Minimum deterrence and morality

We have 450 of these, down from 1000…when we were, you know, safer.

A few weeks ago, the director of the Air Force Association, retired Lt. Gen. Michael Dunn, expressed his concern about deep cuts in nuclear weapons in a post at the AFA’s blog:

I have been seeing “trial balloons” in the press about the Administration’s desire to go to lower numbers of nuclear warheads … beyond those listed in New START. Some of the numbers are as low as 300 warheads…This should worry all of us for several reasons.

General Dunn listed several objections, most of which I and others have dealt with repeatedly. He warns, for example, that a smaller inventory of nuclear weapons “makes it more likely that a potential adversary could deliver a knock-out first strike.” There’s not one shred of evidence from the experience of the Cold War to suggest this conclusion, but it’s long been Holy Writ among nuclear deterrence theorists.

St. Augustine, author of “The City of God” and “The Ethics of Nuclear Operations”

Here, I want to consider an argument that Dunn makes right at the top, in what must have seemed like a deft rhetorical maneuver at the moment: he suggests that going to very low numbers of strategic nuclear weapons would force the United States to adopt strategies that contravene the laws of armed conflict and the Geneva conventions, by making cities and civilians the only possible targets of nuclear weapons — which would be illegal and immoral. (Like strategic bombing during World War II. But let’s not open that can of worms right now.)

In other words, it would be immoral not to keep large numbers of nuclear weapons.

While this seems paradoxical on its face, it’s not actually a ridiculous argument, and it’s especially tempting because it seems so noble in its goals. But the whole “fewer arms means more immoral targeting” position is based not only on paper-thin reasoning, but on so many false assumptions that it’s difficult to know where to start.

But first, a little historical background is in order.

During the Cold War, the United States coyly fudged the question of what we were targeting with nuclear weapons and why. The debate among American strategists back then was between “counterforce” targeting, in which we would aim to destroy Soviet military assets (including nuclear weapons aimed at the United States), and “countervalue” targeting, which essentially meant leveling the USSR’s cities, killing its citizens, and vaporizing its infrastructure forever.

Which one would convince the Soviets we were serious, and deter them from attacking us?

Classical deterrence theory mandated this straddle. If the object is to keep the enemy guessing, and to imply deadly seriousness about war without actually threatening war, then a strategy that finds a military “use” for nuclear weapons makes more sense than a cold-blooded promise to kill millions of people. Threatening mass murder is nuts; threatening to go to war in a methodical and logical way is credible — or so the theory went, anyway.

Who said political cartoonists aren’t subtle? Herblock from the Washington Post, 1962

And to be sure, counterforce targeting seems attractive on a purely instinctive level. After all, who would make such an insane, and therefore incredible, threat to kill millions of people?

Well, we Americans did, actually. But not because we were bloodthirsty loons; rather, it’s because we were so badly outnumbered by our enemy. The United States and NATO spent a good part of the Cold War completely outgunned and overmatched by a numerically superior Soviet Union. Worse, the Soviets enjoyed the immediate advantage of being located right on the borders of Western Europe.

The Kremlin’s war plan in Europe was to seize Germany and move west as quickly as possible, leaving the United States and NATO reeling, in shock, and too terrified to escalate the conflict to nuclear war.

No, seriously, he meant “Massive Retalition” in the positive, life-affirming way

Our initial strategy in the 1950s was less a “strategy” than a desperate threat. As enunciated by the Eisenhower administration, “Massive Retaliation,” as it came to be called in the media, was predicated on America’s nuclear superiority: since we couldn’t match the Soviets (and their then-allies, the Chinese) pound for pound everywhere on the planet, we made a vague threat to use our nuclear weapons against the USSR, no matter what the circumstances. As Secretary of State John Foster Dulles put it in 1954:

If the enemy could pick his time and place and his method of warfare–and if our policy was to remain the traditional one of meeting aggression by direct and local opposition–then we had better be ready to fight in the Arctic and the tropics, in Asia, in the Near East and in Europe; by sea, by land and by air; by old weapons and new weapons.

Therefore, because the U.S. could not commit “to military expenditures so vast that they lead to practical bankruptcy,” we would instead use our nuclear advantage to “retaliate instantly and at times and at places of our own choosing.”

Massive Retaliation was a bad idea, not least because no one — certainly not us — knew what was supposed to trigger nuclear retaliation. President Eisenhower was hardly a nuclear hawk, and (like Ronald Reagan years later) viscerally rejected the notion of nuclear war.  And once the Soviet Union gained the ability to retaliate in kind, the whole idea of just striking Moscow at will as punishment for ill-defined bad behavior evaporated pretty quickly.

