Slocombe - Stability Effects of Limited Missile Defenses

Stability Effects of Limited Missile Defenses:
The Case for the Affirmative

Walter B. Slocombe
Attorney, Caplin & Drysdale, Chartered, Washington, DC, USA
Former Under Secretary of Defense for Policy, US Department of Defense

[Author's Note: This paper was, of course, written before the 11 September attacks. The attacks and their aftermath have underscored the points that the threat of terrorist attacks by non-state actors (with varying degrees of support from nation-states) is both a more urgent problem than that of rogue-state regional aggression and a problem to the solution of which missile defenses are essentially irrelevant. However, neither the attacks, the general problem of global terrorism, nor the increased worldwide effort against terrorism lessen the conceptually and strategically distinct threat to international stability of rogue states armed with missiles and chemical, biological, or nuclear weapons for them. Indeed, the killings on 11 September and from the subsequent anthrax-laden letters (whose source is still entirely unknown as of the date this note was written) serve to remind of the terrible potential costs of a successful state-organized WMD attack. Accordingly, the case for (and indeed the case against) limited ballistic missile defenses is essentially unaffected by the cataclysmic recent events. It does seem possible that greater realization by the US Administration of the value of international support and sympathy - and greater US-Russian cooperation in the aftermath of the attacks - may make more likely the outcome advocated in the paper, that is, development and deployment of limited ballistic missile defenses in the context of US-Russian agreement on an updated framework for their strategic relationship. WBS, 4 November 2001]

The stability effects of missile defenses - or indeed any military measure - can usefully be considered as involving three distinct aspects:

Crisis Effects: Will the existence of missile defenses make it more or less likely that the parties to a crisis will act imprudently, accelerating confrontation out of control?
Arms Race Effects: Will the deployment of missile defenses and, prior to deployment, of a program to develop and deploy them, induce other countries to undertake offsetting military programs of their own that would themselves have negative stability effects, or will it tend to discourage such programs?
Effects on the Political Relationship: Will the existence of missile defenses, or programs to develop and deploy them, worsen or improve general political relationships among nations?

Cold War Analysis

In these terms, the conventional Cold War analysis was straightforward: Missile defenses, at least on a nationwide scale, were a Bad Thing:

In a crisis, they could encourage their possessor to launch a disarming first strike at the other side's strategic forces in the expectation, or at any rate the hope, that the defensive system would be able to defeat a reduced and uncoordinated response by the depleted missile force that survived the initial attack. Correspondingly, the very existence of missile defenses could provide an incentive for those facing them to pre-empt, precisely to avoid the risk of being subjected to such a first strike attempt.
The mere effort of one side to develop defenses would stimulate competitive counter-measures, as the other side took steps to offset the potential advantages it feared its opponent would gain from defenses. These actions might come long in advance of any actual defense becoming operational, and would likely be based on limited, worst-case understanding of the characteristics of the defenses being developed. The overall result would be larger strategic arsenals, development of responsive technologies that were themselves destabilizing, and diversion of resources from other priorities - as well as stimulation of mistrust and conflict.
On the political side, perceiving the opponent as building defenses - with its implication of a design to gain the freedom of action that would come with immunity from retaliation - would confirm general perceptions of irresolvable hostility, and thereby heighten tensions. The potential of defense development programs to stimulate strategic arms competitions could exacerbate these political effects.

It bears noting that, even in Cold War conditions, at least some of these stability effects were more matters of abstract logic and analysis than experience and were most valid where the issue was nationwide defense (including fear of the prospect of such defenses). In terms of crisis stability, for example, the only missile defense programs that ever reached operational capability - the Soviet system around Moscow and the US system near an ICBM field in the northern Midwest - were designed and described as being for protection of command and control and of potentially vulnerable retaliatory forces respectively -both objectives that contribute to crisis stability. Similarly, while some of the US and Soviet programmatic responses to fears about the other's potential defensive systems were themselves sources of instability (notably the deployment of multiple warhead payloads for missiles because of the leverage afforded by a successful preemptive strike against them), others, including increasing survivability of offensive forces (hardening of silos and command systems on both sides, shifts to sea-basing in the US case, and shifts to mobile land basing in the Soviet Union) had positive crisis stability effects.

