Foreign Relations of the United States, 1969–1976, Volume XXXIV, National Security Policy, 1969–1972

14. Paper Prepared in the Department of Defense¹

Washington, undated.

AMENDMENTS TO FY–70 DEFENSE BUDGETS

This paper will address two principal items in the FY–70 Defense Budget: the Sentinel Program and Operations in Southeast Asia.

A. Missile Defense Alternatives

Introduction

In 1967, the United States initiated a ballistic missile defense deployment program called Sentinel. This ballistic missile defense system was composed of the radars and interceptor missiles developed by the Army in its Nike-X development program. These components were designed to defend a variety of missile threats and to be put together in many ways so as to perform any one of several missions or any combination of missions. This paper summarizes the results of a review of several defense missions and of possible arrangements of the Sentinel components to determine if changes should be made. This review also considered deploying no missile defense at this time.

In order to understand the alternatives we have at this time, it will be useful to review the characteristics and operational capabilities of the various Sentinel components.

There are basically two types of ballistic missile defense, area and local (terminal) defense. In area defense, a single interceptor can defend areas of the country several hundreds of miles across. In local defense, the area defended can be 100 miles in diameter.

In Sentinel, a large radar called the Perimeter Acquisition Radar (PAR), has been designed to detect and accurately track missiles at ranges up to 2000 miles. Based on information from the PAR, a Spartan interceptor, carrying a multi-megaton warhead with a lethal radius of up to 30 miles, is launched to intercept the incoming warhead high above the atmosphere and hundreds of miles from the launch site. A much smaller radar, the Missile Site Radar (MSR), is located at the Spartan launch site and is used to guide the Spartan to an intercept by accurately tracking both the incoming warhead and the Spartan. Because of the large area coverage from a Spartan site, only several sites are required to provide protection for the entire United States. PARs are needed around the borders of the U.S. to provide detection and tracking of ICBMs and submarine-launched missiles (SLBMs).

It is possible for an attacker to make use of confusion devices which could be distributed above the atmosphere to hide the exact location of the RV from the area defense. For such attacks, local defense has the advantage over area defense of being able to wait until these devices burn up or separate from the RV on reentry. Because of this necessity to wait, a local defense interceptor can defend only a relatively small area, say up to 30 miles across, when such confusion devices are used. This requires a very fast and small interceptor. The Sprint was developed to carry a nuclear warhead with a yield of a few kilotons and to fly 50 miles at altitudes to 100,000 feet in about 50 seconds. It operates with a MSR which could sort out confusion devices and guides the Sprint to the incoming warhead. To be the most effective the MSR must be close to the point where the warhead comes into the atmosphere. The Sprint, because of its short range, must also be located in or near the defended area. Hence, the MSR and associated Sprints required for each city or military installation (such as a group of Minuteman sites) to be provided a local defense.

A combination of area and local defense can be obtained by placing MSRs, Spartans, and Sprints in a single defense complex. Since the PARs are the eyes of the area defense system, they should be collocated with an MSR/Spartan site and protected by Sprints to insure their survival. This multiple use of the sites and equipment enables significant savings over several single-purpose defense systems.

Alternatives

Four basic alternative ways to combine these defense components for several objectives have been examined. Each will be discussed below.

1. Defense of Cities Against USSR

a. Objectives and Options

(1) Limit damage to U.S. urban/industrial centers in event of nuclear war and enhance national survival and recovery possibilities.

(2) Provide area defense against emerging CPR missile threat and accidental launches.

b. Description

This defense system would essentially be the Sentinel system as originally designed with three major additions: (1) two PARs to give complete radar coverage against SLBMs and Fractional Orbital Bombardment Systems, (2) Sprints to the MSR sites already near the large cities, and (3) new MSR/Sprint sites near additional cities. Such a defense system would have terminal defense for 25 of our key industrial centers with a minimum of about 1000 Sprints as well as an area defense with a minimum of 500 Spartans. The deployment could be started in early 1973 and would be completed in 1977.

c. Costs

The estimated investment cost for such a minimum system would be $11 to $12 billion. The required funding per year, including operations and R&D, is estimated to be:

FY–68	FY–69	FY–70	FY–71
$590M	$960M	$2B	$3B

d. Discussion

Pros

(1) Deterrence of Soviet attacks is a function of our overall strategic capabilities, not only our retaliatory capacity. This system would strengthen our deterrent against the Soviets.