Fast forward to the 1960s. The United States and the Soviet Union are now sitting on thousands of deliverable nuclear weapons. No matter how the numbers were run, a nuclear exchange between the superpowers meant pretty much the end of Life As We Know It, at least in the Northern Hemisphere.

Some months ago, I blogged about the objections of USMC General David Shoup to JFK’s nuclear strategy: he said point-blank that saturating the USSR and China with nuclear strikes — and that meant hitting the Chinese even if they were not initially involved in the war — and inflicting hundreds of millions of casualties in a matter of hours was “not the American way.”

Perhaps worse than being un-American, such strategies weren’t likely to deter the Soviets (or so we worried, anyway), because no one would believe such grandiose threats. A far more plausible path to war would be to tell the Soviet leaders that we didn’t want a nuclear conflict, but that if one were forced upon us, we had a whole list of targets we would march through before we finally got around to burning the entire Eurasian landmass to ash.

Although the Americans called this policy “Mutual Assured Destruction,” or “MAD,” the idea behind it was that this eventual level of destruction would not happen by a single presidential decision, but by a series of decisions, each one militarily sensible in itself, that in their collective effect would drive both sides to an all-out exchange. If the Soviets could see the logic in that position — we hoped — they would be deterred.

But even then, Congressional opponents of MAD felt that this was simply making a mutual suicide pact with Moscow that only one of us might observe if the shit hit the fan a crisis got out of control.

Let me just check my notes, Senator. Yep, 120 million. Give or take.

Then-Secretary of Defense Robert McNamara tried to reassure jumpy legislators by promising to wipe out a lot of Soviet citizens if Moscow got any ideas. “I think,” he said in Congressional testimony in 1967, “we could all agree that if [the Soviets] struck first we are going to target our weapons against their society and destroy 120 million of them.”

This, of course, is a fundamentally immoral threat: it means intentionally targeting innocent non-combatants in urban areas, including children, as a deterrent.

Now, let’s be clear: McNamara wasn’t suggesting this as some kind of good idea. He was saying that in the event of a surprise attack, when all was lost, we could still take the Soviets to Hell with us, and that they had to know that. Thus, they would be deterred.

Essential to all this is that cities would always be the last targets on the list, not the first, and that we would do everything in our power to avoid the senseless killing of civilians, especially if the bad guys would do the same.

Which brings us back to General Dunn’s objections to deep nuclear reductions:

Is there a strategy change that supports further cuts? Those that I speak with say … well … we could just target cities … and 300 is way more than we need. I frequently remind those who support this that the Law of Armed Conflict, Geneva conventions, and other international agreements … and our own moral principles … prohibit the intentional targeting of civilian non-combatants. And … yes, there will be innocents that die in a nuclear attack, but the fact remains that we should never plan to target and kill non-combatants.

In effect, Dunn is saying that we should stick with MAD, and keep enough nuclear arms so that we can target a whole range of military objectives in Russia (and, one supposes, China). Otherwise, with only a handful of nukes, we can only really threaten urban areas, and that would be immoral.

This looks like such a decent and humane argument on its face that it seems almost barbaric to object to it — except that it makes no sense, either on a strategic or moral level.

First, General Dunn is clearly proceeding from the Cold War-era canard that there is some sort of discernable difference between counterforce and countervalue targeting. In theory, of course, there is a difference — but only if a country locates its military assets in the middle of Antarctica.

I’m not even going to crack wise about this map. Some things just aren’t funny.

In reality, a lot of military assets (especially command and control) tend to sit pretty close to where people actually live, and nuclear strategy has always been plagued by this problem of “co-location.”

This makes separating out the destruction of military targets like the U.S. Strategic Command in Omaha from the destruction of your Aunt Sophie and Uncle Herb and their entire Nebraska neighborhood functionally impossible.

It gets even worse with national command centers. Hitting the Pentagon, it could be argued, might be a purely military strike. Unfortunately, it’s located in a huge urban area that also happens to be the national cultural and political capital of the United States. Likewise, the Russian Defense Ministry is right downtown in Moscow, just up the street from the Lenin Library and smack-dab along what would be their version of Fifth Avenue.

If even a very small nuclear war breaks out — and “small” is a relative term in the world of nukes — we’re looking at a lot of damage against urban areas throughout the United States (unless the enemy is completely stupid and doesn’t want to hit the Pentagon, Strategic Command, or things like, say, the Pacific Fleet).

So let’s cut the malarkey: we and the Russians are already targeting each other’s main cities and we always have. In fact, when Dick Cheney — and I dare anyone to call him a wimp — became Secretary of Defense under the first President Bush in 1989, he was appalled after being shown the nuclear war plan that included targeting 500 nuclear warheads against Moscow. (We also were going to rain 69 warheads — sixty-nine, not a typo — on one radar installation outside the city.)

