News



                       Richard L. Garwin
                          P.O. Box 218
                  Yorktown Heights, NY  10598
                         (914) 945-2555
                      FAX: (914) 945-4419


                                             September 3, 1991


  The Honorable Les Aspin       The Honorable Sam Nunn
  Chairman, Committee           Chairman, Senate Armed
    on Armed Services             Services Committee
  2336 Rayburn House            303 Senate Dirksen Office Bldg.
    Office Building             Washington, DC  20510-1002
  Washington, DC  20515

  Dear Chairmen:

  As  you  meet  in September on GPALS and missile defense, we
  hope that you will take  into  consideration  our  views  as
  expressed  in  this  letter.    As  you  know,  we have long
  participated in  both  the  technological  and  programmatic
  aspects  of  national  security,  including many relevant to
  missile defense and  space  weapons.    We  both  have  been
  members  of  the President's Science Advisory Committee, and
  one of us (RLG) also of the Defense Science Board.  Here are
  our current views on defense against ballistic missiles,  in
  relation to the Missile Defense Act of 1991.

  Defense  of  the  U.S.  against  accidental  or unauthorized
  launch of strategic missiles.

  The most important goal of the Missile Defense Act  of  1991
  is  defense  against  accidental  or  unauthorized launch of
  Soviet strategic missiles.   Of course, the  Senate  is  not
  implying  that  the Soviet Union wants to destroy the United
  States with an unauthorized or accidental launch-- quite the
  contrary.  And we have, ourselves, for several years  warned
  of  the problems that could be posed by social and political
  disintegration of the Soviet Union while it still  possessed
  some  30,000  nuclear  weapons.   We advocated a much faster
  pace  to  the  START  negotiations,  and  in  addition   the
  elimination  of the Soviet non-strategic warheads.  We still
  make this recommendation, and hope that you agree.

  However, the rebuilding of Grand Forks is likely to be  seen
  by  the  Soviets  as  a  hostile  act,  however peaceful the
  intention of the Senate.  Reactivation of Grand Forks  makes
  it unlikely that the Soviets will agree to further reduction
  of nuclear weapons, strategic or tactical.  This effect will
  be  much aggravated if we were to renegotiate the ABM Treaty
  as proposed in Subsection b.2 of the Missile Defense Act.

  The enormous political changes inside  Soviet  territory  do
  not  permit the U.S. to assume away all traditional concerns
  of Soviet military planners.   President Boris  Yeltsin  has
  said,  "We  favor  total  elimination  of nuclear weapons in
  Russia" but also "We need  to  maintain  parity  with  other
  nations."    President  Mikhail Gorbachev would say no less.
  Despite vast pending changes  in  the  Defense  and  Foreign
  Ministries,  we  cannot assume that U.S. actions with regard
  to anti-ballistic missile systems will be met with unconcern
  and inaction from the Soviets, whatever the organization  of
  their  military.    Moreover, the political situation inside
  Soviet territory may revert to a more conservative tone.

  Accordingly, we see the present moment as an opportunity for
  the U.S. and the Soviet Union to proceed with removing, in a
  secure and verifiable fashion, some of the excesses  of  the
  assured   destruction   machinery   and   its   hair-trigger
  readiness.  We believe this to be more timely, more  urgent,
  and more practical than attempting actions raising sensitive
  issues  of  the ABM Treaty at the possible cost of derailing
  progress toward reducing the Soviet nuclear  threat  in  the
  present climate of hope and cooperation.

  If  multiple  ABM  sites  or  a  light  nationwide  defense,
  probably permitting nuclear-armed interceptor rockets,  were
  allowed  in  the  U.S.,  they  would  also be allowed in the
  Soviet Union.    This  would  degrade  our  deterrent.    In
  particular,   in  response  to  a  possible  limited  Soviet
  aggression, we could probably not retaliate with just a  few
  nuclear   warheads,  a  measured  response  that  is  widely
  regarded as essential to credibly deter such aggression.

  The ABM Treaty  continues  to  be  basic  to  enhancing  our
  security  through  arms  control.    We endorse the enclosed
  statement by the Arms Control Association.

  Against the threat of accidents and unauthorized  launch  of
  Soviet  missiles, a better protection would be upgrading the
  Soviet capabilities to prevent such launch.  The U.S. should
  offer  the  Soviets  technological  help  in   doing   this.
  Furthermore, the U.S. should encourage Soviet authorities to
  join  in  an  exploration  of  fundamental  changes  in both
  states' nuclear postures that  would  sharply  reduce  their
  present  reliance  on prompt retaliation.  These initiatives
  would  strengthen  cooperation  rather  than  arouse  Soviet
  suspicions.

