Since it assumed office, the administration of President George W. Bush has made missile defense one of its top priorities, giving it prominence in policy, funding, and organization.
First, the administration outlined an ambitious set of goals that extend well beyond the Clinton administration’s missile defense aims. In early January 2002, Secretary of Defense Donald Rumsfeld described the administration’s top missile defense objectives this way: “First, to defend the U.S., deployed forces, allies, and friends. Second, to employ a Ballistic Missile Defense System (BMDS) that layers defenses to intercept missiles in all phases of their flight (i.e., boost, midcourse, and terminal) against all ranges of threats. Third, to enable the Services to field elements of the overall BMDS as soon as practicable.”
Then, in its nuclear posture review, the administration outlined the specific elements of a national missile defense that it wants to have ready between 2003 and 2008: an air-based laser to shoot down missiles of all ranges during their boost phase; a rudimentary ground-based midcourse system, a sea-based system with rudimentary midcourse capability against short- and medium-range threats; terminal defenses against long-range ICBMs capable of reaching the United States; and a system of satellites to track enemy missiles and distinguish re-entry vehicles from decoys.
Finally, to speed implementation, the administration has taken a number of tangible steps. It announced on December 13, 2001, that the United States would withdraw from the 1972 Anti-Ballistic Missile (ABM) Treaty, ostensibly because the treaty was restricting testing of mobile missile defenses against ICBMs. In its first defense budget, the administration requested a 57 percent increase in funding for missile defense—from $5.3 billion to $8.3 billion, of which it received $7.8 billion. Then, Rumsfeld reorganized the Ballistic Missile Defense Organization into the new Missile Defense Agency, cancelled the internal Pentagon documents that had established the program’s developmental goals, and changed the program’s goal from being able to field a complete system against specific targets to simply being able to field various missile defense capabilities as they become available.
All in all, a lot has happened in missile defense in the first year or so of the Bush administration. But have these actions brought the United States any closer to realizing its missile defense goals, especially deployment of a national missile defense? And what elements, if any, of a national missile defense capability might it be possible for the United States to deploy by 2008, as called for in the nuclear posture review?
Despite the Bush administration’s push for missile defense, the only system likely to be ready by 2008 is a ground-based theater missile defense intended to counter short-range targets—i.e., a system to defend troops in the field. Before Bush leaves office, the only system that could conceivably be ready to defend the United States itself is the ground-based midcourse system pursued by the Clinton administration. None of the other elements mentioned in the nuclear posture review as possible defenses against strategic ballistic missiles is likely to be available by 2008.
To understand why, let us examine each of the missile defense programs—starting with the short-range, theater missile defense systems and moving to the longer-range, strategic systems—to see what has happened since the Bush administration took office 16 months ago. The results suggest that the Bush administration should not base its foreign policy on the assumption that during its tenure it will be able to deploy defenses to protect the United States from strategic missiles.
Theater Missile Defenses
Each of the U.S. military services has been pursuing tactical missile defense programs designed to defend U.S. troops overseas. None of these programs was designed to defend the United States against ICBM attacks, and none has any current capability to do so. However, the administration hopes to be able to apply some of the technology from these service programs to a layered national defense capable of defending the U.S. homeland. (For an explanation of the various stages of development discussed below, see the box below.)
The Patriot Advanced Capability-3 (PAC-3) is a tactical system designed to defend overseas U.S. and allied troops in a relatively small area against short-range missile threats (such as Scuds), enemy aircraft, and cruise missiles. Developmentally, it is the most advanced U.S. missile defense system, and a small number have been made available for deployment although testing has not yet been completed.
PAC-3 flight testing began in 1997. From 1997 to 2002, 11 developmental flight tests were conducted, including four flight intercept tests with two or three targets being attempted at once. Most of these tests were successful, but in two of the tests one of the targets was not intercepted. In February, PAC-3 began initial operational testing, in which soldiers, not contractors, operate the system. Three operational tests have been conducted, all with multiple targets. In each, one of the targets has been missed or one of the interceptors has failed.