(2) In the event deterrence fails and U.S. urban/industrial centers are attacked, it would save lives and help ensure a favorable war outcome.

(3) It would also satisfy other missile defense objectives such as protecting against emerging Chinese ICBM threat and accidental launchings.

(4) The area defense aspect of this system would provide some protection for our retaliatory forces; options would exist for increasing this protection by deploying additional Sprints around ICBM sites, for example.

5. It would provide defense of our National Command Authority.

Cons

(1) Our basic strategic objective is deterrence of a nuclear attack on the U.S. and its allies. To meet this objective, we first buy forces that give us a very high confidence retaliatory capability. We also buy conventional forces to handle situations that otherwise might escalate to nuclear war. We believe these forces make nuclear war an extremely remote possibility. If deterrence works, we avoid nuclear war altogether.

(2) We believe that the Soviet Union also places great emphasis on avoiding nuclear war and that they size their strategic offense forces to have a retaliatory capability that could survive an attack by the U.S. Thus, we expect that the Soviets would and could respond to large U.S. missile defense deployments that tend to diminish their retaliatory capability by expanding and improving their offense forces. In the long-run, it does not appear possible to materially reduce the vulnerability of our urban/industrial centers to Soviet attacks, independent of our expenditures on missile defense of cities.

(3) If we desire to protect against the emerging Chinese ICBM threat, accidental launches, or Soviet threats to our retaliatory forces, we could do so at significantly lower costs with different deployments.

(4) A decision to defend our cities against Soviet attacks would stimulate further expenditures in the already expensive strategic arms race, and would adversely complicate possible future arms limitation talks.

(5) There would be adverse domestic political reactions to the deployments in this system.

(6) Allied reaction might well be that the U.S. is retreating toward a “Fortress America” strategy.

2. Area Defense Against Chinese ICBM (Sentinel)

a. Objectives and Options

(1) Provide area defense denying damage against emerging Chinese ICBM attack and guarding against accidental or demonstration launch of a small number of ICBMs from any nation.

(2) Provides some protection for Minuteman and the option for additional defense of these forces when and if needed.

b. Description

The Sentinel system would consist of 16 MSR/Spartan sites providing area defense of the continental U.S. and Alaska. A Sprint battery would protect the island of Oahu. Six PARs would be located across the northern U.S. and Alaska. No radar coverage against SLBMs would be provided. The PARs would be collocated with MSRs and given Sprint protection. Approximately 500 Spartans and 200 Sprints would be deployed. Four of the MSR–Spartan batteries would be located in Minuteman fields to provide a portion of CONUS area defense and the option for later addition of Sprints for local defense. The other MSRs would be located near cities to provide better protection against growth of the Chinese ICBM threat and to provide for Sprint defense of the city should that become desirable in the future. The deployment could be started now with the first site becoming operational in early 1973 and the last early 1975.

c. Costs

The estimated investment cost is approximately $6 billion. This cost plus operating and R&D costs require funding at roughly the following rates:

FY–68	FY–69	FY–70	FY–71
$590M	$960M	$2B	$3B

d. Discussion

Pros

(1) There is evidence that the Chinese could have an initial ICBM force by 1972 and about 20 ICBMs by 1975. The Sentinel system can provide a damage-denial capability against this emerging threat. The effectiveness in reducing U.S. deaths from a Chinese attack after completion of the system is shown below:

U.S. Deaths From A Chinese First Strike in the 1970s

Number of Chinese ICBMs on Launchers U.S. Deaths (Millions)	10	25	50	75
With no Defense	7	11	18	23
With Sentinel	0+	0+	0+	1

(2) This system can provide protection against accidental ICBM launches.