Nor was the impact in terms of fundamental Cold War political relationships necessarily as negative in practice as might have been expected. The fierce strategic competition of the ABM/MIRV buildup of the late 1960s and 70s was concurrent with, and to some degree the result of, what proved in retrospect to be grotesquely exaggerated concerns about missile defenses (the "Tallin"/SA-5 system in the USSR, the Sentinel/Safeguard system in the US). It seems to have had sobering effects. During that era political leaders on both sides with the impeccable Cold War credentials of Nixon and Brezhnev seem to have been convinced that neither side could hope for a decisive advantage even in an all-out strategic arms competition. This perception contributed to opening the door for agreement on limiting defenses and on first restraining and then reversing the buildup of offensive forces. The successive arms control treaties which ensued served a distinctly positive political function by moderating East-West tensions, in addition to making significant contributions to stabilizing the nuclear relationship at a more technical level. This is not to argue that US (or Soviet) ballistic missile defense programs (or any other particular aspects of the strategic buildup of that era) were justified simply because they may have helped induce political leaders to recognize the cost, difficulty and ultimate futility of the competition, and to take difficult decisions fostering mutual restraint. Rather the point is to observe that worsened political relationships are not an inevitable result of the concerns raised by defense developments where political leaders are prepared to consider the issues realistically.

Defenses and Crisis Stability in Post-Cold War Conditions

In terms of the current debate, however, the critical point is less what the potential for defenses meant for stability in the Cold War than to what degree the traditional analysis remains applicable now that the Cold War is decisively over. For these stability effects, at all three levels, came in the particular context of the Cold War, in an international scene characterized by a worldwide confrontation between the US and the USSR and their respective coalitions of allies, clients, and associates. That was not only a world in which two roughly equal, powerfully armed, and fiercely hostile superpowers faced each other, but one in which virtually any local conflict presented a risk of escalation to a confrontation of the nuclear superpowers.

Moreover, for the forty-odd years of Cold War confrontation, there were no significant independent third country players on the nuclear stage. Britain and France - and for that matter Israel -- were firmly in the Western camp. China's alignment was far more ambiguous, but ultimately its nuclear forces - which were (and are) in practice aimed equally at the US and Russia - were relevant only in relationship to their impact on China's relations with the USSR and the US - and theirs with each other.

Changes in the strategic context

It is banal to observe how different that world is from today's, including in those aspects most relevant to questions of the impact of nuclear-armed missiles and defenses against them on stability. The Soviet Union and the international communist movement it led are gone forever, and with their demise the threat they posed to Western societies has also died. For their part, the US (and NATO) have drastically changed their relationship to Russia, substantially reducing their military forces and realigning their foreign policies, treating Russia as, at worst, a potentially difficult major power, but not an enemy, much less a worldwide threat. To imagine that the US or its European or Asian allies have designs on Russia is to indulge in paranoia, not analysis.

But the nearly two decades that saw the end of the Cold War and eliminated any possibility of its revival have also seen the emergence (or in some cases, the survival) of a succession of smaller conflicts. Most of these, dangerous enough in themselves, are essentially local. However, in a few cases, they have a special character because they involve countries -- usefully, if somewhat clumsily, referred to as "rogue states" -- that operate outside normal international standards and aspire to regional dominance or have conflicts with neighbors that threaten broader interests. Without exception, these countries are in the process of trying to develop new military resources, including efforts to acquire the most advanced technological systems, such as high performance aircraft and precision-guided conventional munitions, cruise and ballistic missiles, highly sophisticated air and maritime defense systems, and capabilities for computer network attack.¹

The special problem of rogue State programs to deploy long-range missiles and nuclear (or biological or chemical) warheads for them

For the most part, these new military capabilities are chiefly threats to regional neighbors (and, of course, to the forces of any outside state seeking to assist those neighbors in their defense). However, in three important cases - North Korea, Iran, and (in a state of sanctions-induced suspension) Iraq -- these programs of technological buildup already include development of very long range ballistic missiles and of nuclear, chemical, and/or biological payloads for them.

