Nuclear Deterrence and Four Types of Force: Definitive, Coercive, Catalytic, and Expressive

Journal of Political Risk, Vol. 5, No. 6, June 2017

Victory Day military parade, North Korea, 2013. Source: Stefan Krasowski via Wikimedia Commons.

Captain Robert C. Rubel
USN (Ret)

North Korea’s drive to attain a nuclear warfare capability is currently on the front burner in the Pentagon, and is a driver of tensions in East Asia. This has precipitated plenty of dialogue in the national security community, including on the issue of extended deterrence, the policy of the US that threatens nuclear response if an ally is attacked with nuclear weapons. One input from a former colleague at the Naval War College was the final catalyst that got me tapping on the keys. First, he quoted one of his scholars as saying that the real question concerning nuclear weapons “… is whether strategic nuclear forces have any genuine relevance today in the context of deterrence and warfighting, or whether they’re troublesome legacy weapons of a bygone era.” In a subsequent email he said that he was “interested in deterrence stopping all wars, not just nuclear.” It may be that the general umbrella of nuclear deterrence did suppress some wars that might have otherwise taken place during the decades after 1945, but it is almost impossible to know. However, my colleague’s faith in the utility of the manifold uses of deterrence is not that distant from those who advocate tailored deterrence, which is a scaled or graduated deterrence structure that includes the option of preemptive strikes.[1] Tailored deterrence to some extent reflects the logic behind the DoD concept of flexible deterrent options (FDO), which are defined as “…a wide range of interrelated responses that begin with deterrent oriented actions carefully tailored to produce a desired effect.”[2] In my view, such policies would incur considerable risk, as they ascribe, in an a priori manner, effects on an opponent’s political decision making and strategic planning processes in lieu of any specific intelligence (frequently) and certainly without any historical track record, especially in the nuclear arena. In this short article I will discuss a different way to analyze deterrence and gain insight into the thought processes of my colleague.

I am not a nuclear deterrence scholar, but I spent a number of years at the bottom of the nuclear food chain; a Navy single seat attack pilot who had a nuclear delivery mission if so ordered. I was also the squadron nuclear weapons training officer which meant I got involved with such things as weaponeering, calculating the physical effects of particular weapons, used in certain ways, against specific targets. From this perspective I would first point out that any discussion of nuclear warfare and attendant concepts of deterrence is theoretical and speculative because the weapons have not actually been used since 1945. I could go through the calculated physical effects of a blast of a particular yield weapon delivered in a certain mode, but that would not tell you much about the political effects of the attack, including those attendant to the threat of its use in peacetime or crisis.

There is an underlying logic to warfare that was revealed to us by the Prussian theorist Carl von Clausewitz. At heart it is a duel aimed at getting the enemy to do our will, or failing that, to eradicate him. Clausewitz coins a number of concepts such as progressive interaction leading to extremes of effort, the factors that impede such extremes, which he calls friction, the culminating point of attack and its political cousin, the culminating point of victory. A key idea of his that bears on deterrence theory is that the results of potential battles should be regarded as real, as the estimate of these possible results can affect military decision making. All of these can be seen working in all the wars since he wrote in the early Nineteenth Century. Once, however, the United States and the Soviet Union amassed huge inventories of nuclear weapons (an example of extremes), the prospect of war in the context of mutually assured destruction (MAD) was stripped of the logic Clausewitz unveiled. Even given the lack of mutual understanding of motives, such a situation likely produced overall deterrence. At levels of inventory less than what the Cold War produced, the logic of war is still operative. Thus, in the early nuclear age President Eisenhower could adopt his “new look,” based on the policy of massive retaliation for any Soviet incursion into NATO. Whenever the inherent logic of warfare is available and operative, deterrence becomes a matter of probabilities, just as in actual warfare, as Clausewitz so deftly points out.[3]

