The Last Warrior

by Andrew F. Krepinevich

  While the 1983 SecDef/DCI strategic forces assessment remains classified, most of the NIEs from this period have been released and they offer enough insight to confirm the basic conclusions Marshall had reached about the US-USSR strategic balance as far back as 1976. The February 1983 NIE 11-3/8-82 contained the following assessment of Soviet intentions regarding nuclear conflict: “The Soviets believe that in the present U.S.-Soviet strategic relationship each side possesses strategic nuclear capabilities that could devastate the other after absorbing an attack. Soviet leaders have stated that nuclear war with the United States would be a catastrophe that must be avoided if possible and that they do not regard such a conflict as inevitable.”22 As for the Soviet view of how deterrence might fail Marshall’s view agreed with Moscow. A major US-USSR nuclear conflict would most likely arise out of a NATO-Warsaw Pact conventional conflict preceded by a political crisis in Europe. As for US intentions, NIE 11-3/8-83 found that the “Soviets see little likelihood that the United States would initiate a surprise nuclear attack from a normal peacetime posture; we believe it is unlikely that the Soviets would mount such an attack themselves.”23

  Given Marshall’s long-standing emphasis on the importance of Soviet assessments and calculations, these insights suggested that the United States’ nuclear posture in the early 1980s was probably adequate to deter a Soviet nuclear attack on the continental United States as well as a Soviet-led conventional assault on Western Europe by the Warsaw Pact. Marshall himself had reached similar conclusions in the late 1970s. His view of the strategic nuclear balance reflected the judgment that the Soviets evaluated their own position more pessimistically than did the United States, and would therefore lack confidence in being able to achieve their goals in either a general nuclear exchange or a NATO–Warsaw Pact conflict that escalated to nuclear use.

  That being said, the 1983 and 1984 versions of NIE 11-3/8 both went on to find that the Soviets were making every effort to shift the correlation of strategic nuclear forces more in their favor. This by no means indicated that the Soviet leadership wanted to start a nuclear war with the United States. More likely it reflected the fact that organizations with the USSR’s defense structure were working on their own agendas, which may not have conformed to the leadership’s priorities. Or perhaps Soviet leaders felt, as Marshall suspected, that if they enjoyed a perceived lead in the strategic nuclear competition, it could be leveraged in other aspects of the competition such as in raising doubts among America’s allies regarding the value of US nuclear guarantees. The lack of any Soviet inclination to risk a nuclear war was later corroborated in interviews with Soviet officials such Marshal Sergei Akhromeyev, who from 1984 to 1988 headed the Soviet General Staff. Akhromeyev stated in 1991 that “At no time did the USSR ever intend to make first use of nuclear weapons. In a military sense, the side that attacked preemptively would win, but in practical terms neither side would win. Even to the General Staff it was clear that nuclear weapons were not really military weapons but were political tools.”24

  The SecDef/DCI assessment offered Marshall an opportunity to interact with Weinberger. As Marshall knew at the time, three other areas of US advantage could not be included in the assessment due to their classification. One was the B-2 bomber. In October 1981 the Air Force had decided to develop and build a fleet of B-2s, able to penetrate with high confidence even the most advanced Soviet air defenses.25 A second advantage was the Navy’s edge in submarine detection and quieting. For example, in the late 1970s a Navy nuclear attack submarine had been able to track a Soviet Navaga-class ballistic missile submarine (NATO code-name “Yankee”) for weeks without the US sub’s presence being detected by the Soviet submarine’s crew.26 These two areas, as well as a third that remains sensitive, were so highly classified that even members of Marshall’s staff were not given access to them. Marshall’s only recourse was to meet directly with Weinberger and apprise him of them verbally. He told the defense secretary that these three US advantages made the nuclear balance “much more favorable” than the two-volume SecDef/DCI assessment indicated.27

  After the joint assessment was delivered, Marshall and Rowen discussed whether to begin working together on a second assessment. In light of the heavy burden the effort had imposed on the CIA analysts, Rowen demurred. Neither Casey nor Weinberger pursued the matter. Thus ended the Reagan administration’s efforts to have net assessments done jointly between the DoD and the CIA. From then on, for the rest of Reagan’s presidency, net assessments defaulted to what they had been before under Schlesinger, Rumsfeld, and Brown: a purely DoD enterprise conducted by ONA.

