There were more subtle effects of resource constraints than those that could be measured in rubles or dollars. In the late 1970s, Marshall began supporting Murray Feshbach’s research on Soviet demographics. By 1982 Feshbach was reporting that male life expectancy in the USSR had dropped from a high of 67 years in 1964 to 61.9 years in 1980.6 Influenza, alcoholism, and limited and shoddy medical care were blamed by Western Sovietologists for this unprecedented decline in male longevity in a supposedly advanced nation. In Marshall’s eyes the decline in male longevity was a concrete manifestation of the reality of macroeconomic resource constraints.
A second intellectual theme in Marshall’s career—one that emerged during his collaboration with Loftus—was the importance of understanding the adversary, including the possibility that the opponent could change significantly over an extended period of time. This understanding required appreciating the bureaucratic, budgetary, and historical factors constraining incremental decisions by various stakeholders in peacetime that affected the adversary’s force posture (rather than decisions made in crises or wars). Good net assessments required appreciation of the needs and objectives of the various power centers able to influence force posture decisions as well as a sense of the adversary’s strategic culture. During the Cold War, the recurring inclination of most US analyses was to assume that Soviet force-posture decisions were made by a rational, utility-maximizing central authority with complete information and consistent goals. But none of these elements were commonly found in business firms or other large organizations such as military services. The sheer size and complexity of large organizations, Marshall and Loftus realized, precluded any single central authority from having either enough time or all the information needed to make all the important decisions optimally. Yet that is precisely what the rational-actor model of Soviet decision-making presumed. Beyond this view of decision making being simply mistaken, embracing the rational-actor model encouraged worst-case analyses that painted the Soviets as being “ten feet tall,” to use the popular Cold War euphemism.
Widespread reliance by Western defense analysts on the rational-actor model was also connected in Marshall’s mind with the reasons that both systems analysis and game theory had eventually failed to provide a satisfactory basis for higher-level strategic choices. In the case of systems analysis, James Schlesinger was right to point out that the highest strategic objectives or utility preferences “must in large part be imposed from the outside,” thereby limiting the applicability of systems analysis to higher-level strategic choices.7 The truth of the matter, he insisted, is that “for such goals as deterrence, assured destruction, controlled nuclear warfare, damaging-limiting, to say nothing of welfare benefits, we fall back, not on a firm technical base, but on what may be scientific mush.”8 As for game theory’s ability to answer the complex problems of strategic choice, Marshall’s judgment was that warfare was “just too complex” for such methods to have much utility.9
Given the limits of systems analysis and game theory, Marshall became convinced that in any long-term military competition it was vital to understand the opponent’s assessments, including the weaknesses and proclivities of the adversary, such as the Soviet obsession with territorial air defenses. That is why as the Pentagon’s director of net assessment he invested in Project Eager and John Battilega’s Foreign Systems Research Center, funded Enders Wimbush’s émigré interview project, and supported Feshbach’s research into Soviet demographics and Igor Birman’s work on the Soviet economy. But Marshall also paid attention to what could be learned about the perceptions and assessments of Soviet elites. Here he was influenced by Soviet émigrés such as sociologist Vladimir Shlapentok as well as by American Russian specialists including George Washington University’s Peter Reddaway, the Woodrow Wilson International Center’s James Billington, and Duke University’s Vladimir Treml. All of these individuals and their research efforts contributed to Marshall’s growing understanding of how much Soviet perceptions and assessments differed from those of most US defense officials and academics. Caspar Weinberger’s competitive strategies initiative, which among other things sought to use such US advantages as precision munitions and stealth to devalue the USSR’s reliance on second-echelon or follow-on forces and territorial air defenses grew directly out of the ONA’s sustained efforts over more than a decade to understand Soviet perceptions and assessments.
More generally, Marshall recognized that bounded rationality—the limits of problem-solving routines and man’s capacity to generate alternatives and process information—applied to US strategic choices as well as to those of prospective opponents, thereby constraining the decision-making of individuals and organizations on both sides of the Cold War. Bounded rationality also helps explain why ONA’s 1992 military-technical revolution assessment was unclassified and developed with the active participation of midgrade military officers and civilians from the military services, the Office of the Secretary of Defense, and the intelligence community. Recognizing how difficult fundamental change in large organizations could be, Marshall sought to “infect” midcareer officers and analysts with the lexicon and concepts of the late-twentieth-century revolution in military affairs. Still, he never felt it was his place to prescribe the changes that the US military should make to prepare for a mature precision-strike regime or what many termed information warfare. Those choices, he believed, should be left those officers when they had reached more senior positions.
A third theme that has run throughout Marshall’s career has been his approach to framing or structuring thinking about military competitions. To begin with, there are the actors, which at the most abstract level are states. But particularly from an organizational behavior perspective, it is usually necessary to consider some of the more specific actors and organizations able to influence a state’s strategic choices. In the case of the United States, for instance, influential actors might include the president, Congress, the Defense Department, the Joint Chiefs of Staff, individual military services, or even certain war-fighting communities (e.g., submariners, fighter pilots, special forces operators, etc.). The actors have different strengths, weaknesses, access to resources, vested interests, and patterns of behavior. They also have goals—and strategies for attaining those goals—stemming from their histories and life experiences or cultures. Some of these factors can change more rapidly than others. Intentions and some aspects of technology can change virtually overnight, whereas demographics, differential rates of economic growth, and the earth’s climate not only change more slowly but may not be entirely under the control of the actors.
