The negotiation of the START treaties took place not so many years ago, but most of the original cast of these dramas has moved on to new roles and others have taken their place. Unlike the quick action taken on the INF Treaty, the entry into force of the START treaties was not immediate. During many months of rapid change, this delay has introduced to the contemporary stage a significant number of new players. For that reason, I would like to concentrate my remarks less on the debates in their current style than on the ideas which inspired us in the past and the visions we had then of the future. My assigned task of looking at the relationship of strategic offensive and defensive weapons systems in the context both of further arms reductions and of the proliferation of weapons of mass destruction is actually facilitated by this distance from the current debate.

The Early History of Offense and Defense

Just as arms control is inseparable from national security, so offense and defense are inseparable in the consideration of military strategy. This has always been so. Throughout the history of warfare, one can see periods in which tactics or technology favored the offense or favored the defense, but some optimal mix evolved in each era. Offensive action could apply force for political gain, but it could also be used for defense or for retribution against aggression. Defenses could blunt an attack, but adopting defensive positions on part of the battle front was also a means for both aggressor and victim to concentrate their forces elsewhere. Along with this economy of force role, defenses also provided early warning and attack assessment as each sought to stage decisive action on its own terms.

Even in the age of great fortifications, when the defense was said to be dominant, defensive operations served primarily to delay, dissipate, and channel an attack to a time and location where the advancing forces would be at a disadvantage. The successful defenders of great castles may have, on a few occasions, actually engaged in little combat from behind the protection of their ramparts before a siege was lifted. Ultimately, however, they had to sally forth to reclaim their land after exhaustion, attrition, or fear of diminished prospects for victory had caused the attacker to fall back on its own defenses. Indeed, aggression abroad was not often risked without secure fortifications at home.

This is not to say that the balance between offense and defense has no bearing on the likelihood and intensity of war. It does. During the age of the great fortified cities in Europe warfare was still frequent, but usually limited and highly ritualized with rules of engagement which minimized casualties. As trench warfare demonstrated in World War I, however, increased use of defensive tactics did not always mean that the loss of life was minimized. Likewise, in the world's military histories, bold offensive action is as much associated with limited casualties as it is with massive slaughter and long periods of peace were associated with powerful empires which tolerated no resistance.

In short, strong defenses could be both stabilizing and essential to sound military doctrine, but the price of war was determined more by the causes of conflict, the character of man, and the correlation of forces than by the mere preference of offense or defense dominance. And, finally, although defensive action always played some role, the offense or threat of it brought hostilities to an end. This "spirit of the offense" came to dominate military thinking in the age of Clausewitz. As technology has made weapons more and more destructive, this concept of war as an extension of rational political competition was frequently combined with a more pacific notion that weapons had become so horrible that rational war could not be contemplated. Nobel's dynamite, artillery, the machine gun, the submarine, the Zeppelin, the airplane, poison gas, however, all proved insufficiently horrible to guarantee peace.

This reflection of the extension of violence as the heart of warfare rather than as the basis for peace has inspired many commentators to prefer defense dominance, indeed, to advocate worlds in which all states would have a minimum of offensive force relative to the defenses of their neighbors. In some cases, this distinction between offensive and defensive force has been carried over into distinctions among weapons. One can read of armies that went to war with only swords. One does not read of armies going to war with only shields. One can understand a logic for peace in which the former would be banned and the latter become a safeguard against aggression.

The necessary distinction, however, has not stood the test of time for a number of reasons. Certainly, few defensive weapons have no offensive capability. The soldier with only a shield may sling it at his enemy or use it as a bludgeon. Infantrymen even distinguish between offensive and defensive hand grenades (actually, the offensive grenade has less shrapnel because it is used by troops moving in the open against troops confined in bunkers and foxholes). Second, defensive arms like defense itself serve to complement the offense. Thus, traditional military strategy has also required a mix of weapons which were either predominantly offensive or defensive.

