Orbis, Winter 2004

Thinking (Again) About Arms Control

by Paul Bracken

Paul Bracken (paul.bracken@yale.edu) is professor of management and political science at Yale University.

Arms control has been marginalized over the past several administrations. Even its most dedicated backers have failed to win support for it in key segments of American political life. It has also lost its way conceptually. Whether it was a case of politics undermining arms control or the American political system rejecting a flawed form of it, clearly a fundamental rethinking and redesign of arms control is needed. Instead of trying to make the world over through multilateral treaties as in the 1990s, we should return to arms control’s 1960s roots and again concentrate on lowering the chance of war and reducing the damage if war occurs anyway.

Especially in recent years, arms control has become distanced from these original purposes, coming to serve a broader goal of creating a multilateral anti-nuclear regime. Its scope has come to include other types of arms as attempts were made to turn arms control into a universal regime alongside trade, human rights, the environment, and others. Taken together, these separate regimes aimed to transform the fundamental character of international relations into a global system that institutionalized cooperation.

Commendable as this goal is, arms control has not lived up to its promise. It has failed to prevent the emergence of a second nuclear age: Not only have atomic weapons proliferated, but their spread has created a new nuclear age that is structurally different from the first.

If the nuclear ages that began in 1945 were all of one cloth, then treaty-based arms control, reinvigorated and backed by political leadership, might work again. But the structure of the nuclear age has changed too much for this to happen. The second nuclear age is a world of multi-state, not bilateral, nuclear interactions, and of actors who have radically different strategic cultures than those of the Cold War superpowers. These actors are poor, and this increases their dependence on nuclear deterrence, because conventional force alternatives are expensive.

We need to face up to these structural differences if arms control is not to be consigned to irrelevance. In the first nuclear age, simple two-actor interactions, conservative behavior at the top, and conventional forces used as a firebreak before nuclear use restrained the superpowers from spilling into conflict. These features are all far weaker for the emerging atomic states of the second nuclear age.

As demonstrated by the extraordinary performance of the U.S. military in Iraq, the shift in the first line of American defense from nuclear to conventional arms is another structural change with far-reaching consequences for arms control. The United States no longer needs to rely on nuclear threats, as it did in the Cold War. There are major new challenges, but also opportunities for lowering the chance of nuclear war in the twenty-first century.

The Apparent Choices

Two alternatives form the apparent choices for U.S. arms-control policy in this new world. They are for now more attitudes than strategies, but they fairly describe the choices generated by the debate over arms control to date in the United States.

The first apparent choice is for the United States to return to a reinvigorated version of the treaty-based arms-control regime of the 1990s. Here, we simply go back to what we know how to do: negotiate international treaties in an endless series of diplomatic forums. The United States could launch a diplomatic offensive to strengthen the Nonproliferation Treaty, require tougher inspections, bolster multilateral monitoring institutions like the International Atomic Energy Agency, sign the Comprehensive Test Ban Treaty, ban defensive missiles from its arsenal, and engage Russia for even deeper cuts in strategic nuclear arsenals.

The second apparent choice is also something we know how to do: forget about arms control and turn to military solutions to the challenges of the second nuclear age. After all, there was nothing to physically stop North Korea, Iran, or Iraq from pursuing WMD. And the whole system of multilateral treaties is ultimately unenforceable if military force is ruled out of the equation. The logical conclusion, then, is for the United States to shift from arms control to military strategy. Problems of proliferation would be dealt with by adding bunker-busting nuclear weapons to the arsenal. Virtually all military options are kept on the table, such as the United States’ publicly threatening nuclear preemption, regardless of the political consequences.

Both apparent choices, being things we know how to do, are “easy,” and each has its own core constituencies of supporters and detractors. Yet both have severe limitations: mainly, that they come out of old debates. Their application to the problems of a second nuclear age is barely addressed. Returning arms control to cooperative regime-building through negotiated multilateral treaties will consign arms control to irrelevance. On the other hand, purely military approaches to the challenges of the second nuclear age are unlikely to work in the long run.

The second nuclear age creates important challenges that these “easy” choices miss. This paper argues that returning to the original purposes of arms control offers an important way to meet these challenges.

The Pillars of Arms Control: Old and New

Modern arms control only took shape in the early 1960s, when a loose collection of strategic concepts was linked together to manage the dangers arising from new technologies. It rested on three pillars. First, it wasn’t disarmament. This is the reason the term “arms control” was coined in the first place, to distinguish it from disarmament. Disarmament proposals had been a theater for the worst kinds of political hot air in the first half of the century. In the 1950s, history was repeating itself, with the Soviets grandstanding for “general and complete disarmament.”

