Chapter 1 INTRODUCTION I nn'OVATions in the production, deployment, and application of military power are crucial to international politics. Unfortunately, most assessments of the international security environment fail to incorporate cither the relevance of military innovations or the importance of their spread. For example, in a thirty-year period, from 1850-80, the French Navy became the first to develop shell guns, and the first to deploy a steam-powered warship, an ironclad warship, a mechanically powered submarine, and a steel-hulled warship. These developments should have helped the French Navy gain superiority over its bitter rival, the British Royal Navy, but they did not. Moreover, barely a decade after the introduction of the steel-hulled warship in the 1870s, a new innovative school of naval theorists in the French Navy argued that the future of naval power lay with emerging technologies like torpedo boats and submarines, not the battleship. France was going to jump ahead once again. Yet despite this foresight and demonstrated initiative, most people generally do not consider France a great naval innovator of the period. Why is this? What advantages did it get from its introduction of a series of useful technologies into naval warfare? The real answer is that the French Navy received no advantage. Unlike the U.S. Navy, whose mastery of the technology and organizational practices associated with carrier warfare provided it with a sustainable edge in naval power in the second half of the twentieth century, the French could not institutionalize their advantage. While the French excelled at inventing new technologies, crippling organizational debates prevented the integration of those technologies into French naval strategy. In each case, the French were the first to introduce a new naval warfare capability, while the British Admiralty appeared, in public, disinterested in French developments. Yet in each case the British, who had been carefully studying French advances in private, quickly adopted the new capabilities, improved on them, and used Great Britain's superior industrial production capabilities to eliminate France's ability to gain a relative power advantage from its inventions. A prescient analysis in 1902 of submarine warfare by Herbert C. Fyfe, the "Sometime Librarian of the Royal Institution, London," includes an appendix on the French Navy that expresses French feelings on the matter: "We have seriously believed)" says a writer in the Journal de la Marine, "thai in all the great modifications that have been brought about in the 2 CHAPTER 1 INTRODUCTION 3 construction of submarines is the result of the important changes which the last fifty years of the century have produced in the art of naval warfare. All these changes have been sought out, experimented upon, studied, and finally realized by France, who has also been the first to apply them. These results have established in a brilliant and incontestable manner the skill of our engineers; but our rivals have not only appropriated the results of our labours, but they have not been slow to place themselves on equal terms with us, and finally to excel us in the application of these discoveries____We have been only the humble artisans working for them to establish their superiority." {Fyfe 1902,281) While France was the technological first mover in several cases, it failed to harness its advances into an actual war-fighting innovation in a way that increased France's relative naval power.1 Instead, it was the British Royal Navy that came to exemplify naval power in the mid- to late nineteenth century as it entered an era of naval superiority. The failure of the nineteenth-century French Navy to exploit its technological inventions in sea power yields two important lessons for a general understanding of military power and international relations. First, inventing technologies or even being the first to use them does not guarantee advantages in international politics. There is a big difference between the introduction of a technology on to the battlefield and the full integration of that technology into national strategy, including warfare and coercive diplomacy. It is the difference between the two, in fact, that often determines success or failure in international politics. It is the employment of technologies by organizations, rather than the technologies themselves, that most often makes the difference. Second, in contrast to most prior work on military innovation, which has tended to focus on who innovates and why, it is the diffusion of a military innovation throughout the international system that most determines its influence on international politics. The study of military power is incomplete at best without a theoretically coherent understanding of how states respond to major military innovations, and how the pattern of their responses helps drive the rise and decline of nations as well as the patterns of warfare frequently analyzed by other scholars. By developing a theoretical framework that can bring together empirical topics like suicide bombing and carrier warfare that scholars have tended to study separately, this book presents a new, more efficient, way to think about approaching the diffusion of military power. The introduction and spread of new means of generating military power, sometimes called major military innovations (MMIs), have played a critical role throughout history in determining the global balance of power along with ''litis introductory section in based on both Kyle's and 'Ihroilon f*■ ■ t'|'' '" ''" !•"■« •»« li Navy (I'yl'c VM2; l'JH7,H 11,'12). the timing and intensity of wars.2 The infamous Mongol armies, with their mastery of the composite bow and a new form of cavalry strikes, toppled nations from China to those in the eastern part of Europe because of their leaps in technology and strategy. Hundreds of years later, the German debut of blitzkrieg warfare at the outbreak of World War II helped them rout French forces and consolidate control over Western Europe. But despite their significance in terms of driving change in international politics, the processes that govern the spread of innovations and their effects are little understood in the field of international relations. Several questions about military power remain unanswered: Is it best to be the first mover, to borrow a term from economics, and the first to figure out how to effectively employ new types of military power, like the Germans with blitzkrieg?3 Or is it better to be a follower, learning from the leader, and trying to extend and improve the original ideas, like the Germans with all-big-gun battleships responding to British innovations? How do nonstatc actors fit into this story? Insurgent and terrorist groups have to make decisions about military strategy just like nation-states. How do they decide whether or not to adopt new innovations in how they use force like suicide bombing? This book addresses the broad puzzle of why some military innovations spread and influence international politics while others do not, or do so in very different ways. These patterns are explained with a theory of the spread of military power called adoption-capacity theory. Nation-states have a number of possible strategic choices in the face of military innovations. These include adoption, offsetting or countering, forming alliances, and shifting toward neutrality, as noted in the preface. Adoption-capacity theory posits that for any given innovation, it is the interaction of the resource mobilization challenges and organizational changes required to adopt the new innovation, and the capacity of states to absorb these demands, that explains both the system-level distribution of responses and the choices of individual states. As the cost per unit of the technological components of a military innovation increases and fewer commercial applications exist, the level of financial intensity required to adopt the innovation increases. The rate of adoption decreases and alternatives like forming alliances become more attractive. Similarly, if an innovation involves large-scale organizational changes in recruitment, training, and war-fighting doctrine, the innovation requires a high level of organizational capital for adoption, and fewer actors are likely to adopt it. Some states will have the necessary capacity and interests, while politics will prevent adoption by others. If capacity and interest are lacking, no matter how ; ('li,i|iter .'. ili'ii 11-i ilcliniiij; and ii|ieralioiiali/iiiK military innovations. ' ,'V, i h.i|iii i '. ilf.i iilu-i, in the I ili i Aiiru i .e'', while die llrilish were I lie (irst movers with regard in lie (i'i Inn ili iny, the < n.....in debuted (lie mature innovation 4 CHAPTER 1 intrinsically compelling a new innovation may seem, it will not diffuse throughout the system. Accurately measuring these variations in diffusion also more effectively explains shifts in the balance of power and warfare than traditional theories alone can do. While higher financial requirements generally mean that the adoption patterns will benefit preexisting wealthy and powerful states, higher organizational change requirements can handicap the wealthiest states and upset the balance of power toward newer and more nimble actors. The question of how states deal with periods of uncertainty about military power is of special interest today. Significant global economic turmoil now accompanies ongoing debates about the future of warfare in the information age. International relations scholars have demonstrated that uncertainty about the current and future security environment can be a primaiy cause of conflict (Fearon 1994a; Powell 1999; Smith and Stam 2004). Sharp debates exist between those who believe that the United States should optimize its military for future counterinsurgency campaigns like Afghanistan and Iraq, and those who believe that the United States should focus instead on its conventional capabilities (Gentile 2008; Mazarr 2008; Nagl 2009). An important wild card for both perspectives is the role of the information age in international conflict. The information age is popularly described as the application of information technology to enhance the productivity of businesses and government, increasing the ability of societies to rapidly create and disseminate large amounts of information anywhere around the globe in real time. The information age, like the Industrial Revolution before it, will eventually have a large-scale impact on warfare. While some degree of change is likely inevitable, the details of that change and the consequences are still very much in the air. In particular, the United States currently appears to lead the globe in developing and integrating information age advances into its military forces. But software-heavy developments may come to dominate the information age, rather than expensive physical hardware. The declining cost of computing technology, Internet access, and devices like personal GPS units, along with the dual-use nature of many information age military technologies like precision-guided munitions, mean new capabilities may become available to an increasing number of countries over time. While the United States has led the way in utilizing information technology in its military operations, its lead is far from assured. Peter Singer (2009) has described the way that the robotics revolution will impact the future of warfare, contending that there are risks for the United States as well as potential benefits. In that hypothetical case, the U.S. government's devotion to its tanks, bombers, and carriers could become an albatross that drags down the U.S. military, which might face organizational challenges in transforming il'.i'lf, in I'avor of states that figure out new and better ways to orgunivr rlirli liin cm |i i take ncl vantage of inlbrmalioii age k'i'l«n. di'.i uv.iou (1979, 231-12) of when to debut new weapons in the first place. Whether or not the liiil mover in u keep I lie 'iiiiiuv.il ion secret may be more or less possible depending, on I be 11...... n iiiui turn ii.iiiiui.il politic id onvironmorUi. systems to explain how military innovations spread once they have been introduced into the international system."' The basis of the theory is recognizing the adoption-capacity requirements of an innovation and how the capacities of individual states measure up. Approaching the spread of military power from this perspective sidesteps the traditional debate about whether strategic competition, cultural factors, or norms best explains emulation and allows for the construction of a more powerful new theory. There are many reasons why states are interested in adopting innovations: strategic necessity, international norms, cultural openness, the need for interoperability with allies, and many others. Threats are a vital part of the matrix of factors that motivate nation-states. It is even possible that for the states that initiate military innovations, threats play an important role in their drive and capacity to innovate.11 Prior research, though, has often presumed that for potential adopters, where there is a will to adopt, there will be a way (Elman 1999; Resende-Santos 2007), In the real world, states are sometimes overmatched no matter how well they optimize, and sometimes states do not adopt innovations even when they face large threats. Rather than viewing capabilities as totally fungible depending on state strategy, at least in the short-to medium-term it might be financial and organizational constraints that shape possible strategies as well as the probability of success. Adoption-capacity theory argues that, once states have the necessary exposure to an innovation, the diffusion of military power is mostly governed by two factors: the level of financial intensity required to adopt a military innovation, and the amount of organizational capital required to adopt an innovation. As briefly introduced above, financial intensity refers to the investments required to purchase the physical hardware associated with an innovation, along with the relative ability of states to make those investments. Key to determining the level of financial intensity required for adoption is whether the relevant technology is exclusively military or has commercial applications, and the cost per unit of the physical hardware associated with the innovation, like a battleship or an aircraft carrier, in comparison with previous procurement. The more military oriented the technology and the higher the unit cost, the higher the financial intensity required for adoption.12 '"The notion of relative costs here is complementary to measurements of military balances that incorporate the way new military advances can influence the relative cost of war (Anderton 1992; Powell 1999,110-12, 197-98). James Fearon (1995a, 6-8) in particular explicitly recognizes the way that assumptions about rapid emulation have influenced offense-defense debates. Instead, in his view, what matters is the relative pacing of adoption. " Harry Posen (1984, 1993) argues that threats drive the innovation process by determining whether civilian intervention occurs. "Ibis is related to research on capital intensity, but rather than focusing on the trade-offs between labor and rapilal, financial intensity is more about the way that capital is invested (Gartzke .1(1(11). II) l MAN I'U I INTRODUCTION 11 TllO odlOl li.ill <, tnvolvocl In |mi 1 • i n > ■. mit exactly what determines their propensity to change. 'Ilioiigh IliiH will always he an issue, there are a variety of different ways to mciimire mid evaluate the capacity of military organizations to change. Organl/al.....al 1 apit.il in......npetlccC but powerful way to conceptualize the potential change 1 ap.u ily of a military organization. Three factors in particular, measurable in mtlit.n v organizations prior to the demonstration of a given innovation, appeal hi Ik-.i predict whether or not the organization will have the necessary capacity to adopt. Must, the amount of resources devoted to experimentation is an indicator of the willingness and ability of organizations to consider major innovations. Second, as Mancur Olson (1982) contends in economics, older organizations often become bureaucratically ossified as subgroups of control proliferate, generating an increasing number of veto points that prevent innovations from being adopted. Therefore, the longer military organizations last without experiencing serious upheavals such as regime changes from within or defeats in interstate wars, the worse they should be at integrating innovations.13 Finally, the way that military organizations define their critical tasks plays a vital role in defining the range of the possible for those organizations (Wilson, 1989). The broader the definition of the organization and its purpose, the better it will be at adopting innovations. Al Qaeda's willingness to consider any and all operational methods for attacking the United States and its allies made it nimble enough to adopt suicide terrorism. Al Qaeda defined the means it would use to achieve its goals very broadly. In contrast, when an organization narrowly defines the optimal means to pursue its goals, the chances get higher that pushback from elites within the organization will prevent the adoption of innovations. A textbook case of how a limited view of the means to success can negatively influence an organization is the U.S. Army during the Vietnam War. The Vietnam era U.S. Army viewed using superior firepower as not just a means to an end but rather an end in itself; its ability to employ that firepower bled into how it measured success and failure. In 1965, the army even defined success based on the generation of enemy casualties.14 This made it difficult for the army to master counterinsurgency operations requiring a lower emphasis on lethality (Krepinevich 1986, 5). The army instead preferred search-and-destroy "There is no necessary correlation between organizational age and size. Nevertheless, older organizations are more likely to produce special types of bloated bureaucratic structures that make change difficult. 