Foreign Relations of the United States, 1961–1963, Volume VIII, National Security Policy

151. Draft Memorandum From Secretary of Defense McNamara to President Johnson⁰

Washington, December 6, 1963.

SUBJECT

• Recommended FY 1965-FY 1969 Strategic Retaliatory Forces (U)

I have recently completed my review of the long-range nuclear delivery forces and their associated support for FY 1965-FY 1969. The program recommended will form the basis for the preparation of the FY 1965 Budget. This memorandum summarizes the main factors I have taken into consideration in determining United States requirements for these forces.

I believe we should adopt, for planning purposes, the force structure summarized in the table on page two. Where they differ from my recommendations, the forces proposed by the Air Force are shown beneath mine in parentheses. In particular, I recommend:

1. Approval, in FY 1965, of an increase in the total Minuteman force level of [less than 1 line of source text not declassified] missiles (to [less than 1 line of source text not declassified]), at a total procurement cost of $250 million, of which $167 million will be funded in FY 1965.

2. Retrofit of [less than 1 line of source text not declassified] of the [less than 1 line of source text not declassified] missiles in the first five Minuteman Wings with the Improved Minuteman at a total procurement cost of $279 million of which $130 million will be funded in FY’65.

3. Approval of an extensive ICBM reliability improvement program at a 1965 cost of $115 million. The eventual costs will depend on the extent of the program. They are now estimated to be about $537 million over FY 1965-FY 1969.

4. Phasing out of 27 Atlas “D” ICBM’s in FY 1965 instead of FY 1968, phasing out 27 Atlas “E” ICBM’s in FY 1967, and phasing out 54 Titan I ICBM’s in FY 1968, for a total estimated savings of $209 million over FY 1965-1969.

5. Retrofit of the five A-1 Polaris boats with A-3 missiles. Disapproval of the proposed retrofit of the thirteen A-2 boats with A-3 missiles, for a FY 1965-69 saving of $425 million, of which $110 million is realized in FY 1965.

6. Disapproval of the recommendation of the Chief of Staff of the Air Force for procurement of 355 additional Hound Dog missiles at a cost of $388 million.

7. Disapproval of the recommendation of the Chief of Staff of the Air Force for expenditure in FY‘65 of $78 million on development of a new manned strategic bomber.

8. Continuation of a related classified program discussed in a separate enclosed memorandum.

In addition, I recommend provision in the FY 1965 Budget for: studies of alternative advanced manned strategic aircraft; continuation of conceptual studies leading to an advanced ICBM system and an advanced sea-based deterrent system; and continuation of the development of the MRBM, which will be discussed in my memorandum on Research and Development.¹

RECOMMENDED AND SERVICE-PROPOSED FORCESab	End-Fiscal Year
1961	1962	1963	1964	1965	1966	1967	1968	1969
Bombers:
B-52	555	615	630	630	630	630	630	630	630
B/E-47	900	810	585	450	225
B-58	40	80	80	80	80	80	72	72	66
Total	1495	1505	1295	1160	935	710	702	702	696
Air-Launched Missiles: Hound Dog	216	460	580	580	551	522	522	493	493
Surface-to-Surface Missiles: Atlas	28	57	126	126	99	99	72	72	72
Titan		21	68	108	108	108	108	54	54
Minuteman			150	600	800	750	610	480	*
							(780)	(700)	(620)	(*)
Imprv. Minuteman						200	390	620	*
						(170)	(550)	(780)	(*)
Polaris	80	96	144	288	464	560	656	656	656
Total	108	174	488	1122	1471	1717	1836	1882	1982
							(2086)	(2182)	(2182)
Other:
Quail	224	392	392	392	392	392	392	392	392
KC-135	400	440	500	580	620	620	620	620	620
KC-97	600	580	340	240	120
RC-135
		3	3	4	13	13	13	13
RB-47	90	45	30	30	30
Thor	60	60	60
Jupiter	45	45	45
Regulus	17	17	17	17	9
PACCS: KC-135				17	17	17	17	17	17	17
B-47			18	36	36	36	36	36	36	36
Alert-Force Weapons:c Weapons	1074	1512	2184	2689	2765	2785	2753	2854	2930
							(2966)	(3109)	(3100)
*	*	*	*	*	*	*	*	*
Megatons							*	*	*

a The forces proposed by the Secretary of the Air Force, where different from the Recommended Forces, are shown in parentheses.

b The Multi-Lateral Force (MLF) is still under discussion with our NATO Allies. Assignment of nuclear weapons of the U.K., France, or a “Multi-Lateral Force,” to NATO in accordance with the terms of the Nassau Pact, may lead to adjustments in the U.S. force structure.

c Bombers have flexibility in the choice of gravity bombs and yields. For purposes of this table, average planned loadings of 1.5 weapons and [less than 1 line of source text not declassified]MT for the B-47, 3.3 weapons and [less than 1 line of source text not declassified]MT for the B-52’s, and 3 weapons and [less than 1 line of source text not declassified]MT for the B-58’s are assumed. Beginning in FY 1962, 50% of the B-52’s, 55% of Polaris, and 85% of the ICBM’s on operational launchers are assumed to be on alert or on-station. An additional 12% of the Polaris force would be intransit to patrol areas.

* Entry not declassified.

The estimated total obligational authority required to procure and operate these forces is shown in the following table.

