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



Previous Newsletters

this issue

Contact the Editors


What Are Nuclear Weapons For?
Michael May

From the very beginning of the nuclear era, concerned people from all walks of life have debated what should be done about weapons so destructive that a single one had wiped out the heart of a city and killed over a hundred thousand people. Should they be banned, and, if so, how could that be done in a world where deadly threats to states still existed and where the knowledge needed would inevitably (and did in fact) become ever more widespread?   If not banned, could they be considered as just another, if more destructive, tool of war, to be used as military need dictated? Or did they fit a special category, that of making the prospect of war so frightening as to deter war itself if there was a possibility of their involvement? If this last notion corresponded to reality – a doubtful notion since, fifty years earlier, the machine gun had been thought to be so destructive as to make war obsolete, only to be subsequently used in the two most destructive wars in history – who should be entrusted with those weapons? More realistically, since there was and is no supranational authority, who would in fact come to have them?

Governments seem to have settled in the main on the third of those early ideas, that of nuclear weapons serving as deterrents. The other two ideas have not gone away, however. Given that governments do not always show rationality, that they often do not know what to be rational about, and that they seldom show much concern for the human race as a whole, the  case for nuclear disarmament has remained strong. At the other end of the spectrum, using nuclear weapons to win wars has also retained its adherents. And the question of who should retain these dangerous deterrents, while answered in principle by the 1970 Nuclear Non-Proliferation Treaty (NPT), has not received universal support, as witness the four nuclear-armed states outside the treaty. Nowhere do these questions come into sharper focus than in attempting to cast light on the current question of what should the future U.S. nuclear weapons stockpile look like.

Stockpile issues are the topic of this essay. Stockpile issues however cannot be discussed without considering what the nuclear weapons are to be used for. The US needs one stockpile if nuclear weapons return to the forefront of military doctrine, for instance if lower yield weapons are to be used tactically to destroy underground installations, and another if they are to be kept in reserve indefinitely to deter the use of nuclear weapons by other states. It needs one stockpile if the US plans to keep nuclear weapons into the next century and another if it plans to lead an international movement toward nuclear disarmament on an aggressive time scale, as several prominent former officials propose. These and similar questions have to be resolved whatever the current weapons programs may be,  but the current Robust Replacement Warhead (RRW) program poses them with particular sharpness.

Thus, what yields shall the replacement warheads have? The same as the warheads they replace or some other yields? What configuration requirements shall they meet? The requirements imposed by the current bombs and missiles, or requirements imposed by some not yet designed earth penetrating or other new missile suited to new missions? Yield, robustness, and dimensions are among the defining parameters for a weapon design,  and weapon designers cannot proceed without either getting answers to such questions from the relevant authorities or guessing at what those answers will be.

So far, designers are investigating safer, more robust alternatives to existing warheads for Cold War missions.[1] There is only limited interest in nuclear weapons programs in the government at present. What Washington seems to require of the programs is to keep the US stockpile in operating condition and safe, and otherwise just keep going along at moderate levels of expenditures without creating adverse publicity. It may be that this will continue and the RRW program will simply replace existing warheads. But this prospect is at odds with the thrust of, not only the 2002 US nuclear posture and subsequent national strategy documents, but also with the thrust of the postures of four of the five nuclear weapons states.  These four  have stated that their nuclear weapons are not only for ultimate deterrence, but also to deter or punish any state assisting WMD terrorism.[2] Cold War stockpiles are not well designed for the latter goal. One way or the other, whether now or later, by default or by design, these choices will be made.

What are these choices and what are their consequences? The most fundamental choice is whether the US will commit itself to 1) using nuclear weapons only as ultimate deterrent against aggression that might destroy or severely damage the country or its allies and move steadily meanwhile to arms reductions and other cooperative arms control measures; or 2) whether, as the 2002 US nuclear posture and the new nuclear postures of several other nuclear-armed countries imply, nuclear weapons will be used not only to deter but actively to prevent nuclear proliferation, chemical or biological attacks, and state support of WMD terrorism. Deterring and actively preventing terrorist use of WMD and nuclear proliferation are important goals. Nuclear deterrence and the potential use of nuclear weapons to destroy underground or otherwise protected facilities could assist in reaching those goals. At the same time, minimizing the incentives for any state to acquire or use nuclear weapons is also an important as well as a complementary goal. While the main incentives to acquire nuclear weapons stem from a state’s evaluation of its security needs, its domestic politics and its history, the policies of the US and other nuclear-armed states also figure in the acquisition decisions. For any given state, these two classes of incentives may be in conflict.

If the nuclear-armed states – by and large powerful countries with effective conventional militaries and relatively well-protected by size and alliances – consider nuclear weapons as valuable tools of policy and warfare, security establishments in the many countries that are more vulnerable than they are will consider nuclear weapons more seriously. Nuclear weapons have large political and economic costs, but they are credible deterrents, which can destroy the bases for power projection as well as deny an otherwise superior aggressor the benefits of victory. Indications from the most powerful states, those that set the standards for military excellence, that they themselves need nuclear weapons for their security powerfully reinforces the argument for acquisition that stem from a state’s assessment of its security and from its politics.