The purpose of these capabilities is neither of the two classic Cold War nuclear missions -- deterrence of nuclear attack nor preparation to launch a deliberate pre-emptive disarming attack. Rather, the purpose is to use the threat of nuclear (or other WMD) missile attack as leverage against outside intervention against regional aggression.² A state possessing such capability and considering an attack on a neighbor might convince itself that it could, by brandishing its long-range nuclear armed missile capability, either prevent or sharply limit the support its immediate victims would receive from the US and other extra-regional powers.

At the moment, North Korea, Iran, and Iraq are the only nations that combine potential for regional aggression and rejection of normal international norms with clear evidence of a determination to develop long-range missiles and WMD for them to carry.³ However, it is impossible to be confident that these three exhaust the list of possible problems of this character. Given the political uncertainties of much of the world, and the increasing availability of the relevant technologies, it would be a brave forecaster who could be confident that, over the next decade or so, there will not be new "rogue states" to add to the list, perhaps a "Talibanized" regime come to power in one of the several states in the Islamic world that now combine massive internal problems, extraordinary corruption and weakness in the governing regime, and frustrated national pride and ambition to offer a fertile field for a takeover by such a regime.

Since regional aggression threatens broader international security, the rogue state problem is a security and stability concern, not just for the US, its closest allies, and the prospective immediate targets, but for the world community - and for Russia, which has at least as much long-term potential for tension with the rogue states to its south as does the US and is in a considerably more exposed geographical position with respect to them.

Elements of a response to the rogue State problem

To be sure, the larger part of the response to the "rogue states" problem has nothing to do with missile defense. It is one thing to maintain that limited ballistic missile defenses are a prudent measure in response to the rogue state problem; it is quite another to argue that they can be a complete answer. Indeed, it would seem that those who argue most strongly the case that the "rogue state" regimes are so dangerous a problem as to require missile defense against them have a particular responsibility to support every plausible avenue of diplomatic and political resolution, if they are to be consistent in arguing the urgency of the basic problem. There is, in all three immediate cases, a hope for an agreed and peaceful transition out of the current problems - the "sunshine policy" in the case of North Korea, the promise that pressures for internal reform in Iran will eventually lead to a moderated international policy, and the possibility that a different regime will come to power in Iraq that would realize that genuine compliance with the post-Gulf War UN resolutions is in Iraq's interest.

Moreover, even if political efforts fail, or do not succeed quickly, there is much besides building ballistic missile defenses to be done about the rogue state problem. In particular, tightened controls on exports of relevant technologies (and in the case of Iraq, maintenance of UN sanctions) can still be of some help in restraining the missile and the related nuclear/biological/chemical weapons programs (and can be very important in preventing new problems from emerging). And there are military answers as well - including clear commitments to assist in the defense of targets of aggression by rogue states, and maintenance of both the military capabilities and the political coalitions that are essential to such defense. The prospect that a direct defense against conventional attack would succeed is itself a deterrent to aggression. Consequently, maintenance of theater military capabilities, including defenses against theater-range ballistic and cruise missiles, makes aggression less likely.

Finally, with respect to responses to the rogue state problem, whatever else the term "rogue state" is intended to mean, it certainly does not imply that the leaders of these nations, or those internal elements in the security forces and political system whose continuing support is essential to their power, are immune to the influence of deterrence by the threat of retaliation. It is generally believed, for example, that Saddam Hussein was influenced not to repeat in the Gulf War his use of chemical weapons in the Iran-Iraq war (and in his internal control efforts) by the US warning of overwhelming response with its clear, if veiled, threats of use of nuclear weapons.

All that said, there is still a reasonable possibility that deterrence could fail, and therefore a case for developing and deploying defenses capable of defeating the sort of limited missile attack of which a rogue state might be capable.⁴ The architecture of such a defense is still very much uncertain, and the exact scale and pace of the program still more so, for technical, financial, and political reasons. However, there is little doubt that the US will in fact proceed over the next five or ten years to deploy a limited national missile defense, probably based on some variant of the ground-based system currently in test, meanwhile conducting an expanded program of research on other defense technologies.