The coming of the thermonuclear age reopened this debate once more. Early on, fear of the society-destroying capability of nuclear weapons led to great investments in air defenses to defeat aircraft armed with nuclear weapons. Defensive interceptors themselves were even armed with nuclear weapons. Early declaratory nuclear policies stressed damage limitation, but defenses against ballistic missiles fell well behind the accumulation of huge arsenals of nuclear warheads on the intercontinental ballistic missiles of the superpowers although perhaps not behind those of lesser nuclear powers such as China. The absence of large-scale defenses in the face of overwhelming offensive nuclear capability highlighted the ultimate vulnerability of both sides. The expense of nation-wide defenses to counter such large threats and the certainty that they would not be leak proof increased pressures to limit offensive arms. In this context, the United States and the Soviet Union began their negotiations on strategic arms limitations (SALT).

The centerpiece of the SALT I package in 1972, however, was the Anti-Ballistic Missile (ABM) Treaty, a treaty which itself limited defensive not offensive arms. The ABM Treaty was justified through argumentation that mutual vulnerability was stabilizing. Although the original goal of a treaty capping the growth of offensive arms was not achieved, an Interim Agreement on offensive arms did limit numbers of silo launchers, submarine launch tubes, and even ballistic missile submarines. It did not limit warheads, however, but the existence of the new ABM Treaty was said to reduce incentives to deploy more warheads. This incentive was sweetened when the 1974 Protocol to the ABM Treaty halved the number of permitted defensive interceptors and deployment sites and also when the United States closed its only ABM site a few months after it had finally become operational.

Interestingly, during the initial SALT negotiations, it was the Soviet Union, far more than the United States, that questioned why one would want to limit defenses. And it was the United States which stressed linkage between the future of the ABM Treaty and further reductions in nuclear arms, albeit, in the opposite direction from that Moscow has proposed in resent years. Yet, the consequent SALT II, like SALT I, permitted and codified a massive increase in strategic warheads despite the scarcity of ABM systems and despite the emergence of large numbers of gray area theater nuclear weapons such as the Soviet SS-20 Intermediate Range Ballistic Missile and the Backfire bomber.

As NATO prepared to respond to the SS-20 with its own INF missiles, the West became polarized over nuclear modernization. At the risk of some oversimplification, one could say that one school believed that enough was enough whatever the Soviet Union had. The other school sought to redress the imbalance it perceived. The first school became supportive of a freeze on modernization. The second group proposed a dual track of modernization and the negotiations of reductions to enhance stability. The debate was over offensive arms. Both sides advocated fewer, although they disagreed on how to achieve their goal.

At the height of the nuclear freeze movement, I participated in a debate in a church in San Antonio, Texas. The Speaker of the U.S. House of Representatives at that time, himself a Texan, had just appeared and announced his support for a nuclear freeze. I was a junior U.S. government official defending the NATO deployments against the freeze when my debate opponent, a retired U.S. Army major general, changed the subject briefly. What the world really needed, he said, was defenses against missiles. The audience, clearly in favor of the freeze, roared in approval of strategic defenses. This was some weeks before President Reagan announced his Strategic Defense Initiative in March, 1983. The freeze debate faded away as the United States revisited the question of the role of defenses. Political polarization did not disappear, but new constellations of vociferous advocates and opponents did appeared including hawks together with doves on each side of the issue--Edward Teller and Freeman Dyson favoring defenses, while mainstream thinkers and even the uniformed military seemed split on the issue.

The Debates in the 1980s

The debates of the 1980s were fascinating, although initially there was confusion, misinformation, and rhetoric on both sides of the question. Sometimes there was not much clear thinking even on the theoretical level. Let me give you just one example, the debate over Mutual Assured Destruction (MAD) versus Mutual Assured Survival (MAS), again at the risk of oversimplification. If you took the people who thought they favored each of those positions, set them down, and asked what nuclear targeting doctrine was associated with their concept of defenses, the most common answer for both sides was countervalue targeting, or as some would say, city-busting. Absent absolutely leak proof defenses, both sides were still talking about populations being targeted with tremendous loss of life and destruction. Those who favored defenses were arguing, in essence, that defenses might save millions of lives. Those opposed to defenses favored greater certainty of the most massive destruction to enhance deterrence. The bottomline for both sides was an emphasis on the targeting of population per se.

