F.A.S. Public Interest Report

Journal of the Federation of American Scientists (F.A.S.)

Article Contents

The Threat Technological Readiness Arms Control Propsects PIR Nov./Dec. 1999 Main Page FAS PIR Index

National Missile Defense: Rushing to Failure

" ...we will ...determine whether to deploy a limited national missile defense ...when we review the results of flight tests and other developmental efforts, consider cost estimates, and evaluate the threat. In making our determination, we will also review progress in achieving our arms control objectives, including negotiating any amendments to the ABM Treaty that may be required to accommodate a possible NMD deployment."

In June 2000, the Clinton Administration plans to decide whether to proceed with the deployment of a National Missile Defense (NMD) system. This decision marks the culmination of an effort begun in April 1996, when Defense Secretary Perry decided to upgrade the national missile defense research efforts from a technology-demonstration effort to a deployment-readiness initiative. Under this so-called 3+3 program, three years of development would be followed by a three year deployment effort, leading to an operational capability in 2003. In January 1999, Defense Secretary Cohen announced a reorientation of the developmental efforts toward fielding the system in 2005 rather than 2003, assuming a deployment decision was made in June 2000. However, the program retains the option of an interim deployment of 20 interceptors by 2003, using prototype hardware.

Under current plans the U.S. would deploy by 2005 an initial NMD system intended to defend all 50 states against a few tens of warheads accompanied by simple penetration aids. This NMD system would include 100 ground-based interceptors based in Alaska, with site construction beginning in 2001. The Space Based Infrared System (SBIRS)-High satellite network being deployed to replace the existing Defense Support Program (DSP) satellites would provide initial detection of missile launches. An X-Band phased array radar at Shemya in Alaska and upgraded versions of five existing ballistic missile early warning radars would track incoming warheads.

Construction of the interceptor launch site in Alaska is slated to begin in April 2001. Construction of other system elements, such as the Ground Based Radar (GBR), is scheduled to be accomplished over a period of 36 months. The unclassified public record fails to explain why the relatively simple interceptor launch facility will require five years of construction, while the evidently more complex radar facility will require only three years. Plausibly, the early construction of the interceptor facility may be required to support the 2003 deadline for an interim capability.

Subsequent deployments by 2010, possibly in sequential phases, would include a second site with 100 interceptors and additional radars in order to destroy up to a few tens of ICBM warheads with complex penetration aids. The architecture would include the SBIRS-Low satellite constellation (formerly known as Brilliant Eyes) to discriminate warheads from sophisticated penetration aids. Although such guidance might exist in classified documents, the Clinton Administration has not provided a consistent detailed public rehearsal of the factors in the June 2000 decision. From various public statements, at least four criteria are relevant to the Administration's decision-making:

1. whether the threat warrants the deployment;
2. whether the development effort has sufficiently matured the technology;
3. progress in achieving arms control objectives, including revisions to the ABM Treaty; and
4. cost estimates.

The Administration's rehearsal of these criteria has taken a rather mechanical checklist approach, particularly regarding costs. The primary budgetary question has been simply whether the future year's defense plan contained the projected budgets required for deployment, disregarding the presumably more important question of whether such expenditures represented a wise investment from the public purse. Absent the eye-popping costs associated with Reagan's Star Wars schemes, the cost of NMD deployment, though non-trivial, has remained an afterthought. The other three criteria, however, remain somewhat more problematic.

"...we continue to base our NMD efforts on the assessment ...that North Korea probably will test the TD-2 this year... Iran could test an ICBM that could deliver a several hundred kilogram payload to parts of the U.S. in the latter half of the next decade, using Russian or other foreign technology and assistance."

The threat debate has dominated the missile defense debate and began with the November 1995 National Intelligence Estimate Emerging Missile Threats to North America During the Next 15 Years which concluded that "No country, other than the major declared nuclear powers, will develop or otherwise acquire a ballistic missile in the next 15 years that could threaten the contiguous 48 states and Canada." This judgment provided scant support for the near-term deployment of an NMD system and provoked a series of further assessments. The evolution in threat assessments culminated with the 15 July 1998 report of the Commission to Assess the Ballistic Missile Threat, chaired by former Defense Secretary Donald Rumsfeld.

If anything, however, the threat has matured more slowly than envisioned. In 1994 the Intelligence Community judged that the North Korean medium-range Taepo Dong-1 could be tested in 1994 and deployed as early as 1996, and the longer-range Taepo Dong-2 would be flight tested in the mid- to late 1990s. In fact, the Taepo Dong-1 was not flight tested until late 1998, and remains undeployed, and the Taepo Dong-2 remains untested.

