F O R U M O N P H Y S I C S & S O C I E T Y
of The American Physical Society 
October 2007
Vol. 36, No. 4

ARTICLE

What Are Nuclear Weapons For?
John S. Foster, Jr.
Keith B. Payne

Recently there has been a round of initiatives advocating the long-term elimination of nuclear weapons and/or their near-term reduction to small numbers.[1]  With the end of the Cold War, many thoughtful people understandably ask why the United States should continue to maintain nuclear weapons.  What role do they serve?  Couldn't we defeat most plausible adversaries with our conventional forces alone?

A problem with these recent initiatives is that they tend to emphasize the risks of maintaining U.S. nuclear capabilities, while generally ignoring or dismissing superficially the risks of giving up those capabilities.  The postulated benefit from giving up U.S. nuclear capabilities typically is presented in terms of the contribution such a move supposedly would make to global nuclear non-proliferation.  When the continuing possible need and requirements for deterrence are broached, it typically is to assert that most or all deterrence needs can be satisfied by U.S. non-nuclear forces.  For example, “…the United States , now the world’s unchallenged conventional military power, can address almost all its military objective by non-nuclear means.  The only valid residual mission of U.S. nuclear weapons today is thus to deter others from using nuclear weapons.” [2]  This now common assertion is a logical non-sequitur: whether or not U.S. conventional weapons can satisfy most U.S. military objectives tells us precious little about what may be necessary to satisfy U.S. deterrence objectives.

As a matter of fact, the on-going development and deployment of new nuclear weapons in Russia and China and the spread of mass destruction weapons to rogue states make effective deterrence as important now as it was during the Cold War, and nuclear weapons are likely to continue to be critical to effective deterrence.  And, while superficially counterintuitive, the net effect of U.S. nuclear capabilities almost certainly is a positive and essential contribution to nuclear non-proliferation.  The following provides a brief elaboration of four reasons why nuclear weapons remain critical to U.S. and allied security.   

To address the question “What are nuclear weapons for?” requires that we examine the multiple roles served by nuclear weapons.  We need to look beyond the military characteristics of U.S. nuclear weapons and address the broader spectrum of national defense goals that they serve.  These goals - deterrence, assurance, and dissuasion - reflect our long-standing core objectives of protecting the United States and allies, working to limit the proliferation of nuclear weapons and other weapons of mass destruction, and steering potential adversaries away from military challenges and competition.

There should be no desire to rely on nuclear weapons per se; precision conventional weapons and defensive capabilities may rightly assume a relatively greater role, as was emphasized in the 2001 Nuclear Posture Review.  There is, however, a continuing need for nuclear weapons to support these overarching U.S. defense goals of deterrence, assurance, and dissuasion.  None of these roles for nuclear weapons follows from a “war-fighting” policy orientation, or presumes the actual military employment of nuclear weapons, or entails a requirement to do so.  The value of nuclear weapons for these traditional core goals of deterrence, assurance and dissuasion resides in their continued role as a withheld threat.  Identifying these roles for nuclear weapons in the new strategic environment was a focus of the 2001 Nuclear Posture Review (NPR).  

Deterrence:   The value of effective deterrence did not end with the Cold War; it remains essential to national security, and nuclear weapons remain essential to effective deterrence.  By helping to prevent war and the need to use force, nuclear deterrence does not represent a disdainful “trap” as some commentators have claimed.  Nuclear weapons are an enormously valuable tool of deterrence in the contemporary strategic context and should be given up only after long and careful consideration.  As Winston Churchill observed, “Be careful above all things not to let go of the atomic weapon until you are sure and more than sure that other means of preserving peace are in your hands!”[3] 

Strategic nuclear weapons that can threaten an adversary’s valued targets from afar are, and are likely to remain, essential for holding particularly well-protected targets at risk for deterrence purposes.  These targets are, for all practical purposes, invulnerable to non-nuclear threats and are likely to remain so for the foreseeable future.  For example, during the 1991 Gulf War many hardened Iraqi facilities were destroyed but some bunkers were, “virtually invulnerable to conventional weapons.”[4]  Similarly, according to statements by Clinton Administration senior officials in 1996, the Libyan chemical weapons plant located inside a mountain near Tarhunah could be threatened with destruction only by nuclear weapons.[5]  

The potential importance to effective deterrence of the U.S. capability to hold these types of targets at risk from afar is suggested by the attention and resources some adversaries devote to protecting and shielding them.  Adversaries unsurprisingly seek to protect what they value.  And, as Dr. Harold Brown, Secretary of Defense during the Carter Administration, emphasized when in office, U.S. deterrence threats should be capable of holding at risk those assets particularly valued by the adversary.  In some important cases U.S non-nuclear threats cannot do so and can promise little deterrent effect. 

