SALT II: VERIFICATION

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP88-01315R000400370001-0
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 6
• Document Creation Date: December 16, 2016
• Document Release Date: December 16, 2004
• Sequence Number: 1
• Publication Date: January 1, 1979
• Content Type: NSPR

EPORT
Approved For Relea QUARTERLY 2u y 8-01315R0004003706
Summer 1979
SALT 11.0 VE
HCATON
While President Carter and Soviet President Brezhnev were meet-
ing in Vienna on June 15-18, 1979, the Ohio Arms Control Seminar
stepped ahead of events to simulate part of the ratification process
of the second Strategic Arms Limitation Treaty (SALT II). Spon-
sored by the Mershon Center's Force and Polity Program, the Ohio
Arms Control Seminar (OACS) has grown during its five years of
existence to include forty-nine members representing eighteen Ohio
educational institutions.
At its annual June meeting, twenty-six members of the seminar
staged a condensed version of the hearings on SALT It before a
mock Senate Foreign Relations Committee. (A narrative account of
the hearings will be published later by the Mershon Center.) The
seminar members, other participants from the Arms Control and
Disarmament Agency, Harvard's Center for Science and Interna-
tional Affairs, Headquarters U.S. Air Force, and local media rep-
resentatives uniformly praised the expertise of the witnesses and
"senators." Although the seminar members are not by training
arms control specialists, they have clearly proved that the in-
tricacies of strategic weapons technology and deterrence theory are
learnable.
While recent public opinion polls show that American public
awareness of SALT Ii has risen throughout 1979, there is no com-
plementary evidence that the public grasp of SALT issues has
shown increased expertise. Yet, the fate of SALT II will rest at least
in part on technical questions that require public education. In no
area is this truer than in the area of treaty compliance or verification.
As the members of OACS anticipated. the verification question
became a critical question during both the simulated and real Senate
hearings. The seminar's concentration on verification profited as
well from the concern of Senator John Glenn (Democrat-Ohio), a
member of the real Senate Foreign Relations Committee. (Senator
Glenn actually attended the simulation and discussed his concerns
with the participants.) The burden of defining the verification issue
fell to Dr. Thomas W. Milburn, Mershon Professor of Psychology
and Public Policy, and his research associate, Mr. Kenneth H.
Watman; a doctoral candidate in political science. As the cor-
nerstone of their testimony, Milburn and Watman prepared the
following paper, which won uniform praise from all the simulation's
participants. It is with great pleasure that the Force and Polity
Program now presents this paper to the wider audience it richly
deserves.
ALLAY R. MILLETT
Director, Force and Polity Program
and Chairman, Ohio Arms Control Seminar
Mershon Center
The Ohio State University
At the heart of any international agreement, especially an
arms control pact, is the ability of each side to verify the
compliance or noncompliance of the other. Without this abil-
ity, almost any agreement is bound to collapse under the
weight of mutually reinforcing suspicions. Therefore, as the
Senate prepares to vote on the proposed SALT II agreement,
we are obliged to examine carefully the charges that the
Soviet Union will be able to evade SALT's provisions with-
out detection. We wil9-ptpllga@'~dti~Qq+bEi-a2~kfl&rdMt'st
fundamental principles of SALT II verification and by look-
ing to the relationship between our technical capabilities and
cog 5,q C T
the provisions of the agreement.
The first and most self-evident principle for verification
is that we do not trust the Soviets. Indeed, it is precisely
because we do not trust them that verification, a substitute
for trust, is incorporated in the agreement. So let us put aside
this red herring as simply irrelevant to the issues at hand.
