The “security dilemma” is a central concept in the academic study of international politics and foreign policy. First coined by John Herz in 1950 and subsequently analyzed in detail by such scholars as Robert Jervis, Charles Glaser, and others, the security dilemma describes how the actions that one state takes to make itself more secure—building armaments, putting military forces on alert, forming new alliances—tend to make other states less secure and lead them to respond in kind. The result is a tightening spiral of hostility that leaves neither side better off than before.
If you’ve taken a basic international relations class in college and didn’t learn about this concept, you may want to contact your registrar and ask for a refund. Yet given its simplicity and its importance, I’m frequently struck by how often the people charged with handling foreign and national security policy seem to be unaware of it—not just in the United States, but in lots of other countries too.
Consider this recent propaganda video tweeted out from NATO headquarters, responding to assorted Russian “myths” about the alliance. The video points out that NATO is a purely defensive alliance and says it harbors no aggressive designs against Russia. These assurances might be factually correct, but the security dilemma explains why Russia isn’t likely to take them at face value and might have valid reasons to regard NATO’s eastward expansion as threatening.
Adding new members to NATO may have made some of these states more secure (which is why they wanted to join), but it should be obvious why Russia might not see it this way and that it might do various objectionable things in response (like seizing Crimea or invading Ukraine). NATO officials might regard Russia’s fears as fanciful or as “myths,” but that hardly means that they are completely absurd or that Russians don’t genuinely believe them. Remarkably, plenty of smart, well-educated Westerners—including some prominent former diplomats—cannot seem to grasp that their benevolent intentions are not transparently obvious to others.
Or consider the deeply suspicious and highly conflictual relationship among Iran, the United States, and the United States’ most important Middle East clients. U.S. officials presumably believe that imposing harsh sanctions on Iran, threatening it with regime change, conducting cyberattacks against its nuclear infrastructure, and helping organize regional coalitions against it will make the United States and its local partners more secure. For its part, Israel thinks assassinating Iranian scientists enhances its security, and Saudi Arabia thinks intervening in Yemen makes Riyadh safer.
Not surprisingly, according to basic IR theory, Iran sees these various actions as threatening and responds in its own fashion: backing Hezbollah, supporting the Houthis in Yemen, conducting attacks on oil facilities and shipments, and—most important of all—developing the latent capacity to build its own nuclear deterrent. But these predictable responses just reinforce its neighbors’ fears and make them feel less secure all over again, tightening the spiral further and heightening the risk of war.
The same dynamic is operating in Asia. Not surprisingly, China regards America’s long position of regional influence—and especially its network of military bases and its naval and air presence—as a potential threat. As it has grown wealthier, Beijing has quite understandably used some of that wealth to build military forces that can challenge the U.S. position. (Ironically, the George W. Bush administration once tried to tell China that pursuing greater military strength was an “outdated path” that would “hamper its own pursuit of national greatness,” even as Washington’s own military spending soared.)
In recent years, China has sought to alter the existing status quo in several areas. As should surprise no one, these actions have made some of China’s neighbors less secure, and they have responded by moving closer together politically, renewing ties with the United States, and building up their own military forces, leading Beijing to accuse Washington of a well-orchestrated effort to “contain” it and of trying keep China permanently vulnerable.
In all these cases, each side’s efforts to deal with what it regards as a potential security problem merely reinforced the other side’s own security fears, thereby triggering a response that strengthened the former’s original concerns. Each side sees what it is doing as purely defensive reaction to the other side’s behavior, and identifying “who started it” soon becomes effectively impossible.
The key insight is that aggressive behavior—such as the use of force—does not necessarily arise from evil or aggressive motivations (i.e., the pure desire for wealth, glory, or power for its own sake). Yet when leaders believe their own motives are purely defensive and that this fact should be obvious to others (as the NATO video described above suggests), they will tend to see an opponent’s hostile reaction as evidence of greed, innate belligerence, or an evil foreign leader’s malicious and unappeasable ambitions. Empathy goes out the window, and diplomacy soon becomes a competition in name-calling.
To be sure, a few world leaders have understood this problem and pursued policies that tried to mitigate the security dilemma’s pernicious effects. After the Cuban missile crisis, for example, U.S. President John F. Kennedy and Soviet Premier Nikita Khrushchev made a serious and successful effort to reduce the risk of future confrontations by installing the famous hotline and beginning a serious effort at nuclear arms control.