In addition, there is no doubt that some opponents who were not deterrable via U.S. non-nuclear threats were in fact deterred by what they interpreted to be nuclear threats.  This deterrent effect is a matter of adversary perceptions, not our preferences: Whatever we believe about the lethality of U.S. non-nuclear weapons and what should be their deterrent effect, and whatever our hopes might be about how adversaries should think and behave, the actual behavior of past adversaries, including Khrushchev, Mao, and Saddam Hussein, has shown beyond doubt that there can be a profound difference between the deterring effects of nuclear and non-nuclear weapons.  In some cases, given the adversary’s views and the context, only nuclear deterrence works.  To assert that nuclear weapons now are unimportant is to suggest either that deterrence is no longer important, or that the future will be much more benign than the past,  and that we will not again confront such opponents armed with dangerous weapons.  There is every reason to reject both propositions.  U.S. policy with regard to nuclear weapons should not be based on optimistic hopes that so contrast with the actual past behavior of foes.  Given past experience, the burden of proof is on those who now contend that nuclear deterrence no longer is necessary to preserve the peace.      

The question is not whether we “want” to rely on nuclear weapons for deterrence.  It is whether we are willing to accept the risk of deterrence failure that would be introduced by our inability to threaten some adversaries’ highly-valued targets that may be essentially impervious to non-nuclear weapons and/or our inability to threaten nuclear escalation in response to a severe provocation.  The risk of deterrence failure flowing from such inabilities can not be calculated with precision.  Because multiple contemporary opponents possess nuclear and/or biological weapons, the consequences of deterrence failure could be measured in thousands to millions of U.S. and/or allied casualties.  The risk of deterrence failure following from U.S. abandonment of nuclear capabilities may be low or high depending on the opponent and context.  But even low-probability events deserve serious consideration if they have potentially severe consequences.  The move to reliance on non-nuclear weapons to hold enemy targets at risk would carry the increased risk of deterrence failure, and the probability may not be low. 

Assurance:   Nuclear weapons are essential to the U.S. extended deterrent.  This “nuclear umbrella” is central to the basic U.S. defense goal of assurance.  This is not a trivial goal.  The assurance provided to allies by U.S. security commitments, particularly including the U.S. nuclear umbrella, is key to the maintenance of U.S. alliance structures globally.  It is part of the basic security considerations of countries such as Japan , South Korea and Turkey .  The continuing role of U.S. nuclear weapons for this purpose may not be the preference of those in the United States who would prefer that the U.S. umbrella be non-nuclear.  But what does or does not assure allies is not decided by U.S. commentators or U.S. political preferences, but by the allies themselves.  The United States can decide if the assurance of allies is a worthy continuing goal, but only our allies can decide whether they are sufficiently assured.  In this regard, available evidence points strongly to the fact that nuclear weapons remain critical to the assurance of key allies.  For example, the recent responses by Japan and South Korea to the North Korean nuclear test of October 9, 2006 demonstrated explicitly that U.S. nuclear weapons are viewed by allies as critical to their confidence in the U.S. extended deterrent.  The discomfort felt by allies and friends in the Middle East given the prospect of Iranian nuclear weapons points in the same direction.

We could decide that we would prefer to withdraw the nuclear umbrella and provide non-nuclear extended deterrence.  But, with the nuclear proliferation of North Korea and the apparent Iranian aspirations for nuclear weapons, and the rapid growth of China’s nuclear arsenal, the response of key allies to the U.S. withdrawal of its nuclear extended deterrent coverage would create new and potentially severe problems, i.e., nuclear proliferation by U.S. friends and allies who would likely feel too vulnerable in the absence of U.S. extended nuclear deterrence.  Japanese leaders have been explicit about the extreme security value they attach to the U.S. nuclear umbrella, and they have suggested that Japan would be forced to reconsider its non-nuclear status in the absence of the U.S. extended nuclear deterrent.  Thus, ironically, nuclear non-proliferation is tied closely to the U.S. preservation of its extended nuclear deterrent.  This point is contrary to the typical contention that U.S. movement toward nuclear disarmament promotes nuclear non-proliferation.  Precisely the reverse linkage may be more the reality:  U.S. movement toward nuclear disarmament will unleash what some have called a “cascade” of nuclear proliferation among those countries which otherwise have felt themselves secure under the U.S. extended nuclear deterrent and therefore have chosen to remain non-nuclear.  We should be extremely careful before moving in a direction that carries the risk of unleashing this “cascade,” such as deciding that U.S. nuclear weapons are unnecessary for assurance and moving toward a non-nuclear force structure. 

Dissuasion:   The goal of dissuasion involves discouraging opponents and potential opponents from militarily challenging the United States .  Dissuasion does not involve the use of force, but it does require the force structure and technology base necessary to discourage opponents from anticipating success in competing militarily with the United States .  A past example of dissuasion was the Soviet decision to scale back its deployment of ballistic missile defense in the 1960s:  The Soviet leadership understood that the U.S. strategic offensive potential could overwhelm Soviet defenses.  The maintenance of U.S. nuclear capabilities and a viable nuclear infrastructure may be necessary to discourage some opponents from choosing to engage in a nuclear competition in arms. 