14It is even possible, based on Scott Gartner's work (1997), to argue that the focus on overwhelming firepower might have influenced the army's choice of body counts as lis dominant indicator, or metric for success, during the Vietnam War. missions where it could apply maximum firepower and generate the largest number of casualties (Gartner 1997,130-31). The speed and extent of an innovation's spread therefore depends on the relative financial and organizational requirements. Those requiring less to adopt will spread faster than those that require more. Adoption-capacity theory shows, however, that the levels of financial intensity and organizational Capital required to adopt an innovation not only significantly influence the rate and extent of its spread throughout the international system but also drive its affect on international politics. Since it is generally easier to adopt the physical technologies associated with an innovation than the overall system of fighting, innovations featuring especially high levels of financial intensity are likely to ,piead, albeit slowly. In particular, financially intense innovations requiring organizational changes that sustain, rather than disrupt, previous critical tasks are likely to spread gradually but consistently, benefiting the preexisting strongest and wealthiest states in the international system. While preinnovation major [lowers lacking the financial capacity to adopt are likely to slip and become second-rate powers, the innovation is unlikely to reorient systemwide power balances. In contrast, innovations requiring disruptive organizational transformations but relatively reasonable financial investments, like blitzkrieg, the (ierman combination of the radio, airplane, tank, and other motorized vehicles, will spread haltingly, with only a few states adopting the full innovation, and most acquiring some of its technical components but not adopting the new system of warfare. Innovations requiring large degrees of disruptive organiza-lional change most clearly create strategic openings for power transitions and generate larger first-mover advantages. New powers that master the necessary organizational changes can gain an advantage over their potentially bigger (hough less nimble major power opponents. Essentially, new major military innovations can create discontinuities in international politics, ushering in the risky situations described by Robert Powell (1999, 85, 199) where the actual balance of power sharply diverges from the distribution of benefits in the international system, because the system has not vet caught up to the new power realities. If a rising power develops a new innovation, it may gain an enormous edge in its drive to the top. In response, status quo powers that can quickly mimic and adapt to new military innovations or respond with their own new innovations have the best chance of limiting the disruptive impact of the innovation as well as maintaining their relative power level in the face of a challenge (Gilpin 1981,60-61,161-62). Sometimes, however, new major military innovations confront major powers, but for financial or organizational reasons they cannot adopt in the short- to medium-term. This presents a major power with a fundamental choice: continue posturing as if it is ft major power, or recognize the writing on the wall and seek an alternative strategy that may involve making iln inteie-Us Mth.idiuiy to those of another likely adopter. When match choline the liimiei |mill, like the AustroT lungarian 12 CHAPTER 1 INTRODUCTION 13 Empire did before World War I, it can destabilize the international system by accentuating informational gaps in national analyses of likely war outcomes. The resulting gap between beliefs and reality arc a common cause of war because they make miscalculation and escalation more likely on all sides.15 Adoption-capacity theory is also useful for explaining the behavior of non-state actors, as chapter 6 highlights. Like conventional militaries, insurgent and terrorist groups must make decisions about resource allocations and the organization of their forces. Financial intensity and organizational capital are useful metrics for understanding the strategic choices of terrorist groups in the suicide terrorism era. Those groups with critical tasks based in particular operational methods and that existed long before the beginning of the modern age of suicide terror faced substantial hurdles to adopting the innovation. The PIRA and ETA never adopted, for example, while it took Fatah, a key part of the Palestinian Liberation Organization (PLO), nearly twenty years to adopt. In contrast, groups with younger organizational ages and less defined critical tasks, from the broad tactical setup of the Tamil Tigers to the cell-based network of Al Qaeda, were more easily able to take advantage of suicide tactics, providing them with a new weapon. Just as the tacit knowledge required to effectively operate aircraft carriers creates significant organizational obstacles for countries interested in adopting carrier warfare, the availability of instruction in suicide methods or direct geographic proximity to suicide terrorists has constituted a tacit barrier to entry for some terrorist groups. This is not to say that variables like national liberation movements, religion, and/or fighting for popular influence are not motivating factors for the adoption of suicide terrorism. Rather, adoption-capacity theory can explain both the groups that have adopted and why other groups do not adopt suicide terrorism—something prior work has rarely addressed. Prior theorizing on terrorist strategy, like that on military innovation, has tended to focus on what drives the interest of terrorist groups in suicide bombing, implicitly assuming that the desire to adopt suicide terrorism is enough to make it happen (Pape 2005). Applying adoption-capacity theory to the case of suicide bombing shows the web of interconnections between groups and the flaws in trying to predict terrorist group behavior without an understanding of the broader linkages between groups. It is important not to overstate the scope of the theory. There are a variety of reasons why states are interested in innovations, why states adopt innovations, and why states become more or less powerful. Hopefully this book can make a contribution to ongoing debates in the academic and policy worlds about what types of changes are more or less likely to occur in periods of uncertainty about military power. 1!Tlns argument builds from work on bargaining, Informal.........»1 w,n In |>,uii> ul.ir, mt I'V.mm 199Sb. Importance for the Future of Warfare yvllile this book is mostly focused on the military innovations of the past, looking forward to the future is useful both as a demonstration of the theory nul to show its relevance to ongoing policy debates. Adoption-capacity theory • .in help explain the way different types of warfare in the future will provoke ililli'ient types of reactions on the part of responding actors, and benefit or ■ Ir .nivantage different states.This can help provide a frameworkfor discussion l>\ bowing how the likely implications for the security environment depend mi particular assumptions about the future. AI present, there is a spirited debate occurring around the world about the i \ pi", of wars most likely in the future. In the United States, the debate about Mu- utility of "network-centric warfare" has given way to one about whether the I lllltcd States should focus its limited resources on institutionalizing the hard- ■ n ncd counterinsurgency lessons of Afghanistan and Iraq, or refocus the U.S. military back toward conventional warfare (Boot 2006, 8-9; McMastcr 2008, "i .!H). Hiddle maintains that there is not enough evidence to overturn the prominence of the "modern system" of warfare—the use of firepower, cover, nul i onccalment to take and hold territory in conventional land engagements. I t)i lilting on the dangers of allowing technology to dictate force structure, he i in 1 that the modern system is the "orthodox" approach to war, and that »1 linl.it-. and policymakers should be careful before embracing "heterodox" ap- I.....h Ins (Hiddle 2007b, 463-64). He also notes that many instances viewed as ......u n insurgency campaigns, like Lebanon, have actually revolved around the ipplii .ii ii mi of conventional modern system principles (Biddle and Friedman ■JIHIH). thinking that the information age will make a difference in future warfare ■ m mil mean excluding the human clement or skill on the battlefield. Nor I n mill believing in the human element along with the importance of tactical I ■> * ■ 111 Icncy or skill mean ignoring the way that the information revolution may Itipi the realm of the possible in warfare (Gray 2006). One popular and per-i m' perspective, promoted most clearly by Frank Hoffman (2007), argues • Ii n (lie lii lure of warfare will be "hybrid," demonstrating facets of both regu-I u nul Irregular wars, but in an operating environment characterized by the ml,n ination age. 11 i lir Information age, like the Industrial Revolution before it, is likely to 111 , wide reaching and complicated effects on society, determining its impact ,,ii 11 iľ loi uiily environment matters whether the most probable future combat .......n in', me potential U.S.-China scenarios in East Asia, land-heavy wars • ■ llii Middle Kast, or quasi-pcace enforcement operations around the globe, li i pofiniblc tli,U the most likely wars of the future are irregular campaigns Ii iimiiii), kind loiccs, bul lhal there arc ako ',i|',nilicant possible contingencies In nli in;', llie heavy use of naval and ,ui I.....■>< 'Hie impact of the information 14 CHAPTER 1 age on each of these might be different, just as it might be different for states and nonstate actors.16 The U.S. military struggled during the early part of the last decade learning how to fight against insurgents in Afghanistan and Iraq. Yet many who think about the future of military power argue that at the very least, the U.S. military will indefinitely maintain and deepen its conventional military superiority (Brooks and Wohlforth 2008, 27-35; O'Hanlon 2000,168-69). Acclamatory statements that the U.S. military has already mastered the information age arc surely overstated; the United States leads the world in the application of information technology to its military operations, but U.S. military operations over the last decade suggest that the United States has far from mastered the information age. There are several areas where disruptive changes could influence the trajectory of warfare. At present, the U.S. military has made great strides in precision warfare— the use of new communications and guidance technologies to hit targets more accurately as well as at greater distances than ever before. These capabilities are cited by both counterinsurgency and conventional war advocates as important for the future, although there are disagreements about their relative effectiveness at present. Whether or not these advances are properly categorized as an MMI is a matter of debate. But the initial demonstration of U.S. precision-guided munitions in the Gulf War may have represented the starting point of the ticking innovation clock, like the debut of aircraft carriers by the Royal Navy and/or the introduction of the tank by the Royal Army in World War I, rather than representing a completed, fully functional MMI (Welch 1999, 122). Linear advances In precision warfare up to this point have extended the edge of the U.S. military at conventional operations. Utilizing the innovation requires expensive platforms like bombers and ships, meaning there is a high level of financial intensity required to adopt. Conducting counterinsurgency operations has proven challenging for the U.S. military, however, due to the large-scale organizational challenge of shifting the armed services away from focusing on overwhelming firepower. Precision warfare, under most foreseeable circumstances, will initially remain costly even as the reliance on major weapons platforms like bombers decreases. The financial intensity required to implement anything like what the U.S. military does today is so high that even a small decrease in unit costs will not allow many more states to actively seek military dominance. Moreover, as long as the "While critics of network-centric warfare, like Frederick Kagan (2006, 3S9-90), and scholars studying warfare in the information age, such as Peter Dombrowski and Eugene Gholz (2006, 4-6), are certainly right that the information revolution will not necessarily lead to one particular optima] force structure outcome, that doesn't mean the changes wrought by the information age are irrelevant or that the information age is unlikely to matter at all. INTRODUCTION 15 ......platforms for using precision warfare are linked to the platforms of today, tOCI uiting, training, and organizing modern militaries will look similar. Advances in areas like robotics and information technologies such as computing could shift the military power status quo, though. If advances in muni-i mi ľ, and especially unmanned vehicles begin to make the expensive launching |tl iilorms that sit at the core of the U.S. military irrelevant, it could risklarge-I ilc changes in military power balances. If a cargo ship or cargo plane is sud-ill ulv just as good for launching a missile at a target as an F-15, the financial iiiicii'.ity requirements for implementation will drop and the organizational • i|ni,il requirements will increase.17 In such a situation, militaries may also have to recruit differently—recruiting unmanned aerial vehicle (UAV) pilots v\Iih excel at video games instead of "fighter jocks," for example—and train I" i iplc to conduct different tasks, since they will be operating mostly with joy- in kis rather than in actual battle spaces with the enemy. I ľ rising U.S. military capabilities illustrate a path away from the financially lOt'Onse platforms that currently help ensure U.S. dominance, while also re- !......tg militaries to organize themselves differently to best take advantage of Unliable capabilities, countries such as China and India could find it increasingly possible and attractive to militarily compete with the United States. This tiiw level of financial intensity and high level of organizational capital required in lake advantage of the information age in the area of conventional war would • Ikii lower the barriers to entry for potential competitors to the U.S. military. I Ithcr states in the international system could then acquire the necessary capabilities to begin effectively reducing the military edge of the United States Unlcsa the country continues innovating to stay ahead. Possible areas for de-| < lopment include not only robotics, with UAVs as the most obvious manifes-......n, but cyberwarfare as well. The resulting situation could cause major lull', in the global balance of power as some states benefit and others, unable i......plcment, are increasingly left behind. Adoption-capacity theory princi- I'li ■. help explain the microfoundations underlying the concern by authors like M,i\ Hoot (2006) and Singer (2009) about the way the information age could boomerang, allowing other countries to catch up to the United States in the Inn)', run.18 liľ.i as it will influence warfare between states, the information age is likely in Influence the trajectory of actions by nonstate actors. The commercial spread ' Till* could occur because of increases in range and the miniaturization of tasks currently con-ihlClwl by lUpport planes like the Airborne Early Warning and Control System into next-generation nil««ll<'«. ' I lip, i'i (i lie whether one wishes to call this sort of development the second stage of precision „i H, oi i.......tiling as yet undclcmiinctl but related to the combination of materials, informa- i.....n , liiiiilnj'1v,.iiid L'uminuiiiniliniit (!s;i|J,,Hi n age and the ways it may influence the future of warfare. The conclusion h '.riibcs some crucial issues that are often absent in debates about the future ■ I \ merican defense strategy, and how adoption-capacity theory suggests that lie information age may portend a much greater level of risk for U.S. conven-Ll .ii.il military superiority than some previous authors have envisioned. MASARYKOVA UNIVERZITA FakuItacocial.i-.cn i u JoStova 10 ÚQ2.03 B O DIFFUSION OF MILITARY POWER 19 Chapter 2 A THEORY OF THE DIFFUSION OF MILITARY POWER War is a harsh teacher, Thucydides tells us. We either learn from others who are better at fighting than we are or we die. Yet there are puzzles. France knew after World War I that Germany remained economically and demographically stronger than France. Knowledge of the emergence of blitzkrieg warfare in Germany was available. In the 1930s, however, the French Army planned for the same kind of slow, methodical, defensive war that World War I had taught, even though the logic of its alliance system called for an offensive against Germany in the event of a German attack on Poland. Why? Egypt suffered a massive defeat in 1956 at the hands of the Israelis. It received massive financial and military technological assistance from the Soviet Union. It had every incentive to beat the Israelis at their own game, and the material means to do so. But it suffered a massive defeat again in 1967, for exactly the same reasons. France and Egypt should have learned from those countries that most threatened them, yet they did not. Why? Scholars interested in military power have devoted serious attention to the question of how states try to gain advantages over other states through the creation of new ways of generating military power, also called military innovations. Barry Posen (1984) and Stephen P. Rosen (1991) have theorized about the conditions in which militaries are most likely to innovate, disagreeing about whether innovations happen as the result of pressure from actors outside military organizations, military mavericks within the system, or changing promotion patterns that empower those with innovative ideas.1 This prior theorizing about military innovations in international politics, though, relies on clear implicit ideas about the way military power diffuses. By making those ideas explicit and engaging in empirical testing to determine which of them more accurately depicts world politics, this book explains how change in the international security environment occurs, advancing our understanding of military innovations beyond the first stage of the innovation life cycle. Since we care about military innovations because of their consequences for international politics, it is necessary to focus on what happens after the initial innovation is demonstrated, not just on the innovation process. For example, blitzkrieg is more than an interesting historical footnote precisely because 'For a recent review article, sec Grissom 2006. For examples of some key works on military innovation, sec Avant 1994; Evangelista 1988; Kier 1997; Muliiikeii 2002; Pierce 200'l; Poncil 1984; Rosen 1991; Sapnlsky 1972; Ziak 1993. (J et in any used it to invade France and France proved unable to effectively ii ipond. It is in the way states respond to innovations, the mechanisms gov-> 11111 ig the diffusion of military power, that the importance, or lack thereof, of Innovations for international politics emerges. Diffusion and Military Innovation What Is Diffusion ? Hir spread of military power works by means of a phenomenon known in in iiiv fields as diffusion. The defining text on diffusion research in the social Ii in cs, Diffusion of Innovations by Everett Rogers (2003,11), describes dif-hi .inn as "the process by which (1) an innovation (2) is communicated through certain channels (3) over time (4) among the members of a social I ilrin." Critical to diffusion research is the pattern by which successful in-|l< iviil ions generally spread throughout a population. Perceptions that an in-linviilion is effective create the framework within which diffusion does or I i not occur, meaning there is inevitably some linkage between innovation mil 11 i M Vision. Normal innovations often diffuse in the shape of an S curve— in insight applicable across a wide variety of disciplines and areas, including i ii iiliurc, consumer products, public policy adoption, and military technol-fwy "Ilicre are three essential stages of diffusion for "normal" products after a tfljnrn'ing event or debut. In the first stage, diffusion is slow as risk-acceptant Ii 11n 'i, or curly adopters, implement the innovation. In the second stage, once itilhiMou lias reached a critical mass (analogous to a tipping point in game (In iiiv literature), the rate of diffusion speeds up and most of the set of actors isiili 1114- ability to implement the innovation will do so.These are known as oi mi adopters, In the third stage, laggards, or later adopters, implement the Innovation, The total number of adopters, or percentage of adopters out of the 1.....11 >l potential adopters,graphed over time, is known as a cumulative adopter ill luUilion. innovations also often mutate as they spread and adopters alter them to the iHiilmii'i of their particular situation. For example, Geoffrey Herrera and tlixin.is Mnhnken (2003,242) show how, in the pattern of responses to Napo-' 11 Ii w.uTai'c and Prussian open-order tactics in the nineteenth century, coun-trll (ended to adopt some key components of the innovations but not others, Hid modify the innovation depending on their requirements. | in n he, of when the diffusion of ideas is more or less likely have also grown in ......i ili-i .ulrs, but (he notion that there are interactions between domestic or- inl Minns niul developments abroad is well established in the social sciences. I'mi I I l.ivul'ri work on dill'iision in the licit] of economics has also been significant. See, for lllipli , I lau.l I'lHN. 20 CHAPTER 2 Scholars ranging from Alexander Gerschenkron (1962), who wrote about competition and industrial development, to Peter Gourevitch (1978), with his argument about the way international politics conditions domestic economic decisions, have demonstrated the important connections between new practices abroad and decisions made at home. This research also shows the significance of developing theories that take into account the possibility that not all diffusion is efficient: some individual actors may not benefit from adoption, and some innovations are not relative improvements.3 Research in political science and related fields on diffusion has included work on domestic policy innovations (Berry and Berry 1990; Walker 1969), social policies (Mintrom and Vergari 1998), tort reform (Canon and Baum 1981), and tax policy (Berry and Berry 