The Last Warrior

by Andrew F. Krepinevich

  The commission’s report, Discriminate Deterrence, appeared in January 1988. The more detailed working group reports followed in October. From Marshall’s perspective, his work on the future security environment with Wolf raised two fundamental issues. First, their working group concluded that the Soviets were correct in their assessment that the advent of new technologies would revolutionize the conduct of war, not merely make current forces marginally more effective.87 Their finding made its way into Discriminate Deterrence, which concluded that the “further exploitation of microelectronics, in particular for sensors and information processing,” combined with low-observable (stealth) technology, extremely accurate conventional weapons, and improved means of locating targets, would revolutionize war.88 The implications of this conclusion became a central focus of ONA’s work during much of the 1990s, after the Soviet Union had collapsed.

  Second, in 1988 the USSR was still seen as the United States’ major competitor going forward, and the long-standing habit of seeing the future through the lens of that rivalry had the effect of pulling the commission members toward a short-term focus. Most refused, as Marshall later noted, “to admit publicly that the Soviet had had it.”89 To his frustration, Iklé was unwilling to state in Discriminate Deterrence just how weak Marshall believed the Soviet Union had become.90

  Notwithstanding the apparent distance between Marshall and Weinberger, and despite Iklé’s desire to have ONA focus on projects other than net assessments, work on a number of the Cold War balances continued during the 1980s. Two years after the joint DoD-CIA net assessment of US and Soviet strategic nuclear forces, ONA analyst Dmitry Ponomareff updated the executive summary of this balance.91 It was ONA’s third and final look at the US-Soviet competition in intercontinental nuclear forces. In 1984 Marshall forwarded to Weinberger three more assessments: the US-Soviet maritime and power-projection balances, and an East Asian assessment undertaken by Stephen Rosen. All were fairly lengthy reports, and thus not the sorts of documents that Weinberger was inclined to read from front to back, if at all, which the absence of any feedback seemed to confirm to Marshall. A similar fate awaited the NATO–Warsaw Pact military balance assessment that was forwarded to Weinberger in 1986. This was the last major assessment ONA completed during the Cold War.

  While Marshall could not compel an incumbent defense secretary to pay attention to ONA’s assessments, his thinking on various areas of military competition nevertheless found other, less direct ways of entering the public debate on national security. In particular, Marshall’s “hidden hand” approach could be found in his mentoring of promising young scholars and analysts. At RAND Marshall had been a mentor to researchers such as Roberta Wohlstetter and Graham Allison. Once in the Pentagon he continued to mentor both members of his staff and others through ONA’s contract research. Over time his gentle guidance produced a cadre of individuals who understood enough about net assessment to appreciate its value and endeavor to apply this analytic discipline in their own work on defense issues.

  Inside ONA these individuals came to be loosely referred to as members of St. Andrew’s Prep. Years later, when asked about his most important contribution as ONA’s director, Marshall declared his major achievement to have been the influence he had on the people who had served on his staff. Because the majority of ONA’s net assessments remain classified, the work outside ONA by some members of St. Andrew’s Prep whom Marshall mentored provides perhaps the best window currently available into how net assessment worked in practice. Their work also reveals how, through them, Marshall was able to influence the debate over US strategy even when Weinberger was apparently paying scant attention to ONA’s balances.

  Three of the more distinguished members of St. Andrew’s Prep were brought to his attention in the late 1970s by Sam Huntington. The three—Eliot Cohen, Aaron Friedberg, and Steve Rosen—were among Huntington’s best students at Harvard. Their writings are particularly useful in understanding Marshall’s thinking and on how he viewed military competitions since each of them spent time in the Office of Net Assessment working under Marshall.