(3) This system can also provide through qualitative and quantitative improvements a damage limiting capability against an improving Chinese threat in the late 1970s.

(4) It provides some limited protection for Minuteman sites, bomber bases, and command-control centers from ICBM attacks.

(5) It provides options for adding terminal defense to Minuteman sites; to some cities.

(6) By enhancing U.S. deterrence, it strengthens the credibility of our commitments to defend our allies against nuclear intimidation.

(7) It lessens China’s ability to drag the U.S. and the Soviet Union into a nuclear war.

(8) This level of ABM defense may strengthen our position in entering possible future arms limitation talks.

(9) It provides some protection against small Soviet attacks and complicates their targetting.

(10) It provides all of the above yet does not deprive the Soviets of their second-strike capability, whatever way they might measure it.

(11) It can become operational by the end of 1972.

Cons

(1) Our overwhelming strategic offensive forces and our conventional force capabilities are sufficient to deter Chinese nuclear attacks on ourselves and on our allies.

(2) An anti-Chinese oriented ABM system might overemphasize Chinese nuclear capabilities. The increased fear of Chinese nuclear attack, coupled with the awareness of their vulnerability to such attacks, would cause concern to our allies.

(3) It might also suggest to other nations that we think the Chinese might act irrationally, thereby adding to the above concerns.

(4) It might keep Asian countries from adhering to a nonproliferation treaty by drawing attention to the threat and causing them to raise demands for their own defense, possibly as a step toward developing their own offensive nuclear capability.

(5) The Soviets may perceive this limited ABM system as a first step towards U.S. deployment of a larger system, and may begin to take offensive counteractions to hedge against such a possibility.

(6) The Soviets have slowed down their ABM deployments, although R&D has been speeded up, and have expressed strong interest in discussing limitations of both defensive and offensive systems. Insofar as we would get committed to the full deployment of this system, this might complicate any agreement we might seek to negotiate with the Soviet Union on ABM limitations.

3. Modified Sentinel

a. Objectives and Options

(1) Provides defenses for our Minuteman sites, SAC bomber bases, and our National Command Authority and its communications against a Soviet attack. Additional defense of Minuteman can be provided when and if needed.

(2) Provides coverage of our more heavily populated areas against an emerging Chinese ICBM threat with the option to include defense of Hawaii and Alaska.

(3) Provides protection against the accidental launch of a small number of ballistic missiles from any power.

(4) Provides further options to (a) accelerate protection of urban/industrial centers against an emerging Chinese ICBM threat, (b) incorporate protection of the Combat Operation Centers at Colorado Springs and Omaha against a moderately heavy attack, and (c) incorporate new generation radars and missiles from R&D programs to provide improved capabilities should the threat dictate.

b. Description

The Sentinel system as designed would be rearranged to provide for the above objectives. Complete radar coverage against ICBMs, SLBMs, and FOBS would be provided and the MSR/Spartan sites would be moved away from large cities to locations that provide the best protection of our bomber bases against surprise attack and SLBMs. One PAR site at Alaska is eliminated and two PARs are added in Southern California and Northern Florida, respectively. Five MSR/Spartan sites would be eliminated, three from the interior of the U.S. and one each from Alaska and Hawaii. There would be a total of about 12 MSR/Spartan locations. The four MSR/Spartan sites in the Minuteman fields would be able to provide a portion of area defense and would preserve the option to add Sprints for local defense of Minuteman. A MSR/Spartan/Sprint site would be located at Washington, D.C. to protect the National Command Authority and its communications. A few Sprints would be added at each of the radars to provide some additional defense against attack. About 450 Spartans and 200 Sprints are needed for this system. Due to the requirement for new detailed site selections and evaluation analyses, the first site would not be operational until late 1973; the deployment would be completed in early 1975.

c. Costs

The estimated DOD investment cost of this system would be roughly about that of the Sentinel, or about $6 billion.

