| ||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||
| Copyright © 2009. National Academy of Sciences. All rights reserved. Terms of Use and Privacy Statement |
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 182
Page 182
APPENDIX H
Theater Missile Defense, National IBM Systems, and the Future of
Deterrence
Richard L. Garwin, IBM Thomas J. Watson
Research Center
Mch has changed over the decades in regard to the desirability
of and the capability for defense against ballistic missiles. Now
attention to ballistic missile defense (BMD) in the United States
is directed toward theater missile defense (TMD), nominally for
defense against short- and intermediate-range ballistic missiles
(up to 3,000 km or so) armed with nuclear, explosive, chemical, or
biological warheads.
Defense is proposed to extend to U.S. forces abroad, allied
forces, and the capitals or cities of friendly or allied nations.
In addition, it is proposed to use the same technology to have a
light defense of U.S. territory itself against "accidental launch"
by Russia, or against small attack by a so-called "rogue"
state.
But much has remained the same.
CONTEXT
For three decades the United States has had no defense against
the nucleararmed ballistic missile force of the Soviet Union,
relying instead on deterrence by threat of nuclear retaliation.
Repeatedly we made an effort to structure an effective defense, but
the sole deployment decision (Safeguard) ultimately promised no
effective defense even of our strategic offensive force, but rather
provided a testbed to perfect the antiballistic missile (ABM)
software.
Nevertheless, had it been technically feasible to deploy a
highly effective and durable ABM, we would have chosen to do so.
And had the Soviet Union been able to field an "effective" ABM, the
United States would have responded by expanding its missile force
to overwhelm the defense, enhancing the force with penetration aids
to defeat the defense, or underflying or bypassing the defense with
cruise missiles, bombers, or other nonballistic missiles. Of
course, a truly dominant defense that could not be overcome,
underflown, or otherwise bypassed would be another matter
entirely.
The no-ABM case is simple to analyze—on both sides. Its
perceived problems include the clear reliance on deterrence of
attack by threat of retaliation rather than on protection in case
of attack—hence vulnerability to accidental or unauthorized
launch. However, deterrence by means of retaliation can be defeated
only by defenses that counter all the means of destruction and
penetration available to the other side. It is essential to
understand that military systems have to deal with an intelligent
adversary rather than a predefined threat; the perfect ABM may fail
catastrophically, or it might be bypassed.
OCR for page 183
Page 183
An antiballistic missile defense with the goal of denying the
other side's deterrence by means of nuclear retaliation against the
society itself—population and industry—has an
extraordinarily difficult task. Fundamentally, this difficulty
arises because the enormous destructive power of a nuclear weapon
means that a half-megaton weapon could kill a half million people;
a weapon plus delivery system costing $10 million to $100 million
could destroy value on the order of $500 billion (assuming $1
million per life). Another metric compares how much it costs the
offense to overcome the defense and, ultimately, the cost exchange
ratio between offense and defense—not for holding the damage
precisely constant, but retaining a similar magnitude of
damage.
In general, defense can be very costly if the requirement is to
maintain near-perfect protection against a responsive adversary.
The offense, for instance, can choose the specific target; can
exhaust the local defense with warheads, dummies, or decoys; or can
use enough weapons to leak through the imperfections if not to
overwhelm the defense; or can attack the defense specifically ("the
eyes of the system," for instance, which are often more fragile or
more visible than the targets themselves).
This essay is not itself a book on the future of deterrence and
warfare, centered on missile defense; rather, it is a sketch of the
current situation regarding tactical and strategic missile defense,
with indications of the relationship to deterrence and warfare.
BACKGROUND
This essay is informed by the author's involvement with
strategic offensive and defensive forces since 1952 and with every
successive generation of proposal or deployment of ballistic
missile defenses.
Indeed, the revolution in microelectronics, radio frequency
technology, and signal processing has wrought a revolution in the
reliability and effectiveness of radar detection of objects in
space or in the atmosphere, and this has been augmented by major
advances in optical detection capability both in the visible and
the infrared (IR).
So it is commonplace in the United States or elsewhere to read
about or to see videos and photographs of test intercepts taking
place in the vacuum of space, or in the atmosphere.
In the 1950s or thereabouts, effective intercept could be
conceived only with a nuclear-armed interceptor, and the one
strategic ABM system briefly deployed in the United States
(Safeguard) was equipped with low-yield nuclear warheads on its
short-range interceptor. The exo-atmospheric interceptor was to be
equipped with a multimegaton warhead not only to compensate for
inaccuracy in intercept but also to be able to destroy spaced
warheads and decoys.
OCR for page 184
Page 184
But just as the detection capability has improved, so has the
ability to conceal the target (stealth), to mimic the target
(especially with the aid of antisimulation), and to jam the
detection radars or the fuze of the interceptor itself. These
techniques have been explored quite thoroughly by the United States
for strategic missiles, and some of them for aircraft, but it is
not clear how well perfected they are for short-range or tactical
ballistic missiles.
In addition to the pure BMD systems, initially derived in any
case from the army surface-to-air missile systems (SAMs),
dual-purpose systems have been in favor especially for theater
defense. Thus the Patriot that saw service in Israel and Saudi
Arabia in 1991 was fielded primarily as a SAM system and not an ABM
system.
Much has been written about the performance of the Patriot,
beginning with the claim of essentially hundred percent
"effectiveness" in countering the Scud missiles launched by Iraq.
My own judgment, largely based on a close reading of the analyses
of Dr. T.A. Postol and his critics, is that few, if any, warheads
were destroyed or disabled by Patriot interceptors.
This would be of little import if the success of the Patriot
against Scuds were not used as a baseline by many in arguing for
the effectiveness of future TMD and BMD systems. To the extent that
the baseline is important, these arguments fall on their face. But
even if the Patriot did not work at all, this does not mean that
future systems would be ineffective.
