The Information Revolution and International Stability: a Multi-Article Exploration of Computing, Cyber, and Incentives for Conflict

The Information Revolution and International Stability: A Multi-Article Exploration of Computing, Cyber, and Incentives for Conflict

by Jacquelyn Schneider

B.A. in Economics-Political Science, May 2005, Columbia University M.A. in Political Science, May 2013, Arizona State University

A Dissertation submitted to

The Faculty of The Columbian College of Arts and Sciences of The George Washington University in partial fulfillment of the requirements for the degree of Doctor of Philosophy

May 21, 2017

Dissertation directed by

Charles Glaser Professor of Political Science and International Affairs

The Columbian College of Arts and Sciences of The George Washington University certifies that Jacquelyn Schneider has passed the Final Examination for the degree of Doctor of

Philosophy as of January 27, 2017. This is the final and approved form of the dissertation.

The Information Revolution and International Stability: A Multi-Article Exploration of Computing, Cyber, and Incentives for Conflict

Jacquelyn Schneider

Dissertation Research Committee:

Charles Glaser, Professor of Political Science and International Affairs, Dissertation Director

Martha Finnemore, University Professor of Political Science and International Affairs, Committee Member

Caitlin Talmadge, Assistant Professor of Political Science and International Affairs, Committee Member

iii

Dedication

The author wishes to dedicate this work to the most interesting man in the world—my husband.

Acknowledgments

I am extremely grateful for the support and advice I have received over the last five and a half years of this process that it took to get to this point. It has been two moves, two babies, and two graduate schools to get here and I truly could not have done it without this amazing group of people. In particular, I want to thank my chair, Charlie Glaser, not only for his comments on this dissertation, but for his unwavering support of my sometimes overly-committed and overly-enthusiastic forays into both policy and academic diversions.

He allowed me to take opportunities, even when I perhaps wasn’t ready, and for that I am extraordinarily grateful. I know that quite often these were leaps of faith and it meant the world to me that he was willing to take them. I also found at GW a strong cadre of women that were inspirational not only for their academic accomplishments but also for their ability to balance family and professional life. I was lucky enough to have two of these women on my committee. Martha Finnemore was an enormously influential part of my committee and

I am very grateful that she was willing to read through and provide guidance on my heavily security-focused pieces. Some of the best parts of these explorations were because of her advice and conversations. Caitlin Talmadge was a constant source of guidance over the last three years and her keen comments on all of my work as well as her forthright and caring advice on work-life made her a role model for me. Thank you to both these women for showing me that you can be rock stars at work and rock star Moms. Finally, I need to thank profusely Henry Farrell and Robert Jervis who graciously agreed to act as readers for this dissertation and who provided incredibly useful feedback to improve the piece.

I found at GW a department full of inspirational colleagues and scholars who played a huge role in my intellectual development over the last three years. I first want to thank

Kimberly Morgan who took a chance on a transfer student and made my transition to GW so seamless. I also want to thank the scholars at the Institute for Security and Conflict Studies

including Janne Nolan, Stephen Biddle, Elizabeth Saunders, Rachel Stein, and Alex Downes for their support both academically and personally. At ISCS I found a group of scholars that were challenging, insightful, and interested in building the next generation of security scholars. It was a place that I grew as a scholar and much of that was due to the fantastic group of colleagues I found at the Institute, including Alec Worsnop, Jane Vaynman, Sara

Bjerg Moller, Bryce Loidolt, Andrew Bell, Ryan Baker, Daniel Jacobs, Michael Joseph,

Meredith Blank, Frank Smith, Andy Leven, Joseph Torigian, Inwook Kim, Daniel Krcmaric,

Olivier Henripen, and Tyson Belanger.

I also want to thank faculty and colleagues at outside universities that were so instrumental in the development of me as a scholar and both directly and indirectly this dissertation. First, I want to thank Mikki Kittilson, Sheldon Simon, Carolyn Warner, and

Reed Wood for their mentorship at ASU and the efforts that many of them took to help me transfer to GW. I also need to thank Stephen Walker who has been my biggest advocate over the last five years. He has cheered me on, commented on drafts of a series of papers, and written countless letters for me. He has inspired some of my best work and I am extraordinarily thankful that I have the honor to be in his scholarly lineage. Finally, I would be remiss if I did not thank the colleague and scholars who have been so instrumental to the evolution of my work on cyber including Erik Gartzke, Jon Lindsay, Joshua Rovner, Fiona

Cunningham, Stacie Pettyjohn, Herb Lin, Rose McDermott, and Austin Long.