TOTAL OBLIGATIONAL AUTHORITY BY FISCAL YEARa	(Billions of Dollars)
1962	1963	1964	1965	1966	1967	1968	1969	1965-1969	Prev. Approved	9.11	8.53	7.29	5.07	4.32	3.59	3.30	2.99	19.27
SecDef Recom.			7.53	5.34	4.35	3.81	3.16	2.76	19.42
SecAF Proposed			7.52	6.09	5.34	3.81	3.24	2.79	21.27
O/SAF Proposed
		25.12

a These and all other cost estimates in this memorandum are preliminary, and are subject to further refinement. The figures for General LeMay’s proposal exclude the cost of a new manned strategic bomber.

The Strategic Retaliatory Forces I am recommending are considered to be adequate by the Chairman, Joint Chiefs of staff, the Chief of Staff, U.S. Army, the Chief of Naval Operations, and the Commandant of the Marine Corps. The results of their review of my recommended forces were summarized as follows:

• “8. Recognizing that, in the time frame considered, it is not possible to assure the limiting of damage, in loss of life, to the United States to a level below the criterion suggested by the Secretary of Defense, we consider that:
  • “a. A vital first objective to be met in full by our strategic nuclear forces should be the assured capability of destroying singly, or in combination, the Soviet Union and the Communist satellites in Europe as national societies. In combination with theater nuclear forces, they should be able to impose adequate punishment on Red China for nuclear or nonnuclear aggression.
  • “b. [13 lines of source text not declassified]
• “9. It is our view that the strategic force structure set forth in your draft memorandum for the President provides appropriately for the foregoing objectives.”²

The Chief of Staff, USAF, recommends [less than 1 line of source text not declassified] Minutemen by 1969 and procurement of an additional 355 Hound Dog missiles, enough to equip the entire B-52 force. The 1965-1969 cost of the Minuteman program proposed by the Chief of Staff, USAF, would be approximately $10,064 million, as opposed to about $4,757 million for my recommended program.³ The additional Hound Dog procurement would cost approximately $388.5 million. This proposal was not included in the program submission of the Secretary of the Air Force.

Within the objective of a force of [less than 1 line of source text not declassified] Minuteman missiles by end-FY 1969, I am recommending an increase in the force level of only 50 missiles in the FY 1965 Budget, instead of the 200 previously planned, for several reasons. First, there has been some reduction in the Soviet ICBM force projected for the late 1960’s. NIE 11-8-62, published 6 July 1962,⁴ estimated 300-600 operational ICBM launchers in mid-1967. The range is now estimated to be 335-525. Second, an FY 1965 increment of 50 missiles should reduce the risks of extensive modification which can arise from difficulties discovered in the Improved Minuteman development cycle and permits a more orderly deployment schedule. Moreover, because of the reliability improvement program, plus a $368 million cost overrun for FY 1963 and FY 1964, and other cost increases, the total obligational authority required in FY 1965 for Minuteman is now increased by $274.8 million over the amount previously approved. Finally, the over-all force effectiveness of the recommended force of [less than 1 line of source text not declassified]basic and [less than 1 line of source text not declassified] Improved Minuteman, is greater than that achievable with the previously approved force level of [less than 1 line of source text not declassified] basic and [less than 1 line of source text not declassified] Improved Minuteman missiles. In terms of target destruction capabilities, the recommended force provides [less than 1 line of source text not declassified] percent greater effectiveness than the previously approved force.

I also recommend that we continue development and procurement of the Post-Attack Command and Control System, which includes 17 KC-135B airborne command posts and 36 B-47 airborne communications relay aircraft. This system permits the maintenance of one SAC command post in the air at all times, plus one on a high state of alert at each of the SAC alternate headquarters. These aircraft can launch and control the SAC force, including launching of the Improved Minuteman in the event that its Launch Control Centers have been destroyed.

Based on advice from the Joint Chiefs of Staff and General Power, I have cancelled plans for the construction of a Deep Underground Headquarters for SAC. While there is a requirement for an increase in the survivable command and control capability for SAC, a re-examination of the operational considerations and associated costs have led me to conclude that this center could not be justified at this time. I recommend that the funding previously approved for FY 1965 through FY 1968, totalling $106 million, be deleted from the program.

The following section describes in greater detail the basis for my recommendations. I shall review first our strategic objectives, the Soviet Bloc nuclear threat, our target destruction capabilities, and general nuclear war outcomes. I shall outline in more detail in Appendix I,⁵ pages 24-29 the key decisions to be made this year.

II. General Basis for Force Level Recommendations

General Nuclear War Objectives

The objectives for our Strategic Nuclear Forces can be summarized under three distinct headings. These objectives provide quantitative tests of the adequacy of our posture.

“Assured Destruction” of the Soviet Union

An essential test of the adequacy of our posture is our ability to destroy, after a well planned and executed Soviet surprise attack on our Strategic Nuclear Forces, the Soviet government and military controls, plus a large percentage of their population and economy (e.g. 30% of their population, 50% of their industrial capacity, and 150 of their cities). The purpose of such a capability is to give us a high degree of confidence that, under all foreseeable conditions, we can deter a calculated deliberate Soviet nuclear attack. The calculations made to test this ability are our best estimates of the results of possible Soviet calculations of what we could do to them in retaliation if they were to attack us. This calculation of the effectiveness of the U.S. forces is not a reflection of our actual targeting doctrine in the event deterrence fails. I will call this objective “Assured Destruction.”

“Damage-Limiting” Forces

Beyond the force required to meet the test of “Assured Destruction,” additional forces may be justified if they could further reduce the damage to the U.S. in the event of a Soviet attack by an amount sufficient to justify their added costs. Such forces might help to limit the damage to the United States both by destroying some of the Soviet nuclear delivery systems, and by disrupting the coordination of the rest, thereby easing the task for our defensive forces.

