Michael KreponDe-alerting and De-legitimization

In the September/October 2010 issue of Foreign Affairs [“Smaller and Safer, A New Plan for Nuclear Postures”] Bruce Blair and his co-authors argue that de-alerting is an important step to prevent nuclear terrorism as well as a tool to de-legitimize nuclear weapons:

These [high readiness] postures also perpetuate a mutual reliance on nuclear weapons that lends legitimacy to the nuclear ambitions of other nations. When more states go nuclear, intentional use becomes more likely, and deficiencies in nuclear command and warning systems multiply the risk of accidental or unauthorized use or terrorist theft.

Bruce has been right for a very long time that the state of launch readiness for U.S. and Russian forces is excessive. Reductions in prompt-use capabilities would be welcome as confirmation of the distance we have all traveled since the Cold War ended. Another long-standing argument for de-alerting – that the high state of launch readiness is dangerous – was especially true during the Cold War and immediately before and after the Soviet Union’s demise. This argument is less convincing now that Washington and Moscow have nothing to fight about and since Russia has stabilized.

What about the argument that de-alerting can help reduce the threat of nuclear terrorism? This threat is near or at the very top of everyone’s list of nuclear nightmares, so any preventive measure would seemingly be welcome. Nevertheless, I have my doubts about the validity of this argument.

Take Pakistan, the state with nuclear weapons facing the greatest internal security threats. There have now been two commando-style raids with insider help against important military compounds. Army Headquarters in Rawalpindi was attacked in October 2009 and the Mehran naval base in Karachi was attacked in May of this year. These patrol-sized assaults took approximately eighteen hours to quell. Commando raids with insider help are a very different ballgame than truck bombs, meriting greater security measures at sensitive sites. Solutions to this problem do not include de-alerting, because military authorities normally maintain nuclear forces in a de-alerted state. Fissile cores may be separated from warheads, warheads are separated from launchers, and guarded as if they are crown jewels.

Another worrisome scenario in South Asia is that separated warheads and launchers are moved during periods of great tension. The tradeoff for Pakistan is that warheads in central storage sites are harder targets for internal security threats but easier targets for India. Conversely, warheads in motion are harder targets for India and less difficult targets for internal threats.

During crises, Rawalpindi and New Delhi employ changes in readiness to signal resolve and their interest in Washington’s crisis management. Nuclear-related alert rates on the subcontinent, along with the deployment of strike corps to fighting corridors and associated measures, are key elements of crisis behavior. All parties know exactly what to look for. Nevertheless, these moves can be overdramatized and misunderstood, even when warheads remain separated from launchers.

For these and other reasons, de-alerting is a weak remedy for nuclear dangers in Pakistan. Ditto for India, where fissile cores may be separated from warheads, warheads are separated from launchers, and military leaders are kept distant from civilian control. In the future, when India and Pakistan deploy nuclear weapon capabilities at sea, the remedy of de-alerting will become more useful, but also very hard to put in place with any degree of assurance.

The argument that de-alerting can be a useful instrument to de-legitimatize nuclear weapons is also questionable. Pakistan, India and other states possessing nuclear weapons aren’t inclined to emulate the United States and Russia in ways that begin to suggest the de-legitimization of their deterrents.

The Obama administration prefers the term “maximizing presidential decision time” to de-alerting. Team Obama found value in reducing force levels but not launch readiness in its 2010 Nuclear Posture Review. In my view, de-alerting U.S. and Russian forces is an idea that has been eclipsed by a far better idea — dismantling nuclear warheads and their delivery vehicles. Progress, whether by de-alerting or arms reduction, will occur incrementally, over considerable resistance. At this juncture, incremental reductions make far more sense that incremental de-alerting.


  1. Andrew Tubbiolo (History)

    Am I mistaken in understanding that the Russians are keeping their larger stockpile in order to address the fact that NATO is right on their border and Russia lacks any real allies in Europe? The question it seems is how do you give relief to the Russians without compromising the former Warsaw Pact nations? Is there really a way to make the Russians feel secure enough to shrink their nuclear arsenal and accept the fact that the vast majority of European conventional forces are arrayed against them in one form or another?

    With the Chinese about to embark on a very large nuclear power program, and Russia and The United States about to enter another cycle of deep cuts, at what point in the near future will the Chinese arsenal become a good sized fraction of the Russian and American stockpile? How will that impact alert level, and the ability to de-alert, or cut?

    • Scott Monje (History)

      The Russians also justify their stockpile on the low state of readiness of their conventional forces. After years of talking about it, a reform of conventional forces has begun, although some say that it is haphazard and not well thought out.

      I understand that Chinese forces, so far, are maintained at an extremely low alert level, dismantled even.

  2. John Hallam (History)

    Well, first of all I should declare my interest in this. It was my efforts at getting together with Doug Mattern (who died recently) a declaration urging de-alerting in 2004, signed by 44 nobels, that persuaded six governments to put the issue up to the General Assembly in 2007, 2008, and 2010.

