Foreign Relations of the United States, 1981–1988, Volume XLIV, Part 1, National Security Policy, 1985–1988

131. Memorandum From the Undersecretary of Defense for Policy (Ikle) to Secretary of Defense Weinberger¹

I–24902/86

Washington, August 29, 1986

SUBJECT

• Early Benefits from SDI—DECISION MEMORANDUM (U)

(C) We need to reexamine whether the SDI program could not yield some early benefits, consistent with the long term goals of SDI. Under the current SDI policy, such benefits would not be pursued unless and until they were subsumed under a formal decision to deploy a full SDI system. Thus, with the present program, the country might have to wait some 12 years—or more if slowed down by Congressional cuts—until the money invested in SDI began to show any return.

(C) Through all these years without any benefits, it could become increasingly harder to obtain Congressional support for a continually growing SDI budget. Even if future Presidents supported SDI as strongly as does President Reagan, the SDI budget would become an easy target, to be raided in behalf of DoD projects with more immediate benefits or which enjoy stronger institutional support among the services.

(C) The hurdle imposed by the ABM Treaty compounds the difficulty of promoting an increasingly expensive program with very distant benefits. Thus, we have become defensive about SDI by stressing that it is “purely a research program” in compliance with the ABM Treaty. This stance concedes to our critics the offense; they can point out—correctly—that the ultimate objectives of the research are contrary to the objectives of the Treaty. Instead, the argument should have remained focussed on the failure of the Treaty to advance its stated objectives, that is, to facilitate reductions in offensive arms.

(C) In addition, we have accepted unjustifiably difficult criteria for deciding whether or not the eventual results of this “research” program warrant a decision in favor of deployment:

— The defenses must be guaranteed in advance to be “cost-effective at the margin” against countermeasures and attacks we have only imagined the Soviets might develop. We require of no other weapons system that it be so resilient as to render all conceivable future Soviet countermeasures ineffective.

— The key test of defenses is not that they be effective against likely Soviet attacks, but against an irrational mass attack aimed at our cities. Never mind that the current US strategy also does not protect against such irrational attack. In essence, the test of SDI has been shifted from the President’s sensible and potentially attainable goal of making missiles “impotent and obsolete” to the unattainable goal of precluding, under all circumstances, any missile attack on a US city.

— We allow the opponents of SDI to postulate that the Soviets would bend every effort to defeat SDI, at any cost they could remotely afford. Yet, we allow the same people to argue the Soviets would benignly settle down on some refurbished SALT regime without trying to defeat our “retaliatory” deterrent forces. Thus, against our assured destruction forces the Soviets are supposed to remain placid (contrary to our experience over the last 15 years); however against our defensive shield they are supposed to embark on a new arms race.

(C) In considering early benefits of SDI we must guard against erring in the opposite direction: we must not settle for a deployment option that would deflect the thrust of the President’s program and lock us into a defense that merely protected our strategic forces. This would serve to buttress MAD instead of leading to the desired transformation of the strategic order. However, this risk in the opposite direction should not drive us to reject any and all early benefits. In fact, by failing to identify any early benefits, we make it easier for the opponents of SDI to fill the void and claim that the protection of strategic forces is the only early benefit, indeed the only foreseeable benefit of SDI.

(C) Several options for early benefits merit further consideration:

(1) The initial stages of a layered area ballistic missile defense (see Annex A). This deployment should not be seen as a final deployment configuration or as a point defense system.

— Such a defense would significantly enhance Soviet attack uncertainty against U.S. military assets and provide population protection against light attack or accidental launches. By demonstrating to the Soviets in the near term our commitment to see SDI through to its logical conclusion, such an initial deployment also could have an early and beneficial influence on future Soviet development and deployment decisions and approaches to arms control. Unlike other potential near term benefits, this would not be ABM Treaty compliant.

— Sensor elements of this initial deployment will augment and enhance current ballistic missile attack warning and assessment capabilities.

— Some features of this initial deployment can pose not only a deterrent threat to Soviet strikes against U.S. and allied satellites, but offer an elementary capability to defend satellites against current and some near term Soviet ASATs.

(2) Defense of NATO allies in Europe; components/technology derived from this initial deployment could have utility for enhancing air defenses and an ATBM system.

— The utility of such components would clearly be greater for European NATO than for North America. In fact, air defense deployment for North America would likely be made only in conjunction with a decision to deploy ballistic missile defenses; even then, fiscal realities and priorities for ballistic missile defense may dictate a pause before significant air defense deployments in North America are made.

(3) Other near term benefits could be identified that do not provide defense against missiles but other useful functions. The Boost Surveillance and Tracking System, if characterized as a follow-on to DSP for ballistic missile early warning, may be an example.

(C) Whatever near term benefits are identified and pursued, the elements of the initial deployment sequence should comply with some overarching general principles:

— They should be able to be incorporated effectively into an evolutionary ballistic missile defense architecture that ultimately achieves the President’s SDI goals;

— They should not delay benefits achieved by fulfilling the ultimate goal of SDI;

— They should be militarily useful (even if it were later decided not to proceed with deployment of a full SDI system). When not involving the deployment of ballistic missile defenses, they would satisfy other important military requirement.

— As more advanced elements are deployed, initial elements should be capable of performing important, but alternate roles and missions.

— Aside from their role in ballistic missile defense, they should satisfy or complement other important strategic military requirements.

(U) Jim Abrahamson, who has been out of town as we completed this memo, agrees with the general thrust here expressed. He and I propose that SDIO, in cooperation with ISP, prepare a briefing for you, for the second half of September, to flesh out the specific technical and cost aspects.

Briefing after Mid-September:²

Fred C. Ikle³

Annex A

Paper Prepared in the Department of Defense⁴

Washington, undated

INITIAL OPTION (U)

I. System Functions:

— Enhance ballistic missile attack warning and assessment to increase US confidence in attack characterization and response selection by substantially increasing the number and capability of ballistic missile surveillance assets.

— Establish an initial satellite defense capability to protect critical US assets against co-orbital and some direct ascent ASATs and create a global, real-time anti-satellite capability to deter attacks on US satellites with ERIS and the initial deployment of space-based kinetic kill vehicles.

— Deny Soviet confidence, disrupt attack timing, and restrict allocation flexibility for attacks on critical US military assets with the deployment of terminal, late mid-course and boost-phase interceptors.

— Protect critical military targets from upper tier leakage with high altitude endo-atmospheric interceptors.

— Defend NATO allies and US forces in Europe through the provision of late mid-course and terminal defense SDI technology and by the boost-phase interceptor deployment.

II. Estimated Date of Initial Deployed Capability:

FY92/FY93

III. Estimated Costs: (R & D and Investment)

Boost	$24B
Late Mid-Course (Area)	11B
Terminal (Point)	16B
BM/C3	5B
	$56B + O&S

IV. Description of System Elements in 1996:

— Upgraded ground-based early warning and ABM radars

• BMEWS
• PARCS
• PAVE PAWS
• Cobra Dane

— Upgraded ground-based spacetrack radars and electro-optical sensors

— Upgraded DSP to provide stereo track data

— [1½ lines not declassified]

— 1000 ground-based exo-atmospheric interceptors

— 500 ground-based endo-atmospheric interceptors

— [1½ lines not declassified]

1. Source: Reagan Library, Fred Ikle Files, Arms Control (President Gorbachev)—1986–1988. Confidential. A stamped notation indicates Weinberger saw the memorandum on September 2.
2. Weinberger indicated his approval of this recommendation on September 2.
3. Ikle signed the memorandum “Fred” above this typed signature.
4. Secret.