The Last Warrior

by Andrew F. Krepinevich

  For over a decade the SESC/NESC continued to provide annual studies of the net capabilities of the USSR “ . . . to inflict direct damage upon the United States and to provide a continual watch for changes that would significantly alter those net capabilities.”5 However, in December 1964 Defense Secretary McNamara advised President Lyndon Johnson that these studies had outlived their usefulness because they did not, in his opinion, provide “a basis for planning guidance.”6 In March 1965 the NESC was judged to have served its purpose and disbanded.7

  Within three years recommendations began to surface advocating the creation of a replacement for the NESC.8 In 1968 Lieutenant General (Ret.) Leon Johnson, who had directed the NESC from 1961 to 1964, recommended that an organization be established to provide the president with “across-the-Government analysis of our relative nuclear strategic strength with the Russians.” His proposal, however, fell on deaf ears.

  By this time there were two major incentives for reestablishing a government-wide net assessment capability. First, the USSR was approaching rough nuclear parity with the United States. Second, discounting US spending on operations associated with the Vietnam War, the Soviet military was beginning to outspend the Department of Defense. Hence the United States could no longer solve strategic problems in its military competition with the Soviet Union by simply throwing money at them as it had done in the past. Instead, careful net assessments were needed to understand exactly where the United States stood in various areas of the competition, including whether the USSR was not only outspending the Pentagon but also more efficient in converting resources into military capability.

  By the fall of 1969 Andrew Marshall was in his twenty-first year at RAND and the United States was entering the fifth year of major combat operations in Vietnam. American forces in South Vietnam now numbered over 500,000. The previous year’s Tet offensive, launched by the Communists, had proven a military failure but a strategic success. While the offensive had decimated communist forces in South Vietnam, it had turned US public opinion against the war.

  FIGURE 4.1A.US–USSR Strategic-Nuclear Force Ratios.

  SOURCE: Natural Resources Defense Council (NRDC), “Archive of Nuclear Data from NRDC’s Nuclear Program,” http://www.nrdc.org/nuclear/nudb/datainx.asp

  FIGURE 4.1B.US and USSR Military Spending.

  SOURCE: Donald H. Rumsfeld, “Annual Defense Department Report FY 1978,” January 17, 1977, p. 3.

  Nor was the situation in Europe, the Cold War’s front line, all that encouraging. The previous year had seen the fragile blossoms of political liberalization in Czechoslovakia known as the Prague Spring brutally stamped out by an invasion spearheaded by Soviet Army troops. Only months later West Germany’s chancellor, Willie Brandt, increased US anxieties further when he advocated a Neue Ostpolitik (new eastern policy) that sought a rapprochement with the Soviet Union and its Warsaw Pact allies.

  In the Middle East Egypt and Israel were engaged in a low-level conflict following the latter’s spectacular victory in the June 1967 Six-Day War. The subsequent period of hostilities between Israel and Egypt, known as the War of Attrition, had been proclaimed by Egyptian leader Gamal Abdel Nasser two months after Nixon’s inauguration. Periodic artillery duels, commando raids, air strikes, and brief small-scale incursions became common as both sides faced off along the Suez Canal. Meanwhile Egypt and Syria were being rearmed by the Soviet Union, which also sent military personnel to fly combat missions, man surface-to-air antiaircraft missile batteries, and engage in other military activities.

  As the United States began searching for ways to end the Vietnam War the Soviet Union continued its broad military buildup. At home the American public evinced growing skepticism and in some quarters open hostility to pursuing the activist role in the world advocated by President Kennedy when he had declared in 1961 that the United States would “pay any price, bear any burden, meet any hardship, support any friend, oppose any foe, to assure the survival and the success of liberty.”9 Richard Nixon had been elected president in November 1968 due in part to his assurance that he had a plan to end the Vietnam War on terms acceptable to the United States.

  Confronted with an array of security challenges, President Nixon saw the need for a shift in US strategy. As was the case with President Eisenhower, under whom Nixon had served as vice president, he was prepared to be personally and persistently involved in crafting a new strategy. Among other things this would require good intelligence regarding adversaries’ intentions and capabilities—most of all those of the Soviet Union. In his first months in office, however, the new president found the products provided by the US foreign intelligence community sorely lacking. Henry Kissinger, a Harvard professor Nixon had selected as his national security adviser, shared his views on the shortcomings of US intelligence products. Both agreed something needed to be done to address the problem, and soon.

  In September 1969 Marshall and his wife, Mary, were preparing to set off on an extended vacation in Europe. Since their time in France during the mid-1960s when Marshall was at NATO headquarters, they had periodically returned, spending a month or so touring the country, sampling the French cuisine, and enjoying a break from day-to-day life back home. The couple’s planning was abruptly interrupted by a call from Kissinger. Could Marshall stop in Washington and meet with him on his way to Europe? He agreed.