1992). Internationally, Beth Simmons and Zachary Elkins (2004) have studied the diffusion of financial policy, while others have looked at network effects in post-Communist states (Kop-stein and Reilly 2000), Latin American social sector reform (Weyland 2004), international environmental regulations (Busch and Jörgens 2005), elections (Blais and Massicotte 1997), the impact of foreign direct investment on the spread of technology (Borensztein, De Gregorio, and Lee 1998), market-based infrastructure reforms (Ilenisz, Zelner, and Guillen 2005), gender mainstreaming (True and Mintrom 2001), and the spread of democracy (Cederman and Gleditsch 2004; Ward and Gleditsch 1998), among other topics. Within the area of international conflict, some scholars have modeled the spread of warfare as a contagion process, drawing on theories from epidemiology (Simowitz 1998; Siverson and Starr 1991; Starr and Siverson 1998). Other studies of war and diffusion have included region-specific studies of Africa and Latin America as well as more general studies of the spread of coercive practices in international disputes (Bremer 1982; Hammarstrom 1994; Starr and Most 1983). Studying the Diffusion of Military Poiver Much of the existing research on the spread of military power comes from those who contend that strategic competition drives the diffusion process, states act primarily to maximize their own external security, and the most important determinant of behavior is the international environment. Applying his general insights on isomorphism, or convergence over time, to military technology, Kenneth Waltz (1979,127) writes, "Competition produces a tendency toward 1 David Strang and others have written about a wide variety of topics ranging from the spread of decolonization movements to the spread of poison pills in the corporate world to the Student shantytown movement in the campaign for divestment from South Africa (Smile 1999; Slrang 1991! Strang and Meyer 1993; Strang and Spule 1998). DIFFUSION OF MILITARY POWER 21 lire sameness of the competitors ... [a]nd so the weapons of major contenders, and even their strategies, begin to look the same all over the world." Posen's work (1993) on the Prussian response to mass mobilization and Joao Rescnde-Santos's study (2007) of South American rnilitaries of the late nineteenth and early twentieth centuries similarly show the way security challenges drive interest in adopting innovations.4 Colin Elman (1999, 55-57) maintains that while security maximization can take the form of much more than just adopting an innovation, meaning Waltz's predicted isomorphism is unlikely to occur, the security environment plays a primary role in influencing state choices.5 Others evaluate the way that norms derived from "world culture," rather than competitive pressure, influence how states emulate (Farrell 2005,451-52) Of argue that the cultural tolerance of foreign ideas in general predicts the successful adoption of innovations (Goldman 2006, 70; Pollack 2002).6 Research On the relationship between domestic politics, organizational factors, and military innovation has also played a central role in describing how militaries make decisions (Avant 1994; Evangelista 1988; Goldman and Eliason 2003a; Mahnken 2002; Zisk 1993). This book attempts to use prior theorizing as a building block to explain how, given the distribution of geopolitical interests and national attributes, the financial and organizational requirements for adopting an innovation determine how it spreads. Prior research has generally addressed one piece of the diffusion puzzle at a time. In contrast, I take the existing literature as a jumping-off point to build a more cohesive theory that describes how innovations diffuse, I iow states respond to innovations, and how those patterns impact interna- I ional politics. The question that mostly occupies the previous literature is what makes countries interested in responding to innovations, while the key question examined in this book is what determines the way these innovations end Up spreading in general and actually influence the international security environment.7 Adoption-capacity theory makes clear the distinction between interest in responding to an innovation and the substance of that response.8 'While some argue that studying emulation differs from studying diffusion because diffusion is descriptive while emulation is about the learning process (Resende-Santos 2007), in reality studying diffusion means evaluating the process by which an idea spreads, and emulation is one way that can occur-—as a subset of diffusion. ' For an assessment of how strategic competition can influence the innovation process, as opposed to the diffusion process, sec Goldman 2007. ''Theo Farrell's incredibly detailed work (2005,470) on the Irish Army and its doctrinal choices Her independence is an excellent example of llie way that norms can sometimes influence the i iinleut of military strategy, ■' For an ex. option, sec (ioldinau'ii work (200 I) on (he development of carrier warfare, which iI ■i|'li».-■■ i i lie llupi il l.mir ol i ap.li ily. " Ihe ilr.iiii. i..... belwcen iiltcinplitig i......plcinriil .in .......v.ilion and actually doing so has I" in pi ii hi nl i mi in i be < ontext nl i lie ill lien in . In le......n ill..... unli i ing military reforms and mllllilllrx ii.m.illv iiiiplrnir.....iB lliem (hi.....n I'lTl, ['),/ml, ]<)<>],',) 22 CHAPTER 2 DIFFUSION OF MILITARY POWER 23 Second, instead of focusing solely on whether or not states adopt innovations, adoption-capacity theory makes predictions about all the possible responses for a given state, only one of which is adoption. It also shows the limits to the assumption made by scholars like Resende-Santos about the preferability of emulation. Third, adoption-capacity theory uses the same theoretical premises to make predictions about how states will respond to innovations, the system-level distribution of responses, and the effect on international politics. Linldng together how states respond to innovations and the consequences of those responses allows for broader insights into international politics. I return to the prior literature again below to look at the distinction between adoption-capacity theory and previous research. Military Power and Major Military Innovations Defining Major Military Innovations When the production of military power changes, meaning the character and conduct of warfare change in some measurable way, it is called a military innovation; the bigger the change, the bigger the innovation. MMIs are defined for the purposes of this book as major changes in the conduct of warfare, relevant to leading military organizations, designed to increase the efficiency with which capabilities are converted to power.9 They are sometimes, but not always, closely related to changes in the technology used by military organizations. This book is concerned with new ways that military organizations generate power—and respond to innovations in the production of military power—not simply technological inventions.10 Given the possibility for misunderstandings about terminology that seem endemic to discussions of military innovations, some clarification is necessary. The "conduct of warfare, relevant to leading military organizations" refers to the way that leading states, sometimes called major or great powers, organize their militaries and plan to fight wars. The notion of changing the character of warfare refers to shifts in the core competencies of military organizations, or 9 Where scholars most often disagree is about which shifts in military power should count as innovations. The definitional distinction between innovations being designed to help states get ahead and actually helping states get ahead is crucial (Gray 2002; Krepinevich 1994). 10This move is similar to the one made by Robert Gilpin (1981,60). Approaching the question of shifts in power from the perspective of the diffusion of innovations also moves beyond the offense-defense balance paradigm, which makes sense since it is the relative balance of forces and beliefs about battles that influences behavior in militarized situations not something inherent about particular technologies. Also, the offense-defense halain r lileralure underplays the rule of politics in determining behavior (I ,'tcher 2005). 'i I lilts in the tasks that the average soldiers perform.11 Including "designed to increase the efficiency with which capabilities are converted to power" in the definition expresses the idea that all innovations are not the same and all are 110t successful. While the goal of major military innovations is to produce in-i leases in military capabilities, whether that occurs is independent of the defi-Miiion. This distinction is necessary to avoid tautology in the definition and measurement of the object of interest, the military innovation, and dependent variables like power balances and war. How Major Military Innovations Emerge (>nce we have determined what a military innovation is, we can move on to analyzing the way they debut in the international system and subsequently diffuse. Studies of the diffusion of consumer durables in the business world '.how that an incubation period often occurs between when new products ■ [iter the commercial market and when the product matures and sales take off. pi ir example, the data on the spread of color televisions and compact disc play-ris shows a large gap between the initial development of each product and •.•.hen sales increased in the broader population (Goldcr and Tellis 1997, 257; M.iliajan, Muller, and Bass 1990, 3).12 The idea of an incubation period also in,ikes sense in the military context, helping to distinguish between tcchnologi-i ,d invention and a military innovation, because the specific technical capabili-in". that represent the public face of a military innovation are often introduced II lira, if not decades, before militaries develop them further and figure out how i......c them to produce military power differently. In the case of the tank, for BXtimple, the technology was first used by the British at the Battle of Cambrai III 1917, but the mature innovation, which included the tank as well as a series ■ tl hi her technologies like radios and ah power, did not debut until the use of I ilil/krieg by the German military in World War II (Welch 1999,122). In the Intervening incubation period, the Germans were able to determine how best to Utilize the new technology and reform their operational concepts to implement combined arms warfare.13 Ibis does not exclude the possibility that other actors might innovate. It is just to point out ili ii an innovation reaches critical mass in a social system when key actors arc forced to adapt in "in' way or another.Tlie innovation well could come from below. U.ijiTv Kuhli, Donald I.chmann, and Jae Pae (1999,138,141-42) speculate that the incubation period may function dillcrently for discontinuous versus continuous product innovations. 1 Pii iniiiii'iil historian Kmcst May Muds (ierinany's success particularly impressive in comparison with I Vance's failure. I le describes I'Vance's military organization as "sclerotic" and unable to take ItlvitntllgO ill .in Otlge In the numbers oflruopn and i .inks mi llie western front in 1940 due to mas-n i Intelligence I'uIIiiich hy the Allii". (May 20110, III, -I'il ';.',), May's work also shows that simple mi i 'in mi nr. nl Inn i i mi 11 n ■ hilltlefll lil III I I.....p i minl'i in i lie founts of particular technologies 24 CHAPTER 2 DIFFUSION OF MILITARY POWER 25 Separating the incubation period from the takeoff period of a mature innovation reveals the importance of determining the demonstration point for each major military innovation. The demonstration point for each MMI occurs when the potential of its full capabilities is reasonably known in the international system through an action by a first mover, rather than the capability merely being the subject of internal exercises or debates." This is when the clock begins ticking on potential adoption by other states.15 An MMI exists as a discontinuous change based in part on the perspective of the states that view the demonstration of the innovation, not necessarily as a discontinuous change for the first-moving state.16 For example, while many German officers viewed blitzkrieg as the logical culmination of German military advances in the 1920s and 1930s, most French and British military officers saw it as a radical departure from previous patterns of warfare on its debut (Mahnken 2003,253). Each innovation has a unique life span; different innovations have relevance in international politics for different time periods. New developments, potentially even in response to a major military innovation, could trigger a new MMI in a short period of time, constraining the time period of relevance for an innovation. For instance, while the principles of warfare in the age of sail stayed relatively constant for hundreds of years, the mechanized battlefieet system that replaced the sailing navy in the late nineteenth and early twentieth centuries was itself replaced just a few decades later by carrier warfare at the onset of World War II, Nuclear weapons have maintained their hold as a critical element of global power since 1945, a period of over sixty years, while blitzkrieg warfare supplanted trench warfare after only a few decades.17 Given that innovations can matter for varying period of times and different areas of warfare such as land or sea, it makes sense to study the innovation as the unit of like tanks or planes do not always predict battlefield success. While the Germans had 76 first-line divisions and 26 second-line divisions in May 1940, at the outset of the war, the Allies had 93 first-line divisions and 30 second-line ones. Moreover, while the Germans had 2,439 tanks deployed on May 10,1940, the Allies had 3,079 (476-77). Instead, to supplement those measures it is also important to understand the way that organizations employ force—a conclusion this book draws as well. Tie point here is not that quantitative measures are not useful, just that more sophisticated models ate sometimes necessary, and it is crucial to incorporate how militaries use raw materials. 14 Many innovations debut in wartime. It is possible for an innovation to debut through the announcement of a new capability, revelations about doctrine or new technologies, or in other wavs. Warfare, however, is the most likely demonstration point, because states may save new innovations for wartime and because the rest of the international community is more likely to he paying close attention (Axelrod 1979). "This distinguishes between diffusion processes and miiiiiIi........n drvrlnp.....iu>, hy multiple states. It is theoretically possible that more than one M........uld .......11.....lite Mine innovation, meaning there is more than one fint mOVCI "'For more on ditl'ciini; pen cptiini'i, i.iv [rn In \'I'/U 1V I'm' an ullriiiiilive prinpn live, mt MhMIi '(HI! .m kilysis, rather than trying to build something like a comprehensive data set that crosses several innovations. ( )utlining Possible State Responses to Major Military Innovations ()nce a military innovation debuts, those states interested in responding have a v.tricty of options to choose from. Presume for a moment that states will generally respond to innovations in ways that they believe will maximize their foreign policy interests, however defined. The decisions of the state are influenced by a series of incentives and constraints that will shape its eventual re-i.|innse strategy. The geostrategic environment is one of the most significant factors determining the range of states interested in an innovation.18 Other factors, ranging from international norms to cultural openness to the need for Interoperability with allies and many others, may also influence interest in re-fi|iiHiding to an innovation. Vet it makes sense to separate whether or not a state needs to respond to an innovation from the content of that response. That states may attempt to maximize their security does not necessarily suggest that they will adopt an innovation. Instead, as Elman (1999,55-57) argues, while maximization could take lite form of adoption, it could also take the form of relying on alliances and/or Irving to counter an innovation through alternative means. While the question tli.il some realists and others have traditionally asked about emulation may ef-Ict lively answer why states are interested in responding to innovations, exist-mr, research has difficulty explaining what predicts the content of responses to innovations. Writing on the spread of military innovations or military technology traditionally views the strategic choices of states in response to innovation us a yes or no question—either innovations are adopted or they are not.1'-1 In reality states can pursue a much broader range of options, sometimes simultaneously, and those options may be preferable to emulation/adoption in some i in uinstances. Implicit assumptions about the fungibility of capabilities have also driven tlic threat-based story about emulation—the idea that any state can adopt an innovation if it faces a large enough threat. But if capabilities were entirely hiugihlc, it would make something like military victory logically impossible 11 both states Optimized their militaries, since each state, facing a threat to its lltrvival, would continually innovate and parry the thrusts of its opponent. The ien! work! strongly suggests otherwise. To give an extreme case, no matter how a country like Belgium optimized prior to the Second World War, it 11 inItl not dclciit (rcrmany. especially in the short- and medium-term, we can "Sit ilir wink bv ........I'nM'ii, ami Krcmlr Kanins cited above. "Sit, loi i'Kiimplf, Kmcmle SuntiM 2007. 26 CHAPTER 2 DIFFUSION OF MILITARY POWER 27 reverse the traditional story and assert that capabilities form a constraint that influences national preferences by defining the range of the possible. While the "rational shopper" approach assumes that, essentially, "where there is a will there is a way," the reality is more complicated.20 If capabilities are not entirely fungible and the optimal response to an innovation is not necessarily adoption on the part of a threatened state, something more is required to unpack the diffusion process.21 Research on domestic politics and military organizations suggests that, like the strategic environment, domestic political and military organizational factors are vital in shaping the way military organizations behave 22 Poscn's work (1984) on military innovation, for example, indicates that both the external security environment and organizational factors can play important roles. Figure 2.1 highlights the range of potential responses that states have when thinking about how to deal with a new military innovation.23 One set of possible responses involves external actions, or changes that a state can make to its foreign policy to deal with the potential repercussions of another state adopting a military innovation.24 A state could decide that the implications of the innovation make the achievement of national foreign policy goals no longer possible, requiring a lessening of overall foreign policy as-sertiveness or even a shift toward neutrality. This could lead to what Paul Schroeder (1994,117) describes as "hiding" in international politics or potentially "transcending": pushing for international institutions or other means to deal with a situation in which a state no longer has the relative power to protect itself. Another possible external response is to attempt to mitigate the costs of nonadoption by allying with a likely adopter (Walt 1987).25 One option is balancing, creating or joining an alliance against the state that initially debuts the military innovation with another state or a group of states likely to successfully adopt or counter the innovation. The converse of balancing is bandwagoning, allying with the pioneering state that generated the innovation. Alliances boost the relative power of the involved states, because these states can work together 20"Rational shopper" is taken from Goldman and Ross (2003,380). 21 John Nagl's study (2005,219) of U.S. and British counterinsurgency strategy in Malays, and Vietnam comes to a similar conclusion, arguing that the optimization of U.S. forces for conven tional war after World War II made it difficult to adapt to countcrinsurgency requirements m Vietnam. 