  Again, the long-standing impression of most who worked in ONA during the Weinberger period, including Marshall himself, was that the secretary of defense paid little attention to the office. The irony about this impression is that Weinberger and Carlucci were initially concerned about strategy, if only to counter outside critics who argued that the new administration had no guiding strategy for allocating the additional defense dollars. After meeting with Carlucci on this issue, Marshall recommended the creation of a small core group of five or six people split between military officers and outsiders reporting directly to Carlucci.28 Weinberger approved Marshall’s suggestion and created the Strategic Concepts Development Center (SCDC) at the National Defense University in 1981, appointing Phill Karber as the center’s first director. In this capacity Karber reported directly to Weinberger and the chairman of the Joint Chiefs of Staff, and enjoyed “all the access one could possibly want” to the defense secretary.29

  Having been close to Marshall since the early 1970s, Karber had had considerable exposure to net assessment. During Karber’s two years as Weinberger’s strategy adviser, Marshall and Karber discussed ONA’s strategic and European balances as well as Marshall’s thinking about competitive strategies. As head of SCDC, Karber was able to provide a back channel for conveying some of these assessments and ideas to Weinberger.

  SCDC’s most sensitive project was a war game conducted during two weeks in June 1983. Called Proud Prophet, the game’s scenario included a year-long rolling mobilization of NATO and Warsaw Pact forces, followed by a conflict that started in the Mideast and spread to Europe. The game saw intense nuclear counterforce attacks—US strikes against Soviet nuclear forces—the result of NATO’s doctrine that called for a deliberate escalation of a conventional conflict between NATO and the Warsaw Pact—as well as large-scale Soviet nuclear strikes in Europe. The exercise ultimately ended in a massive strategic exchange, despite both sides desperately trying to avoid general nuclear war. The actions of two states committing mutual suicide despite the overwhelming desire not to would not have surprised Roberta Wohlstetter or Graham Allison. Nor did they surprise Marshall.

  The participation of Weinberger and Joint Chiefs of Staff chairman General John Vessey in the exercise was concealed from virtually all of the more than three hundred participants except for Karber, who ran Proud Prophet 83, and members of his small SCDC staff. In the game, Weinberger consciously tried to execute the existing US military strategy, something that no previous defense secretary had been willing to risk testing even in a war game. In the end the game went, in the argot of the time, “nuclear big time” because Weinberger and Vessey “faithfully implemented the prevailing US strategy,” which called for horizontal escalation and employing limited nuclear strikes.30

  The game culminated with the “Blue” or US/NATO side’s decision to execute the full Single Integrated Operational Plan (or SIOP)—the Pentagon’s targeting plan for general nuclear war. This decision was personally made by Weinberger playing the role of the US president.31 The unprecedented direct participation of the secretary and the chairman in the game was deemed so sensitive by Weinberger than he directed that those aware of his and Vessey’s participation not to reveal it for twenty-five years. Thus, it was not until 2008 that Karber was able to talk openly about what had occurred behind the scenes during Proud Prophet 83.

