Michael KreponInferred vs. Demonstrable Deterrence

The Stimson Center will release an edited volume in September comparing and contrasting deterrence against nuclear and space warfare. Here’s a preview.

Nuclear deterrence and space deterrence have common elements as well as distinct differences. No difference is more striking than the visibility of deterrence capabilities in one domain and its lack of visibility in the other.

The advent of nuclear weapons was advertised with spectacular effect. The mushroom cloud instantly became the symbol of the “atomic age.” Ever since, nuclear deterrence was widely presumed to be strengthened by visible displays. Tests of warhead designs came with exclamation points when carried out in the atmosphere, and continued to leave a distinct impression once driven underground. Missile flight tests repeatedly affirmed vigilance and readiness. Some states still parade nuclear-capable missiles on national holidays.

In contrast, capabilities to harm space assets have been tested only occasionally in dramatic ways and have mostly been pursued quietly or by indirect methods. Consequently, space warfare capabilities rarely make headlines, unlike actions signaling nuclear deterrence, which are the subject of intense public and media attention.

While nuclear deterrence rests on deployed or readily deployed capabilities, the weaponization of space – defined here as the placement of dedicated war-fighting capabilities in this domain – has yet to occur. The nuclear superpowers deployed nuclear weapon delivery vehicles carrying thousands of warheads, many ready for launch on short notice. In stark contrast, US and Soviet military capabilities specifically designed to harm satellites were rarely deployed, had limited operational utility, and were subsequently mothballed. Anti-satellite tests during the Cold War were surprisingly rare phenomena.

There were, to be sure, years of heightened military friction and competition is space, especially in the 1970s, prompted by Soviet ASAT tests, and in the 1980s, after President Ronald Reagan announced his Strategic Defense Initiative. These brief periods of intensified competition did not result in arms races in space and ended with whimpers, not bangs. Both superpowers were extremely careful not to cross each other’s red lines in space, no less than on the ground and at sea. Just as Washington and Moscow learned not to play with fire in particularly sensitive zones after the Berlin and Cuban missile crises, so, too, they reached tacit and diplomatic agreements not to create havoc with satellites.

Why was demonstrable deterrence deemed to be crucial for nuclear capabilities, while inferential deterrence sufficed for space? One reason was that missile systems designed for one domain could be adapted for use in the other. Over time, reassurance found a place alongside deterrence to keep the Cold War from becoming hot. The nuclear superpowers figured out how to cooperate as well as compete in both domains.

The United States and China now find themselves in an intensified competition over the military use of space, punctuated by tests explicitly demonstrating ASAT capabilities during the George W. Bush administration. Will this competition continue unabated, or will the Cold War track record of surprising restraint and inferential deterrence be extended?

Success in the past was based on a mutual recognition that attacks on satellites invited uncontrolled escalation and horrific consequences. Inferential deterrence was also reinforced by language protective of satellites embedded in diplomatic accords. Military planning and doctrine will continue to be offensively-minded: that’s what major powers do. Will cooler heads continue to prevail between Washington and Beijing, as they did between Washington and Moscow?

The linkages between nuclear capabilities and space assets haven’t ended with the Cold War. Can a formula for success built around inferential deterrence and reassurance, which has belied worst-case projections, be replicated?


  1. Allen Thomson (History)

    “…a mutual recognition that attacks on satellites invited uncontrolled escalation and horrific consequences.”

    Would your shoe boxes have any official(ish) statements to that effect? I heard the opinion expressed casually every now and then back in the day (1973 and after), but never in any circumstances that indicated that it was canon.

    • krepon (History)

      Thanks for checking in. Will take a look.

  2. iam he (History)

    No nation wants or needs Nuclear Weapons except to deter sociopathic belligerent governments from attacking.

    attacking what? that is 100% completely up to the nation being attacked.

    and yes that would include vital interests that are in space.

    Deterrence: inaction caused by the perception of danger.

    what is the minimum danger it would take to deter the US.

    this is the most important issue..

  3. iam he (History)

    Iran is a good example of inferred nuclear deterrence.

    she does not have a nuclear weapon… but she has the means and know how to make a nuclear weapon, and the possibility she may already have a secret nuclear weapon…

    thus she may not have a nuclear weapon yet she has a nuclear deterrent….. if only the means to make one.

  4. John Schilling (History)

    I don’t see how this subject can be meaningfully addressed without noting the fundamental moral difference between space warfare and nuclear warfare. Nuclear war inevitably results in the fiery, horrific deaths hundreds of thousands to hundreds of millions of people. In most plausible scenarios, that’s the whole point.