22 See, for example, the work by Evangelista, Zisk, and Mahnkcn cited above. ^Tliis figure is not meant to suggest the paths are all exclusive, just Id show the diflcrcnl |m i bilitics. For a realist take, sec Rescndc-Santos 2007, 69, "This rubric assumes that a state has some lime I.....I.ipl llir results might he different il a state faces a severe,short-term threat, 2*This is similar In Powell's irgUfflOM (1999| l'/ti 79) thai ulalfh lomctlmci may turn to alll antes as a response strategy when |«ili ntlal mil....... \« ........in leases in iiiiIk.iiv i.ip.ilull Hi", t )n Imw stales make jIIi.iiii c i linn i < i' .11 •,. i null In i i. men 1990, Attempt to balance with likely adopter against first mover Attempt to bandwagon with first mover I I Ml 1. Potential state responses to major military innovations. 1 • lit) more than they could do alone.36 Alliances may also help a state more ' ipltll) gain access to the knowledge necessary to implement the innovation, • Hi" i Ihrough buying time to build the capacity to adopt or through direct as- ......' from the first mover (Goldman and Ross 2003,375-79).27 For exam- |ih , in 1945, when the British Navy reengaged the Pacific theater of World II, 1(8 alliance with the United States facilitated greater power projection • di ncvcral mechanisms, including technology transfers.The British Navy ii id ils alliance with the United States to learn more about adopting car-Mi > tv,u litre,2" I" Itltlllion to external response strategics, states may also pursue internal •■"hi m i lunges. One option is to try to adopt part or all of the innovation, ......I', llie innovation diffuses from country A to country B. Partial adoption '• M Involved adopting technological or operational but not organizational 1 ol .in innovation, since it is generally much less disruptive to adopt I 11 Ulrttt'liil 10 mi anonymous reviewer for making this point clearer. 1 ir| limenl here makes sense even if the halancing/bandwagoning terminology is flawed, as HIVI nbsi i vrd, lin anse states generally view themselves as balancing even if an external i Q ii would lUggetl Otherwise (Schroeder 1994,119). ...........I ibis example, see chapter 3. Essentially, alliances help boost the power of states ll..... I" combine their resources, transfer technology/knowledge, and free up re- ■' 1 ItC might hive iii spend elsewhere, among other mechanisms. For more on alli-Lll I.I Otlg, and Mitchell 2000; heeds 2003; Levy 1981; Morgan and Palmer 2003; I ' ii, 28 CHAPTER 2 DIFFUSION OF MILITARY POWER 29 technologies than to change the way an organization thinks about employing military force. Some states may lack the capacity to fully adopt in the short-term but will build or buy some of the technological components. This introduces a time-delay element. Alternatively, some states might have the financial means to fully adopt and recognize that adoption makes strategic sense, but be unable to adopt for organizational reasons. This type of state might think that adoption would trigger domestic revolts or upset key military social hierarchies, outweighing the military benefits from adopting and making effective implementation unlikely. For example, in response to defeats at Ulm and Austerlitz at the hands of Napoleons armies, the Austro-Hungarian Army reorganized itself into divisions and corps, adopting the French organizational form. Despite the best efforts of Archduke Charles, however, fears of internal instability from an armed populace caused Austro-Hungarian emperor Francis I to veto the more widespread reforms to the recruiting system required to truly adopt mass mobilization, the crux of Napoleonic warfare (Herrera and Mahnken 2003,214-15). Additionally, some states have the raw financial and organizational capacity to adopt an innovation, but do not do so because the financial and nonfinan-cial (frequently domestic political) costs of adoption are viewed as higher than those of nonadoption, especially when an external alliance strategy is possible. In the same Napoleonic warfare case, despite possessing the capacity to adopt, the British Army instead pursued an alliance strategy. Army and political leaders thought that the probability of extensive British land engagements on the continent was sufficiently low that the domestic costs, in the form of public upheaval, from shifting to mass mobilization through conscription outweighed any benefits. The existence of a viable alliance strategy made paying the costs of adoption less attractive to Great Britain (Herrera and Mahnken 2003,216-17). Another internal response strategy is countering an innovation, also some times described as offsetting (Goldman and Ross 2003, 288-89). Countering is defined for the purposes of this book as an internal military response thai excludes adopting the innovation, yet can range from trying to neutralize its impact with inexpensive tactics drawn from existing forces and operational plans to the search for another new military innovation to counter the lira one. An example of countering through existing military means was the cam ouflage practices used by the Serbian military in response to the U.S.-led nil' campaign over Kosovo in 1999 (Thomas 2000,13-15). A case of a countering strategy that became an innovation in its own right was the devclopmenl nl the trace italienne, the squat and broad-angled bastion fortress. It came aboui ,v, a response to increasing cannon quality that gave attacking armies the ability to batter down the tall, thin walls of most fortresses in early modern Kuiopi (Parker 1996,9-11). i 'I'innnstration Hiuiinunicates H liimiation about MH|tiiri!inentsfbr adoption .....v.tration i m ijor military min ivation I '.iiiimstration ' .11111111nicat.es iiMlion about i mllilo alliance partners and i alii'f foreign |m1i( y options Military Organization evaluates financial intensity and organizational capital requirements in comparison with capabilities Military informs Executive of adoption and countering possibilities Consideration of potential alternatives by Executive choice of dominant strategy and other policy paths Executive Branch evaluates external responses based on alliances or substantially changing foreign policy m 1 .'„"Ideal" state response to demonstration of major military innovation. In general, revelation of an "average" (MMI) triggers two simultaneous pro-iii an "ideal" state.29 The military evaluates the requirements for adopt-iliľ innovation in relation to its current and potential capabilities, along llll I lie possibility of a countering strategy, while the executive determines ■ i" putľiilial utility of external policy solutions like alliances or a shift to neu-illll In states based on civilian control of the military, the military reports I In I he executive, who then weighs the relative costs and benefits, proba-illi continuing advice from military leaders, of the different strategic ihihlies. In the case of an average middle power with an advanced indus-'I It r .nid medium-sized economy, the most likely choices are partial i nim, especially of some technological components, but a larger emphasis m alliance-based response. The character of the alliance-based response i mli..... ihe relationships between the responding state, the first mover, 1 illiri potential adopters. Given the current international security envi-h nl, il ihe responding state is Canada and the first mover is the United ■jimi u, liatulwagoning is likely. If the responding state is Japan and the first ■ i In (lie People's Republic of China, balancing is probably more likely, m ' 1 j'.iaphically depicts an idealized version of this response process, tilvpn 30This assumes a somewhat "normal" distribution of capabilities in the international system al the time the innovation is demonstrated. While capabilities in the international system are rawly normally distributed in a statistical sense, there is usually a sufficiently broad range of intereitl and capabilities such that the assumption is useful for understanding the effect of an innovation. 31 This makes sense when thinking about the spread of military power because military organ i.m tions are not infinitely pliable, especially in the short-term. So while the theory deals with i.tr.llrgli choices, it does not deal with them in a game theoretical fashion that allows for mixed cqulllbl I urns. If adoption-capacity theory is incorrect, military organizations will end up being mU( 11 "..... elastic than predicted, suggesting that future research efforts might more fruitfully fin iin nil llii issue differently. |».u licular resource mobilization requirements involved in attempting to I .i major military innovation.32 The higher the cost per unit of the hard-*.m .issociated with an innovation and the more the underlying technologies ■ lusively military oriented, the higher the level of financial intensity i ii'il to adopt the innovation. Ilir cost per unit of technologies is important. Lower unit costs can allow .......«' experimentation by a military organization, making it easier to test i quipment and determine its feasibility for full deployment. For example, 'IV ' in production methods made possible by the Industrial Revolution ■ I lower the cost per unit of the rifle, which had existed for over two hun-iltul years, but never systematically made its way on to the battlefield because i 11« previously high unit costs. Alternatively, the production of one B-2 I'i'i, ci isting between one and two billion dollars per plane on average, is so iiiitiusly expensive that experimentation is not possible on the same scale ' ll i 'dO'ia; U.S. Air Force 2006a).33 He underlying technology driving a military innovation can take many bin it is usually either an essentially commercial technology, meaning He businesses have economic, nondefense incentives to develop the tech-\, i if an essentially military technology, meaning it was invented for mili- ii .iM nis and will arouse little initial interest from businesses outside of de-• imliactors. Underlying commercial technologies will generally require i nei capital investments for a military organization than underlying mili- I ■ I.....logics, because market competition creates incentives for private Hi mi In pay some of the product development costs.34 For instance, the can-li | •!"ilined during the early modern European period wrere solely military iiiienis, requiring a higher relative capital investment than something like ' "li",nI in the mid- to late nineteenth century, which nondefense com-■ i 'I uileiesls bad an incentive to build.35 i "ill intensity is somewhat related to the economics literature on "capital intensity," 11 ii 11 in In 11 ii' level ol financial investment necessary to accomplish a task in comparison with 'I......atcrial investment in the form of labor. For some recent examples, see Arai 2003; iii H......if and Machin 1998; Gartzke 2001; Sattinger 2004. In contrast, financial intensity I | Mill) .i I ii ml Mir overall investment level. While it is possible in the military power area that ......' i......rations that require more labor and less capital to implement, and vice versa, ll ......iml\ ii'li-rs In the structure of capital investments. Nevertheless, it is certainly true iHMn\.iii"ir. requiring high levels of financial intensity to implement generally require higher Ml In. Ii, potentially necessitating a trade-off with labor given limited budgets. II up. "I experimentation that is more likely to result when the cost per unit is low may ii ting itplllovcr effects on organizational capital as well. Experiments can help acclimate ' u 11' 111 i I < > i ii-vv lech oologies and organizational practices. I".......* 1»' nsii...... such .i'. sis r<\y or control of production why the military would want to 1 l' ""I pii'ilm i''......<■ nl'its own weapons.'Ihe relative cost, however, will be higher. I' 1 ■..... tli.it ' armonj originally developed from editing procedures used for bells, but unlike 1 '' 1 I nun invented, iheiv w.u in. ir.i.ui in prmlui c i .unions except for military purposes. 32 CHAPTER 2 The financial intensity level of each innovation is derived using budget data to determine the cost per unit of the technology in comparison with the cost per unit of the prior dominant technologies, and using available data on production processes to determine whether the underlying technology is mostly commercial or mostly military. Depending on the case, the financial intensity capabilities of each potential adopter are then measured by comparing factors like the unit cost of the technology for the first mover to the defense budget of potential adopters, the gross domestic product (GDP) and defense spending of the first mover to that of potential adopters, and the relative industrial capabilities of the first mover and each potential adopter, taken from national capabilities data.36 Table 2.1 shows how these factors determine the relative level of financial intensity required for adoption. MMIs that require high levels of financial intensity for implementation, meaning the cost per unit is high and the underlying technological basis is military, not civilian, are harder to adopt than those requiring low levels of financial intensity. Financial Intensity Hypothesis: The greater the financial intensity required to implement the innovation, the slower the spread of the innovation at the system level and the lower the probability that a state will attempt to adopt the innovation. Organizational Capital Just as the financial requirements for adopting an innovation can vary, the organizational requirements for adopting an innovation can also vary. For example, in the late nineteenth and early twentieth centuries the British Navy confronted two large innovations in naval warfare. The first, battlefleet warfare, was the culmination of naval industrialization. Epitomized by the Dreadnought, the first all-big-gun battleship, battlefleet warfare required changes in the education, recruitment, and training of naval personnel. Instead of training soldiers to climb the rigging, for example, soldiers had to learn advanced math to plot gun trajectories. And instead of mastering sails, they had to take apart and put together engines. But the core of the innovation involved doing what the British Navy had always done: bringing its guns to bear in battle to destroy the capital ships of the enemy. This made the innovation organizationally easier to implement. The second innovation, submarine warfare, represented a more disruptive change. The potential shift in naval power away from the gun and capital ship, and toward the torpedo and submarine, contradicted centuries of thinking about the production of naval power and specifically challenged the core competency that had defined British naval supremacy: gun ''Data often drawn from the Correlates of'War Project Malrriiil I 'im.i1iiltlir'i data nrl t.'OIIM DIFFUSION OF MILITARY POWER 33 Table 2.1 Financial Intensity Drivers I '>!,Inlying basis of tht technology Civilian Military Cost per unit Low Low financial intensity Medium financial intensity High Medium financial intensity High financial intensity Pflttlcs between capital ships. The innovation required not just changes in edu- ■ Hnni, recruitment, and training but also wholesale shifts in force structure 11 id plans for the use of force. This made it much more difficult to implement, 11MI Is one of the reasons that Admiral John Fisher's attempt to transition the lltiliiib Navy away from the all-big-gun battleship and toward flotilla defense felled (Lambert 1995a, 1999; Sumida 1995). IIiiis, different military innovations can place different demands on organi-idons as far as the degree and type of transformation required. Organizational i iipttal is an intangible asset that allows organizations to change in response to pen eived shifts in the underlying environment. For militaries, organizational I tplcal represents a virtual stockpile of change assets needed to respond to ■ lltnges in the character of warfare. For those who study firms, one simplistic M) to measure organizational capital is to write an equation that subtracts the Vtilut' of the physical assets held by a firm from the value of the company as l>i ti eived by investors, measurable in its stock price. The difference represents I lie intangible value of the company, or the value not captured by its physical mots,17 [flion Cummins (2004, 4), from the Division of Research and Statistics in ilii Federal Reserve System Board of Governors, provides one useful way to ilrline organizational capital in the context of information technology: 'in what exactly is organizational capital? As a purely mechanical matter, I define organizational capital as an adjustment cost from IT investment, defined as the difference between the value of installed IT and that of niitri'.i.illi'd IT. Suppose a company purchases database software. By itself, 11,itabasc software does not generate any value. At a minimum, the soft-u,nr must be combined with a database and, perhaps, a sales force. Orga-iii/.itionul capital defines how the database is used and, consequendy, how i >ll ware investment creates value. ilu > an Include thing! like brand reputation, financial accounting issues, perceived knowl- ■ 'i • !'•> ii U'l.ili", lii Inline iiiiiov.uioii'i, li.....hi , ,i|iil.il in general, and other issues (Bond and ...............MKID) Sir.,1,.,, Lev and [()*■'. 