For much of the national security community, however, the focus was different. That community recognized the ultimate countervalue effects of a strategic nuclear exchange, but this community focused more on its own differences, differences concerning the impact of strategic defenses on the military balance and thus stability. Here most experts also fell into two schools. One school basically believed defenses favor the aggressor. Here's why: He who launches his missiles first will overwhelm an opponent's defense with numbers. If an aggressor conducts a disarming first strike against an adversary's retaliatory force, and the remnant of that retaliatory force then faces the alerted defenses of the aggressor, the aggressor has gained leverage in both offense and defense. Hundreds of computer runs were made based upon this assumption. Thus, they often concluded that even if the offense and defense were equal and symmetrical on both sides, defenses would be destabilizing.

On the other side of this issue, experts were doing their computer runs. And their approach was different: "He who shoots first in order to disarm has a harder targeting requirement than he who simply must retaliate in order to inflict unacceptable pain." If the initiator of the war must have high confidence of counterforce success in detail to avoid unacceptable retribution, defenses can so complicate the disarming first strike that under almost all calculations, they are stabilizing.

In summary, the nuclear policy debate in the 1980s seemed bogged down in debates over perfection. The primary public debate concerned whether anything less than perfect defense was sufficient--that is, whether to defend anyone if everyone could not be defended, and against every threat. The primary debate among defense intellectuals was whether even the most imperfect defenses might encourage too much nuclear self confidence to be stabilizing.

Compromise Conceptualized

In the middle of this debate, the United States was confronted by the Soviet Union in bilateral negotiations even as research and development programs were going forward. When the nuclear arms talks resumed in Geneva in 1985, the Soviet Union sought linkage between the INF issue, the START issue, and the so-called Defense and Space issue. The United States recognized that there were interrelationships, but did not want any one negotiation held hostage to another. The United States and the Soviet Union agreed to a format that covered both sides' interpretations of what the proper relationship should be. This resulted in odd shaped tables and strange protocols. The gist of the Soviet Union's position was that there could be no START Treaty unless the United States solved Soviet problems with INF and Defense and Space. Early on, we were able to agree to proceed with the INF Treaty, and later we were able to work out a form of delinkage on START.

I want to remind everyone in the room, however, that the U.S. position was always that everything was interrelated. The U.S. did not think there ought to be any formal linkage of agreements, but in fact, in the context of those negotiations, the Reagan Administration at various times had conceptualized a number of compromises across agreements. The U.S. position usually stressed delinkage of most issues, but the U.S. position sometines included variations of the so- called "grand compromise"-- "you give us something on offense, and we'll give you something on defense." Sometimes, the U.S. position also had certain aspects of what I call the "green light" compromise, according to which the United States would not accept certain provisions in an offensive agreement unless it were given something favoring defenses." One finds examples of all of these approaches in the U.S. negotiating position, sometimes all at the same time.

GPALS Initiative in 1991

New political circumstances, geopolitical agreements, and strategic calculations appeared rapidly at the end of the 1980s. In January 1991, in the context of improved relations between the United States and the Soviet Union, President Bush proposed a different, much more limited approach to strategic defenses. This lead people to rethink what would be needed in the post Cold War era to enhance security and still have a stable relationship with the then Soviet Union. President Bush proposed the so- called GPALS system (Global Protection Against Limited Strikes) which was downscaled tremendously from the Phase I Joint Chiefs of Staff requirements for the original Strategic Defense Initiative, Those requirements, at least in their original absolute numbers, had already been achieved by START I.