[Heading and 1 paragraph (13 lines of source text) not declassified]

The following analysis reviews alternative U.S. forces in terms of their ability to accomplish these objectives. The conclusions I have reached are as follows:

1. The forces I am recommending are clearly adequate for the objective of “Assured Destruction” under any reasonable definition.

2. The prospects for “Damage Limiting” by counterforce attacks may not hold great promise in the latter part of the 1960’s if the Soviets harden and disperse their ICBM force and build up their missile submarine force as we now expect them to do. I believe that the recommended forces accomplish what might reasonably be able to be done from this point of view, and that the extra capability proposed by the Air Force would make a contribution to “damage limiting” too small to be justified in the light of its extra cost.

3. A “Full First-Strike Capability” does not appear to be feasible during the time period under consideration with the weapon systems projected for both sides, unless the Soviets choose to buy strategic forces that are both smaller and less well protected than we now expect. In any case, 200 or 750 extra Minuteman missiles, as proposed by the Secretary and Chief of Staff of the Air Force, respectively, would not significantly improve the outcome of the war for us.

Projected Soviet Strategic Posture

The following table summarizes current estimates of Soviet strategic forces in mid-1967 and 1969. For comparison, estimates for mid-1967 made last year are also given.⁶

SOVIET STRATEGIC RETALIATORY FORCES AND DEFENSIVE FORCES	Based on NIE of 6 July 1962	Based on NIE of 18 October 1963
Mid-1967	Mid-1967	Mid-1969
Low	Med	High	Low	Med	High	Low	Med	High
Operational ICBM Launchersa
Soft Launchers (2 per site)
1st & 2d Generation	150	200	250	150	200	250	150	200	250
Very Large ICBM’s				10	20	30	25	40	50
Hard Launchers
2d Generation (3 per site)				100	115	125	100	115	125
Follow-on (possibly one per site)	150	250	350	50	75	100	100	175	250
Tyuratam				25	25	25	30	30	30
Total (rounded)	300	450	600	325	435	525	400	560	700
Operational IRBM/MRBM Launchersa	Soft (4 per point)	550	650	650	600	600	600	500	500	500
Hard (2 per more/point)				100	125	150	200	225	250
Total	550	650	650	700	725	750	700	725	750
Missile Submarinesb
Ballistic
Nuclear	25	28	30	19	24	28	23	30	36
Diesel				41	43	45	41	43	45
Cruise
Nuclear	24	24	24	14	17	20	18	23	28
Diesel				20	25	30	20	25	30
Total Missiles:
Ballistic	75	84	90	173	194	212	185	212	236
Cruise	144	144	144	144	177	210	168	213	258
Bombers/Tankers
Heavy
Bison		70		80	90	95	70	80	90
Bear		35		70	85	95	60	70	85
Total		105		150	175	190	130	150	175
Medium
Badger		500		400	475	525	200	200	300
Blinder		250		200	250	275	200	275	350
Total		750		600	725	800	400	525	650
Total—Bombers/Tankers		855		750	900	890	530	675	825
Defensive Forces
ABM Defended Areas	N/A	N/A	N/A	2	2	2	2	2	15
SA-2	600	600	600	1100	1100	1100	1100	1100	1100
SA-3	400	600	800	460	460	535	500	500	650

a Soft launchers, but not the hard launchers, are believed to have missiles and capacity for refire.

b The NIE is noncommittal on the total number of Submarine missiles. Best estimates have been used for this table.

Projected Soviet Strategic Posture (Contd)

Although projections of Soviet forces in the late 1960’s are necessarily subject to uncertainties, development and deployment patterns have made possible the identification of some broad trends.

ICBM’s and IRBM’s

The Soviets now appear to be deploying ICBM’s in both a soft configuration with two launchers per site and a hard configuration with three launchers (silos) per site. One additional missile is probably available to each soft launcher, but not to the hard launchers, for a refire capability. Our own experience suggests their silo hardness would be in the range 100 to 300 psi. Although NIE 11-8-63 of October 18, 1963, estimates that by the late 1960’s the Soviets will be deploying only one instead of three silos to a site, there is as yet no evidence of such a dispersal.

During the past year, the Soviets apparently have encountered difficulties in the development of the SS-8 and have curtailed its deployment. The SS-8 was suspected to be a large payload missile, 25,000 lbs., but further analysis has indicated that its payload is similar to that of the SS-7 (2000-5000 lbs.). While no new missile development has been identified, the estimate includes the possibility that, if test firings begin in the next few months, a new ICBM capable of delivering very high yield warheads might be operational during FY 1966. No statement is made to imply that such test firings are likely to begin. As a follow-on to the successful SS-7 program, the Soviets may either improve this missile further, or develop a new standard size missile using improved liquid propellants, with initial deployment in 1966-67.

By 1965, the Soviet IR/MRBM force will probably level off at 700-750 launchers. Currently a refire capability for soft launchers is available, and this force is deployed in a four launcher per site configuration. The Soviets are also hardening some of their IRBM sites, and while currently believed deployed in a two launcher per site configuration, three launchers per site is possible. There is no evidence to indicate that follow-on systems are being developed. However, a mobile deployment of a new system is possible.

SLBM’s

The Soviets have under development a 700 n.mi. submerged-launch ballistic missile for their submarine force. Existing submarine launched ballistic missiles have only a 350 n.mi. range and the submarine must surface to fire. At present, Soviet submarines carry at most three ballistic missiles. Although likely, there is no evidence that the Soviets are building a new class of submarine carrying more missiles. In addition to ballistic missiles, the Soviets also have a large number of submarines (both nuclear and diesel powered) capable of launching cruise missiles. It is believed that the cruise missiles will be deployed in support of anti-shipping warfare; however, the possibility that they could be used against land targets is not ruled out.