The second principle of verification is adequacy. We
must make clear precisely what constitutes adequate verifi-
cation. It is tempting to equate adequacy with our ability to
detect and identify the slightest violation of the agreement
regardless of how minute or peripheral, but this would be an
oversimplification. SALT is an international agreement be-
tween sovereign powers for which there is no impartial judge
with compulsory jurisdiction over the parties. The purpose of
verification is to ensure for us the benefits of SALT that flow
from the compliance of both parties, not to win technical,
legal points. Verification has no intrinsic value outside the
value of SALT itself. Therefore, the test of adequacy is the
level of verification that allows us to protect the substance of
SALT and to protect ourselves, a quite different matter from
our ability to detect every nuance of every action taken by the
Soviets that might constitute less than the most precise com-
pliance. This means we must be able to detect a covert Soviet
attempt to develop and suddenly deploy new or old weapons
in numbers that would be strategically significant. Short of
this worst-case situation, our verification abilities must be'
adequate to detect less drastic though still significant be-
havior and trends that would indicate a Soviet intention to,
abandon the restraints of SALT and to provide a timely
enough warning that we can react to protect ourselves. This
perspective can be abstractly stated as follows: Adequacy is,
that level of verification which makes the chances of timely 1
discovery high enough to deter a significant violation. We
will show that the U.S. means of verification are more than
adequate for this purpose.
The third fundamental principle of SALT verification is
relevance. By this we. mean that, in the context of an arms
control agreement, we are not interested in every charac-'
teristic of the Soviet strategic forces; not all those charac-
teristics are relevant to the verification of SALT. Rather, we
are interested only in those qualities of the Soviet strategic;
forces that come within the ambit of a provision of the agree-
ment. Therefore, the requirements for SALT verification,!
both in terms of precision and confidence levels, are less;
stringent than the requirements of our military intelligence.
We believe the issue of relevance is the key to the concerns;
expressed over the loss of our Iranian monitoring stations.
This matter will be explored below in more detail.
The fourth fundamental principle concerns what stan
dard of evidence is required to support a suspicion of viola-
tion. Here again, as with adequacy, we must avoid legalisms.!
We do not require evidence beyond a reasonable doubt. Our:
evidence can be more realistic; we can act on the basis of a
cl suspicion
ic?18tZ9c~e~v instance 9f ambiguous PJ Ads fthatocoould tempt'
 Approved For Release 2005/01/12 : CIA-RDP88-01315R000400370001-0
:pie Soviets to cneat. Our evidence need not meet the stan-
d :rds of the International Court of Justice, but may rise only
t,, that level which stimulates our sense of uneasiness. We
should remember that the Soviet response to our queries is
part of the verification process and that it will serve as a basis
inferring Soviet intentions. If the other side is forthcoming
about bur concerns, we can have greater confidence that the
agreement has not been violated. If the Soviets act in such a
way as to su_}est a violation, we have a number of alterna-
tives open to us. These include the threat of a reciprocal
violation of restraints limited to the sector of the observed
suspicious 'behavior: a demand for immediate rectification
or, in the alternative. a showing of our mistake: a consulta-
tion to reduce the ambiguities; or a warning that we are
alerted and sensitive to further hints of noncompliance.
We believe the U.S. system of verification more than
satisfies the requirements set by these four fundamental prin-
ciples. But the process of detection, of course, rests on our
technical ability to discover what is transpiring covertly on
the other side. It is these technical factors and their relation-
ship to the terms of the agreement that we now consider.
Sensor Technology
This description of sensor technology applies to both
aircraft and satellite reconnaissance. Although a designer of
sensors might wish to monitor a broad electromagnetic spec-
trum, he is actually restricted to just a few frequency ranges.
1. The high-energy radiation in the X-ray and gamma-
ray regions is of limited use because sources of this radiation
are not connected with activities on earth relevant to SALT.
However, they are used to detect nuclear explosions in
space.
2. The ultraviolet region is useful for observing the
earth's ionosphere since atmospheric nuclear explosions and
missile launchings disturb the ultraviolet characteristics of
the upper atmosphere. Indeed, each type of explosion and
missile leaves particularand distinctive ultraviolet signatures
which can be used to identify the source of the disturbance.
The ultraviolet frequencies cannot be used to observe the
earth from space since the atmosphere severely attenuates
radiation in this range.
