Michael KreponWin, Lose or Draw?

Several chroniclers of The Bomb have concluded that the term “nuclear strategy” is an oxymoron. The dilemmas of crafting a nuclear strategy and planning to execute are best dealt with on paper. But even on paper, dilemmas are hard to finesse or work around — including the most basic question of what the user hopes to accomplish.

Secretary of Defense Caspar Weinberger got into a lot of hot water when passages of the Reagan administration’s Fiscal Year 1984-1988 Defense Guidance” were leaked to the print media. The most eye-catching phraseology was the following:

The primary role of United States strategic nuclear forces is deterrence of nuclear attack on the United States, its forces and its allies. Should such an attack nevertheless occur, United States nuclear capabilities must prevail even under the condition of a prolonged war.

This language made tsunami-like waves because it appeared in the context of a significant U.S. strategic modernization program and was presaged by other memorable quotes by Reagan administration officials, none better than T.K. Jones’ unforgettable comment on the utility of civil defenses: “If there are enough shovels to go around, everybody’s going to make it.” Equally mortifying was the opinion of Eugene Rostow, Reagan’s ACDA Director: “We are living in a pre-war and not a post-war world.” [Wonks interested in more of the same might read Robert Scheer’s With Enough Shovels. Scheer, the Oriana Fallaci of nuclear interviewing back then, had an amazing facility to elicit memorable quotes.]

Needless to say, Weinberger got beat up on Capitol Hill for this language. His initial riposte, typically, was not to give an inch. Two samples: When asked whether he had a plan to prevail should the Soviet Union start a nuclear war, Weinberger replied, “Yes, of course. Consider the alternative.” Another memorable retort was, “You show me a Secretary of Defense who’s planning not to prevail and I’ll show you a Secretary of Defense who ought to be impeached.”

Many awkward questions arose from the objective of prevailing in a nuclear war with the Soviet Union, which was a reprise of the earliest U.S. war plans. Which side would win if both adversaries sought to prevail? And who would call a time out to figure out the score? The resulting domestic political and diplomatic damage around the word “prevail” was so considerable that even Weinberger issued a public retraction (of sorts) in a letter to Scheer’s paper, the Los Angeles Times, published August 25, 1982:

I am increasingly concerned with news accounts that portray this Administration as planning to wage protracted nuclear war or seeking to acquire a nuclear ‘war-fighting’ capability. This is completely inaccurate… It is the first and foremost goal of this Administration to take every step to ensure that nuclear weapons are never used again, for we do not believe there could be any ‘winners’ in a nuclear war. Our entire strategy aims to deter war of all kinds, but most particularly to deter nuclear war. To accomplish this objective, our forces must be able to respond in a measured and prudent manner to the threat posed by the Soviet Union… But it does not mean that we endorse the concept of protracted nuclear war, or nuclear ‘war fighting.’ It is the Soviet Union that appears to be building forces for a ‘protracted’ conflict.

Readers would be forgiven for scratching their heads trying to figure out how to compete effectively in a measured and prudent way against a foe planning to win a protracted nuclear conflict.

If winning is a problematic goal, should the United States plan for a tie? An accidental nuclear war can theoretically be played to a draw, if command and control is good enough and if national leaders can withstand popular sentiment for tie-breaking vengeance. Several Hollywood screenwriters paid off mortgages considering these odds. (Stanley Kubrick did not believe in the likelihood of tied scores.) And what about the deliberate use of one or more nuclear detonations? If the stakes are so high between two nuclear-armed states as to result in a crossing of this threshold, do the stakes somehow become lower after the mushroom clouds appear — and does command and control remain robust enough — so that a “draw” is feasible? And how do we measure a draw?

A little while ago, Jeffrey pointed us to the Clinton administration’s de-classified PDD-30, a presidential decision regarding U.S. nuclear posture, signed on September 21, 1994. Here’s what President Clinton decided on the role of nuclear weapons in U.S. security strategy:

the United States will retain strategic nuclear forces sufficient to deter any future hostile foreign leadership with access to strategic nuclear forces from acting against our vital interests and to convince it that seeking a nuclear advantage would be futile. Therefore, we will continue to maintain nuclear forces of sufficient size and capability to hold at risk a broad range of assets valued by such political and military leaders.

