Establishing an agreement imposing restrictions on the prior use of nuclear weapons

The war in Ukraine has brought the use of nuclear weapons into the realm of possibility. Comprehensive nuclear war is no longer just a theoretical possibility, and such an event would involve the entire human race. To follow Clemenso’s sarcastic quote about the importance of being left to warlords, the use of nuclear use and the growth of nuclear deterrence strategies is directly related to our lives we will ignore that and leave it to military and security experts. In this column, I examine nuclear deterrence strategies as a story in which everyone can participate.

The logic of mutually convincing destruction is that “if nuclear power is attacked by nuclear weapons, it will retaliate and destroy the other party and no one will survive.” So, no one is going to start a nuclear war ”- is dangerous. There is a vulnerability and distortion in this logic. For an analysis of conflict and nuclear deterrence within the framework of game theory, see Shelling (1960), Zagare (1992) and Craig (1999).

First, vulnerability: the logic of mutually convincing destruction is correct in terms of game theory, but the assumptions behind the findings are very strong. The assumption is that, under perfect information, players are fully aware of their opponent’s behavioral objectives and will act perfectly rationally according to their own behavioral objectives. In reality, however, such an assumption is untenable. The world knows that information, intentions and everything else about Russian President Vladimir Putin’s statements and actions are opaque.

The distortion of the logic of mutually convincing destruction is beyond the means of achieving the goal of the conflict. According to the logic of mutually convincing destruction, each country’s goal is to ‘survive in the world’ and each country treats the ‘world’ as a tool (i.e. hostage). But the balance between the goal (the existence of individual nations in the world) and the means used to achieve it (the destruction of the world) is not balanced.

This is to say that the Tokyo-controlled gang and the Osaka-controlled gang are fighting to increase their power, and if the Osaka gang loses, it will kill all 20 million Tokyo residents. Such a threat, if fulfilled, would make any goal (in this case, an extension of gang power) meaningless.

First, there is no validity in a game in which the parties to the conflict have (largely undisputed) the right to determine the existence of the whole world. It is equally unjust to assume that gangs have the right to kill or not kill the entire population of cities.

Such insecurities and distortions need to be addressed and stable nuclear deterrence considered. To overcome insecurity, nuclear forces must be prevented from climbing the ladder of nuclear growth because climbing that ladder increases the likelihood of stress and unforeseen events.

To overcome the distortion, we also need to define a game that involves the entire world (the rest of the world) outside of the parties to the conflict as players, who have nothing to do with the conflict but will suffer huge losses due to the use of nuclear weapons. .

The victims of nuclear war from the rest of the world are not the only ones currently living in non-partisan countries. All future generations are victims that countless generations do not exist – not just humans but every living creature on earth. Whatever the cause, motive or cause of the conflict between the two parties, it cannot support endangering the future of all living things on our planet.

In 2,000 years, the current invasion of Ukraine would seem like a futile struggle, not a threat to the very existence of humanity – just as the causes of the Punic Wars 2,000 years ago are not important to us today. Therefore, for the whole world, the main goal is to prevent conflicting parties from climbing the ladder of nuclear growth.

As a mechanism to prevent any party from climbing the first step of the nuclear ladder, we may consider creating an agreement that imposes restrictions on the prior use of nuclear weapons. Such an agreement would “impose immediate, unconditional and maximum sanctions on any nation launching a nuclear pre-emptive strike.” All non-military sanctions may be used, including economic sanctions and other enforcement measures, such as exclusion from the international framework.

It is like establishing a new ideal that ‘nuclear pre-emptive strike, a very limited one, is a criminal act against humanity because it threatens the very existence of all human beings not related to the conflict in question.’

Note that denying the use of nuclear weapons is not consistent with the current Japanese government’s ideology. While it is unlikely that the Japanese government will make such a proposal in the current circumstances, it makes sense to start with a blank slate to discuss new criteria for nuclear deterrence, including theoretical possibilities, in response to the tail risk of nuclear use.

Consider the effectiveness of the new agreement in the figure. Figure 1 shows the usual bilateral game of nuclear inhibition. In the conflict between nuclear power A and B, A decides whether to launch a strike with limited use of nuclear weapons (LS) (NS) then B decides whether to retaliate (C) (NC), and then A counterattack (C). Decides whether or not (NC). If both countries choose C, there will be a comprehensive nuclear war.

Figure 1 Bilateral game on preventing the use of nuclear weapons

Note: This is a broad format of a three-stage sequential game, in which State A moves first, State B second, and then State A moves again.

If both countries are rational, in a state of equilibrium, state A imposes limited nuclear use (LS), but state B does not counterattack (NC) and nuclear growth does not occur. Theoretically, nuclear war is avoided everywhere. In the real world, however, once nuclear weapons are used, even to a limited extent, uncertainty spreads and the possibility of nuclear war everywhere is no longer zero.

Figure 2 shows what would happen if all the countries of the world joined an agreement imposing restrictions on the prior use of nuclear weapons. In this case, the benefit of a pre-emptive nuclear attack by State A would be negative as restrictions would be imposed by the rest of the world for the limited use of nuclear weapons. In Figure 1, without restrictions, it would be impossible to stop the limited use of nuclear weapons by the A state. However, as shown in Figure 2, if it is already known that countries around the world will impose restrictions on the use of nuclear weapons. State, State A will not use nuclear weapons in the first place because there will be no benefits from limited nuclear use in the first place. Thus, the first step of molecular growth is prevented. Since there will be no nuclear use in equilibrium, restrictions are not required and as a result, the approving countries will not have to pay any price.

Figure 2 If several countries participated for approval on the pre-use of nuclear weapons

Note: (X, Y) is the benefit of (State A, State B).

Although the Nuclear Non-Proliferation Treaty was enacted in 2021, the destruction of nuclear weapons is currently unrealistic because it cannot be achieved without the consent of all states that have nuclear weapons. The agreement does not provide for sanctions against the use of nuclear weapons. If an agreement on restrictions on the prior use of nuclear weapons can be drawn up, restrictions can be imposed even if the signatory state is a non-nuclear state. Thus, this agreement will create a new international norm with high credibility. Establishing such an agreement is tantamount to creating an international standard of equality in which all citizens of the world share equally the power to determine the future of the planet.

Author’s note: This column has been reproduced with the permission of the Institute of Economics, Trade and Industry Research (RIETI). The original article appeared in the Nikkei Shimbun on June 15, 2022 and was translated by RIETI with some additional information.


Craig, MR (1999), “Molecular Prevention in the Developing World: A Game-Theoretical Treatment”, Journal of Peace Research 36 (2): 141–67.

Shelling, TC (1960), Strategy of struggle, Cambridge MA: Harvard University Press.

Zagare, FC (1992), “NATO, Rational Growth and Flexible Response”, Journal of Peace Research 29 (4): 435–54.