  Rosen had graduated from Harvard in 1974, and not long after earned his PhD at that university. Early in his career Rosen was hired by Marshall as a civilian assistant. After a few years he left ONA to become the director of political-military affairs on President Reagan’s NSC staff and later served on the faculty at the Naval War College in Newport, Rhode Island. Whatever position Rosen held, he was never far from Marshall or his fellow members of St. Andrew’s Prep. He participated with Marshall in the CILTS work on the future security environment in the late 1980s and, along with fellow ONA alumni Eliot Cohen and Barry Watts, on the secretary of the Air Force’s Gulf War Air Power Survey following the 1991 Persian Gulf War. Rosen later earned tenure at Harvard as the Beton Michael Kaneb Professor of National Security and Military Affairs. During his summers he often led one of the panels of experts exploring an issue of particular interest to Marshall at one of ONA’s two-week “summer studies” at the Naval War College. In the early 1990s his book Winning the Next War would play an influential role in ONA’s effort to assess what Marshall suspected was a discontinuous shift in the character of warfare and, correspondingly, the need for military innovation in preparing for it.

  Aaron Friedberg had trailed Rosen at Harvard by four years, graduating in 1978. Like Rosen, he then earned a PhD from Harvard. Friedberg first met Marshall in April 1979 when working as a twenty-three-year-old intern at the journal Foreign Policy. When Huntington left his position on President Carter’s NSC staff as director of long-range planning to return to Harvard, he was succeeded by Fritz Ermarth, who shortly thereafter hired Friedberg to do a classified history of US nuclear doctrine. Friedberg needed a place to work, and Ermarth was able to get Marshall to provide him with some space at ONA. In 1987 Friedberg joined the faculty at Princeton University, and in 1999 he was appointed professor of politics and international affairs. Like Rosen, Friedberg has combined scholarship and public service, including service as deputy assistant for national security affairs and director of policy planning on the staff of Vice President Richard Cheney during 2002–2005. As in the case of Rosen, once Friedberg had worked for Marshall, he found himself permanently drawn into the ONA director’s orbit.

  Starting in the late 1980s Friedberg began focusing much of his formidable intellect on an issue of great interest to Marshall: how dominant powers preserve their position in the face of challenges from rising powers, a topic Friedberg addressed in his 1988 The Weary Titan: Britain and the Experience of Relative Decline, 1885–1905. As the subtitle suggests, the book explored how British statesmen became aware of unfavorable shifts in Britain’s relative power at the turn of the twentieth century and how they sought to mitigate them. Over a decade later Friedberg returned to the subject of a dominant nation’s being challenged by a rising power. As with Marshall, it had become clear to Friedberg that China would pose a growing challenge to US primacy, and he endeavored to assess how the People’s Republic of China (PRC) would seek to assert itself and how the United States might best respond.

  The third Huntington protégé that came to Marshall’s attention was Eliot Cohen. Cohen was a contemporary of both Friedberg and Rosen while an undergraduate at Harvard, where he also earned his doctorate. Like Rosen, Cohen would later teach strategy at the Naval War College. After serving a brief stint in the secretary of defense’s policy planning staff at the beginning of the Bush administration in 1989, Cohen left the government to become the Robert E. Osgood Professor of Strategic Studies at the Johns Hopkins University’s Nitze School of Advanced International Studies. Along with Friedberg and Rosen, Cohen has combined his pedagogical career with frequent, albeit temporary, forays into public service. Following the 1991 Persian Gulf War Cohen was tapped by Air Force secretary Donald Rice to lead the Gulf War Air Power Survey, an assessment of air power’s performance in the war. The multivolume survey was widely praised as one of the most insightful and impa
rtial studies of air operations since the US Strategic Bombing Survey at the end of World War II. During George W. Bush’s administration Cohen returned to serve on Defense Policy Board, which advises the secretary of defense. In 2007 Cohen was appointed to serve as counselor to Secretary of State Condoleezza Rice, a position he held until the transition to the Obama administration in 2009.

  In 1983, shortly after earning his doctorate, Cohen, who had also been commissioned a lieutenant in the Army Reserve, was assigned to the Office of Net Assessment to fulfill his reserve duty requirements. As was the case with all new arrivals to the ONA staff, Cohen was instructed to sit and read—to self-educate himself in the workings of the office and the analytic discipline of net assessment. He soon impressed Marshall, and for years afterward was invited to serve his reserve duty time working in ONA.