A year ago, PAC-3 was planned to begin full-rate production at the end of 2001. However, problems with system reliability and difficulties in flight intercept tests have delayed that schedule. This means that full-rate production likely will be delayed until more stressing “follow-on” operational tests can be conducted against targets flying in a wide range of altitudes and trajectories. In March, Lieutenant General Ronald Kadish, who heads U.S. missile defense programs, testified to Congress that the full-rate production decision would be made toward the end of 2002 (before operational testing has been completed), representing a delay of about a year since last year. The full system will be deployed once all operational testing has been completed, perhaps around 2005.
A future version of PAC-3 is being considered for terminal defense of the United States. However, PAC-3 was not designed to counter long-range threats, and no flight intercept tests have been conducted to demonstrate how it might be incorporated in a terminal defense layer. Further, the ground area that can be defended by PAC-3 is so small that it would take scores of systems to defend just the major U.S. cities. A version of PAC-3 that could be effective in a national missile defense is probably a decade away.
The Theater High Altitude Air Defense (THAAD) system is designed to shoot down short- and medium-range missiles in their terminal phase. THAAD would be used to protect forward-deployed troops overseas as well as nearby civilian populations and infrastructure. THAAD is to defend a larger area against longer-range threats than PAC-3, but it is not designed to protect the United States from ICBMs.
From 1995 to 1999, 11 developmental flight tests were performed, including eight in which an intercept was attempted. After the first six of those flight intercept tests failed, the program was threatened with cancellation. Finally, in 1999, THAAD had two successful flight intercept tests. The THAAD program has not attempted an intercept test since then, instead focusing on the difficult task of developing a new, more reliable, higher-performance missile than the one used in early flight tests.
A year ago, full-rate production was scheduled to begin in 2007 or 2008, but because there were no intercept tests in 2000 or 2001, that schedule has likely slipped two years or more. In fact, no flight intercept test is scheduled until 2004, and it is therefore unlikely that the first THAAD system will be deployed before 2010.
The Bush administration is considering THAAD for use in a layered national missile defense system. Conceptually, THAAD might be used in conjunction with PAC-3 as part of a terminal defense, or it could be deployed overseas to intercept enemy missiles in the boost phase. However, in its current configuration THAAD is incapable of performing these missions—even once it has met its Army requirements for theater missile defense—and therefore a role for THAAD in national missile defense is probably more than a decade away.
Navy Area Theater Ballistic Missile Defense
The Navy Area Theater Ballistic Missile Defense was the sea-based equivalent of PAC-3. The Navy Area system was being designed to defend forward-deployed Navy ships against relatively short-range threats. But in December 2001 the program was cancelled because its cost and schedule overruns exceeded the limits defined by law. (Ironically, the cancellation came just one day after President Bush announced that the United States would pull out of the ABM Treaty because its missile defense testing was advanced enough to be bumping up against the constraints of the treaty.)
The Navy still wants to be able to defend its ships against missile attack, and the program will most likely be restructured and reinstated once the Navy decides on a new approach. In the meantime, the Navy Area program is slipping with each day that passes. As with PAC-3, the Bush administration has considered extending the Navy Area system to play a role in the terminal segment of a layered national missile defense. However, at this point the program is too poorly defined to allow speculation about when it could accomplish such a demanding mission.
Navy Theater Wide
The Navy Theater Wide program was originally intended to defend an area larger than that to be covered by the Navy Area system—that is, aircraft carrier battle groups and nearby territory and civilian populations—against medium-range missiles during their midcourse phase. In this sense, Navy Theater Wide is the sea-based equivalent of THAAD.
In January, the Navy Theater Wide program conducted its first successful flight intercept test, but a dozen or more developmental flight tests will be required before it is ready for realistic operational testing. About a year ago, full-rate production was scheduled for spring 2007, meaning that the system could be deployed before the end of the decade.
But since then, the Pentagon has given new priority to a sea-based role in defending the U.S. homeland. Navy Theater Wide was not designed to shoot down ICBMs, but the Bush administration has restructured the program so that it aims to produce a sea-based midcourse segment and/or a sea-based boost-phase segment of national missile defense.
Either mission will require a new missile that is twice as fast as any existing version of the Standard Missile, which the system now uses; a new, more powerful Aegis radar system to track targets; a new launch structure to accommodate the new, larger missiles; and probably new ships. As a result, the Navy Theater Wide program requires a great deal of new development. It is unlikely that Navy Theater Wide will be ready for realistic operational testing until late in this decade, and it will not be ready for realistic operational demonstration in a layered national missile defense for several years after that.