  Defense of the U.S. against Third World ballistic missiles.

  We  believe  that  if a Third World state wishes to threaten
  the United States with nuclear weapons, it  would  far  more
  likely  try to introduce these on a cargo ship, a commercial
  airplane, or on trucks, than on an ICBM.  The same  is  true
  for   chemical  or  biological  weapons.    Defense  against
  ballistic missiles would seem to  make  an  unlikely  threat
  even  less likely, and the potential adversary would move to
  the  delivery  means  most  likely  in  the   first   place.
  Furthermore,  a  Third  World  nation capable of deploying a
  nuclear-armed ICBM would almost surely be  able  to  provide
  elementary   countermeasures   against   ballistic   missile
  intercept in space; this would make an  effective  ABM  much
  more difficult to achieve, as is discussed in the Appendix.

  However  unlikely, a threat to the U.S. from a nuclear-armed
  ICBM in the hands of a Third World nation  would  be  highly
  objectionable.    In  the  spirit  of  the Non-Proliferation
  Treaty, the United States  might  get  together  with  other
  like-minded   nations,  and  with  the  United  Nations,  in
  agreeing that this will not be tolerated.  As in the case of
  the Iraq War, the  resulting  coalition  might  empower  the
  United  States to destroy such ICBMs on their launch pads it
  not before.

  Theater missile defense.

  In addition to  the  planned  improvements  to  Patriot,  we
  support   R&D   on   systems   to  protect  forward-deployed
  expeditionary forces against theater ballistic missiles, but
  believe the systems considered have not addressed  the  most
  likely and effective near-term threats.

  In  particular,  deployed  forces  will  in  the near future
  require protection against  multiple  warheads  (biological,
  chemical,  high-explosive)  on  a  single missile, dispersed
  early enough so that 10 or 20 or more effective interceptors
  would be needed to nullify a single  missile.    This  would
  require  emphasis  on  destroying  the enemy missiles before
  they can be launched.

  In the Appendix we provide a  more  complete  discussion  of
  Theater  Missile Defenses.   Here we urge great caution that
  the contribution of the ABM Treaty to our  security  not  be
  impaired by premature action or legislation that will permit
  or  even  challenge  the  Soviet  Union  to test and perhaps
  deploy defenses that would not  be  in  our  interest.    We
  suggest  that  the  broad and considered counsel of Generals
  Colin Powell and Lee Butler be sought on this matter.

  With great appreciation of the work of your  two  committees
  under your leadership, we are

  Sincerely yours,


  Hans A. Bethe                 Richard L. Garwin
  Cornell University            IBM Research Division and
                                  Columbia University


  (Affiliation given for identification only)

  Encl:
     07/00/91  Arms  Control  Association  background paper on
       ABM deployments and amendments.  (070091.ACA)
     09/02/91  Technical  Appendix  re  Missile Defense Act of
       1991, etc.  (090291MDAA)


  RLGP:rlg:Q246LA:090391..LA
*******************************************************************
                                 APPENDIX

                                    TO

                        LETTER ON MISSILE DEFENSE


                                    by

                            Richard L. Garwin

                          IBM Research Division
                     Thomas J. Watson Research Center
                               P.O. Box 218
                        Yorktown Heights, NY 10598

                              (914) 945-2555

                                  (also
                      Adjunct Professor of Physics,
                           Columbia University)



                            September 2, 1991


       ABSTRACT  AND SUMMARY. It seems imprudent to suggest that we
       want to modify the ABM Treaty of 1972 in such a  way  as  to
       permit  the  Soviet  Union more capability against strategic
       ballistic missiles.  Protection of the entire world  against
       unauthorized   or  accidental  launch  of  Soviet  strategic
       ballistic missiles could better be handled by launch-control
       improvements than by deployment of an ABM system.

       As  for  protection  of  the  United  States  against   some
       third-world  ICBMs  armed  with  nuclear  warheads,  100 ABM
       interceptor missiles at Grand Forks is an excessive number--
       perhaps  20  would  be  more  appropriate,  but   our   best
       protection would be constant vigilance against a threat from
       the emergence of such missiles and international (coalition)
       diplomacy   and   potential  military  action  against  such
       missiles.

       For protection of our deployed forces, much  more  attention
       should  be paid to intercept of warheads (and bombs) at very
       low altitude, and  especially  to  the  destruction  of  TBM
       before launch.