    Space warfare, at least in the near future, likely results in zero human casualties. Only the destruction of pieces of machinery that have been deployed in a realm not subject to national sovereignty. Indeed, as with “cyberwar”, it is unclear whether “war” is really the appropriate term. And any such conflict, whatever we call it, will be essentially invisible to the general public.

    Space deterrence, then, is going to be based on purely utilitarian calculus. What do we gain from deploying or using capability X, compared to what we might lose when an adversary sees that as license to use closely-related capabilty X’? And, for that matter, does the adversary’s own cost-benifit analysis mean that they are going to roll out X’ regardless?

    Nuclear deterrence invites a fundamentally different sort of moral calculus in which utilitarian cost-benefit analysis is seen as largely unthinkable by just about everyone with anything to say about it and certain outcomes Absolutely Must Be Avoided No Matter What.

    It is not surprising that the methodologies of deterrence are somewhat different at these opposed extremes.

    • krepon (History)

      Your basic assumption seems to be that nuclear warfare and space warfare are watertight compartments. I don’t share this assumption. Not when satellites are critical is so many ways for nuclear operations.

      In the skimpy literature on space deterrence and in discussions on this topic, one can read and hear about escalation ladders for space warfare. Start small, nothing to get totally upset about, just disrupt, reversible effects, then move on up the ladder to more critical satellites, etc.

      Sound familiar?

      My sense is that space warfare is as inherently dangerous, messy, hard to fathom, and as difficult to manipulate and control as Herman Kahn’s escalation ladder for nuclear warfare.


    • Cthippo (History)

      I’m with John on this one.

      I can’t see a nation responding to an attack on it’s space infrastructure with lethal, much nuclear, force. The closest terrestrial equivalent I can think of would be shooting down a drone.

      Despite the utility of satellites, they are fundamentally unmanned machines operating outside anyone’s sovereign territory. To respond to the destruction of such a device by killing actual people somewhere just seems like so over the top as to be unreasonable.

    • John Schilling (History)

      Satellites may be critical to nuclear operations, but so are e.g. SOSUS arrays. I don’t recall ever reading or hearing about how if the Soviets had dropped depth charges on some hydrophones on the Pacific seabed, we’d have wound up nuking Vladivostok.

      The difference between lethal and non-lethal force is as close to a watertight compartmentalization as we are going to get in this business. There is an escalation ladder in space warfare, as there is in any area of international affairs, but in this case the highest rung on that ladder is barely in the middle of Kahn’s more generalized ladder – and there is a conspicuously large gap over the middle ground that Khan populates with marginally lethal steps. And it’s the marginally lethal steps that are the most ambiguous and most prone to provoke emotional overreaction.

      Now that I think about it, the SOSUS -> nuke Vladivostok step is more plausible than any nuclear escalation of space warfare. Because the former case does have room for those ambiguous steps in the middle. A USN destroyer “nudges” whatever Russian ship is tearing up the hydrophones. They light up the US destroyer with fire-control radar. Someone lights off an antiship missile, and a few dozen sailors are dead in international waters. You started it. No, you did. But there are flag-draped coffins coming home, live on CNN.

      W/re space warfare, maybe some satellites are turned to scrap. Then either some more satellites are turned to scrap, or we’re launching strategic missiles at Baikonur and killing thousands of Russians (many of them civilians) on Russian soil. Nothing much in between, so a whole lot of rungs on Khan’s ladder are missing.

      The people who I have seen claim likely or even plausible escalation from space warfare to nuclear warfare, have not to my knowledge been speaking from a position of either authority or inside knowledge, and their speculation is at odds with my own understanding of the situation.

      As with Allen, I’d be very interested in any exceptions to that.

  5. Mark Gubrud (History)

    Space weaponization, as a review of the literature employing this term would show, is not limited to the placement of weapons in space; it also includes the development, testing, production and deployment of weapons that are prepared for “pop-up” placement in space and of weapons that are prepared for attacking objects in space, i.e. ASATs.

    Whether space weaponization “has yet to occur” is thus subject to debate, although certainly the worst, and what one would expect in the case of an unmoderated, technology and resource-limited space arms race, has not happened. There are a lot of things we could have done that we have wisely held off on.

    However, the US has deployed and plans to continue the expansion of BMD systems which are fully, operationally, and immediately ASAT-capable (if they work at all). I have heard claims from people close to official sources that China has also deployed some number of its ASAT/BMD interceptors, which are thought to be primarily intended for the ASAT mission, deep in its interior, protected from easy air attack. This is said to be mission #1 for proposed US prompt global strike weapons.