This language is far less incendiary, but it still begs fundamental questions: What is sufficiency? And sufficient for what purpose? The requirements of preventing disadvantage are presumably less onerous than prevailing, but with the demise of the Soviet Union, the requirements of prevailing are less onerous, too.

Many supremely confident and smart people have tried to answer these questions. Notwithstanding their efforts, “nuclear strategy” remains an oxymoron.


  1. archjr (History)


    Great post. Just to stir things up, I would define sufficiency as the minimum amount of land-based nuclear weapons, protected by missile defenses, to pound the hell out of the bad guy if he attacked first. Or, a survivable second strike capability. Verified by treaty, of course.

  2. A Complete Stranger (History)

    I am reminded of the recent wave of Republican responses to the Obama Administration’s Nuclear Posture Review. Far too many of them complained that the document didn’t say that “the primary goal of US nuclear weapons was to defend the US.” They didn’t seem to realize that the goal of our nuclear strategy, as defined by the NPR, is to prevent the US or large portions of its population from being destroyed by nuclear weapons and that reducing their “saliency” is a core part of that strategy. Only then can we build the international concensus needed to fight horizontal proliferation. After all, even the Republicans believe that an attack from an emerging nuclear power is much more likely than an attack from Russia. And I, for one, would rather prevent new nuclear powers from emerging rather than deter them after they have acquired nukes.

  3. Carey Sublette (History)

    The consequences of nuclear explosions on U.S. soil are so dire the only rational strategy is one of deterrence – trying to prevent it from ever happening.

    But if deterrence were to fail, then the logical objective should be to minimize the consequences of failure. That is, bring the conflict to an end as quickly as possible with as little damage to the U.S. as possible so that we (and the Soviet Union/Russia/China) can get to work rebuilding our shattered societies.

    Any other objective should be considered treason.

    If we lose New York, I would rather not “prevail” if we must also lose Chicago to do so.

    • Mark Gubrud (History)

      Sorry for playing Devil’s Advocate, but your comment seems to suggest that, should deterrence fail, the next move should perhaps be surrender. Did you have a different suggestion in mind?

    • George William Herbert (History)

      I can think of worse fates than cities obliterated; Hitler or Japan actually winning WW 2 would have been.

      We seem to have worked active genocidalism out of international relations (the vague threats by Iran don’t count; if they were clearly serious “clear and present danger”, Israel would have nuked them already).

      But it could always come back. We certainly have not worked irrational behavior or lack of human rights respect out of the international systems. Those who feel free to mistreat their own populace or minorities could escalate mentally at some point. Very peculiar logics apply to leaders inside of crisis situations as well.

      Al Qaeda winning? What would you give up to prevent that? They literally want to end western civilization. At some point, there’s no compromise, no matter how painful it is, even if the painful is cities full.

      These are not pleasant subjects to think about, and dwelling on them leads to paranoid responses in areas not nearly as dire and much more likely. But one cannot categorically reject thinking about them. Genocides happen; small ones still are attempted today. Barbarians – in the “tear down all of civilization” sense – still exist today. These are existential threats to western civilization. Those, we have to consider the unthinkable about.

    • Carey Sublette (History)

      I have long been puzzled that many people who by all measures should be considered well informed about nuclear weapons nonetheless do not seem to have internalized just how different the unconstrained hostile use of nuclear weapons would be from any previous conflict in history.

      Unless the weapons are being deliberately fired on targets specifically chosen to cause little damage, a Kahnesque “signaling exchange” rather than an actual war where the objective is to destroy an important part of the enemies war-making capability, the damage level suffered by all parties in the exchange jumps immediately beyond even the levels of destruction and death seen in WWII.

      In this context Mark’s comment about “surrender” strikes me as flippant, even foolish (though it may only be the fictional devil and not the actual Mark who is being flip and foolish). Particularly for the United States with no land borders across which an enemy could invade, what do you suppose “surrender” could mean in any practical sense? Please be explicit.