       The  protection  of  allied cities against Theater Ballistic
       Missiles (TBM) poses a particularly  difficult  problem,  in
       view of the fact that attack effectiveness is increased by
       the simple and low-cost measure  of  splitting  the  payload
       among  multiple  RVs,  both  for high-explosive and chemical
       payloads.  Destruction of the  threatening  warheads  before
       they can be launched is the best remedy here, too.

       In  all  of  this, U.S.   space observation capabilities are
       important, as proved by the  effectiveness  of  the  Defense
       Support Program infrared warning satellites for detection of
       missile  launch.   U.S. capability to operate in this medium
       should be preserved and expanded by negotiation of a ban  on
       space weapons and on antisatellite tests.


       Q245MDAA             090291MDAA DRAFT 6             09/06/91
              Views of the author, not of his organizations


CONTENTS

INTRODUCTION. . . . . . . . . . . . . . . . . . . . . DEFENSE OF THE U.S. AGAINST THIRD-WORLD NUCLEAR WEAPONS. . . . . . . . . . . . . . . . . . . . . . . DEFENSE AGAINST ACCIDENTAL OR UNAUTHORIZED LAUNCH OF SOVIET STRATEGIC MISSILES. . . . . . . . . . . . . . DEFENSE OF DEPLOYED U.S. FORCES AGAINST NON-NUCLEAR THEATRE BALLISTIC MISSILES. . . . . . . . . . . . . DEFENSE OF ALLIED CITIES AGAINST ATTACK BY NON-NUCLEAR TBM. . . . . . . . . . . . . . . . . . DEFENSE AGAINST A LIMITED BUT DELIBERATE NUCLEAR ATTACK ON OUR RETALIATORY FORCE BY A RESURGENT SOVIET MILITARY. . . . . . . . . . . . . . . . . . SUMMARY. . . . . . . . . . . . . . . . . . . . . .

INTRODUCTION.

This Appendix is based on personal experience in both the technological and programmatic aspects of national security, including many relevant to missile defense and space weapons. These involvements range from the first hydrogen bomb, through strategic missiles and reentry vehicles, to x-ray lasers, missile guidance systems, strategic offense and defense and countermeasures, to current evaluation of SDIO programs. I have also specifically reviewed the lessons of the Gulf war in intercepting SCUDs, and have considered their relevance for defense against theater missiles and strategic ballistic missiles. This presentation is somewhat technical and the discussion abbreviated; the intent is to provide background for independent views on the various purposes and types of missile defense. In a March 1968 Scientific American article, "Anti-Ballistic-Missile Systems," Hans Bethe and I explained the capabilities and difficulties of the Sentinel "light" ABM System that the Johnson Administration was proposing to deploy against a potential Chinese nuclear-armed ICBM threat. Although the Nixon Administration and industry argued strongly for deployment of the Safeguard ABM, and the Congress acceded, the system was dismantled in 1976 after about a year of operation. Many arguments had been advanced for the necessity of deploying Safeguard, but the nation could have saved some $10-12 B by looking more closely at the arguments before authorizing deployment. Today, in 1991, the world has changed, in technology and in political alliances. Not only has the technology available for strategic defense changed, but also the nature of the threats. What can be done about the threat of ballistic missile attack? What should be done? In "The Missile Defense Act of 1991" authorizing the deployment of a 100-interceptor ABM system at Grand Forks, ND, the Senate responds to a fear of nuclear destruction in the United States which should be addressed. But the proposed remedy does not stand up to analysis as a way to get protection, without raising additional dangers. First, one must distinguish sharply between nuclear weapons and chemical warheads, in destructive power, in impact on a defensive system, and in the ability simply to penetrate a defense. Second, there is a big distinction between the Soviet threat (even the accidental or unauthorized nuclear threat) and a third-world nuclear threat. We will address the proposed capabilities under five categories and then return to summarize. 1. Defense of the U.S. against a primitive third-world ballistic missile threat. 2. Defense against unauthorized or accidental launch of Soviet ballistic missiles. 3. Defense of deployed U.S. forces against (non-nuclear) theater ballistic missiles. 4. Defense of allied populations against non-nuclear theater ballistic missiles. 5. Defense against a limited but deliberate Soviet military threat against our retaliatory force by a resurgent Soviet military.

DEFENSE OF THE U.S. AGAINST THIRD-WORLD NUCLEAR WEAPONS.