    I think the reason space weapons news attracts less attention than nuclear weapons news is just the fact that people see nuclear weapons as directly threatening themselves and everything they value, while space is somewhere out there and (for all the possible inconvenience and disruption) few tears would be shed over the loss of unmanned satellites.

    The message that seems to have been lost is the close relationship between nuclear and space security.

    John Schilling’s comments exemplify this; since he is an expert in his own right on all of this, I can only ascribe his position to denial of the received wisdom that space warfare is a likely stepping stone to nuclear catastrophe. But that wisdom is exactly why the US and USSR held off, and it is what has kept this issue smoldering.

    Here are some points I like to make about the relationship between space and nuclear weapons/war:

    * Space weapons are harder to make than nuclear weapons, as the historical timeline illustrates. As follows from this, all actual space weapons powers are actual nuclear weapons powers and all potential space weapons powers are potential nuclear weapons powers.

    Jihadis may be able to jam or hijack unprotected communications links but that’s about it, and they probably have better things to do. Even a threshold nuclear power like North Korea does not really have the ability to pose a credible space-strike threat. HAND, if they could do it, would not have easily predictable, militarily useful outcomes.

    * Space weapons threaten strategically important military assets (including early-warning satellites but just as importantly surveillance, communications, and navigation on which the US and increasingly China and other major powers depend for their ability to fight and to defend vital interests. Space warfare would be, and would be perceived as, strategic warfare.

    * Unexplained satellite malfunctions and ambiguous events in space are potential crisis destabilizers.

    * If space attacks are seen as less dangerous than nuclear weapons use, we have the usual paradox of something that can be perceived as a signal/warning or equally as a first move — a breach of the nuclear firebreak (if there is one).

    * Although a resource-limited space arms race has never broken out, a slow-motion space arms race (defined in terms of an action-reaction cycle, e.g. we develop/deploy BMD/ASAT, so China follows, we test maneuvering microsats with ASAT potential, so China follows) has progressed and remains poised to accelerate. Current efforts toward robotic on-orbit servicing/refueling/salvage and toward space debris removal are shortening the time available to establish the kind of sunshine, reporting and monitoring regime that is needed to prevent the simultaneous development and deployment of the same technologies, in different and distinguishable forms, as space weapons.

    Think we’ll be able to contain the “second nuclear age” and move toward global zero for nuclear weapons meanwhile a global space (and robot) arms race rages out of control? Think again.

    • j_kies (History)

      John speaking as an expert has informed opinions, you are free to disagree but your views are also subject to critical examinations given available data.

      I am not aware of actual ‘space warfare’ advocates as the kinetic negation of space assets guarantees the generation of debris fields that effectively preclude the use of that orbit or lower orbits. As the primary user of space for both civil and national purposes, the US has an asymmetric disadvantage in such a conflict. This issue is the primary reason that the US advocates for a responsible code of conduct for spacefaring nations to mitigate the generation of space debris. Any nonsense you read from ‘space control’ advocates should be ‘taken with a large grain of salt’, they are arguing for better space situational awareness to allow them to attempt to protect the assets from collision hazards.

      I believe the US views space war in the same light as 60 kg hand grenade with a 15 m lethal radius, the first people killed aren’t the enemy.

  6. Fred Miller (History)

    Once you’ve developed the automobile, is the tank inevitable? Once you’re building airplanes, can you prevent the next war including air combat?

    The militarization of space can be stopped, but only with a careful and robust effort.

    Can a war in space be limited to space? History isn’t reassuring. We tried to fight a war in Afghanistan, and it found it’s way to New York and Washington DC.

    Since 1945, we’ve managed to keep our wars from going nuclear, but we have not managed to keep them from escalating to far greater cost than their authors imagined possible. Cheney and Rumsfeld assured America that the Iraq war would last “weeks, not months”, and Britain went into the Falklands to protect a few thousand sheep and a modest fishery, but lost seven ships and hundreds of men.

    Our great grandparents were assured that dynamite would end all war, because it was so powerful. Later, TNT, poison gas, and aircraft were all predicted to end war.

    While nobody is predicting that spacecraft will end war, I am sceptical that space war will be limited to space. War will go where ever there are valuable targets, and people are valuable targets.

    A war that starts in space may not become a nuclear war, but that’s not very reassuring: conventional war is deadly enough. The Rwandan Genocide killed perhaps a million, mostly with machetes. If space war ignites a war on earth, the toll in human suffering will not be limited by lack of lethal devices.

  7. kme (History)

    The mutual vulnerability of space assets shares a lot with undersea communications cables, which are easy to cut and hard to defend. It’s somewhat comforting that we’ve managed to so far avoid a cable-cutting war.