U.S. policy initiatives also have reduced the threat. Under the terms of the September 1999 U.S.-North Korea agreement, the DPRK will refrain from testing long-range missiles of any kind during discussions to improve relations. And the U.S. announced the easing of certain sanctions related to the import and export of many consumer goods. In response to continuing Iranian efforts to acquire missile technology, Russia has created the institutional framework to implement newly enacted non-proliferation policies, along with new export control legislation covering sensitive technologies.

Nonetheless, the Rumsfeld Report presented a more alarming view of the nature of the threat and the limitations of the Intelligence Community's ability to predict how rapidly it might change. This report also proceeded from rather different premises than those that normally inform intelligence assessments and examined merely the possibility, instead of the probability, of long-range missile threats.

Threat probability is a product both of technological possibility and the political intentions of adversaries. American military planning presently focuses on Major Theater Wars in Southwest Asia (with Iran or Iraq) and on the Korean peninsula. While analysts may argue about the absolute probability of nearly simultaneous wars in these theaters, few would dispute that these are by far the most probable arenas. By the singular standards of the Rumsfeld Report, the United States should be giving equal weight to the possibility of major wars with Canada and Mexico.

This indifference to the political intentions of adversaries extends to the intention to use long range missiles, should they be acquired. In the aftermath of the Gulf War, an article of faith is that the United States now faces "non-deterrable" threats. As Defense Department spokesman Ken Bacon noted on 22 November 1999, "... in the new global environment of smaller, more radical states, deterrence may not work with the same effectiveness that it has over the last 40 years...." Although highly classified intelligence assessments might make a compelling case for this observation, their arguments have yet to emerge into the public realm.

Deterrence Failures?

The public record reveals deterrence failures, but these are failures to seek deterrence, rather than failures to achieve deterrence. Deterrence failed with both North Korea and Iraq, but in both instances the failure was on the American side and consisted of not clarifying probable American responses to aggression. Dean Acheson neglected to include South Korea within America's security perimeter, and April Glaspie indicated American indifference to Iraqi action against Kuwait. More recently, however, the United States has provided ample warning that any use of weapons of mass destruction against American interests would provoke a massive response. As Ted Warner noted at a DoD News Briefing on 20 January 1999, "...we have made Saddam Hussein aware and we have made the leadership of North Korea aware that we will prosecute a war to a victorious conclusion, and that any use by them of these types of weapons will lead to a devastating response."

A review of the Gulf War provides ample evidence for the robust operation of deterrence. Each of the nuclear weapon states clearly stated that Iraqi use of weapons of mass destruction would provoke a decisive response. Although Iraq had a variety of chemical weapon delivery systems, it did not use them during the war. And after the war, interrogation of Iraqi generals made it clear that fear of coalition response influenced their decision-making.

"...one thing that has changed literally over the last year... is the reality of how difficult this job is... The reality of how tough it is to try to do missile defense and how tough it is to try to get hit-to-kill technology... It is still high risk because we're doing things that we don't do for normal programs in the Department of Defense. You will find no programs at all that have the limited amount of testing and the aggressive schedule that we've embarked upon here even with this revised program... "

For the past five years the missile defense debate has been cast as a conflict between the White House and Congress over the desirability of missile defense. More accurately, Congress and the Pentagon have fought over the feasibility of missile defense. Congressional Republicans have consistently urged early deployment. Pentagon resistance ultimately derives from the view that the NMD program should obey the same "fly before you buy" rules that govern other defense acquisition programs.

By June 2000, BMDO plans to conduct four intercept tests involving surrogate boosters and kill vehicles, rather than the actual deployment hardware. An integrated system test of all NMD components is scheduled for May 2000. However, the actual booster for the kill vehicle and the exoatmospheric kill vehicle will not be tested until several years later.

The June 2000 Deployment Readiness Review will have two options: procure long lead items to support fielding a system by 2003, or continue testing and wait until tests of some of the actual components are completed. Tests are planned for early fiscal year 2001 for the first booster, and early fiscal year 2003 for the final configuration of the kill vehicle.