We should recall that most third-world nations adhere to the Non-Proliferation Treaty, are friendly toward the United States, or certainly have no quarrel with the U.S. that calls for the deployment or use of nuclear weapons. The prospect that some 15 nations of the world possess or will soon possess ballistic missiles is not at all equivalent to a forecast that they will have ICBMs armed with nuclear warheads. The DSP (Defense Support Program) satellites provided immediate detection of the launch of each SCUD fired by Iraq during the Gulf War. There is no doubt that these same DSP infrared detectors would provide immediate detection of the launch of strategic ballistic missiles anywhere in the world, and could readily provide trajectory information on such launches. With that information alone, one could launch from Grand Forks a small missile to provide intercept near apogee, with the homing done by autonomous infrared detectors on the interceptor. For reliability, probably two such interceptors would be launched against each of the incoming ICBM weapons. Thus, a deployment of 100 interceptors is excessive against any foreseeable threat, although the interceptor population could be increased with time if the threat developed. But there are two problems-- effectiveness and alternatives. As for effectiveness, a hit-to-kill intercept outside the atmosphere can be defeated by the simplest countermeasures, and it is best to understand this before authorizing deployment rather than afterwards, as with Safeguard. These countermeasures have long been published, and consist, for example, in deploying a large balloon or umbrella around or attached to the reentry vehicle. The purpose is not to hide the existence of the reentry vehicle (RV) but to conceal its precise location. Hypervelocity impact of the interceptor with this balloon or umbrella at a distance of 20 meters or so from the RV would cause no damage to the nuclear warhead, even if the interceptor were loaded with explosive and pellets. If such countermeasures seem implausible, remember that we are talking of a hypothetical third-world nation that is (by assumption) capable of building and operating ICBMs with nuclear warheads, in an era in which millions of automobiles are equipped with "air bags" that inflate in less than one tenth of a second. The message is that: any deployment at Grand Forks desired to protect against third-world threats must, from the first, be effective against the warhead concealed in a balloon or umbrella; there should be a few tens of interceptors, not 100; and the defense could be deployed without any ground-based radar, relying upon DSP and its analogues. Whether one either needs to deploy or should deploy such a defense is another question. A final point is that a nuclear-armed near-apogee intercept would be far more reliable than a hit-to-kill intercept and much less vulnerable to attack countermeasures. A few nuclear-armed intercepters might be included both for the real effectiveness that they might add and for their impact on anyone designing a small nuclear-armed ICBM threat against the United States. But defense against third-world nuclear weapons can be done in other ways, probably much more reliable and available earlier. First there should be no tolerance of a threat to the United States by third-world strategic ballistic missiles, especially when paired with nuclear weapons. There should be every expectation that these would be destroyed on their launch pads or before, and the United States should work with like-minded nations of the world to see that this is done, preferably with the support of a coalition, as was active in the Gulf. The nuclear threat to the United States from newly nuclear nations is probably far more significant from nuclear weapons carried on cargo ships, on commercial aircraft, or on trucks, than from ICBMs. Against such threats, a missile-defense system obviously has no effectiveness at all, and if the deployment of an ABM system were taken to signify a reluctance to destroy an ICBM threat before launch or to retaliate in case of use of any nuclear weapons (and thus to deter, if possible, the development and use of nuclear capability), then such deployment would be highly disadvantageous. A non-nuclear third-world ICBM could carry a chemical or biological warhead, but the threat is overwhelmingly greater from the dispersion of pre-delivered such agents at ground level. As with third-world nuclear-armed ICBMs (not theater-range ballistic missiles), deployment of a missile defense might make an unlikely threat even less likely, by moving an adversary toward the more likely and effective delivery means. In any case the effectiveness of ICBM-delivered chemical or biological agents would be increased if the missile payload were distributed among many small reentry vehicles dispersed just after boost phase of the offensive missile, which would provide dispersion of the agent difficult to achieve otherwise; against such a threat none of the proposed systems would have any effectiveness, except for an international commitment to destroy such weapons before they could be launched.

DEFENSE AGAINST ACCIDENTAL OR UNAUTHORIZED LAUNCH OF SOVIET STRATEGIC MISSILES.