In addition to the Patriot, the Soviet (now Russian) SA-10,
SA-12, and S300 systems could have some capability against
ballistic missiles. How good a capability? And how good is good
enough?
THE PROBLEM
The problem is not to "hit a bullet with a bullet," a problem
that was solved long ago. That requires only detecting the incoming
"bullet" at a sufficient standoff to be able to get an interceptor
into its neighborhood before impact (or perhaps before reentry into
the atmosphere), and guiding the interceptor so that it arrives
with sufficient precision to a point on the trajectory of the
incoming warhead (and at the correct time) to collide with or for
the interceptor warhead to explode so as to destroy or disable the
incoming warhead.
Intercept is easier if the interceptor can climb along the
inverse trajectory, in which case precise control of its velocity
is less important to the success of the intercept. Otherwise, for
"crossing intercepts" the interceptor must be steered as a function
of its velocity and drag to make the intercept at one point or
another along the trajectory of the incoming missile.
A major problem for intercepts outside the atmosphere is that it
costs very little for the offense to provide "penetration aids" or
countermeasures, particularly simple against non-nuclear
interceptors.
For instance, a large balloon surrounding the reentry vehicle or
missile would deny the interceptor the ability to detect precisely
the location of the
OCR for page 185
OCR for page 186
OCR for page 187
OCR for page 188
OCR for page 189
OCR for page 190
OCR for page 191
OCR for page 192
OCR for page 193
OCR for page 194
OCR for page 195
OCR for page 196
OCR for page 197
OCR for page 198
OCR for page 199
OCR for page 200
Representative terms from entire chapter:
missile defense
Page 185
vulnerable warhead within the balloon.1 A smaller balloon around the warhead
accompanied by other balloons at some distance (either tethered or
free flying) would require the defense to destroy all of these with
interceptors, or to have some preliminary balloon-destroying
interceptor followed by an assessment and intercept of the real
warhead, or to attempt to discriminate between the balloon
containing the warhead and the balloons that are empty. These
particular penetration aids are simple only when the light and
heavy objects are all in "free fall" and they would be inapplicable
to maneuvering portions of the trajectory or within the
atmosphere.
The "counter countermeasure" of the defense could be this
required enhanced discrimination capability, but that is an
extremely fragile option. At the very least, electrical heaters on
the balloons could mimic the residual heat from a warhead to
confound infrared sensors, as a counter counter countermeasure to
this response.
But a much more powerful penetration aid is to be found in the
technique of "antisimulation" in which the warheads themselves are
given a broad spectrum of observables, in order to make them easier
to mimic by inexpensive decoys.
Of course one could imagine an adversary with sufficient
blindness and specific limitations in technology to be able to buy
or make ballistic missiles and their warheads, but with a peculiar
inability to make these penetration aids.
I don't think so.
In view of the ease of countering intercept outside the
atmosphere, most of the serious proposals for intercept deal with
the incoming warhead during the reentry phase, when light balloons
are stripped away by atmospheric drag, and the dynamic pressure
makes it more difficult to mount penetration aids on the warhead
itself.
But such endo-atmospheric intercept poses its own problems for
the defense (especially for interceptors not armed with nuclear
warheads) since the trajectory of the threat is affected by drag,
and because the threat has now the option of substantial maneuvers,
by interaction with the very dynamic pressure that causes the drag.
If the missile is very accurate, the preservation of that accuracy
while incorporating "substantial maneuvers" can be a serious
problem, however.
Similarly, the extrapolated position of the interceptor is
affected by its own drag, and the control is not so simple as it is
in the exo-atmospheric intercept. Still, successful
endo-atmospheric intercepts have been made in tests, either with
fragment-kill warheads or hit-to-kill warheads. These latter make
use of the fact that each gram of an interceptor at 6 km/s closing
speed has some four times the energy of a gram of high
explosive.
The Iraqi Scuds demonstrated this major problem associated with
endoatmospheric intercept, in that the range extension of the Scud,
done by the Iraqis
1 Such a balloon, of
itself, would not reduce the effectiveness of an interceptor armed
with a nuclear warhead, but it would totally defeat an interceptor
that was effective only in actual collition with the offensive
reentry vehicle.
Page 186
themselves, involved a lengthening of the missile tank, which
led to instability and breakup on reentry. Thus the incoming
high-explosive warheads were maneuvering in a tight helix, while
the Patriot had no specific software to help it make an intercept
under those circumstances. Furthermore, it is clear that the fuzing
option for the Patriot was far from optimum for the closing
velocities that were involved.
The designer of the Soviet SA-10 system remarks that his system
does have a more flexible fuzing option, as would any future
system.
NEAR-TERM OPTIONS FOR U.S. THEATER
MISSILE DEFENSE
The widely deployed 1960s-origin U.S. Army Hawk SAM system can
have an option against TBM. In the 1960s it was proposed to upgrade
the Hawk system for defense against Soviet SS-9 ICBMs that might be
used in nuclear attacks on Minuteman silos. And they would have
worked for that limited purpose. In fact, the Ballistic Missile
Defense Office (BMDO—successor to the Strategic Defense
Initiative Office [SDIO]) is expected to spend some $60 million to
upgrade the ABM capability of the Marines Corps HAWK air defense
system.
WHAT IS THE THREAT?
The military effectiveness of inaccurate high-explosive (HE)
warheads against our military forces in the field is negligible.