  8. Mark Gubrud (History)

    It may just be that I am short of time, but I am going to make a point of not engaging this “Whose opinions are expert opinions” game. I think my comments above and elsewhere speak well enough to my own knowledge of the subject. I’ll respond to substantive points.

    So, I agree with j_kies about the US “asymmetric disadvantage” if space warfare including kinetic destruction gets started, but that does not seem to stop the US from leading the way in the development, testing, production and deployment of space weapons, while stubbornly refusing to engage any ideas for space arms control.

    In fact, one does not have to get very deep “inside” before learning that electronic space warfare is an operational capability of the US military and that it is (reportedly and probably, to some extent) actually in use. And that the possibility of kinetic attacks on e.g. Chinese satellites, especially those that would be used to target US Naval vessels for air or missile attack, is discussed as an option in certain contingencies.

    Also, “kinetic” means anything that moves, not the same as kill by “kinetic energy” or hypervelocity collision, which always produces large amounts of debris. So “kinetic attack” can include coorbital ASATs (e.g. “maneuvering microsats”) with non-debris creating kill mechanisms, another technology thrust brilliantly pioneered by the US under GW Bush and now (surprise, surprise) being copied by China.

    So, j_kies is right, the policy that the US has been pursuing since at least the mid-1990s (with deeper roots) is indeed very, very stupid, short-sighted and self-defeating.

    • j_kies (History)

      Unless we actually have a secret government that doesn’t bother to talk to the rest of us, the US is not pursuing an offensive ‘space war’ policy. Certain development programs were examined (F-15 ASAT, USA ASAT) and those were quashed appropriately as being counter to inherent US interests prior to ‘real acquisition’ or pursuit as a weapon system. While BMD systems have inherent capabilities, (and hazards) we are diligent in forcing the kinematics of the impacts to be on the ‘downward leg’ so as to preclude inserting debris into even temporary orbits. Significant efforts are spent to constrain test actions to preclude accidental ‘painting’ of objects with radars and lasers (the DOD clearing house job jar) as we don’t desire to send inappropriate / unintended messages to the owners of those objects.

      If you have information as to nefarious ‘space war’ capabilities in development or test – please share so we can diligently quash those as well.

  9. krepon (History)

    I suspect that quotes from US officials linking space warfare to nuclear warfare can be found by dedicated researchers, but they are not in my shoeboxes. The closest I can come is a statement by Robert MacFarlane, then the Assistant to President Reagan for National Security, dated August 1, 1984:

    “We have expressed the view that the problem of weapons in space cannot be considered in isolation from the overall strategic relationship.”

    The context for this remark was US-Soviet negotiations, not warfare, but this connection hard to de-link in both domains.

    As for the impending arms race in space due to BMD interceptors, I would note that BMD interceptors have been around for a very long time, and we have so far avoided an arms race in space. This is no cause for complacency, because BMD interceptors are more capable now than in the past, and because they have gained prominence by being tested, most dramatically, in an ASAT mode in 2007 and 2008. Using BMD interceptors in this way to engage in space warfare would be ruinous to operations of all kinds in low earth orbit, as j_kies rightly notes. Self-defeating acts of this kind tend to be self-deterring, as well. Again, no cause for complacency, and more reason, in my view, to push for realistic diplomatic initiatives protective of satellites, like the International Code of Conduct for responsible space-faring nations.

    As for escalation control and the barriers between warfare in space and nuclear warfare, Allen’s prompt for me to go back to my shoeboxes has led to the rediscovery of a wonderful quote by Kurt Gottfried and Richard Ned Lebow. It’s from their article in Daedalus, which devoted two volumes to space warfare-related topics in 1985:

    “ASATs possess a considerably greater capacity for transforming a crisis into a war, and for enlarging wars, than they do for assisting in military missions or enhancing deterrence.”

    Arguments to the contrary, i.e., that ASAT operations are entirely separable from nuclear dangers, undermine deterrence in both domains, even if they are true. In order to be true, the compartmentalization of space warfare from the potential for nuclear weapons’ use in confrontations between major powers, two conditions (at least) must be met: (1) space warfare is “limited” and does not seek decisive, across-the-board advantage; and (2) space warfare can be a tidy business, unlike warfare in any other domain. Both of these assumptions/conditions strike me as being strange.


  10. Jonah Speaks (History)

    “Will cooler heads continue to prevail between Washington and Beijing, as they did between Washington and Moscow?”