The Airborne Laser (ABL) is a program to develop a high-power chemical laser that will fit inside a Boeing 747 aircraft. It is the most technically challenging of any of the theater missile defense programs, involving toxic materials, advanced optics, and the coordination of three additional lasers on-board for tracking, targeting, and beam correction. The first objective of the program is to be able to shoot down short-range enemy missiles. Later, it is hoped the ABL program will play a role in national missile defense by destroying strategic missiles in their boost phase.
The ABL has yet to be flight-tested. About a year ago, full-rate production of the ABL was scheduled for 2008. The plan was to build seven aircraft, each estimated to cost roughly $500 million. At that time, the first shoot-down of a tactical missile was scheduled for 2003. Recently, the ABL program office announced that the first shoot-down of a tactical missile had been delayed to late 2004 because of many problems with the basic technology of high-power chemical lasers—about a one-year slip since last year and about a three-year slip since 1998. Accordingly, full-rate production probably cannot be started before 2010, and the cost will likely exceed $1 billion per aircraft.
Assuming all this can be done, it is important to note that the ABL presents significant operational challenges. The ABL will need to fly relatively close to enemy territory in order to have enough power to shoot down enemy missiles, and during a time of crisis it will need to be near the target area continuously. A 747 loaded with high-power laser equipment will make a large and inviting target to the enemy and will require protection in the air and on the ground. Finally, relatively simple countermeasures such as reflective surfaces on enemy missiles could negate the ABL’s capabilities.
Deployment of an ABL that can shoot down short- and medium-range tactical targets is not likely before the end of the decade, and the Airborne Laser will not be able to play a role in national missile defense for many years after that.
National Missile Defense
The Bush administration hopes to build a layered national missile defense that consists of a ground-based midcourse system, expanded versions of the theater systems discussed above, and, potentially, space-based systems. The Bush administration does not use the phrase “national missile defense” because it was the name of the ground-based midcourse system pursued by the Clinton administration and because the Pentagon’s plans to defend the country are now more robust. But national missile defense is a useful shorthand for any system that is intended to defend the continental United States, Alaska, and Hawaii against strategic ballistic missiles, and it is in that sense that it is used here.
For all practical purposes, the only part of the Bush national missile defense that is “real” is the ground-based midcourse system. It is real in the sense that six flight intercept tests have been conducted so far, whereas versions of the THAAD or Navy Theater Wide systems that might be used to defend the United States have not been tested at all. Space-based systems are an even more distant prospect. For example, the Space-Based Laser, which would use a laser on a satellite to destroy missiles in their boost phase, was to be tested in 2012, but funding cuts have pushed the testing date back indefinitely. Deployment is so far in the future that it is beyond the horizon of the Pentagon’s long-range planning document, Joint Vision 2020.
As a result, despite the Bush adminis-tration’s attempts to distinguish its plans from its predecessor’s, Bush’s layered national missile defense is, in effect, nothing more than the Clinton system.
Since 1997, the ground-based midcourse program has conducted eight major flight tests, known as IFTs. The first two, named IFT-1A and IFT-2, were fly-by tests designed simply to collect target information. The next six tests, IFT-3 through IFT-8, were all flight intercept tests. IFT- 4 and IFT-5, conducted in January 2000 and July 2000 respectively, both failed to achieve an intercept, which became a principal reason why, on September 1, 2000, President Bill Clinton decided not to begin deployment of ground-based midcourse components, such as a new X-band radar on Shemya Island in Alaska.
Another year passed before the next flight intercept test, IFT-6, was conducted. The intercept was successful except that the real-time hit assessment performed by the ground-based X-band prototype radar on the Kwajalein Atoll in the Marshall Islands incorrectly reported the hit as a miss. IFT-7, conducted in early December 2001, was also successful. Until then, all of the flight intercept tests had had essentially the same target cluster: a re-entry vehicle, a single large balloon, and debris associated with stage separation and decoy deployment. Then, in IFT-8, conducted on March 15, 2002, two small balloons were added to the target cluster. This flight intercept test also was successful and marked an important milestone for the ground-based midcourse program.
However, despite these recent successes, there have been significant delays in the testing program. Several of the flight tests were simply repeats of earlier tests, and as a result IFT-8 did not accomplish the tasks set for it in the original schedule. In short, the testing program has slipped roughly two years—i.e., what was originally scheduled to take two years has taken four. This is not to say that the program has made no progress but rather that key program milestones have receded into the future.