(As a case in point, the cost of perfect defense against this
negligible threat is high—illustrating the strong dependence
of cost of defense vs. demanded effectiveness.) Furthermore, the
threat of precision HE-armed theater missiles to our forces in the
field could be countered by intercepts at a kilometer distance
rather than by interceptors that need to cover the entire area, at
substantial expense and uncertain results. The threat of chemical
or biological weapons to deployed U.S. forces is not much greater
than that of HE-armed weapons, in view of the available defensive
clothing and decontamination measures, and there is, for the
present, no significant threat of nuclear-armed missiles. Against
biological warfare (BW) and chemical warfare (CW), passive
protection can be so effective that it would have a very
substantial effect of deterrence by reducing the value of such
weapons.
However, the threat of biological and chemical weapons against
friendly cities is far from negligible, although delivery by
ballistic missile is neither the greatest nor the most urgent
threat, and passive protection can do much there as well. We cannot
always count on people getting it as wrong as did Aum Shinrikyo in
its (chemical) attack on the Tokyo subways.
As shown by the destruction produced by a 2-ton explosive blast
in Oklahoma City, substantial human and property damage could be
done to modern cities by even high-explosive armed Scuds or other
missiles, although their impact would likely cause less damage on
average than that carefully
Page 187
placed van bomb. Nevertheless, a 300-kg Scud warhead destroyed
the Ministry of Education building in Riyadh on January 25,
1991.
Credibility and Responsibility
During the Desert Storm operation against Iraq in 1991, it was
announced by President George Bush that the effectiveness of the
Patriot missile in intercepting Scuds was almost 100 percent. It
seems to me that not only President Bush but also the Defense
Department and the U.S. Army must have believed this, and once the
words are out of the mouth of the President, there is a substantial
establishment devoted to establishing their truth or reality, as
was the case following the announcement of the Strategic Defense
Initiative by President Ronald Reagan on March 23, 1983.
Over the years since January 1991, I have discussed Patriot
performance with several diplomats and military officers of
friendly countries and have learned that they overwhelmingly
believe that Patriot did not conduct successful intercepts, but
that it was a "political response to a political weapon" and in
this regard was "very successful."
And one can hardly disregard historical facts presented, for
instance, by Alexander H. Flax.2By
late summer 1944, only one in every seven V-1 "buzz bombs" launched
by Germany against England survived to their targets, but the V-2
rocket attacks had begun. During July and August, Allied air forces
expended one-fourth of their total tonnage on missile-related
targets, and General Dwight D. Eisenhower recounted, "It seems
likely that, if the German had succeeded in perfecting and using
these new weapons six months earlier than he did, our invasion of
Europe would have proved exceedingly difficult, perhaps
impossible."
According to Eisenhower, with the chosen ground attack routes
into Europe, "In this way we would, incidentally, quickly clear the
area from which the V-1 and V-2 had been consistently bombarding
Southern England."
So weapons that might be better ignored (because more effective
military use could be made of the resources expended against those
weapons) did divert major resources and did affect decisions of the
military leaders substantially.
Perhaps the claim of a highly successful Patriot system kept
Israel from responding militarily against Iraq, which would have
complicated the military and political situation, to say the least.
But in the U.S. democracy we would be deceiving not only our
citizen-bosses but also our leaders themselves if we did not tell
the truth in such matters.
Although Winston Churchill remarked that "in wartime, truth is
so precious that she should always be attended by a bodyguard of
lies," that bodyguard is stifling not only to democracy but also to
the national security unless used only where strictly necessary. In
the case of Patriot as an antimissile system, insufficient
provision was made for gathering information on its
effectiveness.
2 Personal communication, April 14, 1995.
Page 188
When an aircraft is intercepted by Patriot, it normally crashes
into the ground, but since that is the purpose of the ballistic
missile, it is not so easy to tell the difference between an
intercepted missile and one that has not been touched.
THE BIG PROBLEM FOR CITY DEFENSE
A problem for ballistic missile delivery of chemical weapons
(CW) or biological weapons (BW) comes from the inherent difficulty
of disseminating CW or BW from a reentry vehicle from a missile
with a range of 500 to 3,000 km. It is far simpler to improve the
military effectiveness by early release of submunitions in the
ascent phase. In this way, submunitions weighing only a few
kilograms or so would be released by the hundreds, to have the
dispersion desired in the target area. Although desirable to the
offense from the point of view of military effectiveness, this
would also be an effective counter to any nonnuclear defense except
that operating before launch or during the ascent phase.
Indeed, the early release of submunitions totally counters the
performance of nonnuclear TMD systems thus far proposed, except
those that involve fastacceleration interceptor missiles launched
from close to the ballistic missile launch site so as to be able to
destroy the missile during powered flight. This could be achieved
by ground-emplaced interceptors (GEIs), or by launching the
interceptors from orbiting aircraft, or in principle by powerful
lasers in low Earth orbit, or by a large number of Brilliant Pebble
interceptors in orbit.
Effective launch-phase intercept is not a simple task, since it
requires intercept often within 40 s after launch. Limited in
initial acceleration by the necessity to moderate the heating and
dynamic pressure it would encounter in the atmosphere, a GEI
nevertheless could make an effective intercept if placed within 50
km of the launch site of a typical Scud, for instance.
An air-launched interceptor of 8 km/s could move 300 km during
that time, but a stealthy air vehicle or assumed air superiority
might be required. Russia might not agree in principle that
launch-phase intercept for TMD is compatible with the 1972
U.S.-Soviet ABM treaty, but possibly could agree on a specific
system that would be permitted by an amendment to that treaty and
that might be available to both sides.
