Deconstructing Deterrence

Deterrence is a peculiar concept. Its effectiveness is measured by actions that don’t take place. Actions that are often called deterrence may have nothing to do with the actions that don’t take place.

It’s hackneyed to start with a definition, but this is a confused enough subject that let’s do it. Here’s what Google gives me, from Oxford Languages:

de·ter·rence  /dəˈtərəns/  noun  the action of discouraging an action or event through instilling doubt or fear of the consequences. “nuclear missiles remain the main deterrence against possible aggression”

To keep it simple, let’s assume two actors. Actor One may be contemplating an action that Actor Two finds undesirable. So Actor Two takes an action to instill doubt or fear of the consequences in Actor One’s mind.

A great deal of mindreading is involved in deterrence. Actor Two believes that Actor One is contemplating an action. Actor Two cannot know this unless Actor One states that they are, and even then, Actor One may be bluffing. When Actor Two takes their action, they have no way of knowing whether it prevents Actor One’s action or if Actor One never planned to take that action. In outline,

  1. Actor One considers taking an action that may be a threat to Actor Two
  2. Actor Two takes an action to demotivate Actor One from that action
  3. Actor One does not take that action.

This sequence represents deterrence only if both actors interpret their actions the same way. If Actor One considers their action defensive and Actor Two considers it a threat, then the situation is what political scientists call a security dilemma. If Actor One takes that action in point 3 in an escalatory spiral. That’s political science language again. Arms races are an example. This small difference from deterrence is why it’s important to understand whether an action called deterrence is in fact that.

In another small difference, if Actor One is indifferent to Actor Two’s action, the result may look like deterrence, but it may be more akin to snapping one’s fingers to keep the elephants away. You don’t see any elephants, do you? Snapping fingers must work. One of the difficulties of a situation in which inaction is the desired result.

Let’s make that concrete. During the Cold War, the United States and Russia built up their nuclear arsenals competitively. At first, the motivation was that a war might be fought with them, but with intercontinental ballistic missiles, it became obvious that the end of such a war would be total destruction. As both sides built significant numbers of nuclear weapons, the justification became deterring the other side from a first strike.

As an American, I learned that we would never carry out a first strike. I suspect that young Soviets learned the same thing about themselves. If neither side intended a first strike, then the first act of mindreading – interpreting the action that the other side intends – failed.

Have we gone 75 years or so without a nuclear explosion in war because both the Soviet Union and the United States had enough nuclear weapons to destroy the world several times over? Or was it because both sides felt that making a better life for their people took precedence over a nuclear war today? Or was it some of both?

The concept of deterrence is used very sloppily, not least by the US Department of Defense. What is attractive about it is that it imputes agency.* To claim deterrence means that one’s actions have caused another to modify their actions, making the deterrer more powerful. We often see claims of deterrence that depend only on the deterrer’s state of mind. This modifies the sequence:

  • Actor Two claims that Actor One presents a threat
  • Actor Two asserts that particular actions or equipment will deter that threat
  • Actor Two takes those actions or acquires that equipment

This is, of course, the budget sequence for the Department of Defense. The claim in point 4 may have more or less support, but seldom is a case made publicly that the adversary sees the proposed action as having deterrent force. But Congress buys the deterrence argument easily, and the funds flow.

Nuclear deterrence is a subset of deterrence, which refers to overall military postures. Throughout the Cold War, the media became accustomed to thinking of deterrence only in reference to nuclear matters. Puzzlement results over why Russia attacked Ukraine and why the United States or NATO doesn’t push Russia back. A desire for agency translates to wishes about how deterrence works.

President Joe Biden is heading up an innovative information offensive against Russia. Like any offensive in a war, it has had mixed success. I plan to write more about it. For this post, I’ll focus on one aspect, denying Russia the opponent Vladimir Putin wants.

Putin’s motives in attacking Ukraine are complex. His history says that Ukraine and Russia are historically connected in a way that demands governmental unity. He has claimed a nonexistent NATO buildup as a threat to Russia. He has stated that Ukraine would be a friend of Russia if not for the malign influence of NATO. Although he has decided to punish Ukraine, NATO (and the US and EU, which tend to be conflated in Putin’s writing) is a prime enemy.

Biden’s declaration that the United States will not fight in Ukraine explicitly undercuts Putin’s perception of a threat, as in point 1. This open and explicit declaration pre-empts any Putin claim that the United States is deterred, as does the American arming of Ukraine. The decision not to fight is made by the United States, for its own reasons.

Biden’s declaration puts the responsibility for the carnage in Ukraine fully on Putin. Uncertainty about whether the United States would enter Ukraine would allow Putin to regard the US/NATO as his opponent. Facing the US/NATO in the war he started would put him on an equal footing with them, something he very much desires. Biden’s statements make that impossible. Biden’s declaration is embedded in a broader information offensive that has revealed Russian plans and has likely deterred (yes) and delayed several Russian actions in the war.

If the US/NATO remains uninvolved in any direct way, the motivation to escalate to nuclear use also declines. All escalation now is Putin’s responsibility. Russian doctrine is that nuclear weapons will only be used in case of existential danger to the Russian nation, something that is difficult to argue in a war with Ukraine.

Russia has the initiative to escalate to war crimes, which it has used, or to end the fighting, in which it is the sole maintainer.


A few hours after I posted “Deconstructing Deterrence,” this exchange occurred on Twitter.

This is precisely the kind of thing my post is aimed at. By valorizing the abstraction “deterrence” and assuming that both sides view the actions similarly in sequence 1-3, one may manage to accuse another of circular reasoning while saying “Deterrence works only if one deters.”

What Chipman is describing is an escalation of threats, not unlike the famous game of chicken, in which the “winning” move is to throw the steering wheel out the window.

We can’t continue to do this.


*In reviewing this post, Martin Pfeiffer (@nuclearanthro) made the excellent point that a claim of deterrence is also a moral claim. That point deserves a great deal more thought.

Graphic from Jeffrey Lewis

Cross-posted to Lawyers, Guns & Money