The estimated total funding requirements, including operations and R&D, would be:

FY–68	FY–69	FY–70	FY–71
$590M	$750M	$1.5B	$2.0B

d. Discussion

Pros

(1) It allows some protection of our Minuteman ICBMs against the Soviet missile threat.

(a) Although we can maintain Assured Destruction with high levels of destruction on the Minuteman force, we are concerned about maintaining an expensive force that might become vulnerable. Such forces, in a period of extreme crisis, may invite an attack rather than deter one if the enemy knows he can probably destroy the force. Therefore, we should protect the Minuteman or be prepared to phase out land-based missiles, or be prepared to develop other alternatives.

(b) Even though the Soviets are not expected to have an adequate force (an accurate MIRV) to destroy Minuteman for several years, we must maintain options against the possibility that they could. Therefore, this option, by providing a radar network as a base, allows us to make follow-on decisions at an appropriate time.

(c) We have investigated several alternatives for protecting Minuteman against a “greater-than-expected” Soviet threat: (a) ABM defense, (b) adding or relocating Minuteman in superhard (3000 psi) silos, and (c) combinations of above. We have examined these options against the accurate Soviet MIRV threat and have compared the near term and relative costs to keep about one-third of the Minuteman force surviving, shown in the table below:

	Costs ($Billions)
Force Option	R&D	Investment	Level-Off Annual Cost2
ABM Defense3	0	5.9	0.6
New MM in Hard Rock Silos (HRS)	0.3	8.8	0.5
New & Relocated MM in HRS	0.3	10.2	0.5
ABM and New MM in HRS3	0.3	7.0	0.5

ABM defense of Minuteman is expected to be less expensive in the initial years and probably less expensive over all than the other force options identified above. In addition, defense would be much cheaper against lesser threats such as the development of a smaller number of large accurate MIRVs for their large missiles. This is something the Soviet technicians might develop as a product improvement without a decision by their national leaders to adopt a major damage limiting strategy. However, we would not want to rely exclusively on ABM defense of Minuteman. We are uncertain about the effectiveness of a heavy defense against a heavy Soviet attack and the sensitivity of the defense to smaller MIRVs and penetration aids. To counteract this sensitivity we would have to add even larger levels of defense resulting in a larger cost than that shown in the table above. We would need lower levels of defense and smaller numbers of hard rock silos if we deployed a combination and hedged against a greater uncertainty in the threat and in the effectiveness of the defense and hard rock silos. Such a force would be extremely difficult for the Soviets to destroy.

(d) The table below compares the protected Minuteman force with other alternatives for maintaining Assured Destruction. We have assumed the Soviets have deployed a full greater-than-expected threat consisting of accurate MIRVs, improved low altitude air defenses, and heavy ABM defenses of their cities in an attempt to remove our Assured Destruction capability. This is the largest plausible threat we plan against.

Costs to Maintain Assured Destruction (Soviet Greater-Than-Expected Threat)
	Costs ($ Billions)
	R&D	Ten-Year System Costs
Minuteman-Defended and in Hard Rock Silos	0.3	10
700 Additional Minuteman in HRS	0.3	11
20 Additional Poseidon Submarines	0	14
New Land-Based ICBM (WS–120A)	2.6	14
New Sea-Based ICBM (ULMS)	2.0	14

We can protect Minuteman with ABM defenses and hard rock silos cheaper than we can add Poseidons or hard rock silos alone. The new ICBMs, sea-based or land-based, show no clear cost advantage until the greater-than-expected threat becomes stronger than currently estimated.

(2) It provides an effective means of preventing our bomber force from becoming vulnerable to a surprise Soviet SLBM attack.