The players or states, in turn, are embedded in the prevailing international security environment: the Cold War, whether it be the second nuclear age, or the conditions flowing from such things as Russian revanchism or China’s rise. In this context nations develop over time a complex set of goals and seek to attain them by competing individually or by forming coalitions. The geo-strategic context of a competition can, of course, change, sometimes rapidly and radically, as happened in 1991 when the Soviet Union collapsed to the surprise of almost everyone. But the broader point is that opposing states “need to be treated entirely symmetrically” within this framework.10 Note, too, that there is no reason to restrict the actors to states. Terrorist organizations, such as al Qaeda, its affiliates, and its offshoots are increasingly relevant actors because, since the 1950s, technology has dramatically increased the damage that small groups of dedicated individuals can inflict on nations and societies.
A fourth theme that emerges from Marshall’s intellectual journey is the importance of looking at particular issues from the perspective of the long-term-competition framework that he developed during 1969–1970 after succeeding Schlesinger as RAND’s director of strategic studies. This framework had antecedents, especially in George Kennan’s analysis of the sources of Soviet conduct, from which the guiding principle of containment emerged. But when Marshall resurrected the idea of viewing the US-Sovi
et rivalry as an extended competition, he was struck by the change in perspective that resulted from this shift in focus. Up to that point in time, the standard analyses of the US-USSR strategic nuclear balance had largely focused on quantifying the immediate requirements for deterrence and the likely deaths and destruction the United States would suffer if deterrence failed. For both Pentagon decision makers and strategic analysts this focus tended to narrow attention to specific force posture and programmatic choices. Insufficient attention was paid to the more complex, evolving goals that both sides developed and pursued over the course of their extended military competition.
What Marshall realized was that embracing a long-term-competition framework allowed US strategists to broaden their focus from trying to quantify deterrence by calculating arsenal exchanges to, instead, viewing the rivals’ relationship as a series of moves in peacetime aimed at improving the United States’ position relative to the USSR’s. This perspective would focus attention on leveraging areas of the United States’ enduring comparative advantage, exploiting enduring Soviet weaknesses, imposing disproportionate costs on the USSR, and otherwise complicating the Moscow’s efforts to compete effectively. The genesis of Weinberger’s competitive strategies initiative is not difficult to discern in this framework which, as Marshall later recalled, transformed the way in which he subsequently viewed many defense issues, not just the US-Soviet strategic nuclear balance. The long-term-competition framework has been central to Marshall’s approach to net assessment ever since.
A fifth intellectual theme that has persisted throughout Marshall’s career concerns what might be termed the nonrational aspects of human behavior. This theme can be traced back to Herbert Simon’s conclusion that evolution had adapted organisms to make “good enough” decisions rather than “optimal” ones.11 This view of human decision making was reinforced by Marshall and Loftus’s realization that Soviet organizations often made suboptimal decisions about the USSR’s strategic-nuclear forces—such as the location of their bomber bases—that could not be explained by the rational-actor, utility-maximizing model of human behavior.
Marshall’s conclusions about human decision-making led him to begin exploring the then-emerging literature on the behavior of business firms and, in the 1960s, to initiate research into organizational behavior as a way of transcending the limitations of traditional systems analysis. This line of thought was reinforced by reading and discussing with Schlesinger works on the evolutionary roots of human behavior, including Robert Ardrey’s The Territorial Imperative, and, later, meeting Ardrey through the anthropologist Lionel Tiger, who became a regular participant in Marshall’s Newport summer studies. Marshall’s conviction that human behavior has inherently irrational components was reinforced by Tiger and Robin Fox’s The Imperial Animal, which argued that evolution had wired “identifiable propensities for behavior” into the human genome.12 Man, Tiger insists, is an “action animal” rather than a “thinking animal”; human behavior is dominated by the ability to make quick, astute choices to survive. Further evidence of human irrationality was provided by the seminal work on judgment under uncertainty by Daniel Kahneman and Amos Tversky.13 Among other things, they found that even experienced researchers trained in statistics were prone to fallacies in making intuitive judgments about probabilities in more intricate and less transparent problems.14 Given all this evidence, Marshall agreed with Fox’s conclusion that that the human race is as likely to have a future of peace and nonviolence as one of chastity and nonsexuality.15 Fox’s conclusion in 1991 was that war is neither an institution like slavery that can be outlawed, nor a disease that can be cured, “but part of the normal human nature.”16 Hence Marshall’s belief that war is an integral part of human nature. The use of military force might be controlled but never banished.