As for active defense by counterforce, a recent report3 argues that the highest payoff comes
from counterforce attack against garrisons, depots, and command and
control facilities; the second highest payoff from attack on
transportation infrastructure and industrial facilities; and the
lowest payoff from attack on mobile missiles themselves and missile
launchers. This report considers the boost-phase intercept
alternative, but emphasizes that it should not be viewed as primary
or the preferred solution but constitutes a "mid-term to long-term
capability." Although boost-phase intercept is politically
difficult, I
3 Air Force Studies
Board, Counterforce Options Against Tactical Missile Systems (U).
National Academy Press, Washington, D.C., 1994
(Classified).
Page 189
emphasize that no midcourse or terminal capability such as those
now proposed for TMD or even for national ABM systems will handle
the motivating threat of BW and CW from bomblets dispersed on
ascent; so it is boost-phase or nothing.4
The United States proposed to interpret the ABM treaty as
permitting any system that had not actually been tested against
targets exceeding 5 km/s—an easing of constraints that
I am sure would lead the U.S. Congress to abandon the strategic
arms reductions, in view of the ABM capabilities that would then be
permitted and projected for Russia. A BMD capability deployed in
the United States to protect against "accidental launch" would need
to handle incoming reentry vehicles (RVs) of a full 7 km/s reentry
speed, but a system designed for those speeds and fully tested only
against a target of 5 km/s would not be inhibited by lack of
testing against the RVs of 7 km/s.
Testing does not "develop" the missile system; it just
challenges and perhaps verifies the model that was used to design
and develop the system. Thus, the radar tracking of 7 km/s objects
is verified independently of any intercept, and the IR detection by
the interceptor (or the radar detection by the interceptor) is a
function of interceptor speed but not adversely affected by
increased target speed. As for an all-up system "proof test," that
would not be available even for incoming RVs of a full 7 km/s
reentry speed, unless those RVs were supplied for test by the
adversary!5
Because Russia is more threatened by accidental launch and
intermediaterange missiles launched from its neighbors than is the
United States, a decision by the United States to proceed with such
a system would result in a comparable system in Russia, which would
then cause havoc with the assured penetration of strategic
ballistic missiles launched from British, French, or Chinese
forces, unless those forces were modified or greatly augmented. To
the extent that an ABM system depending on exo-atmospheric
intercept by nonnuclear armed interceptors is deployed,
countermeasures are relatively simple, and it is for that reason
that such a system is not even very useful against accidental or
limited attack. A commitment to ABM would, however, force those
operating strategic retaliatory forces to provide effective
penetration aids against endo-atmospheric intercept; for the U.S.
missile forces, such "penetration aids" have been additional
missiles and warheads. Indeed, there is little doubt that Russia
would deploy a system of nuclear armed interceptors which would add
many warheads to the Russian nuclear armory.
In the context of a theater opponent countering U.S. "high-tech"
conventional military capabilities, a recent article6 states that the "explosion of a single
high-altitude low-yield nuclear weapon could destroy $14 billion
worth
4Or preboost phase, or deterrence of launch, or
passive defense.
5 They should be asked also to supply decoys
and jammers that might automatically accompany any attack, even an
accidental or unauthorized one.
6R.C. Webb et
al., "The Commercial and Military Satellite Survivability Crisis,"
Defense Electronics, August 1995.
Page 190
of low-Earth-orbit satellites that would transit through the
enhanced radiation belts produced by such a nuclear event.'' Of
course, the signing of a universal comprehensive test ban treaty
would tend to prevent and certainly make illegal such an act, and
one must try to find a way to make it explicit that such damage
would result in dire retribution for anyone who caused it, without
making excessively clear to potential troublemakers the magnitude
of the damage that could be caused this way. Such retribution would
be all the more legitimate if the United States and the other
nuclear states emphasized their commitment not to use nuclear
weapons first, so that this postulated use of nuclear weapons would
damage the entire international security system, as well as the
specific target of the nuclear attack.
The Threat to the United States
That there are serious objective dangers to the United States is
indisputable. In 1969 a panel of the President's Science Advisory
Committee (PSAC) on chemical and biological warfare was asked
specifically to review for President Nixon the utility of a ban on
biological weapons. Indeed, President Nixon soon issued an
executive order eliminating not only U.S. use of BW, but also
possession, manufacture, and even R&D on biological weaponry.
This was followed quickly by the negotiation with the Soviet Union
of a treaty banning BW in the same way, leading to the
international Convention on the Prohibition of the Development,
Production and Stockpiling of Bacteriological (Biological) and
Toxin Weapons and on Their Destruction, signed April 10, 1972,
which entered into force March 26, 1975. The Soviet Union
has apparently not fully complied with the convention, and the full
force of international resolve has not yet been turned to
implementation of the convention and to its buttressing by means of
effective societal verification.
Any use of BW by the Soviet Union against the United States was,
presumably, in any case deterred by U.S. possession of nuclear
weapons and their delivery capability, whether or not nuclear
retaliation was specifically threatened in the case of BW attack.
There is no reason to believe that such deterrence would not still
work against Russia, or against most states contemplating use of BW
or CW. By the same token, it is hard to see how one could deter by
threat of retaliation the use of BW by terrorists. Indeed the very
aspect of BW that makes it so ineffective against combat troops may
paradoxically greatly increase its effectiveness against civilian
populations.
This is the substantial duration (hours or days) between the
ingestion of the agent and the outbreak of the illness, giving time
for a modest crew to spread BW agent widely. Furthermore, although
most of the BW agents contemplated by major power for use in
warfare were infectious but not contagious,7terrorists could perfectly well use
highly contagious natural agents. The main point is
7 That is, one or a few "bugs" could cause an
infection in humans, but the disease would not spread with
substantial probability from human to human.
Page 191
that terrorists and nations (in the modern world of relatively
open borders, international travel, and mixed societies) would
hardly rely by choice on ballistic missiles for delivering BW as a
terrorist weapon against population centers of the other side.