    This news link says maybe yes:
    ‘Landmark’ Space Policy Shift As China, Others Agree To Space Code of Conduct Talks, By Colin Clark on July 23, 2013 at 12:48 PM

  11. Andrew Tubbiolo (History)

    I think a lot of this calculus depends on what’s being attacked, who’s attacking, and who’s being attacked. In 2003, I think the US could have been deterred from invading Iraq if the US armed forces could have been convinced that the Iraqi conventional forces could have extracted the cost of an armored division from the US military. Two more real world examples are Libya and Syria. Libya was borderline go, and with Syria we get to see the US deterred on a day to day basis.

    As for space, I think it also matters. If GPS were turned off or negated, it would be like the USAF and the USN losing a vast portion of their air arms. Effective delivery of munitions would revert to mid 1990’s levels. The current batch of battle staff might react to that as if they lost the physical aircraft. Esp during a real fighting war. Likewise if communications relays were taken out, you might lose the ability to operate drones and be forced to employ manned aircraft. There again the US military might view the loss of the satellite as the loss of the system the satellite controls or enhances.

    Not to mention what happens to our ability to use our SLBMs in a first strike role against ground targets during a nuclear crisis? Would loss of GPS, or other space based systems put pressure on the US during a crisis to employ it’s SLBM’s in a counterforce role before it loses the ability to do so?

    • j_kies (History)

      If you examine the dates when the current US nuclear deterrent forces went operational you might come to the appropriate conclusion that those forces were judged to be fully effective without access to space assets for guidance or other issues. Substantially all elements of that deterrent are as insensitive to losses of surveillance and communications capabilities as the designers could feasibly make them. While the deterrence isn’t likely to be perfectly decoupled from things like space assets, the designers certainly tried. Its a deterrent, its not a warfighting tool.

  12. Bradley Laing (History)

    —-New conventional bomber? or new nuclear capable bomber?


    In other news, Kehler said he “personally” believes the US needs to replace the SR-71 Blackbird — a system capable of penetrating hostile airspace and performing intelligence, surveillance and reconnaissance (ISR) — with something. He hinted that replacement might be associated with the Long Range Bomber program (also known as Long Range Strike). But the basic problem is “we can’t yet make final decisions” because of the current budget mess. Now that’s something that must have the Chinese smiling: America can’t afford to build a system needed for anti-access/area denial operations.

    • Cthippo (History)

      One and the same.

      Think B-52 / B-1B replacement, or something like that anyway. Strategic range bomber capable of carrying either nuclear or conventional payloads including heavy cruise missiles. Air Force has wanted one for years, at least since they only got a handful of B-2s.

      Which brings up an interesting question…

      Seems like the B-2 would make a really good ISR platform, assuming that it really is as stealthy as claimed. I have never heard of this potential capability being mentioned, but it’s a logical role.

    • Allen Thomson (History)

      > Seems like the B-2 would make a really good ISR platform, assuming that it really is as stealthy as claimed. I have never heard of this potential capability being mentioned, but it’s a logical role.

      As I understand it, the original mission concept for the B-2 had it using its integral ISR capabilities, in particular its stealthy synthetic aperture radar, to fly stealthily around over the USSR looking for mobile ICBMs to nuke. Like most nuclear war-fighting ideas, I’m glad we didn’t find out how well that would have worked.

  13. jeannick (History)

    Problem , any action in near Earth orbit might result in an avalanche of debris junk ,
    any identification is going to go ape-sh.t
    not the best situation to make cool reasoned decisions

  14. M.V. "Coyote" Smith (History)

    As a former Air Force Missile Launch Officer and commander of an ICBM squadron, I need to reassure everyone that satellites are nice to have, but not required. As pointed out above, our nuclear triad was created long before satellites began contributing to military operations. In some cases, some satellites speed-up attack warning or command and control communications, but only by a matter of a few seconds or minutes compared to other non-space systems. Redundancy upon redundancy was built into every aspect of our nuclear operations from start to finish–including attack detection and command and control. It is a system-of-systems in which satellites make an important contribution, but we are far from dependent upon them. Even if we suffered a complete take down of our space systems our nuclear forces would be comparatively unaffected. Snow storms in the Midwest cause more disruption to nuclear operations than losing satellites. That also illustrates the beauty of the redundancy of the triad; subs don’t mind the snow like ICBMs and bombers.

    The bottom line is that taking out all of our satellites will not force a president’s hand to advance our nuclear posture or calculus. A president may decide to do that, but that is not driven by the loss of satellites in relation to nuclear forces. We have done all we can to ensure that a president always has options other than nuclear. A president’s hands are not tied to an nuclear escalation ladder.