(a) Our Strategic bombers are a major component in our retaliatory force because (a) they force the Soviets to pay large costs for a balanced defense against bombers and missiles, (b) they hedge against the unexpected failure of missile forces, (c) they are useful for nonnuclear conflicts, and (d) they allow us to quickly increase our force size by simply increasing the alert rate. Bombers are vulnerable to a surprise Soviet attack, since they rely almost exclusively on tactical warning for survival. We have adequate warning of Soviet ICBMs and FOBs through current BMEWS and 440L system and are taking steps to improve this warning with a new surveillance satellite (Program 949). However, against a surprise SLBM attack, even if we get warning at near the time of launch, the missile flight time is so short to some bases that a significant portion of our bombers and tankers may be destroyed before they can be launched.

(b) There are four alternatives to decrease the vulnerability of the strategic bomber force against SLBMs: (1) dispersal, (2) airborne alert, (3) improved ASW forces, and (4) active defense of the bases. We can disperse the bomber force to reduce the take-off time by putting only two bombers and two tankers on each of 250 airfields (all would not be on alert). However, this would cost $200–400 million per year and depressed SLBM trajectories would still make the take-off time marginal. We are not certain about our ASW capability and probably would not rely on it completely. Airborne alert is difficult to maintain over a long period of time since additional crews and increased aircraft maintenance is required. We estimate airborne alert for 40% of our bombers would cost $800–900 million per year.

(c) We have initiated three steps to increase bomber survivability. The first step includes a limited bomber dispersal plan which includes about 67 alert bomber bases, thus increasing the number of targets required for the SLBM. The second is a new satellite warning system which gives tactical warning at nearly the time of missile launch to increase the warning time. With these new plans, a significant portion of the strategic bombers can survive until the SLBM force expands to that projected for about 1973.

(d) ABM defense of the bomber bases against new long-range SLBMs with a good warning system can provide additional time to launch the bombers. In the years after 1973, this defense, with the new warning system and dispersal, significantly increases the bomber survivability as shown in the table below, and reduces the dependency of the bombers on tactical warning for survival. Such a defense could be completed by 1975.

Capabilities for Protection of Strategic Bombers

	FY–70	FY–71	FY–72	FY–73	FY–74	FY–75	FY–76
Threat
Number of Soviet SLBM on-Station (Surprise attack)	0	1–2	2–4	4–7	6–11	8–15	10–15
	% of Alert Bomber Force Surviving Against High NIPP
U.S. Force
Minimum Energy SLBM Trajectories
Satellite Basing with 949	100	95	90	60	40	30	30
Satellite Basing with 949 and ABM Defense	100	95	90	60	80	85	95
Depressed SLBM Trajectories (should this threat develop)
Satellite Basing with 949	100	90	60	35	20	15	15
Satellite Basing with 949 and ABM Defense	100	90	60	35	60	85	95

(3) Since missile and radar sites would not be located in large cities, it would not be perceived by the Soviets as a first step towards a major U.S. ABM program. Because of these reasons, this system is not expected to complicate strategic arms talks.

(4) It would provide nearly the same level of protection of our population against Chinese ICBM threats, small attacks from any nation, and accidental launches as the Sentinel system. However, Hawaii and Alaska could be at risk if we did not exercise the option to defend them.

(5) It would protect our National Command Authority; and maintains the option to protect the COCs in Colorado Springs and Omaha.

(6) It does not call for deployment of ABM interceptors in any major cities, thereby reducing domestic criticism.

(7) It could provide defense against the total threat; ICBMs, SLBMs, FOBS, and growth threats.

Cons

(1) Even after we deploy the system, there is a possibility that the Soviets might develop means to degrade it, e.g., warheads with small radar signatures, depressed trajectories, and other penetration aids. Thus, the effectiveness of the system may become uncertain, or without improvements, become degraded. Such uncertainty is not consistent with our requirement for the maintenance of a high-confidence retaliatory capability, especially if we require a reliable capability in each element of our strategic forces, i.e., our bombers, ICBMs, and SLBMs.

(2) We will have a high-confidence retaliatory capability in our ICBMs and SLBMs, only if the Soviets do not deploy a large ABM system. Thus, even if the bombers can be destroyed in a surprise attack, we still would retain an adequate deterrent.