REGIONAL MISSILE DEFENSE IN RELATION
TO THE ABM TREATY
The problem of regional missile defense, as discussed above, is
not only one of effectiveness against the regional threat, but the
impact of TMD on the ABM treaty deserves attention, given the
magnitude of the regional threat compared with the essential nature
of the ABM treaty in limiting strategic offensive arms.
There is qualitative impact as well as quantitative impact. The
ABM treaty was crafted not only to prevent the existence of an
effective nationwide defense against nuclear-armed strategic
ballistic missile forces, but also to provide a substantial "buffer
time" before such a defense could be deployed. A situation in which
an ABM does not exist but in which it could be deployed the next
month, would be worse, in reality, than the gradual deployment of
such a defense. The very prospect of an ABM defense effective
against the existing strategic offensive force would call forth
penetration aids, multiple warheads, and expansion of the strategic
force, until the ABM were overcome, if, indeed, it was cost
effective to overcome it. But can an ABM be "overcome"?
The current generation of political leaders and strategic
analysts seem to ignore the insight of the 1960s that led to the
adoption of the ABM treaty and which is valid today, in essence. It
recognizes that some 400 nuclear weapons reaching their targets
would surely destroy the United States or the Soviet Union (and
fewer would now be required to destroy Russia) as a functioning
modern society. That number penetrating would constitute effective
"assured destruction" and the prospect of receiving such a
retaliatory strike would essentially nullify any desire to have a
first strike or to dictate political surrender to the other
side.
The large force of more than 10,000 strategic nuclear warheads
on each side appears to have grown to that level not because that
many were deemed essential for the assured destruction role, but
because a large fraction of the strategic warheads might be
destroyed before they could be launched in retaliation, and an
additional factor entered to compensate for some ABM system that
might be built before the force could be further expanded.
In addition, there was still a residue of the 1962 McNamara
mission of "damage limitation" by which nuclear warheads beyond
those required for assured destruction would be used to destroy the
strategic offensive force of the other side. Furthermore, the
calculus of destruction before launch and the penetration of ABM
systems is subject to a very great "offense-defense asymmetry" of
conservatism that in itself could account for the positive feedback
and essentially divergent numbers of nuclear weapons, increasing
almost without limit.
Page 192
On the other hand, the agreement to strictly limit ABM, and to
provide an effective buffer time during which the strategic
offensive force could later be modified, laid the basis for the
reduction of strategic warheads to the committed level of 3,000 to
3,500. Playing an important role, although not very often
explicitly acknowledged, was the recognition also that the
destruction of even vulnerable land-based "MIRVed" missiles before
they could be launched in retaliation was not feasible, in view of
the possibility of "launch under attack (LUA)" or "launch on
warning."
So this calculus also drove the two sides to a capability of
launch under attack, although the number of strategic offensive
warheads did not decline to reflect fully the reality of LUA.
The problem of destruction before launch (DBL) was especially
severe because the United States had voluntarily chosen to respond
to the potential of a Soviet ABM system (or to grasp the
fruit of technological advances in nuclear weapons and missiles) by
deploying multiple warheads on our land-based missiles, thereby
unilaterally introducing the potential for the Soviet Union to
destroy three Minuteman-3 warheads with a single accurate Soviet
nuclear warhead. The Soviets followed (in view of the fact that
MIRVs were never put on the table in the initial SALT
negotiations), thereby incurring on their side a tremendous
vulnerability, especially in view of the greater reliance placed by
the Soviet Union on intercontinental ballistic missile (ICBM)
weapons in contrast to submarine-launched ballistic missiles
(SLBM).
Under START, the multiplier of DBL due to the self-imposed
vulnerability of MIRV will disappear with the elimination of
land-based MIRVs on both sides, but on the other hand the number of
warheads on either side will also be much reduced, and there will
be greater sensitivity to the effect of an ABM system.
Some in the U.S. defense community still want to rescind the ABM
treaty. They tend to believe in national security on a unilateral
basis, as reflected in the 1980 presidential campaign literature of
Governor Ronald Reagan, which stated that President Ronald Reagan
would have a three-point program to build nuclear weapons to disarm
the Soviet Union, and if the Soviets tried to respond, it would be
so costly that they would destroy themselves economically. Russia
is now in substantially more dire economic straits than was the
Soviet Union, and the appeal of this program may be substantially
larger to a small but influential minority in the United
States.
Unfortunately, there is much misinformation, and even technical
misinformation provided to the Russian legislature, that could lead
to substantial missteps by the United States and by Russia.
For instance, a study8 paid for
by the BMDO and released publicly in February 1995 has been claimed
to counter the analysis of Professor T.A. Postol of the
Massachusetts Institute of Technology and his colleagues that
argues that
8 Lee, Laura T., et al.
"The Abuse of Footprints for Theater Missile Defenses and the ABM
Treaty" (U), SPARTA, Inc., McLean, Va., September 1994.
Page 193
THAAD9 has significant
effectiveness against strategic ballistic missiles, if it is
effective against missiles of 3,000-km range.
Unfortunately this BMDO-sponsored study has no "study" behind
it—just the briefing charts, as explained to me by BMDO staff
and the contractor. Furthermore, the results are wrong,
although it is more difficult to determine that they are wrong if
there is no written analysis that can be evaluated.
The key claim of the Sparta study is that for the missiles of
range against which THAAD is to be deployed (up to 3,000 km), the
ground-based radar in the terminal area can see the missile during
its ascent phase, where the radar cross section is large because
the missile is essentially broadside to the radar beam; the ICBM,
however, is below the horizon in its boost phase and then presents
a small enough nose-on angle to the radar so that it cannot be
seen. The data shown for radar cross section vs. angle, however,
and the sketch of the trajectory make it very clear that the
3,000-km missile is well below the horizon during any high
cross-section phase of flight. Even shorter-range missiles need
never present an aspect angle greater than 45 degrees, out to
which, according to the cross-section data shown in the BMDO study,
the cross section is very low.