The November 1999 report of the NMD Review Group, headed by retired Air Force Gen. Larry Welch, states " ...by June 2000, BMDO will have demonstrated the 'feasibility' of [an] NMD system, but not the 'readiness to deploy' of the system... The demonstration of readiness to deploy will not come until 2003 at the earliest when the integrated [Ground-Based Interceptor or GBI]... is to be demonstrated." Under current plans, a Critical Design Review will take place in 2001 to establish the overall NMD system configuration. Five GBI intercepts tests are planned prior to a 2001 Defense Acquisition Board (DAB) decision on the NMD radar configuration. And 11 intercept tests are planned prior to a 2003 DAB decision on the operational design of the GBI.

The Welch panel recommended that the June 2000 Deployment Readiness Review be recast as a deployment feasibility review, with a focus on subsequent readiness assessment. Regardless of the characterization of this review, any political commitment it makes to deployment will be followed by additional decisions to provide content.

"We will seek to amend the treaty if necessary, and we will work in good faith to do so. We have amended the treaty before and we see no reason why it cannot be amended again. The ABM Treaty also provides, of course, for right of withdrawal with six months notice if a party concludes it's in its supreme national interests."

For nearly two decades the 1972 Anti-Ballistic Missile (ABM) Treaty was a centerpiece of the Cold War geopolitical landscape. Along with the SALT-1 agreement, it initiated an era of negotiated strategic arms control that ratified the equality of the two superpowers. Although the end of the Cold War vitiated the premise of the geopolitical equality of Moscow and Washington, it has yet to negate the rough equality of their nuclear arsenals, or the salience of these arsenals to relations between these countries. The present implausibility of a general nuclear exchange between America and Russia must surely have relaxed the tight coupling between offensive and defensive forces that drove the offense-defense reaction cycle capped by the ABM Treaty.

The end of the Cold War has not entirely invalidated the strategic arms control process, however much its progress has slowed or its premises have evolved. Engagement in bilateral arms control remains a key signifier of Moscow's place in the world and an important measure for cooperative threat reduction.

American deployment of NMD requires both a fundamental revision of the premise of the ABM Treaty and extensive revisions to the Treaty's details. Reflecting the limited range of the interceptors available in 1972, the Treaty permitted limited regional defenses (of two regions, later amended to one region), while banning comprehensive defenses of the entire national territory. The categorical ban on nationwide defenses in Article One of the Treaty is elaborated on in the subsequent articles and associated texts of the Treaty.

At a minimum, the initial NMD deployment would require elimination of the Article One ban and a revision of the Article Three limitations on permitted deployment areas to allow deployment of interceptors in Alaska (versus the permitted deployment in North Dakota, which would be the location of a second site). The deployment of large phased array X-Band battle management radars in Alaska would also require revision of Article Three (which requires such radars to be co-located with the interceptors), and the deployment of such radars at Thule, Greenland and Fylingdales, UK would require relief from the Article Nine ban on the deployment of ABM components in other countries. The various elements of the SBIRS satellite networks would require relief from the interrelated Article Five ban on ABM systems or components which are space-based and the Article Six ban on giving non-ABM systems capabilities to counter strategic ballistic missiles or their elements in flight trajectory. The existing public record clarifies neither the precise modality of any modifications nor whether the Administration will seek additional changes.

The point at which the United States would be in material breach of the Treaty, absent treaty changes, is somewhat uncertain. Under the 28 October 1976 Procedures Governing Replacement, Dismantling or Destruction, and Notification Thereof, for ABM Systems and Their Components, "The beginning of any construction or assembly work, other than earthwork (excavation), associated with the building of replacement ABM launchers (above-ground and silo) shall constitute initiation of deployment of these launchers, after which they shall be subject to the limitations provided for in Article III of the Treaty." The provisions concerning radars would appear to permit more advanced construction activity, since according to this agreement the "beginning of any construction or assembly work associated with the building of antennas (arrays), ABM radar antenna structures, or antenna pedestal supports which are not parts of ABM radar buildings shall constitute initiation of deployment..."

Ground-breaking for the Alaska interceptor site could begin in April 2001, while construction work on other components such as the X-Band Radars would appear to begin in the 2003 time-frame to meet the 2005 deployment goal. One of the outstanding puzzles posed by the construction schedule in the November 1999 NMD Environmental Impact Statement is the discrepancy between the five years required for construction of the interceptor site, versus the three years needed for the presumably more complex radar facilities. These construction times are independent of location, so the shorter construction seasons in Alaska cannot explain this peculiarity. The early ground-breaking contemplated for the interceptor site may be predicated on meeting the 2003 interim capability with 20 prototype interceptors. Thus, construction beyond excavation would occur in fall 2001. Meeting the 2005 deadline might delay this construction until sometime in 2003.