You guys are scaring me. Do you have any idea how hard that is to do?

I still believe that the former Vice President is a decent man and a patriot, despite what his opponents say. But when you’ve managed to freak out Dick Cheney with a war plan that is too wantonly violent, you’ve probably crossed a line somewhere.

Even a plan that doesn’t try to hit Moscow, Washington, or Beijing is still going to kill a lot of people with purely “military” strikes. There’s no way around it: a “limited” strike against American or Russian nuclear sites would generate tens of millions of casualties. Even a limited strike on China’s tiny nuclear force would kill upwards of two million people, not counting fallout, food supply problems, and the myriad other problems that will come in the wake of a nuclear war. Just look at the mess the Japanese are still dealing with after the Fukushima reactor fire – and then multiply that by a factor of thousands.

General Dunn’s argument comes down to this: Keep lots of nuclear weapons on hand, so that you can observe a more “moral” and less illegal strategy of pretending not to target civilians, because otherwise deterrence will rest on the potential murder of children and other non-combatants. If you target 40 million civilians, it’s immoral; if you kill 400 million as incidental deaths during an all-out war…well, you did your best.

Of all the last-ditch arguments against deep cuts in nuclear weapons, this is probably the worst and most disingenuous of them all. It’s also one that’s been overtaken by events: the current New START Treaty was negotiated between Moscow and Washington on the premise that it leaves both Russia and America capable of destroying between 150 and 300 urban targets in each country, and that such a level of carnage is enough to keep both sides from doing anything…rash.

Do we look like we’re joking?

Personally, I think the number of nuclear weapons that deters each side is probably lower. (Maybe closer to, say, one, but that’s an argument I’ll present more fully this winter in my new book.) Certainly, the Chinese seem to think so, as Jeffrey Lewis has argued in his book on China’s deterrent strategy, and our British and French allies openly refer to their own nuclear arsenals as minimum deterrents meant to inflict pain, rather than military losses, on anyone who attacks the UK or France.

General Dunn asks if there’s a strategy that supports going to far lower numbers of nuclear weapons. As I’ve argued before (and will again when my book is released this winter), the simplest strategy is to reject the use of nuclear weapons in all circumstances except a threat to the actual existence of the United States.

Looks like we’ll be putting in some overtime today…

In that case, it’s hard to imagine that a threat to destroy 100 Russian or Chinese cities won’t be enough to deter even the most implacable enemy — and if it’s not, we’re in a lot worse trouble than we could ever have imagined even during the Cold War.

To argue for keeping high levels of nuclear weapons in the name of a humane and more just targeting strategy, then, is nonsense.

An exchange of 3000 weapons, no matter where they were targeted, would have obliterated both sides; an exchange of 300 weapons, or even 30, will destroy Russia, America, or China as they currently exist.

Real political leaders know this, which is why they’re loath even to think about nuclear war. As former White House advisor McGeorge Bundy wrote in a 1969 quoted hundreds of times since:

Think-tank analysts can set ‘acceptable’ levels of damage well up in the tens of millions. They can assume that the loss of dozens of great cities is somehow a real choice for sane men. They are in an unreal world. In the real world of real political leaders….even one hydrogen bomb on one city of one’s own country would be recognized in advance as a catastrophic blunder; ten bombs on ten cities would be a disaster beyond history; and a hundred bombs on a hundred cities are unthinkable.

He was right then, and he’s still right today.

We won. Apparently.

If General Dunn and others think that high levels of nuclear weapons are essential to American security, that’s their right. After being on the same side of that debate in my youth, I now believe they’re wrong.

But let’s at least keep the debate honest. The use of nuclear weapons, in any capacity, will be a moral horror, and no one in those circumstances is going to much care about the legal niceties once the missiles start flying. Arguing against nuclear reductions by appealing to some sort of legal code that makes their use, or the threat of their use, somehow more acceptable in a “military” capacity is a moral and strategic dead-end that does a disservice to the debate.

Print Friendly


  1. At some level, all nuclear deterrent threats are incredible in character, even the one you mentioned in class about a “demonstration strike” of one: I burn one of your cities to the ground, and you then either surrender, take just one of mine and we call it even Steven and maybe even something else happened?, or, we launch something more, maybe all of them: massive retaliation had that feature for a long time before they did the SIOPs. 3,400 warheads at 7200 megatons from the Elbe to the Yalu and Red River, and 500 million dead, easy.
    Of course John Steinbruner might be right that to use one equals to use all, since to launch one at even “only” C4RSI would kill a lot of people, and more importantly, you would lose Command, Control, and Communications, and then like you said in class as to what happens in the White House in a preemptive strike on U.S. C4RSI,”Have a nice day everybody and good luck! I’m out of here and let everything fly and God Bless the United States of America.” Or something like that. You said it better in class I am sure.
    On the other hand, if that became understood to be the case, that no one would really do that, actually use a nuclear weapon for sure, then Great Power Conventional Wars would be permitted again, along with cyberattacks too, would they not, instead of the wars by proxy we have now. [Edited for length by TMN]

