« An Answer for GregB: An Islamic Reformation? | Main | Israel rebukes America over treachery toward Mubarak. »

31 January 2011


Feed You can follow this conversation by subscribing to the comment feed for this post.

David Habakkuk

William R. Cumming,

That stated the high level nuke war theorists ultimately reached a consensus that those with the most weapons after the nuclear war ended won the war.

That was a view that the Soviets ceased to hold in the course of the late Sixties and early Seventies – at which time they abandoned the belief that a successful preemptive attack might make it possible to prevent a nuclear war being anything other than an utter and irretrievable catastrophe.

This was obscured by the tendentious reading of documents in the 2005 study 'A Cardboard Castle' edited by Vojtech Mastny and Malcolm Byrne, but was actually clearly established in a 1995 study by the Pentagon contractor BDM Corporation, which was declassified in 2009.

(See http://www.gwu.edu/~nsarchiv/nukevault/ebb285/index.htm )

From the summary on the National Security Archive website:

The Soviet military high command "understood the devastating consequences of nuclear war" and believed that nuclear weapons use had to be avoided at "all costs." In 1968, a Defense Ministry study showed that Moscow could not win a nuclear war, even if it launched a first strike. Although Soviet ideology had insisted that survival was possible, no one in the leadership believed it. In 1981, the General Staff concluded that "nuclear use would be catastrophic." [I: 23-24, 26; II: 24 (Danilevich), 124 (Mozzhorin)] This does not support arguments made by Richard Pipes in the late 1970s that the Soviets did not believe that a nuclear war would result in "mutual suicide" and that the "country better prepared for it and in possession of a superior strategy could win and emerge a viable society."

Before the 1970s, Soviet military officials paid no attention to the environmental impact of nuclear war, but they began to recognize that "drastic effects on climate" would be among the catastrophic effects of nuclear war (See document 5, "Stockholm Roundtable," p. 65). According to Dr. Vitalii Nikolaevich Tsygichko, a Senior Analyst at the Academy of Sciences, the author of the study, Mathematical Model of Soviet Strategic Operations on the Continental Theater (see document 4, with summary attached), and a former member of the General Staff, military analysts discussed the idea of a "nuclear winter" (although they did not use that term) years before U.S. scientists wrote about it in the 1980s. [II: 39 (Danilevich), 137, 139, 142 (Tsygichko).

A brief personal reminiscence. At the start of 1989, when I was working for BBC Radio, a colleague and I interviewed the military figure most closely associated with the so-called 'new thinking', General-Mayor Valentin Larionov, then about to retire as Professor at the General Staff Academy. He was a scholarly man with steel teeth.

At the start of the interview, he said that to understand the 'new thinking', one had to go back to the realisation by the Soviets in the Seventies that they could not win a nuclear war. Subsequently he talked a great deal about a Soviet theorist of the Twenties, Aleksandr Svechin, whom he said had been 'repressed' under Stalin.

Unfortunately at that time I had not come across what was then the Soviet Army Studies Office at Fort Leavenworth, which subsequently became the Foreign Military Studies Office. It had been founded in 1986, I understand, with the leading Western authority on the military aspects of the war in the East in 1941-5, Colonel David Glantz, playing a key role, and employed two pioneering historians of the evolution of Soviet conceptions of 'operational art', Dr Buce W. Menning and Dr Jacob W. Kipp.

It was only some years later that I came across an English translation, published in 1991, of Svechin's 1927 study Strategy, with an introduction co-authored by Larionov, and an extremely informative discussion of Svechin by Kipp.

From the editor's introduction, by Kent D. Lee, I learned that Larionov 'comes to us distinguished as perhaps the most outstanding Soviet military intellectual of the postwar period', and that he had been 'one of the fundamental forces' behind the study of Military Strategy originally published under the name of Marshal Sokolovskii in 1962 – the classic statement of the strategy to win a nuclear war by preemption. (In fact he compiled it, as well as being co-author.)

So it was a sometime leading advocate of nuclear pre-emption who had become the leading advocate of the liquidation of the contingency planning for an offensive into Western Europe – which was a key part of the point of his drawing on Svechin.

Had I been aware at the time of what Dr Kipp had already published on the evolution of Soviet strategic thinking in the interwar years, and the way that Svechin had been marginalised by Tukhachevskii, I would have been better able to understand what was being said to me.

But I suspect my ignorance was shared by most figures in American and British intelligence. One of the extraordinary things about the way Western 'strategic studies' developed in the Cold War, I think, was how marginal the serious scholarly analysis of the history of Soviet military thinking was.