  When Marshall arrived for their meeting, Kissinger told him of Nixon’s frustration with the intelligence reports coming into the White House, recounting how Nixon’s attorney general, John Mitchell, had read some of the National Intelligence Estimates (NIEs) provided by the Central Intelligence Agency and thought them lacking in logic, substance, and even in grammar and composition. Kissinger had his own complaints. One area of analysis that he personally valued was personality profiles of foreign leaders and key officials that could help him develop his negotiating tactics with the Soviets over strategic nuclear arms. But the CIA was proving little in the way of this sort of intelligence. Kissinger also judged the intelligence products being sent to the White House to be even worse than what one could find in the national press, describing them to Marshall as “lousy.”10

  Having summarized the new administration’s collective dissatisfaction with US foreign intelligence, Kissinger asked Marshall whether he would be willing to come to Washington to address the issue. Marshall told Kissinger he would get back to him with an answer shortly. He wanted to broach the subject of spending time in Washington with Mary. She was not keen on the idea of leaving their friends and home in California. Mary finally relented, however; a few months in Washington were something she could endure. So Marshall accepted Kissinger’s offer with the proviso that he would begin working on the intelligence issues after he and Mary returned from their planned vacation. Kissinger agreed.

  Despite this arrangement with Kissinger, the Marshalls’ time in France was soon punctuated by phone calls from Kissinger’s military assistant, an Army brigadier general named Alexander Haig, asking why Marshall had not arrived at the NSC. Marshall took Haig’s hectoring in stride and, after returning to the United States, calmly went about wrapping up some obligations at RAND. At last, in early December 1969, he appeared at the NSC. Marshall’s first task was to provide Kissinger with an assessment of the flow and quality of intelligence coming into the White House, a project the two men agreed would take only a couple of months.

  Marshall began his assessment by going over to the White House Situation Room, the focal point for the intelligence coming in for the president and Kissinger. There he interviewed people representing the various US intelligence organizations, including the CIA, NSA, the Defense Intelligence Agency (DIA), and the intelligence arms of the military services. Where did their intelligence come from, Marshall wanted to know, and how did intelligence officials decide what was important enough to send to the White House?

  Marshall found that there was no uniform approach. Among the agencies, only the NSA seemed to have a logical method for decid
ing what to send to the president. Every few months the NSA leadership would meet with its representative in the Situation Room to review current intelligence priorities, adjusting the topics as they deemed appropriate to changing world conditions and whatever seemed of interest to the president and Kissinger at the time. The NSA leadership would then inform its staff to prioritize their efforts to obtain intelligence regarding these topics.

  The CIA presented a startling contrast to the NSA. The Agency (as it was often called) made little effort to ascertain what topics might be of most interest to the president. Marshall was told that the CIA tracked issues by using stories run in the New York Times as the gauge for what was important and what was not. Given this sophomoric approach, and his discussions with the CIA representatives, Marshall concluded that the Agency’s senior managers made little or no effort to ensure that their products addressed Nixon’s interests and needs. Indeed, they seemed almost hostile toward the president.

  Marshall next began examining the CIA’s premier intelligence product, the President’s Daily Report. As the name suggests, it was dispatched to the White House the first thing every morning. He gathered up the reports that had been provided during Nixon’s first six months in office, sat down, and began reading through them. He knew that the president had a strong habit of making marginal notes on everything he read. As Marshall worked his way through the reports he noticed that the president’s notes became fewer and fewer until, finally, there were none. Nixon, Marshall realized, had stopped reading the reports.

  Marshall discovered that Nixon’s disinterest coincided with the introduction of a competing daily analysis generated by the National Security Council. These analyses were drawn from multiple intelligence sources, and there was only about a 60 percent overlap between what the CIA was providing and what the NSC staff was ginning up on its own. The NSC staff also provided the president with intelligence on a wider array of issues. Most important, Marshall found these reports bore Nixon’s margin notes.

  Marshall went back to the CIA. Why, he asked, did the Agency persist in sending Nixon reports he clearly did not read? Why did the Agency not even attempt to adapt its reports to fit the president’s needs? His queries ran into a bureaucratic stone wall. Marshall came away with the impression that the Agency’s managers believed their expertise should dictate what Nixon needed to know; hence, they were providing him with only the intelligence they thought he should be interested in. And despite the clear evidence that Nixon was not reading the President’s Daily Reports, the Agency continued sending them. The report was the CIA’s premier product. It would be developed as the Agency saw fit.11

  Marshall also took a look at another key Agency product, the National Intelligence Estimates. The NIEs provided intelligence on particular issues and were the product of research and analysis by the National Intelligence Council (NIC). Prior to 1973 the Office of National Estimates (ONE), formed in 1950, had produced these estimates. The office originally drew upon a wide range of talent, to include intelligence officers but also “outsiders” such as former military personnel, ambassadors, and members of the business world and academia. By 1969, however, the CIA’s institutional antibodies had progressively reduced the number of outsiders in favor of the Agency’s career intelligence officers. Moreover, the NIC had become a career dead end, populated mostly by people awaiting retirement. The office had lost the objectivity and responsiveness that it had had in 1950s and early 1960s, and Marshall discovered the quality of its analyses had declined as well, just as Mitchell had said.12