So in this case one should not trust the material published by
BMDO, on which BMDO policy, that of the Department of Defense
(DOD), and presumably U.S. national security policy are based. Of
course, one might point to errors in the analyses of some critics
of DOD programs, but that seems to me quite irrelevant.
"Force on force" criteria for the acceptability of TMD advanced
by an unnamed government official, and quoted in a Washington
Times newspaper article of May 10, 1995, imply that "unless
there is some kind of significant, meaningful, major, material
capability against the other guy's strategic force, then that is a
permissible TMD." This implies that only systems with capabilities
against the entire strategic force are limited by the ABM treaty,
so that one-on-one tests should not determine the criteria for
regional or strategic systems.
Evaluation of the effectiveness of various systems must consider
as a primary element "sensor integration"—even so simple as
launch detection satellite cueing. For instance, SDIO Director Dr.
Henry Cooper revealed that U.S. "Defense Support Program"
satellites detected every Scud launched by Iraq during Desert
Storm. And the U.S. Navy has recently discovered a substantial
capability of a fleet of vessels against cruise missile or even
theater ballistic missile attack, by taking seriously the
integration of sensors on the various ships. Even in the early
1970s, the DOD testified about its concern with "SAM upgrade," one
version of which was the networking of the Soviet SA-2
surfaceto-air missile sites, to provide a coherent ABM capability.
It is not a simple job to establish the absence of such links.
9 The Theater High Altitude Air Defense interceptor, of which
the U.S. Army proposes to buy 1,442 missiles.
Page 194
Such an interpretation would entirely destabilize the strategic
scene, not least by forcing major actions in the near term by
Britain, France, and China, and also forcing Russia to modify its
strategic force. Russia would be able to mobilize the resources to
restore the effectiveness of its strategic force only by portraying
the United States as an enemy bent on disarming strike, and by far
the easiest way for Russia to increase the effectiveness of its
planned forces is immediately to stop the START process.
On December 28, 1995, President Clinton vetoed H.R. 1530, the
National Defense Authorization Act for Fiscal Year 1996, on the
grounds, first, that it
requires deployment by 2003 of a costly missile
defense system able to defend all 50 States from a long-range
missile threat that our Intelligence Community does not foresee in
the coming decade. . . By setting U.S. policy on a collision course
with the ABM Treaty, the bill would jeopardize continued Russian
implementation of the START I Treaty as well as Russian
ratification of START II—two treaties that will significantly
lower the threat to U.S. national security, reducing the number of
U.S. and Russian strategic nuclear warheads by two-thirds from Cold
War levels. The missile defense provisions would also jeopardize
our current efforts to agree on an ABM/TMD (Theater Missile
Defense) demarcation with the Russian Federation.
The December 4, 1995, edition of The Washington Times
includes the text of the U.S.-Russian "Agreed Framework" covering
theater and regional antimissile systems, which was agreed to on
November 17, 1995. The article reporting on this document notes
that the original is classified SECRET and presumably reflects the
Administration position on modifying the ABM treaty to permit
certain types of TMD.
Revival of the Strategic Defense
Initiative?
In May 1995, a letter from the leaders of several defense
contracting corporations stated that spaced-based chemical lasers
were essentially ready for deployment as a test system with a
4-m-diameter mirror and that within a few years an entire system of
12 SBLs with 8-m-diameter mirrors could be deployed at a cost of
$15 billion. There is substantial pressure behind such deployments,
with these lasers claimed to have an effective range of 3,000 km,
each one claimed to defend effectively against missiles launched in
the 10 percent of the surface of the Earth within its field of
view, so that a constellation of 12 SBLs would provide an effective
defense against a small threat, and 25 SBLs would provide a very
substantial defense.
The letter claims that the lasers could be deployed each with
fuel for something like 200 effective "shots," and that the primary
purpose would be to destroy missiles in their boost phase. Of
course, these lasers would as readily destroy ICBMs as theater
missiles in boost phase, and they would strike at the heart of the
strategic reductions that we have in process.
Page 195
However, just as was the result of the analysis in the early
days of the SDI, however effective such lasers might be, they could
be destroyed as they were being deployed, by simple antisatellite
(ASAT) measures.
There would be no reason in the 1990s for Russia to use the
co-orbital ASAT that the Soviet Union tested some 22 times and that
was described, for instance, in our article in Scientific
American.10 Instead,
Russia would surely use a directascent ASAT, equipped with either a
small nuclear warhead or a pellet warhead to destroy the rather
fragile SBL, without imposing the requirement of sufficient
accuracy to destroy by kinetic energy collision of the ASAT homing
head itself. SDI proponents formerly argued that the deployed
constellation of SBL would be mutually protecting, but such systems
are not operational as soon as they are put into orbit, and the
exchange ratio between the cost of an SBL and the cost of a
direct-ascent ASAT is surely such that no such weapons in space
could survive.
Nevertheless, the launch of space-based lasers to provide an
effective component of strategic or theater defense would lead to a
strategic confrontation that would not be optional but
mandatory.
A Truly Cooperative Defensive
System?
During the SDI program, there was promise of a defensive system
that would "benefit" both the United States and the Soviet Union.
President Reagan seemed sincerely to advocate a system that would
actually protect both the United States and Soviet Union,
destroying equally Soviet or U.S. missiles if they were launched.
However, U.S. Defense Department personnel made very clear that not
only was this not their goal but also that they would not even
"share technology" with the Soviet Union. In a September 1986
debate in Dallas, Texas, the DOD representative chose his words
very carefully to say that we would "share the benefit of defensive
technology."