Cooperation on Defenses;
ABM Revision Talks- - 1992

The initiation of the GPALS program was followed by a series of rapid and major international developments. The completion of START I and a few weeks later the failed coup in Moscow suggested that cooperation between Washington and Moscow should be enhanced. In September 1991, President Bush called for cooperation on defenses. The United States also announced that it would be eliminating all of its tactical nuclear artillery and many other tactical nuclear weapons. In October of 1991, President Gorbachev announced that he too believed cooperation on defenses should be discussed. By January of the next year, President Yeltsin of Russia made a bold proposal that the United States and Russia work together to bring about a cooperative, global protective system. In that same positive environment, the United States agreed to talk also about START II, a step Washington had believed was premature prior to the recent political changes. START II began to weigh very heavily upon everything we were doing. START II was seen as a way of strengthening the foundation for a cooperative future both in limiting offensive arms and cooperative defenses. Obviously some form of interrelationship would emerge, given the history of the negotiations and also the new opportunities for cooperation.

In the final statement of the June 1992 Summit, Russia and the United States agreed that a group of experts, the so-called Ross-Mamedov group, would discuss cooperation on early warning, cooperation on technologies for defense, nonproliferation, and the legal basis for a Global Protection System (GPS), including any changes which might be necessary to retain the existing treaties, including the ABM Treaty. (Note that the United States had decided to adopt the Russian name, or GPS).

The most important Ross- Mamedov session was probably that of September 1992. At that meeting, on behalf of the United States, I presented the case to the Russian delegation for amending the ABM Treaty. Ambassador Robert Joseph subsequently presented this proposal in its detail at the Standing Consultative Commission. The U.S. view was that circumstances have changed, politically and technologically, and that we now have an opportunity for a new relationship. An important part of this relationship is rethinking the question of whether we should begin cooperating in defending both of our populations, rather than collaborating to maximize their vulnerability. We talked about what we thought needed to be done about early warning, technology cooperation, and nonproliferation. We accepted and emphasized a multifaceted approach to the problem.

We made clear that defenses would play an important role in the future, and we made specific proposals to amend the ABM Treaty. We proposed that it permit more than the 200 interceptors that were permitted by the original ABM Treaty. As I highlighted in my remarks at the time, the ABM treaty does not ban defenses. In fact, it explicitly provides, as signed in 1972, for 200 interceptors, plus additional test sites. Thus, in its original form it already envisioned as many as perhaps four or more places where a country might have interceptors, although only two of those were to be operational deployment sites.

We talked about the changes in technology which made it increasingly difficult to maintain distinctions between early warning, command and control, surface- to- air missiles and theater ATBMs on the one hand and similar ABM systems on the other hand. We stressed the need to look at the whole- - at what a BMD system really is. The inevitable increase in the capabilities of non-ABM systems was feeding ever more contentious debates over distinctions that were also very difficult to verify. The electronics revolution is radically altering the meaning of many of the boundaries sought by the ABM Treaty. This led the United States to propose that sensors run free- - that we would agree that with respect to sensors, since they're so important for so may vital functions such as early warning, national technical means of verification, and conventional forces, not to make them an issue between us.

With respect to numbers, of course we had a position proposing several hundred ground based interceptors. I should note that Russia has 100 interceptors already while the United States has none. The United States was willing to forego a decision on the question of space- based interceptors, if we achieved an agreement for near term ground- based systems along our line of several hundred- - maybe six, seven, eight hundred- - not that far from the Russian number which was 100 and not far from the 200 permitted by the ABM Treaty in 1972. So in a sense, we were haggling about the numbers, although we had in mind a certain level of effectiveness that we wanted to achieve by the technologies that we had available. That level of effectiveness seemed compatible also with the Russian concept of a Global Protective System

Discussion of amending the ABM Treaty was complicated also by the changes in the political circumstances of that time. One signatory to the bilateral treaty, the Soviet Union was gone, and the existing ABM system of the former Soviet Union no longer was solely within the sovereign bounds of a single country. There were a series of basic fixes to the ABM Treaty that we thought would be necessary to make it viable and effective, and our position was that we were prepared to do this, in the context of getting an agreement on defenses that was in the interest of both sides. This history demonstrates that the United States did engage very specifically on how to work together with Russia in the context of the ABM Treaty. Circumstances had changed. The ABM Treaty was broken, but the United States was prepared to agree to fix it if it in the context of cooperation on defenses.