  As his review continued, Marshall sensed an enormous gulf separating the CIA from the administration’s senior policy makers.13 Apart from the bureaucratic and institutional priorities undermining the intelligence community’s ability to provide the White House with valued information, the Agency seemed to have a very different view of the world and of the major forces shaping that world than did the senior policy makers it was supporting. Nixon and Kissinger thought strategically, on a global scale, with both the near- and (especially) longer-term consequences of their decisions in mind. Marshall felt that Nixon could see both the dangers and, just as important, the opportunities residing in crises and their relationships to America’s long-term security. He found, however, that members of the intelligence community typically focused only on the prospective dangers arising out of a crisis. They also seemed to infuse personal beliefs into what was supposed to be a rigorous analysis. Marshall’s research suggested that if a situation arose in which Americans were being killed, the Agency’s analyses tended to argue the crisis needed to be resolved as soon as possible, even at the expense of sacrificing longer-term US vital interests. Whereas people like Nixon and Kissinger emphasized the long-term consequences that would flow from their decisions, the Agency appeared to conduct its analyses as though there were no long-term consequences.

  As Marshall reflected on the situation, he recalled his experience in chess. The situation was a little bit like the game of Kriegspiel he had played during his early days at RAND, only with a major variation. In lieu of the referee in Kriegspiel who alone has knowledge of the location of both players’ pieces, Nixon and Kissinger had to rely on the CIA for information regarding their opponent’s position and activities. But the Agency was failing to provide the information on the Soviets that Nixon and Kissinger most valued and needed to know. The Agency didn’t seem to understand the rules as Nixon and Kissinger saw them, or understand what the president and his security adviser were trying to accomplish.14

  After assessing the quality and flow of intelligence into the White House for six months, in May 1970 Marshall submitted his report to Kissinger. Consistent with his work at RAND, his paper was long on descriptions of the problems in the intelligence community and short on recommendations. On the positive side of the ledger, as far as Marshall could tell the CIA was not actually withholding information from the White House. He also found that its analysis of hard data and factual reporting on Soviet forces—their numbers, general locations, and types of equipment—was good. Where he judged the intelligence forwarded to the White House to be poor was in the Agency’s use of overly simplified assumptions about Soviet behavior, including frequent reliance on a “model of the Soviet government as a single unified actor pursuing an easily stated strategy.”15 This criticism echoed Marshall and Loftus’s earlier concerns about the tendency of most RAND analysts to assume Soviet behavior could be understood as stemming from the actions of a unitary rational actor. Overreliance on this model of Soviet decision-making undermined the intelligence community’s ability to understand the motives behind past and recent Soviet decisions and use that understanding to better estimate the USSR’s future strategy and overall military posture.

  Marshall’s report did offer a few modest suggestions, including the need for clearer communication of White House needs to the intelligence community, and the desirability of encouraging the CIA to generate different intelligence products to address those needs, particularly as they related to decisions with long-term consequences.16 Yet Marshall noted that the problems were so fundamental—and the political and bureaucratic resistance to taking them on so formidable—that the odds of effecting the necessary reforms were low.17

  Although Marshall had completed the main task that had brought him to Washington, leaving the NSC and returning to California proved more difficult than he anticipated. Kissinger now wanted him to conduct a follow-on study on the USSR’s growing ICBM forces. They settled on doing an in-depth analysis of the Soviet Union’s new heavy R-36 ICBM (NATO code name the SS-9 Scarp) to illustrate the sort of intelligence assessments that the president and his security adviser desired from the Agency. Kissinger’s underlying motivation for choosing this topic stemmed from his and Nixon’s hopes of reaching arms-control agreements that would constrain the USSR’s buildup of its strategic nuclear forces. So for the rest of 1970 and well into the following year Marshall found himself in Washington, supervising a CIA effort to
lay out the design and bureaucratic history of the SS-9 and its antecedents.

  US efforts to negotiate an arms agreement on nuclear forces between Washington and Moscow had begun over two years before Marshall arrived at the NSC. In 1967 President Lyndon Johnson had suggested the two superpowers engage in negotiations to limit their nuclear forces. Although both sides had agreed to talks in the summer of 1968, full-scale negotiations did not begin until November 1969, after Johnson had left office. President Nixon’s point men for the negotiations, called the Strategic Arms Limitation Talks (SALT), were Kissinger, based in Washington, and the head of the Arms Control and Disarmament Agency (ACDA), Gerard Smith, who met with Soviet negotiators in Helsinki, Finland. Progress was slow, so slow that by the fall of 1970 Nixon and Kissinger had become concerned that the Soviets might be using the talks primarily as a way of providing cover for the USSR’s continued nuclear buildup while giving ammunition to US critics of the Nixon administration’s efforts to modernize America’s nuclear forces.