This is the most momentous goal of the Missile Defense Act of 1991. Of course, the Senate is not implying that the Soviet Union wants to destroy the United States with an unauthorized or accidental launch-- quite the contrary. And we have, ourselves, for several years warned of the problems that could be posed by social and political disintegration of the Soviet Union, while it still possessed more than 30,000 nuclear weapons. We wished and urged a much faster pace to the START negotiations, and the elimination of all Soviet non-strategic warheads; we still make this recommendation. Unauthorized launch is as likely to be an entire submarine load of SLBMs as a single missile. Similarly, a whole squadron of SS-18 missiles might be accidentally launched as well as a single one. But is it not worth while to defend against a single accidental or unauthorized Soviet strategic missile, even if there are other possibilities against which we would have no adequate defense? Yes, depending on cost, effectiveness, and alternatives. The proposed light missile defense would not, in fact, provide significant protection against the most likely accidents and it would seriously impede (or reverse) reductions in the Soviet nuclear threat. The U.S. has long considered a limited and discriminating retaliatory capability essential to the credibility of our deterrent of limited Soviet aggression. This would involve the delivery of one or a few nuclear warheads against selected targets. Therefore, if a light nation-wide defense were deployed in the Soviet Union, the U.S. would need either to reinforce these hypothetical small-scale options, or provide effective countermeasures (penetration aids) against that defense. Which course we would take depends upon the nature of the defense. There is no reason to believe that the Soviet Union plans to use its strategic nuclear weapons any differently-- especially no reason to believe that it would not take all feasible measures to retain the capability of detonating one or a few nuclear warheads on U.S. soil. The deployment by the U.S. of a hit-to-kill ABM system would without doubt drive the Soviet Union to enhance the countermeasures that it already has, in the nature of objects tethered to or associated with the reentry vehicles, and in particular of umbrellas or balloons to conceal the location of the actual warhead. So the problem is not only that the U.S. might be faced with an accidental launch of 10 SS-18s, each with a complement of 10 real warheads, but that we could not destroy even a single errant Soviet missile if the Soviet force responded to the limited defense as we would respond to such a light nationwide Soviet missile defense. Furthermore, as has been well documented in Congressional testimony by T.A. Postol, a single site at Grand Forks would not do a good job in protecting even the 48 contiguous states from a single Soviet SLBM launched from near U.S. shores. What to do? The severity of the problem of mid-course intercept is addressed by Ambassador Henry Cooper in his "Special Defense Department Briefing, SDI and Global Protection Against Limited Strikes," of 02/12/91. Significantly, speaking of the budgeted funds for the "Endo/Exoatmospheric Interceptor," EEI, to intercept the RV as it reentered, he said, This makes the case for the greater effectiveness of ground-based interceptors in comparison with space-based weapons. But if the "mid-course discrimination problem" cannot be solved, space-based weapons will not intercept in midcourse, but if it can be solved, then ground-based interceptors will do a better job. The Director of the SDIO also argues that the space-based interceptor (Brilliant Pebbles) is cost effective and affordable, as well as survivable, at a launched cost of $2-3 million each. He explicitly argues that this is a "cost-trade that's reasonable" against even the single-warhead Soviet SS-25. This is surely a conclusion that will provoke a response from the Soviet Union (former Secretary of Defense Caspar Weinberger indicated that it would be the "worst strategic nightmare" he could imagine). It is just that judgment that will drive analysts in the Soviet Union to regard their retaliatory capability as being at risk by a U.S. deployment of a "defense against accidental launch" whether or not it includes space weapons. In reality, there is no way in which such a population of Brilliant Pebbles and Brilliant Eyes could be made survivable against a specialized Soviet non-nuclear ASAT that would destroy them over the months that would be required to build such a population.(1) A deployment of 40-kg Brilliant Pebbles could be defeated much more cheaply than it would cost to put up, and it would thus impair rather than add to our security.(2) The answer in this case is clear. If the U.S. is proposing to spend some billions of dollars to protect ourselves from a Soviet unauthorized or accidental launch, we could surely work with the Soviet Union to have them deploy additional safety systems, such that their nuclear warheads cannot explode unless provided with the proper PAL code. We repeat, it would be worth a lot of money to us to see that the Soviets do this right, and it can be accomplished more quickly and surely more reliably than a defense against unauthorized or accidental Soviet launch, and without the impairment of our own deterrent that would arise from a similar Soviet nationwide defense. In a New York Times OP-ED of 07/31/91, Senator Nunn states that the Senate Armed Services Committee "is requiring that the interceptor system it proposes be kept well below a level that would threaten this philosophy" (of preserving the other side's retaliatory force effectiveness after a first strike and in the presence of the planned defenses). But the problem has always been the "conservative" evaluation of one's own capabilities against what the other side might eventually have. History is littered with "limited" programs that escaped the bounds set by their originators-- for instance the MIRVed missile itself remorsefully cited by both Robert McNamara and Henry Kissinger. It is quite the same here. Once the United States begins a deployment of ballistic missiles against Soviet forces (even with the avowed purpose of countering unauthorized or accidental launch), the Soviet Union has no reason to believe that this will stop with the planned number, especially since the Missile Defense Act of 1991 emphasizes the expansion to multiple sites, not permitted by the ABM treaty. Finally, it is unlikely that the Soviet Union would accept a U.S. deployment beyond the strict limits of the 1972 ABM treaty without doing their own thing in the ABM field, which could begin with a more widespread deployment of ground-based nuclear-armed interceptors. And that would be the end of U.S. or Soviet strategic force reductions under START and any follow-on to START.