Stability Considerations

In terms of stability, therefore, the issue is whether such defenses would have, on balance, sufficiently large positive effects, and sufficiently small and controllable negative effects to be advisable. In particular, it is necessary to consider whether positive effects on the "rogue state" problem would be offset by negative effects on the relationship between the nuclear superpowers.

It is a critical element of the current issue that the defenses that the US is now considering are, as even their most avid advocates acknowledge, of a distinctly limited character - particularly compared to the SDI-Star Wars concepts of the 1980s. Whatever their technological content, they would be designed to deal only with the kind of relatively small and unsophisticated missile forces that rogue states must be expected to be able to deploy within the next decade or two. The stated objective of the US system is to have high confidence of defeating an attack by a few tens of missiles, equipped with relatively limited anti-BMD countermeasures. (There is, of course, a substantial debate about the technical feasibility of the programs underway achieving even this limited goal. The merits of that argument are beyond the scope of this note. However, for purposes of analysis of the stability effects, it is appropriate to assume that the defense will in fact be capable of doing what it is supposed to do.) Such defenses have very different stability implications from the massive national-scale defenses that were the focus of concern in Cold War conditions.

As for Crisis Stability, the most important implication is the potential of such defenses to reduce the risk of crisis produced by regional aggression. The purpose of rogue state long range nuclear armed missiles is not, obviously, to attempt to disarm the strategic forces of the US (or, mutatis mutandis, Russia) but rather to serve as leverage, in support of regional aggression, by using the threat of attack on distant intervenors' homelands to block or limit the assistance given to the immediate victims of aggression. Correspondingly, the purpose of defenses able to negate such attacks is to remove this source of leverage or blackmail. Defenses would thereby supplement deterrence by threat of retaliation.⁵

In some sense, the most dangerous situation for international stability would arise if a local aggressor mistakenly believed his missiles would deter an effective resistance, and went ahead and attacked. Even if neither his missiles nor the other side's missile defenses were ever used (because, e.g., deterrence worked, or the intervenors limited their response to less than the attacker's threshold for an actual missile attack⁶), the consequence would be a costly and avoidable war. Insofar as limited ballistic missile defenses would make major regional aggression less likely, they would contribute importantly to avoidance of precisely those crises which are overwhelmingly most likely under current and foreseeable conditions.

Nor would limited defenses have significant negative effects in the wildly unlikely event of a US-Russian crisis in which nuclear options were seriously considered. Even assuming very large cuts in Russian strategic nuclear forces - whether as a result of economic pressures, agreed arms control measures, or adjustment of Russian military priorities toward rebuilding its conventional capability - Russia will have retaliatory forces that would totally overwhelm any plausible US defense.⁷ It is no exaggeration to say that the 20 or so warheads a US defense could intercept - even if it worked perfectly - are dwarfed by uncertainties about Russian weapon reliability, alert levels, and schedules for taking weapons off-line for maintenance, testing, and training. Moreover, Russia will - precisely because of the scale and sophistication of its strategic arsenal and the infrastructure that supports it - have options for tactical and technological measures to defeat limited defenses that no third nuclear power -much less a rogue state - could dream of.

The situation with respect to arms race stability is more complex.⁸ Against the defenses the US is actually considering deploying, Russia⁹ has no need to take offsetting measures - and any measures that were at all plausibly related to the scale of the defenses would have no significant international stability effects.¹⁰

It is worth emphasizing that most of the arms race stability concerns about ballistic missile defenses in the Cold War era applied chiefly to efforts to build national-scale defenses, and to the related concern that more limited systems (or even sophisticated air defenses) had some potential to be - or appeared to be - the precursors of such large scale defenses. The ABM Treaty prohibits defense of national territory and contains elaborate provisions against creation of such a defense under the guise of other programs. Nonetheless, it was a critical part of the concept of the Treaty when negotiated that it does not ban defenses per se, but only defense of national territory, and that it explicitly permits defense of national capitals (and the command authorities they contain) and (since 1974, "or") a portion of a nation's land-based offensive missile capability. That such defenses were permitted was the product not only of the fact that the programs involved were well along and supported by powerful interests in the respective superpowers, but also of acknowledgment that they did not represent any significant threat to the other side and were, in fact, conceptually consistent with a mutual recognition of the inevitability of mutual vulnerability.