Since the chance of war was believed to be directly related to the number of weapons, disarmament aimed to either cut the number of weapons or eliminate whole weapons classes. Reductions were seen as a general good that needed little additional justification given the menace of the new technologies.

Arms control, in contrast, analyzed the relationship between strategy and the goals of reducing the chance and damage of war. The founding books of the field were sophisticated and thoughtful, focusing on complicated interactions: how nuclear weapons would actually reduce the chance of war, the crisis dynamics between two opponents, and tacit and explicit agreements to eliminate dangerous behaviors.

The second pillar of arms control was making a sharp distinction between ends and means. As Hedley Bull wrote in 1961:

It is commonly assumed that the only important questions that arise in connection with disarmament or arms control concern how it may be brought about. But the question must first be asked, what is it for? Unless there can be some clear conception of what it is that disarmament or arms control is intended to promote, and to what extent and in what ways it is able to do so, no disciplined discussion of this subject can begin.

As an example, nuclear weapons were known to be highly destructive but were also seen to be beneficial. Without a nuclear deterrent, the defense of Europe and other areas would have been far more difficult and risky. From a disarmament perspective, the tactical nuclear defense strategy was a disaster, because it increased the number of nuclear weapons. Its application to Europe legitimized, rather than abolished, these weapons. From an arms control perspective, in contrast, this strategy made perfect sense. It lowered the chance of war in Europe. The Soviets knew that any attack by them could escalate into nuclear conflict. Arms control (the means) called attention to the safe operation of the nuclear deterrent (the ends), and thereby lowered the chance of war.

The third pillar of arms control was creating a middle position between the extremes. In the early 1960s the debate over U.S. strategy was between those who believed that the arms race was more dangerous than the Soviet Union and those who believed that the Soviet regime was so malevolent that the only way to deal with it was through military power. The first group called for nuclear disarmament under a strengthened un, the second for thousands of atomic missiles and cobalt (dirty) bombs.

One of the most striking features of early arms control is the way it brought people together, both between the two superpowers and within them, in common debate. It was respectable to take a middle position and seek additional alternatives. The discussion highlighted interests shared by the superpowers and the fact that disarmament might even increase the chances of war.

In recent years the pillars of arms control have badly eroded, and in some cases broken down altogether. Arms control in too many instances became a Trojan horse for discussing disarmament without saying so, with proposals put forward under its auspices for the reduction or elimination of nuclear weapons, land mines, cruise missiles, defensive missiles, chemical and biological weapons, small arms, and even military satellites.

To be sure, there is nothing wrong with eliminating biological weapons. But the conflation of arms control with disarmament for the purpose of expanding the scope of negotiated treaties confuses two distinct fields. Since negotiated treaties were a pillar of arms control, treaty withdrawal automatically was a setback, regardless of its effects on the probability of war. The means-ends distinction of arms control emphasized by Hedley Bull, in which treaties were embraced as a way to lower the chance of war, tended to vanish.

There is also nothing wrong with building a better world order. But labels matter. Calling this “arms control” caused many problems. Identifying arms control with multilateral treaties overlooked that there was much one could achieve without treaties, and even without any mutual agreements. These offer important opportunities and shouldn’t be disregarded.

By morphing from the original purposes of arms control without openly saying so, arms-control proponents also made it difficult to sustain a middle ground position on complicated questions of military strategy and national security. One was either an arms controller or an arms supporter. The gradual disappearance of a middle ground conversation was one of the most harmful consequences of the shift in the meaning of arms control.

The Need to Rethink Arms Control

Arms control is not doomed to fail, nor should it be abandoned altogether. It has made tremendous contributions, many of which had nothing whatever to do with formal treaties. The installation of a hotline between Washington and Moscow was an acknowledgment by each that new technologies made the strategic environment dangerous in unprecedented ways. While the hotline hardly prevented reckless behavior, it profoundly changed the interpretation of that behavior on the occasions when it was used. The hotline was an acknowledgment of this condition, something that was far more valuable than any signed treaty.

The massive investment by the two superpowers in the command and control of nuclear forces was also strongly shaped by arms-control thinking. Looked at purely militarily, the purpose of command and control was simply to get word to the forces. But arms-control thinking showed that there was a lot more to command and control than this. If the enemy thought that your command structure was shaky, it might be forced to consider actions that it would not otherwise wish to take, such as initiating a first strike. Hardening the command-and-control system-and letting the enemy know this-drove down the chances of war. There were no “command-and-control treaties” here, but rather an understanding that the two sides had common interests. Wanting to ban command-and-control improvements because they acerbated a nuclear arms race wasn’t a good way to think about the problems of the first nuclear age, and arms control said so.