As is likely obvious from the dissertation, I have been heavily influenced by the experiences and opportunities I have had through Cyber Command and the Naval War

College. I want to thank Sean Keenan, Jen Silk, Paul Jacobs, Shannen Parker, John Zerr,

James Vandevelde, Michael Warner, Emily Goldman, Bart Gray, and Guy Walsh at Cyber

Command for their support. I also want to thank the Naval War College for not only giving me the opportunity to use the wargaming data, but for also shielding me from outside work so that I could finish the dissertation over the last six months. Big thanks go out to Andrew

Winner, Hank Brightman, Peter Dombrowski, and Chris Demchak.

I have saved for almost the very end my thank you to one of my biggest supporters and sources of inspiration—Julia Macdonald. I met Julia in 2012. I was in the early stages of my first pregnancy, attempting to hide my emerging baby bump while camping out in the

Cornell dorms at SWAMOS. It turns out that SWAMOS would be a huge turning point in my life and that is because it introduced me to the best co-author and friend you could ever imagine. She has been an eternally positive and uplifting person in my academic life who is just as willing to read a draft as to promote my successes and watch my kids. She is my academic other half and I cannot wait for all the amazing research we are going to do together.

Finally, I want to thank my amazing family who has supported me through this long rollercoaster journey. Huge thanks go out to my parents who gave me the drive and thirst for knowledge and to them, my in-laws, and my brother who have all stepped in to help Jeff and I raise these awesome kids. I could not have done it without you. And I absolutely could not have done this without my amazing husband, Jeff Schneider. He moved across the country (twice) so that I could follow my dreams. He has always believed in me and without his love and support I would never have been able to start or finish this journey.

Thank you.

vii

Abstract of Dissertation

The Information Revolution and International Stability: A Multi-Article Exploration of Computing, Cyber, and Incentives for Conflict

Over the last twenty years, the digital technologies of the information revolution have fundamentally impacted the nature of modern warfare and introduced significant uncertainty about capabilities, intentions, and the limits of conflict. How do these digital advances and the explosion of cyber technologies on and off the battlefield impact international stability? Has the information revolution fundamentally altered crisis dynamics such that states are more likely to inadvertently or intentionally escalate crises to war? This dissertation utilizes a multi-article, multi-method format with case studies and experimental analysis to examine these questions. The first piece looks at military revolutions and international stability and argues that a capability/vulnerability paradox of military revolutions suggests de-stabilizing effects of the information revolution. The second article examines offense-defense in U.S. doctrine since the information revolution and finds that the information revolution technologies invested in by offensively minded doctrinaires has created a path-dependency for further offense dominance in the information revolution. Finally, the last piece analyzes data from six years of crisis wargaming to generalize U.S. decision-maker perceptions of escalation and cyber operations. Together, these pieces seek to solve fundamental policy questions about how cyberspace operations and technology may impact current and future crises and their potential for conflict eruption.

viii

Table of Contents

Dedication ...... iv

Acknowledgments ...... v

Abstract of Dissertation ...... viii

List of Figures ...... x

List of Tables ...... xi

Chapter 1: The Capability/Vulnerability Paradox and Military Revolutions ...... 1

Chapter 2: Offense-Dominance in the Information Age ...... 68

Chapter 3: Cyber and Crisis Escalation: Insights from Wargaming ...... 115

Bibliography ……………………………………………………………………………….160

List of Figures

Figure 1. Corps 86 Acquisitions Program.…….…………………………….…………………….96

Figure 2. JOAC/ASB Family of Plans…………………………………………………………....102

Figure 3. U.S. Air Sea Battle Strategy……………………………………………………………104

Figure 4. Joint Operational Access Concept……………………………………………………...106

List of Tables

Table 1. List of all Military Revolutions.…………………………………….…………………….7

Table 2. The Capability/Vulnerability Paradox.……………………………………….………….14

Table 3. Characteristics of Military Revolution Resources………………………………...……15

Table 4. Characteristics of Most Dangerous Capability/Vulnerability Paradox………………...16

Table 5. Cases of Military Revolution and the Capability/Vulnerability Paradox…….…………18

Table 6. Summary of Wargames and Cyber Activity…….…………...……………...…………134

Table 7. Cyber Escalation Ladder…………...………………………………………………...…150

Chapter 1.