To the contrary, if the nuclear-armed states move away from nuclear weapons, by reducing their numbers and salience, and especially if they find credible ways to support the security of non-nuclear states, as NATO did for instance, the arguments for acquisition are weakened. The success of the NPT among the majority of nuclear-capable but non-nuclear weapon states for the past thirty years attests to the importance of credible security arrangements that guarantee, as well as possible, the security of peaceful states that abide by the NPT.

Actual first use of a nuclear weapon by any country would have an even more pronounced effect on incentives to proliferation. First use of a nuclear weapon by anyone would have both direct effects, which would depend on location and circumstances, and indirect effects. The latter cannot be fully assessed ahead of time but they include the cost of giving the world a demonstration of the effectiveness of nuclear weapons as well as the cost of abandoning the sixty-year old taboo against use of those weapons. The last time the weapons were used, in 1945, nuclear weapons hastened the end of the war, helping to save many casualties and ending the war on more advantageous circumstances for the United States, by limiting Soviet advances, and for Japan, by sparing it the split occupation that Germany endured. But most important, despite some very close calls, those two uses did not lead to further use.

We cannot count on this today. The world today is vastly different from the world of 1945: it is more familiar with the needed technologies, it is wealthier, and with wealth comes greater availability of materials and facilities.

And it is different politically. No longer do two colossi, located far away from each other, stride the political world. The West may be dominant economically and in some measures of military power, but neither the West nor any other major power can determine whether states in the rest of the world will go nuclear nor whether nuclear capable states will go to war with each other. The world today is more similar to the usual historical situation than the post-World War II world was: a number of states in disparate situations, many of which have reason to be insecure, many of which can acquire the latest weapons. No one can predict the consequences of using nuclear weapons in such a world, but first use today seems unlikely to be a last use, especially if that first use turns out to be militarily effective.

The above argues that the continued reliance of major powers on nuclear weapons and the expansion of nuclear missions can influence other states toward acquiring these weapons. Proliferation heightens the danger of both nuclear war and nuclear terrorism: more governments with conflicting interests will have the weapons and more nuclear weapon materials and essential components will be available in different places and under different security. First use would have unpredictable consequences, including an increased likelihood of further use. The right stockpile, in this view, would therefore be a stockpile that contributes minimally to proliferation, a decreasing stockpile, kept in the background of policy, and not a stockpile aimed at greater utility in a variety of circumstances.

Total nuclear disarmament is another question, however. The world political situation described argues that the consequences of abandoning nuclear deterrence entirely are also unpredictable and could also be dangerous. The current unsettled and more broadly nuclear-capable world argues against bringing back nuclear weapons to the forefront of military doctrine but it also argues against precipitate nuclear disarmament – unless of course it can be done verifiably, universally and, most important, irreversibly. All three of those conditions are difficult to meet but the last is the most difficult. So long as war is possible, a war to the finish between two disarmed but nuclear capable adversaries could easily lead to an exceedingly dangerous race to rearm. The distinctive feature of nuclear weapons that makes them much more difficult to ban than chemical and biological weapons is that chemical and biological weapons are terror weapons, effective against civilians and unprepared military but not against a prepared modern military. On the contrary nuclear weapons are effective military weapons: with a few missions and relatively little expense, they can defeat even large scale conventional attacks, especially those attacks that require force projection at a distance and therefore air bases, ports of debarkation, logistical centers, and other vulnerable and costly targets. They are in fact partial equalizers against the might of the United States especially, given the long-term U.S. priority on force projection.

The most important argument against complete nuclear disarmament is that powerful states have historically initiated the most devastating wars. For at least two centuries before the advent of nuclear weapons, those wars spread devastation while doing little but adjust boundaries in minor ways. Since the advent of nuclear weapons, the world’s most powerful states have avoided full-scale war with each other. No one can demonstrate that the advent of nuclear weapons ended (at least so far) war among the most powerful states, but no other political mechanism has changed – not the way political leaders come to and hold power, not the limited rationality in the world’s capitals especially as regards to hostile countries, not the security dilemma generally – while the fear of nuclear devastation has been a major consideration in instilling realism and caution in disputes. That fear has not been salient since the end of the Cold War and it is tempting to think that relations among those states have changed in some other fundamental way, perhaps owing to economic interdependence. What those other ways are however does not readily come to mind. In particular, most of the same states were economically interdependent in 1913.

So a world without nuclear weapons is a distant prospect, as distant perhaps as a United Nations or other accepted authority that could and would guarantee all states against wars to the finish. More likely is a world where nuclear weapons are a background and diminishing presence, where nuclear-armed states do more to move toward fulfillment of their disarmament obligations under the NPT, and where meaningful security assurances are extended, as it becomes possible, to states that may otherwise seek nuclear weapons for security.  Evaluating how much more likely presents a mixed picture.