3. The visible spectrum is the electromagnetic region of
human vision and provides a wide window for observing the
earth. Without going into great detail about such things as
focal length and film emulsions, let us say that the new
generation of observation satellites, Big Bird and the KH-l 1,
cart approach the atmospheric limit on ground resolution of
sia inches to 1 foot from a distance of 100 miles. It must be
said that this level of resolution is usually degraded some-
what by atmospheric, turbulence and cloud cover. Other
areas of the electromagnetic spectrum are not affected by
these problems, as we will relate. and so compensate for the
toss of visible-light acuity. Television, film, and digital
methods are used to transmit visible-light data. Television
has the advantage of providing real-time reconnaissance but
its resolution is somewhat less than what is possible with
film. Film, however, must be recovered from the satellites
and processed so there is an inevitable time lag between
reconnaissance and interpretation. The digital transmissions,
are of a slightly lower resolution than film but are instantane-)
ous like television. %Tu~h progress has been made with low-1
light technology so that photography of high resolution is.
possible under moonlight, twilight, or even starlight.
4. All objects with temperatures above absolute zero
emit electromagnetic radiation continuously. At tempera-
tures below 500? F, most of this natural radiation falls in the
infrared portion of the spectrum. A wide assortment of films,
b d v ! d 'n reco ni-
e o
e
b
g
p
es have een e
tenses, and television tu
tion of this fact. Infrared reconnaissance has two distinct
advantages over visible light: It can he used at night and,
because infrared radiation is emitted by objects and not re-
flected, it can penetrate camouflage or earth. The price paid
is that the resolution of infrared optics is approximately an
order of magnitude less than that for visible light. Short and
long wavelength infrared sensors are carried on early warn-
ing satellites to detect the hot gas plume emitted by missile
launchings. These sensors are also carried by observation
satellites to detect underground missile silos, penetrate
camouflage, and track missiles in mid-course in both day and
night.
5. Unlike visible light and infrared, radar has all-
weather capabilities. In the past, this type of sensor has
suffered both from the need for very large antennae and the
problems of atmospheric attenuation. However, the side-
looking radars installed on observation satellites can now
produce high resolution images by exploiting the motion of
the vehicle to make a small antenna perform like a large one.
6. Radio frequency sensors are also used by the new
generation of Rhyolite satellites to intercept ICBM
telemetry.
The technical abilities of U.S. sensor technology are
important. But equally important is to keep in mind the
redundancy and complementarity of each of these methods
of remote sensing. Each has weaknesses and strengths which
interact synergistically to make systematic evasion ex-
tremely difficult.
Observation Satellites
Observation satellites have two major tasks: area sur-
veillance and close-look. The area surveillance involves the
surveying of very wide areas (all of the Soviet Union or
China) with sensors of moderate resolution. Those data are
then quickly transmitted to earth so that particular areas of
interest can be identified and high resolution, close-look
photographs taken from a lower orbit. The area surveillance
data are usually transmitted electronically; close-look films
are sent back to earth in reentry capsules which are re-
covered in mid-air by specially equipped aircraft.
During the 1960s and early 1970s, the United States had
to use two separate types of satellites to carry out the two
tasks. However, since June 1971, an entirely new fourth
generation of observation satellites, the Big Bird type, has
been introduced to perform both missions. This series carries
both area surveillance and close-look visible-light cameras,
an infrared optical system. side-looking radar, and multi-
spectral sensors. In addition, Big Bird is equipped with a
television camera with a zoom-type. telephoto lens which
transmits real-time images to earth via synchronous com-
Approved For Release 2005/01/12 : CIA-RDP88-01315R000400370001-0CONTIiTTJED
 Approved For Release 2005/01/12 : CIA-RDP88-01315R000400370001-0
munications satellites. Big Bird has the ability to change its
orbit to conduct the area surveillance and close-look mis-
sions and can maneuver to take advantage of breaks in cloud
cover. In the last I3 months. Big Bird has begun to be re-
placed by a fifth generation of observation satellites, the
KH-t I series. carrying even more sensitive sensors and more
efficient methods of information transmission.
In the context of SALT, the observation satellites are
best suited for verifying quantitative provisions. From satel-
lite pictures analysts can'detect new silo construction and
missiles being transported to their deployment sites. Multi.
spectral and infrared photography enable us to detect and
penetrate camouflage and to monitor nighttime activities.
Submarine construction yards can be kept under constant
observation and a count kept of the submarines. Above all,
observation satellites provide timely warning of activity in
violation of SALT by monitoring Soviet transportation net-
works. manufacturing facilities, and power generation with-
out which it is impossible for modern strategic systems to be
constructed and tested. The real-time reconnaissance capa-
bility of both Big Bird and the KH-11 allows analysts to
closely examine suspicious activity. While photography cans
not penetrate buildings, infrared and multispectral tech-
niques can reveal an extraordinary amount of inside activity.