Defenses and Further Offensive Reductions: The Legacy of Reykjavik

Permit me now to jump to the future. Increasingly, as we approach the millennium, in the context of the NPT extension, we are hearing more and more about attempting to go to zero nuclear warheads, or to very low numbers. And emerging again and again in the debate, and not on a partisan or ideological basis, is the view that you cannot go to deep reductions without defenses. This was actually was one of the key issues at Reykjavik, and what the debate over what was proposed at Reykjavik was really all about.

We have already given much thought about the offense- defense relationship, but we need to get beyond frozen positions. I have tried to give you a sense of some of the key initiatives from the past which were designed to get us beyond stalemate. Today, as we try to go beyond linear thinking about how you safely move towards further reductions, traditional patterns of partisan politics and ideological splits are starting to fragment. So, perhaps it would be a good idea if everyone engaged on the issue of the offense-defense relationship revisit the question through a fresh process. We should revisit our assumptions, determine the real constants and variables for our age, and think anew. To do that, we will have to put aside our current mindsets, our current coalitions, and our current interest groups to determine if there isn't a path which brings us together.

Current Trends: Four Assertions

With this discussion of past and future as a foundation, let me turn to the question of the present just briefly. It isn't my primary focus, but I want to make four assertions about the present in reaction to what I have heard here and in Moscow recently. These are four assertions you can accept or reject.

First, if it were left to the U.S. and Russian military, START II would have entered into force already.

Second, if available material resources, i. e., budgets, were comparable on both sides, the ABM Treaty would not be as big an issue as it is today. There is actually a strong latent view within Russia that it ought to have defenses against ballistic missiles. In fact, they do. They have 100 ABM interceptors.

Third, unfortunately- - and I hope not increasingly, many of the issues that are being raised about START II and the ABM Treaty are really being used as vehicles for expressing uncertainty about the geo-strategic future, uncertainty about where we, the United States and Russia, are in our relationship to each other. This includes also uncertainty about where we think we ought to be. We need to answer the question of what it means to say the Cold War is over.

Fourth, the substantive uncertainties about the ABM Treaty or START II are really being greatly amplified by contextual uncertainties, most of them of a domestic political nature. We have important new or reinvented players in Washington and Moscow. Some of them know these issues well, but many do not. There is a tendency to see many decisions made on the basis of a simple interrogation: "If my domestic opponent is in favor of it, I must be against it," or vice versa. We have a similar problem on the international front to which I alluded earlier; namely, that whatever you think of the arguments on their merits, the legacy of the ABM Treaty and the legacy of Cold War deterrence debate are giving us vocabulary that is not always helpful, as we try to discuss a proper U.S.- Soviet relationship. In a way, our very words, including words I've used today such as a "mutual hostage relationship", poison the water. We need fresh language reflecting our real objectives, language which doesn't carry so much baggage.

We're experiencing manifestations of the "Ifft rule." Ed Ifft is famous for saying, "it's not that our positions are different, it's that they're the same at different times." Some believe that this is a description of a fickle or frivolous basis for negotiations. I don't interpret the rule that way. Rather, it reflects the reality that as circumstances change, what we should do can change.

If you go back to the mid- 1980s, for example, the Soviet Union put out many feelers to see if we would be willing to settle the ABM dispute by agreeing to 200 ground based interceptors--or 300, or 400. And it was in the United States that voices said, "Wait a minute, we'll never get an environmental impact statement through. Our future is in space. This is a Soviet trap to get us to try to deploy some missiles that we can't deploy politically while they build a large ground based system. We will lose." Our positions have been the same at different times, but there remains in the domestic debate today in Russia and the United States, the Cold War remnant of, "if it's good for the other side, it must be bad for us." Again, we need to find a way to break out of that mindset.

