ENGINEERING RISK AND THE ABM TREATY Testimony
by
Henry F. Cooper
to the
House Subcommittee on National Security, Veterans Affairs, and International Relations
Committee on Government Reform
September 8, 2000Mr. Chairman, thank you for the opportunity to provide my views on the implications of recent ballistic missile defense testing and how the Anti-Ballistic Missile (ABM) Treaty affects America’s efforts to build effective missile defenses.
As you requested, I have focused my testimony on two questions: 1) What is the impact of test results to date on technology maturity and deployment schedules? 2) What is the relationship between ABM Treaty provisions and current proposals to design, test, and deploy an effective missile defense system?
In addressing these questions, my perspective is based on experiences as an engineer, who spent seven years actively engaged in the arms control process—ranging from developing Reagan Administration space policy, to backstopping our negotiations with the Soviet Union, to serving as Ambassador and Chief Negotiator at the Geneva Defense and Space Talks. I carried that combined engineering and diplomatic perspective to my job as Director of the Strategic Defense Initiative (SDI) during the Bush Administration.
From that perspective, I see these questions as distinct, but closely related—and the second as far more important. The ABM Treaty and associated political constraints have retarded—if not precluded—the development, testing and deployment of the most effective, least expensive defensive systems, systems that could be built soonest.Testing, Engineering Risk, and Technology Maturity
I wish to make three points: 1) Developing, testing and deploying effective ballistic missile defenses is an engineering problem using a mature technological base; 2) No significant engineering design flaws have been identified by recent development testing—and none are expected; and 3) The U.S. aerospace community has repeatedly met more daunting engineering challenges than posed by the task of building effective ballistic missile defenses. Consequently, there is little doubt, from an engineering perspective, that effective defenses can be built. Better management practices can us help build them sooner, and the removal of political constraints—especially those associated with the ABM Treaty—can help us build better defenses sooner for much less money.
Mature Technology Base. Key proof-of-principle experiments conducted 15 years ago demonstrated that the hit-to-kill intercept concept works, provided the kill vehicle can be delivered into a “sufficiently small error basket” in the vicinity of the target “ballistic missile or its elements in flight trajectory” (to use the terminology of the ABM Treaty). Using pre-SDI technology in 1984, the Army’s Homing Overlay Experiment (HOE) launched from Kwajelein Island in the South Pacific a Volkswagen-size kill vehicle to intercept a Minuteman missile launched from Vandenberg AFB in Southern California. Shortly thereafter, the Air Force successfully intercepted a dying low-altitude satellite with its Miniature Homing Vehicle launched from an F-15—also using pre-SDI technology.
Early on, the SDI program instituted a major technology demonstration program that placed priority on dramatically reducing the size and weight of critical propulsion, sensor, data processing, and other electronic subsystems to enable an effective hit-to-kill interceptor system—however based. Notable among these demonstration programs was the Delta series. After only 18 months from conception to liftoff in 1986, SDI’s Delta-180 experiment successfully completed an intercept in space between two uncooperative small space vehicles employing independently thrusting rockets—winning wide applause from the aerospace community. Another most notable experiment in this highly successful series was Delta 183 (or Delta Star) in 1989, which, over a nine month period, gathered very important signature information that is most pertinent to today’s efforts to accomplish intercept in boost-phase and midcourse in the face of responsive countermeasures. Also in 1989, the Army’s ERIS program repeated the HOE experience with a much lighter interceptor kill vehicle, using mid-1980s technology. And there have been numerous other experiments that demonstrate the maturity of the basic technology underpinning current programs to develop ballistic missile defenses.
The technology underpinning current efforts was significantly advanced throughout by the SDI program, leading to the possibility of much lighter weight, more capable sensor, data processing, and propulsion components of modern kit-to-kill interceptor designs. Much of this technology was demonstrated and space qualified on the award winning 1994 Clementine mission, which returned to the Moon for the first time in 25 years, provided over a million frames of optical data in 15 spectral bands, and discovered water in the polar areas. Regrettably, President Clinton used his short-lived line item veto authority to kill the Clementine follow-on deep space probe that would have space qualified today’s technology that could have been exploited by ballistic missile defense programs for all basing modes.