More important, even when the nature of inside develop-
ments remains ambiguous, the sensors can tell analysts that
something of concern might be happening. This will enable
U.S. political authorities to pursue the matter with the
Soviets at the SCC. So, even when the cameras do not
penetrate a deception they will inform us that an effort to
deceive is taking place and will provide enough basis for a
query. Cloud cover has been a problem in the past, but the
side-looking radars now deployed have sufficient resolution
to reduce the ambiguity caused by clouds. Other methods of
in'elligence supplementing the observational techniques are
discussed below.
Surveillance of Missile Tests
The ability to monitor Soviet missile tests is particularly
important in verifying qualitative provisions of SALT.
Therefore, we will not discuss the monitoring systems solely
concerned with war-fighting, such as the DEW line, Semi-
Automatic Ground Environment Back Up Interceptor Con-
trol, and AWACS.
1. Line-of-sight radars have been stationed in Turkey
and Iran to track Soviet missile tests emanating from Kapus-
tin Yar and Tyuratam. In order to track the Soviet vehicles
along the length of the test track, radars of great range and
precision have been deployed at Shemya. Alaska in the Aleu-
tian Islands. Johnston Island, Midway Island, Kwajalein
Atoll, and Bikini Atoll. In addition, radar picket ships and
aircraft are deployed to cover the target areas both in the
Pacific and Soviet Kamchatka. These radars provide the
following sorts of data: First, they can detect the existence of
a test. Second. the trajectory of the test vehicle can be deter-
mined and, from that determination, the range and region of
impact can be inferred. Third, the size and shape of the
missile and reentry vehicle can be determined based on flight
path characteristics and acceleration. Fourth, new missiles
can be detected since every missile currently deployed has a
unique radar signature. Fifth, the sequence and frequency of
the tests allow us to monitor the progression through the
development, test, and deployment cycle of a new weapons
to line-of-sight radar. Such data as the detailed structure of
the reentry vehicle, its ballistic coefficient, and degree of
maneuverability are collected from sensors located close to
the final impact area. These sensors will be discussed below.
2. Unlike. line-of-sight radars, over-the-horizon radars
are not constrained by the curvature of the earth. By reflect-
ing off the ionosphere, OTH radars can penetrate to great
distances. The back-scatter variety of OTH uses the Doppler
characteristics of a signal reflected by a missile to determine
its velocity and acceleration. The forward-scatter variety
detects disturbances in the ionosphere caused by the
ionized jet of gas emitted from rocket engines. Since each
type of missile disturbs the upper atmosphere differently,
forward-scatter OTH can be used to identify a missile by its
signature. Of particular interest is the forward-scatter OTH
called System 440-L deployed in the Far East and Western
Europe. Although the radar was designed as an early warning
system, it has also detected an extremely high percentage of
Soviet and Chinese missile tests.
3. Like the OTH radars, satellite systems that were-
developed to provide early warning of an attack also detect
and track missile tests. These satellites are equipped with
short-wave infrared telescopes and receivers to detect the
exhaust plumes of missiles lifting off, and to pick them up as
they emerge from the earth's atmosphere. To reduce the
danger of false alarms, a teleVt,trih camera is also carried
which is automatically directed at any suspicious infrared
source, The images are then transmitted in real-time enabling
observers to see the object. Once the reentry vehicle gape,
rates from the missile, long-wave Infrared sensors mounted
on aircraft and ships take over to provide accurate trajectory
and warhead data.
4. The most accurate information about reentry vehl?
Iles comes from shipboard sensors located close to the im
pact area. These radar and infrared devices can detect MIRY
testing, the weight and ballistic coefficient of the warhead,
and improvements in accuracy.
5, Last year a now type of satellite, the Rhyolite series,
was placed in geostationary orbit in orderto intercept missile
telemetry from Soviet tests. So far, four of these satellites
have been deployed to intercept data from the liquid fuel
ICBM tests at Tyuratam and the solid fuel ICBM tests at
Plesetsk. Telemetry data are used to help determine charac-
teristics of the test missiles. For example, fuel flow data-
contained in telemetry are one way of calculating range,
throw weight, and launch weight.