      In a very real sense even if deterrence fails, it continues to operate – unless command and control have collapsed. If Russia has lost St. Petersburg, would they really be indifferent to losing Moscow also? Or would the prospect offer powerful incentives for agreeing to immediate war termination? (This is a particular case of the well established concept of “intra-war deterrence”.)

      If command and control has collapsed, trying to engineer immediate war termination is even more important, if also much harder. And avoiding C2 collapse, which jeopardizes the ability to terminate the war, is an extremely important reason to terminate early.

      The thing about SLBMs and ICBMs is that you can always decide to launch more later, if you really haven’t had enough of the global holocaust yet. But you can’t unkill millions of people.

    • Mark Gubrud (History)

      Given the levels of overkill, vulnerability of a substantial fraction of nuclear forces and existence of accurate weapons to threaten them, and promulgation of first-strike and counterforce doctrines, that obtained during the Cold War and still do; given the possible deliberate or accidental disabling of attack warning systems; and given the apocalyptic frame of mind that decisionmakers would likely already be in at the 11th hour of a crisis over some stupid issue (and what issue would not be stupid in this context), it is hard for me to have any confidence in intra-nuclear-war deterrence or rationality on the part of national decisionmakers once they came to the realization that — HOLY F!@%ING $#!+ — the nukes were about to actually go flying, let alone that some already had.

      Maybe once we’ve made real progress toward abolition, starting with US and Russian cuts to around 300 warheads on survivable platforms only, there might be some hope of cooler heads prevailing in the event of a showdown or even a hot war that leads to the use of one or two Bombs. Maybe.

      My comment, Carey, might seem flippant but I meant to point to the absence in your comment of any explanation of how to “bring the conflict to an end as quickly as possible” after an actual nuclear attack, given that all efforts to prevent it from getting to that level had already failed.

      Telling the other side that we’d just give in on whatever the question might have been, assuming we could still get them on the phone, might work. Sticking to our demands and trying to negotiate acceptable terms of a truce under conditions of a global apocalyptic panic seems more uncertain of success, and certainly represents another objective than immediate war termination.

      Especially given that the most likely condition for first use of a nuclear weapon by a major opponent is that our armed forces are already engaged with theirs, and given that the first use is most likely to be against military forces, which would put immense pressure on the president to authorize immediate nuclear retaliation against their military forces, which would lead to another round… All this occurring under intense time pressure.

      I just do not think we can expect nuclear deterrence to continue to hold against incentives for and vulnerability to preemption once nuclear deterrence has already failed.

    • George William Herbert (History)

      Mark, it’s entirely possible that the second nuclear war will be to the bitter end of the first combatant to run out of atomic weapons (and great loss to the one remaining). But if one assumes a priori that any mutually nuclear war must and absolutely will run amok to final armageddon without any control or hope of standing back, you essentially put yourself into a position where anything nuclear destroys the world, because nobody is prepared to do anything other than that.

      I would rather not be there. It’s stupid not to acknowledge that that’s a real risk, and could well be what happens, and adopt ones plans (including disarmament, to whatever degree that factors into it). But one must prepare for the alternative for there to be any chance for it. That’s what warplanning is. Have options already thought out and on the table.

  4. Andrew Tubbiolo (History)

    In the end, I think it depends on how you start out. What kind of militarist are you? Are you a Tubbiolo ™ Type 1 Militarist, or a Type II? Type I militarists think that there is salvation for the individual and society via combat and use of force of arms. Tubbiolo ™ Type II militarists look at Type I militarists and are scared silly such that they either become arms controllers or arms builders with the hope that they can repel or contain the Type I’s via force of arms. Almost all societies’ military forces are manned by type I’s while governments are made up of a mix of type I’s and II’s.

    The question of strategy I think is what type of militarist is in charge of your nations policy and civilian/military offices? This flows. More so in the West than anywhere else and is complicated by the fact that most Western Type I’s claim to be Type IIs. But either way, each new government inherits what its predecessor leaves for it, and are limited to what they can do by term limits or limits in law. It also depends on what you are going up against. Is your partner in deterrence a Type I or a Type II? It all matters….