In the email string there were some comments by others about the substitution of precision guided conventional weapons for nuclear weapons, a thread of logic that permeated the 1980s Soviet literature on the revolution in military affairs.[4] There are certain operational situations in which precision conventional weapons could be substituted for nuclear weapons, from a weaponeering perspective, but precision, including cyber, is qualitatively different from what happens when a nuclear weapon explodes. The original idea for substituting precision guided weapons (PGMs) for nuclear weapons was part and parcel of the Army’s development of AirLand Battle in the wake of Vietnam. The context was the massive Red Army presence east of the Inter-German Border and the huge stockpiles of nuclear weapons on both sides that created the condition of mutually assured destruction (MAD). Both sides realized that all-out nuclear warfare was essentially mutual suicide, so some conventional means had to be found for deterrence and warfighting in the NATO Central Region. The context and the logic of that era no longer exists today, so trying to extend the substitution logic to various scenarios today is dangerous in lieu of some very detailed analysis.

However, for all their power, nuclear weapons have limitations. I spent several decades in the wargaming business as a faculty member at the Naval War College. In some games, my students would enter a scenario thinking that nuclear weapons were a final solution to a particular military problem; at a certain point of frustration they wanted to “nuke ’em” when conventional ops were stymied or especially when Red used a nuke. However, when I asked them to give me a targeting strategy, either counter-force or counter-value, they often could not produce a viable one for the situation in question. Part of the problem was often residual effects on nearby friendly nations.

In my military mind there is a way to at least partly untangle the deterrence issue, respecting always that looking back, there is no way to prove that someone did not take an action based on something we did or had, so the estimation of deterrent effects is necessarily speculative. I start with the notion that human conflict logic operates the same with nuclear weapons as it does with any other kind of violence. Thus we can utilize the basic logic of the use of force to help us speculate more productively. As both a fleet strike planner and a war college planning and decision making instructor I felt compelled to look beyond simple destruction of targets to the uses of such destruction. I came to the conclusion that there are four and only four ways that force can be used, and I believe this framework scales from the tactical level to the strategic.

The first way to use force is definitive (I am in debt to James Cable and his book Gunboat Diplomacy)[5], in which force directly settles the issue. The Israeli strike on the Osirak reactor is an example. No enemy decision making is required, the only issue being the ability to overcome enemy defenses. A comprehensive first strike counter-force strategy for nuclear weapons might fit into this category if it was indeed able to destroy all the enemy’s nuclear weapons.

The second way is coercive, upping the pain or threat level until the enemy capitulates. This is a common strategy when, for whatever reason, definitive use options are not available. Nuclear force in general might fit into this category if indeed the threat of such use causes the enemy to not do something he otherwise would have done.

The third way is catalytic, using force to precipitate second or third order events. Saddam Hussein’s use of Scud missiles on Israel used this logic.  The apparent idea was for the Scud attacks to precipitate Israeli involvement in the war.  This would then force Arab nations out of the coalition against him.

The final way is expressive, in which there is no specific objective or defeat mechanism versus the enemy; the motive being simply a desire to harm or perhaps to achieve domestic purposes such as to mobilize popular support or at least demonstrate will. Reprisals fall into this category. The 1983 Lebanon air strike was an example.  Public pronouncements from the Reagan Administration notwithstanding, there was little to no prospect of the strikes being definitive, coercive or even catalytic.  The idea was to “send a message.”

Any use of force in any context will fall into one of these categories, even cyber. The only possible exception might have been an all-out nuclear exchange between the Soviet Union and the US, which, as an exercise in mutual suicide, would have been devoid of any logic.

The reason for establishing these categories is that they allow one to clinically analyze the prospects that a particular use of force would have for obtaining desired results. Any use of force tends to have political side effects that are hard to predict in advance. Too often, and I might offer the lead up to the 2003 invasion of Iraq as an example, decision making about whether to use force – go to war – is attended by unwarranted optimism, if not mystical thinking, propelled by the claims of advocates of new technologies or warfare concepts. In the case of Iraq, the concepts of network-centric warfare, rapid decisive operations and effects based operations (producing shock and awe) instilled a distorted view among key Bush Administration officials of how the use of force would work. In earlier eras the claims of air power advocates spawned counter-productive decision making. My framework is meant as an antidote to those intellectual pathologies. I am not arguing for pacifism; I am just trying to help avoid intellectual errors in operational and strategic planning.