Indeed, I am very critical of the Clinton Administration’s dramatic reduction of investments in such demonstration programs that I believe are vital to sustaining the U.S. technology edge, especially in the inevitable measure-countermeasure competition associated with building and sustaining effective ballistic missile defenses. On my watch, SDI invested over a billion dollars a year in such technology programs—now they receive only a few tens of millions a year. In the 1960s, the Advanced Ballistic Re-Entry Systems (ABRES) program invested around $200 million a year—corresponding to over $1 billion in current dollars—in such activities to assure U.S. reentry vehicles could penetrate Soviet defenses. This program was abandoned in the 1980s because that technology work was being accomplished under the SDI program. The nation needs to reinstate such a program to assure American space defense technology remains second to none in today’s world.No Engineering Design Flaws. Although there have been many “failures” in recent test programs, none have identified fundamental design flaws associated with the “hit-to-kill” defense interceptor development programs.
Consider the Army’s Theater High Altitude Area Defense (THAAD) program, which became a Major Defense Acquisition Program (MDAP) on my watch as SDI Director. It had a well-advertised string of 6 failures before achieving a successful intercept. Every one of these failures was traced to a failure in engineering discipline and/or quality control in aspects of the design that were mastered years—and sometimes decades—ago. Most of these failures occurred in the test sequence of events before the interceptor’s kill vehicle could fully exercise its endgame sequence—and in the exceptional cases, the endgame failure was traced to well understood and previously tested sub-components. Following this string of failures, THAAD has entered engineering and manufacturing development, which should lead to a production decision now planned as soon as in 2006.
Except for continuing political constraints associated with the ABM Treaty, I anticipate that the Administration’s National Missile Defense (NMD) program may have much the same experience. Contrary to President Clinton’s September 1 pronouncements associated with his decision not to commit to deploy a radar in Alaska, the much repeated and media-hyped “two-out-of-three” test failures have nothing to do with the maturity of the underlying NMD technology nor the soundness of the NMD engineering design. As was the case with THAAD, these two failures were not engineering design failures; they were engineering discipline failures that precluded meaningful tests of the NMD kill vehicle.
(It is frustrating to have to relearn such lessons over and over again, but such engineering discipline failures should not be confused with engineering design failures. These failures have perhaps identified an industrial base problem that your committee might consider further, Mr. Chairman. Perhaps with all the mergers and downsizing of key defenses companies over the past decade, we have lost the service of competent senior engineers before they could pass on their expertise to the next generation. If so, we have to pay to reinvent what we learned long ago.)
But in neither case—THAAD nor NMD—should any test failure to date be used to infer anything about the eventual effectiveness of the defense system. Separate testing of the ultimate production interceptors must establish the basis for estimating system effectiveness. Comments to the contrary are wrong—and when made by alleged scientists or engineers betray either a lack of professional expertise or irresponsible rhetoric intended to further an ideological cause.
Retired Air Force General Larry Welch has led numerous reviews of BMD development programs over the past several years, and—to my knowledge—he has found no fundamental engineering design flaws in spite of his oft-quoted criticism of a “rush to failure” attributed to cause many of the breakdowns of engineering discipline. With all due respect for General Welch and his committee, I consider this to be unhelpful rhetoric from an engineering perspective. It is possible to do things rapidly and succeed—in spite of many failed attempts along the way. Rushing and failing do not necessarily go together.
I do not recommend recklessness, of course. But much that is important to assuring highly reliable and effective systems can be learned from failures in an intense test program—whereas the current political hype demanding success even for early tests in a complex system development program creates a risk adverse atmosphere that slows the development program, actually adds to the costs and risks, and is generally contrary to sound engineering practice.History Encourages Confidence in Engineering Success. To emphasize this point, one needs only to consider a few of the many historic examples of successful aerospace programs that had numerous initial test failures before establishing impressive records of success as operational systems. The highly successful manned space program was plagued by numerous very public test failures before a successful launch capability was achieved—just as had been the case in the unmanned space program before it. So it was also with the earlier development of the Air Force’s Intercontinental Ballistic Missiles (ICBMs) and the Navy’s Submarine Launched Ballistic Missiles (SLBMs).
Of particular interest is the Navy’s Polaris program in the late 1950s, which overcame many problems far more daunting than those facing engineers now developing ballistic missile defenses. For example, the Navy had to design a special submarine with numerous launch tubes (16 in the first SLBM-carrying submarines), each with larger escape hatches than ever before attempted. Navigation errors at sea were then often on the order of 10 miles—much larger than could be tolerated for launching a militarily useful SLBM. (In 1959, an error of thousands of meters was discovered in the location of Australia.) And at the outset of the Polaris program, solid rocket technology would not provide sufficient thrust and control to lift and deliver an existing warhead to the desired ranges. Inertial guidance systems, underwater communications systems, and reliable electronic controls had to be invented to make viable the Navy’s Strategic Submarines (called SSBNs) and their SLBMs.