  5. archjr (History)


    I completely agree with you. My theological argument is very simple: an ability to respond is an ability to deter. God forbid the theory should ever have to be proved. By the way, this theory also covers other actors besides Russia, and is inherent in China’s nuclear weapons policy, which we should adopt. I was a big skeptic of no-first-use in the early 80’s, but am not so sure that now it couldn’t be an important first step towards global zero.

    I have always thought that PD 59 was perhaps the most sensible approach, even though it was issued during 1980 at the end of the Carter Administration. Brown subsequently testified before Congress (in the Committee I worked on, in the context of the freeze) “If we build [nuclear weapons], they build; if we don’t build, they build.”

    This is not an argument for building more nuclear weapons: it is only a recognition of the fact that the only utility of nuclear weapons is deterrence, as you suggest. It is also an acknowledgment that the potential utility of nuclear weapons is largely theoretical, and linked to questions of conventional weapons balances and probably stupid, if necessary, planning exercises for those countries bold enough to develop nuclear weapons. I am sure Iran’s leaders have thought carefully about this, which explains their actions for the last several decades. So I maintain what the US needs is the ability to bounce the rubble with a second strike, until and unless we can get everyone to give them up. There are lots of interim steps we can take until we all figure out how to do that, and they can reinforce the stability of terror/deterrence, and you are doing a great job of seeking that outcome.

  6. P.E.T. (History)

    I can’t agree. This is just more political partisanship that solves no issues or changes any minds.

  7. The Donkey (History)

    “The consequences of nuclear explosions on U.S. soil are so dire the only rational strategy is one of deterrence – trying to prevent it from ever happening.

    . . . .

    Any other objective should be considered treason.”

    That states the case too strongly

    E.g., extended deterrence is not “treason.”

    Neither is a preemptory counterforce strategy against a rogue state without “second strike” capability — or a non-state actor — who credibly threatens the imminent use of WMD against the US or its allies, or implements a plan for the imminent use of such weapons.

    Not saying such strategies are without their problems, but they are valid options.

    • Carey Sublette (History)

      You cut out the paragraph to which my comment about “treason” referred (“But if deterrence were to fail…”), and introduced an unmentioned variant of deterrence (to which I did not object) rather than the notion of “warfighting” in a general war which I did.

      The notion of preemptive attacks on foes with limited means, which does not entail the risks of “warfighting” with Russia or China, is also outside the scope of my remarks entirely.

  8. Mark Gubrud (History)

    I would not say that “nuclear strategy” is an oxymoron.

    Without too much eloquent exposition, I would just like to suggest that a good nuclear strategy for the United States to pursue now would be unilateral deep fractional cuts while challenging Russia to do the same, then move under a new, strongly verified, US-Russia treaty to an interim “minimum” deterrent posture consisting of maybe 300 survivably-based warheads to be used only for deterrence of nuclear attack, while also addressing Russian concerns about non-nuclear strategic threats by adopting strong qualitative and quantitative arms control of conventional weapons.

    Then engage the entire world in serious pursuit of a further round of reductions, to say less than 100 nukes per posessing state, then, as confidence grows, to say 30, then to 10, then to global zero, not in some distant unforeseeable future, but within a decade of having achieved the first step (US-Russia minimum deterrence), and possibly sooner.

    All this would of course require strong new international institutions to implement the arms control measures as well as to resolve other pressing global issues, and this would need to include a reinvigoration of the UN and renewal of commitment to support and abide by its Charter; and of course the outbreak or continuation of raging conflicts over resources, land claims, migration, climate disruption, or the failure of the US and other great powers to refrain from wars of aggression in the future, would be likely to delay or derail progress towards these goals. I’m not saying any of this will happen, just that it would be a good nuclear strategy.

    As for what happens if deterrence fails along the way, it seems to me that situation will be much easier to navigate and survive after deep reductions than in the present confrontation which is still wired up for hairtrigger first-strike and counterforce warfare, let alone the one that may soon develop if we allow the space arms race to take off, the high-tech/robot arms race to run amok, and the nuclear arms race to renew, this time with who knows how many players in the game.