I likened this to the slave owner who "shared the benefit" of
slavery. The slave owner obtained the profits, and the slave was
protected and fed and housed, to some extent, so long as his
product was sufficiently valuable to the owner. If one side has
acquired a good defense (especially when combined with its
strategic offensive force), it will be a tranquil (and compliant)
world until the other side catches up, makes an end run, or
miscalculates. That same kind of shared benefit, and tranquility,
obtains when only one side has a disarming force against the other;
and both sides know it.
Brilliant Pebbles Resurgent?
The x-ray laser, cherished development of Edward Teller and
Lowell Wood and their colleagues at the Lawrence Livermore National
Laboratory, seems truly dead and will surely remain so, so long as
there is a ban on nuclear testing.
Not so for their next enthusiasm, "Brilliant Pebbles."
10 Garwin, Richard L., Kurt Gottfried, and
Donald Hafner. 1986. "Antisatellite Weapons," Scientific Amerixan,
Vol. 250, No. 6, June.
Page 196
One does not actually need to have an explosive warhead to
conduct an effective intercept in space. Any significant crossing
angle (with a low-orbit satellite moving at 8 km/s) would lead to
relative velocities of 5 to 15 km/s, and the kinetic energy of the
interceptor in the frame of the more massive strategic offensive
weapon would correspond to many times the explosive energy per gram
of high explosive. Indeed, the two are equal at a relative velocity
that will give a kinetic energy of some 4,000 J/g, or about 2.8
km/s. At 10 km/s relative speed, each gram of interceptor has 12
times the kinetic energy of a gram of high explosive.
So although it had long been considered to use pellet warheads
or for that matter orbiting pellet clouds to destroy objects in
space, the public relations associated with the advocacy of SDI
called now for "smart rocks"—as if one would be using a simple
rock, but "smart" enough to be guided to a collision with the
target.
Some went one step farther, claiming to increase the
intelligence by making the rocks "brilliant" and reducing the size
by the miracle of modem consumer electronics so that these were no
longer smart rocks but "Brilliant Pebbles." Instead of a few
kilograms as was originally proposed, the mass in orbit grew to on
the order of 100 kg, for which one can make quite a reasonable
interceptor, in principle. Thus was born the proposal to orbit
something like 5,000 Brilliant Pebbles (BPs), to destroy strategic
weapons during their boost phase. Of course, the BPs would need to
be assigned to the boosting weapon and conduct an intercept with
the precision necessary to strike the missile during boost phase.
The BP would need to be self-guided, and there are counters to
this, on the part of the ICBM itself.
However, in 1991 I published a paper,11 and distributed widely a more
extensive version,12contrasting the requirements for a
direct-ascent ASAT to destroy Brilliant Pebbles with the
requirements for the Brilliant Pebbles themselves. In every way the
ASAT job is simpler. The nation that wants to destroy a
constellation of BPs can take its time in doing so, and it can do
so with very small homing interceptors supported by ground-based
radars or lasers with a view of the engagement taking place in low
Earth orbit—a capability that the BP itself cannot call
upon.
Furthermore, the ASAT itself need detect the BP satellite and
provide guidance from a distance of only a few kilometers, or for
that matter a few hundred meters, given the accuracy with which the
ASAT can be guided to the predicted position of the BP with the aid
of ground-based radar or lasers.
The ASAT guidance and homing system need survive only for a few
minutes, whereas that of the BP needs to survive for years in
space, and the same is true of the power supply for the ASAT, which
could be batteries,
11 Garwin, Richard L. 1991. "Defense Is Easier
from the Ground," Op-Ed piece, Space News, March 11-17.
12 Garwin, Richard L. 1991. "Are Brilliant
Pebbles the Counter to Stretched Scuds?", February.
Page 197
whereas the BP would need to have a solar supply. Instead of a
refrigerator for its infrared sensor (if any), the ASAT could carry
liquid nitrogen or even liquid helium for the few minutes of its
flight.
Thus it is clear that the ASAT job of wiping out the BPs is very
much easier than that of putting up the BPs in the first place.
Instead of Brilliant Pebbles,
Brilliant Eyes?
Brilliant Pebbles, of course, would be a clear violation13 of the ABM treaty. As a result, it
was proposed to deploy a system of "Brilliant Eyes," fewer than are
necessary for a BP constellation, and the nominal job of which
would be to refine the trajectory observations of warheads in
midcourse, so that terminal ABM systems could work more
effectively. I have been unable to see why the sensor of a terminal
ABM system could not be given the capability to make an intercept
without the trajectory "refinement" available from BP (if such
could be obtained), nor do I understand why an "optical probe"
launched from the terminal area on detection of a ballistic missile
launch would not be a better idea than a Brilliant Eye.
I note, however, that Edward Teller proposed in the SDIO era
that Brilliant Eyes should have all of the capabilities of
Brilliant Pebbles, including rocket engines and homing systems that
could boost the interceptor and conduct an intercept, but they
would be deployed without the fuel and so would be "legal" under
the ABM treaty.
I don't know any other judgment that such a system would be
legal under the ABM treaty, any more than the Krasnoyarsk radar was
legal. After all, that radar could have been maintained unlinked
from the rest of the strategic defense system, or its beam could be
held low enough to constitute a space track or early warning rather
than ABM system, but there was no way in which such limitations
could be verified or enforced. The same is true of BMDO claims that
Brilliant Eyes are acceptable if they lack a direct communications
link to interceptors.
Similarly, since the ABM treaty has the goal of providing the
time buffer before deployment, to build and launch and test
Brilliant Eyes that have a capability of Brilliant Pebbles would
presumably not be allowed either.