In any event, absent appropriate revisions to the Treaty, the U.S. would be in material breach of the Treaty's provisions within the next few years. Further, the Clinton Administration has approached the Russians on a take-it-or-leave-it basis, and that the American proposals are not subject to further "improvement" in the negotiating process. As Ambassador John Holum noted on 9 November 1999, the U.S. " ...determined from the beginning of this process... that we wouldn't try to play negotiating games with Russia, that we wouldn't come in and jack up our negotiating position in order to be able to give away concessions later on in the process."

The Russian position, publicly, is equally non-negotiable. According to a Russian Foreign Ministry news briefing on 20 October 1999: "Russia is not engaged in any bargaining over this treaty. We are not conducting any negotiations on any amendments to the ABM Treaty, especially amendments that would alter its key provision banning any deployment of national ABM defenses or creating any basis for such defenses."

Russian objections to American NMD deployment and the associated revisions to the Treaty probably fall into several categories. Although these layers of objections may be progressively revealed, notably, Russia fundamentally opposes a revision to the Article One ban on nationwide systems, rather than articulating detailed objections to other Treaty modifications.

Troublesome Components of the NMD System

Some specific components of the proposed NMD system, and the associated Treaty revisions, may prove more troublesome than others. As originally signed, the Treaty allowed two sites and 200 interceptors. A reversion to this posture as proposed by the Clinton Administration's NMD plan would seem to do no great violence to the original conception of the Treaty. Although the Treaty originally envisioned a bright-line distinction between permitted regional defenses and prohibited nationwide defenses, this was in part an artifact of the limited range of then-extant interceptors.

Since then, interceptor ranges have grown from hundreds to thousands of kilometers. While these ground-based components of the American NMD system could, with some difficulty, be shoe-horned into the traditional Treaty framework, the space-based sensor elements may appear rather more troublesome. Many of the Treaty's provisions are in the nature of predictability measures, providing a buffer of several years between the breach of the Treaty's provisions and the achievement of significant operational capabilities. The SBIRS satellites, initially intended as part of the limited NMD system, would prove equally useful for a much larger number of interceptors and could provide significant strategic capabilities for theater anti-missile interceptors. Totally apart from the implausibility of a general nuclear exchange between America and Russia, the Russians should have little concern about the impact of initial NMD deployments on the credibility of their nuclear deterrent. Even assuming perfect discrimination, an implausible assumption given robust Russian countermeasures, the initial 200 interceptors could counter no more than a few dozen incoming warheads.

The Clinton Administration has proposed a sequential negotiating approach, which envisions achieving initial Russian agreement to the first tranche of NMD deployment, followed by subsequent negotiating rounds to accommodate subsequent deployments. For the Russians this must surely raise the prospect of an open-ended process in which the Treaty is eventually revised to accommodate more extensive deployments. Building on the technological base of the SBIRS sensor satellites, risk-averse Russian planners could easily envision scenarios in which an extended American NMD system could provide a not-implausible damage limitation capability against a small and disorganized Russian retaliatory strike.

Russian (and Chinese) objections to American NMD deployment would, however, appear to be far more fundamental than encompassed in a narrow technical bean-count of the ratio between incoming warheads and the number of interceptors rising to meet them. These objections flow from the Russian-Chinese Joint Declaration on a Multipolar World and the Establishment of a New International Order, adopted in Moscow on 23April 1997. Although little-noted at the time, in the wake of the Kosovo War the anti-hegemonist entente between China and Russia has become more explicit and active, ultimately finding concrete expression in their joint sponsorship of a General Assembly resolution endorsing continued compliance with the ABM Treaty.

Russia (and China) are concerned not simply with the potential impact of American NMD on their own deterrent postures, but more generally with the projection of American diplomacy backed by force. There would seem to be broad agreement in Washington, Moscow and Beijing that U.S. NMD is a key element in consolidating America's role as the sole remaining superpower. There is evidently less agreement as to whether this further consolidation of American hegemony is a good or bad thing.

In 1967, Defense Secretary MacNamara delivered a famous speech that outlined the various hazards and shortcomings of missile defense and concluded by announcing the Administration's decision to deploy the Sentinel missile defense system. Although a poor defense against incoming missiles, this decision provided a more robust defense against Republican's use of missile defense in the 1968 presidential election.

From this perspective, the Clinton Administration's calculus adds up. Beginning with the 1992 campaign, the Clinton-Gore political apparatus has actively sought to avoid providing Republicans with foreign policy campaign issues. Embracing Republican positions has proven effective, as demonstrated in the 1996 decision to produce more B-2 bombers. During the 1996 presidential campaign, Bob Dole placed great stock in missile defense as a campaign issue. The utter failure of this effort has evidently not discouraged Republicans from hoping that it will be a winning issue in the 2000 election.