34 CHAPTER2 DIFFUSION OF MILITARY POWER 35 New military technology, like the new database software described by Cummins, does not produce value on its own. It is only in combination with organizational methods that technology has value for producing military force. In this way, organizational capital is the ?zo«tcchnological aspect of force generation, demonstrated through doctrine, education, and training. Organizations with a high degree of organizational capital are much more able to take advantage of new innovations and transform themselves successfully for the future than organizations with a low degree of organizational capital. The preceding discussion raises a measurement problem: if organizational capital consists of relatively intangible organizational assets, how should these assets be measured? The demands on military organizations in periods of uncertainty vary in ways potentially similar to variance in the demands on firms during periods of rapid change. Lynnc Zucker and Michael Darby posit that successful firms take advantage of organizational assets and a culture of innovation to stay ahead during periods of uncertainty. Firms, even leading ones, are not necessarily immutable. For example, the fact that some leading pharmaceutical firms radically shifted their identities in response to the biotechnology revolution while others failed demonstrates that different firms respond in different ways to external demands (Zucker and Darby 1996a, 960, 971; Zucker and Darby 1996b). Organizational capital is a critical element of firms' success, helping businesses survive during periods of change by giving them the institutional capacity to shift practices without being blocked by bureaucratic obstacles. Because it is difficult to copy business processes, as opposed to specific technologies, organizational assets are difficult to duplicate. Darby and Zucker find a ten- to fifteen-year gap between the beginning of the biotechnology revolution and widespread knowledge of the practices necessary to take advantage of the revolution for business purposes. In between, tacit knowledge of how to conduct specific research was held by key scientists working for a few firms and others in their immediate vicinity, making the knowledge "naturally excludable" from outside firms (Darby and Zucker 2003, 2). These human resources then became the basis by which an existing incumbent firm stayed ahead when further change appeared likely (Zucker and Darby 1996a, 12709; Zucker and Darby 1996b, 965-67). Even detailed case studies on business processes have frequently been insufficient to enable copying by other firms. Erik Brynjolfsson, Lorin Hitt, and Shinkyu Yang (2002,144) explain the difficulties involved: Intel, for example, has adopted a "copy exactly" philosophy for any chip fabrication plant built after the first plant in each generation. Wholesale replication of even seemingly insignificant details lias proved more reliable than trying to understand which characteristics really matter. Going from the plant level to (he linn level only eomplii ales the imitator's lusk, A useful set of alternative operationalizable markers from this business innovation literature can help us measure both the level of organizational capital I k isscssed by a military organization and the baseline level of organizational i ttpital required to implement a given MMI. Three measures of the level of mganizational capital possessed by a military organization are the critical task d k us of an organization, experimentation resources, and organizational age. CRITICAL TASK FOCUS I Vpieally measured by statements of intent and planning documents, the criti-■ .il task of an organization is that which the organization views as its main "cmd"—what it seeks to achieve. While not always referred to as a critical task in literature on organizational effectiveness and the integration of innovations, I lie notion of a critical task captures the central operating principle of organiza-Iurns. Having a critical task helps organizations frame and justify their actions, pioviding a central theme for motivating workers. As business innovation II liolar Clayton Christensen (1997,168) writes, "Over time, however, the locus nl I he organization's capabilities shifts toward its processes and values." Critical l.e.k focus is the extent to which the means by which an organization achieves il'. goals are conflated with the goals themselves (Wilson 1989, 25-26). Many successful firms come to define their business processes and values based on 11 ic strategies that initially led to success. The embedding of a particular vision nl 'a critical task within the organizational architecture then shifts from where ,i business allocates resources to the metrics used to make decisions to the pro-i esses used to design metrics. One example is the failure of the Digital Equip-HHiii Corporation (DEC) in the shift from minicomputers to the personal * omputer.The organizational process at DEC prioritized reinforcing the high-mnrgirt minicomputer business, even though it also had high overheads. Since i lir emerging personal computer market looked miniscule in comparison, with low overheads and profit margins, DEC's organizational decision making l>lunled it from seeing the shift in the market. As Christensen (1997,169-71) i [escribes, eventually DEC failed, not because it lacked the resources to compete in (he personal computer arena, but because the business processes designed to maximize profits in the minicomputer arena embedded an inappropriate critical i.il. al the core of the business. In general, businesses with narrow critical tasks associate particular ways of doing business with the goal of the business. Businesses with broader critical talks are those that delink the goal of the organization from the means used hi achieve that goal. Kim Clark shows how firms often use their experiences in build specific types of knowledge and skill. They identify key problems . < 11 > I solve ihem through the creation of efficient business practices. Still, when i Kogenous technological slim I.-, incur or customer preferences change overnight, siiivessful businesses i mi liud themselves hamstrung by the organization!] < Inn.lie lliev pieviciiislv > idled H 'l.uk I'WO t 'ompetenee destroying 36 CHAPTER2 DIFFUSION OF MILITARY POWER 37 innovations, like the tlrvoloptnnil of private airlines or mass production processes for cement, can dh„idv,inl,ip,c linns that previously held large segments of market share (Tuslmiaii and Anderson 1986). The critical task locus liir ,i firm thus becomes myopic at times not because a firm is poorly managed hut precisely because smart managers optimize their entire organization, from production processes to decision-making hierarchies, to grow their market share in it particular business area (Christensen 1997,98). Firms frequently create strategic group identities to reduce uncertainty through the use of predictable metrics and planning processes. Especially when these identities form around production processes or specific ways of achieving market share, they make it more difficult for firms to respond to radical innovations in core practices (Henderson and Clark 1990). In contrast, the broader the critical task focus—meaning the less specifically a firm links its organizational Identity to particular production methods—the easier it is to incorporate new ways of achieving organizational goals—that is, adopting an innovation—without encountering too many bureaucratic stumbling blocks to implement the change. The more specifically a military organization defines its critical task, the harder it should be for the military to adopt an innovation. Entrenched interests within the organization will be more likely to rebel on the grounds that a proposed innovation is outside the scope of acceptable activities. An organization with a broader critical task focus will find it easier to justify, within the organization, how the innovation fits within the regular "actions" of the organization. One often-cited example of differences in critical task focus comes from the U.S. Marines and U.S. Army. The critical task of the Marines is described broadly as being a "warrior," and the Marines envision all Marines as soldiers with rifles first rather than identifying each Marine by their specific task. This has made the Marines more receptive to a wider variety of core operational concepts including amphibious warfare, trench warfare, counterinsurgency, and urban warfare, since they all involve the activities of a warrior. The army, in contrast, as James Q^Wilson (1989, 43-44) has pointed out, has traditionally defined its critical task as massing firepower to win conventional wars, making it more difficult to justify the incorporation of new war-fighting methods like counterinsurgency (Krepinevich 1986, 5). The army preference for operations emphasizing overwhelming firepower also likely influenced the army's decision to use casualties inflicted on the enemy as its dominant indicator, or key metric of success, during the Vietnam War (Gartner 1997).33 The army's operational focus on overwhelming firepower drove its strategy, rather than the other way around. This is related to work by John Nagl (2005, 216-17), who 58 Beginning in 1965, General William Westmoreland based I IK, mrategy mound «tweli and ■ destroy missions designed lo use overwhelming firepower In gem hilf i um lid I Ii'a (( uiiluri 1997, 1,10 '1 D.'lhis eh|ilaiialii»i is very much in line wilh udiiplu......| <>i < lly ll.....i uses the U.S. Army in the Vietnam War to argue that the way organizations view themselves and their critical mission can hinder learning even in the face nl strong evidence suggesting current methods are failing. EXPERIMENTATION 'the prescriptions for firm success outlined by Christensen use the organizational commitment to experimentation as an accurate way to measure organiza-lional capital. Christensen contends that the firms best prepared for disruptive innovations are those with ongoing experimentation efforts and/or spin-off companies that can avoid substantial losses if the spin-off fails. Spin-offs with loose ties to the established company, because they measure success in relation to their new competitors, rather than the rest of the company, are best posi-linited to highly value the small profit margins likely to accrue at the beginning of a transformative period (Christensen 1997,176-77).39 [ Iigh overall investments in experimentation may make militaries more re-■ eptive to innovations; experiments demonstrate an institutional willingness to think outside the box, meaning the organization builds in the capacity to integrate innovations. The George W. Bush-era Office of Force Transformation in the U.S. Defense Department was one example of a subunit focused on experimentation. Within the Marine Corps, the Marine Corps Warfighting I ,nb is another locus of experimentation. In fact, while a key finding in the business innovation literature is the relationship between corporate "skunk-works" and success in the face of innovation, the term itself is derived from the work of a defense contractor. When Lockheed designed the P-80 Shooting Star jet plane during World War II under high secrecy, a circus tent with a top-secret incubator was positioned next to a plastics factory. Rogers (2003, 149) writes, "The strong smells that wafted into the tent made the Lockheed R&D workers think of the foul-smelling 'Skunk Works' factory in Al Capp's I ,\'\ Abner comic strip. The name stuck and came to be generalized to similar high-priority units that have been created by various companies since."40 ORGANIZATIONAL AGE (>ionizations generally acquire some degree of rigidity over time, independent of domestic politics and overall political centralization, throwing up barriers to transformation that undermine organizational capital levels. While business innovation research demonstrates that firm identities are not entirely ii m mi table, the conventional wisdom—that it is difficult for large organizations ™ I lei ause ihe market size is smaller for emerging component markets, the profit margins are iiltcu smaller. * Experimentation is a better proxy than research and development. Organizations may pump 11tilM nl' resources into research ami development, but if they do not invest in the right areas, they inn mil up promoting iiurcmenliil improvements to ihe last great thing, rather than the next i, n Hinir [( 'liil'.li'll'.en 199/) 38 CHAPTER2 DIFFUSION OF MILITARY POWER 39 to transform to face new business challenges, especially when the challenge is disruptive to their fundamental business model—is the conventional wisdom for a reason. Olsons research shows that the age of an organization often predicts its productivity level.4' Based on a number of case studies and a wide variety of data from studies of businesses to studies of cross-national post-World War II economic growth, Olson's work (1982, 76-77) demonstrates that older organizations many times become more stagnant, making them ill equipped to deal with transformational change.42 Olson (1982, 31) argues that collective action incentives decrease as group size increases, because the benefits from collective action are more diffused, making it harder for larger groups to act in pursuit of common interests. Attempts to solve these problems by creating subgroups to promote change often backfire, fueling the development of bureaucratic sublayers of inefficiency by reformers and corresponding interest groups that mobilize to block reform, requiring special interest groups to help people navigate the bureaucracy. The result is an increasing number of veto points in the overall organization with the capacity to block change. Interest groups that have cartelized for the purposes of distributing benefits will have incentives to block resource redistribution, especially to new groups forming in response to changing conditions. Ibis can delay the adoption of new technologies and the more efficient distribution of resources (Olson 1982, 64-65). The theoretical extension to the military realm is obvious. Well-defined military service cultures often militate against change, because service groups will defend existing distributional bargains about the spending of defense dollars against attempts to reallocate. The basic prediction is that as organizations age, the level of organizational capital declines.43 These three measures of organizational capital are potentially operational-ized by using internal documentation (when available) to code for the critical task focus of militaries, budget data to code for the commitment to experimentation, and the time gap since a regime change or a major war in which a *" For some critiques of Olson, see Coates and Heckelman 2003; McLean 2000; Unger and Van Waarden 1999. 42 For example, Olson argues that the destruction of Germany and Japan during World War II allowed the two countries to start anew economically, resetting their organizational ages. This made it easier for them to take advantage of innovations in the post-World War II period (Olson 1982,76-77). "There are additional indicators likely unrelated to the transformational capacity of organizations but that may matter, including the centralization of authority, receptivity to information, civil-military relations, and organizational size (as opposed to organizational age, which is more about whether the bureaucracy is bloated than its absolute size, though they are MinuTunes related). The importance of centralization, civil-military relations, and iuli nutation in probably subsidiary questions to the issues of erilictl tasks and oiganiAilioii.il ngi , in i tin pn .In linns are relatively similar (i.e., the expansion of veto point. a. m, .mm i...... i . Iiulii ihi ..... ly|>PK <>f dysliini linns as those piedli led In a i ivil unlit.u V nqiliiiilmnl I In qm ........I ■ ■<, mm • n mm nil iii/r hlo-lv depend', nit the i li.n.ti let ni lite lullll.u I niin.i.iii....... I III I I il mi iilln I lui lulu ••late lost, whichever occurred more recently, to code for organizational age.44 When evaluating the organizational capital requirements for a given innovation, I determine the level of organizational capital possessed by the first mover(s) for each innovation, which is a good proxy for the requirement since I here is sometimes uncertainty about the exact measurement of the three indi-i dtors. Combined with general measurements, a given innovation is assessed .i'. requiring a certain amount of organizational capital to adopt.45 Then, the organizational capital required to adopt an innovation is compared with the organizational capital possessed by individual states to determine the propen-■•ity of each state to adopt the innovation. Organizational Capital Hypothesis: The greater the organizational capital required to implement the innovation, the slower the spread of the innovation at the system level, and the lower the probability that a state will attempt to adopt the innovation. It is important to note that this understanding of organizational capital is not without its limitations. Studying organizational change is difficult, and the wide variety of factors that determine the way organizations make decisions makes it impossible to encompass all. of them in any particular theory. The I luce factors described above—critical task focus, experimentation, and organizational age—are designed to proxy for the type of characteristics likely to exist in organizations more or less able to adopt new military innovations. Predicted System-Level Diffusion Pattern While the particular international context in which an innovation arises matins a jjjreat deal, the best general predictors of the way it is likely to diffuse throughout the international system arc the relative levels of financial inten-ilv anil organizational capital required to adopt the innovation. Combining II if predictions about financial intensity and organizational capital highlights ilic way that different requirements for adopting an innovation should influ-i m i' the way it spreads throughout the international system. "When possible in the empirical chapters, the regime change data comes from Polity while the u.u Ins', data comes from the COW Project (Marshall and Jaggers 2002; Sarkees 2000). Major .. it . arc measured as wars where the demand involved regime change (Horowitz, Simpson, and Blum 2009). An alternative measure is wars that arc followed by regime change, though this con-ll it. . ilie two measures. 1 I king I he lirsf mover as a model for the required level of organizational capital is imperfect, Inil it dues Hiulextnali/c the adoption requirement-.. 11 is possible that over time, the level of orga- ui/ iiiuii.il i.i11■ i.11 required in adopt tome Innovation! will decline.This is an important limitation 1'iiln In .1 iiiihi'i test Yet II i. only nut-nfi.ivci.il u.iv nl measuring organizational capital. If the In hi mover let It-ads In an tXMtOftllOQ "I iln idoptloil icqiiticiiieiils, it will show up in the em- I M n 11 i lmpii i ., -.mi e lite adoption Capadt) il.....) will pitdli I u Hindi '.mallei' and iinne limited upn .id ul tin n nun ,il n hi tli in wli.i I a. In 111 > in . UK 40 CIIAITKR2 DIFFUSION OF MILITARY POWER 41 Table 2.2 S\ i, in level I illusion Predictions Level of organizational capita/ required to implement major military innovation Low High Level of financial intensity required to implement major military innovation Low High Rate/extent fast Rate/extent medium Kale/extent medium Rate/extent slow State-Level Predictions Moving from system-level predictions to individual states requires acknowledging additional complexity. One key factor is whether knowledge concerning how to adopt has diffused to a particular state. While the adoption-capacity requirements can predict how an innovation will spread in general, forecasting the decisions of specific states requires understanding their particular motivations as well. One could argue that the system-level predictions, when applied to individual states, are tautological because they predict that those states with the capacity to adopt will do so. Yet threats and internal politics may lead states toward choices that maximize their utility for preference functions other than making the most of short-term military power. STRATEGIC CHOICES In general, states responding to a major military innovation are likely to choose a dominant response strategy that consumes the bulk of their time and effort from the range of possible choices outlined above—though they may simultaneously pursue a few different paths.*' The dominant strategy is an attempt by the state to maximize its utility in response to a given innovation, though states can and do make bad choices. The payoff for adopting a particular strategy is a function of state certainty about the substance of the innovation, depending on how effectively information about the innovation has spread, the interest a state has in adopting given its geopolitical position and domestic political configuration, and its adoption capacity, its ability to adopt the innovation based on the combination of its financial and organizational resources. While other factors also influence responses to innovations, these factors should generally be the most critical. As Posen's work (1984, 233,239-40) on the intcrwar period in Europe suggests, large-scale geopolitical threats could increase the urgency with which ""In an idealized state, while initially states will have debates over the appropriate response to an innovation, with advocates existing for a range of possible responses, evenlu.ilh one emerges lor a variety ol reasons as the dominant response even if elements of otliein i Ofll mm CO 1 tllti Al tentatively,lee Kcnciulc-Snnti.....ilinent internationally that influences allies and potential adversaries, and might alwi pint he ihr Irani bud op tinn in a comparative sense, given the relative costs cil'thr vatmu . ......I y \i- t ill' in it im . ■"This builds on Gilpin 1981,59-62. when some states rapidly rise while others decline, are critical times in global politics because they may be the dry timber through which major fires, large-scale wars, can start.