  It is not difficu
lt to see why Weinberger was so sensitive about the war game. Karber’s postmortem of the exercise for the secretary and chairman reviewed how the “Red” (Soviet–Warsaw Pact) team had countered each of the “Blue” (US-NATO) team’s escalatory options. Reportedly, Weinberger then turned to Vessey and said, “Our strategy is bankrupt”—a judgment that could have gravely undermined US national security had it appeared, for example, in the Washington Post or the New York Times.32 Weinberger’s recognition of the defects in key elements of US defense strategy had considerable impact. SCDC’s Paul Bracken, now a professor at Yale University, participated as a neutral observer able to talk to both sides in the game. Bracken maintains that Proud Prophet 83 led Secretary Weinberger and General Vessey to undertake major changes in US strategy, including banishing from Pentagon war planning such strategies as launch on warning, conventional horizontal escalation, the early use of nuclear weapons, and tit-for-tat nuclear exchanges.33 As for ONA’s influence on Proud Prophet 83, Karber maintains that Marshall was primarily responsible for the issues explored in the game.34

  What troubled Weinberger the most about Proud Prophet 83 was that it showed US Cold War defense strategy, if executed, could not prevent escalation in conventional, much less nuclear, conflicts with the Soviet Union. War between the two superpowers incurred a high risk of leading inexorably to a nuclear Armageddon. Weinberger was hardly the first to reach this conclusion. Still, the strategic problem remained. Weinberger was particularly concerned about the European balance, given that US General Bernard Rogers, the Supreme Allied Commander, Europe, stated in the fall of 1983 that if NATO was attacked conventionally, he could only sustain a defense “for a relatively short time,” after which he would have to ask for authorization to use nuclear weapons.35

  To Weinberger this meant that to avoid crossing the nuclear threshold in the event of a European conflict, the conventional capabilities of the United States and NATO had to be strengthened. This entailed, in the defense secretary’s mind, sustained increases in US and allied defense spending. In FY 1981 the Carter administration’s last defense budget, DoD’s total obligational authority (TOA)* was $175.5 billion (in current dollars). By FY 1985, DoD’s TOA had grown to $276.2 billion, a 57 percent increase.36 Spending on strategic nuclear forces had more than doubled, from $12 billion to $26 billion; and the $47 billion added to the conventional, or “general purpose forces” budget represented a 70 percent increase37—in all, an unprecedented peacetime commitment to strengthening the country’s military.

  Weinberger’s concerns over US military strategy persisted into Reagan’s second term. The defense secretary continued looking for better ways to prevent the United States from having to resort to nuclear war in contingencies such as a Soviet invasion of Western Europe. Unaware of his concerns, critics of the administration accused Weinberger not only of having no strategy for the Reagan defense buildup, but also of seeking a capability to fight a nuclear war.38 The latter charge, he later wrote, was “nonsense,” declaring, “No one who has received as many briefings on nuclear weapons or participated in crisis exercises as I have could hold any doubts about the absolute necessity of avoiding nuclear war.”39 But how, especially in the wake of Proud Prophet 83, was nuclear war to be avoided?

  This time Weinberger turned to Graham Allison, then dean of Harvard’s Kennedy School of Government, for help in answering this question as well as in countering public criticism that the DoD’s defense strategy could be encapsulated in one word: “More.”40 In mid-1985 Allison began dividing his time between Harvard and the Pentagon as Weinberger’s special assistant on strategy. Given an office near Weinberger’s, Allison began spending two or three days a week in Washington in the summer and one day a week after classes resumed in the fall.

  Marshall had known Allison for two decades. They had been so close that when Allison married in 1967, the Marshalls had hosted a reception for the newlyweds. And of course Marshall had been a mentor to Allison when the latter served as a junior member of the May Group. When Allison arrived at the Pentagon it was natural for him to explore whether Marshall could help with the strategy issues that concerned Weinberger. Marshall immediately offered the idea he had suggested to Iklé four years earlier: cost-imposing or “competitive” strategies.41

  Weinberger quickly embraced the idea. In his February 1986 annual report to Congress he announced that competitive strategies would be a major DoD theme for the rest of the Reagan administration.42 And it was. When Carlucci succeeded Weinberger as defense secretary in November 1987, he continued the effort to institutionalize competitive strategies in the Department of Defense.43