A truly cooperative defensive system could hardly be
objectionable, but we are not ready to deploy such a system. It
would need to be accompanied by a regime that would make illegal
the launching of missiles from one nation against another, which
might indeed then lead to the disappearance of ballistic missiles
entirely. But an effective ABM system and a commitment to upgrade
it and to keep it effective would need to be operated by the United
Nations and would require an operating committee and a voting
procedure, all of which basis would need to be laid before a system
was developed and deployed. I am not
13 Article V: "1. Each Party undertakes not to
develop, test, or deploy ABM systems or components which are
sea-based, space-based, or mobile land-based."
Page 198
saying that this is simple or that it can be achieved with
confidence, but only that this groundwork must be laid before
defenses can be developed without destabilizing the world.
A space-deployed defensive system, protected by international
law and by the might and power of the nations subscribing to
international law, would be quite a different consideration from a
unilaterally deployed system. So that is of interest for the long
term, although serious consideration may result in the rejection of
such a system as infeasible, impractical, or undesirable.
One near-term and noncontroversial contribution to stability can
be obtained by silo-cover sensors deployed cooperatively. That is,
the United States would provide a small package for each Soviet
silo cover, with the function of continuing to transmit a signal
that cannot be simulated, so long as the sensor remains attached to
the silo cover and the silo cover has not moved. The actual
transmission would be handled by a Russian-supplied relay box.
Russia would supply similar sensors to the United States. Each
sensor would have its own cryptographic key (or a "public-key"
system could be used) so that it would continually encrypt the time
and the serial number of the sensor, so long as it remained
attached to an unmoved silo cover.14High
Russian officials explain that (like the Soviet rocket forces
before them) their normal posture is one of responding when the
first few nuclear explosions occur on Russian territory. Their
other real option is to launch on warning of attack, before any
explosions have occurred. And they insist that "delayed
retaliation" is not an option for them. Hence the reliable
assurance that U.S. missiles have not been launched is very
important to the prevention of a massive launch of Russian
strategic forces.
CONCLUSIONS
In the meantime, the consequences of unilateral deployment of
space weaponry are so severe that I believe that it is necessary to
revive the U.S.Soviet talks on banning antisatellite weapons and
extend them to the banning of all space weapons as well as ASAT
test and use. The ban on space weaponry would not affect
communication satellites, navigation, satellite imagery, launch
detection systems, or other satellites that are not actual weapons.
Such a treaty could be agreed to between the United States and
Russia and then opened for revision and subscription by all of the
nations of the world, in analogy to the Biological Weapons
Convention.
In addition to the traditional "national technical means of
verification" (a euphemism for "satellite reconnaissance")
authorized in the ABM Treaty of 1972 and in later treaties, new
treaties ought to make provisions for societal verification, by
which the text of the treaty is published widely in the states
party, domestic law is established that makes it illegal for
individuals to perform
14 A sensor embodying these characteristics is
deployed in cooperative monitoring of nuclear material stockpiles
and may be viewed at the Cooperative Monitoring Center, Sandia
National Laboratories.
Page 199
those activities that the state has agreed not to perform, and
the permission and responsibility are given individuals to report
to a verification commission a state's violation of the
agreement.
For the real threat of use of CW and BW against friendly cities,
the most effective approach is to pursue vigorously the entering
into force of effective bans on BW and CW and to have it understood
that any violation of such a treaty (especially the use of BW or
CW) would lead to the most severe response by the community of
nations. A response with nuclear weapons could not be precluded.
Passive defense should be emphasized, particularly for ships and
the military, and should be considered for civil populations under
particular threat.
Active defense of ships against cruise missiles and ballistic
missiles should take advantage of the fact that only a very small
region needs to be defended, if intercept takes place no farther
than 5 km from the ship. This is a very different system from those
that are discussed, which try to obtain a theater-wide
capability.
Against accidental launch of strategic systems (far less likely
under conditions of nontargeting and reduced readiness than it was
formerly) cooperative control measures are far superior to BMD, and
cheaper, too. The silo-cover cooperative monitor should permit
reduced alert levels under normal circumstances.
Against a blackmail or rogue nation strategic threat,
destruction of the strategic missile before launch should be
considered, but a light ABM even against one or a few missiles is
not a realistic option, in view of the necessity to intercept above
the atmosphere, where countermeasures against nonnuclear intercept
are quite feasible. A commitment to an effective light nationwide
defense would (and technically should) lead to the use of
nuclear-armed interceptors, which would be very similar to the
classical ABM systems.
Of course, it is our actions that will influence the
world, not our desires. Hence it is important to have some kind of
understanding of the impact of various candidate actions on the
world. By "actions" I mean not only development and deployment of
weapons, but also speeches, negotiations, deception, and so on.
Our actions can have direct effects, but also indirect effects
when others are moved to take or not to take actions of their own.
In some cases, indirect effects can be much larger than direct
effects, and they may come earlier as well. My own judgment is that
it is not in our national security interest to interpret the ABM
treaty as limiting subsystem performance only if it has actually
been demonstrated against strategic-class reentry in actual tests.
The effects of such an interpretation on our own security have not
been thought through.
Thus, the United States should go ahead with dual-capable
(aircraft, cruise missiles, and theater ballistic missiles) systems
such as the PAC-3 upgrade of the Patriot, with remote firing of
such interceptors from displaced radars and should rely for system
performance on "launch-point cueing."
In general, there should be increased emphasis on passive
defense against CW and BW and on a balanced defense against
all threats. But we should not
Page 200
confuse the wish for effective defense and the capability for
effective defense, which confusion can jeopardize the uneasy
security provided by deterrence against major potential
threats.
Items in cart [1]