™ Analyzing how the rise and decline of states influences the international security environment has been an important subject in international relations for decades (Copeland 2000; Gilpin 1981; Kugler and Lemke 1996; Organski and Kugler 1980; Waltz 1979).51 Understanding the difliision pattern of the dominant innovation of a time period and the match of adoption requirements to the capabilities of states can help us more accurately explain power transitions. In contrast to his general contention about the inevitable decline of existing powers, Robert Gilpin (1981, 161-62) writes, for example, that existing powers can reinvent themselves and stay on top, pointing to Chinese and British developments over time. He does not explain, though, why it is that they can reinvent themselves when facing some challenges, but face "institutional rigidities" (189) at other points that hasten their decline. So what explains their eventual decline? Most narrowly, adoption-capacity theory can explain the mechanism by which these transitions sometimes occur. Military innovation diffusion is part of the causal process governing power transitions.52 The predictions made by adoption-capacity theory about the way adoption requirements and state responses will influence the spread of a military innovation also lead to predictions concerning the importance of innovation diffusion for the international security environment. Further, differentiating between financial and organizational requirements for adopting innovations and the implications for international politics can also explain the process by which new states take the lead for Gilpin, or how the costs of war shift for Powell. More broadly, at the system level, adoption-capacity theory explains both why some shifts in relative power occur and how. Some new major military innovations constitute disruptions in international politics that can generate large power disparities. First movers that innovate and generate new ways of producing military power can gain significant advantages; the exploitation of these advantages then can usher in power transitions, exposing status quo powers that can become overmatched paper tigers. Rising powers that become first movers are especially likely to experience large gains in relative power, hirst-moving existing powers or existing powers that can rapidly emulate or '"''flic phrase power transition is used to describe increases and decreases in relative power, not the power transition "school" (Organski and Kugler 1980). Subsequent discussions of shifts in power should hold whether or not traditional power transition dynamics or Powell's argument (1999, 85, 199) about the divergence between the distribution of power and the distribution of benefits holds. 1,1 In this section, the trims enisling and rising power refer strictly to relative capabilities in the period prior to the intiniliu lion ol tlit' innovation, not intentions. 1J h'or an example ol Wink lli II 'I......lilt uss the allcel of innovation difliision on power, sec i loldnwn and An.h. | I 44 CHAPTER 2 DIFFUSION OF MILITARY POWER 45 otherwise adapt to an MMI can make relative power gains, or help stave off disadvantageous power shifts. Innovations more generally trigger strategic responses on the part of the rest of the international system, with success and failure in the world of the new innovation determined by the match between the financial and organizational requirements for adoption and the financial and organizational capabilities of interested states, modified by their alliance postures and other options. New major military innovations sometimes present major powers with challenges they simply cannot meet, if they lack the financial or organizational capacity to adopt. These shifts in relative power and periods of instability with declining powers do not occur because a state makes a mistake. Rather, sometimes the financial and organizational requirements for the next major military innovation are just beyond the capability of a major power to implement. Adoption-capacity theory is a protorationalist explanation for change in response to innovation; rather than looking to pathologies or other reasons why states fall by the wayside in the international power game, it is simpler to look at what it would have taken for a state to "keep up," and whether the state was simply unable to do so. Especially when preexisting powers desperately want to stay on top, they may end up with an attribution bias that causes them to exaggerate their capabilities. It is in these situations that the most dangerous power transitions occur. These faltering states are the most likely to engage in foolhardy military strategies that make escalation more probable. Alternatively, the recognition that their relative military power will only decline for the foreseeable future could create incentives for risky military gambles as leaders enter a "use it or lose it" mind-set. Fear of structural military decline could also cause a general heightening of tension that leads to war as a declining power covers for insecurity about their capabilities with greater levels of belligerence at the same time that potential adversaries are gaining military confidence due to successful adoption of the new innovation. Strong incentives might exist for national leaders to try to gamble for resurrection with regard to the reputation of the country as a global power, especially given the potential personal cost to the leader if the leader is blamed for overseeing national decline (Goemans 2000).53 While this possibility exists, it is also extremely difficult to test since it requires not only knowing what a country did and why but getting detailed information on the risk perceptions of leaders as well. Rather than decline being inevitable, as it is for those that presume the diffusion of knowledge will inevitably harm system leaders, shifts in power and their timing are better understood as a function of the adoption requirements of the key military innovations of a period.54 Not all innovations will benefit rising "On the Other hand, the risk of removal from office if the stale gels involved in a w.u uul loses could iiidin c i .tutiihi. u Ronald RogOWlkl (l'>m, 735) makes tllll point about < !ll|........1 ''" ......" nl r. uill powers and hurt established ones. An international configuration that might seem ripe for change could hang on for decades or longer because the dominant form of military force favors the existing power. It is only the emergence of a new innovation requiring a high level of organizational capital to adopt that drives a nail through the coffin of the existing power. Or as Ronald Rogowski (1983, 735) hints, it is necessary to combine the organizational insights of ()lson with Gilpin's arguments about shifts in relative power, which adoption-capacity theory does. This allows us to predict not only that a shift in the distribution of power is likely but also the timing and character of that transition. So how do different adoption requirements for innovations, when summed over the international system, actually translate into changes in how an innovation will influence the international security environment? There are two key ways that the diffusion of military power matters for the overall international security environment: through the implications of the system-level distribu-lion of responses, and from first-mover advantages and power asymmetries for particular states. The former is more about the relative balance of power between states, while the latter is more about the implications for the relative timing, intensity, and geographic scope of warfare. diffusion and system-level consequences Returning again to the business innovation literature, Christensen highlights I he way that differences in the adoption requirements for innovations in the business world influence the probability of firm success and failure. He differentiates between sustaining innovations, or those that improve current busi-i less practices, and disruptive ones, or those that make following prior business practices likely to lead to firm failure instead of success (Christensen 1997, xiv-xv).5S Christensen studies the disk drive industry and its transition through several different stages of innovations. He finds that in every case of a sustaining' innovation, like thin-film disks and heads, established firms like IBM led the way. Yet disruptive changes almost always caused major transitions in the industry. By the time the new technology became clearly more efficient than (he old one and generated new markets of sufficient size for big firms to become interested, it was too late for entrance by most of the established firms (13-15). Essentially, leading firms failed when confronted with innovations I hat required a new way of doing business, but succeeded when confronted with innovations that enhanced the previous way they did business. That insight forms the basis of the relevant predictions about MMIs.56 Fo-i using on variations in the financial and organizational requirements for adopting an innovation yields four ideal types of cases: requirements are either ' 'Ihcse ideas build on some of the concepts contained inloseph Schumpeter (1942). ''•Others have also used I'luistcuscii to make arguments about military innovations, including 11.....hiovv.fi and < iliol/ (.'lldn, IK,.'(, ><)) and Terry Pierce (2004). CIIAPTKR2 DIFFUSION OF MILITARY POWER 47 high financial-high organizational, low financial-low organizational, high financial-low organizational, or low financial—high organizational. Innovations requiring low levels of financial intensity and organizational capital to adopt are not terribly interesting from a relative power perspective. Nearly all affected states should be able to quickly and efficiently adopt the innovation, since it presents neither significant financial barriers to entry nor extremely complicated organizational procedures. One example of this might be the rapid adoption of chemical warfare by the entente powers following Germany's use of poison gas in the Second Battle of Ypres in World War I. Other possibilities are more interesting. MMIs requiring a large financial investment but not much organizational transformation are likely to benefit pre-innovation great powers and/or the system leader, since they are the wealthy states, and best positioned to simply build or buy the relevant technology and plug it into their military. Preexisting major powers will merely incorporate the new technologies and some doctrinal changes into their existing organizational structures, meaning they can rapidly build up in the new area of capabilities. Nonmajor powers will face huge financial barriers to entry even if adoption is possible from an organizational perspective. An example of this, discussed in greater detail below and in chapter 4, is the spread of nuclear weapons, especially in the first decades after World War II. In contrast, innovations requiring high levels of organizational capital to adopt but low levels of financial intensity can generate disruptions in the balance of power due to the mctamorphic organizational shifts required to take advantage of the new ways of war. Rebecca Henderson (1993, 250, 254-55) divides innovations into categories based on the relative degree of economic/ technological and organizational change necessary for implementation, studying innovations in the photolithographic alignment equipment industry, a subset of the semiconductor industry.57 She finds that the types of innovations that caused the most industry turnover were technologically incremental but organizationally radical, such as contact printers and scanning aligners. The failure of firms like Parker-Elmer when faced with organizationally radical changes like stepper aligners highlights the primacy of organizational change as a causal variable (261-62). Organizational inertia made the efforts of leading firms less effective than those of market entrants. While leading firms were not absent when radical innovation occurred, their ability to take advantage of those types of innovations was lower than those of entry firms. When innovations were organizationally incremental, however, leading firms capitalized because their existing business structure was reinforced by the new innovations, giving them a larger advantage over smaller firms (267-68). ' ' Photi>li ill ,l ■ ■■mi......III. I'll I I' il H I i I li ii. I. i i hi 1993, ||6) When new military innovations require high levels of organizational capital lo implement, Hendersons research and Christensehs work on firm failures suggest that just as those firms that arc most successful in existing markets i iftcn flounder in the face of disruptive innovations, having the ability to heavily and successfully invest in the military capabilities that enhance combat effec-I iveness during one period may handicap the capacity of a military organization to transform to deal with the next era. These firms struggle for similar reasons to those described by Jervis when assessing how decision makers integrate new information. Jervis explains how preexisting beliefs shape the way that decision makers integrate new information. Their worldviews, which have served them well in the past, can blind them to changing circumstances, since ihcy see things through the lens of their past experiences and beliefs. There-lore, they apply the same cognitive map to a new circumstance even though it is now inappropriate (Jervis 1976,143). In the military context, since these types of innovations will require doing business differently, rather than doing it better, research and development segments of leading military organizations are more likely to dismiss early signs of the innovation as irrelevant to core competencies, placing them at a tacit knowledge disadvantage if/when they attempt to catch up later. The low required financial intensity will decrease the barriers to entry in the international system, potentially allowing new states to compete by at least partially wiping the military slate clean. This will threaten the position of existing powers as some likely try to resist the change, while others recognize that an adoption-capacity gap means the state must choose an alternative Strategy and accept a period of relative decline. Many argue that combined arms warfare, also known as blitzkrieg, falls into this category. As Ernest May (2000, 476-80) and others have pointed out, the other armies of Europe possessed much of the same, if not more, raw military equipment: radios, airplanes, Links, and motorized vehicles. It was the way Germany put together those technologies into mancuverable, deep-strike packages that was inventive. Ihe (rcrman departure from the French strategy of static defenses and trench war-lare required a significant shift in training and operations, making maneuver warfare difficult to adopt for many militaries (Boot 2006, 233-36). France, in contrast, represented a military with a low level of organizational capital due to its high organizational age; this is one reason that its victory in World War I locked in flawed lessons about future wars that prevented it from grasping the potential of maneuver warfare (May 2000, 449-50). Finally, shifts requiring high levels of organizational capital and financial intensity, precisely because their spread rate and extent is likely to be constrained, should iheuielii ally have the largest effect on the balance of power because cotinlricM llial du lake advantage of the MMI will gain long-term asymmetrical powei iuh ......Majm powers that lack the capacity to adopt the new MMI an al-n il...... i...... pinue In making strategic errors when 48 CHAPTER2 DIFFUSTON OF MILITARY POWER 49 facing a challenge. 1 Ihr > ,u i in warfare case, described in more detail in chapter 3, exemplifies litis I.in- bul poWOJ fill high-high case. Table 2.3 outlines I he idea 11/cd relationship between the diffusion of a given MMI in a normally distributed international system (an important qualifying assumption) and its impact mi global power. Regardless of one's theory of change in the balance of power, the above examples demonstrate that shifts in the ability to produce power matter. Military innovation diffusion represents one causal process governing the timing and character of power shifts that have been previously explained on the basis of other factors. The degree to which leading countries take advantage of new forms of military power helps predict long-term trends relating to the international balance of power. First-Mover Advantages Economists have argued about the relative advantages a firm achieves when it is the first to introduce a new technology or marketing strategy into a given market, making it the so-called first mover (Bain 1956; Schumpeter 1942). While the notion of first mover generally refers to the ability to bring a new product to market in a way that captures large amounts of market share, it is useful to break down first-mover advantage into its component parts. There is both a technological and an organizational component to first-mover advantages. The technological advantage refers to hardware; firms break ground directly in the area of a given product, like semiconductor improvements, or introducing entirely new products, such as the personal computer. The organizational advantage is when firms take a given technology, and figure out how to produce and market it in ways that give the firm an advantage. The production line, which Henry Ford developed for the Model T, exemplifies an organizational first-mover advantage. There are some clear advantages to being a first mover in terms of long-term rents and business viability (Milner and Yoffie 1.989, 244). One analogy between the military and business world comes from patent law, which influences the ability of firms to maintain a monopoly on production knowledge and extend their first-mover advantage (Mykytyn, Bordoloi, McKinney, and Bandyopadhyay 2002,1). Industries like pharmaceuticals that depend on secrecy regarding production processes have shown that patents are critical to the long-term viability of first-mover advantages (Lieberman and Montgomery 1988, 43-45). Military organizational processes are somewhat (though not entirely) comparable to the special knowledge a company has when it invents a product. iH'Iliese slates may or may not lie in decline prior U> ilu..... < I "I il» .......ration Adoption now 11 v thoory Ii ngnotdc on tli.n point Table 2.3 Relationship between the Spread of Military Power and the Balance of Power Level of financial intensity required to implement major military innovation Low Irrel of rganizational tepital required H implement major military innovation High Low Diffusion fast Impact short-term and not likely to be structurally significant Relatively shorter first-mover advantage Key example: chemical warfare • Diffusion medium • Risks reordering global power balance in ways likely to threaten existing powers • Relatively longer first-mover advantage ' Key examples: suicide terrorism, Napoleonic warfare, combined arms warfare (blitzkrieg) High ' Diffusion medium • Risks reinforcing existing global power balance • Relatively shorter first-mover advantage (except in extreme cases) • Relatively benefits existing powers • Key example: nuclear weapons • Diffusion slow • Likely to have substantial and long-lasting disruptive impact on balance of power • Relatively longer first-mover advantage • Key example: carrier warfare I ,ikc businesses, MMI first movers have a natural incentive to maintain se- I us y about new technological capabilities and their organizational processes, since cither or both could be based on knowledge that the first mover wishes to prevent from leaking. But once a demonstration occurs or enough information spreads, it is difficult to maintain secrecy for all but the most complex innovations. Like firms choosing research and development strategies, military organizations have to determine whether to allocate limited resources toward reinforcing competence in current military areas, or potentially becoming a hr.I mover in new military technology or innovative organizational processes II .ieherman anil Montgomery 1.988,54). ' I lie length of the fust-mover advantage for a given innovation affects its consequences for international politics for two reasons. The relative power of a first-inovet stale should increase liisler vis ,i vis peer slates during its first-mover p. nod ih.tu m any oilier one. Additionally, dlO substantive effect of a Specific military innovation should also In i ipn i,illy pionoiiiit ed tin Inst movers, lor 50 CHAPTER 2 DIFFUSION OF MILITARY POWER example, some innovations, like Prussian open-order tactics in the late nineteenth century, involve taking advantage of breakthroughs in transportation, such as the railroad. In that case, the content of the innovation, combined with the first-mover advantage, produced an enormous military edge for Prussia against France in the Franco-Prussian War. But the innovation quickly—albeit partially—spread, to some extent negating Prussia's first-mover advantage over the medium-term (Herrera and Malmken 2003,226).59 Lire adoption requirements for an innovation influence the length of the asymmetrical advantages that adopters receive, especially first movers, changing the way an innovation matters for the security environment. If an innovation is relatively easy to adopt, an actor will be unable to hold on to its first-mover advantage long enough to establish a dominant monopoly; adoption will be relatively easy for both early adopters and those that follow. The length of the first-mover advantage will decline, and the relative impact on the international balance of powrcr will also be smaller. Other states in the system will mimic the first mover quickly enough to negate many of the advantages to being first. In contrast, innovations that are more difficult to adopt, especially in terms of organizational requirements, will provide greater first-mover advantages to early adopters, helping them achieve larger international political advantages. Innovations based on new knowledge are more difficult to adopt, either because expertise at prior key competencies makes acquiring the new knowledge harder or because they are breakthroughs in new areas where others lack knowledge. These innovations are also entry enhancing, meaning the door is open for massive market upheavals (Darby and Zucker 2003, 2, 8; Tushman and Anderson 1986). First-Mover Gains Hypothesis: First movers should experience relative gains in power proportional to the length of their monopoly over the MMI and its relevance in international politics. The length of the first-mover advantage will be inversely proportional to the diffusion rate of the innovation. First-mover advantages might not apply in all cases, however. Late movers have an edge in some business environments because they can take advantage of technological improvements to innovations that reduce the uncertainty and risk associated with a given product (Lieberman and Montgomery 1988, 47-48; TellisandGolder2002).60 59 It is also important to distinguish between those states that follow the first mover, like Germany with all-big-gun battleships, and true late movers, like North Korea with nuclear weapons. "Late-mover effects" refers to the latter, not the former. '""This concept is related,though, not quite the same, as the arginuenl dial < irrm licukmn (1962) makes about the advantages of "backwardness," (lerschcnkron's assertion ll mON iboul development than (bout the response to Innovation winners.'there iv it polrnlliil iUttUtvAnltgS In being a lin.l movor in a military area dial relates I.....