DEFENSE OF DEPLOYED U.S. FORCES AGAINST NON-NUCLEAR THEATRE BALLISTIC MISSILES.

In this regard, especially, we refer to the Issue Paper of the AAAS Program on Science and International Security, "Ballistic Missile Defense After the Kuwait War," which contains useful descriptions of the Patriot deployed defensive system, as well as of various other candidates for theater missile defense (TMD) in the development or in the definition stage. Although in the NATO-Warsaw Pact confrontation in Europe there was little merit in modifying the Patriot interceptor to give it some capability against Soviet short- and intermediate-range ballistic missiles,(3) the situation is very different against adversaries with limited numbers of missiles. In Saudi Arabia and Israel, the technical performance of the Patriot system against the Iraqi al Abbas "stretched SCUD" 800-km missile probably surpassed expectations, although analysis is hindered by the absence of any engagement recording capability on the deployed Patriot systems. Nevertheless, although Patriot made "intercepts," it very likely did not detonate the incoming warhead in some cases, its job being complicated by fuzing problems and by the tight helical trajectory of the incoming warheads, in view of the breakup of the al Abbas missile. The problem of defending our deployed military forces against theater ballistic missiles (TBM) is not one of defending against nuclear warheads but against missiles armed with high explosive or chemical weapons. Clearly it makes sense to build upon the widely available Patriot.(4) The Patriot Advanced Capability 3 ("PAC-3") modification will substantially improve Patriot performance against aircraft and cruise missiles, as well as greatly increasing the area defended ("footprint") against theater-range ballistic missiles. There are several advanced missiles under development-- ERINT, THAAD, etc.-- which would give a greatly expanded footprint, and choices will soon need to be made. Navy AEGIS cruisers with their capable radars could use interceptors from their own Vertical Launch System (VLS) or, significantly, could command the launch of ERINT or Theater High Altitude Air Defense interceptors based in the target region, providing a theater defense capability which could be rapidly deployed. For defending military forces, one could ignore almost all SCUDS or al Abbas missiles, which because of their inaccuracy of 1-5 km would for the most part land harmlessly. Thus, it would be most useful to intercept at an altitude of 1 km or so, after it had been determined that the particular warhead was threatening a target of value-- leverage for the defense that is not available in defense against aircraft. It should be noted, however, that a low-drag RV on a 800-km missile would reenter at 3 km/s, at which speed the kinetic energy of an inert body is equal to that in the same mass of high explosive (HE). Nevertheless, such energy is not so damaging as that from HE, only in part because the RV slows in the atmosphere. It is far easier to discern the target of a reentering ballistic missile than of an aircraft picked up by a Patriot radar at a distance of 50-100 km. The aircraft is maneuverable, and it could attack any target within the region; furthermore, the aircraft is far more fragile than the bombs that it will eventually deliver, and it costs typically a lot more than the $0.6 M Patriot interceptor. From all of these points of view, it makes good sense to destroy such aircraft as soon as they can be intercepted, with the additional benefit that more time remains for additional attempts at intercept (shoot-look-shoot). There is typically a big difference between the radar "appearance" of an aircraft that has not been effectively intercepted and one that has been killed by a Patriot interceptor, the principal difference being in the Earth-intercept trajectory of the destroyed aircraft itself. Unfortunately, there may be little difference between the radar appearance of the "destroyed RV" and one that has escaped destruction, since pieces of the RV may continue on the original trajectory. Thus, shoot-look-shoot may be far less useful for TMD than for anti-aircraft systems, and relatively more costly. Enthusiasm for TMD should be tempered by the recognition that there are sure-fire countermeasures to such systems, well within the capability of the suppliers or operators of such missiles. Among the simplest and most effective is the recognition that the military effectiveness of high-explosive warheads (or chemical warheads) is increased by dividing the warhead into a substantial number of packages, accounting for the popularity of cluster munitions or "bomblets." For instance, a missile with a 500-kg reentry vehicle (RV) might instead be equipped with 20-kg RVs. To have even a chance of nullifying an attack by one such missile, 20 successful intercepts would be required, involving 20 interceptors. All that is required to negate the defense system is for the warheads to separate earlier in the trajectory than an effective intercept could be made. Of course, it might be argued that some targets cannot be injured significantly by such small RVs, so that the military effectiveness of the attack would be reduced; in such cases, clearly, the defense is particularly simple, since one need only wait until an ultra-short-range radar at this hard target detects a large warhead about to strike, so that an intercept could be made at a distance of 100 meters! We urge that such automatic, self-protection systems be considered seriously, in view of their likely simplicity and low cost. Most military targets, however, would be more at risk from a given TBM payload incident as sub-munitions, and the choice and pace of investment in defensive systems must be guided by this fact. Again, this is a significant difference between attack by TBM and attack by aircraft-- the TBM sub-munition falls to its target without significant additional cost, whereas to divide an aircraft into sub-munitions for delivery over 50-100 km (cruise missiles, in fact) involves substantial difficulty and additional cost. Additional implications for TMD of our deployed forces will become clear as we now consider protection of allied cities against attack by TBM.