Furthermore, the Navy was the last Service to propose a ballistic missile program—in 1955. Its initial 10-year plan was to begin testing in 1958, deploy a liquid fueled missile on a freighter in 1963, begin testing a solid fueled missile from a submarine in 1963, and begin undersea operations in 1965. A year after the Navy’s ballistic missile program was officially begun in 1956, a major acceleration (by five years) came in response to Sputnik in October 1957.
In its first year, 22 flight experiments examined various technology options—leading to a specific set of design objectives for the first solid propellant SLBM. In the first Polaris development test series from October 1958 through September 1959, there were 12 failures in the first 17 attempts to launch the first prototype SLBM. In the next year, through September 1960, there were 12 failures out of 40 flight tests to assess production-related hardware. And the Navy deployed its first operational Polaris submarine, the George Washington—and its complement of SLBMs—in four years after program initiation; and the SSBN force went on to become the most survivable leg of the nation’s triad of nuclear deterrent forces.
Among the technical challenges confronting today’s engineer designing a ballistic missile defense system, only the midcourse discrimination problem comes anywhere close to many of the difficulties faced and solved by the Polaris team in the late 1950s. Here again, an intense test program is required to solve this important problem with confidence—and such a program should be reinstated for this purpose.Managing for Success. Another lesson from a review of past successful rapid development programs, such as Polaris, is the importance of streamlined management—which is far from standard Pentagon acquisition management. As I noted in my January 20, 1993, End of Tour Report as SDI Director, “I deplore the current wasteful [acquisition] process which needs a major overhaul of all DoD programs—and especially complex revolutionary acquisition programs such as SDI. Interestingly enough, the 5000 series of acquisition regulations permit some innovation, but these provisions are exercised rarely if at all because, in large measure, those, particularly at the staff level, who exercise oversight have no accountability for programmatic success—it seems everyone in OSD can say no and hold up the entire process (and someone seems inevitably to be so disposed) while essentially everyone in OSD must say yes to progress to the next step.”
To illustrate the waste, I had my staff keep track of the costs and time spent in running the gamut of six months of meetings in 1991 leading up the Milestone I Defense Acquisition Board (DAB) for THAAD. The SDI staff and our executive agents (Army) spent about 75,000 government labor hours, over 250,000 contractor hours, and over $22 million creating over a ton of supporting documents, giving briefings, and supporting meetings intended to reach consensus with three staff levels of Pentagon bureaucrats (in up to 35 offices) to address—and invariably readdress—over 900 supposed issues, often with conflicting programmatic changes (sometimes within the same Pentagon office). In short, program managers honed their diplomatic skills for numerous roundtable debates rather than rather than their technical and management skills needed to direct a sound development program.
I assure you things have not improved during the Clinton Administration. For example, I understand that the committee preparing the Pentagon’s report to congress on the potential of a sea-based NMD system has seven—repeat seven—co-chairs. This is in the “Executive” Branch? Is it any wonder that this report, due last March, still has not been sent to Capitol Hill? Practically every aspect of acquisition management in the Pentagon fails well-known principles of Management 101.
Sadly, we have grown accustomed to this wasteful bureaucratic process and now budget both time and money to accommodate its excesses. Needed is focused management, with assigned responsibility, resources, and accountability—like existed for the Polaris program, which performed so well over forty years ago. That is precisely why Retired Vice Admiral J.D. Williams and I recently recommended forming such a focused Navy organization to expedite building the most effective sea-based defenses afforded by American technology.
If fully funded under such a streamlined management structure, our proposed approach (attached to my testimony as Appendix A) could lead to an initial operational sea-based capability as early as in 2003 to begin protecting Americans at home—as well as our overseas troops, friends and allies—against North Korean missiles. This initial capability, which would cost less than $2.5 billion more than is currently budgeted for the Navy Theater Wide program, could be significantly improved by 2005 for an additional $5 billion to provide block improvements to the initial operating capability.
The Navy should be still able to find engineers who are as good as were the Polaris engineers forty-five years ago. Surely they can perform as well in building and improving a Navy Theater Wide program—which has many fewer technology hurdles to overcome.
