There is nothing social about using social networks to help slaughter innocents and disseminate murder porn. But that kind of activity is merely symptomatic of the fundamental challenge that social media poses to national security. The key to winning cyberwars has more to do with securing physical space than dominating cyber space.

Social networks are powerful. For starters, they can scale very quickly. The number of people that one person can reach can be pretty impressive. President Obama made news when he set the world record for the number of Twitter followers he attracted in one day. Yet he was humbled just days later when Caitlyn Jenner eclipsed the Tweeter-in-chief’s record in barely four hours. (Still, neither can match rock star Katy Perry, who has more than seventy million followers on Twitter.)

Of course, it doesn’t require a hit single to establish a formidable presence online. Social networkers with very little reach can also have outsized impact. Their contributions may go “viral,” picked up and retransmitted by others. In 2010, for example, a seat belt video by a local safety advocacy group in rural England garnered one million views on YouTube in less than two weeks. So far, it has been seen by over nineteen million people in more than 120 countries.

Most social networks conform to what is called the “power curve,” with a few contributors dominating the preponderance of activity on the network. This “high ground” of influence is called “broadcast mode.” Those who can get a dominant influencer to convey their message have a great competitive advantage when it comes to driving the conversation on the network.

There is also, however, a second high ground on social networks. That space is on the other far end of the curve called “conversation mode.” Something really interesting happens when networks scale down to very small groups. The level of participation among the members is more balanced. This allows for more high-quality conversations.

Terrorists have already figured out that the advantage is to work both ends of the curve. Thus, they are thrilled when an execution video goes viral and grabs the world’s attention.

On the other hand, extremists also like to lure individuals into small group conversation where they can attract new recruits or radicalize the other discussants. “Some of these conversations occur in publicly accessed social networking sites,” noted Michael B. Steinbach, the FBI’s Assistant Director for the Counterterrorism Division recently before one Congressional Committee, “but others take place via private messaging platforms.”

“As technology advances,” Steinbach added, “so, too, does terrorists’ use of technology to communicate.” And there is little doubt they are having success. Terrorist and extremist groups use the Internet and the social networks it carries to conduct all kinds of activities, from recruiting, fundraising and propaganda to intelligence gathering and mission planning. And this leads to dangerous actions. At least six Islamist terror plots have been thwarted already this year. That’s a noticeable uptick in frequency.

But skillful use of the Internet doesn’t make the terrorists ten feet tall. The Internet is neutral. It confers no inherent advantages to any state or nonstate group. Counterterrorist operations can go online and conduct the same activities on the same systems. Or, they can use the terrorists’ online presence against them by gathering intelligence about the group, disrupting their online activities, or just trolling them to annoy and distract.

But it’s important to recognize that the terrorist operating online is not the root of the problem. Fixating on terror talking would be like trying to lower a fever and ignoring the underlying infection.

What makes any network powerful is when the online community links to a physical community or a “human web” that has a propensity to act.

What if Katy Perry went online and tweeted that everyone should slaughter their pets and string them up on the streetlamps? A ludicrous hypothetical, yes. But equally ludicrous is the idea that, in such a scenario, Fluffy, FiFi and Fido would be in jeopardy. That’s because Ms. Perry’s social presence isn’t linked to a group with common cause in physical space.

On the other hand, when messaging connects with a physical group, like the terror team inspired by the Boston bombing to replicate the attack in Canada, bad stuff does start to happen.

The real “center of gravity,” that which turns terror talk into threat, is the physical network. Taking down the real-life terrorist community is the best way to diminish the danger posed by online actors.

Rather than chasing shadows on the Web, the main thrust of an effective counterterrorism strategy ought to be directed toward crushing the bad guys. This approach to winning online probably goes for other malicious online competitors as well. Rather than get into a symmetrical struggle of “our” electrons versus “theirs”—an effort that usually involves sweeping censorship, massive data mining and continuous online propaganda—better to look at asymmetrical responses that can cripple a competitor’s human web. When the physical threat is diminished, the online threat is little more than an angry troll.

James Jay Carafano, the author of “Wiki at War: Conflict in a Socially Networked World,” is a vice president of the Heritage Foundation. He directs the think tank’s research on national security and foreign affairs.