    I have been organising workshops at the UN on it ever since 2006.

    And the last workshop we organised in 2010, had a principal co-author of the Foreign Affairs paper, Colonel Valery Yarynich, speaking at it.

    Well, It seems to me that Mike Krepon has had a change of heart since he saw Fail Safe!

    The reason the Obama administration went ‘cold’ on de-alerting is the same reason it went cold on many other issues it should not have gone cold on – loss of nerve.

    Senior US and Russian military figures continue to insist that large numbers of nuclear weapons on high alert (whether or not we call that ‘hair trigger’ is an almost meaningless question), of itself makes compressed decision-making times inevitable.

    Undoubtedly, increasing presidential decisionmaking time is at the very core of the de-alerting issue.

    And as Colonel Valery and other military from both Russian and US missile forces argue, those ridiculously short decisionmaking times continue to pose a potential threat to us all.

    The Blair/Esin/Valery/Matt Mc Kinzie paper shows, I believe convincingly, that a key argument against de-alerting, namely that there would be a ‘re-alerting race’ is plain wrong, and that it is possible to show with some statistical rigour that the danger of complete global catastrophe with nuclear forces off high alert is, surprise surprise, less that with them on high alert.

    Therefore, we will just have to keep right on arguing for measures that do in fact increase decisionmaking time. And such measures, however they are talked about, amount to the very conceptual core of de-alerting.

    John Hallam
    PND Nuclear Flashpoints project

    • John Hallam (History)

      Well, obviously I believe that de-alerting, or increasing decision-making time or whatever you want to call it, must take place in tandem with reductions in arsenals. However, as long as thousands of warheads are capable of being launched within a couple of minutes, and as long as those warheads remain vulnerable to an ‘enemy’ strike, we have a problem with potentially apocalyptic consequences whether or not the cold war is supposedly ‘over’. And going to lower alert levels, however incrementally, would indeed signal that it was really, truly, over. So far that doesn’t seem to be the case.

      John Hallam

  3. Cameron (History)

    It seems that the impact that de-alerting would have on terrorism is through limiting the spread of weapons. A specific example like Pakistan likely wouldn’t convince the authors, who are relying on superpower standing down of forces to help de-legitimize weapons. After the weapons are de-legitimized then new states will no longer want them, so the conjectured terrorists will have fewer targets.

    I’d respond that states look to nuclear weapons for different strategic purposes, and that preventing the evolution of their viewed purpose could have been impacted by such a change. But the evolution from Cold War standoff to their current use and what their strategic use is becoming has already occured.

  4. Aaron Tovish (History)

    (Please use this proofread version.)

    At the beginning of this piece, Micheal acknowledges two powerful reasons for ‘de-alerting’, but by the end of the piece he seems to have forgotten them. The fact that he finds other arguments for de-alerting are less cogent, should not cancel out the cogency of the other two — which have always been the main arguments anyway. Granted, he argues that the danger posed by high alert is less than it was during the Cold War, but given the extraordinary consequences of failure (something FAR TOO OFTEN overlooked by nearly all ‘wonks’) ‘less’ is never enough.
    But I do think there is value in considering on how reductions could LEAD to retirement of the launch-on-warning option. Reductions should be focused on those weapon systems which are vulnerable to preemption, mainly the fixed land-based systems. Once they are eliminated (either bilaterally or unilaterally), the use’m or lose’m pressure is gone. The attack (if it is not a false alarm!) can be absorbed and then the form of retaliation decided in good time. In this regard, I am not happy with the formulation “maximizing presidential decision time.” A submarine off the coast of Maryland can deliver a low-trajectory strike on the White House in less than ten minutes. There’s your maximum. Any improved posture must assume that command is likely to pass out of the hands of the President.
    As noted parenthetically above, given the catastrophic consequences of the use of just a couple of dozen modern warhead on or near cities, any discussion of a rational retaliation in those circumstances is superfluous. The attack will results in tens if not hundreds of millions of deaths by famine and pandemic in the attackers homeland, even if there is no direct retaliation. Folks, it is time to stop ignoring this reality.

    • Andrew Tubbiolo (History)

      Mr Tovish and Hallman, thank you for your thoughtful responses. I see your points and consider them more than valid. However let me bring up a point. Both of you make excellent points of how the system of alerted arsenals is stressed from time to time and how disaster is averted. It’s a normal human reaction to think that after repeated stress it’s only a matter of time until the system breaks. Which of course it can. However you both give examples that span 30 years of systems failure but happily non failure. Don’t you think that you’ve also made an observation of how robust the human side of the system has been overall?

      Which again brings me to my point. Start at the easiest least intrusive means of disarmament to address the problem. Until the arms control community can address the various core insecurities that drive the keeping of nuclear arsenals, the dream of going to zero will remain a dream. I think a similar logic applies to alert levels. My point was to start at the largest stockpile where the shock to the functionality of the arsenals would be the least. This might have the effect of allowing the politics and military postures time to settle into the new reality and establish a new quasi stable state.