Definitive force is the most reliable method if it is available, but actual opportunities are rare. However on a large scale, even if one overruns the enemy country, we have found that it doesn’t necessarily end things. Clausewitz articulates the concept of the culminating point of victory, the idea that a victory, no matter how complete, has to be subsequently defended in some way.[6] An extension of that idea is that how a war is conducted can affect not only whether victory is possible, but what it would take to defend it. The surprise nature of the Japanese attack on Pearl Harbor could be said to have taken them past their culminating point of victory right then and there as it galvanized the US population to conduct all-out war. The common error is to ascribe definitive effects to a strategy when they are not actually available, or to ascribe definitive effects to one of the other modes. This is only to say that even if a definitive opportunity seems to present itself, side effects and blow backs could occur if the operation is not properly deliberated. Any use of nuclear weapons, regardless of their definitive military effects, are likely to incur these. Finally, if a nation perceives the opportunity to use definitive force, it will be tempted to do so. This seems to be part of the Japanese decision to strike at the US Fleet, much of which was at anchor at Pearl Harbor. Similarly, the decision to invade Iraq to capture weapons of mass destruction, if that was indeed the real reason, could be regarded as an attempt to seize an opportunity to use definitive force. In both cases, the unanticipated side effects produced long and ultimately unsuccessful wars. The danger, to reiterate, is that perception of definitive opportunities could precipitate a decision to conduct preventative war.

The problem with coercive force is predicting what level of pain or coercion would be enough to get the enemy to yield. Mostly we overestimate the coercive effects of our proposed operations and underestimate the enemy’s ability to tolerate threat or pain. When such miscalculation occurs, wars stretch on beyond anyone’s predictions. The difficulty in planning lies in attempting to calculate in advance what amount of coercion is enough. Complicating such attempts are the almost inevitable political side effects of using force as described above. Moreover, even the targets of coercion most likely do not know how they will react. Another trap is ascribing definitive effects to what are necessarily coercive means. A military service or perhaps senior government official is convinced that a particular use of force will coerce the enemy government, but couches justification in definitive terms. Deterrence, at heart, is coercive. It is meant to stop somebody from doing something they would otherwise have done if the threat was not present. Would the Soviets have invaded Western Europe if the US had not posed a nuclear threat? Maybe, maybe not. In any case, both sides built up their nuclear arsenals to the point where their use promised mutual suicide. At that point a reasonable person could say that deterrence on both sides occurred. In today’s world, that logic does not exist, and the basis for deterrence is wrapped up in the intellectual difficulties associated with the coercive use of force.

Catalytic force is alluring for its potential to produce large effects from small causes. However, especially at the strategic level, the degrees of freedom of the problem easily confound calculations. At the lower levels of war catalytic effects may be relatively easy to calculate. At the upper operational level and the strategic level, however, issues can be so complex that it may be difficult or impossible to calculate the linkages between a military cause and a hoped-for second order effect. During the strategic bombing of Germany, Allied targeteers hit on the idea that bombing their housing would reduce the morale of the German work force and thus result in reduced productivity in the German defense industry, notably aircraft building. However, extensive destruction of housing did not have the projected effect on production because the targeteers failed to account for the excess capacity inherent in housing; “de-housed” workers simply stayed with friends and relatives. The moral is that dazzling visions of achieving strategic leverage through the catalytic use of force is a dangerous undertaking unless the potential cause-effect linkages are known and understood. If a strategy is based on catalytic effects, the nation may quickly find itself with a bankrupt strategy, and facing a drawn out war of attrition. Catalytic strategies can backfire and have virtually the opposite effect than was originally envisioned. Schemes for applying precision munitions to achieve strategic aims quickly and cheaply often depend on catalytic effects, whether they clearly state this or not. There is no evident and necessary linkage between destruction and political control, and a failed catalytic scheme may end up causing substantial political damage to its perpetrator. At a minimum, it argues for caution in any scheme to substitute conventional weapons for nuclear.

