Jeffrey LewisMinimum Deterrence Essay Online

The full-text of my review essay on Minimum Deterrence is available on-line.


  1. Distiller (History)

    Would like to comment on minimum deterrence:

    I guess with that concept a premium must be put on survivability (of the whole strategic complex), because in the end it’s a game of chance, and even a numerically greatly superior enemy has always to be kept afraid, that a counter-strike would devastate him.
    (A few ideas about minimum numbers below).

    And then the question of targeting. Aiming strategic weapons at military targets is wasted. Strategic warheads are clearly made to annihilate population centers. THAT is what deterrence is about. Military complexes might be hardened to withstand – or so the military might think and figure they could ride out an attack, but with population centers that’s a whole different story.
    A strategic weapon is foremost a terror weapon.

    The balanced minimum numbers game (for the U.S., taking the lower SORT limit as basis):
    — 18 active SSBNs (21 in total, derived from SSNs to save costs) with 14 missiles each with 3 RVs each (makes 252 missiles with 756 warheads). Reasoning: Survivability of a smaller number of platforms against sneak attacks is questionable (assuming a 33% peace-time patrol rate = 6), plus more platforms give more attack vectors against BMD systems. It’s basically the Carter option.
    — 270 active rail-mobile launchers (300 in total) with 1 missile each with 3 RVs each. (Makes 270 missiles with 810 warheads). Reasoning: Again railcars are more survivable than silos, the C2 problem basically no different from SSBNs.
    — The question of “rogue regimes” and the limited strike option. For that 30 silo based single-warhead missiles with DAY penetrator warheads (like a modified W80). Reasoning: Deterrence might not always work with lunatics, there might be a situation where a tactical quick strike is required. An ICBM is still fastest, the silo position would make it clear that it is not a general (sneak)-attack on one of the other powers.
    — No airborne component, no strategic cruise missile component.

    — PS: As missile a “Joint Strategic Deterrence Missile” (or whatever you want to call it) is worth a thought, since that commonality would save money and make balancing the force easier. (I’m aware of the different requirements of SLBMs and ICBMs). Warhead somewhere around 450kT fixed yield. No dicking around with DAY, no downloaded-MIRVs. Strategic deterrence has to be a clear, no nonsense game.

    — PPS: The question of MIRVs. Six is obviously an optimum number to destroy an area target like a city, but if you want to keep to an upper limit of warheads, then 6-MIRV missiles result in a too low number of missiles, endandering survivability (esp for the SSBN portion). That’s why three in the above dalliance, to provide a balance of launch platforms, missiles, and warheads.

  2. Yossi

    A very good essay, Dr. Lewis!

    Distiller made an impressive analysis of the technical side. I would like to add a layman’s comment on deterrence.

    The main practical problem is the common human tendency to attribute non-humanity to the other side. If they are demons (e.g. fanatical communists, islamists or some new curse) they will certainly be happy to sacrifice their babies in order to kill ours, wouldn’t they?

    Well, our short nuclear history seems to teach us that no ruling elite wants to see a significant part of its country become a nuclear desert and is ready to be quite flexible in order to prevent this.

    We are now told the Russian elite assassinated Stalin so he wouldn’t start a nuclear war with the USA. Lieutenant colonel Stanislav Petrov made a very difficult decision in order to prevent a very dangerous escalation.

    We live today because these Russians made the right thing at the right moment, so maybe they are not so terrible after all but simply good people?

  3. Carey Sublette (History)

    Jeffrey, the crux of your essay is the fundamental unreality of the theories upon which main-stream strategic thinking is based.

    I agree, but think your essay does not go far enough in exposing this.

    The essay raises the question of whether a minimum deterrence force would be “aimed at military targets and the enemy‚Äôs weapons … or would they be be pointed toward population centers – a countervalue strategy? “, and cites legal opinions holding that countervalue targeting is not lawful.

    I argue that this is a hypothetical dilemma of no relevance to deterrence in the real world.

    Certainly there is a crucial legal distinction between legitimate military operations that produce civilian casualties and destruction as a side-effect, and illegitmate ones that cause civilian casualties and destruction as the intended primary result.

    And in any given nation there are some important military targets that are fortuitously removed from civilian areas so that nuclear attack can be contemplated without subjecting a large civilian population to harm.

    But I use the term ‘fortuitously’ advisedly – there are few such targets. Most significant military targets lie close to urban areas because of their dependence on the infrastructure of a civilian economy (or are an integral part of that economy). Even strategic military targets that are deliberately located at great distances from urban areas typically expose downwind cities to deadly fallout hazards due to their hardening, and thus the necessity of high-yield low altitude bursts to destroy them with high degrees of confidence.

    In reality, any substantial nuclear strike against an array of military targets sufficient to cripple a nation’s ability to fight a war (surely the objective of such a strike) would inflict such collateral damage that few nations have ever experienced in any war.

    One does not have to engage in illegal city-busting targeting to achieve deterrence with a “small” arsenal of a few hundred warheads. Any prospective strike against important military targets numbering in the hundreds, or truly critical ones numbering in the dozens, will result in such destruction that deterrence will be achieved, if it can be achieved at all.

    Personally, I am ambivalent bout the term “minimum deterrence” partly because of the semantic games it entices opponents into playing. E.g. if one has a “minimum deterrent”, then if an opponent attacks first and destroys part of it, whatever is left must be “less than the minimum”, etc. The term I have used myself is “simple deterrence”, emphasizing the rejection of notions of “war fighting” that drive warhead counts to arbitrary high levels.

  4. bobbymike (History)

    Minimum deterrence is having enough weapons to deter any possible agressor. As important is to also have a responsive infrastructure to prevent strategic surprise. Platforms should be large and numerous able to be uploaded with 10+ warheads. Another key is to have the most modern force. The US need to rebuild its strategic nuclear strike force.