By mid-1969, the number of submarines carrying ballistic missiles is estimated to vary between 64 and 81, and in the absence of a new class submarine, this force would carry between approximately 185 and 236 missiles.

Long Range Bomber Forces

Although the Air Force believes that the Soviets intend to deploy a new heavy bomber between 1965 and 1967, this view is not shared by other members of the intelligence community. Barring this possibility, there is a projected reduction in both the heavy and medium bomber forces. Evidence indicates that the Bear and Blinder “B” aircraft have available a standoff missile capability. However, the capability for intercontinental attack remains limited, even though the Soviets have given considerable emphasis to Arctic staging exercises and to aerial refueling practice in an effort to overcome range deficiencies of their bomber force. (The range of the medium bombers is shorter than previously estimated.) The NIE suggests that the Soviets could place only 90–115 heavy bombers over target areas in the United States on two-way missions.

Availability of High Yield Weapons

We expect that in 1969 the yields of most Soviet ICBM’s will be in the range between 6 and 13 MT. The Soviets could have available some high-yield weapons (100 MT for delivery by both bombers and missiles) to use as “terror weapons,” or to employ against hard targets or cities.

The Soviet homeland defenses, including civil defense, are discussed in Appendix II, pages 30 to 31.

Strategic Targets in the Soviet Bloc

A projected list of Soviet Bloc targets was derived based on the NIE estimates of the Soviet strategic offensive forces for the mid-1969 period. In addition to cities, the list includes primary military targets which represent a threat to the U.S., Western Europe and overseas theater forces. This projection includes targets which would compromise the Allied Command Europe (ACE) Threat List. The number of these targets which would be attacked by theater forces and would not have to be scheduled for attack by our Strategic Retaliatory Forces is uncertain. The total Sino-Soviet Bloc Target List which is of primary interest to the U.S. is shown on the following page.

SOVIET BLOC TARGETS END-FY 1969

[Table (1 page of source text) not declassified]

[Heading and 4 paragraphs (17 lines of source text) not declassified]

There are, in addition, other unscheduled targets of opportunity which would be attacked in connection with the land battle.

The extent to which U.S. “external” forces (i.e., U.S. forces not under SACEUR’s command) are now scheduled to attack targets on the ACE Threat List is shown below.

[Table (1/2 page of source text) not declassified]

Forces Required for “Assured Destruction”

Returning now to the broad strategic objectives mentioned earlier, I would like to address first the forces required for “Assured Destruction.” The effectiveness of our posture as a deterrent to deliberate Soviet attack depends on the Soviet calculation of what we can do to them in retaliation if they attack us. Although the Soviets would doubtless consider the prospective military outcome of such an attack, they would have to give great weight in their consideration to the number of people we could kill and the amount of damage we could cause to their industry. Therefore, in considering “Assured Destruction,” I shall calculate he destructive capacity of our force on the hypothetical assumption that [less than 1 line of source text not declassified] even though in fact we would not use our forces in that manner if deterrence failed. The key decision we now face is the total size of our Minuteman force. Through the FY 1964 Budget, a force of [less than 1 line of source text not declassified] Minuteman missiles has been authorized. In the calculations that follow, I shall assume that all other forces are held constant at the levels shown from FY 1969 in the table on page 2, and then vary the number of Minuteman missiles as we consider alternative objectives.

In 1969, without any Minuteman missiles, we plan to have about 1,900 weapons and [less than 1 line of source text not declassified] in the alert portion of our Strategic Retaliatory Forces. Assuming that the Soviet Union, using the “Medium” force, attacks our forces with a well planned and executed surprise attack, and that in all cases our expected (i.e., most likely) estimates of operational factors (listed in Appendix III, pages 38-39)⁷ prove to be correct, the U.S. force targeted as hypothesized above could be expected to deliver about 700 weapons and [less than 1 line of source text not declassified] on about [less than 1 line of source text not declassified] DGZ’s.⁸ Recognizing that such estimates are necessarily uncertain, the “most likely” factors imply that such an attack [2 lines of source text not declassified].⁹ On this basis, then, the programmed forces for 1969 appear more than adequate without any Minuteman missiles to meet the test of a capability for “Assured Destruction.” And, with the inclusion of the already programmed force of [less than 1 line of source text not declassified] Minuteman missiles, the force could [less than 1 line of source text not declassified].

But these calculations depend on the assumptions made about many uncertain variables. In fact, there are several key uncertainties and possible alternative assumptions considerably less favorable to ourselves than those underlying these estimates. Because of the critical importance of our ability to retaliate against deliberate surprise nuclear attack, I believe that our force requirements for this mission should be tested against pessimistic assumptions as well as against our best estimates. We want both fully adequate insurance against future contingencies and to be able to deter even an “optimistic” Soviet decision-maker.

We recognize that it is not normally sensible to design the defense program entirely on the basis of pessimistic estimates. In fact, it can be just as dangerous to overstate as to understate enemy capabilities, for it can lead to the pricing out of the market of valuable capabilities that would appear very useful under realistic estimates of the threat, and it can lead to the adoption of strategies of desperation. In some cases, the fact that a capability exists under optimistic estimates may be significant. However, deterrence of deliberate nuclear attack is so fundamental to our whole defense posture that I believe we should require that our retaliatory power to [be] beyond any reasonable question.