  9. MarkoB (History)

    I notice the comment in the above post suggesting that an accidental nuclear war could be a draw in theory if command and control is good enough. I think it was WSEG No 50 which pointed out that in the missile age C2 would be lost after only a few detonations, a point usually erroneously first attributed to Desmond Ball. I think that the big issue with “prevailing” in nuclear war is the concept of “intra-war deterrence,” which I tend to think was revived by Bush (at least in the context of regional contingencies). There was an interesting comment made in the 2010 NPR about survivable command and control; if that had been made in Beijing or Moscow it would be seen, by say the Hudson Institute etc, as evidence indicating the existence of an intra-war deterrence strategy. Just a final point on nuclear strategy. I think there could be a chicken and egg issue at play. My hunch is that what we call nuclear strategy was just a means to justify the large size of the weapons complex.The realists are right about “the requirements of deterrence.” Strategic overkill tells us that the size of the weapons complex did not have much to do with security. Because of this the absurd force levels had to be justified, so that’s why you had “nuclear strategy.” That what we call nuclear strategy largely came out of the RAND Corporation tends to reinforce this view. Even the 2010 NPR has a line in it basically stating that current force levels are over-determined in so far as strict deterrence is concerned. In so far as the citation of PDD30 goes (1994), we should remember that Presidential Guidance (i.e. the NWEP) was still Reagan’s NSDD 13. It was PDD60, for strategic arms control reasons, that finally replaced NSDD 13 in 1997. So the language in PDD 30 does not appear as out there as Cap Weinberger’s statement, yet we might need to conclude that it is NSDD 13 consistent because NSDD 13 was still in force. If the lingo in PDD 60 sounds similar to PDD 30, then, really, how big a change was it? And if Obama hasn’t touched PDD 60…?

  10. John Schilling (History)

    It isn’t an either-or decision; deterrence pretty much requires war planning. Deterrence is based on a potential adversary being unwilling to accept the expected consequences of starting a war. Those consequences are a function of what the adversary believes you plan to do in the event deterrence fails. And it would be dangerous to presume that his intelligence services won’t eventually figure out your actual plans, or a reasonably close approximation thereof.

    With a sufficiently large deterrece arsenal, you might be able to get away with, “we’ll make it up as we go along” as a sufficient plan, even against an adversary with his own large arsenal coupled to a more sophisticated plan. But even then you’ll want to do some serious analysis to figure out what constitutes a sufficiently large deterrent, and that’s close enough to war planning that you might as well finish the job.

    If the plan is to draw down to minimal nuclear deterrents, I’m pretty sure “let’s think carefully about how we’d want to actually use these hundred missiles if it came to that” will get better results than, “we’ll make it up as we go along”. Allow deeper reductions while still preserving deterrence, and allow any failures in deterrence to be managed with less than apocalyptic results.

    If the plan is to bluff, to never actually use the weapons because the consequences are too horrific, then you’re betting absolutely everything on the true plan never being brought to light. George Herbert has already hinted at what “absolutely everything” means in this context, and yes, it’s way worse than just losing New York and Chicago.

    Having the weapons, and a good war plan, and the will to execute it, is far safer than just having the weapons. If making war plans is percieved as politically unacceptable, that’s a serious problem – and one I don’t have a good solution for.

    • Carey Sublette (History)

      A strategy of using intra-war deterrence to bring a nuclear exchange to a very early end, rather than imagining “prevailing” in an extended nuclear war, is a real and valid war plan, not an absence of a war plan. Nor is it a secret plan to “do nothing”.

      Intra-war deterrence cannot work if the initial deterrence threat has been shown to be empty, so the objective of keeping the damage to the U.S. to the lowest possible level definitely requires actual retaliation if deterrence fails.

  11. FSB (History)

    Odd. Lots of comments speaking of Russia as if the Cold War never ended.

    Among my favorite comments on deterrence, now — by Ivan Oelrich of FAS:

    “The basic nature of deterrence is that you might try to seize something of value from me, and I must be able to plausibly threaten to impose costs on you that are great enough to make the prize not worth the fight.