  The competitive strategies concept was not complicated. The idea had grown out of a larger set of Marshall’s ideas about the long-term competition between the United States and the Soviet Union. The pertinent question became: What sorts of strategies should the United States adopt to be a more effective competitor in the continuing peacetime competition?44 Marshall’s answer, going back to 1969, when he had replaced James Schlesinger as RAND’s director of strategic studies, was to develop strategies that could capitalize on the United States’ enduring strengths (those easiest to sustain over time) while exploiting the enduring weaknesses and vulnerabilities of the USSR (those most difficult for the Soviets to overcome).45

  Like Marshall, Weinberger saw competitive strategies as a way to compel the Soviets to perform less efficiently or effectively by (1) moving the competition into areas where the USSR would have to expend far more resources than the US to remain competitive; and (2) creating conditions whereby the USSR would be incentivized to invest disproportionately in less threatening capabilities.46 Marshall had, of course, been urging the adoption of such strategies for at least a decade, as in the case of his arguments for going ahead with the B-1 bomber program.

  Weinberger created three organizations to implement the competitive strategies initiative. At the top was a Competitive Strategies Council, chaired by Weinberger himself, which also included the chairman of the Joint Chiefs of Staff, the service secretaries and chiefs, the undersecretaries of defense, the directors of the National Security Agency and the Defense Intelligence Agency, and the assistant secretary of defense for PA&E. He assigned Allison chairmanship of the Competitive Strategies Steering Group,47 which included Marshall, the assistant defense secretaries for policy and international security affairs along with representatives of the undersecretary for acquisition, the service secretaries and chiefs, and the DIA director. For day-to-day management of the initiative, a competitive strategies office, headed by a colonel, was established within OSD in June 1987.48 Subsequently, two intradepartmental task forces were established. Task Force I explored competitive opportunities in a mid- to high-intensity European conventional conflict. In large measure Task Force I recommended expediting what the Soviets termed reconnaissance-strike complexes based on “smart” conventional munitions, wide-area sensors, and battle networks.49 Task Force II was asked to explore nonnuclear strategic capabilities based on precision conventional munitions and long-range systems.50 However, its recommendations proved much too general and abstract to be given serious consideration for action. Both lines of development would capitalize on US strengths but neither would be brought to fruition before the Cold War ended.

  As sound as the thinking behind Weinberger’s competitive strategies initiative was, its implementation proved difficult. Within the Pentagon, the military services and the Joint Chiefs of Staff began quietly implementing a “paralysis by analysis” approach to competitive strategies to frustrate Weinberger from taking programmatic actions that might disrupt the their own budget priorities. In November 1988, not long after Weinberger’s departure, they succeeded in getting deputy defense secretary William Taft to sign a memo to the steering group chairman that effectively derailed the competitive strategies initiative.51

  The issue that prompted this memo was whether the War Game Committee charged with evaluating specific acquisition programs to implem
ent competitive strategies would be permitted to use special-access programs (SAPs, highly classified programs also known as “black” programs) for its computer-based modeling and simulation analysis. The military services opposed this, and Taft’s memorandum placed a hold on allowing SAPs in competitive strategies work for the remainder of the Reagan administration. This rendered the War Game Committee’s programmatic recommendations too general and abstract to be actionable. Despite this outcome, competitive strategies was by no means the only initiative during the Reagan years that caused great concern in Moscow by threatening to exploit US technological advantages to move the US-Soviet military balance in the United States’ favor.

  Reagan’s Strategic Defense Initiative (SDI) also put pressure on the Soviets, and Marshall’s analysis of this program made a stronger case for the value of pursuing ballistic missile defenses than did the arguments of those who were actually charged with doing so. His ability to conduct this contrarian assessment was only possible because ONA had substantial study funds for independent research. In SDI’s early days many observers, including both critics and proponents, presumed that to be successful any missile shield would have to be able to stop almost all incoming Soviet warheads, as even a handful of thermonuclear weapons falling on US cities would inflict catastrophic damage. Short of meeting that criterion, the conventional wisdom held that SDI represented an expensive road to nowhere.