It is a happy consequence of the complexity of strategic
systems that very extensive and elaborate testing must pre-
cede deployment. This testing is lengthy and highly visible,
as are efforts to conceal it. Therefore, the ability to observe r
these tests is an integral part of arms control, especially
qualitative limits. The visiblity of the tests discourages the'
development of systems that would violate the agreement. In
the event the Soviets are not discouraged, length of the !
testing period ensures a timely warning for us. In more detail,
the U.S. monitoring capabilities enable us to detect the fol-
lowing qualitative improvements: First, new boosters can be
identified by their characteristic signatures as observed by
line-of-sight (minus Iran)and OTH radars in conjunction with
infrared sensors on early-warning satellites. Second, any
appreciable change in the front end of the missile or the i
reentry vehicle can be detected by the down-range sensors. I
CIA-RDP88-01315R000400370001-80=1uE33
system. Certain other ~OfetDal~@feat$~a~96S1'r0~ 112
 Approved For Release 2005/01/12 : CIA-RDP88-01315R000400370001-0
Third, since significant changes in accuracy must be as- What about the deployment of old weapons of an
soclated with structural changes, important CEP Improve- existing type'? As pointed out above, the production and
ments can be detected. Fourth. the presence of maneuvering deployment of major weapons, even existing ones, is a highly
warheads can easily be detected. Fifth,'MIRV testing can be 1 complex and visible procedure. 'Material has to be trans-
detected by the land-based and ship-borne radars as well as
by satellite systems which can observe the MIRV bus. The
number,of warheads can be directly sensed with radars and
also inferred from the movements of the bus.
Verification of Specific SALT Provisions
The new SALT treaty consists of quantitative and qual-
itative restrictions which the Soviets could try to evade in
three ways: First, they could deploy new types of strategic
weapons. Second. they could deploy more weapons of an
existing type. Third, they could change an existing system so
that its capabilities are increased or changed in a way relevant
to SALT. Let us consider the verifiability of SALT, provi-
sion by provision.
1. Combined strategic nuclear delivery vehicles (heavy
bombers, SLBMs. ICBMs, and air-to-surface ballistic mis-
siles) are to be limited to 2250 by 1982. The first method of
cheating, the deployment of new systems, entails a five-stage
process: research, development, testing, production, and
deployment. The U.S. has a fair to excellent ability to detect
covert activity in all but the first of these stages. This means
that the Soviets have to conceal successfully all four of the
latter stages. A moment's reflection will show that. even in a
worst-case situation in which the U.S. ability to detect cheat-
ing is .5 (it is much better than .5, in fact), the chances of all
four stages escaping detection are .06. When one considers
that a major weapons program requires approximately a
,ninimum of ten years from R & D to deployment, it is clear
that the probabilities of a timely warning are exceedingly
hi 2h. Weapons development takes place at a relatively few
centers in the Soviet Union, much as in the United States.
These are closely observed, and the flows of tell-tale mate-
rials can be monitored. Any weapon prototype has to be
transported to the test sites (which are known) and this
movement is easily detected as are the extensive prepara.
tions and radio traffic at the sites. New weapons require
extensive tests for long, visible periods. Production of
sophisticated systems is concentrated at only a few places.
Finally, deployment is easily observed and there is a long
tithe-period between the beginning of deployment and the
point at which a new system becomes operational. Different
weapons are most visible at different stages. Submarine con-
struction. for example, takes place-at a few yards in the open.
Therefore, submarines are most easily detected at the de-
velopment and production stages. But even at the testing
stage, a craft with new performance is detectable by U.S.
naval units. Tests are especially open to detection in several
ways. U.S. line-of-sight radars produce distinctive signa-
tures of reflected microwaves from each major type of Soviet
missile. A new type of missile produces a new signature.
Similarly. OTH front-scatter radars can detect and recognize
the characteristic pattern each type of missile makes as it
passes through the upper atmosphere. Early warning satel-
lites carrying infrared sensors can identify the exhaust
plumes produced by test firings. Finally, as indicated above,
the United States has a complex assortment of sensors to
monitor the length of the trajectory including the impact area.