    If I have a million dollars on my desk and I threaten to rap you on the knuckles with a ruler if you take it, you might not be deterred; if I have an apple on my desk, the same threat might be effective. . . . If the prize one side is trying to seize is the future of the world, that is, the prize is everything, then one must threaten near total pain to make seizing that prize not worthwhile.

    The most basic difference between the Cold War and the world of today is not the lower levels of tension between the United States and Russia (or the Soviet Union) but the much lower stakes involved. When we talk about U.S. nuclear deterrent forces, we have to address what prize might some nation try to seize, even in theory, that is going to take a retaliation of more than 5,000 warheads to make it seem like a bad deal.”

    For that reason alone we ought to be able to negotiate a much lower level of nuclear weapons, and of much lower potency.

    That deterrence has not failed (deliberately or by accident) does not mean that it will always be so.

    archjr’s 1st comment re. missile defense: it does not work, ok? And thus lessens US and global security.

    • Mark Gubrud (History)

      The Cold War may have ended, but the nuclear confrontation was only partly dismantled, and some of its most dangerous elements remain in place.

      With all due respect to Dr. Oelrich, the basic nature of nuclear deterrence is that the emotion of fear is going to override the motives for aggression and the emotions of self-righteousness and anger, and keep everyone studiously avoiding situations that threaten to bring them into hot conflict with another nation possessing nuclear weapons–rethinking positions and seeking to deescalate confrontations.

      This works as long as everyone believes it will work, and it will break down when people believe it is about to stop working, and an actual nuclear exchange is about to begin. That’s when they’ll be thinking about preemptive options, either of a symbolic nature (Hey, you, we really mean it!) or of a substantive counterforce nature, and that’s when their thinking will be the most volatile and prone to bursts of insanity, perhaps on the basis of an accident, outbreak of hostilities, or bad information.

      If one side launches an actual nuclear attack, the anger and fear is going to be so great as to completely unhinge decisionmakers on the receiving end, and if as is most likely, the first strike is a limited one against military forces in theater, the path of escalation will be clear and I do not think it will stop short of global holocaust.

      I’m surprised at your comments here, FSB– it’s as if you buy the old baloney about needing so many nukes because the Commies were out to take over the world and had to be deterred from launching their long-planned war of conquest. I see no reason to believe this was ever anything more than superstructure, as Marx might have put it, or let’s say, a rationalization for the levels of overkill that resulted from an out-of-control arms race fueled by the technology that made it possible, and the fact that the other guy was going to do it, too, and the inability of bureaucracies and political systems to comprehend that enough was enough (until finally, they did begin to understand it, and Nixon said, “Let there be SALT”).

      And if the possibility should exist today to go to much lower levels, that’s not so much because the Red commissars are no longer chafing at the bit to parachute into Wyoming, as it is because people think it ought to be possible, now that the Cold War is over and we’re all such good friends. And because, after 60 years of living in the shadow of the Bomb, we’re tired of it and would like to get on with our lives and urgent work before civilization collapses for other reasons.

    • FSB (History)

      Dr. Oelrich’s comment wasn’t on “the basic nature of nuclear deterrence” it was on the nature of deterrence.

      I am not sure what you are disputing in what he says or what I say. Could you clarify more succinctly?

      I think you take exception to the fact that I said some level of nukes do inspire a level of fear that inhibits risk-taking, but I am not clear what you take exception to.

      I do think some very low level of low-potency nukes would be OK if they kept nations from throwing their weight around. What we don’t need are hundreds or thousands of >100 kT devices.

      A handful of 10 kT devices on subs (as anon says) would be OK until such time as we can figure out how to live peacefully.

    • archjr (History)

      FSB: Please enlighten us on the notion that missile defense doesn’t work. Do you mean in all cases? I think that would be a completely unsupportable statement, but I invite you to try.

    • FSB (History)
    • FSB (History)

      This too on Bush’s MD plan — same applies to the SM3s that have never been tested against just warheads without a huge-ass rocket body attached to make seeking easier (see Lewis/Postol article in ACT for details):


  12. Anon (History)

    A few submarines with a few nuclear weapons will suffice. We can talk how “few” over a few beers.

    Land weapons invite devastating attack on the homeland.