For example, the elimination or steep reduction of U.S. ICBMs could lower the bar considerably for an adversary who might, under such circumstances, consider realistic the possibility of achieving a counterforce strike option against the United States .  Maintaining a set of diverse nuclear retaliatory capabilities serves to discourage the aspiration for any such option.  Consequently, it may be critical for dissuasion purposes to maintain an adequate nuclear force structure, particularly because if the United States were to decide to severely reduce its deployed forces, it would not be able to recover those capabilities easily, inexpensively or quickly.  As the Chinese, by their own statements, move toward greater interest in counterforce nuclear options, keeping the bar high for any possible success in that regard may be critical to future stability.  Moving to a very small number of U.S. nuclear retaliatory capabilities could encourage the Chinese in the wrong direction.  

Defeat:  This fourth reason for the maintenance of U.S. nuclear capabilities does envision the potential military value of nuclear employment.  The NPR referred to this as the goal of “defeat.”  Military war planning and targeting are properly the prerogative of U.S. uniformed “war fighters.”  For the purposes of this article it is sufficient to point out that military threats may emerge that can only be countered with confidence by nuclear weapons.  Nuclear weapons may be the only means available for promptly destroying hard and deeply buried targets, achieving prompt war termination, preventing an adversary from marching on and annihilating civilian centers, or for possibly eliminating nuclear or biological threats arrayed against the United States and allies.  Whether such objectives would be worth the political and human costs of employing nuclear weapons would be a calculation to be made by the president and would likely be shaped decisively by the specific context and the opponent’s prior actions.  If the opponent had, for example, already employed biological weapons (BW) with horrific effect against U.S or allied civilians and military forces, would a President consider a nuclear response appropriate, militarily useful, and/or necessary to prevent subsequent BW attacks?  The possibility can not be excluded that a president would want to have the option for nuclear employment under such circumstances, and that employment could then be vital to U.S. or allied survival.  It takes considerable hubris to claim knowledge that such circumstances assuredly will not arise in the future, and correspondingly that no critical military employment value can be attributed to nuclear weapons.  

There are risks associated with retaining and modernizing the U.S. nuclear arsenal; there are also  risks in not doing so.  Nuclear weapons may be critical for the deterrence of war and the dissuasion of military competition; and, they are critical to the assurance of allies who have indicated that they will consider moving toward their own nuclear capabilities if they conclude that the U.S. extended nuclear deterrent is no longer reliable.  Advocates of the elimination of U.S. nuclear weapons tend to presume they know that adversaries will continue to be deterred by U.S. non-nuclear weapons, that allies will continue to be assured by the same, and that the U.S. “good example” of moving away from nuclear weapons would be emulated.  Again, the burden of proof is on those who make such claims, particularly when considerable available evidence points to the contrary.  Yet, proponents of moving toward nuclear disarmament have not offered any serious net assessment of the consequences likely to follow from their recommended course. 

Finally, there are conditions that should attend any significant U.S. steps toward nuclear disarmament.  The realization of some of those conditions would represent a more dramatic restructuring of international relations than has occurred since the 1648 Peace of Westphalia.  This need not preclude our thinking about steps toward nuclear disarmament, but it certainly should make us wary of moving quickly toward the vision. 

For example, one of the reasons deterrence is so valuable is that it provides incentives for self-discipline in the behavior of states that otherwise can not be trusted to behave peaceably, i.e., the lack of trust that can be attributed to such states is the reason we need effective deterrence.  Yet, movement toward the elimination of U.S. nuclear weapons would at various key points require considerable trust that friends and foes alike were pursuing the same goal honestly and could, at a minimum, be relied on to give up nuclear and biological weapons.  The fact, however, is that not all states are trustworthy, and it is those states that often pose security challenges.  In the past, untrustworthy states included Nazi Germany; now they include North Korea ; in the future there will undoubtedly be comparable candidates.  The appropriate lack of trust that these states will behave benignly and honestly is the reason deterrence is important and why formulas for disarmament remain visions.  The same lack of trust inherent in international relations that creates the need prevents visionary solutions.  Again, the proponents of nuclear disarmament have not begun to suggest how this sturdy barrier to the realization of their vision and like visions in past centuries could be brought down while maintaining our security and the security of our allies.  We all would like to hear and to believe.

Ronald Reagan was a proponent of a non-nuclear vision; he also repeated the motto “trust but verify” and understood that concomitant conditions such as the realization of highly effective active defenses had to precede the vision.  If his vision is to be brandished now in his absence, it should be brandished in its entirety.

John S. Foster, Jr., is Chairman of the Board of GKN Aerospace Transparency Systems.  He was the Director of the Lawrence Livermore National Laboratory (1961-1965) and then Director of Defense Research and Engineering (1965-1973).  An officer of TRW and served on its Board (1973-1994).  Served on the President’s Foreign Intelligence Advisory Committee (1973-1990) and is a member of the Defense Science Board, which he chaired (1990-1993).  Currently consults for the DoD, DoE, U.S. Senate and corporations.