Likewise, development, testing, and production of new
heavy bombers is highly visible. Therefore, the likelihood of
the Soviets violating the 2250 limit b introducing news s -
tems covertly is nil. Approved For Release 2005/01
ported, command and control facilities have to be built, new
holes have to be dug just to mention a few of the required
actions. Attempts at camouflage can be detected by infrared
and multispectral sensors. It is true that the probability of
detection varies directly with the size of the violation. There-
fore, it is possible that very small-scale violations could es-
cape the verification net. To be conservative, let us say 100
ICBMs of an existing type could be deployed covertly.
Missile submarines are large vessels displacing up to
9,000 tons and measuring more than 300 feet. Their construc-
tion sites are known and under constant observation and,
since SALT I, the construction itself has been in the open.
Even if the Soviets were able to build submarines covertly,
the vessels eventually will have to be moved from the
shipyard. From that point on, there are innumerable
opportunities for detection, and the chance of a significant
undetected increase in the Soviet missile submarine force is
nil. The same is true for the deployment of additional sub-
marine missiles.
An increase in the number of heavy bombers is equally
subject to detection. The production lines of the Bear and
Bison strategic bombers have long been closed and their
restart w ould be readily discernible. The production facilities
for the Backfire are monitored and an increased rate of man-
ufacture would be quickly detected. Beyond that, the de-
ployment of additional units would be very hard to hide.
Large planes are not easily handled or serviced. New han-
gars. facilities. and Iona runways would have to be built and
these would be virtually impossible to hide.
. Last. what are the prospects of the Soviets converting
existing systems in a way that would violate the strategic
launcher limit? This is a somewhat more challenging problem
since two Soviet systems, the SS-20 IRBM and the Backfire,
could be made into strategic weapons. The two-stage SS-20 is
an intermediate-range missile and as such is not c,;vered by
the strategic launcher provision. Hbwever, verification is
complicated because the SS-20 simply comprises the first
two stages of the three-stage SS-16, an ICBM. Furthermore,
the mobile launcher for the SS-20 is identical to that fay the
SS-16. It is feared that the Soviets could covertly stockpile
SS-16 third stages posing the threat that many SS-20s could
suddenly be trans;ormed into ICBMs. This has been obviated
by the Soviet agreement not to produce, test, or deploy the
SS-16. The SS-16 has been plagued by problems; its last tests
were several years ago and all were failures. If the Soviets
were to pursue the third-stage stockpiling deception, they
would have to be willing to rely on a virtually untested system
with no record of success. We believe the likelihood of this
course to be nil.
The Backfire bomber in its present configuration has a
limited intercontinental capability; it can fly one-way mis-
sions to the United States. Round-trip strategic missions
would require either tanker refueling or a change in the
bomber's configuration. The Soviets have committed them-
selves to inhibiting the Backfire's use as a strategic weapon.
v ONTINUE
 A proved For dRelease 20t%5
/01/1$
These commitments v i e same m ing as o cor-
pus of the agreement and are expected to limit Backfire
production, deployment, and refueling capability. As indi-
cated above, production and deployment are easily verified.
The tanker restrictions are predicated on the difficulty of
mid-air refueling and the amount of practice and training
required to develop confidence in the procedu-e. Practice
runs can be monitored in several ways including ELINT and
COMINT. The Soviets are faced with attempting wartime
refueling without practice if they want to avoid U.S. observa-
tion. Verifying the range configuration of the Backfire is quite
difficult and skillful camouflage would probably enable con-
figuration changes to- evade detection. The restriction in
Backfire production is to compensate for this.
2. Of the 2250 permitted strategic nuclear vehicles,
neither side is allowed more than a combined total of 1320 of
the following types: launchers of MIRVed ICBMs, launchers
of MIRVed SLBMs, heavy bombers equipped for long-range
cruise missiles, and MIRVed ASBMs. Of the 1320, neither
side is permitted more than a combined total of 1200 launch-
ers of MIRVed ICBMs, launchers of MIRVed SLBMs, and
MIRVed ASBMs. Of the 1200, neither side is permitted more
than 820 launchers of MIRVed ICBMs. - ,
There are four ways the Soviets could try to evade the
MIRV/ALCM limit, all falling into the second category of
cheating: deploying more weapons of existing types.