David Habakkuk,

A very thoughtful summary. On your ending comments, isn't the same marginal analysis of trends in the Middle East also a problem today?

William R. Cumming

Thanks David and for references. It was computer power that allowed other than the nuclear priesthood to understand the climate impacts of execution of the SIOP thus Carl Sagan and others in 1983 SCIENCE article and others on so-called NUCLEAR WINTER. The writings of retired 4-star Lee Butler are also of relevance. Unfortunately, MAD is still the strategic doctrine of both US and RUSSIA. Although some considerations to changing that doctrine. The problem there is almost no NUKE strategy or doctrine that allows any particular state to WIN. The concept behind WAR GAMES with Matthew Broderick in the Hollywood movie. IMO nuclear weapons will be employed somehow and somewhere and just wondering how far up the chain of foreseeable or unforeseeable events that employment will march. I actually do have my own theories of the employment of NUKES but refuse to share them on the basis they might be adopted by someone.

Neil Richardson


"Unfortunately, MAD is still the strategic doctrine of both US and RUSSIA. Although some considerations to changing that doctrine. The problem there is almost no NUKE strategy or doctrine that allows any particular state to WIN. "

What other strategic doctrines did you have in mind? I understand the desire to keep certain thoughts private, but MAD isn't just a strategy. It is just a state of existence. One could conceivably go from MAD to something "better" but like many things in life, it's the transition that will likely pose the biggest threat (e.g., miscalculations). MAD isn't perfect, but it's still better than counterforce, counterforce/BMD, no-first-use or any other alternatives. Uncertainty regarding the usage of nuclear weapons is a good thing IMHO. As Robert Jervis pointed out, because the cost of a potential exchange is so high it makes the strategic calculus relatively simple for policymakers. That has been valuable in preventing a general war.

Will there be a situation where two nuclear states (e.g., Pakistan and India) stumble into a "stability-instability paradox" and a conflict spirals out of control? It's possible, but the fact remains that MAD is a state of existence and that usually forces political leadership to take a deep breath in any crisis. I'm not worried about a bolt-out-of-the-blue nuclear attacks. I'm more worried about command and control especially during transition period toward survivable second strike capability. I would hope that the United States and other nuclear states readily share permissive action link technology to the extent possible with new states.

William R. Cumming

Neil! Thanks for the invitation but still decline. I do think that if the US announced NO FIRST USE it would introduce an interesting dynamic into Nuclear proliferation issues and strategy and tactics. I believe it actually is the unstated policy of the US but as always we like ambiguity even when it does not serve our purposes. IMO renouncing FIRST USE would be a step forwards to lowering world background tensions. And given warning times--almost non-existent, that doctrine is a challenge when US is most likely to be hit by a surface to surface missile being shot off a sub or ship just off our coastline. Essentially a no warning shot.

David Habakkuk


I do very much think that both in London and Washington hazy ideological schema are used to interpret situations which can only be made sense of in terms of very detailed and hard-headed analysis, which characteristically takes one back into history – often a long way back. The crap from Martin Indyk about being 'on the wrong side of history' is representative of the kind of hazy thinking which is all too common, alike in Washington and London.

In an earlier thread you asked whether I thought the the pessimism of the Tsarist official Peter Durnovo about revolution in Russia was relevant to Tunisia and Egypt. I do not know enough about the Arab world to have a considered view. I am hoping it is not.

William R. Cumming,

I read with great interest – and admiration – what General Butler was writing back in the second half of the Nineties, and I much regret that he seems to have ceased talking about these matters, given his wealth of experience.

There is a fascinating interview with him by Jonathan Schell, of which extracts are available on the web in a discussion of the gulf between appearance and reality in nuclear war planning by a leading academic expert on nuclear command and control, Bruce Blair. Actually Blair is a very interesting figure, because before turning academic he had been – back in the early 70s – a Minuteman launch control officer.

In that capacity he was puzzled by the fact that according to officially promulgated 'deterrence' theory, which was based on the notion that strategic stability was assured by a capacity for second strike retaliation, he and his colleagues should have been doing exercises where they rode out a Soviet attack, and then retaliated. However, in almost all the drills he was ordered to carry out, the missiles were fired before the Soviet attack had occurred.

If a suspicious Soviet official had got wind of this – and I would be interested to know if the Soviets were familiar with the ways in which the Minuteman force was exercised – it would have been likely to reinforce the belief we know they were inclined to hold that talk about 'deterrence' was humbug, that what was at issue was 'compellence', and that U.S. contingency planning featured ideas of 'first strike' in a prominent place.