The Capability/Vulnerability Paradox and Military Revolutions: Implications for Computing, Cyber, and the Onset of War

Part 1: Theory

In 1991, the U.S. military shocked and awed the world with the dramatic unveiling of a strategy enabled by a new military revolution. The Information

Revolution,1 or the rise in computing power, made those states that leveraged digital capabilities as tools of modern military power extraordinarily capable. Computing power drove the development of precision-guided munitions, long-range targeting, near real time intelligence surveillance and reconnaissance (ISR), and revolutionary ability to command and control. The information revolution was the backbone upon which network centric warfare became the dominant precept for modern military employment.

And the dominance of weapons, tactics, and strategies enabled and dependent on computing power remains the central characteristic of the acquisition strategies and force postures of most modern militaries. For example, in December of 2015, U.S. Department of Defense Under Secretary for Defense, Robert Work, announced a new acquisition strategy. The “Third Offset,” Work explained, would focus on emerging technologies

1 For a review of the literature on network-centric warfare and the Information Revolution in Military Affairs, see William A. Owens and Ed Offley. Lifting the fog of war. (Baltimore: Johns Hopkins University Press, 2001); Eliot A. Cohen, "A revolution in warfare." Foreign Affairs (1996): 37-54; Eliot A. Cohen, "Change and transformation in military affairs." Journal of Strategic Studies 27, no. 3 (2004): 395-407; Erik Dahl, “Network Centric Warfare and Operational Art,” Defence Studies 2/1(Spring 2002): 17; Arthur K. Cebrowski and John J. Garstka. "Network-centric warfare: Its origin and future." In US Naval Institute Proceedings, vol. 124, no. 1(1998): pp. 28-35; Michael J. Mazarr. The revolution in military affairs: A framework for defense planning. ARMY WAR COLL STRATEGIC STUDIES INST CARLISLE BARRACKS PA, 1994; Edward A Smith, Effects Based Operations: Applying network centric warfare in peace, crisis, and war. OFFICE OF THE ASSISTANT SECRETARY OF DEFENSE WASHINGTON DC COMMAND AND CONTROL RESEARCH PROGRAM (CCRP), 2003.

like unmanned systems, big data, and human-machine teaming.2 The U.S. was doubling down on the capabilities created by computing power in the information revolution.3

However, at the same time that states invest in the capabilities created by the information revolution, they also create new vulnerabilities from the very digital architectures that enable the revolutionary military capabilities created by computing power. As a 2013 Defense Science Board and a 2014 Annual Report from the DoD’s

Operational Test and Evaluation Director warned, “with present capabilities and technology it is not possible to defend with confidence against the most sophisticated cyber-attacks.”4 Paradoxically, the very capabilities that have made states so dominant during the information revolution also create novel vulnerabilities for digitized militaries.

From mission planning software to undersea cables, digital links between satellites, and data networks that distribute both blue and red near real time information, the information revolution creates incentives for first strike that could destabilize conventional balances of power. Indeed, as states continue to develop weapons reliant on computers, so also do states invest in computer network attacks and broader offensive cyber capabilities to capitalize on the vulnerabilities created by otherwise conventionally dominant digitized militaries.

2 Deputy Secretary of Defense Bob Work, “The Third U.S. Offset Strategy and its Implications for Partners and Allies,”January 28, 2015: http://www.defense.gov/News/Speeches/Speech-View/Article/606641/the- third-us-offset-strategy-and-its-implications-for-partners-and-allies. 3 Jacquelyn Schneider. “Digitally-Enabled Warfare: The Capability-Vulnerability Paradox,” Center for a New American Security. August 29, 2016: http://www.cnas.org/digitally-enabled-warfare. 4 DoD Operational Test and Evaluation, FY2014 Annual Report, http://www.dote.osd.mil/pub/reports/fy2014/pdf/other/2014DOTEAnnualReport.pdf.; Office of the Under Secretary of Defense for Acquisition, Technology, and Logistics, Resilient Military Systems and the Advanced Cyber Threat, January 2013, http://www.acq.osd.mil/dsb/reports/ResilientMilitarySystems.Cyber-Threat.pdf.