Another arms-control success was the 1968 Nonproliferation Treaty. True, it failed to stop the spread of nuclear weapons. But the time element has to be taken into account in judging it a success or failure. Many strategists believed that the npt might buy five or ten years in which the spread of the bomb could be halted. In fact, it bought nearly twenty-five years. How many other U.S. government policies have worked this long? If Medicare, to take an example, were successfully reformed for “only” the next twenty-five years, any reasonable person would call this an extraordinary success, even if it began to unravel at the end of this period.

It was never hoped that the nonproliferation regime created by the npt would by itself stop the spread of atomic weapons once and for all. Such an interpretation was endowed on this regime only in the 1990s, as arms control was regeared to promote treaties as a form of global regime-building.

By the mid-1990s the nonproliferation regime was unraveling. India and Pakistan set off multiple nuclear shots. Iran and North Korea both tested long-range missiles that could deliver atomic bombs and had weapons programs in the works. In Israel, Pakistan, and India, the nuclear option was not only pursued, it was institutionalized into both their strategy and their defense technology programs. Israel was deploying missile defenses and building a sophisticated arsenal well beyond the Model T force it had fielded earlier. India was building missile warning radars and mobile nuclear launchers, as was Pakistan. All three countries entered the Dr. Strangelove world of secret launch codes, subtle manipulation of nuclear threats, and hardened command posts.

These three countries were not signatories of the npt. But to discount their behaviors as somehow “not counting” because of this is absurd. By fueling the structural transformations of the second nuclear age, their actions helped break the nonproliferation regime, npt or not.

In East Asia during the same time period, China began to modernize its nuclear forces after decades of neglect. Adding this to North Korea’s acquisition of the bomb and the declining capacity of Pyongyang’s conventional forces, the chance of nuclear war, and the potential damage of war, in Northeast Asia was growing, no matter how impolitic it became to say so.

Treaty-based arms control in the 1990s could not hold back these developments. Its core treaties— the CTBT, START, NOT, and the ABM Treaty— were promoted as foundations of a global anti-nuclear regime, one that would make nuclear weapons irrelevant. But nuclear weapons weren’t rendered irrelevant, as is now all too clear.

Arms-control theorists tried to make the case that these developments mattered little, that the successes outnumbered setbacks. But numbers can be used in many ways. Disarmament proponents pointed out that more countries had signed on to the npt, in order to show that the world was becoming safer. But this criterion ignored the original standard of arms control, reducing the probability of war. The chance of nuclear war was in fact growing, as was all too apparent. The atomic competition between India and Pakistan and the sharp increase in nuclear risks in the Middle East and on the Korean peninsula became impossible to ignore. All things being equal, the more countries on board a nonproliferation regime, the better. But all things weren’t equal, and nuclear war was increasing in probability. The idea that Tehran or Pyongyang would abandon their bomb programs because the United States signed the ctbt, or eliminated its own nuclear forces, cannot be taken seriously. As for the start, it is, if anything, even less relevant to the problems of a second nuclear age.

A number of countries have abandoned their nuclear programs. South Africa, the Ukraine, and Kazakstan gave up nuclear weapons. These are welcome developments. But this doesn’t reverse the growing chance and damage of war in other regions, either. Confusing the two trends-nuclear abandonment by some and acquisition by others-is as dangerous as it is sloppy thinking.

By the late 1990s the second nuclear age had emerged, with its own dynamics that weren’t dampened by 1990s arms control. Despite important successes, a framework that had worked for decades was coming apart. Tightening it up wasn’t going to reverse the spread of atomic weapons to the countries of greatest concern, because the entire basis of arms control in the 1990s was conservative. It relied on multilateral rather than unilateral initiatives, and had a preference for passive rather than active behavior; familiar (e.g. treaties) rather than innovative approaches, and respect of sovereignty rather than interest-advancing goals. It preferred avoiding risks. Such a conservative approach was inadequate to prevent a second nuclear age.”

New Arms Control Opportunities

The new challenges and opportunities for arms control require going beyond the last decade’s multilateral treaty-based approaches, but not necessarily jumping from the first apparent choice to the second. New thinking about alternatives that reduce the chance of war and the damage from it needs to be brought to bear on the subject. Suggested areas for thought toward these ends are set forth below.