(3) If the capability to protect bomber bases is to be maintained in the future in the face of growing threats, additional Spartans would probably be needed. However, the Spartans are also capable of defending cities. Thus, the Soviets might view an increase in the number of Spartans as a destabilizing move on our part. It might lead them to increase the size of their offense force to maintain their retaliatory capability.

(4) SLBM warning time would range from three to 15 minutes. This implies that nuclear release authority for defensive missiles must be predelegated to the ABM defense commander (probably CINCONAD). Otherwise, with only the President giving the nuclear release, the time between warning and release authority may preclude intercept of a large number of SLBMs.

(5) We run some risk of not having the system deployed in time.

4. No Missile Defense Deployment

a. Objectives and Options

(1) Continue reliance on strategic offensive capabilities.

(2) Maintain options to deploy various system now under consideration (SABMIS, Nike–X, etc.) with emphasis on options for protecting retaliatory forces.

(3) Reduce costs and domestic criticism.

b. Description

(1) Cancel Sentinel.

(2) Continue ballistic missile defense R&D.

c. Costs

Sentinel can be cancelled shortly. Non-recoverable costs have been incurred. This would result in roughly a $600 M loss, i.e., if the program had never been started, we could have saved $600 M exclusive of R&D costs. The funding requirements to date for Sentinel and for continuing R&D only would be roughly:

	FY–68	FY–69	FY–70	FY–71
Sentinel and Nike–X	$590M	$750M	$350M	$300M

d. Discussion

Pros

(1) Although ABM defense provides the least costly alternative to the protection of our deterrent force against a Soviet attack, we do not have to deploy an ABM defense. We could rely on other alternatives for force protection.

(2) Our overwhelming retaliatory force would deter a deliberate attack by the Chinese.

(3) The high cost for the defense system could be applied to other pressing national needs.

(4) It might enable us to negotiate a complete ban on ABMs with the Soviet Union and thereby simplify certain kinds of verification problems.

(5) No change in our alliance relationships.

Cons

(1) The pros of all previous alternative defense systems.

Recommendation

The review of the pros and cons have led us to select Alternative 3 (Modified Sentinel). On balance, we feel that defense of our retaliatory forces, protection against the emerging Chinese threat and an accidental launch are essential to U.S. national security.⁴

[Omitted here is Section B, “FY 70 Budget Amendments for Support of Operations in Southeast Asia.”]

1. Source: Ford Library, Laird Papers, Box 27, Safeguard. Secret. No drafting information appears on the paper. This paper was prepared in response to NSSM 23, Document 9. Laird sent it to Kissinger as an attachment to a March 1 covering memorandum. On March 4, Kissinger forwarded it to Agnew, Rogers, Laird, and Lincoln for their consideration prior to the following day’s NSC meeting. Copies were sent to David Kennedy, Mayo, Gerard Smith, Helms, Wheeler, and Richardson. (National Archives, Nixon Presidential Materials, NSC Files, NSC Institutional Files (H-Files), Box H–21, NSC Meeting, FY 70 Defense Budget, March 5, 1969)
2. Costs include operation of Minuteman force. [Footnote in the original.]
3. Assumes Sentinel R&D costs are “sunk.” Investment costs for Sentinel equipment in Minuteman fields are included in defense costs above. [Footnote in the original.]
3. Assumes Sentinel R&D costs are “sunk.” Investment costs for Sentinel equipment in Minuteman fields are included in defense costs above. [Footnote in the original.]
4. The Bureau of Politico-Military Affairs generally endorsed the Defense Department’s recommendation. However, in a briefing memorandum sent to Rogers on March 4, Farley recommended pushing for revisions in the program’s public rationale, including a “delay in the program which gives us additional time to explore seriously a strategic arms agreement with the Soviets. Moreover, the rationale should also stress that implementation is subject to modification, depending on the outcome of negotiations with the Soviets.” (National Archives, RG 59, Entry No. 5000, S/S–NSC Meeting Files, 1969–70: Lot 71 D 175, Box 3, NSC Meeting, March 5, 1969)