On the one hand, there are cooperative international efforts to bring about nuclear disarmament in North Korea and to prevent Iran from acquiring facilities that could be used to make highly enriched uranium. It is noteworthy however that both are countries with which the U.S. has long had hostile relations and of which it requires little else than non-proliferation and an end to support of terrorism. In the cases of countries that were either allied with the U.S. or needed by the U.S. for some strategic purpose, such as Israel and Pakistan, on the other hand, U.S. efforts to prevent nuclear proliferation or to roll it back have either been absent or have taken a relatively low priority. The record therefore is mixed, although the U.S. historically initiated most nuclear non-proliferation initiatives.

Beyond these immediate problems, the nuclear-armed states put essentially zero political muscle behind nuclear disarmament and little toward reducing the number and salience of nuclear weapons. Numbers are being reduced, if slowly, but systems are being modernized in most nuclear-armed countries. France, Russia and the United Kingdom as well as the U.S., have explicitly broadened their nuclear deterrent targets to include states that assist WMD terrorists, along with abandoning their very limited commitments to no-first-use of nuclear weapons. India has no doctrine of no-first-use and has given no assurance that it will not use nuclear weapons against non-nuclear states (so-called negative security assurances). Pakistan ’s doctrine is not so clear but is likely to be patterned on India ’s. Only China has remained with its initial policy of no-first-use and no use against non-nuclear states. On the whole, there is little sign of a concerted approach to dealing with the nuclear danger.

There are steps that the nuclear-armed states can take that would help rebuild the NPT consensus despite the asymmetric nature of the treaty. Such steps include different nuclear postures that take better account of the need to cooperate with other states; agreement of the nuclear-armed powers to the International Court of Justice ruling that the use or threat of use of nuclear weapons would violate international law; more clearly defined negative security assurances, conditional only on adherence to the UN Charter and not on good relations with any major power; actual Entry Into Force of the Comprehensive Test Ban Treaty, a formal step that makes the treaty binding on all signatory parties and which is conditional on the United States and China, among others, ratifying the treaty; more great power support for Nuclear Free Zones. These all would help, in some cases significantly. But they do not go to the demand for nuclear weapons based on local insecurities. 

Must then further and more significant steps to allay the nuclear danger wait upon the difficult and probably distant resolution of conflicts in insecure regions of the world such as the Middle East and South Asia? The major powers come together to an extent on resolving the Palestine problem and the Iran problem. They may yet come together to resolve the Iraq problem. But resolving today’s problem will not be good enough if new problems arise tomorrow, as they surely will. Some general agreement on the conditions for international support, for getting the benefits of civilian nuclear technologies, for dealing with WMD terrorism such as exist in principle now must get the kind of believable great power support that Cold War alliances used to get. Cold War alliances, for all their defects, especially on the Soviet side, did manage to keep most of the then-nuclear capable countries from going nuclear. Security arrangements are not enough but they are necessary if more meaningful steps toward nuclear disarmament are to be taken.

What steps to take where, how to take the first steps, how to get a durable consensus among the P-5 (which have a special obligation) to begin with and among the other nuclear-armed states subsequently, those questions define a long-term agenda for diplomacy backed by meaningful inducements and example. This paper cannot pretend to begin to flush out the details of that agenda. But if the U.S. is to go in that direction, the goal of its stockpile should be limited to maintaining last-ditch deterrence, principally against the use of nuclear weapons by others against U.S. allies and indeed, if the UN Charter principles are to be implemented, against any state.

Michael May is Professor Emeritus (Research) in the Stanford University School of Engineering and a Senior Fellow with the Freeman-Spogli Institute for International Studies at Stanford University. He is the former Co-Director of Stanford University's Center for International Security and Cooperation.  He is Director Emeritus of the Lawrence Livermore National Laboratory where he worked from 1952 to 1988, with some brief periods away from the Laboratory. He served as U.S. delegate to the Strategic Arms Limitation Talks and other arms control and defense advisory positions.


[1] Deputy Secretary of Energy Clay Sell, remarks to the Carnegie International Nonproliferation Conference June 26, 2007While the RRW will not represent a new role for nuclear weapons, and will have no new or different military capabilities, the RRW will have more robust performance margins that will increase reliability and enable us to significantly reduce the size of the overall stockpile.”

[2] Russia : Sokov, Nikolai. “The Russian Ministry of Defense’s White Paper: The Nuclear Angle.” Nuclear Threat Initiative.  October 2003 http://www.nti.org/db/nisprofs/russia/weapons/whitepap.htm France : Chirac, Jacques. Address on France ’s Strategic Forces. Landivisiau, France . January 19, 2006. United Kingdom : Department of Defense, “The Future of the United Kingdom ’s Strategic Deterrent.”  Government of the United Kingdom .  London. December 2006. China : Office of State Control, “ China ’s National Defense in 2002.” Government of the People’s Republic of China . December 9, 2002.



Previous Newsletters

this issue

Contact the Editors