The major uncertainties affecting the assured retaliation capabilities of our Strategic Nuclear Forces can be grouped under four headings:¹⁰

1. Improved Soviet Defenses,

2. Larger or More Effective Soviet Offensive Forces,

3. Lower U.S. Missile Reliability, and

4. Unanticipated Wartime Degradation in U. S. forces.

The effectiveness of U. S. forces for “Assured Destruction” under each class of uncertainties is discussed in Appendix II, pages 31-33.

The effect of making all pessimistic assumptions, however unlikely such a contingency might be, as opposed to making the assumptions we consider most likely, is shown for several alternative Minuteman forces in the following table. In each case, it is assumed that the rest of our forces as shown in the table on page 2¹¹ for FY 1969.

[Table (1/2 page of source text) not declassified]

Granted there are uncertainties, in my judgment these calculations show that to the extent that the prospect [2 lines of source text not declassified]already authorized, in combination with the other planned forces should be clearly adequate for “Assured Destruction,” while a larger force would add very little from this point of view. From a Soviet viewpoint, the potential damage that could be inflicted by a United States retaliatory attack is so severe that they, the Soviets, should be deterred from initiating general nuclear war.

Because of the concentration of Communist China’s industry, [less than 1 line of source text not declassified]Minuteman missiles can destroy about [less than 1 line of source text not declassified] of the industrial capacity or alternatively [less than 1 line of source text not declassified] Minuteman missiles can destroy about [less than 1 line of source text not declassified]. My recommended program is more than sufficient to cover this possible requirement.

“Damage Limiting” Capabilities of Alternative U.S.Forces

In this section are discussed the “damage limiting” potential of alternative Minuteman forces acquired for second-strike [less than 1 line of source text not declassified]. Of course, in the range of conditions most likely to be faced in 1969, almost all of our ICBMs and all of our bombers, and perhaps some Polaris missiles should be available for [less than 1 line of source text not declassified].

The effectiveness of the U.S. second-strike missile attack when applied to [less than 1 line of source text not declassified] is shown below for three alternative forces. Force I represents the FY 1964 Budget to be in place by end FY 1966. Force II is my recommended program including [less than 1 line of source text not declassified] Minutemen by end FY 1969, and Force III is Secretary Zuckert’s proposal including [less than 1 line of source text not declassified]Minutemen by end FY 1969.¹² All Minuteman/Polaris missiles are programmed to utilize a “missile-away” indicator (which allows the commander to assess whether the missiles completed count-down and were launched). Furthermore, the greater flexibility of Improved Minuteman is used to enhance the performance of the basic Minuteman force by a combination of retro-fitting improved missiles in Wing I through V silos and co-location of Improved Minuteman squadrons with existing Minuteman wings.

[Table (1/2 page of source text) not declassified]

[5 paragraphs (1-1/2 pages of source text) not declassified]

Little is known about the Soviet targeting doctrine. The Soviets would probably plan on attacking both military targets and cities. But it is possible that we might be able to deter them from attacking cities or that the war could be terminated before massive city attacks are executed. In the table on the following page, cases are considered in which the Soviet attack against our cities are either executed immediately, delayed, or withheld. With respect to the extent of U.S. active and passive defense, if we have fallout protection and effective defenses, [less than 1 line of source text not declassified] can make the job of our defenses easier and reduce directly the number of weapons falling on the United States. Alternative U.S. [less than 1 line of source text not declassified] are compared under alternative sets of assumptions representing combinations of the above conditions. A case is also shown in which [1 line of source text not declassified]. This case will be designated Force A.

These calculations suggest that the damage-limiting potential of [less than 1 line of source text not declassified] is subject to diminishing returns. In the absence of a nation-wide fallout protection program the U.S. fatalities are great even if urban-industrial areas are not directly involved. For all cases considered, the advantage (as measured in the reduction in fatalities) of second-strike [less than 1 line of source text not declassified] attacks when compared with attacks which do not target [less than 1 line of source text not declassified] is clearly shown. However, the return associated with th. [less than 1 line of source text not declassified] extra missiles proposed by the Secretary of the Air Force is small.

[Heading and 6 paragraphs (1-1/2 pages of source text) not declassified]

U.S. FATALITIES ASSUMING THE SOVIETS INITIATE THE ATTACK¹³

Alternative U.S. Forces	(Fatalities in Millions)
	Force A	Force I	Force II	Force III
Low Soviet Forces
A. Minimum Fallout Protection in U.S.a
1. U.S. cities hit immed.	165	138	138	138
2. U.S. cities attack delayed c	165	100	83	80
3. U.S. cities attack withheld	N/A	67	53	51
B. Augmented Fallout Protection in U.S.b
1. U.S. cities hit immed.	95	88	88	88
2. U.S. cities attack delayed c	95	66	55	52
3. U.S. cities attack withheld	N/A	25	17	16
Medium Soviet Forces
A. Minimum Fallout Protection in U.S.c
1. U.S. cities hit immed	195	183	183	183
2. U.S. cities attack delayedc	195	162	150	145
3. U.S. cities attack withheld	N/A	145	130	127
B. Augmented Fallout Protection in U.S.b1	1. U.S. cities hit immed.	120	110	110	110
2. U.S. cities attack delayed c	120	92	85	79
3. U.S. cities attack withheld	N/A	48	43	40

Force A: Forces only targeted against urban-industrial areas.

Force I: Includes [less than 1 line of source text not declassified] Minuteman, and achieves the target destruction capabilities against Soviet Bloc [less than 1 line of source text not declassified] as previously shown.

Force II: Includes [less than 1 line of source text not declassified] Minuteman, and achieves the target destruction capabilities as previously shown.