Politics Behind Clinton's Decision

Unlike MacNamara's cautionary speech, the recent Clinton public record does little to suggest meaningful reservations on the merits of missile defense. Over the past five years, the main political dispute has been over the timing of the deployment decision, and that controversy has subsided via Clinton's intended June 2000 decision. The White House and Congress broadly agree on the clear and present danger of the threat, the impending readiness of the technology, the necessity of deployment regardless of Russian arms control objections, and the affordability of the proposed deployment.

Both election politics and the internal logic of its own policies will compel the Clinton Administration to an affirmative decision endorsing national missile defense deployment in June 2000. Less apparent is the substantive content of this decision. Evidently, from the Welch Report, any decision in 2000 will be little more than an expression of political intention, and that the more substantial decisions concerning the precise nature of the deployed system will be made by the next president.

It is less than evident that Clinton must place the United States on an immediate collision course with the ABM Treaty. If the start of interceptor site construction in early 2001 is only required to retain the option for an interim capability by 2003, neither the development of the threat nor the maturity of the technology warrant this step. Faced with Clinton's highly visible political commitment to NMD deployment by 2005, the Republican presidential contender is unlikely to make much political hay from a call to deploy a largely untested system by 2003. Given the electorate's demonstrated indifference to this issue generally, such modest distinctions will surely be lost in the heat of the campaign.

Nonetheless, for either political or policy reasons Clinton may choose to authorize the start of construction in Alaska in early 2001. However, this may not result in a material breech of the ABM Treaty by fall 2001. The new President will probably undertake a broad review of national security programs, to include missile defense, and might delay construction in Alaska pending review completion. The new President might also be reluctant to confront difficult choices on the ABM Treaty in early 2001, before his national security team has been nominated and confirmed.

Implications of Material Breach

However, should construction proceed, the full implications of an American material breach of the Treaty in late 2001 are presently difficult to encompass. Both America and Russia will be led by new Presidents and will have elected new legislative bodies. The recent drift towards increasingly adversarial relations may reverse or abate, or become greatly magnified. These broader political considerations will surely condition political reactions to missile defense and arms control.

From a narrow arms control perspective, even a material breach of the Treaty would not mark the end of the road, but rather yet another turn in a long and twisting path. Absent an intent to abrogate the Treaty, the U.S. might plausibly argue that it remains optimistic that subsequent negotiations will reconcile the Treaty's provisions with the newly created facts on the ground. As witnessed by the Krasnoyarsk radar episode, violations of the Treaty have not automatically led to the Treaty's termination, as long as neither side desires that outcome.

The ABM Treaty has accumulated a rather disparate collection of unratified amendments and clarifying declarations of uncertain legal standing, demonstrating a considerable tolerance for ambiguity in the relationship between the letter of the Treaty and actual state practice. In 1997 America and Russia reached agreement on extending the demarcation between theater and strategic ABM systems to include interceptors with velocities of up to 3 km/sec, tested against ballistic target-missiles with velocities up to 5 km/sec. Russia refused to agree to extending the definition of non-strategic anti-missile systems to faster interceptors and targets, and the U.S. made a unilateral determination that such faster systems were compliant. This discrepancy remains unreconciled. More recent Russian complaints about the first NMD interceptor test constituting a breach of Article One have not impacted the overall status of the Treaty. The precise methods for the reconciliation of the explicit provisions of the ABM Treaty with actual American missile defense activities cannot be predicted with certitude. With somewhat greater confidence, one can anticipate an increasing focus on the management of the political and strategic consequences of these activities.

With some creative effort, the Russian leadership may be persuaded to decouple concerns about the missile defense issue from the strategic arms reduction agenda and kindred cooperative threat reduction initiatives. Although the challenges are evident, the means for surmounting these difficulties find ample precedent in decades of prior statecraft. (A companion article examines the perhaps more difficult case of China.)

Nearly two decades after Reagan's Star Wars speech, America appears on the verge of finally deploying an NMD system. During the 1980s, critics charged that missile defenses would prove unworkable, concerns that were validated by the poor performance of Patriot in Desert Storm. Critics have long argued that missile defense will create more problems than it will solve. We are about to obtain experimental validation of these concerns, and increasingly, the task at hand will turn to solving the problems created by missile defense deployments.