Technology is not a problem. Engineering risk is not a problem. Management capability should not be a problem. However, arms control—and the ABM Treaty in particular—is a problem that adds risk, costs, and additional time to the development, testing and deployment of effective ballistic missile defenses. Relief from these constraints is absolutely necessary if effective defenses are to be built.The ABM Treaty Trap
Mr. Chairman, I will try to explain to you and the Committee why the ABM Treaty is itself a threat to U.S. security interests, a threat costing us money and time, while limiting—sometimes severely limiting—the effectiveness of systems we are developing. Unless our policies change, our adherence to the Treaty will cost us lives when ballistic missiles are fired at our overseas troops, friends and allies—and possibly someday Americans at home.
These costs will result because we are precluding the development, testing and deployment of the most effective defenses—which also are least expensive and can be built fastest—to adhere to a strict interpretation of the ambiguous terms of this Cold War Treaty. Because of our attachment to the ABM Treaty, we are taking longer and spending more money to develop less effective defenses than our current technology affords. The as yet unrealized potential of sea-based defenses, discussed in Appendix A, is clear evidence of this bottom line.It’s a MAD, MAD World—At Least for the U.S. You must understand that the ABM Treaty has succeeded in its purpose of blocking the development, testing and deployment of effective defense against ballistic missiles, at least for the United States. This meets the objectives of those who wish to preserve the Cold War Mutual Assured Destruction, or MAD, doctrine of mutual deterrence on which the Treaty is based. The most dedicated MAD advocates have opposed even a very limited ballistic missile defense capability—even for our troops on the battlefield—as violating the ABM Treaty.
Consider the history of the Patriot improvement program to illustrate how profound are these political and policy constraints. In the mid-1980s, several senators—led by Senator Dan Quayle—initiated a program to improve the Patriot air defense system, which had been precluded from having any ballistic missile defense capability since the 1970s. They wanted to give the Patriot battery at least a capability to defend itself against short-range ballistic missiles.
The senators wisely placed this improvement program initiative in the Army rather than SDI, where politics surely would have killed it. Even so, at various hearings, the arms control elite claimed that we couldn’t give Patriot even this limited capability—far less than being developed for today’s theater missile defenses in the wake of the Gulf War. The former Arms Control and Disarmament Agency Director and SALT II Negotiator, Paul Warnke, among others, testified against improving Patriot, claiming it would violate the ABM Treaty if Patriot were given the ability to shoot down even a short-range ballistic missile.
In spite of such resistance, the Patriot improvement program proceeded with the Senate adding money and direction to the proposed budget, while no money or direction was added by the House—and then, as they generally do, the Senate and House split the difference in conference. When Iraq invaded Kuwait on August 2, 1990, we had three or four improved Patriots ready to test—and none to deploy. A courageous Army colonel in Huntsville, Alabama directed the start-up of the production lines on his own authority, without top cover from the acquisition bureaucracy in Washington. (In tribute, I later—after the Gulf War—went to his retirement ceremony, when he retired as a colonel. I thought he should have been made a general just because of his courageous decision to move ahead with Patriot.)
In any case, all the Patriots used in the Gulf War were produced between August 1990 and January 1991—before full testing, and the software was modified on the battlefield. This rapid response illustrates what American industry can do when the adrenaline is flowing.
Whatever the technical effectiveness of Patriot—which is arguable—having this defensive system in Israel was very important politically. If we had not had that defensive option, Israel might not have stayed out of the Gulf War. If Israel had come in, the Arabs might have left the coalition. That was what Saddam Hussein had in mind when he launched Scuds at Israel in the first place, turning normal deterrence theory on its head—he sought to provoke retaliation. (This experience should at least raise fundamental questions about the validity of the MAD doctrine in the post Cold War world.)
So Congress’ collective reluctant investments to provide a ballistic missile defense capability for Patriot turned out to be profoundly important—and no one now claims Patriot violated the ABM Treaty. Indeed, almost everyone now supports building better theater missile defense (TMD) systems. But if you read the fine print, you will discover that we continue to limit the capability of our TMD systems under development, as discussed below, because of concerns about the ABM Treaty.Soviet/Russian Behavior—A Different View? Before discussing further the constraints we place on our development, testing, and deployment programs, I want to emphasize that our Treaty partner, the Soviet Union, did not behave as if they bought the MAD doctrine.
