Salvation or Annihilation; is MAD mad?

Salvation or Annihilation; is MAD mad?

. 8 min read
Suppose the Russians have invaded West Germany, Belgium, Holland, France? Suppose their tanks and troops have reached the English Channel? Suppose they are poised for an invasion? Is that the last resort?
Why not?
Well, we'd only fight a nuclear war to defend ourselves. How could we defend ourselves by committing suicide!
So what is the last resort? Piccadilly? Watford Gap service station? The Reform Club?

So realizes the new British prime minister that nuclear weapons do not work in Yes, Prime Minister, the classic ‘80s British sitcom of dysfunctional government. For close to six decades, MAD (Mutually Assured Destruction) has been at the center of military strategy for the world’s great powers, but as Russia once again heats up the rhetoric around nuclear weapons, it is worth asking whether the logic of MAD actually holds. Because if it does not, the world is in a far more precarious position than the neat balance of MAD would suggest.

The days when the public was prepared for nuclear war with nuclear fallout shelters and nuclear drills has long passed. It can be tempting to assume that the threat of nuclear war (or, for that matter, conventional war between nuclear powers) has also passed. After all, no nuclear weapon has been dropped in wartime in three quarters of a century, and (outside of India and Pakistan), nuclear armed powers have not fought openly since 1953. Meanwhile, the number of deployed nuclear warheads has fallen dramatically since the 1980s. Proponents of MAD would argue that it has played a critical role in stabilizing this status quo.

However, that complacency belies a situation that warrants more concern—the risk of nuclear war remains as present as ever as Russia unleashes large-scale warfare in Europe.  Even if MAD prevents nuclear war, it does not necessarily prevent conventional war between nuclear armed states, a threat promising terrifying consequences for the world.

The Theory

MAD has become military doctrinal orthodoxy since the development of reliable means of nuclear retaliation in the 1960s. The theory holds that a nuclear armed state will not use nuclear weapons in a first strike against another nuclear state provided that the targeted state has the capability to make a retaliatory strike, rendering the first strike self-destructive.

This theory requires a significant number of caveats to be effective. First and most importantly, it is necessary that a credible response can be made. This tends to take the form of one of two main approaches. The United States and Russia both maintain a retaliatory strike capability utilizing strategic bombers to retaliate in the event of a nuclear attack at the direct instruction of their leader. Hence, the United States and Russia operate nuclear triads capable of providing not just the diplomatic nuance of strategic bombers, but also the second strike capacity of intercontinental ballistic missiles and submarine launched ballistic missiles. Under this style of response, the first strike is deterred because the opposing side can, through an early awareness of the strike from radar, order a response before the initial annihilation of its country. Questions remain about the viability of the nuclear triad method, given the advanced surveillance and tracking techniques employed by the nuclear armed powers.

As a result of the vulnerability of strategic bombers in modern war, since the 1960s the more common form of MAD is through a so-called second strike (employed by the United States, Russia, and most other nuclear nations), in which the country maintains an unattackable nuclear capability that can respond in the event of a first strike regardless of the leader’s decision. The classic formulation of this approach is the British government’s letters of last resort, whereby a sealed letter containing instructions of what to do in the event of nuclear war and a first strike is placed in each British submarine, and so the result of a nuclear war is guaranteed to be mutual annihilation since all major nuclear nations maintain undetectable nuclear submarines at sea which are less subject to human temperaments.

A second problem for MAD is the potential for one side to develop the capability to shoot down nuclear missiles, thus enabling it to withstand a nuclear strike unharmed and act as though it were the only nuclear power. The most infamous attempt at this defense was Reagan’s Star Wars program (Strategic Defense Initiative or SDI), though numerous attempts have been made from the 1960s NikeZeus program to Trump’s SDI II program, while the Soviet Union also tested a number of strategies including lasers and space cannons. However, the prohibitive cost of these programs, the difficulty of ensuring complete efficacy, the development of nuclear weapons like hyper missiles that can evade them, and the destabilizing effect of a successful program have so far largely deterred further development of such programs.

An addendum to this description of the theory is to note that a nuclear war is likely to be complete and total in its devastation under MAD. This outcome is not just probable because it is implied by MAD (Mutually Assured Destruction), but because a variety of war games have demonstrated that nuclear conflicts between nuclear powers always escalate to complete destruction. The most famous demonstration was Proud Prophet, a series of war games involving the US Cabinet and real US War Plans in 1983, where attempts at limited, de-escalatory, or tactical nuclear war consistently led to full-fledged salvo and the annihilation of at least half a billion people.

Overall, despite several caveats, MAD appears convincing and has been widely credited with ensuring that the Cold War remained cold.

Does MAD actually work?

MAD is called MAD for a reason. The term “MAD” was coined by the strategist Donald Brennan in 1962 precisely because a doctrine predicated on global annihilation was considered mad. Historically, most of the criticism against MAD has stressed the idea that it may be possible to disable a country’s ability to carry out a nuclear response, thus preventing MAD from working. However, there are more significant issues with the theory as it currently stands.

Essentially, the key problem is that nuclear weapons are disproportionately destructive. Even the hawkish President Reagan disliked their existence for this reason, describing MAD as a “suicide pact” that threatened all of humanity. In other words, the severity of actually carrying out a nuclear strike makes it difficult to believe that MAD could ever actually be employed as has been suggested, since it would require the merciless annihilation of millions of innocent civilians.