DEFENSE OF ALLIED CITIES AGAINST ATTACK BY NON-NUCLEAR TBM.

Aside from precious historical, cultural, and religious sites, modern cities appear both more valuable and more vulnerable to high-explosive attack than was the case in the second world war. They may be less vulnerable to incendiary attack. Nevertheless, one should not count on providing perfect protection of cities against attack by missiles. In particular, sub-munitions would get through any of the proposed defenses, assuming that they are dispensed early enough. This is particularly important for chemical attack, which is much more significant against civilian populations than against military, because discipline and training and equipment can allow U.S. and allied military to operate in the face of chemical attack. Parenthetically, the sooner the United States and like-minded nations can totally ban the possession as well as the use of chemical munitions, the sooner it will be possible to take measures against nations and organizations that would then be criminals if they merely possessed such weapons, even if they had not yet used them. Our capability to detect TBM before launch may improve, particularly mobile TBM launchers, but it should not be imagined that there will long persist a requirement for 30 minutes of readying before launch. Paradoxically, only the U.S., the Soviet Union, and Iraq are legally barred from possessing ground-launched ballistic missiles with range exceeding 500 km; we might use a long-range air-launched ballistic missile for destroying a TBM before launch; we might attempt to extend the ban on ground-launched missiles from a bilateral to an international treaty; or we might work out with the Soviet Union a shared joint stockpile of land-based missiles to be used only for shared international security goals. How about space-based weapons to intercept longer-range TBMs? Clearly, one can fly a weapon to 800 km range while maintaining the ballistic apogee below 100 km.(5) Descending to that altitude, an orbital weapon would experience heating of some 100 watts per square centimeter of frontal area, which would make it difficult if not impossible to use infrared or optical sensing of its quarry. Furthermore, the 1000 Brilliant Pebbles considered for GPALS would not be sufficient to provide more than one or two to a given region during the time a flight of missiles would be near apogee, so penetration would be guaranteed if several TBM were fired in a salvo. In any case, it is perfectly clear that even long-range TBM are better countered by ground-based interceptors than by space-based interceptors. SDIO Director Cooper has stated that the DSP satellites detected every launch of an al Abbas missile; DSP and forthcoming improvements will continue to provide such capability, together with a pretty good indication of trajectory. This will enable a small ground-based interceptor (THAAD, for example) from anywhere within 100-200 km of the target area to be launched to intercept the TBM at apogee, a far simpler job than at-apogee intercept by a space-based interceptor like Brilliant Pebbles. For instance, sensor heating on the ground-based interceptor moving 2 km/s at 100-km apogee is lower by (2/10)**3 (a factor 1/125) than that of a space-based interceptor moving at 10 km/s at intercept. Furthermore, any number of ground-based interceptors could be launched simultaneously to counter a salvo of incoming TBMs. No ground-based radar would be required for such an intercept, anymore than a ground-based radar would be invoked to guide a Brilliant Pebbles intercept. Simply put, if Brilliant Pebbles were expected to work (aside from the heating of its sensor), then the same homing sensor could be used on the ground-based interceptor, with the following aspects that would ease its performance and reduce its cost: o The electronics life in operation need be 5 minutes instead of 5 years. o The sensor need have a field of view of 10 degrees instead of a hemisphere. o The sensor need have a range of 100 km rather than 2000 km. o Batteries can readily power the GBI for 5 minutes instead of solar cells for the BP. Furthermore, the ground-based interceptor to counter TBM would be immune to destruction by Soviet ASAT, unlike space weapons. Nor would such anti-TBM ground-based interceptors imperil the ABM Treaty or interfere with a ban on space weapons or ASAT. The "Global Protection against Accidental Launch System" GPALS is expected to include some tens of Brilliant Eyes, satellites that might be used to track TBMs once the Brilliant Eyes are cued by Brilliant Pebbles. In fact, Brilliant Eyes could just as well be cued by DSP. While it is not at all clear that Brilliant Eyes can do what is required of it, it is clear that whatever it does in the TBM role can more readily be done by similar sensors on an optical probe launched from the ground anywhere in the theater. At most, one such optical probe would be launched per TBM, but its required life would be only a few minutes instead of 10 years. In view of the low-cost "consumer electronics" proposed for Brilliant Eyes and Brilliant Pebbles, it would be eminently affordable to expend an optical probe under such circumstances. Nevertheless, although GBI and optical probes will do a better job of at-apogee intercept than will space-based weapons and Brilliant Eyes, these hit-to-kill intercepts can predictably be defeated by enclosing balloons or umbrellas. We have already discussed this problem in the context of strategic defense.