The more plausible concern in terms of potential Russian programmatic actions arises from the potential of any BMD program to be seen as the precursor of a broader system. Indeed, it is hard to believe that any informed Russian analyst has any doubt about the insignificance for Russia's deterrent capability of the kinds of defenses the US is actually talking about deploying. American advocates of limited missile defense should however acknowledge the more understandable Russian fear that once the US commits to a partial defense, it will inevitably proceed to technologies and scales of deployment that could conceivably put Russian retaliatory capability at risk.

Cold War secrecy and suspicion tended to produce over-reaction to the other side's strategic initiatives for defenses and otherwise, and it can be genuinely difficult from the outside to determine the intended scale and scope of any development program. There are, however, new opportunities now to deal with this problem. Given the breakthroughs already achieved in verification measures in the START, INF, CFE, and other agreements and the potential for still more extensive verification measures, there are ample means to prevent such effects in the very different political context of today. Moreover, the time-scale and technological challenges of developing and deploying missile defenses are such that there will be ample time for Russia to assess the actual character of US actions and not rely on worst-case hypotheses.

In short, the answer to suspicions about the long-term course of the US program is not to persist in the hope of stopping all US NMD efforts, but to agree on a new framework of measures in which that work will take place. That framework would permit limited national defenses and allow development work on new concepts, but also set limits that will assure that those efforts do not threaten the Russian deterrent and provide measures of transparency and cooperation to the same end. The objective should be not to attempt to limit defenses to a particular architecture for all time,¹¹ but to insure that what is actually deployed at any stage does not endanger the US-Russia strategic relationship. Whether that new framework should be, in form, a modification of the existing ABM Treaty, or a wholly new agreement, remains open -and indeed the relative merits of formally legally binding agreements as contrasted to more flexible politically binding agreements are very much matters of detail, not fundamental principle. What is fundamental is that there is still a need for a framework, but that framework must reflect new threats and new conditions.

The issue of long-term intentions lies at the heart of the effects of limited ballistic missile defenses on the stability of political relations as well. It is an understatement to observe that the missile defense issue has become a major irritant in US-Russian political relations.

Russia, on the one hand, sees the US program as a potential threat, not just to its deterrent, but to its status as the strategic nuclear equal of the US. Concerns for the long term shape of the US effort are, of course, increased in so far as advocates of NMD in the US maintain not just that the 1972 ABM Treaty cannot be allowed to give Russia a veto over the US decision, but that the sooner the US withdraws from the Treaty the better and that no agreed limits on defense development and deployment -and perhaps no arms control at all - are still relevant in today's conditions. If Russia were to conclude that the US is determined to build missile defenses without limit and to abandon the very concept of an agreed US-Russian framework of their strategic relationship, the consequences for the overall political relationship could be very far-reachingly negative. Conversely, if Russia were to persist in refusing to make adjustments necessary to permit the US to build limited defenses against new threats, that would have serious negative political effects as well.