Force III: Includes [less than 1 line of source text not declassified] Minuteman, and achieves the target destruction capabilities previously shown.

a Fifty million stocked shelter spaces, but because of limited training for the population, no effective use of these spaces is made.

b The “Augmented” protection program assumed here for the U.S. is a nation-wide fallout protection at a cost of $4 billion and consists of 240 million fallout shelter spaces.

c The delay is assumed to be roughly 1 to 8 hours, long enough for our missile attack but not our bomber attack to arrive.

The following table compares four alternative U.S. forces. Force I is a force posture including [less than 1 line of source text not declassified]Minutemen. Force II is the recommended force and includes [less than 1 line of source text not declassified] Minutemen. Force III is the proposal of Secretary Zuckert and includes [less than 1 line of source text not declassified] Minutemen. Force IV includes the [less than 1 line of source text not declassified]Minutemen proposed by the Chief of Staff, Air Force.

US/WESTERN EUROPE FATALITIES UNDER U.S. INITIATION	And Soviet Counter-City Retaliationa	(In Millions)
U.S. Fatalities	Western Europe Fatalities
	I	II	III	IV	I	II	III	IV
	/(*)/	/(*)/	/(*)/	/(*)/	/(*)/	/(*)/	/(*)/	/(*)/
Low Soviet Postureb	50	40	32	28	90	75	65	60
Medium Soviet Posturec	95	75	65	58	100	90	85	75
High Soviet Postured	162	153	148	138	155	142	138	130

a All surviving Soviet forces, except their bomber force, are assumed targeted against cities. If only military targets are attacked, U.S. fatalities would be under 20 million providing that the U.S. has a nation-wide civil defense program. In the absence of a civil defense program U.S. fatalities could number between 70-90 million. Under first-strike circumstances U.S. reserve forces greatly outnumber those of the Soviet Union.

(*) Number of Minutemen not declassified.

b The results are based on the “low” Soviet force, Optimistic U.S. operational factors and nation-wide Civil Defense program.

c The results are based on the “medium” Soviet force, Expected U.S. operational factors and nation-wide Civil Defense program.

Only if the Soviets elect a minimum force posture would there be prospects of keeping U.S. fatalities at a relatively low level. But in this case, greater numbers of Minutemen beyond the level recommended do not substantially reduce U.S. fatalities. Moreover, the potential damage to Western Europe remains very high. For the two other Soviet postures U.S. and NATO fatalities are high, and additional Minutemen do not hold great promise in reducing these fatalities. For the Medium Soviet Posture, even if we were to augment Force IV with the Nike-X anti-missile system at a cost of $20 billion (assuming the system could be deployed around 23 cities by mid 1969) U.S. fatalities would, nevertheless number about 35 million provided the U.S. has a nationwide civil defense program. However, in the absence of a civil defense program and assuming the High Soviet Posture U.S. fatalities would number about 120 million.

It would seem almost unbelievable that the Soviets would not react if we started building Force IV augmented by Nike-X. What would be the prospects in such an arms race? Studies suggest that, assuming Soviet costs are similar to ours and that they disperse their hard ICBM’s two to a point, and assuming that we wish to assure the survival of 80 percent of our population (i.e., no more than 40 million dead) after a [less than 1 line of source text not declassified] the cost exchange rate is roughly three to one against us. That is, if we attempt to maintain a [less than 1 line of source text not declassified] (and the Soviets attempt to achieve a capability to kill at least 40 million Americans in retaliation) we must outspend them, at the margin, by three to one. And if we set our sights higher than 80 percent, the cost exchange rate becomes even more unfavorable.

What this and other analyses suggest quite clearly about the [less than 1 line of source text not declassified]objective is: first, that if we were to want to make the attempt to achieve such a capability, the most productive increments with respect to our current program would certainly be in civil defense and possibly in anti-missile defenses, not more ICBM’s; and second, that the attempt to achieve a [less than 1 line of source text not declassified]capability, under any reasonable definition of the term is, to the extent that anything is predictable in defense planning, bound to be defeated by diminishing marginal returns.

Other Reasons for [less than 1 line of source text not declassified]

There are other reasons for having strategic forces available for targeting against [less than 1 line of source text not declassified]. The list includes a capability for limited strategic nuclear attacks, forcing the Soviets to devote resources to [less than 1 line of source text not declassified] and placing significant constraints on the Soviet attack planners. These considerations are discussed in greater detail in Appendix II, pages 33-35.¹⁴

Conclusions

The advice of the Joint Chiefs of Staff, the various calculations and studies I have reviewed over the past year, and the analysis described in this memorandum have led me to the following conclusions:

1. The forces I am recommending are clearly adequate for the objective of “Assured Destruction” under any reasonable definition.

2. The prospects for “Damage Limiting” by counterforce attacks may not hold great promise in the latter part of the 1960’s if the Soviets harden and disperse their ICBM force and build up their missile submarine force as we now expect them to do. I believe that the recommended forces accomplish what might reasonably be able to be done from this point of view, and that the extra capability proposed by the Air Force would make a contribution to “damage limiting” too small to be justified in the light of its extra cost.

3. A [less than 1 line of source text not declassified] does not appear to be feasible during the time period under consideration with the weapon systems projected for both sides, unless the Soviets choose to buy strategic forces that are both smaller and less well protected than we now expect. In any case, [less than 1 line of source text not declassified] extra Minuteman missiles, as proposed by the Secretary and Chief of Staff of the Air Force, respectively, would not significantly improve the outcome of the war for us.