  This was the case when, in early 1988, Cohen was drawn into a debate encouraged by the journal International Security. Although the Berlin Wall’s fall was less than two years off, at the time the Cold War seemed to have no end in sight and few disputed that the Soviet threat to Western Europe was both real and formidable. Just how formidable was the subject of considerable dispute. Analysis of the military balance in Central Europe focused on the forces facing each other along a line stretching from Denmark in the north, moving south along the intra-German border down to the West German-Czech border.

  In the event of war Soviet military doctrine called for Warsaw Pact forces to advance rapidly into Western Europe to seize West Germany and the Low Countries by employing blitzkrieg-style operations, with fast-moving armored forces supported by air and missile forces. NATO’s objective was to block the Soviets from achieving a quick victory while not giving up much territory and avoiding escalation to nuclear use. If the alliance’s conventional forces could accomplish these goals, it was hoped that NATO could mobilize its superior industrial base and manpower and compel the USSR to seek peace. Of course, NATO’s ultimate objective was not to fight at all but to convince the Soviets that the risks of taking military action were unacceptably high—that is, to deter them from attacking in the first place.

  With this in mind the editors of International Security commissioned papers to assess the Central European military balance. Three distinguished scholars—Joshua Epstein, John Mearsheimer, and Barry Posen—were engaged.92 Each was asked to present his assessment of the balance. In conducting their assessments, the three reached generally the same conclusion: The balance between NATO and Warsaw Pact conventional forces favored NATO because the Soviets and their Warsaw Pact allies lacked the combat power to win the breakthrough battles on which their campaign strategy depended. Mearsheimer declared, “The chief question in assessing the European balance is whether the Soviets have the wherewithal to launch a successful blitzkrieg against NATO. This question can never be answered with certainty, but there is ample reason for thinking the Soviets cannot overrun Germany with conventional forces.”93 Moreover, his assessment led him to conclude that “contrary to the conventional wisdom, NATO’s forces probably can stymie a Pact offensive in Europe without surrendering much German territory.”94

  Posen was even more optimistic, declaring that his analysis indicated “that the widespread impression of decisive Pact quantitative superiority is a myth” [emphasis in the original]. He concluded that NATO forces “are fully competitive with the Warsaw Pact in Central Europe” and that “it is clear that predictions of early NATO defeat based on Pact quantitative superiority should be treated with great skepticism. NATO actually is in a very good position to defeat a Pact attack [emphasis in the original].”95 Similarly, Epstein wrote that “overwhelming Warsaw Pact superiority has not been demonstrated using serious analytical methods,” and that even based on his “conservative” assumptions regarding its forces, “NATO has the material wherewithal to stalemate the Warsaw Pact” [emphasis in the original].96

  Despite their common findings, the articles triggered a heated debate among the three authors. Although coming to similar conclusions regarding the state of the Central Front balance, each found their colleagues’ methodology lacking.97 After all, the value of an assessment’s findings is heavily dependent on the line of argument used in reaching them—the methodology—and the evidence provided in support of the argument.

  Three members of St. Andrew’s Prep—Eliot Cohen, Jim Roche, and Barry Watts—also had serious issues with the methodologies employed by the three authors and proceeded to make their concerns a matter of public record in two articles, one by Cohen and the other by Roche and Watts.98 Their concerns echoed how Marshall approached the problem of estimating this key military balance, and reaffirmed the apprehensions the ONA director expressed in his 1966 seminal paper on the difficulties of developing accurate measures of relative military power.

  Having served in ONA, Cohen, Roche, and Watts knew that net assessment is very much about identifying and exploiting favorable asymmetries in key areas of the competition, and finding ways to mitigate the effects of those asymmetries that work in the enemy’s favor. Net assessment also relies heavily on reliable data—here, “garbage in” really does result in “garbage out.” It also requires a good understanding of the value—and limitations—of what the intelligence community can provide in the way of “good data.”

