the tone and tenor of national debates with even tiny tweaks to the algorithms that govern its News Feed.

Russian social-media users and allied accounts seek to manipulate opinion and sow dissension in enemy states.

This power has already become a point of concern in domestic politics, but what of war, during which subtle changes to the flow of information might grant an immense boon to one side over the other? In such a charged environment, remaining “neutral” would itself be a momentous choice. As Zeynep Tufekci, an expert on the influence of technology on politics and society at the University of North Carolina, has observed in The New York Times, “What we are shown is shaped by these algorithms, which are shaped by what the companies want from us, and there is nothing neutral about that.”

And finally, what of the most drastic measures: Could the United States government, under great duress in some future conflict or catastrophe, censor or nationalize the social-media industry? Extreme as it may sound, there is ample precedent. During the Civil War, Abraham Lincoln ordered the censorship of telegrams. Twelve days after Pearl Harbor, Franklin Roosevelt formally established the U.S. Office of Censorship—its official motto was “Silence speeds victory.” Would such control of social media be advisable? Would it even be possible?

None of these issues has an easy answer. Yet these are the dilemmas that will come to define the social-media age as it confronts the timeless challenge of war. National leaders will have to reckon with a social-media environment that seeds violence through vast digital networks and a public that has never spoken with so loud and so immediate a voice. And they will face new kinds of conflict shaped by the internet’s next iteration.

The emergence of a truly interconnected world has long been hailed as a step toward cross-cultural cooperation and global enlightenment. As societies communicate more freely, the thinking has gone, empathy will be nourished, the truth will be easier to find, and many causes of conflict will wither. Thanks to the mobilizing power of social media and the resultant “wisdom of crowds,” citizens will exert more direct control over their governments, helping solve disputes without need for violence. The age of social media, in other words, should be an age of peace and understanding.

The same was once said of the telegraph.