Throughout the Cold War, the Soviet Union spent as much on strategic defenses as on strategic offense—unlike the U.S., which spent an order-of-magnitude more on offenses. Furthermore, the U.S. dismantled its major air defense system and mostly invested its “strategic defense” dollars in strategic warning systems to enable a confident retaliatory strike should the Soviet Union attack the U.S.—rather than on either active or passive defenses that would become important if MAD deterrence ever failed. On the other hand, the Soviets invested heavily in ballistic missile defenses at and beyond the limits of the ABM Treaty (e.g., the Krasnoyarsk radar), air defenses, hardened command and control systems including massive deep underground centers, a major civil defense program, etc. Such major investments suggest that the Soviets did not adopt the U.S. view of the desirable advantages of vulnerability. Russia continues to invest in maintaining and modernizing many of these activities today.
Whatever the U.S. did to constrain its air defense systems for so many years (as illustrated by the above discussion of the tortured birth of Patriot’s limited ballistic missile defense capability), the Soviets view differed from the U.S. view of the ABM Treaty, and they fully exploited its ambiguous terms in designing and building their SA-5, SA-10, and SA-12 air defense systems. According to information that former Defense Intelligence Agency analyst Bill Lee has assembled, former senior Soviet officials have revealed that these systems were internetted into a territorial defense and were intended to have ballistic missile defense capability, from the beginning. The Soviets pursued this design before the ABM Treaty was signed and they never deviated from it. They now market current versions around the world as being superior to Patriot. (Could this competitive entrepreneurial spirit be related to the loud Russian protests about U.S. theater missile defense development programs?)
The Russians continue to exploit this command and control architecture that integrates into their territorial defense widely dispersed radar, including the LPARS—the large phased array radar systems on the periphery of the former Soviet Union. The Clinton Administration refuses to use this same architectural approach for its theater missile defense systems, because it continues to judge it would lead to violations of the ABM Treaty—the Russian precedent notwithstanding.
Just to emphasize how asymmetric is the U.S. and Soviet—now Russian—treatment of Treaty ambiguities and our penchant for persisting with our restrictive views even when we know the Soviets/Russians do not, consider another related case history. In 1991, after the Soviet Union collapsed, a U.S. group visited one of the LPAR sites and discovered it was tied into the command and control system for the Moscow ABM system. Washington was quite concerned about this discovery because the U.S. had precluded such an approach in the design of its own ABM systems—because of ABM Treaty considerations.
After interagency review, President Bush’s 1992 compliance report to Congress indicated our concerns, but stopped short of calling the Soviet/Russian practice a violation of the ABM Treaty. After a year of interagency deliberation, President Bush’s January 1993 compliance report to Congress concluded that the Soviet (and Russian) practice was not a violation because of the ambiguities of the treaty. This determination was followed by committee-drafted, lawyerly gobbledygook describing what could and could not be done consistent with our unilateral interpretation of the ABM Treaty.
Presumably, this painfully laborious process should have eliminated interpretation differences that permitted the Soviets and Russians to do what the U.S. had thought (for 20 years) was prohibited by the Treaty. But behold, the Clinton Administration continues, today, to deny even our theater defenses the same command and control architecture the Soviets/Russians have used since before the Treaty was negotiated. And the Administration has delayed building low-altitude space sensor systems, which could support both theater and national defenses.
Why? If we used this architecture to tie a host of sensors into our sea-based and mobile ground-based theater missile defense systems, they could also defend the United States of America. But that would violate Article V of the ABM Treaty, as President Clinton’s primary advisor on arms control, Bob Bell, stated last year would be the case if this approach were adopted by the Navy Theater Wide system.
Only recently, I might add, were Aegis cruiser radar modified to look up and out into space to see ballistic missiles—via a software change. They didn’t need to look into space to see cruise missiles—against which it is permissible to defend. After the Aegis radar software was modified, it was able to track the launches out of China toward Taiwan in 1996, when China sought to intimidate the Taiwanese during their first free election. We had no interceptor to shoot them down, of course—and we still don’t, because of policy choices made by the Clinton Administration. If programs I left fully funded in 1993 had been continued, I am confident that Aegis cruisers would now be capable of shooting down such Chinese missiles if there were a future repeat performance. Hence, it is not hard to understand why China is so vocal in opposing U.S. missile defense programs.