Finally, expressive use of force is the most dangerous for the user as it can blow back in any number of unanticipated ways, and is often ascribed to have effects of one of the other modes under the rubric “this will fix’em.” The defining characteristic of the expressive use of force is the lack of a clear military objective. Even some terrorist acts have specific political or military goals, and can be otherwise categorized. Expressive force is used to vent anger, present a non-specific threat or to merely harm the enemy. Reprisals fit into this category many times, if their use is reflexive and not clearly directed. Often, war crimes may be forms of the expressive use of force – the manifestation of anger or hatred. This is not to say that the expressive use of force is always wrong; it may end up having salutary effects. It is, however, frequently the refuge of those who have no other means or ideas, and are hoping for the best. The danger in this is that the expressive use of force can generate reactions that are out of proportion to its original strength and intent. Those who advocate “sending messages” lack a true understanding of the use of force. The simple old axiom used by many gun owners seems appropriate here: don’t point your gun at someone unless you intend to use it. “Using it” in the strategic sense means having a plan and capability to use force in a definitive, coercive or catalytic manner. In the nuclear realm, setting off one nuclear weapon to “send a message” might very well constitute an expressive use of force, with all the hazards involved.

Nuclear weapons do not escape from this framework, and any counter-force or counter-value targeting strategy must fall into one of the categories. Recalling Clausewitz’s assertion that the prospective results of potential battles must be regarded the same as actual results,[7] we must admit that the framework applies as much to the matter of deterrence as any actual use of force. If the errors just described that are commonly attendant to the actual use of force are at all present in devising deterrent policy and strategy, in addition to the historical problem of proving why a nation did not do something, strategic risk soars. This is where the notions of tailored deterrence and flexible deterrent options produce risk; they require the opponent to conduct a calculation of the outcome of a possible battle using the same frame of reference as our planners. This does not seem very likely.

Whatever the other elements of deterrence theory, common sense dictates that a secure second strike capability is a valid element of any approach to nuclear strategy. As a strategy to deny an opponent the possibility of the definitive use of force to disarm in a first strike, it takes a big part of war logic off the table. The desire for such a strategy seems to me to be at least part of what is behind China’s actions in the South China Sea. A few years back I talked with a couple of retired People’s Liberation Army Navy (PLAN) officers who intimated that they wanted the South China Sea to be a bastion for their nuclear-powered ballistic missile submarines (SSBNs). Since their ships are not quiet, they need an array of support to be survivable, which includes air bases on islands. Thus, our ballistic missile defense (BMD) initiatives and quiet attack subs have frustrated both Russia and China. Both, I think, have resorted to mobile rail and highway launchers as a substitute, but both would probably prefer a true Trident-like submarine capability.

However, the Chinese seem to have other uses in mind for a secure nuclear capability. One PLAN flag officer with whom I spoke said China just wanted to be able to get one nuclear weapon through our defenses so that they had some leverage against US interference in their local matters. That is a first strike, coercive, counter-value strategy that is akin to what North Korea, as a nation possessing a few nuclear weapons would likely adopt. As a coercive strategy conducted below the absolute level of the US-Soviet standoff, it is vulnerable to the miscalculations and problems described above. Among the consequences of failure is falling into a “put up or shut up” situation. This would make nuclear use more likely, depending on the state of mind and political situation of North Korea or Chinese leadership at the time. Another adverse side effect from their perspective is that it has generated a kind of security dilemma for them as it in part stimulated our BMD efforts.

Beyond the framework I have just discussed, deterrence, nuclear or otherwise, is, in a sense, the resort of the weaker party to a dispute. I had an article about this in Orbis in November 2012. There is no room to go into that logic here, but suffice it to say that if, as I assert, deterrence is the resort of the weaker party to a dispute, nuclear weapons would easily be seen, like the Colt 45 in the old west, as the equalizer. There are any number of nations that see themselves as the weaker party, so, as Paul Bracken has argued in his book The Second Nuclear Age, nuclear weapons will not become obsolete anytime soon. This seems to put to rest the question of whether nuclear weapons are a legacy of a bygone age, but they are indeed troublesome, especially if one accepts my framework of the uses of force and the dangers and risks attending each.