First, the Soviets could build more submarine missile
tubes or landbased missile silos to accommodate more
MIRVed missiles. As indicated earlier, either of these two
courses is readily observed.
Second, the Soviets could try to substitute MIRVed for
unMlRVed missiles. This possibility exists because the
Soviets have a number of unM.IRVed missile types and silo
fields containing unMIRVed missiles. This is also true for
some Soviet missile submarines. Detection depends upon
our ability to know which missiles are MIRVed and which
silos contain them. This is precisely the rationale for the
counting rules upon which the United States.has insisted.
Both sides have agreed that if a missile type has been tested in
a MIRVed mode, all missiles of that type will be counted as
MIRVed. But, would it be possible for the Soviets to substi-
tute covertly MIRVed for unMIRVed missiles in existing
silos? The answer is "no" because of the known dimensions
of the missiles and !?te requirements for MIRVing. Silos that
contain. missile types counted as MIRVs under the rules look
,i4flif1:4ntly different from :silos that do npt. Further, silo3
cc ;tteinln MIRVed missllea require clearly different
,:t7rnm.and-and-control facilities which ure easily detected,
All of these constraints are equally true of stubmarine missile
Third, the Soviets could try to evade the limit by jurrep-
titiously substitutin9'MIRVed for unMlRVed warheads or by
increasing the number of MIRVs on existing MIRVed
warheads. As with some of the other items, this method of
CIA-RDP8,8-01315 R000400370001-0
sequences. archer, these alterations cannot be accurately
predicted by simulation. Therefore, a program of testing is
required for all such charges and MIRV testing is open to
easy detection. For these reasons, the MIRV counting rules
limiting the SS-18 to ten MIRVs, the SS-19 to six MIRVs, the
'SS-17 to four MIRVs, all SLBMs to 14 MIRVs, and all new
ICBMs to ten MIRVs are verifiable.
Fourth, the Soviets could try to cheat by placing more
ALCMs on their bombers than permitted. The arithmetic of
the subiimitsows either side to deploy up to 120 heavy
bombers armed with ALCMs. The average number of
ALCMs per new bomber must equal 23, the average number
per existing bomber must equal 20. These conditions heavily
favor the United States since the Soviets lag behind us in
ALCM development, but assutne for the sake of argument
that they do not. The Soviets have never mounted cruise
missiles internally; they are all externally mounted and visi-
ble. Installation of internal launchers would be a major effort,
probably requiring the flight of eligible bombers to central
facilities for conversion. The United States mopitors such
activity quite closely. In addition, the dimensions of the
Soviet ALCM are different from those of gravity bombs, so
the bomb bay doors would have to be altered for internal
mounting and this is detectable. More difficult to verify is the
payload of ALCMs. It is not possible to distinguish between
an ALCM carrying a nuclear warhead and one that is not, at
least not by external observation. Again, this is really the
Soviets' problem rather than ours since they are not expected
to have precision-guided ALCMs (nuclear or nonnuclear) for
the duratioh of the treaty.
5. In exchange for the ALCM provisions, the treaty
allows the Soviets 308 modern large ballistic missiles.
MLBM is defined as any missile larger than the SS-19 (8,000
tons) with an upper limit of the size of the SS-18 (16,000
tons). As indicated above, the U.S. capability to monitor
Soviet testing enables us to determine with great accuracy
the SALT-related characteristics of a missile. In addition,
deployment of a missile violation of the MLBM limit would
be easy to observe as evidenced by the SALT I debate over
the substitution of the SS-1I with the SS-19.
6. The treaty bans rapid reload systems for both sides.
This can easily be verified both directly and indirectly. Di-
rectly, loading a large missile weighing many tons with relia-
bility requires elaborate, visible equipment near the silos.
Indirectly, additional missiles have to be stored quite near to
the silo fields to make the reload rapid. They have to be
transported to the area and sheltered there. Such activities
are very difficult to conceal.
The protocol contains several temporary prohibitions to
remain in force through 1481.