      Let’s face it, we’re not at the opening of a new age that will allow for total de-alerting of the arsenals. It’s not about to happen. We’re about to start talking about it. Maybe. If it is going to happen it’s going to be a decades long process, and it’s going to happen during a time when the global balance of conventional forces is in flux. I think going slow and feeling your way forward will in the end not get us there any slower after you figure in all the time and effort that will go into doing it all at once and setting up the inspection regime, treaty terms, and sorting out the various local politics. We’ve lived with what we have in one form or another now for 60 years. Overcoming that inertia is going to be very slow.

  5. Andrew Tubbiolo (History)

    Could there be a spectrum of relaxed posture? I fully appreciate the consequence of an alerted weapon being launched and detonating on target as an accident. Bad as that would be it would not be the end of the world. However launching a full bore counter force strike under a compressed decision cycle enforced by the existence of a large number of alerted forces would be doomsday. Since we’ve lived in a world where the ‘classical’ counterforce exchange is the least likely as viewed from the point of view of the political crises we have and are facing, perhaps those are the forces we can start a slow gradual grinding de-alert process. A slow gradual approach would allow time to tweak-observe-analyze and repeat the process. Is there a reason to do it all at once?

    • John Hallam (History)

      The deadly paradox of the whole de-alerting debate is that the ‘accidental launch’ of a single weapon is the LEAST likely scenario, and, as it is ‘not the end of the world’, is the scenario we have to worry about least of all.

      (Perhaps I should qualify that by noting that in spite of what I have just said, in 1979 and again sometime in the very early 80s (80 or 81), practice launch sequences at Minuteman missile silos turned into the real thing and would not stop. In each case it was not a case of crazed launch personnel doing unauthorized launches: On the contrary, the officers concerned had heavy military vehicles, as I understand, driven onto silo doors to make the launch physically impossible. Full marks for initiative.)

      But to return to my original theme:

      It is precisely the legal, authorized, launch of a full ICBM salvo by miscalculation and panic, in a highly compressed decision-making time-frame, and in the absence of proper information (which is simply not to be had in so short a time), which is the ‘doomsday nightmare’.

      And there are a disturbingly large number of times we’ve come close to it:

      In the early 80’s in the US three times, combat computers at NORAD indicated hundreds, or thousands, of incoming Soviet warheads – resulting in the launch of the ‘doomsday plane’, and minuteman crews being ordered to be launch-ready, while nuclear – armed fighter bombers were taxied to the edges of runways with motors running.

      From 1985, all US ‘near misses’ have been classified.

      In 1983, Colonel Stanislav Petrov of the Soviet missile forces had his now famous brush with the apocalypse, resolving that he had a false alarm amidst wailing sirens and flashing lights. In Col Valery’s classic Russian understatement ‘I think Colonel Petrov acted responsibly’.

      And in 1995, the Norwegians launched their famous weather research rocket, having notified the Russians by fax some weeks previously that they were going to do it.

      Unfortunately the fax seems to have been mislain and russian perimiter radar did exactly what it was trained to do, namely assumed the rocket was a US submarine-launch first strike aimed at taking out the Kremlin.

      The alert went right up to Yeltsin, who it seems fortunately, did not believe what his high command was telling him.

      It was this incident that impelled Yeltsin and Clinton to sign an agreement to establish a ‘joint strategic stability centre’. That centre has now been announced no less than three (maybe four) times by the two governments, but has never actually become operational.

      Finally I want to note that the CONSEQUENCES of the ‘ultimate’ bad-hair day at STRATCOM or Kosvinsky Mt – ie the use of a number of thousands of 300Kt-1Mt weapons, to a large extent for city-busting – is a mass extinction event.

      The peer-reviewed work done in 2006 By Toon and Robock makes this perfectly clear.

      The Swiss government is now taking this up and doing a formal study on consequences of large-scale nuclear weapons use.

      Given the utterly apocalyptic consequences of glitches in nuclear command and control systems, lowering operational readiness is a matter of urgency, as is the complete elimination of these weapons.

      John Hallam

    • Aaron Tovish (History)

      Andrew, I think one should be worried that the alerted missile being launched would be perceived by the target nation as something requiring a launch-on-warning response, etc., etc. It is hardly material how ‘less likely’ such scenarios are as long as they are not actually and completely ruled out. As long as disaster of this magnitude is built into the systems and procedures, it will not help a whit to say after the fact, “Gee, I didn’t think it would actually happen.” Postponing action is just plain negligent. It is time to stop pretending one can be diligent within a construct which is inherently reckless. Aaron

  6. bobbymike (History)

    I thought I read an excerpt from a speech given by Gen Kevin Chilton at one of the Strategic Deterrence Conferences that de-alerting is really a fallacy.

    The missiles are in the tubes ready for launch and there is really no way than to dangerously take them offline (they may not come online when needed)to “de-alert” them.

    They are ready to launch or not ready to launch there is no middle ground, to paraphrase Chilton.

Pin It on Pinterest