START II Compromises

When I first became active in arms control negotiations, the one fundamental rule about domestic politics was that you never took a treaty to Capitol Hill in election year. But in 1987, we broke the rule. It wasn't all that easy, but it wasn't all that hard. We got the INF Treaty ratified. Here we are again, in a much more difficult world, in the middle of an election year in Russia as well as in the United States. And friends of mine in Russia say to me, "Well the problem is that START II was negotiated from weakness, and our side gave too much to you." I remember it a little differently, however. In fact, I remember how much we gave to the Russian side that would have been unthinkable in previous years. I think about the separate SLBM limit that we'd never agreed to before, the bomber counting rules which reversed a fundamental U.S. approach to stability. I think of the intrusive inspection of bomber bases and special limits on bombers, and how, again and again, on issues like the SS- 19, silos dismantlement, and simplified verification we allowed issues to be reopened in order to address Russian concerns.

We used to say there could not be further reductions until after START I had entered into force and after vast new improvements in verification were achieved. Instead, at Russian insistence, we agreed to act almost instantaneously on START II and, basically, to use the START I verification rules. It was in the interests of both countries for us to exhibit this flexibility, but these concessions, or compromises, or flexibility by the United States, would not have taken place in fact, if the situation had not changed in Russia. If the previous regime had been in power in Moscow, we probably never would have shown that flexibility. There probably would not have been a START II Treaty.

So, when you think about the START II Treaty, remember that the United States was actually very forthcoming. We thought it was important to a new, better relationship. And if we were wrong, that's going to have tremendous impact at home and aboad. Yes, Russia is having an election, but so is the United States. In this election year, both sides need to be very, very careful. To our Russia colleagues, I would say don't ask our president to go to the Congress and to look as if he's cutting deals with a foreign government blocking the aspirations of the elected officials of the United States. The Congress expects the president to come to them and to work out a united U.S. position. The Congress expects him to work together during negotiation of that position. Neither Russia nor the United States will gain from an end run of their own political processes. At a minimum, there must be a very close consultative process.

Recommendations

What is my recommendation? I think we need to do some new thinking in a less polarized way that brings all the players, including some new players, to this process. There are certain things that our countries have agreed already to do. Let's do them. START II is, I think, essential. If we want to keep our relationship on track, moving in the right direction, START II must enter into force. We can and should, however, commit to a fresh look at the questions related to offenses and defenses. This probably ought to be done after both countries' elections.

This new process probably ought not be a negotiation initially, or a formal government- to- government process by itself. It may require a Track II process, and it should have a certain number of legislators from the United States and Russia. An informal process- - perhaps initially off the record and anonymous- - is necessary. Opinion leaders with diverse views must rethink these questions of what we mean by "the end of the Cold War" and what we should do about offense and defense after the Cold War. How do we think about balancing weapons if the Cold War is really over, and how do we get beyond that? If we can't do that, we're in for trouble.

A Prediction

Now, let me make one prediction about the future. My own view is that further defenses will be deployed. They're already deployed in Russia. They will be deployed in the United States. Putting together the coalition necessary will take longer than advocates recommend, and this will continue to result in greater development costs. The operational system itself, however, will inevitably cost less, not more than has long been assumed. National missile defense will cost less than what many people think because smaller threats are of increasing urgency and because dual-use technologies which leverage defense are advancing. The world of electronics is going in a direction that drives many defense associated costs down.

The decision to deploy nationwide defenses, however, will not be made in Moscow or Washington based upon an accountant's estimate of affordability. It will be made when citizens demand that they be defended. The event that will probably cause this to happen may not even have anything to do with Russia, and it may not be based on an initial threat against the United States. It may well be that theater ballistic missiles, armed with a weapon of mass destruction, strikes someone else's forces or cities. The world will suddenly change the way it evaluates this equation. Much of the current debate will be washed aside by the force of events.

Defenses are not an alternative to a multifaceted approach including reductions, nonproliferation, and controlling smuggling, but my own assessment is that we will be living for some time in a world in which a multifaceted approach is not a substitute for defenses against ballistic missiles. I believe that a new look undertaken without the blinders of past political divisions will reveal that cooperating in defending the people of Russia and the United States against ballistic missiles will be seen as necessary for the security of both and a powerful foundation upon which to build a more viable arms control and non-proliferation regime.