The Obama administration did something similar when it negotiated the nuclear deal with Iran, which it saw as a first step that that blocked Iran’s path to the bomb and opened up the possibility of improving relations over time. The first part of the deal worked, and the Trump administration’s subsequent decision to abandon it was a massive blunder that left all the parties worse off. As the former Mossad chief Tamir Pardo has observed, Israel’s extensive efforts to convince then-U.S. President Donald Trump to withdraw from the deal was “one of the most serious strategic mistakes since the establishment of the state.”
As the writer Robert Wright recently pointed out, then-U.S. President Barack Obama’s decision to not to send arms to Ukraine after the Russian seizure of Crimea in 2014 showed a similar appreciation of security dilemma logic. In Wright’s telling, Obama understood that sending Ukraine offensive weapons might exacerbate Russian fears and encourage the Ukrainians to think they could reverse Russia’s earlier gains, thereby provoking an even wider war.
Tragically, this is pretty much what happened after the Trump and Biden administrations ramped up the flow of Western weaponry to Kyiv: The fear that Ukraine was slipping rapidly into the Western orbit heightened Russian fears and led Putin to launch an illegal, costly, and now protracted preventive war. Even if it made good sense to help Ukraine improve its ability to defend itself, doing so without doing very much to reassure Moscow made war more likely.
So, does the logic of the security dilemma prescribe policies of accommodation instead? Alas, no. As its name implies, the security dilemma really is a dilemma, insofar as states cannot guarantee their security by unilaterally disarming or making repeated concessions to an opponent. Even if mutual insecurity lies at the core of most adversarial relationships, concessions that tipped the balance in one side’s favor might lead it to act aggressively, in the hopes of gaining an insurmountable advantage and securing itself in perpetuity. Regrettably, there are no quick, easy, or 100 percent reliable solutions to the vulnerabilities inherent in anarchy.
Instead, governments must try to manage these problems through statecraft, empathy, and intelligent military policies. As Jervis explained in his seminal 1978 World Politics article, in some circumstances the dilemma can be eased by developing defensive military postures, especially in the nuclear realm. From this perspective, second-strike retaliatory forces are stabilizing because they protect the state via deterrence but do not threaten the other side’s own second-strike deterrent capability.
For example, ballistic missile submarines are stabilizing because they provide more reliable second-strike forces but do not threaten each other. By contrast, counterforce weapons, strategic anti-submarine warfare capabilities, and/or missile defenses are destabilizing because they threaten the other side’s deterrent capacity and thus exacerbate its security fears. (As critics have noted, the offense-versus-defense distinction is much harder to draw when dealing with conventional forces.)
The existence of the security dilemma also suggests that states should look for areas where they can build trust without leaving themselves vulnerable. One approach is to create institutions to monitor each other’s behavior and reveal when an adversary is cheating on a prior agreement. It also suggests that states interested in stability are usually wise to respect the status quo and adhere to prior agreements. Blatant violations erode trust, and trust once lost is hard to regain.
Lastly, the logic of the security dilemma (and much of the related literature on misperception) suggests that states should work overtime to explain, explain, and once again explain their real concerns and why they are acting as they are. Most people (and governments) tend to think their actions are easier for others to understand than they really are, and they are not very good at explaining their conduct in language that the other side is likely to appreciate, understand, and believe. This problem is especially prevalent at present in relations between Russia and the West, where both sides seem to be talking past each other and have been surprised repeatedly by what the other side has done.
Giving bogus reasons for what one is doing is especially harmful, because others will sensibly conclude that one’s words cannot be taken seriously. A good rule of thumb is that adversaries will assume the worst about what you are doing (and why you are doing it) and that you must therefore go to enormous lengths to persuade them that their suspicions are mistaken. If nothing else, this approach encourages governments to empathize—i.e., to think about how the problem looks from their opponent’s perspective—which is always desirable even when the opponent’s view is off-base.
Unfortunately, none of these measures can fully eliminate the uncertainties that bedevil global politics or render the security dilemma irrelevant. It would be a more secure and peaceful world if more leaders considered whether a policy they believed was benign was unintentionally making others nervous, then considered whether the action in question could be modified in ways that alleviated (some of) those fears. This approach won’t always work, but it should be tried a more often than it is.