The Navy also is actively developing an internetted sensor architecture—called the Cooperative Engagement Concept (CEC)—to provide a wide-area cruise missile defense. It is important to note that this CEC architecture being used to increase wide area coverage against cruise missiles is not planned for use to increase wide area coverage against ballistic missiles. According to the ABM Treaty lawyers, Navy Theater Wide ballistic missile intercepts must be accomplished by a system using only co-located radar and interceptors on a given cruiser. If target-track information were passed to another Aegis cruiser—say one that might get a better shot at a threatening ballistic missile—then that other Aegis cruiser might become too capable. It might even be able to shoot down missiles launched at the United States, and become a sea-based ABM system—which is banned by Article V. If that happened, the lawyers would stop all development, testing and deployment—even for TMD applications.
Thus, to avoid this “undesired” result, the Clinton Administration has dumbed-down our theater defenses—otherwise, the U.S. would not be able to build anything. If something is not done to remove this policy constraint, the captain of a cruiser in the Sea of Japan will someday be able to shoot down a missile launched from North Korea toward Tokyo—but not if that same missile is launched toward Seattle. I am confident that the American people will not understand such ludicrous policy constraints if that day ever comes. (Appendix A discusses how appropriately enabled sea-based defenses can defend the U.S. against North Korean missiles.)How we got an Ambiguous Treaty. During the original ABM Treaty negotiations, ABM systems under development included nuclear-armed interceptors and their launchers and associated radar—these elements of then-existing ABM systems were called “components” in the ABM Treaty jargon and limited by the terms of the ABM Treaty. Other essential system elements were not included as components, because Treaty verification would not have been possible by “National Technical Means.” Furthermore, such difficulties led to significant ambiguity as to how to apply the Treaty to systems without such components—such as lasers, non-nuclear interceptors, key supporting sensors other than radar, etc. These “other” or “future” ABM systems were not discussed in any specific sense. In particular, the Pentagon was conducting very secret R&D on lasers at the time, and the negotiators were forbidden even to use the word “laser” in their discussions with the Soviets. They dreamed up the euphemism “systems based on other physical principles” to describe defensive systems using “future” technologies—without specifying what constituted “future” technologies.
A review of the negotiating record in the mid-1980s (because the SDI program was considering all possible technologies) led to the “broad vs. narrow” interpretation debate. The “broad” interpretation argued the only place future technologies were authoritatively dealt with in the Treaty was in Agreed Statement D, which referred to “systems based on other physical principles.” Consequently, all future ABM systems and components (using technology other than that used in ABM systems being developed in 1972) could be developed and tested, but limitations had to be negotiated before deployment.
I yield only to Dr. John Norton Moore (Walter L. Brown Professor of Law and Director of the Center for National Security Law at the University of Virginia) and Judge Abraham Sofaer (Legal Counselor to the Secretary of State in the Reagan Administration) as spending more time during this debate than I reviewing the negotiating record and “practice of the parties.”
Professor Moore’s 4,000-page report is still under lock and key at the State Department—and I urge the Committee to look into making at least its conclusions part of the public record.
Judge Sofaer concluded the broad interpretation was fully justified—as did I. Certainly, the Soviets never agreed to the “narrow” interpretation in 1972, and they and the Russians after them have exploited the ambiguities of the Treaty and negotiating record (and consciously violated clear restrictions as well) whenever it served their interest to do so. But in 1985, Senator Nunn and other Senate Democrats won the political debate, and the Reagan and Bush Administrations were constrained to pursue R&D bound by the “narrow” interpretation—while claiming the “broad” interpretation was fully justified. Then in early 1993, the Clinton Administration renounced any validity of the broad interpretation and adopted the narrow interpretation as U.S policy.
Consequently, Article V of the ABM Treaty—which, in this “narrow” context, deals with future systems whatever may be the meaning of Agreed Statement D—prohibits development, testing, and deployment of mobile ABM systems and their components, whether based in space, in the air, on the ground, or at sea. So mobile ABM systems can’t even be developed or tested under the narrow interpretation. Under the broad interpretation, they could have been developed and tested, but limitations had to be negotiated before deployment.
This constraint is why there is so much contention about building, for example, sea-based theater missile defenses intended to protect our overseas troops, friends, and allies. It is an engineering challenge to build forward-based wide-area sea-based defense with no capability also to defend the United States under all scenarios. (See Appendix A for a discussion of the inherent potential of sea-based defenses in the Sea of Japan for shooting down North Korean missiles.) But if engineers are not creative enough to do so, then the ABM Treaty lawyers will prohibit them even from developing and testing sea-based theater defenses for our overseas troops already threatened by ballistic missiles from several rogue states.