A sixth intellectual theme that has influenced Marshall’s approach to analysis in general and net assessment in particular concerns the difficulties of thinking about the future. Net assessments were supposed to look ahead and highlight emerging strategic problems and opportunities that could be leveraged to gain competitive advantage. Nevertheless, his long-standing conviction has been, and remains, that it is not possible to predict either the strategic outcomes of actual conflicts or the course of military competitions. There is simply too much uncertainty. Despite the desire of senior leaders for accurate forecasts of what the future holds, the stubborn fact that the future is unpredictable has to be faced. As the economist Douglass North has written, the reality is that in human affairs we “cannot know today what we will learn tomorrow which will shape our tomorrow’s actions,” and the world itself is “nonergodic,” meaning that statistical time averages of future outcomes can be and, more often than most people appreciate, are persistently different from the averages calculated from past observations.17 Some of the most concrete evidence for unpredictability in protracted competitions comes from astrophysics. We now know, for instance, that the clockwork view of the planetary motion suggested by Isaac Newton’s mechanics is wrong. The movement of the planets around the sun is in fact chaotic. If one looks far enough into the future—say 100 or 200 million years—computer models lose “all ability to predict planet trajectories.”18 And in strategy and international affairs as in planetary science, unpredictability can take us by surprise literally overnight.
Over the years, Marshall has advanced three main suggestions for how net assessment can look ahead to identify emerging challenges and opportunities for the United States, given that the uncertainty of the future cannot be eliminated. First, some aspects of the future are surer than others over time spans of years or decades. Examples include demographics, long-term macroeconomic trends such as differential rates of economic growth, and some aspects of technology. In the latter case, in 1997, four years before al Qaeda’s successful attacks on the World Trade Center and the Pentagon, Martin Shubik pointed out that since the 1950s, the technology of killing, combined with advances in finances, communications, and logistics, had increased the lethality of a small organized group (say ten to twenty trained, dedicated individuals) by orders of magnitude.19 Creating large-scale death and destruction are no longer the exclusive purview of nations with huge economies, large militaries and weapons of mass destruction.
Second, the use of scenarios can be valuable not as predictions so much as tools to help decision makers envision alternative futures and think through how to respond to particular situations or challenges. Here Marshall often pointed to Pierre Wack’s adaptation of the theories of futurist Herman Kahn to create the scenario approach that prepared Royal Dutch/Shell’s managers to cope better than their industry rivals with the abrupt quadrupling of oil prices precipitated by the October 1973 Arab-Israeli war.20 Scenario planning, Marshall maintains, is a viable way, in Wack’s formulation, of “reperceiving” reality in turbulent times or, as Peter Schwartz later put it, of taking the long view in a world of great uncertainty.21
Third, since the 1950s Marshall has viewed war-gaming as a useful way of exploring higher-level strategic problems. In 1950 the RAND mathematicians Merrill Flood and Melvin Dresher first posed the prisoner’s dilemma, a fundamental problem in game theory that demonstrated two people might not cooperate even if it was in the best interests of both to do so. Marshall recalls that early in Flood’s RAND career he observed that human beings are very good at playing games, and well-structured games can generate insights into reasonably effective strategies for coping with situations that elude the closed mathematical solutions of formal game theory.22 Flood’s observations led to a number of insightful competitive games played at RAND in the 1950s. Herbert Goldhamer was an early innovator in political or crisis games, and RAND’s Strategy and Force Evaluation (SAFE) game was used to explore US and Soviet alternative budgetary options for planning and programming force postures.
Marshall has continued to fund war games as a way of gaining insights into the strategies competitors might adopt in specific situations, usually articul
ated through scenarios that included both well-established trends along with possible discontinuities. One of the more successful examples of scenario-based war games that provided strategic insight was ONA’s Strategic Defense Initiative (SDI) series in the mid-1980s. It revealed that ballistic missile defenses with far less than near-perfect effectiveness might make the US-Soviet nuclear balance more stable. Due to the high probabilities of success—upward of 90 percent—Soviet planners insisted upon, ONA’s SDI games suggested even leaky US missile defenses would require Soviet planners to target so many warheads against the US ICBM force that they could not be confident of executing an effective counterforce strike against even one leg of the US nuclear triad.
A final intellectual theme throughout Marshall’s career, which is also reflective of his personality, is his realization that one should be modest about one’s expectations of what can be accomplished in large organizations such as the Defense Department. Inevitably there are a lot of less-than-ideal, if not deleterious, activities and choices on defense issues that are made by the military services, the Joint Chiefs of Staff, OSD, and Congress. Only so much of this dysfunctional behavior can be prevented, an outlook captured in what has been perhaps Marshall’s best-known saying: “There’s only so much stupidity one man can prevent.”
Because most of the assessments produced by ONA have been highly classified and written primarily for the use of the secretary of defense, any understanding of what Marshall’s office has accomplished over the last four decades has been limited, even within the Defense Department. This, combined with Marshall’s aversion to publicity and his refusal to define what he sees as an ever-evolving form of analysis, has led even some of those close to him to define net assessment simply as “what Andy Marshall does.” Similarly, a recurring question many in the Pentagon and the strategic studies community have asked about ONA has been: What did Marshall ever accomplish?
The Last Warrior Page 35