In fact, this would have been a quite erroneous interpretation. The problem was that the academics who devised the theories of strategic stability did not reckon with awkward facts like the difficulty of devising a command and control system which would be resilient under all-out thermonuclear attack – a problem which pushed the people responsible for contingency planning to a posture of launch on warning or launch under attack.

The problem and the outcome, was clearly stated by General Butler in the interview with Schell, which was published in 1998:

“Part of the insidiousness of the evolution of this system … is the unfortunate fact that, whatever might have been intended by the policymakers (who, incidentally, had very little insight into the mechanisms that underpinned the simple words that floated onto a blank page at the level of the White House), in reality, at the operational level, the requirements of deterrence proved impracticable…. The consequence was a move in practice to a system structured to drive the president invariably toward a decision to launch under attack…. Launch under attack means that you believe you have incontrovertible proof that warheads actually are on the way….. Our policy was premised on being able to accept the first wave of attacks. We never said publicly that we were committed to launch on warning or launch under attack. Yet at the operational level it was never accepted that if the presidential decisions went to a certain tick of the clock, we would lose a major portion of our forces… Notwithstanding the intention of deterrence as it is expressed at the policy level – as it is declared and written down – at the level of operations those intentions got turned on their head, as the people who are responsible for actually devising the war plan faced the dilemmas and blind alleys of concrete practice. Those mattered absolutely to the people who had to sit down and try to frame the detailed guidance to exact destruction of 80 percent of the adversary’s nuclear forces. When they realized that they could not in fact assure those levels of damage if the president chose to ride out an attack, what then did they do? They built a construct that powerfully biased the president’s decision process toward launch before the arrival of the first enemy warhead.” (Jonathan Schell, The Gift of Time, Metropolitan Books, 1998, pp. 191-194).

(These extracts, with a useful discussion by Blair, are at http://www.cdi.org/blair/launch-on-warning.cfm )

How the command and control problems were handled with the tactical nuclear arsenal – in particular given its obvious vulnerability to being overrun in the case of a Soviet offensive – I do not know.

But having once been a believer in academic 'deterrence' theory, I know think it is a fundamentally flawed and actually very dangerous pseudo-science.

Postscript: I only saw your comment about first-use after drafting the above. I strongly agree, and think that this indeed even from a narrowly self-interested point of view this would make sense for the United States. Back in 1989, it was already perfectly evident that technological developments were in the process of shifting the conventional balance in Europe decisively towards the United States – and accordingly that actual or potential enemies of the United States were likely to be tempted to rely on first-use, on grounds similar to those used to justify NATO's reliance confronted by the conventional power of the Warsaw Pact. So it was very much in the interests of your country, as of mine, to seek to delegitimise strategies of 'deterrence'. We chose to do precisely the reverse.

William R. Cumming

Thanks David for citations and discussion. Very helpful. Disclosure: I was a Pershing 1 LCO.
Pershing unclassified range 600 miles and TOF about 6 minutes from launch. A ballistic, solid fuel missile about same size as a NAZI V-2 as pointed out to me several times by FRG officers and NCO's. V-2 of course liquid fueled. I watched many times FRG nuclear capable units exercising. Fantasy land.

Neil Richardson

Dear Mr. Habbakuk:

"Conventional balance" shifts periodically. Maintaining conventional deterrence is a rather tricky business IMHO. As for Blair and Gen. Butler's comments, I am not certain when these drills were conducted. Was it during the 1970s when there was a clear shift toward counterforce (e.g., PD-59)? I don't doubt Gen. Butler's words but Minutemen were "use 'em or lose 'em" leg of the triad given their vulnerability. That was the reason why the United States misinterpreted the Soviet nuclear doctrine (they had their own reasons for relying heavily on the land based leg). However, why would a President or the surviving NCA be pressured into launching them in any scenario other than a preemptive disarming strike which would inevitably fail?

I do appreciate the concerns about LOW, but it seems to me both the US and the Soviet Union exercised restraint in crisis even when one side theoretically possessed "first strike capability" as well as local conventional superiority (e.g., the Cuban Missile Crisis).

Personally I don't have a problem with NFU. It might even be a good idea if we are certain that it would lower the rate of "normal accidents" in terms of command and control and maintaining readiness. However, I'm not certain if the threshold for nuclear deterrence is that high enough for one side to expect or even contemplate a counterforce strike against the US.

FB Ali

The discussions of the theory and practice of MAD and deterrence between the West and the East are interesting. At present, it seems to me, whatever the existing theory and practice, the chances of either side using nuclear weapons are very small.