The pace of successful testing will be one of the primary determinants of how quickly the United States can field a national missile defense. If the ground-based midcourse system has three or four successful flight intercept tests per year, as it has during the past year, it could be ready for operational testing in four or five years. If those operational tests also were successful, then whatever capability had been demonstrated in all those tests—which would probably not include the capability to deal with many types of decoys and countermeasures or the capability to cover much of the space through which an enemy missile could travel—could be deployed by the end of the decade or even by 2008.
However, the ground-based midcourse system has difficulties beyond the testing pace of its interceptor. The system requires a new, more powerful booster rocket than the surrogate currently being used in tests—a task that was thought to be relatively easy. That new booster was to be incorporated into the continuing series of flight intercept tests to make those tests more realistic and to be sure that the new booster’s higher acceleration did not adversely affect other components or systems on board.
But development of the new booster is about two years behind schedule. Indeed, on December 13, just hours after President Bush announced U.S. plans to withdraw from the ABM Treaty, a test of the new booster had to be aborted and the missile destroyed in flight for safety reasons because it flew off course. Flight intercept tests that were to have used the new booster have come and gone without it. Indeed, development of the booster is so far behind that the Pentagon recently issued another contract for a competing design.
Equally problematic is uncertainty over how the system will track enemy missiles in flight and distinguish targets from decoys. One approach is to use high-power radars operating in the X-band (that is, at a frequency of about 10 billion cycles per second). A prototype X-band radar on the Kwajalein Atoll has been part all of the ground-based midcourse flight intercept tests so far, and technically, X-band radar progress has been one of the most successful developments in missile defense technology.
A year and a half ago, Lieutenant General Kadish testified to Congress that establishing an X-band radar in Alaska was the “long pole in the tent” for missile defense. This meant that the X-band radar was critical to a ground-based midcourse system and that if that radar was not built soon, the program would start slipping day for day. Then, as now, there were many other developments that would take as long or longer than building an X-band radar at Shemya, but the Pentagon’s official position was that construction needed to start in the spring of 2001 at the latest. Nevertheless, Clinton deferred taking action on the radar.
Surprisingly, the Bush administration has not requested funding for an X-band radar at Shemya in either of its first two budgets. This may be because the administration views such an installation as inconsistent with the ABM Treaty, which the administration has said it will not violate while the treaty is still in effect. Or the administration may not have requested funding because the Missile Defense Agency has been exploring “portable” X-band radars—that is, X-band radars deployed on ships or barges.
Some defense analysts believe that the Space-Based Infrared Satellite (SBIRS) program could be used in place of the X-band radar to assist a national missile defense. SBIRS—which would consist of two sets of orbiting sensor satellites, SBIRS-high and SBIRS-low—is designed to detect the launch of enemy ballistic missiles and could be used to track and discriminate among them in flight. However, the program has significant technical problems.
SBIRS-high, which will consist of four satellites in geosynchronous orbit and two satellites in highly elliptical orbits, is to replace the existing Defense Support Program satellites, which provide early warning of missile launches. A year ago, the SBIRS-high satellites were scheduled for launch in 2004 and 2006, but recently those dates have slipped roughly two years because of problems with software, engineering, and system integration. A year ago, realistic operational testing was scheduled for 2007; now, it may not occur this decade, which means that full deployment may not occur this decade. SBIRS-high is also well over cost and is in danger of breaching the legal restrictions covering cost growth.
SBIRS-low is to consist of approximately 30 cross-linked satellites in low-Earth orbit. A year ago, the launch of the first of these satellites was scheduled for 2006, but SBIRS-low has slipped two years because of a variety of difficult technical problems. The developmental testing program for SBIRS-low is very challenging, and realistic operational testing will probably not begin this decade. This could delay deployment of the full constellation of SBIRS-low satellites until the middle of the next decade. SBIRS-low is also dramatically over budget and was threatened with cancellation in the latest round of congressional appropriations.
For now, the administration has been saying that it will upgrade an existing radar on Shemya called Cobra Dane. Under this plan, the Cobra Dane radar would become an advanced early-warning radar with some ability to distinguish among targets. But the Cobra Dane radar operates in the L-band with about eight-times poorer resolution than a new X-band radar would have, raising questions about the effectiveness of any national missile defense using it.