  13. Spruce (History)

    Win? Draw? Lose? I have hard time seeing how any of those concepts apply if the nuclear weapons start to fly. If you plan to “win” a nuclear conflict, I would like to see what you think is a win. Is it LeMay’s two Americans and one Russian criteria? Is it total submission of the enemy – regardless of the cost to yourself? Or what?

    I can think of only one reasonable goal for planning for the conduct of nuclear war: survival. The damage caused to food and energy production, sanitation, healthcare, communication, and other systems that the modern society depends on by even a limited nuclear exchange is so serious that it will immediatly put the survival of the state into question. And after that “win or lose” really becomes irrelevant.

    Of course, at that point the best chances for survival might be to totally destroy the enemy. And that is also the problem in the concept of intra-war deterrence. If the nuclear weapons have been used, the opponent has already shown that he considers the conflict worth using nuclear weapons. Also, whatever caused the situation to reach that point must have utterly destroyed the trust between the sides. How would you expect your enemy to trust anything about say about stopping the war in this situation?

    Another problem with the intra-war deterrence is the reaction times. Loss of a single city does not invite a pause to try to stop the exchange. Instead, it invites immediate counter-force strike within seconds (at least way faster than one could hope to even contact the enemy for talks) to minimize the incoming damage. Which, in turn, invites second-strike retaliation – and I think you can see where that leads. Everything from practical experience to behavioral psychology tells that in that kind of stressful situation with stakes that high people won’t be talking to each other. Not that they would have even time to talk as the time pressures means that they would be under pressure to decide on launching their own weapons RIGHT NOW.

  14. Carey Sublette (History)

    Spruce is quite correct in observing that the notion of “winning” (or in Cap’s phrase “prevailing”) is meaningless in a nuclear exchange.

    It is like playing poker when the house takes half the pot in each hand – everybody’s stake rapidly vanishes so everyone is a loser; the only way to cut your losses is to get out of the game, if you got suckered into playing in it in the first place.

    I find Oelrich’s quote remarkably inapt. His phrasing it as “prize denial” sounds as if he is stuck in the early 1950s when “deterrence” meant nuclear retaliation against a Soviet conventional attack. Ever since the USSR developed a significant nuclear strike capability the only meaningful interpretation of “deterrence” is preventing nuclear attack by threat of retaliation in kind. There is no ‘million dollars on the table’.

    Most of the remarks here so far also seem to assume that “first use” will be a deliberate well-informed decision. I agree that the situation where first use is most likely is at the height of some pre-existing crisis, possibly an actual shooting war, but it is very unlikely that any sane person will think that they will end up better off by starting a nuclear exchange. “First use” would almost certainly be due to a breakdown in decision making, command, or control — it would be the result of a mistake of some kind.

    This is one of the reasons that stopping the exchange immediately should be the intent of any nuclear plans — you both got suckered into the game, you need to convince your opponent to cash out now.

    And a major reason why nuclear war-fighting plans, far from being “prudent”, are profoundly wrong headed is that they give your opponent rich opportunity in a crisis to misread your intentions.

    It may turn out, if “first use” ever comes about that things will devolve rapidly into a global holocaust – but the fact that this is a very real possibility is exactly why nuclear strategy and planning must focus on halting this process in every possible way, and not toy with notions of “prevailing”.

    Oddly two popular entertainments have got this exactly right. One is “Fail Safe”, where one U.S. city is sacrificed deliberately to save all the rest (and which would have been lost in the imminent general exchange anyway); and “WarGames” where the tag line is “A strange game. The only winning move is not to play.”

    (Dr. Strangelove is also exactly right in a different way – it examines disastrously wrong-headed approaches.)

    • FSB (History)

      In my view, Oelrich’s comment cuts to the heart of what would be worth fighting over with nukes, now that the Cold War is over.

      Nothing really.

      But given that nukes are there, we can certainly sit down with other nuclear armed nations and work towards having 10s of 10 kT nukes in subs only. The “deterrence” created by the rest of the nukes is inconsequential. (See e.g. Jeffrey’s essay on minimal deterrence).

Pin It on Pinterest