What should be of concern is the real possibility of a nuclear exchange between Pakistan and India. In both countries, despite such decisions being theoretically in the hands of the political leadership, in practice they are effectively in the hands of the generals. The latter, in both countries, have never shown much evidence of the knowledge and sophistication required to handle such decisions wisely. The chances of a miscalculation leading to a spiral of escalation are very real. The tensions between the two countries provide many points for conflict and escalation.

A study of the consequences of such a war estimated that, apart from the horrendous casualties in these two countries, it would result in a ‘nuclear winter’ across the globe for several years, thus crippling global agriculture and threatening around a billion people worldwide with starvation.

David Habakkuk

William Cumming – I had noted with much interest that you had been a Pershing I LCO.

I am just disappearing for a long weekend, so I cannot do either F.B. Ali's comments or those of Neil Richardson the justice they deserve. But there will be occasions to return to these matters.

But briefly. What F.B. Ali says about the weakness of political control in India and Pakistan, and also the scepticism about the ability of the generals to handle decisions wisely, I find extremely alarming. However, I also think that nuclear weapons confront military planners with dilemmas to which they are no good answers. The vulnerability of command and control is one case in point. There is a basic tension between the requirements of what I understand is called in the jargon 'positive control', and what is called 'negative control' – the requirement to ensure that your arsenal cannot be disabled – and 'negative control' – the requirement to prevent unauthorised launch. At one end of the spectrum, if both the legal authority and the physical ability to launch the arsenal are concentrated in the hands of the political leadership, there is vulnerability to decapitation. The further however that the authority and even more physical ability is decentralised, to obviate this risk, the greater the risk of accidental and inadvertent launch, particularly in the event of a war going out of control. All one can do is attempt to balance the requirements, and no balance is particularly satisfactory.

I would be most interested to have some picture as to how these dilemmas are confronted by the Pakistan and Indian military.

As to the question of the stability or otherwise of conventional 'deterrence', I am going to have the temerity to disagree with Neil Richardson about this, and hope to produce a post on the subject at some point in the near future. The view to which I have come is that in fact a conventional balance between the Soviet Union and the United States would probably have been far stabler than the nuclear balance.

The basic point can be put very briefly, in the terminology Aleksandr Svechin took from the German military historian and theorist Hans Delbrück. In a world without nuclear weapons, both the United States and the Soviet Union would have had to plan for a war of attrition – in which there would have been a near certainty of strategic warning prior to the outbreak of hostilities. It was only nuclear weapons which opened up the possibility of a war of destruction, in which there was no certainty of strategic warning.

A war of attrition would have been an most unattractive prospect not only for the leadership of the United States, but also – for rather different reasons – for that of the Soviet Union. Among them was the immense disparity in military-industrial potential in favour of the United States, the strength of American strategic air power (even absent nuclear weapons) and naval power, and the demonstrated capabilities of the Americans and British for large-scale amphibious operations. Also relevant are the internal vulnerabilities of the Soviet system, which the Germans had exploited so successfully in the First World War, and failed to exploit in the Second.

FB Ali

David Habakkuk,

The principal danger does not arise from the two countries’ command and control systems, but from their policies, and how they would be implemented in a crisis. This is an issue worth considering, but this is not the best place for that ‒ buried under all the later posts.

I will try one of these days to write a post on the subject for the blog, and look forward to reading your views and comments.

William R. Cumming

Thanks David H. and General Ali.
Clearly nuclear command and control and nuclear surety isse are very important as we close in on the expansion of the nuclear club over the next 30 years. My comments were directed to the concepts developed by "rational" minds and as PL always points out correctly it is completely uncertain as to whether "rational" minds are in leadership positions--politically or militarily in some or all of the nuclear powers. My belief is the notion that whether conventional or nuclear war the belief in the concept of "winning" may well depend on a view of history that does not comport with my own. And I do believe that there will be NUDETS at some future time on the face of the earth. A tragic outcome for all human kind.

David Habakkuk

F.B. Ali,

I very much look forward to your post, as I am shamefully ignorant about the nuclear policies alike of India and Pakistan.

William R. Cumming,

I would agree with everything you say. Indeed, I think that the attempt to 'model' nuclear relationships on the basis of the interactions of 'rational' actors has been fraught with danger. The problem is not simply that one cannot have much confidence in the 'rationality' with which these matters are handled. It is also that if the kind of crisis erupted where the possibility of nuclear threats being implemented was real, this might occur so rapidly that people would hardly have time to think. Obviously one hopes that crises can be managed, but the really dangerous crises may be inherently uncontrollable.

The comments to this entry are closed.

My Photo

February 2021

Sun Mon Tue Wed Thu Fri Sat
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
Blog powered by Typepad