What the initial optimistic proponents of the Information Revolution may not have realized was that in their fervor to leverage computing, they failed to take into account the fact that military revolutions quite often bring with them not only new capabilities but also new vulnerabilities. This balance between the capabilities and the vulnerabilities created by military revolutions may have significant implications for conflict initiation. As with all revolutions—whether they are societal, technical, or military—with massive changes for individuals, militaries, and states comes the potential to destabilize international interactions. Understanding the lessons of these past revolutions and their impact on the onset of war can help us explain and predict the coming impact of the implication of the information revolution on international stability.5

In this piece, I examine the relationship between military revolutions and conflict initiation in order to draw lessons that may help predict the future impact of the information revolution on international stability. In examining historical cases of military revolution, I identify a capability/vulnerability paradox that suggests the degree of capability created by a military revolution combined with the excludability of associated vulnerabilities explains the potential for instability from a military revolution. This historical approach towards understanding military revolutions and conflict helps correct a sizeable omission in the current literature on the impact of computer technologies, cyber, and war.

This literature has generally bifurcated into two separate types of arguments.

One, which focuses on digital technologies and conflict, concludes that the emergence of

5 This piece uses a concept of stability linked to the onset of war. When states are more likely to attack one another, the international system is less stable. Similarly, dyads and crises between dyads are less stable when there is a greater potential for armed conflict. Therefore, for the sake of this analysis probability for conflict initiation (or onset of war) are synonymous with crisis stability.

digital technologies on the battlefield increases military effectiveness but has little to say about the subsequent impact on conflict.6 The second, which focuses on cyber as set of operations conducted within the cyberspace domain, tends to analyze cyber separate of conventional operations and therefore has been unable to come to an understanding about the role that these cyber operations may have on conflict initiation.7 This research seeks to solve some of the limitations of our current understandings of the importance of the information revolution by exploring these perspectives together as both vital components of the information revolution. In exploring both sides of the cyber phenomenon as the information revolution, I find a paradox—common to other military revolutions—in which the information revolution creates states that are militarily extremely capable and yet extremely vulnerable because of the increasingly excludable nature of information.

This creates a dangerous incentive for conflict initiation.

Below, I review the literature on military revolutions and the onset of war. I then introduce the theory of the capability/vulnerability paradox and illustrate the paradox through four military revolution case studies: levee en masse, the land war revolution, the mechanization revolution, and the nuclear revolution. Finally, I apply the historical analogy and theoretical paradox to the information revolution in order to draw

6 Paul T Mitchell and Jeffrey Bardzell, Eds.Network Centric Warfare: Coalition operations in the age of US military primacy. (NewYork: Routledge, 2008); Cebrowski and Gartska, "Network-centric warfare: Its origin and future"; Owens and Offley, Lifting the Fog of War. 7 R.A. Clarke and R. Knake, Cyber War: the next threat to national security and what to do about it (New York: HarperCollins, 2010); Gary McGraw, “Cyber War is Inevitable (Unless we Build Security In),” Journal of Strategic Studies 36, no. 1 (2013): 109-119; Thomas Rid. "Cyber war will not take place." Journal of strategic studies 35, no. 1 (2012): 5-32; Jon R. Lindsay. "Stuxnet and the limits of cyber warfare." Security Studies 22, no. 3 (2013): 365-404; Erik Gartzke. "The myth of cyberwar: bringing war in cyberspace back down to earth." International Security 38, no. 2 (2013): 41-73.

conclusions about the role the information revolution, computing, and cyber will have on international conflict dynamics.

Military Revolutions and Conflict Initiation

What is a military revolution? Andrew Krepinevich provides a useful definition and the established understanding for what is more colloquially known within U.S. defense policy circles as revolutions in military affairs.8 For him, military revolutions are

“what occurs when the application of new technologies into a significant number of military systems combines with innovative operational concepts and organizational adaption in a way that fundamentally alters the character and conduct of conflict. It does so by producing a dramatic increase—often an order of magnitude or greater—in the combat potential and military effectiveness of armed forces.”9 Military revolutions are often instigated by technology but—unlike innovations—change the way in which states fight wars so completely that the technological innovation creates new operational concepts, tactics, and ultimately balances of capabilities. Military revolutions, by their very nature, make states more capable. In fact it is the increased military effectiveness that comes with military revolutions that makes these innovations necessary for states

8 The term revolution in military affairs (or RMA) was in vogue in the 90s and early 2000s especially within the U.S. defense sector. In this analysis I am using the more general term military revolution because it is less connected with the particular RMA defense policies of that time period. However, military revolution as I am using it is interchangeable with RMA. 9 Andrew Krepinevich, “Cavalry to Computer: The Pattern of Military Revolutions,” National Interest (1994): pg. 30. For other definitions of military revolution and RMA see, Colin S. Gray, Strategy for Chaos: Revolutions in Military Affairs and the Evidence of History, (Frank Cass: London, 2002); Geoffrey Parker, The Military Revolution: Military Innovation and the rise of the West 1500-1800, (Cambridge: Cambridge University Press, 1996); and Macgregor Knox and Williamson Murray, The dynamics of military revolution 1300-2050, (Cambridge: Cambridge University Press, 2001); Steven Metz and James Kievit, Strategy and the Revolution in Military Affairs: From Theory to Policy (Carlisle: Strategic Studies Institute, 1995).