There is an interesting analogy between the US position now on the ABM Treaty and the Russian position in the mid-90s on the CFE Treaty's limits on Russian deployments in the Caucasus. Russia argued that it faced genuine security threats in that region that required certain levels of military deployments in the parts of the Russian Federation subject to the Flank limits. Russia acknowledged that those deployments were inconsistent with the strict terms of the Flank limits, but argued that the political and strategic situation had changed in fundamental ways since the limits had been agreed to, and asserted that the measures it proposed to take were in no way inconsistent with the basic spirit and purpose of the CFE Treaty, only with certain terms whose effects were totally different from those contemplated when they were agreed. Rather than seeking to scrap the CFE Treaty altogether, Russia offered the other CFE partners the option to adjust the treaty to meet Russia's requirements in this specific respect - which, it argued, could be done without violence to the basic concept of the CFE Treaty, the security interests of the other parties, or their reasonable expectations of the Treaty's effects when it was signed. The other CFE participants, with varying degrees of reluctance, acknowledged that the Russian concerns were real and that it was neither reasonable nor in the other parties' long term interest to risk the collapse of the whole CFE structure by trying to insist on the original terms when there was an alternative of the limited adjustments Russia proposed. The result was a protracted and difficult, but eventually successful negotiation that adjusted the CFE Flanks regime to preserve its essence while permitting Russia to do what it judged necessary for its security interests. The political lesson is that when an arms control agreement conflicts with fundamental security concerns of a party and can be modified to accommodate those concerns without significant impact on the security interests of other parties, it is not only the party seeking modification, but the other parties as well, that have an interest (and from an international perspective, an obligation) to seek accommodation and adjustment, avoiding the extremes of insistence on the strict letter of past agreements reached in very different conditions on the one hand, or total destruction of the basic framework on the other.

Central to managing the problem so as to address the reality of the rogue state issue while also addressing the reality of the potential of defenses to poison general US-Russia relations is recognition of the limited nature of what is at stake:

For the US, a defense against rogue state missile blackmail is perceived as a high national security priority. Building limited defenses for the US and its allies does not, in itself, threaten any critical Russian interest.
For Russia, the equivalent vital interest is maintaining both its nuclear deterrent in technical terms and its nuclear superpower status in broader strategic terms.
For the world at large, as well as for the two nations directly concerned, preserving the principle that the US-Russian nuclear relationship proceeds in an agreed framework is important to overall international stability.
The US has no legitimate interest in seeing Russia's deterrent called into question. Maintaining a solid agreed framework for the management of the US-Russian nuclear relationship is in no way inconsistent with US interests.
And Russia has no legitimate interest in seeing the US open to rogue state blackmail. Indeed, President Putin has acknowledged that the problem of missile proliferation is a genuine one, however much he continues to object to the US NMD program as part of the solution.

This interrelationship of real interests suggests the possibility of an agreement that would serve the interests of both nations, and the world community as a whole:

The US must treat Russian concerns about the long-term implications of an unrestrained BMD effort as real, and acknowledge the legitimacy of Russian interest in assuring that its deterrent is not threatened and that a framework for the US-Russian nuclear relationship remains "a cornerstone of international stability." It should not be reaching out for the opportunity to withdraw from the Treaty, but instead making imaginative proposals for a new framework.
Russia must acknowledge that the US has a legitimate interest in having an effective defense against limited missile attacks, and in pursuing future technologies to maintain that defense against emerging threats, and that updating the framework - not scrapping it - is necessary to reflect current conditions.
The two nations should seek to agree on:
1. Modification or replacement of the ABM Treaty by a new agreement permitting limited defense of national territory and that of allies, as well as more flexibility in work on new technologies, with measures for greatly expanded transparency - and, eventually cooperation.
2. Strengthened controls on exports of missile and WMD-related technologies.
3. Substantial cuts in offensive forces.
4. Continued cooperation (including US financial assistance) on the mechanics of reductions and safeguarding retained systems.
5. A serious start on the long-term problem of expanding the framework of limitation, transparency, and cooperation beyond strategic nuclear forces to include non-strategic forces, nuclear infrastructure, and stockpiles.
6. Cooperation on measures to strengthen control and communications, including expanded information exchanges, maintenance of early warning capabilities, and perhaps ultimately, cooperation on active defense

FOOTNOTES

1. Since these buildups usually include short-range ballistic missiles (to a deplorable degree imported either as technology or finished product from some of the countries most vociferously opposed to missile defenses), they create a requirement for other countries in the region (and their potential extra-regional supporters) to develop and deploy defenses against them. While there are some important differences between defenses against what are "strategic" missiles in ABM Treaty terms and against other sorts of missiles, there are also important overlaps, e.g., in boost phase systems. Even if the US were giving no consideration whatever to national missile defense, the 1972 Treaty's vague distinction between severely limited defenses against "strategic" missiles and unconstrained defenses against other ballistic missiles, aircraft, cruise missiles, and, for that matter, satellites would be under severe strain in the coming years.