■gain/mlnnisl men.....I \n II I tu i t\m\\\r, milriniolille I /.ite-Mover Gains Hypothesis: Late adopters will face lower barriers to adoption due to more available information about the innovation, giving them a relative power edge over first and early movers once adoption occurs. War 'I ho diffusion pattern for major military innovations can also influence a broad i.ingc of factors related to how states select into wars and fight. Rapid changes in military effectiveness may also affect the overall utility of military power as it method of accomplishing political objectives. geography I he effect of geography on international interactions, especially military interactions, is well-known (Diehl 1985; Vasquez 1995). While most disputes are hetween contiguous states because they are fighting to control territory, the lechnology and thinking about the use of military force undergirding most military operations should not be ignored. If it takes years and/or decades to i each the land of an adversary, it will be much harder to fight a war than if an adversary is right next door. MMI diffusion, presuming the MMI is successful (which is not necessarily the case), should substantially alter the geography of military incidents by allowing states that experience MMIs to militarily interact with a broader geographic range of states. The effect should be more pronounced when both Males in a dyad have experienced an MMI, but should also exist in an asymmetrical relationship. Even if a particular innovation is not directly related to the technology of I ravel, major military innovations may give a state the ability to conduct more deadly military interactions in a wider geographic range. For example, the rifle and machine gun allowed for the projection of lethal force with fewer troops than were previously necessary, increasing the geographic scope of expedition-(try forces.61 'To the extent that MMIs shift outward the geographic range of the possible ill a military sense, they should introduce more systemic instability, meaning In I mover Henry Ford continued to use the same business model to produce his cars even after developments in the industry should liave caused changes (Abernathy and Wayne 1974,109). So I he organizational model thai led (o his success also led to a form of bureaucratic lock in due to its high organizational age mid narrow critical lusk focus. 1,1 lor instance, lake two Klalen lli.il have normal economic and diplomatic relations but cannot iii(ci.m i rapidly due lo iv'oj'.nipl......u.iiamU Ifan MMI makes it possible to effectively project military I......I o.k 1 a) n mm h gn 11• l . 11■• I,......Inn \\,v. pirvn ni'ily possible, one or both slates will bavr (he ability (o I......■mslibh mm!.....In m ilm ,n-> Ih......uld fundamentally shift stare-to- n(.i11■ n laiuni'iliip'i 52 l'IIAE>TER2 DIFFUSION OF MILITARY POWER 53 periods with MMIs are more likely to have more militarized interstate disputes (MIDs) and wars on a systemwidc level, not just for those states experiencing an MMI or competing with a state experiencing an MMI. SPEED In addition to widening the geographic range of states with which a given state can interact, some MMIs may also enhance the speed of those interactions. If states can more quickly mobilize military forces, they can issue credible threats more often and in a wider variety of situations, including economic and diplomatic cases. MMIs that allow for the projection of force faster and in greater depth could still allow states greater coercive options in regular interactions, meaning they should influence the way that states behave even if both states have achieved an MMI. It is possible that rapid mimicry may erase the speed-based impact of an MMI, meaning that neither side gets an advantage once the first-mover advantage has been countered. Still, MMIs should enable states to issue credible threats in a shorter time frame than was previously possible. Also, a state that has experienced an MMI could have the ability to respond to diplomatic overtures much more quickly than was previously possible. The end result is that the rate of military interactions between countries should increase at the sub-warfare level (i.e., low-level militarized disputes) because of the increase in the opportunities for interaction offered by an MMI.62 INTENSITY Diffusion patterns will also influence the rate and extent of casualties, or war intensity. If slower-spreading MMIs lead to more asymmetrical wars between MMI adopters and nonadopters, those wars are likely to be shorter. Yet the wars themselves will probably feature much higher relative casualty rates suffered by the nonadopter, controlling for war length and engagement size. In contrast, wars fought during periods of rapidly spreading MMIs are likely to be especially intense as states deploy heightened military capabilities against each other and raise the overall level of violence without changing the balance of forces, meaning the war is likely to be both bloodier and longer.63 63 Alternatively, MMI adoption could increase the number of choices for an adopting state in a given militarized situation, lengthening the decision-making process. 6i,As described above, innovations are themselves rarely offensive or defensive; the question is whether one side has an asymmetrical advantage. If an innovation tends to make attacks easier if the deiendcr has not adopted, though, versus making defense easier if the other side has not adopted, it could lead to some advantages. This line of thinking shows the impni i.iiu e of organizational plans for using technology, rather than technology in and ol ilsell 'In liunliigy only at quires an ollensive or defensive rhai'.n In in the hands of mililiiry oigam/al.....i Si i t ...Mum.......| Andres i'l'm Military Interaction Hypothesis: In general, states that experience an MMI will have more frequent, varied, and intense military interactions with a broader range of states than those that have not experienced an MMI. Answering Objections ' I here are several potential questions raised by adoption-capacity theory.64 One clear criticism is that this story only deals with the "supply-side" of the equation, ignoring the issue of why states might be interested in adopting innovations, litis is an important related issue, but this book is designed to build on existing work and deal with a distinct question. As noted above, there has been a BCCat deal of prior work, much of it good, on why states are interested in adopting innovations. Adoption-capacity theory begins at the point that diffusion Ittirts, after the point at which the international system, or norms or other facial's, influence whether or not a state feels compelled to respond to an innova-lion. It is variation in the adoption requirements for innovations in relation to I he capacity of a state to adopt that determines the content of state responses. I lit- geostrategic environment can make a country want to respond to an innovation, but state capacity to adopt will be a critical input into the character of that response.That analysis helps a state figure out whether adoption, using ex-loi nal alliances, or shifting to neutrality, for instance, might be the best response. And of course, those choices are not necessarily exclusive. ('ritics of the idea that organizational capital matters might protest that rational national leaders should recognize that their military organizations are i issi (led and adjust national policy accordingly to boost the organizational cap- iI a 11 if their military, giving them the capability to adopt necessary innovations. I )omesttC political constraints, however, often prevent leaders from enforcing large-scale organizational changes on a national military in the short-term. Most national leaders, even in nondemocracies, depend on support from at least one and sometimes more major nonmilitary constituencies in addition to i.npport from the military. Barbara Geddes's work (1999) explaining the re-ItrnirrtS on leaders in different types of nondemocracies shows that few leaders have true dictatorial power to quickly enact large-scale changes. The process of building support takes time and bureaucratic political capital. Given that leader, want to stay in power, they are unlikely to push for changes that they think will decrease their chances of political .survival, especially if ejection from of-liic may make their personal survival less likely as well (Goemans 2000). So if i he '.1 km I term risk to the Icudot t.....10 "iionupiinial" response is lower than if ''hi.....'.i.i....., tin i. ,ui pun mi lib i) nil' ml qui ......i iľ|i.tiiling how power transitions fttnc- .....i .nul lilt' vv......i ill.H.....i...... 'i ih ii ■>>> i *' hulí d I.....i......I .Hi d n .iKiiiiH. 54 CHAPTER 2 the leader pushes for adoption, most leaders will rationally decide not to push for adoption. Rosen's work (1991) on military innovation also shows that external pressure from politicians is often not enough to cause militaries to change their organizational setup, especially pressure related to war fighting. For example, many link the operational struggles of the U.S. Army in the Vietnam War in part to its failure to reform despite pressure from civilian leaders. In the early 1960s, U.S. president John F. Kennedy directly ordered the army to focus on and train for counterinsurgency warfare. Despite this high-level pressure, the army failed to significantly change, instead opting to cosmetically modify its training programs and define counterinsurgency as a lesser and therefore easier form of conventional warfare (meaning training for conventional war would also prepare the army for counterinsurgency warfare). As a result, the U.S. Army entered the Vietnam War much less able to fight an insurgency than it would have been if external civilian pressure had been enough to cause change (Bacevich 1986). Another potential critique is that perceptions of innovation effectiveness, rather than adoption capacity, predict the distribution of responses to an innovation. If countries only adopt those innovations that work, does this provide evidence against adoption-capacity theory? This question misconstrues the point of adoption-capacity theory. It is certainly true that perceptions of the effectiveness of an innovation create the framework within which diffusion does or does not occur. For example, strategic bombing is an innovation that did not diffuse in part due to perceptions of ineffectiveness. No theory can explain everything. The point of adoption-capacity theory is that within the range of those innovations that states might be interested in adopting, wanting to adopt an innovation is not enough to predict the distribution of responses or the affect on international politics. Moreover, if knowledge of innovation effectiveness is all that is necessary to predict diffusion, there should be either a lot more diffusion of innovations than happens empirically—since most MMIs do not spread widely, especially initially—or all actors with the capacity to adopt an innovation should adopt. But instances like the British Army in response to Napoleonic warfare, which had the capacity to adopt mass mobilization but chose not to, show that countries do not automatically mimic successful innovations just because they can (Herrcra and Mahnken 2003, 216-17). Also, if perceived effectiveness is the key factor, the pattern of responses to innovations should not change based on the adoption requirements, meaning empirical variation in the number of states that adopt, counter, and/or form alliances shows that automatic adoption of successful innovations docs not occur. Instead, it is the relative financial and organizational rcquiremeiils lor adopting an innovation that best predict the distribution ol'tcsp.....,es and many individual actors. DIFFUSION OF MILITARY POWER 55 Finally, in many cases, only a small number of actors either adopt an innova-llun or have the ability to adopt. These relatively small sample sizes, rather ih.in representing a problem for the theory, help illustrate the difficulties in-t. ihvd in adopting especially complicated innovations. If innovation adoption I i11 easy and many states could quickly adopt major military innovations, even BjUDUgh how wars are fought might change, a given innovation would be un-lil i-lv to change the international security environment much in a relative . n.e. The lack of innovation diffusion in some situations, despite clear strate-i. need, demonstrates adoption-capacity theory at work. 'Ibis book represents an attempt to fuse together some important aspects of " \ i. his research into a more coherent explanation for the diffusion of mili-iiiv power. Therefore, when assessing alternative arguments, in many cases nieces of them overlap with adoption-capacity theory, although pieces of them diverge as well. Additionally, in some instances I explain what alternatives I ii mi the literature might look like even though the literature docs not explic-uh address the same questions. That is a crucial caveat on this section, but it iI ' i means I am trying to build the strongest possible alternatives for compari-mhi. Drawing from these preexisting theories means this section focuses more mi attempts to derive orthodox and exclusive versions of alternative explana-.....is than anything else.65 Reluming to the existing research on the diffusion of military power within I he international relations literature, as described above it is possible to derive iI nee competing schools of thought: a more or less neorealist set of arguments .....i crning the critical role threats and strategic competition play in governing i he diffusion process; a norms-based contention that innovation adoption oe mis when states try to gain status or legitimize their existence, not as a strategic incisure to increase relative power; and a cultural assertion about the importance of cultural similarity in allowing responding states to adopt innovations. \ . explained above, adoption-capacity theory both builds on and moves be-Vi uid this research, but it is essential to discuss some of the distinctions. Strategic Competition I lie broadest version of this realist line of argumentation is that states engage in security maximization in response to military innovations, with geopolitical pnssure based on the particular situation of a state determining the particular ii.iliniial response to an innovation. The behavior of military organizations in n •.piiuse to innovation is determined by systemic pressure.66 Rcscnde-Santos's I. ii example, while intcrscrvicc competition undoubtedly plays a critical role in the process in mnr ii ".pis Is, there is not an account of its relationship to the diffusion process writ large. More- iivti, the .....In .lauding of organizational capital presented here incorporates some elements of mil i i• iv11 c rivalry. i ,..1.1111.i11 and Kli.ium (2001b,X) also make this point. 56 CHAPTER 2 DIFFUSION OF MILITARY POWER 57 recent book (2007) on South American militaries in the late nineteenth century maintains that the severity of external threats explains the extent of emulation and that countries attempt to emulate successful innovations in relevant areas of warfare.67 The strategic competition alternative is powerful because most instances of military diffusion probably have some strategic aspects at their core, whether it is due to perceptions of threats motivating interest groups to promote particular types of changes or global norms causing the interpretation of events to occur in such a way that it makes certain innovations more likely to diffuse than others. Indeed, when applied to the decisions of particular states, adoption-capacity theory similarly contends that the geopolitical position of a state helps predict its behavior—though in a less mechanistic fashion. Nonetheless, there are several reasons to think the picture offered by the strategic competition perspective is incomplete, especially in the case of Re-sende-Santos's research (2007, 296, 300), which is the most recent realist work on the topic. First, a strict strategic competition approach excludes the notion that financial and organizational capacity play important roles influencing the response to innovations. As described above and as the subsequent empirical chapters will show, however, factors internal to military organizations play a critical part in determining how militaries approach war and organize themselves to fight. Case studies ranging from British strategy in the Boer War to U.S. naval developments in the 1920s and 1930s to Soviet military innovation in the post-World War II period show the significance of internal military organizational factors in driving the innovation-adoption process (see, for example, Avant 1994; Farrell 1998,408-9; Rosen 1991; Zisk 1993). Second, strategic competition can predict potentially which states might be interested in adopting an innovation, but it does not predict which of them will actually succeed in adopting; it can only predict interests, not outcomes, since it cannot explain capacity (for a similar point, see Goldman 2003). For example, Resende-Santos (2007, 9) exphcitly excludes whether or not attempts at adoption succeed from his theory. While a neorealist might respond that sufficiently motivated states will always just build the capacity to adopt an innovation, the discussion above shows that sometimes the organizational or financial barriers to change may be sufficiently large that adoption becomes unrealistic. Adoption-capacity theory predicts both the strategy a state will adopt and whether or not attempts at adoption, if they occur, will succeed. Third, a pure strategic competition approach risks tautology: if the theoretical prediction is that states adopt military innovations when they feel strategic pressure, one could argue that by definition, anytime a military adopts '" Wall/ (1979,')?., 106, 1.27) suggests that emu latum rapidly m > urn in ii-.