  2. I am not following this;

    - there are more enemies now than there were before (i.e., they are geographically separated so that one delivery will not be enough to cover multiple dispersed targets – such that you may actually have to use some of the 50 outside of Moscow);

    - the planet is more populous than before by quite a lot;

    - the interests of our enemies and our “elites” are becoming more closely aligned (you know those pow wows at Davos, the NAFTA love fests, etc.);

    - our “elites” reside primarily in large Democratic controlled cities;

    so what’s the problem with having lots of options?

    Please note that an earlier exchange (which may have been one sided if early enough) would have meant acceptable losses and no troubles for American dominance today; instead we chose to die a slow death;

    I am betting that Cap (unlike, apparently, Cheney) was not frightened one bit;

  3. If we count neighboring Canada and Mexico (the U.S.’s first and third largest trading partners) as”enemies,” per GrigoriL’s comment about “NAFTA love fests,” then things are really out of control. I do not think Stephen Harper intended Canada’s War of 1812 commemorations to provoke such fear south of the border.

  4. I guess I’m not following the “multiple dispersed targets” comment. Is the idea that we have to keep a few hundred strategic weapons so we can hit every place in the world where the regime doesn’t like us? Sounds like Randy Newman’s old song, “Political Science,” where “no one likes us anyhow, so let’s surprise ‘em and drop the Big One now.”

  5. PS: Alex, back some years ago, Tony Blair apologized to Congress for the destroying the Library of Congress. So I think we’re ok with them and Canada on the 1812 thing ;)

  6. As to deterrence being stable, if during or right before a war, one Trident returns to homeport, and uploads to 8, 192 total, maybe 12 in a pinch, but say 8, and then, with the usual penetration aids, it can take out the Trans-Siberian corridor, plus Kazan’s ties to Russia, and Tsaritsyn, and Russia is back to pre-Ivan IV land. I say that’s 30 targets, like Novosibirsk, Chita, Baikalsk, places like that, and mainly airblasts, and that with 3 per target, it would still leave ninety for China airblasted over cities for optimum kills of population, flattening Beijing and the North under some theories of aftermath, and that’s the real world main concern, deterring Russia and China. Unless France is going to go loopy or Britain, and then … I mean that’s crazy talk there, 1812?, and so it’s hard to see how you’d really want to do that as to Russia and China as to drawing that firepower in the first place, and so, nuclear deterrence should hold.
    Although, what it does show is a difficulty of current delivery systems/platforms as to going to 300. If a Trident is supposed to be 92 under the 24×3 allocation in New Start, to be able to upload, or even say be in a port and have a “fire” on a submarine, almost like a hint, before downloading those warheads, maybe 300 isn’t very realistic, at least with any subs. Bombers, that’s only 20 if there’s an attempt to mislead. A sub is easily over 100 waheads, which might be tempting to some people, on a C4RSI strike, if you could get the subs of course, if pretty out there too, given what one 24×8 Trident strike would look like if targeted for maximum political-demographic effect, even around the world. I mean if you just wanted to “kill them all and let God Sort it Out/Start Over Again,” two Tridents at maximum warhead loads are pretty close to being able to do that globally, save for the survivalist types.

  7. Don- Not likely that a Trident is going to come home and suddenly change its total load of warheads, especially during a crisis. This is what I mean about getting carried away with wargaming scenarios: the kind of widespread damage you’re talking about is so apocalyptic that it has no foundation in anything like a strategy.

    If you really want evidence that none of this matters, look at the New START counting rules, which count a bomber as “one” weapon. Why? Because it’s easy. No one really cares how many warheads are on a bomber — not us, not the Russians.

  8. If deterrence was failing, like in a crisis, then it seems to me that one way it would be signalled would be by abrogating arms treaties, in which a return of Tridents to port would be one way to do that, if a dangerous one, as is putting bombers in the air, or opening silo doors.
    Hopefully, there will never be a real deterrence failure, or rather near failure.
    The bombers it was my understanding have always counted oddly, as they have dual roles, or at least that was the case in SALT I is that correct?

  9. We had disagreements about how to count some of their bombers due to range. (Are they theater or strategic?) Abrogating a treaty in the middle of a crisis is pretty much a prelude to war, but again, hard to swap out all that hardware so quickly. It would be highly provocative and dangerous to even suggest it during a crisis.