Attachment¹⁵

APPENDIX II: SOVIET DEFENSIVE POSTURE AND OTHER “ASSURED DESTRUCTION” AND “DAMAGE LIMITING” CONSIDERATIONS

[Here follows discussion of the Soviet defensive posture.]

Other “Assured Destruction” Excursions

In the memorandum the capability of U.S. forces to inflict high levels of damage on the Soviet Union was calculated on the basis of expected U.S. operational factors. In problems of long range defense planning, we always encounter substantial uncertainties about the factors that influence force requirements. Many of these uncertainties can be described by the range of likely possibilities. For purposes of planning and analysis, it is often useful to describe this range by three estimates: an “optimistic” estimate, an “expected” or most likely estimate, and a “pessimistic” estimate. The pessimistic estimate is the estimate least favorable to ourselves of those which are consistent with the available evidence. For example, we have so far achieved about a [less than 1 line of source text not declassified]success rate in Minuteman ICBM tests. Based on this and our experience with other missile programs, and assuming that we continue reliability testing and development, it is my judgment that a reliability of less than [less than 1 line of source text not declassified] under operational conditions by 1969 is effectively ruled out by the data. And that is our pessimistic estimate. In fact, by 1969 we expect to achieve 75 percent, and possibly as high as 85 per cent reliability under operational conditions.

The point here is that the relevant test is, “Is it consistent with the available evidence?” and not “Is it conceivable?” A 1969 reliability of [less than 1 line of source text not declassified] or less, is conceivable. But it is so unlikely that it can be ruled out of the range of practical possibility. This point requires emphasis because there is a mistaken tendency sometimes to test our forces against the most adverse circumstances conceivable.

The major uncertainties affecting the assured retaliation capabilities of our Strategic Nuclear Forces can be grouped under four headings:

1. Improved Soviet Defenses,

2. Larger or More Effective Soviet Offensive Forces,

3. Lower U.S. Missile Reliability, and

4. Unanticipated Wartime Degradation in U.S. Forces.

First, the Soviets may improve the protection of their cities and population beyond what we now expect. Our first calculation of expected damage was based on the assumption that the Soviet populace use whatever fallout protection is now available, but do not have a nation-wide fallout protection program. Moreover, although we expect the Soviets to deploy ABM defenses only at Leningrad and Moscow, they might deploy such a defense at as many as 10 or 15 cities. Such a defense would cost us from $12-15 billion. Assuming a nation-wide fallout protection program and ABM defenses for 15 cities, but not assuming Soviet offensive forces larger or more effective than anticipated, [3 lines of source text not declassified].

Second, the Soviets may prove to have larger and more effective strategic forces than those we now anticipate. The calculation of expected damage shown above was based on the assumption that the Soviets would have the “Medium” force of 560 ICBM’s by 1969. The high NIE estimate of the 1969 Soviet force is now 700.¹⁶ Moreover, there is uncertainty about Soviet missile accuracy and reliability. The foregoing calculations were based on the assumption of [less than 1 line of source text not declassified] missile reliability (forces peaked for attack) and a [less than 1 line of source text not declassified] CEP as suggested by our intelligence estimates. However, I believe it prudent to insure against a reliability as high as [less than 1 line of source text not declassified] and a CEP of as low as [less than 1 line of source text not declassified].¹⁷

Finally, there may be unanticipated wartime degradation in the performance of our forces. There are several possibilities. First, we have assumed that our warning systems work and that we successfully launch the B-52 Alert Force. However, our force will be concentrated on about [less than 1 line of source text not declassified]; the Soviets will have missile launching submarines that can attack our bomber bases with very little warning; there are ways in which they can reduce the warning we get from their ICBM’s (e.g., long-way around or low angle trajectories); and we may be slow in responding. We should consider the possibility of [1-1/2 lines of source text not declassified].

Another possibility is that, for one of several reasons, we might be unable to launch our ICBM’s before a Soviet follow-up bomber attack. One reason for this might be disruption of our high-level command or communications. Another might be that fear of the effects of gamma radiation on the guidance systems, if we expose the missiles soon after an ICBM attack, would cause us to want to hold the missiles in their silos until the radiation pulse has passed. Therefore, we should consider our retaliatory capability on the assumption that the ICBM’s must ride out the Soviet bomber attack. Yet another uncertainty we face is the precise blast resistance of our ICBM silos. [7 lines of source text not declassified]

The results shown in the memorandum combined all four groups of pessimistic assumptions.

Some other possible reasons for possessing counter-military forces are as follows.

Limited Strategic Nuclear Options

Besides the ones already mentioned, there are other reasons for having strategic forces available for targeting against Soviet military forces. One is to give us the possibility of executing limited strategic attacks. Admittedly our understanding of this range of possibilities is very limited and inadequate at this point in time. But it will doubtless become more important relative to other forms of thermonuclear war as both sides deploy secure well-protected nuclear delivery systems. What I have in mind here are controlled deliberate demonstrations of intent to escalate to strategic nuclear war, to back up ultimata, and to do so in such a way that the war might be able to be controlled and brought to a close.

Assume, for example, that a conflict over Berlin, or elsewhere in the NATO area, has expanded into a large scale non-nuclear war and that we are losing. Although I do not believe that our loss at the non-nuclear or small-scale tactical nuclear level is as inevitable as commonly supposed, it is clearly a possibility that must be considered. In these circumstances, it may be desirable to have, as an alternative, the possibility of a [1-1/2 lines of source text not declassified]. The recommended force clearly provides us enough missiles for such contingencies.

Forcing the Soviets to Harden

Another reason for a [3-1/2 lines of source text not declassified]. Forcing the Soviets to harden, or otherwise protect, their forces from U.S. missile attack has at least two advantages for us. The first is that it forces the Soviets to incur a very heavy cost whence, out of any given budget, they can afford fewer forces.