DEFENSE AGAINST A LIMITED BUT DELIBERATE NUCLEAR ATTACK ON OUR RETALIATORY FORCE BY A RESURGENT SOVIET MILITARY.

No "limited" strike could significantly degrade the ready silo-based force or the submarine-based missiles at sea. If it were advantageous to the Soviet military to destroy Washington by "limited nuclear attack," there are ways to do so for which missile defense would be irrelevant. Our command and control system for the retaliatory force is supposed not to collapse under heavy attack, so it is not clear why a "limited" strike should be more dangerous. Any such deliberate limited nuclear strike would be well equipped with countermeasures against the proposed system and could be countered only by a defense that is likely to call into question the effectiveness of the Soviet second-strike retaliatory force, which would with high probability be the end of negotiated reduction of the Soviet nuclear threat.

SUMMARY.

In sum, it seems imprudent to suggest that we want to modify the ABM Treaty of 1972 in such a way as to permit the Soviet Union more capability against strategic ballistic missiles. Protection of the entire world against unauthorized or accidental launch of Soviet strategic ballistic missiles could better be handled by launch-control improvements than by deployment of an ABM system. As for protection of the United States against some third-world ICBMs armed with nuclear warheads, 100 ABM interceptor missiles at Grand Forks is an excessive number-- perhaps 20 would be more appropriate, but our best protection would be constant vigilance against a threat from the emergence of such missiles and international (coalition) diplomacy and potential military action against such missiles. For protection of our deployed forces, much more attention should be paid to intercept of warheads (and bombs) at very low altitude, and especially to the destruction of TBM before launch. The protection of allied cities against TBM poses a particularly difficult problem, in view of the fact that attack effectiveness is increased by the simple and low-cost measure of splitting the payload among multiple RVs, both for high-explosive and chemical payloads. Destruction of the threatening warheads before they can be launched is the best remedy here, too. In all of this, U.S. space observation capabilities are important, as proved by the effectiveness of the Defense Support Program infrared warning satellites for detection of missile launch. U.S. capability to operate in this medium should be preserved and expanded by negotiation of a ban on space weapons and on antisatellite tests. ---------------- 1 Roger D. Speed, "ASATs vs. Brilliant Pebbles," Lawrence Livermore National Laboratory report UCRL-ID-103669 (March 1990) 2 Richard L. Garwin, Nature, 344, 301-302, 22 March 1990. 3 (because the $600,000 cost of a Patriot interceptor considerably exceeded the cost of a SCUD or other short-range missile, so that in a large-scale attack the defenses could have been predictably exhausted, by SCUDS without nuclear warheads) 4 The U.S. Army reportedly has 53 Patriot batteries already fielded, with a total of almost 1700 launch tubes, and orders of Patriot worldwide amount to 190 batteries and 10,300 missiles. 5 D.C. Wright and L. Gronlund, "Underlying Brilliant Pebbles," Nature, April 25, 1991.