Agreed Nuclear Worlds. Serious consideration must be given to which countries will be considered legitimate members of the nuclear club and which will not. I call this an “agreed nuclear world.” Acknowledgement as a legitimate nuclear state might be tacit or explicit, but need not be instituted in a treaty.

Who decides who is a legitimate nuclear power, and what might an agreed nuclear world look like? Both questions are contentious. But not asking them overlooks the most important question in international politics today. The question of “who decides” brings to the front what is often forgotten in arms-control discussions: power. It is pointless to design arms-control proposals without acknowledging that the United States has more power than any other state. Overlooking the unique U.S. power position is like overlooking the Monroe Doctrine in the nineteenth century or the Brezhnev Doctrine in the twentieth. They are facts of the international system.

Logical answers to the second question, what an agreed nuclear world might look like, are either zero, one, or five nuclear states. That is, nuclear weapons could be abolished altogether; the United States alone should be allowed to have them; or only the five nuclear states enshrined in the npt should possess them. But we shouldn’t confuse familiar debates with the real challenges of the second nuclear age. Some nuclear worlds are better than others because the chances of a nuclear disaster are lower in some than in others.

We cannot know everything about the future. But we do know some things. Some worlds have distinctly lower probabilities of war than others. A nuclear North Korea and Iran under their present regimes is a lot more dangerous than a nuclear Israel and India, and it’s important to say so. The reason for this has to do with the nature of their regimes, not merely their possession of atomic weapons. It’s important to say this, too. Creating stable nuclear worlds should be as much the purpose of U.S. arms-control strategy as was trying to limit Soviet heavy missiles in the first nuclear age.

Including Israel and India in an agreed nuclear world could well undermine the npt. But so what? It was the view that arms control is an interlocking system of treaties with foundational pillars that got us into problems in the first place. Acknowledging India and Israel as nuclear states and encouraging their responsible behavior offers a lot of benefits to the United States. It is not in the United States’ interest to have either one test or build larger bombs, or to rob their conventional forces of resources to pay for nuclear ones.

Of course, the concept of agreed nuclear worlds described here only touches on a complicated subject. Other countries would have to be considered. This should be done not only on a case-by-case basis where the character of the political regime is considered, but also on a systems basis, asking the question whether it would raise the risk of war in a given region.

No First Use of Nuclear Weapons. The United States declined to declare a no-first-use pledge on nuclear weapons during the first nuclear age because to do so was thought to increase the chances of Soviet conventional attack of Europe. Clearly, this is now moot. Today, acceptance of no-first-use could offer significant benefits, since conventional forces have a strategic capacity once possessed only by America’s nuclear arms.

A “no-first-use” declaration by the United States calls for much more serious consideration then it has received. It could, for example, be extended to no first use of any WMD, to include chemical and biological weapons. This would communicate that the United States would employ nuclear weapons only under the gravest conditions, that is, when another nation had broken a norm that has been in effect since 1945 against using these weapons. It would delegitimize nuclear weapons, and chemical and biological ones as well.

There are many variants of a no-first-use pledge that need further discussion. In one, the United States could sign a no-first-use treaty with other nations that included WMD and every other state suspected of possessing WMD. That is, a group of countries led by the United States could prohibit first use of WMD by themselves and everyone else. This could even be backed with statements to the effect that no-first-use implied a guaranteed second use by Washington.

Such a policy could not prevent the use of WMD, but it would bolster America’s commitment to an important international norm. One problem with conventional warfare options is that, depending on the circumstances, it could take a long time to effect a retaliatory response adequate to the provocation. At the same time, since these are declaratory pledges, Washington wouldn’t be locked in to any particular course of action.

The United States would be giving up very little compared to its gains. It is not in America’s interest to be a superpower that threatens first use of nuclear weapons when it has built, at great expense, a superlative conventional capability, and when India, China, and Israel have in one form or another made their own no-first-use pledges.

An American no-first-use pledge would make it significantly more difficult for any of these countries to break theirs. For example, were U.S. relations with China to worsen, it would be much harder for Beijing to break its pledge in the face of an American no-first-use pledge. To do so would single it out as having broken a pledge, with all the negative political consequences thereof.

No Surprises Compacts. The danger of surprise attack in the first nuclear age arose from the enemy. No one believed that allies such as Britain and France would take precipitous nuclear actions that could worsen a crisis. nato’s tight control of military force minimized this possibility. China never brandished its nuclear weapons in ways that made Moscow nervous, although this was considered a danger by some.

