In late June, after Iranian-backed militias launched a drone attack against U.S. troops in Iraq, U.S. fighter jets responded by dropping bombs on the militias’ facilities in Iraq and Syria. A Pentagon spokesman said the bombing was meant as “a clear and unambiguous deterrent message”—i.e., don’t attack us again, or we’ll attack you again.
And yet, on Wednesday, less than two weeks after the bombings, the same militia fired 14 rockets at an Iraqi air base, which was hosting U.S. forces.
It seems the “deterrent message” didn’t get through.
Then again, deterrent messages often don’t get through. Or they get through, but the recipient isn’t deterred. Many countries, very much including the United States, frequently issue threats, mount attacks, or take other punitive measures as a way of pressuring another country (or militia or some other entity) to stop doing what it’s doing or not to do it in the first place.
Sometimes the pressure works; sometimes it doesn’t. However, people in power—here and elsewhere—don’t seem to be studying what works and what doesn’t, or why it sometimes works and sometimes doesn’t. If they were asking these questions in a serious way, they wouldn’t keep doing the same thing over and over, to little avail.
Small-scale attacks (whether from bombs, rockets, missiles, or drones) are classic examples. Israeli leaders would have stopped the rocket attacks from Gaza long ago, if their incessant shelling of Hamas and other terrorist groups had successfully communicated the “deterrent message.” But no, Hamas keeps firing back.
Sanctions are another tried-and-not-so-true method of deterring adversaries from continuing to do adversarial things. Sometimes the sanctions work—especially when the “stick” is coupled with a “carrot.” For instance, the talks that led to the Iran nuclear deal were brought about, in large part, by U.S. sanctions coupled with a proposal to lift those sanctions if Iran dismantled its nuclear facilities. Sometimes they don’t work. For instance, the threat to impose sanctions on Russia and China, in response to their cyberattacks, has had little if any effect.
A distinction should be made between deterrence-through-the-use-of force and more passive or existential forms of deterrence. An example of the latter is deploying troops to a region as a warning (here is my army; if you invade this territory, I will fight back). Since the mid-20th century, this has been expanded to include nuclear deterrence (here are my nuclear weapons; if you strike me, I will strike you back) and, in a variation on that, extended deterrence (if you attack my allies, even if just with conventional weapons, I might strike you back with my nuclear weapons).
This sort of deterrence works—until it doesn’t, at which point big, potentially catastrophic wars happen—and it can work for a long time because the country threatened with nuclear retaliation is loath to take a gamble, as the consequences of miscalculation are too enormous.
However, this sort of deterrence doesn’t seem to prevent smaller forms of aggression. During the Cold War, the United States and the Soviet Union held each other at bay with their mutual threats of annihilation—but those threats didn’t deter either superpower (or their proxies) from invading smaller countries or from engaging in acts of political subversion. Nor do mighty nuclear arsenals necessarily make small powers, such as the Iranian-backed militias in Iraq and Syria, cower before Washington’s will. A nuclear threat against them wouldn’t be believed. Nuclear deterrence seems to deter only nuclear war—or possibly a large-scale conventional war. (It could be argued, for instance, that India and Pakistan would have gone to war with each other far more frequently and fiercely if both sides did not have nuclear weapons.)
Even in the realm of nuclear weapons, it is not at all clear how many nukes a country needs to deter another country from aggression. U.S. (and presumably Russian) officials invoke (largely theoretical) calculations to justify the need for thousands of nuclear warheads. However, an otherwise puny and impoverished country like North Korea has been able to deter its enemies from invasion—and even, on occasion, compelled them to provide economic assistance—because it has, at most, a few dozen atomic bombs and seems desperate enough to use them if pushed against the wall.
“Cyberdeterrence” is a big topic these days, as President Joe Biden—like a few presidents before him—has threatened Russia and China with “consequences” if they keep launching or sponsoring cyberattacks or ransomware attacks against U.S. industries, power plants, waterworks, or other “critical infrastructure.” But the intrusions continue. (It should be noted that the U.S. mounts “cyberoffensive operations” against other countries too, which complicates matters.)
Some have tried to apply the principles of nuclear deterrence to conflict in cyberspace, but there is no analogy between the two. From the start of the nuclear age, a bright, bold line has been drawn between the use and nonuse of nuclear weapons, so any use of nukes—even a “limited” number of “small” nukes—could bring on Armageddon. However, cyberattacks of various types and magnitude are mounted thousands of times a day. In terms of deciding whether and how to retaliate, where should a president draw the line between a serious threat and a mere nuisance? Dozens of special commissions have attempted to parse the distinctions and to outline various orders of response, but no administration has set an official standard. In other words, in the cyber domain, the basic question of deterrence hasn’t yet been addressed: What specifically are we trying to deter?
This leads us back to the sort of deterrence that militaries try to deal with routinely: how to stop attacks in a war zone or in a hot zone that isn’t quite a war zone—for instance, attacks like those mounted by the Iran-backed militias in Iraq and Syria.
The key element of deterrence is to threaten or impose a penalty that exceeds whatever gains the militia might attain by continuing the attacks. To figure out what deters, we would first have to figure out what the militias want, how much effort they need to exert to get what they want, and whether we can live with the situation if they get it.
Here’s a cautionary tale. Thomas Schelling (who died in 2016) was the master deterrence strategist. His books The Strategy of Conflict and Arms and Influence are classics in the field. Schelling wrote eloquently—at times, all too casually—about sending messages with force. There are, he once observed,
enlightening similarities between, say, maneuvering in limited war and jockeying in a traffic jam, deterring the Russians and deterring one’s own children … between the modern balance of terror and the ancient institutions of hostages.
He also wrote: “War is always a bargaining process.” One must use force in a way that exploits “the bargaining power that comes from [the] capacity to hurt,” to cause “sheer pain and damage,” in order to pressure the enemy to avoid further pain and damage.
Back in 1964, senior officials in Lyndon Johnson’s administration laid plans to step up military action in the Vietnam War. Secretary of Defense Robert McNamara’s closest adviser on Vietnam, John McNaughton, was a close friend and former Harvard colleague of Schelling, and he drew on Schelling’s work to devise a strategy that would affect the “will” of the North Vietnamese army, to “deter” it from further fighting. Nobody knew how to do this, so McNaughton consulted his old friend, Tom Schelling. (This episode of history, which almost no one has cited, comes from my 1983 book The Wizards of Armageddon and is based on my interviews with Schelling.)
The two men spent more than an hour discussing what the U.S. could command the North Vietnamese to stop doing that they would obey as a result of our bombing.
In the end, Schelling—the master theorist of deterrence strategy and limited war—could not come up with a single plausible answer when faced with a real limited war.
McNamara and Johnson began their bombing campaign anyway. Schelling did tell McNaughton that, whatever they decided to do, it would work within three weeks or not at all. On March 24, 1965, almost three weeks to the day after the bombing began, McNaughton wrote McNamara: “The situation in Vietnam is bad and deteriorating.” That fact never changed, but McNamara and his successors kept trying anyway. (McNaughton died in an airplane crash in 1967.)
The point is this: In wars, big or small, sometimes it’s not clear how to deter adversaries from doing or not doing what you want them to do or stop doing. Figure out that problem before you start dropping the bombs. In any case, stop talking loosely about sending “deterrent messages”—because if you keep talking that way, and the militias aren’t deterred, our messages on myriad matters will be taken less and less seriously everywhere.