1. Both sides are prohibited from testing and deploy-
ment of new types of ICBMs with the exception of one new
ICBM for each side. Newness is defined as five percent
difference in volume or performance between the new missile
and the Minuteman III or the SS- 19 with a ceiling of the
che:a?ng di:pends upon the Soviets' willingness to rely on
unt,stcd systems. Substituting a MIRVed for an unMIRVed
w.irheid or increasing the MIRV number completely alters
the flight characteristics of the warhead and the missile. At
intercontinental ranges, small errors have very great con.
CONTiNurm
Approved For Release 2005/01/12 : CIA-RDP88-01315R000400370001-0
 Approved For Release 2005/01/12 : CIA-RDP88-01315R000400370001-0
,.;.r creases the risk and subsequent costs of violations. Studies of
Ti-13. This provlsian can he veriClad through our ability to
monitor the edits of Sovlat rnkk-llei and, from those torts, to past Soviet risk-taking behavior appear more relevant than
refer missile chartt4tatirtics. doctrine-based assertions about Soviet motives either to de-
2, Both sides are prohibited from deployln,; and witlIng fend against invasion or to dominate the world. Such is the
mobile iCSM launt:hers, iu4h activities are higitly vl+ibla case because opposite motives can lead to the same behavior.
b'
k
f
l
and easily v,,-rifled, One possible umbl uity is this SS-20
IRI3M mobile launcher which is identical to the S5-16 ICBM
launcher. "Phis problem Is resolved in the same way this
third-sf:Ige problem was: The SS-20 mobile launcher is no
{trategic threat if the SS-16 ICBM cannot be used.
3. GLCMS and SLCMS are limited to a range of 600
kilometers. A:;ain, this is much more a problem for the
Soviets th;trr for us since they are .tpproximately ten years
behind in cruise missile development. The ranges of cruise
missiles cannot bo verit?e:d because, unlike other strategic
missiles, cruise missiles need not be tested at even near full
range for the military to have confidence in its performance.
The United States is in the midst of flight testing highly
advanced, compact cruise missiles and these are not
scheduled for significant deployment until after the protocol
expires. The Soviets have not reached even this stage, so it is
difficult to see how the problems of SLCM and GLCM range
verification threaten the advantages of SALT II for the
United States.
It has been emphasized that the United States has many
overlapping, redundant means of checking Soviet com-
pliance with the SALT II treaty. Given the costs to the Soviet
Union of being detected in a clandestine strategic weapons
development prohibited by SALT II-ranging from curtail-
ment of trade and technical aid to a vast increase in the arms
race or total U.S. abrogation of the treaty-what is the likeli-
hood of such a violation? In general, the more effective the
verification, the less likely it is to be needed because in-
creased probability of discovery of forbidden activities in-
e, Is am rguous In
e,
or examp
A preemptive military sta
that it may serve defensive or offensive ends. The Soviets
have recently been responsive and cautious: U.S. responses
to Soviet actions that might have constituted violations of
SALT I led either to Soviet explanations the United States
found satisfactory or to cessation of the behaviorin question.
Moreover, assessment of Soviet risk-taking behavior over a
number of historical cases has led U.S. scholars ranging in
ideology from moderate to right-wing to conclude that the
Soviets are cautious, conservative risk-takers. The Soviet
military establishments in particular have preferred to ad-
vance only under conditions of the most favorable odds, the
better to absorb the inevitable uncertainties associated with
t military engagements. The immense size of the stakes in
Soviet-U.S. relations would appear to increase Soviet cau-
tiousness and perhapsto decrease the probability of attempts
I to violate the agreement that representatives of both nations
have signed.
In summary, all strategic systems are highly complex
and react in nonpredictable ways to engineering changes. To
employ missiles with any degree of confidence, extensive
testing, i.e., ten to thirty tests, is necessary, and extensive
testing tends to be highly visible. In interpreting actions of the
Soviet Union the United States is interested in patterns and
trends as contrasted with isolated events. The confidence
level necessary to rely upon a strategic weapons system is
quite high for both the United States and the Soviet Union,
and the Soviets have shown themselves to be cautious, low
risk-takers in situations of importance to them. e.g., national
security. SALT H. of course, is a matter of relevance to the
national security of both nations. is
Approved For Release 2005/01/12 : CIA-RDP88-01315R000400370001-0