2. Such leverage, that is, restraining American response to conventional attacks on its allies, was a purpose of Soviet nuclear forces during the Cold War, at least in Western eyes. No doubt Soviet analysts also considered that US nuclear forces had the objective of restraining Soviet responses to American provocations, e.g., against Cuba. An implicit rationale for US (and even more for British, French, and Chinese, not to mention Israeli) nuclear forces was to convey a threat that they would be used, in extremis, to dissuade potential opponents with a conventional force advantage to stop short of fully exploiting an initial conventional success.

3. Not all countries with nuclear and missile programs are "rogue states." The extensive programs of India and Pakistan do not (at least under current political conditions) present significant strategic problems for nations other than the two of them, however serious the implications for non-proliferation policy and for stability in South Asia. Their relevance to the "rogue state" problem is - and particularly in the case of Pakistan - chiefly as potential source of technology and materials. Similarly, it is, happily, the case that not all rogue states have serious missile or WMD programs. Whatever ambitions Milosevic had for Serbia in that regard vanished with his ouster from power. Libya and Syria seem, at least for the moment, to be focusing on shorter range missile capabilities and chemicals, though progress in those efforts could lead - as they did with North Korea, Iran, and Iraq - to work on long-range missiles and nuclear and biological weapons.

4. It is also, unfortunately, true that effective defenses against ballistic missiles would not eliminate all possibilities for rogue states to threaten use of WMD. Other delivery means, from cruise missiles to terrorists, are, in principle, available. A comprehensive program to counter the rogue state threat (and the distinct, and more urgent, non-state terrorist threat) must therefore include measures aimed at these dangers as well. But, so long as there is absolutely no defense against ballistic missiles, using them as the delivery device has overwhelming advantages for a rogue state attacker - including very short interval between decision and execution, absolute maintenance of central control over the weaponry until the very last moment, and guaranteed success (if the missile works). These advantages are such that, at a very minimum, the possibility of alternative attack modes does not permit the conclusion that the ballistic missile threat is so marginal and easily substituted for that it can safely simply be ignored.

5. To advocate defenses against such blackmail attacks is not to deny the potential of deterrence by threat of retaliation, but rather to argue for the utility of a backup if deterrence fails. There is no question that successful deterrence is better in every respect than defense. However, deterrence could fail - particularly if the issue were a threat to the continuation of a rogue state regime in power, following a repulse of its immediate attack. If Iraq or North Korea were to initiate another regional war, there would be powerful voices in the US and elsewhere for going beyond simply repelling the attack, the outcome in 1991 and 1953, to eliminating the problem. (Secretary Rumsfeld recently spoke, in the context of US strategy for a major regional war of "conclud[ing] it on the basis that you may wish to go to [the enemy's] capital, you may wish to occupy a country for some brief period.") Faced with so thoroughgoing a defeat, a regime that valued its own survival over that of its population might well not be deterred from using whatever means it had to stave off disaster. (Indeed, NATO had - and nominally still has - a formal doctrine of readiness to use nuclear weapons if conventional defense failed, and the new Russian military doctrine similarly holds out the option of use of nuclear weapons in the event of attack on Russian national territory.) And once deterrence fails, successful defense is more attractive - militarily, morally, and in terms of costs on both sides - than a retaliation that would inevitably inflict massive casualties on individually innocent people, even if the targets were limited to military objectives.

6. It is sometimes argued that instead of defenses, the US should rely on supplementing deterrence by a policy of pre-emption against the missile capabilities of a regional aggressor. That is, of course, an option. However, even leaving aside the obvious intelligence and operational difficulties of relying on pre-emption, such a policy would have all the disadvantages of pre-emption as a strategic doctrine in the superpower context, albeit at somewhat lower levels. Those disadvantages include most obviously the pressure for early and overwhelming commitment of massive force.