|hiiim- in innovation anil that security masiini/ation is only really a i|iieslitm nl |>ii.,ii I li...... 11'>')')), however, • lis If.ll I '■ mi innovation it demonstrates the "capability," broadly defined, to adopt the ii.....vation and the state must have faced a strategic threat, while anytime an innovation is not adopted it demonstrates the lack of capability or need, In.unify defined, to adopt. Threats do clearly matter and influence the way dial actors view the potential risks involved in making defense planning i linices. But threats alone are insufficient; questions of the relative benefits or . ,i . of innovations arc something that several theories could potentially explain, meaning it is inappropriate to describe all approaches that argue states frOlgh costs and benefits as "realist" (Zarzecki 2002,67). In addition to these arguments, there are other important distinctions between adoption-capacity theory and Resende-Santos's claim, He argues that 11 ii in I lies emulate military systems with proven success in warfare, using the Si it ill) American adoption of the German military model {Resende-Santos .'007, 80-81). But the international response to the Dreadnought shows that l.ui'.i'-scale responses to a new innovation, both emulation and alternative iiiaicgies, can occur even before use in warfare (see chapter 5). Adoption-i .ipacity theory is also focused on something smaller since it concentrates explicitly on the response patterns to the debut of a particularly large and crucial 11 huh of military innovations, rather than a situation where an established 11 nide! has existed for decades. Additionally, while he only looks at whether or mil stares emulate, adoption-capacity theory predicts the alternative response on.i logy of states, which could be emulation, but could also be something else. I'iu.illy, adoption-capacity theory makes predictions about the systemwide re-Iponsc to an innovation, while Resende-Santos addresses how individual states might respond. International and Domestic "Norms" Another perspective on the diffusion of military power maintains that shared in n ins of appropriate behavior determine the way that states respond in the International environment and that military affairs are no exception (DiMag-\\n i ami Powell 1983; Finnemore 1996). One phrase often used to describe this approach to studying military organizations is "sociological institutionalism" (I bill and Taylor 1996). I ),uia Eyre and Mark Suchman apply sociological institutionalism to mili-i ii y leclmology purchases in the developing world, contending that the spread ..I military technology is based more on perceptions of appropriateness by minor powers than genuine security concerns (Eyre and Suchman 1996, 96; '.in Imiun and Eyre 1992; Wendt and Burnett 1993).TheoFarrelTswork (2005) mi Co] i verge net' due to norms, ilea tiheil above, also fits into this category.68 ""'Micro uro wveral stUiantH Ol CltOH argument*. lO lutilni.....ipetltlve Isomorphism, or mimicry il.......iiintpcllllw pit' inte, wlih h 1' ml" i" prpilli llmm •ImiUi l!ir limitation with this cultural alternative is that it is only focused on the i|iii",lion of innovation adoption; it does not make predictions about strategic 11 ii (ices in general. The idea that cultural openness may influence the interest |C(Lt!CS have in adopting innovations is also consistent with adoption-capacity ilicni'v, since adoption-capacity theory does not exclude other motivations for adoption. finally, the strong cultural perspective also risks tautology in some cases be-cmiae it is difficult to evaluate variables like openness to cultural diversity without evaluating if the state in question actually adopted the particular innovation being studied.71 This risks the independent variable (the level of acceptable cul-mial diversity) being coded on the basis of variation in the dependent variable (whether or not the innovation is adopted). This is clearly problematic; it both artificially creates the appearances of a causal linkage where one may not exist and coiild lead to attribution bias because it incorrectly identifies the actual lOlircs of cultural diversity, failing to independently define that variable. *" Another possibility is that culture may sometimes serve as a proxy for substantive debates, Mritll < iillur.il language serving as a way of explaining material concerns. The point is not that cul-i in i' ill ii-. iiiil matter, just that it is important not to view culture as necessarily static. '( roltliii.m (21106) explicitly H'sti llicc .iigiiiuciits against each other in her work on the Otto-iii.in I'liipivr ami Meiji Restoi'illion, linduifj, llnil polltll al culture, in the form of elite beliefs, influ- .....n the receptivity nl'ii society In mllllui') litnov......ill. Adoption-capacity theory attempts to plWlds Ul explanation for IMUO | llkr win 'In- ......1,1 hurIi! be more or less receptive to in- h'r, ,ii ii111m iii u glvan cma. 1 < mM.....n'n wink (.Kllld) ruplii lib In ■ Ij i ii mi i linn p" .iliihn l'[[A]>TER2 DIFFUSION OF MILITARY POWER 61 Adoption-capacily theory can also arguably describe the conditions in which cultural, openness is likely to occur, especially with regard to military organizations. Militaries with high levels of experimentation, young organizational ages, and critical tasks less defined by operational methods have higher levels of organizational capital, making it easier for them to be culturally open and adopt innovations. This is' a more complete argument explaining innovation capacity because it allows for concrete predictions about when cultural tolerance is likely to occur, along with falsifiable indicators such as the commitment a military has to experimentation. Case Selection and Next Steps This book employs two primary research methods—qualitative cases studies and quantitative analysis—to determine how states respond to new major military innovations, and how those responses affect international politics. The case studies are necessary because, as discussed above, differences in how long innovations remain relevant and the possibility of overlapping innovations in different areas of warfare, like land and sea, make a comprehensive data approach untenable. Additionally, innovations occur in part as an action-reaction process. New innovations displace, or sometimes even exist alongside, the dominant weaponry and strategies of previous periods. This means attempts to measure the diffusion of innovations are often "right censored." We rarely get to see the full distribution of responses to an innovation, over time, since new innovations arise to replace old ones while states may be trying to build the capacity to adopt or develop a long-term response strategy. In other cases, we might get to see the full distribution. But the potential for right censoring makes it difficult to evaluate the diffusion process. Instead, the unit of analysis is the innovation itself. Process tracing is used for each case to establish the existence of the innovation and evaluate the extent to which diffusion processes governed the specific strategic choices of key states. Quantitative analysis provides a bigger-picture perspective for some of the cases, showing trends only comprehensible when the reactions of all states are aggregated together at the system level. The methodological combination increases confidence in the accuracy of the results by ensuring that both the micro- and macrolevels are consistent with the theory. Table 2.4 shows a range of military innovations that scholars have claimed "qualify" as major military innovations from 1800 to the present. Since many scholars diverge in their thinking about which cases count, this study uses the universe of major military innovation candidate cases as its range'.7'- Kven if "The dclwtľ wit wlin h j 111111 it 11.......v ,n Kinu i.....>t m mi Inporttfll Ihuo Mini dflMrvoi recogn! linii:'Ihn bOOk I 'i" 01)1) i ÖUnlllP íl i......II iiiiinhi i •■! , , , ......| ihn, ,u, ,||,M(,|<-mimin nvt'l whii h Table 2.4 Range of Possible Major Military Innovations, 1800-Present 1'lining Revolution Nineteenth century Napoleonic revolution/levee en masse Nineteenth century Strategic communications/mobility Nineteenth century Professional staff personnel and procedures Nineteenth century Prussian open-order tactics: Railroads/rifles/telegraph Early twentieth century Batdefleet warfare Nineteenth to twentieth Tactical fires (machine gun and artillery) centuries Nineteenth to twentieth Medical centuries Nineteenth to twentieth Fortifications (trenches) centuries Twentieth century Chemical weapons twentieth century "The modern system" Twentieth century Total industrialized war Twentieth century Blitzkrieg ' 1 wenticth century Carrier warfare Twentieth century Tactical air attack Twentieth century Air warfare (bombing) Twentieth century Submarine warfare ' 1 'wenticth century Nuclear weapons 'Twentieth century Mao Tse-tung's "people's war" Twentieth century Unconventional war/suicide terrorism ' 1 wenticth century Micro electronics/genetics engineering 1 ,HCC twentieth century Information war/network-centric warfare 1 ,;ite twentieth century Fourth-generation warfare scholars disagree about which cases meet their particular definition, they gen-ti. ill v agree that the set of cases under consideration represents the most important military innovations. The existence of the scholarly debate itself also indicates that it makes sense to select cases from the broader set.73 'Ihe major military innovation cases evaluated in this study are the following',; early twentieth-century battleflcet warfare, mid-twentieth-century carrier < ľ.e'i hIiiiuIi! count lint this hook represents n first step toward a more integrated theory of inno-v.1111111 diffusion, 1 Some of the cases on llii't Hat overlap hcciHMc scholars disagree in the wiiy they categorize an innovation. Tor example, Miilne si holms lneak u|> innovations like Prussian open-order tactics into liner li'ilinoloji/uiilly ilelinnl i umeniu n........i.il......i i.iilinails, rifles, and the telegraph. Ihc cod- liiK kbuiod on Information ntl.....1 b) tin lulimt mul Aim.....ler Holand {lW7).'lhe hold cases ..u il. , c inveied hy ihtu IhhiI In ill | 'I' 62 CIIAI>TIÍR2 warfare, nuclear weapons, and suicide terrorism. As table 2.5 shows, this selection of cases maximizes variation on the two key independent variables: financial intensity and organizational capital. It also allows for significant time variation, over a period of a century, and cases that focus on both nation-states and nonstate actors. Carrier warfare is one of the only (strategic bombing maybe another) major military innovations requiring high levels of both financial intensity and organizational capital to adopt. Even during the interwar period, when the battleship was still considered the most important element of the modern navy, aircraft carriers had already overtaken the battleship in terms of the cost per unit. Aircraft carriers are as complicated to operate as they are expensive, especially the adoption of carrier task forces, the crux of the carrier warfare innovation. Operating a floating airfield and the ship itself, plus coordinating with support ships, is simply a much harder set of tasks than lining up the big guns of a battleship and firing.74 In addition to this operational complexity, the critical task of carrier warfare, the massed air strike, replaced the five-hundred-year stranglehold of the big gun over naval warfare. So carrier warfare represents a disruptive innovation requiring enormous levels of organizational capital to implement, alongside its high cost. Nuclear weapons, like carrier warfare, require high levels of financial intensity for adoption, but in contrast they only require low levels of organizational capital. Nuclear weapons are the only weapon in history where the mere possession of one unit, one nuclear bomb, regardless of anything else, substantially impacts international security, as the unresolved North Korean and Iranian nuclear controversies highlight. There is no other weapon where just threatening to develop a single decades-old version vaults a country onto the priority list for high-ranking U.S. government officials. The uniqueness of nuclear weapons as a symbol of power for coercive purposes in and of themselves, and the potential destruction in the case of use, makes studying the spread of nuclear weapons and the strategic choices made by states in response to nuclear weapons especially interesting. It requires a low level of organizational capital to adopt, despite its organizational complexity, because there is no particular organizational form that is necessary to acquire and deploy nuclear weapons. The suicide terrorism case is distinctive because it involves an attempt to apply a theory designed to predict the behavior of traditional military organizations to terrorist groups—nonstate actors. Suicide terrorism represents the premier military "innovation" among tactics by terrorist groups (and arguably all nonstate actors) over the last twenty-five years.75 The chapter focuses on the 71 While battlcflcct warfare was not easy to adopt either, carrier w.u I'.ni' was harder. ' Depending on hmv one thinks about guerrilla warfare, and whether i niicnl nonsl.ile violent Ii luven li-li Is are substantially similar In or dif Infill (nun past |;iii'i I ilia Will l,i........ill .,".....dein" rinn ill,i w, ii I, in- intild .it. i i|ii,ilili DIFFUSION OF MILITARY POWER 63 Table 2.S How Variation in Adoption Capacity Drives Case Selection Required level of Required level of financial intensity organizational capital liattlcfleet warfare (chapter 5) Medium Medium t tarier warfare (chapter 3) High High Nuclear warfare (chapter 4) High Low Suicide terrorism (chapter 6) Low High terrorist group as the unit of analysis, looking at the factors that drive some terrorist groups to adopt suicide terrorism but others to stick with prior tactics. Some revisions are necessary to account for the differences between states and nonstate actors, yet the core theoretical contentions are pertinent. Terrorist groups still face organizational challenges and financial constraints, meaning adoption-capacity theory is applicable. As referenced above, the adoption of suicide terrorism requires a low level of financial intensity. There is also no real "signature" technology underlying the innovation. Delivering explosive destruction through the necessary death of the transmitter, the suicide bomber, is a different operational paradigm from something like kidnapping or hijacking, however. As an innovation that rc-tjitires high levels of organizational capital but only low levels of financial intensity to adopt, the suicide terrorism case is a nice contrast to the nuclear weapons one. By showing the portability of the theory beyond simply those MMls mostly of interest to great powers, this case highlights the utility of adoption-capacity theory for thinking about strategic choices in response to all types of military innovations and perhaps for outside the military realm as well. bin-ally, since no cases truly fit in the "low-low" box, further diversification along the key independent variables was conducted by selecting an innovation iequiring medium levels of both financial intensity and organizational capital in adopt. The battlefieet warfare case involves the creation of the all-big-gun steel battleship, the culmination of the fifty-year-long operational shift away From the wooden-hulled sail navies that had dominated naval warfare for centuries. Battlefieet warfare also involved skyrocketing procurement costs for the new ships because the fast rate of warship obsolescence in the late nineteenth and early twentieth centuries due to rapid technological changes meant states had (o consistently ramp up their naval procurement budgets. Give that prein-iiovalion spending was already high a( the unit cost level and the underlying basis of the previous ei.i had .dsn been military rather than civilian, the finan-. i.il Intensity required lot killlclleel wail.tie was medium. Modern battlcflcct w.u fare forced enoi nun is i Ii.iiijm ......i tut nip, a in I training, because the core ii 1 .....nlin led In thr nvi i 'it',' nl'o Ii id ' It uit',1 il hum climbing the rigging 64 CHAPTER2 and trimming sails to operating steam boilers and using automated fire control systems for the big guns. Still, battlefleet warfare was also a sustaining innovation because while it required large-scale changes in recruitment, education, and training, the critical task of most navies remained the same: seeking a decisive battle in which it could deliver the maximum possible number of shells on to enemy ships. Thus, the organizational capital required for adoption was medium. Each chapter proceeds by outlining the essence of the innovation and its emergence in the international system, then applying the theory to predict the strategies states will choose in response, whether those that attempt to adopt the innovation will succeed or fail, and how the overall distribution of choices influenced key aspects of the international security environment. Depending on the available data, the empirical chapters feature different combinations of qualitative and quantitative methods. The suicide terrorism and nuclear weapons chapters include quantitative statistical tests for the diffusion of the innovation, while the battlefleet and carrier warfare cases rely more on process tracing based on primary and secondary source research. Conclusion Adoption-capacity theory is a new approach to studying the introduction and spread of major military innovations. It builds on existing research on military innovations and emulation to produce a framework for explaining the way that military innovations spread throughout the international system. It also helps us understand, conceptually, the way that the uneven pace of military innovation diffusion functions as one of the major driving factors underlying shifts in relative power and warfare. The next four chapters empirically test adoption capacity on the cases described above: carrier warfare, nuclear weapons, battlefleet warfare, and suicide terrorism. Chapter 3 CARRIER WARFARE As the predominant form of naval power, aircraft carriers are one of the i barest symbols of military strength on earth. Short of the atomic bomb, noth-m;'; signifies the power of a great nation like the possession of a fleet of aircraft i arriers, able to control the oceans and project power across great distances. It was the transition from battleship to carrier warfare that helped shepherd in the era of U.S. naval supremacy, and the complexities involved in adopting the innovation have ensured U.S. naval superiority ever since. The United States i urrently possesses an overwhelming advantage in naval power fueled in large part by its dominance in carrier warfare. While the United States presently operates eleven fleet carriers, no other country operates more than one fleet i arrier—and only one other country, Great Britain, operates more than one i arrier of any kind. Only the U.S. Navy, the Royal Navy, and the World War II i'm Imperial Japanese Navy have ever adopted the core organizational practices associated with carrier strike operations, the crux of carrier warfare. This Ih curious given the natural desire of states to protect their own security by building up their military forces with the best available technologies and practices. If the effective operation of aircraft carriers is a critical element of naval | h iwer, why haven't more countries adopted the innovation? ' I his chapter builds on work by Goldman and others on the development of i,i trier warfare during the interwar period and the early years of World War II, when carrier warfare matured. The chapter focuses on what happened next: the post—World War II gap between the diffusion of aircraft carrier technology .itid ihe diffusion of carrier warfare as an operational practice—the carrier ■.Hike operations almost simultaneously implemented as the core element of naval operations by the U.S. and Japanese navies after the Battle of Midway. ( hie explanation for the nondiffusion of carrier warfare and the slow diffusion of carriers themselves involves geostrategic requirements: states simply have not needed carriers. Another possibility is that the failure of carrier warfare to 11Hfuse in the post-World War II era lies somewhere between security requirements and the difficulties, financial and organizational, involved in adopting. "Hit* high financial and organizational requirements for adoption caused most naval powers to evaluate the cohIn and benefits of their naval strategy differ-rnlly than they did iu r<'h|niii«e to the u.iv.il innovations of the past, driving a linger proportion til ',1,11c, 1111 In i] 11 in I ol'I lie niiv.tl power game, bandwagon, or hy to counter U.S. naval mpn mat \ through iilleninlivc means.