Getting back to my colleague’s desire for deterrence to be an implement to end wars, my one overriding strategic principle for nuclear weapons or anything else is Don’t Shoot First; nothing good comes of it. One could make the case that both the Russians and Chinese are adhering to it, adopting “gray zone” tactics to achieve their goals. “Gray zone conflict is best understood as activity that is coercive and aggressive in nature, but that is deliberately designed to remain below the threshold of conventional military conflict and open interstate war.”[8] Whether or not gray zone operations can be deterred is at the moment an open question, although one group of researchers have put together a set of recommendations for doing so.[9] Their recommendations are similarly pitched below the level of open conflict, so I will have to devote more thinking as to whether the use of force framework is applicable to the analysis and development of deterrent strategies in that arena.

Getting back to North Korea, where this article started, we can see that as a nation potentially possessing only a limited number of nuclear weapons, it must craft its policies in the context of Clausewitz’s military logic and is vulnerable to the intellectual errors I have discussed.  Its recent track record is to use ballistic missile launches to “send a message” to the United States, Japan, and South Korea.  It is hard to come up with a realistic coercive or catalytic objective for these launches, which gives us some insight into the thought processes of the regime.  The danger is that such an approach would be adopted in case conventional war broke out, making their use of one or more nuclear weapons more likely.  As discussed previously, a deterrent policy of using one or a few nuclear weapons to cover for conventional weakness is inherently a first strike (that is, first use of nuclear weapons during a conventional war) strategy that would have to be based on one of the four uses of force, with all the pitfalls that have been identified.  I doubt seriously that North Korean planners are any more aware of this use of force framework than are U.S. and allied planners.

The bottom line to all of this is that in my view deterrence is not what many folks think it is. I am an adherent of the Roman Vegetius’ adage that if you want peace, prepare for war. That tends to generally take away from potential aggressors the perception of low hanging fruit, but beyond that, things get complicated. Keeping in mind the framework for the uses of force does not completely defog our windshield, but at least we can see some obstacles and hopefully the stop signs.

Captain Robert C. Rubel USN (Ret) served thirty years active duty in the US Navy. In the first twenty years he served as a light attack/strike fighter pilot. His last ten were mostly spent on the faculty of the US Naval War College, teaching planning and decision making, and wargaming. After retirement from active duty spent thirteen years as a civilian faculty member at the Naval War College, serving as Chairman of the Wargaming Department and later as Dean of the Center for Naval Warfare Studies. He currently serves as an advisor to the Chief of Naval Operations on fleet design and fleet architecture. JPR Status: Opinion.

[1] Air Force Studies Board of the National Research Council, U.S, Air Force Strategic Deterrence Analytic Capabilities, (Washington, DC: National Academies Press, 2014), Appendices D and E, pp. 118-155. See also, Michael Raska, “Tailored Deterrence: Influencing North Korean Decisionmaking,” The Diplomat, April 26, 2016,

[2] Department of Defense Dictionary of Military and Associated Terms, Joint Publication 1-02, May 2017, p. 92.

[3] There are too many ideas discussed here for individual citations, but I use the 1976 translation of Clausewitz’s opus magnum On War by Michael Howard and Peter Paret cited below.

[4] Mary C. FitzGerald, Marshall Ogarkov and the New Revolution in Soviet Military Affairs, Center for Naval Analyses Research Memorandum, 14 January 1987.

[5] James Cable, Gunboat Diplomacy 1919-1979, (New York, St. Martin’s Press, 1981). Cable’s categories for the use of force are oriented on limited use of naval force in peacetime. I modified them a bit to accommodate the use of force in actual war.

[6] Carl von Clausewitz, On War, edited and translated by Michael Howard and Peter Paret, (Princeton, NJ: The Princeton University Press, 1976), Book Seven, Chapter Twenty Two, pp. 566-573.

[7] Ibid., Book Three, Chapter One, p.181.

[8] Hal Brands, Paradoxes of the Gray Zone, February 5, 2016, (Philadelphia: Foreign Policy Research Institute),

[9] Michael Green, Kathleen Hicks, et. al., Countering Coercion in Maritime Asia: The Theory and Practice of Gray Zone Deterrence, (Washington, DC: Center for Strategic and International Studies, May 2017), pp. 278-284.