Bear in mind that in 1972, we never agreed with the Soviets on the meaning of many important words of the Treaty—not even for such operative words as “development.” Unilateral statements have governed our R&D actions—in the case of “development,” we declared unilaterally that we could test anything short of full scale prototypes of ABM systems and components before crossing the line of development. But this definition seems contentious still, based on what comes out of Pentagon discussions associated with compliance of R&D on modern and advanced ABM systems.Impact on Development and Testing of “Effective” Defenses. The important legacy of such constraints is that, when planning and conducting R&D on mobile ABM systems, engineers have to design experiments to prove out concepts without “developing” a mobile system or “component”—as noted above, another term defined in 1972 only for 1972-vintage ABM systems. But, they are driven to more complicated, higher-risk experiments—if major experiments are done at all.
Consider space-based interceptors. Any competent engineer, not constrained by legal interpretations, would put an interceptor on a satellite launcher in space, have it orbit the Earth awaiting launch of a target rocket from Earth, and then commit the interceptor to destroy the target. That would be a straightforward experiment. But, according to our unilateral interpretations of “development” and the “narrow interpretation” of the ABM Treaty, such a test would constitute prohibited development.
So what can engineers do? To intercept in space, they can design more complicated experiments, for example, using two satellites in orbit to perform a co-orbital intercept—like the 1986 Delta 180 experiment. (Anti-satellite, or ASAT, systems are not limited by the Treaty—indeed, the Soviets had an operational co-orbital ASAT at the time.)
To intercept more realistically “from space” an accelerating (“boosting”) rocket from Earth, engineers can creatively employ ground-based interceptors, permitted by the Treaty. They can launch a ground-based interceptor into a lofted trajectory and, after it reaches apogee—its highest point above the Earth—and is on its way down, then they can launch a separate target rocket from the ground and intercept it with a probe launched from the pseudo “ground-based interceptor” as it returns to Earth from space. Although more complex than a straightforward test from space, this experiment is feasible—at least conceptually.
The problem is that the early commitment of the first part of this elaborate “ground-based interceptor” experiment could cost $300 million and the target would cost much less—say $3-5 million—and failure of the target would result in a failed experiment overall. Many hundred-million-dollar experiments have been lost because of million-dollar target failures—and such targets involve 30-year-old technology not at issue in the experiments.
Thus, this Treaty-driven experiment is not prudent, from a technical risk point-of-view. Much better would be the far simpler test that would launch an interceptor from a satellite after the target rocket is launched. If the target didn’t work, the experimenter would simply take another few orbits and launch another target rocket before conducting his several hundred-million-dollar intercept experiment. Obviously, that experiment could be conducted with much less risk than the Treaty-driven one.
Test failures of advanced concepts because of failures of well-understood technology are not unusual—as demonstrated by the THAAD and recent NMD intercept test failures discussed previously. At the same time, numerous experiments, such as the one I described, have been made more complicated and costly because of Treaty interpretations.
I want to emphasize that a space-based interceptor system still offers the most mature, least expensive, and most effective near-term global missile defense capability from a technical point-of-view—for both theater and national missile defense. Space technologies were the most mature and effective SDI technologies.
This fact is not realized by many—apparently including Defense Secretary Bill Cohen, who told the Senate Armed Services Committee on July 25, 2000, that the Bush Administration in 1991 had chosen to build a ground-based NMD system “that could be developed more rapidly” than a space-based system because “technology challenges” made space-based systems “too complicated.” As I pointed out in my letter to SASC Chairman Senator John Warner (Appendix B), this impression is simply untrue. Space-based interceptors were the most advanced of all SDI programs.
In particular, the Brilliant Pebbles program, begun in 1987, was the first SDI program to become a formal Major Defense Acquisition Program (MDAP) in 1991—a stage not reached by the ground-based NMD system until two years later—after which it was immediately scuttled by the Clinton Administration in early 1993, and only recently has it been resurrected. Had Brilliant Pebbles been sustained and the ABM Treaty problems solved, a global defense could have begun operations as early as in 1996—for substantially less than half the likely cost of the first ground-based NMD site.
In the 1980s and early 1990s, technology flowed from space to ground, not the other way around. Since then, industry has exploited SDI-developed technology to build low-altitude satellite systems for communications, remote sensing and other functions. Industry wants as little as possible to do with the government, which only slows them down—and time is money.