We do not have detailed studies of the cost of protecting our forces, but there are many indications that it is high. For example, we maintain [7 lines of source text not declassified].

[1 paragraph (13 lines of source text) not declassified]

These factors suggest that the program of protection of our Retaliatory Forces from nuclear attack has roughly doubled their cost. If the same factors apply for the Soviets, and we have no reason to suppose the contrary, forcing them to harden would halve the number of weapons they can deploy.

Moreover, we have found it very difficult to harden our large liquid fueled missiles to the point that we can have high confidence they will survive ground shocks and still operate. This is one of the factors that led us to Minuteman. A Soviet liquid-fueled missile capable of delivering a 100 MT warhead would have a gross weight of approximately 660,000 pounds. This represents a weight twice that of Titan II, and two and one-half that of Atlas F. While this Soviet missile could be hardened, assurance of making the silos withstand high overpressures could be difficult. By forcing them to harden, we may force them to go to smaller missiles with lower payloads.

Finally, forcing the Soviets to harden may be desirable from the point of view of creating a more stable posture, reducing their incentive as well as their ability to make a pre-emptive strike against us. However, in view of the fact that the recommended U.S. force provides us with between a [less than 1 line of source text not declassified]numerical superiority in ICBM’s alone [less than 1 line of source text not declassified], I do not believe that a further increment in our forces is required for this purpose.

Constraints on Soviet Attack Planner

Our possession of a [less than 1 line of source text not declassified] puts significant constraints on the Soviet attack planner. In the cost of their bomber operations, the Soviets can get substantially more out of their intercontinental bomber force if they can stage it through peripheral staging bases. But this is a very vulnerable operation. If they have to take seriously the possibility that [less than 1 line of source text not declassified]they must plan a far less vulnerable operation which would deliver fewer bombers to the United States. [5 lines of source text not declassified] However, it is clear that the recommended U.S. forces will provide enough missiles for this purpose.

0. Source: National Archives and Records Administration, RG 218, JCS Records, JMF 7000 (3 Jan 64) Sec 1A. Top Secret. This draft memorandum is Tab I of “Department of Defense Draft Memoranda for the President: Recommended FY 1965-1969 Defense Programs,” dated December 19. An earlier version of this draft memorandum, dated August 31, is ibid., JMF 7000 (31 Aug 63) Sec 1. Another, almost identical to that printed here, is marked “Final Draft” and is dated November 13. (Department of State, Central Files, DEF 1 US)
1. This memorandum was another tab of the December 19 compendium; see Document 152.
2. The quotation is from Appendix A to a November 7 memorandum from the JCS to McNamara. (National Archives and Records Administration, RG 218, JCS Records, JCS 1800/753-3, JMF 7000 (31 Aug 63) Sec 2)
3. [text not declassified]
4. Document 96.
5. Entitled “Basis for Recommendations on Particular Weapons Systems,” not printed.
6. The NIE of July 6, 1962, referred to in the table below is NIE 11-8-62, Document 96. The NIE of October 18 is NIE 11-8-63, Document 144.
7. Entitled “Strategic Targets in the Soviet Bloc, U.S. Force Laydown, and U.S. Operational Factors,” not printed.
8. These Designated Ground Zeros (DGZ’s) include [less than 1 line of source text not declassified]USSR military and government control centers in major urban areas, [less than 1 line of source text not declassified] and [less than 1 line of source text not declassified]defense suppression targets located in corridors on the way to the target cities. [Footnote in the source text.]
9. The Soviet surprise attack on the U. S. to which this is a response would have killed 195 million Americans in the absence of an effective nationwide fallout protection program, 120 million if there were a nationwide fallout protection program, in both cases assuming the Soviets immediately target cities. If the Soviets do not target cities, U. S. fatalities would, nevertheless, vary between 75 million (nationwide fallout protection program) and 180 million (no nationwide fallout protection program). U. S. fatalities in a nuclear war are discussed in more detail in the section on “Damage Limiting” below. [Footnote in the source text.]
10. The range of optimistic, expected, and pessimistic estimates for U. S. operational factors discussed in this section is tabulated in Appendix III, pages 38-39. [Footnote in the source text.]
11. Reference is to the first table above entitled “Recommended and Service-Proposed Forces.”
12. [Here follows a footnote in the source text that is a table with the costs of Minutemen forces for FY 1964 through FY 1969. The total for Force III for FY 1964-1969 was [text not declassified].]
13. The November 13 draft cited in the source note above gives figures for the four items marked N/A in this table. In order, they are 135, 49, 180, and 75.
14. The results are based on the “high” Soviet force, Pessimistic U.S. operational factors and no nation-wide Civil Defense program.
14. These pages are included in the portion of Appendix II that is printed below.
15. Top Secret. The portion of Appendix II printed here comprises pp. 31-35.
16. The Air Force dissent speaks of 750-1000 ICBM’s under certain conditions. Although this possibility cannot be altogether ruled out by the available evidence, if the Soviets were to embark on such a building program, we would find out about it in time to expand our own forces enough to offset it. We therefore do not need to buy insurance against that possibility now. [Footnote in the source text.]
17. It is also possible that the Soviets might have multiple guided re-entry vehicles for some of their ICBM’s by 1969, although there is no evidence that the Soviets have begun development on such a capability. Our own studies indicate that this would be costly and difficult, and not an efficient way to expend [expand?] our force. Therefore, I believe we can rule out this possibility for 1969. [Footnote in the source text.]