In sum, the only element of a “layered” national missile defense that exists on anything but paper is the ground-based midcourse system pursued by the Clinton administration. Accordingly, it is nearly impossible to predict when, if ever, an integrated, layered national missile defense with boost, midcourse, and terminal phases might be developed. As noted above, given the most recent pace of testing, some part of the ground-based midcourse system could be deployed by the end of the decade or possibly by 2008.
However, the capability such a system would have would be marginal and probably would not be able to deal with many types of decoys and countermeasures or to cover much of the space through which an attacking ICBM might fly. The Bush administration has said it will deploy test elements as an emergency capability as early as possible, but such a deployment would be rudimentary and its capabilities would be limited to those already demonstrated in testing. It would likely not be effective against unauthorized or accidental launches from Russia or China, which might include missiles with countermeasures. It also would not be effective against launches from Iraq, Iran, or Libya since those countries are to the east, out of view of a radar on Shemya.
During the first year of the Bush administration, all U.S. missile defense programs—both theater and national—have slipped. In general, the shorter-range tactical missile defense systems are further along than the medium-range systems, and those medium- range systems are further along than the longer-range systems intended to defend the United States against ICBMs.
PAC-3 is the most developmentally advanced of any U.S. missile defense system, but full deployment will not likely take place before 2005, and realistic operational testing will continue for many years after the first Army units are equipped in the field. The THAAD program has slipped two years or more and will not be deployable until 2010. The Navy Area Wide program has been cancelled, and the Navy Theater Wide program has slipped two years or more and will not be deployable in a tactical role until the end of the decade. If the Pentagon restructures the program so that its priority is boost-phase or midcourse defense against strategic missiles, it will likely take longer. The Airborne Laser has slipped one year and will probably not be deployed as a theater missile defense before the end of the decade.
SBIRS-low has slipped two years and doubled in cost and probably will not be deployed before 2008.
For all practical purposes, national missile defense is technically not much closer than it was in the Clinton administration. There have been no flight intercept tests of the boost-phase or terminal-phase elements suggested by the Bush administration, and developmental testing could take a decade or more, depending on the pace of testing and the level of success in each test. The only element that can be flight-intercept tested against strategic ballistic missiles today is the ground-based midcourse system. Part of that system could be deployed by 2008, but elements fielded before then will have only a limited capability.
Thus, while making foreign policy, the Bush administration would do well to consider that probably only a limited-capability version of PAC-3 will be fielded during its tenure and that an effective, layered national missile defense will not be realized while it is in office. It would make little sense to predicate strategic decisions on a defense that does not exist.
It is important for Congress and the American public not to be frightened into believing that the United States is—as some missile defense proponents like to assert—defenseless against even a limited missile attack by a “rogue state” such as North Korea. Powerful and effective options exist, both military and diplomatic.
In Afghanistan, U.S. attack operations with precision-guided weapons have been highly effective. Those same precision weapons would be effective against an enemy ICBM installation. In fact, given current capabilities and the ever-improving technologies for precision strike, it would be fantasy to believe any national missile defense system deployed by 2003 to 2008 would work better and provide greater reliability at a lower cost than the precision-guided munitions used in Afghanistan.
On the diplomatic front, in 1999 former Secretary of Defense William Perry made a series of trips to convince North Korea to stop developing and testing long-range missiles. He was remarkably successful. Although Secretary Perry would not say that North Korea was no longer a threat, it was obvious that the North Korean threat had been moderated. Secretary of State Madeleine Albright was able to build on his trip the next year to secure a pledge from Pyongyang to halt flight testing of missiles. Dollar for dollar, Secretary Perry has been the most cost-effective missile defense system the United States has yet to develop. The most straightforward route to missile defense against North Korea may be through diplomacy, not technology.
Many decision-makers in Washington—and, from what one reads, the president himself—seem to be misinformed about the prospects for near-term success with national missile defense and the budgets being requested for it. It takes 20 years to develop a modern, high performance jet fighter, and it probably will take even longer to develop an effective missile defense network. Taking into account the challenges of asymmetric warfare, the time it can take to develop modern military equipment, the reliability required in real operational situations, and the interoperability required for hundreds of systems and subsystems to work together, it would be highly unrealistic to think that the United States can deploy an effective, layered national missile defense by 2004 or even by 2008.