  Extending all the way back to Marshall’s master’s thesis that found Klein’s economic model severely wanting, he has always urged those engaged in crafting net assessments to be extremely wary of the ability of models of any kind to predict the behavior of highly complex phenomena, whether it be the factors shaping a nation’s economy or those influencing a war’s outcome. And as those familiar with Marshall’s approach to net assessment know, “Not everything that can be counted counts; and not every thing that counts can be counted.” That is, Marshall advised his staff that the impulse of many political scientists to quantify the main elements of military balances risked ignoring important qualitative and intangible factors. Shaped by their time spent under Marshall’s tutelage, Cohen, Roche, and Watts applied these and other Marshall insights in their critiques of the Epstein, Mearsheimer, and Posen assessments.

  For his part, Cohen began his critique by stating that the “Optimists”—the term by which he referred to Epstein, Mearsheimer, and Posen—discounted qualitative “fundamental asymmetries between West and East with respect to coalition unity and purpose” that could significantly affect war outcomes in Central Europe and, therefore, the balance of forces. As an example, Cohen noted that “Unlike the Soviet Union, the United States cannot dictate to its allies the fact, let alone the timing, of their mobilization, nor their choices of arms and doctrine.”99 Yet NATO’s ability—or inability—to mobilize all its members’ forces promptly once warning of Warsaw Pact mobilization had been received could significantly influence the outcome of a NATO-Warsaw Pact conflict.

  Cohen acknowledged the Optimists’ point that the armed forces of the Soviet satellite states, particularly those of East Germany, Poland, and Czechoslovakia, might not prove reliable. Yet Cohen noted that their reliability would be most likely to come into question if NATO forces were winning the war, not merely trying to stave off a Warsaw Pact offensive, a factor left unaddressed by the Optimists.100 Cohen also pointed out that the Optimists generally relied on a single scenario to inform their assessment: a mobilization race between the two coalitions that assumed the Soviets would only strike at the moment when the balance of forces was most heavily in their favor. Yet Marshall would never approve an assessment that self-limited itself to a single “canonical” scenario. At the time at least two other scenarios would have engaged Marshall’s attention. One emerged from President Carter’s pronouncement in January 1980 following the Soviet invasion of Afghanistan that the United States would commit itself to the defense of the Persian Gulf region against external aggression. The question that soon arose was how the United States would react if the Soviets launched an incursion into Iran. Would Washington mobilize and move forces
earmarked for NATO to the Persian Gulf region instead? If so, what did it portend for the Central Front balance?* The concern arose that a Soviet attack on Iran could be used as a diversion to draw US forces away from a coming Warsaw Pact offensive in Europe.

  Another alternative scenario was the so-called Hamburg Grab in which the war began as envisioned by the Optimists; however, the Warsaw Pact’s goal was not to defeat NATO outright but to grab enough NATO territory to fracture the alliance politically. A short, limited offensive to seize Hamburg, a major city in West Germany roughly forty miles from the border with East Germany, would also minimize the risk that NATO would escalate the conflict by employing nuclear weapons.101 Indeed, it might be possible to seize the city before NATO could mobilize its forces. Once Hamburg had been overrun, the Soviets could offer a cease-fire, calculating that NATO would prefer a negotiated end to the fighting in light of the obvious difference between the value of Hamburg and the cost (to include the prospect of nuclear war) of retaking it. If the offer were accepted, it was feared that NATO members—West Germany in particular—would conclude their pledge to consider an attack upon one of them as an attack upon all was hollow. This could create political fissures in the alliance that the Soviets could exploit.

  Cohen’s critique turned next turned to the issue of mobilization. Because both NATO and the Warsaw Pact could significantly enhance their combat forces by mobilizing those not in a high state of readiness, mobilization was an important factor in assessing the Central Front balance, and each of the three Optimists addressed this issue. While Posen asserted that he assessed the situation “more realistically” than did others, Cohen found that many of his assumptions about how quickly the two alliances could mobilize reserves and bring them to bear—including the entry of French ground forces—yielded an advantage to NATO.102