7. One occasionally sees arguments based on the premise that at some point the Russian arsenal will have deteriorated so badly in material condition, personnel, command and control, and readiness that an initial US strike would leave only a few score Russian warheads surviving - perhaps in the range of what a limited defense can handle. Suffice it to say that there is no reason whatever to believe that the Russian force faces such a crisis of vulnerability, and that if it did, the US refraining from building a limited defense would do little or nothing to ease concerns. If 90% of the Russian force is vulnerable - which is not the case, but is the premise of such scenarios - the uncertainties in that estimate dwarf any effect of defenses.

8. The analysis of the effects of defenses on military programs ought not to be limited to considering what Russia might do in response to a US deployment, but should include consideration of the effects on the military programs of current and potential rogue states. Realization that the US has programs that would negate the blackmail potential of ballistic missile forces would tend to reduce the impulse for countries to acquire or continue them. (It is, for example, plausible that one reason North Korea has indicated a willingness to negotiate away its missile program is recognition that its continuation will produce far greater disadvantages than advantages.) On the other hand, some rogue states would no doubt pursue anti-defense countermeasures to maintain the potential of their missiles. The subject of countermeasures is highly controversial and far beyond the scope of this paper. Suffice it to say that the experience of those who have tried to build countermeasures, as contrasted to giving briefings or conference papers about them, is that developing countermeasures that work is itself a complex, lengthy, and expensive business. A rogue state staking its power on a technological competition of its countermeasures against US counter-countermeasures would be operating at a huge comparative disadvantage.

9. It is often argued that a major disadvantage of the US deploying even a limited NMD is that China will respond by substantially increasing its nuclear missile force, producing an "arms race in Asia" as India and perhaps other nations respond. Whether it is conducive to stability in the Asia-Pacific region - much less US and Russian specific national interests -- for China to have the capacity in a crisis to threaten the US (or Russia) with a nuclear attack is a debatable issue. China has resolutely rejected participation in most international arms control efforts, so the issue of departing from an agreed framework of a nuclear relationship does not arise. Moreover, it does not follow from the reluctant US-Soviet acknowledgment of the inevitability of their mutual vulnerability that mutual vulnerability among all significant nuclear states is conducive to international stability. In an important sense, however, all this is rather beside the point. To be sure, the missile defense the US is likely to deploy would have a substantial capability against the Chinese deterrent, if China simply maintained its current force of some 20 fixed-silo ICBMs. China, however, is already in the process of developing a new generation of ICBMs that will be mobile (and therefore far more survivable) and seems likely to scale its deployments to what it judges necessary to hedge against a possible US (or Russian) missile defense, whatever the US actually does. A US deployment may somewhat accelerate the Chinese program; it will not prevent China from having a survivable deterrent, and it is unlikely to have much effect on the ultimate size or shape of that deterrent. For the US, the practical question is whether it is better to be vulnerable to rogue state missile attack as well as Chinese - or just to Chinese (and of course Russian).

10. It is sometimes argued that the problem of Russian response to a US deployment of a national missile defense can be dismissed simply on the ground that Russia's economic problems make any significant response impossible. Certainly the obsolescence of key elements of the Russia strategic nuclear force and the rudimentary character of most development programs would mean any actual buildup in Russian force levels would take considerable time, though retirements could be deferred and systems kept nominally in the force despite serious operational limitations. (Generally speaking, the same would also be true for any US effort to increase offensive force levels.) More broadly, economic constraints - and strongly competing military priorities - would tend to make Russian policymakers forego countermeasures that were based on illusory dangers. It is equally true that Russia would be at a far greater economic and technological disadvantage than was the USSR in a renewed all-out arms competition. However, for Russia (as for other nuclear powers) strategic forces have the highest priority for defense resources and it would be a delusion to doubt that, once the Russian political and military leadership decided action had to be taken to maintain Russia's deterrent, they would somehow find the resources.

11. Despite the contrary claims both of BMD zealots in the US and of extreme opponents of any changes in the 1972 Treaty, that agreement does not purport to impose limits valid for all eternity. Instead, it explicitly provides for adjustments in light of changes in strategic circumstances. Of course, such changes can only come by mutual agreement, but the Treaty contemplates there will be a mutual effort to reach agreement, not either a fixed determination to have no agreement at all or a fixed refusal to contemplate any changes.