In a world of multiple nuclear actors this is no longer the case. “Allies” are much less controllable. It isn’t hard to imagine that some of America’s allies could take actions that would endanger the United States. During both the 1991 and 2003 wars against Iraq the U.S. government had to think through the implications from a possible Israeli escalation. As Washington takes a greater role in the India-Pakistan dispute, either one of these nations could act in ways that would surprise Washington. This situation is a far cry from the strategic calculations of the first nuclear age, where independent French or British nuclear actions were unimaginable.

“No-surprises” compacts should be thought out to lower the chance of war arising from complicated interactions, giving consideration to different perceptions of national interest. Problems arise when states embark on alerting actions or troop movements, because the interpretation of behavior by each actor can be radically different. This was a problem in the first nuclear age. It’s a much bigger problem in the second, because of the very different strategic cultures involved.

Neither the content of a no-surprises compact nor the nations it might apply to can be specified as yet. And that is exactly the problem. No one is thinking about a very important issue. For at present, the U.S. national interest in the context of arms control has not been identified well enough to do this. Some nations-including North Korea-should clearly be excluded from such a compact, because it could be interpreted as legitimizing their possession of atomic weapons. But which countries, and what content, should be included is something in far greater need of elaboration and thought. A severe crisis is not the time to play catch-up.

Missile Defense and Security Guarantees. Despite complaints from many critics that U.S. withdrawal from the abm Treaty destroyed one of the pillars of arms control, the pursuit of national and theater missile defense creates major new arms-control opportunities. The nature of these opportunities so far has received scant consideration. Security guarantees, wherein Washington supplies missile defense protection to other countries in return for agreements by them not to build their own nuclear deterrent, could be an important mechanism to stop the further spread of atomic weapons.

U.S. missile-defense protection of other countries could range from a mutual understanding between them to a provisional accord arranged as one important part of a larger military relationship. It could also be an ad hoc crisis management tool, to enlarge the set of options available to the president. A sea-based theater missile defense would be highly mobile and could be shifted to different parts of the world with greater ease than conventional forces.

The political significance of extending U.S. missile defense to protect another country could be as important as its military worth. Security guarantees made to other countries would have more credibility if they went beyond extended deterrence to include active population and urban defense. Most thinking about American security guarantees focuses on reversing military actions taken against Washington’s allies. If North Korea invades the South, the United States will respond to protect Seoul. Yet there are added dimensions to Washington’s security guarantees. Projecting defense, that is, using a U.S. missile defense shield which shot down ballistic missiles aimed at an ally’s urban industrial complexes, should be a pivotal feature of future military alliances.

Missile defense could also offer a way to manage certain low-level threats. In the Middle East, for example, several countries are likely to feel threatened by a growing Iranian missile capacity. Yet a massive, bolt-from-the-blue attack seems unlikely. Much more likely is the implicit threat embodied in fielding a few dozen missiles which might or might not contain WMD warheads. Moving defensive missile forces into a region during a crisis could offer a way to respond with great political significance.

Of course, any such missile defense system can be overwhelmed. But this compels an enemy into a more massive type of attack that significantly increases the likelihood of a retaliatory U.S. response. Missile defense links active protection of an allied nation’s population to the likelihood of triggering the American security guarantee. The larger the attack, the more probable is a U.S. response.

Missile defense, conceived as central to arms control, offers a way for important allies like Japan, Canada, and those in Europe to cooperate with the United States. If missile defense is seen, or worse, if it is positioned, as “anti-arms control”-based on the view that the abm Treaty was a pillar of arms control-it makes cooperation difficult. Binding arms control to missile defense creates an entirely new way to look at the subject, one that is much easier to justify to allies. And ironically, a weapon banned by arms control in the first nuclear age might become a positive, stabilizing cornerstone in the second.

Conclusions

This paper is not an argument for junking arms-control treaties. Where they make sense, multilateral treaties play an important role and should be strengthened as appropriate. The problem is that arms control in recent years has come to be identified exclusively with this approach and with the notions of cooperative security that go along with it. If multilateral treaties had made nuclear arms irrelevant in the world, and if the risks of war were not increasing in key regions, it would make perfect sense to stick with it.

But this hasn’t happened. More active, innovative, and forward-thinking arms-control approaches are needed— in conjunction with treaties— to manage the dangers of the second nuclear age.

You may forward this email as you like provided that you send it in its entirety, attribute it to the Foreign Policy Research Institute, and include our web address (www.fpri.org). If you post it on a mailing list, please contact FPRI with the name, location, purpose, and number of recipients of the mailing list.

If you receive this as a forward and would like to be placed directly on our mailing lists, send email to FPRI@fpri.org. Include your name, address, and affiliation. For further information, contact Alan Luxenberg at (215) 732-3774 x105.