In the meantime, policymakers should be careful that U.S. foreign and security goals and policies are not dependent on something that cannot work now and probably will not work effectively for the foreseeable future. A case in point is President Bush’s decision to abandon the ABM Treaty with Russia. That decision was certainly premature given the state of missile defense technology and likely could have been avoided or postponed for many years if not indefinitely.
This is not to say that missile defense technology ought not to be pursued—only that it should be pursued with realistic expectations. Policymakers must be able to weigh the potential merits and costs of missile defense based on a sound understanding of both the technology and the possible alternatives. No one weapon system can substitute for the sound conduct of foreign policy, and even a single diplomat can be effective on a time scale that is short when compared with the time that will be required to develop the technology for national missile defense.
Stages of Development
|Missile defense, especially national missile defense, is the most difficult program ever attempted by the Department of Defense—much more difficult than the development of a modern jet fighter like the F-22 Raptor, the Navy’s Land Attack Destroyer (DD-21), or the Army’s Abrams M1A2 tank complete with battlefield digitization, endeavors that all have taken 20 years or more. Each new major weapons system must proceed through several stages of development, which are listed below. Most U.S. missile defense systems are currently in developmental testing and are therefore not close to deployment.
Research and Development (R&D): The period during which the concepts and basic technologies behind a proposed military system are explored. Depending on the difficulty of the technology and the complexity of the proposed system, R&D can take anywhere from a year or two to more than 10 years.
Engineering and Manufacturing Development (EMD): The period during which a system design is engineered and the industrial processes to manufacture and assemble a proposed military system are developed. For a major defense acquisition such as a high-performance jet fighter, EMD can take five years or more. If substantial difficulties are encountered, EMD can take even longer.
Developmental Testing: Testing that is performed to learn about the strengths and weaknesses of proposed military technologies and the application of those technologies to a new military system in a military environment. Generally, developmental testing is oriented toward achieving certain specifications, such as speed, maneuverability, or rate of fire. Developmental testing is conducted throughout the R&D and EMD phases of development and becomes more stressing as prototype systems evolve and mature.
Operational Testing: Testing that aims to demonstrate effective military performance against operational requirements and mission needs established for a system. Testing is performed with production-representative equipment in realistic operational environments—at night, in bad weather, against realistic threats and countermeasures. Military service personnel, not contractors, operate the system, which is stressed as it would be in battle. Operational testing of a major defense acquisition system typically takes the better part of a year and is usually broken into several periods of a month or two to accommodate different environments or scenarios. If substantial difficulties are encountered, several years of operational testing may be required.
Production: The phase of acquisition when a military system is manufactured and produced. Early on, during “low-rate production,” the quantities produced are typically small. Later, after successfully completing operational testing, a system may go into “full-rate production,” where the rate of production is designed to complete the government’s planned purchase of the system in a relatively short period of time, about five years.
Deployment: The fielding of a military system in either limited or large quantities in military units. The first military unit equipped may help develop tactics, techniques, and procedures for use of the new system if that has not already been done adequately in development.—P.C.
Flight Stages of Ballistic Missiles
|All ballistic missiles have three stages of flight.
The boost phase begins at launch and lasts until the rocket engines stop firing and pushing the missile away from Earth. Depending on the missile, this stage lasts three to five minutes. During much of this time, the missile is traveling relatively slowly although toward the end of this stage an ICBM can reach speeds of more than 24,000 kilometers per hour. The missile stays in one piece during this stage.
The midcourse phase begins after the propulsion system finishes firing and the missile is on a ballistic course toward its target. This is the longest stage of a missile’s flight, lasting up to 20 minutes for ICBMs. During the early part of the midcourse stage, the missile is still ascending toward its apogee, while during the latter part it is descending toward Earth. It is during this stage that the missile’s warhead, as well as any decoys, separate from the delivery vehicle.
The terminal phase begins when the missile’s warhead re-enters the Earth’s atmosphere, and it continues until impact or detonation. This stage takes less than a minute for a strategic warhead, which can be traveling at speeds greater than 3,200 kilometers per hour.—ACA
Philip Coyle, a senior advisor at the Center for Defense Information, was assistant secretary of defense and the Pentagon’s director of operational test and evaluation from 1994 to 2001.