First is the case of using nuclear weapons in a first strike capacity. While MAD is, strictly speaking, a theory that prevents the employment of nuclear weapons as a first strike, it is often implicitly used to suggest that the threat of nuclear (first) strike could be used to prevent an event of the severity of a nuclear strike. If MAD is a legitimate strategy in the geopolitical toolkit, its usage is theoretically not limited to nuclear retaliation, but is also applicable to conventional attacks. Such a scenario plays out in Yes, Prime Minister. However, the problem is that the loss of Europe to the Russians is not as bad as a mutual apocalypse, so nuclear weapons could not actually be used.

This logic extends to straight-up occupation of another country. Existence under occupation is better than mutual annihilation, so even if—as the Yes, Prime Minister advisor suggests—Russian tanks drove up the M1 motorway, it would not be rational to initiate a first response. As Winston Churchill noted, only “lunatics or dictators in the mood of Hitler when he found himself in his final dugout” might be expected to launch a first strike under occupation. Carrying on a war under occupation places a heavy burden on a population in the pursuit of victory, but that is different from the promise of total destruction under MAD; hence, only the former is reasonable.

These two examples have an important consequence for the specific terms of MAD itself. Afterall, if you knew that a full strike had been launched against your own country, and there was nothing you could do to prevent your national obliteration, the morally correct response is not to retaliate and needlessly slaughter millions of the opponent’s citizens when victory is inevitably lost. As Reagan put it when proposing SDI, “is it not better to save lives than to avenge them?” Similarly, Edward Teller, a member of the Manhattan Project, wrote for the HIR as early as 1985 that “the MAD policy as a deterrent is totally ineffective if it becomes known that in case of attack, we would not retaliate against the aggressor.”

While MAD makes sense as a threat before nuclear weapons are launched, as soon as they are launched, the rationale switches against launching them. The problem that arises from this logic is that if you are not, as few can be, willing to exterminate millions for the sake of extermination, then MAD no longer functions as a deterrent. While steps like the British letters of last resort or the Russian dead-hand system have been taken to partially automate the process of retaliation removing human emotion, there is nothing stopping these methods from falling short (in the British case, you could write to do nothing, and in the Russian case, the system must be—as it rarely is—turned on).

This dramatically changes nuclear policy, since there cannot be a circumstance in which MAD acts as a rational deterrent to nuclear war, nor can MAD act as a deterrent against opportunistic exercises in conventional power, from the invasion of Crimea to incursions into the Taiwanese Air Defense Identification Zone. It is the mistaken belief in MAD, and not MAD itself, which maintains world peace in an uneasy equilibrium.


Russia’s invasion of Ukraine has provoked significant nuclear concerns after Putin stated on April 20, 2022 that he “will use all the means at our disposal,” viewed as a threat to use nuclear weapons. However, analyzing the situation in light of MAD’s potential shortcomings helps explain why the invasion was able to happen, along with potential responses.

MAD is seen as the ultimate deterrence against foreign misbehavior (Russian in this case). However, as history has shown, the scale of nuclear destruction means that it cannot be a credible threat. As Timo Koster, NATO’s Director of Defense Policy and Capability, put it, “a massacre is taking place in Europe and the strongest military alliance in the world is staying out of it. We are deterred and Russia is not.” This dynamic helps explain the Russian invasion—the fact that Ukraine was a European democracy under Western protection could not protect it because it can never be worth it to enter a nuclear war with Russia over Ukraine.

For the same reason, the West could never actually launch a nuclear war in response to a conventional attack on Western protectorates like the Baltics or Finland. Concerningly, this also applies to Russia using tactical nuclear weapons against Kyiv, or even NATO allies in the Baltics. NATO could protest, rally international support, and strengthen the sanctions regime, but it could never actually retaliate against Russia, for that would be self-destructive.

That said, NATO’s limited ability to respond goes both ways. At the conventional level, it suggests that the West has a greater ability to maneuver than it realizes. For example, despite Putin’s aggressive postering and threats, NATO has a far greater ability to help the Ukrainians than the threats of MAD would suggest. Russia’s conventional forces are humiliated and bruised, making a direct retaliation against NATO unlikely. This weakness effectively gives NATO a free hand to assist Ukraine far beyond what is currently being done. Even if NATO does not want to sacrifice soldiers, it could still provide Ukraine with as much military equipment and intelligence as Ukraine needs, as well as take direct actions against Russia such as supporting internal resistance operations and coups.

At a wider level, there might be far more scope for brinkmanship amongst the world’s great nations—free from the mistaken fear of an imminent apocalypse. Heated disputes may arise in trade and diplomacy spheres, but (at least initially) not accompanied by nuclear posturing. In the longer term, a MAD-free world suggests that tactical nuclear weapons are possible, since as long as they are limited, a full strike could never be contemplated. However, since this relies on knowing your opponent does not mistakenly believe in MAD this outcome likely remains distant since MAD game theory suggests that tactical uses of nuclear weapons always escalates to mutual destruction.

Embracing a MAD-free world is difficult. It will likely involve more conflict and more brinkmanship freer from the fear of accidentally triggering nuclear apocalypse. However, it is also a world where the West is free to stand up for itself, not deterred by a theory only one side really follows. And it is a world no longer reliant on a devoted but mistaken belief in MAD.

Cover photo courtesy of National Nuclear Security Administration / Nevada Field Office.