So space technology is mature, but our missile defense programs are not fully exploiting it because of policy and political constraints, mostly associated with the ABM Treaty. Such constraints are inhibiting even our use of space-based sensors that can greatly improve the effectiveness of theater and national missile defenses.A Way Out? If the United States is to build effective defenses in the future, something must be done to alleviate the influence of the ABM Treaty. My preference used to be a negotiated solution that mandated a cooperative regime with Russia—and others—to build a global defense to protect free peoples around the world against ballistic missiles launched by rogue leaders. And there was once hope that we might achieve such an agreement.
In 1992, in the same U.N. speech in which he proposed the deeper reductions that became START II, Russian President Boris Yeltsin proposed that SDI be redirected to take advantage of Russian technology and to build a global defense for the world community. The Bush Administration made progress toward agreement on this agenda—and our allies and friends supported the U.S. position at that time. But no agreement was reached—and the Clinton Administration abandoned those talks on a cooperative regime to build effective defense. It instead declared its allegiance to strengthening the ABM Treaty, which is based on the MAD confrontational model of the Cold War.
It may be possible to revive talks on such a cooperatively regime—which would be in Russia’s interest as well as ours—if cooperation is truly Russia’s objective. Russian President Vladimir Putin’s recent proposal to build a joint boost-phase defense for NATO might provide grounds for working together on a global defense. But I urge the U.S. powers that be to state clearly—unambiguously—their intention to build the most effective defenses that current technology permits with or without Russia’s cooperation. If Russia cooperates, we can work together to assure that the global defenses serves our mutual interests. Otherwise, we should declare ourselves free on the ABM Treaty and build the best defenses we can as soon as we can.
ConclusionsMr. Chairman, I explained why I believe recent test results have little, if any, bearing on evaluating the maturity of ballistic missile defense technology, essentially all of which already passed “proof-of-principle” tests—in many cases over a decade ago.
In my opinion, there is no technical reason for not establishing schedules for development, testing and deployment of a comprehensive, layered national missile defense system. We know that the hit-to-kill and associated system technology will work—development, testing and deployment is purely an engineering task. Furthermore, establishing a definite deployment goal puts rigor and discipline into engineering development scheduling in a way that reduces risk and management time to accommodate the uncertainties associated with dealing with the excesses of the Pentagon’s acquisition bureaucracy.
The recent decision by President Clinton should be taken in this context. A lack of technological maturity clearly was not the issue that led to delaying a national missile defense deployment decision, which responsibly could have been made at a much lower level years ago—as indeed it was on my watch as SDI Director during the Bush Administration. This recent decision was dominated by international politics—especially surrounding the ABM Treaty and the concerns of Russia, China, and various U.S. allies.
The ABM Treaty has achieved its objective of blocking the building an effective defense against ballistic missiles—at least for the United States. When in 1993 the Clinton Administration declared the ABM Treaty the “cornerstone of strategic stability” and withdrew from the negotiations with Russia then seeking to establish a cooperative basis for jointly building a global defense—as proposed by then-President Boris Yeltsin, it assured that ongoing programs would not build an effective national missile defense. Furthermore, because of the Treaty, our theater missile defenses have not been permitted to be all they can be.
Finally, Mr. Chairman, I wish to emphasize that the Clinton Administration’s programs are not responsive to the growing threat of ballistic missiles that can be used to blackmail or attack the United States. In particular, the bipartisan Rumsfeld Commission authoritatively and unanimously warned in 1998 that the United States might be so threatened within five year—by 2003—by one of several nations. None of the Clinton Administration development programs will do anything to end by that date America’s vulnerability to even a single ballistic missile. This is a dangerous situation, which should be urgently addressed.
I believe we should put behind us the era of dumbing down systems to be consistent with our perceptions of the constraints imposed by ambiguous Cold War Treaty terms. We should declare our intention to abandon our policy of adhering to the ABM Treaty, and to build the most effective ballistic missile defenses we can as soon as we can. If Russia wishes to cooperate—perhaps, as Russian President Putin has suggested, on building a boost-phase intercept capability to protect all of NATO and other allies and friends—then that would be a welcome development. But we should end out adherence to the ambiguous terms of the ABM Treaty, as is our every right.
If that step is taken, I believe that sea-based and space-based defenses are the most effective defense that we can build quickly—beginning operations as early as in 2003. I call your attention to Appendices A and B for the reasons why I believe a “first from the sea, and then from space” strategy is called for to provide the most effective, least expensive defenses—which, perhaps surprisingly, could begin protecting America so quickly.