that wish to be viable competitors in the international system. As Knox and Murray argue, military revolutions “alter the capacity of states to create and project military power. And their effects are additive. States that missed the early military revolutions cannot easily leap-frog to success in war by adopting the trappings of Western technology.”10

Krepinevich identifies ten military revolutions, with the information revolution as the possible 11th (see Table 1). While all of these periods are similar in that they ushered in a fundamental change to the conduct of war, their differences are also illustrative.

First, for most of these revolutions, technology enabled or instigated innovation.

However, many revolutions were catalyzed by or fully implemented not solely because of technology but also because of societal advancements. The most expansive military revolutions are those that occur in concert with societal revolutions because these revolutions affect both the integrity of civil society and international security. For example, levee en masse was a military revolution only possible because a societal revolution allowed states to harness democratic and politically active populations to create revolutionarily large militaries. In this case, society was the catalyst for military revolution. But for other revolutions, changes in society run concurrently with changes in the conduct of war and both are facilitated by technological innovation. For the rise of land warfare, it was the overall industrialization of society through technological innovation that enabled capabilities, like railroads and telegraphs, to revolutionize war.

What is also distinct about these revolutions is their temporal variance. In general, the revolutions are not immediate reactions to technological change. Instead, some occur

10 Knox and Murray, pg. 7.

over decades, others in just a few years. Additionally, some of these military revolutions occurred in war, others in peacetime, and others received their genesis in war but would not show their effect until peace. Some of these revolutions are offensive (for example the rise of mechanization) while others, like the revolution in fortification during the 15th and 16th century, are defensive.

Table 1. List of military revolutions11

Scholars have spent a sizeable amount of effort explaining the evolution of these military revolutions and the variables that influence when and how a military revolution occurs.12 They have even devoted significant study to how these military revolutions

11 Krepinevich, “Cavalry to Computer: The Pattern of Military Revolutions.” 12 Gray, Strategy for Chaos: Revolutions in Military Affairs and the Evidence of History; Parker, The Military Revolution: Military Innovation and the rise of the West 1500-1800; Knox and Murray, The dynamics of military revolution 1300-2050; Williamson R. Murray and Allan R. Millett. Military innovation in the interwar period (Cambridge: Cambridge University Press, 1998); Wim Smit and John Grin, Military technological innovation and stability in a changing world: politically assessing and

impact military effectiveness and balance of capabilities between states.13 Despite this rich literature on the development of military revolutions and their impact on military effectiveness, little research has been conducted specifically on military revolutions and their impact on conflict initiation. When do these military revolutions increase the chance for conflict and when do they stabilize?

Though the literature on military revolutions and conflict initiation is sparse, there has been extensive research on the variables that influence the onset of war. What might this literature on military capabilities and conflict initiation suggest about the relationship between military revolutions and conflict initiation? The simplest explanation is that states that invest the greatest resources in their military will also be the most likely to win conflicts and, subsequently, generally the most likely to deter conflicts. Balance of power theorists who generally aggregate military resources argue that larger imbalances in resources will lead to short-term stability with the greatest chance of conflict occurring as a rising state begins to threaten the material lead of an existing hegemon. In this world, a state that invests more material in their military is more capable than a state that invests less. For most balance of power theorists, one unit of military material has the same effect on military balance as any other unit of military material.14 Military revolutions therefore are only impactful to stability in so much as they influence states to

influencing weapon innovation and military research and development. (Charlotte: Virginia University Press, 1992); Metz and Kievit. Strategy and the Revolution in Military Affairs: From Theory to Policy. 13 Allan R. Millett and Williamson Murray, eds. Military Effectiveness: Volume 2, The Interwar Period (Cambridge: Cambridge University Press, 2010); Stephen Rosen. Winning the next war: Innovation and the modern military. (Ithaca: Cornell University Press, 1991); Max Boot. War made new: technology, warfare, and the course of history, 1500 to today. (New York: Penguin, 2006). 14 Kenneth Waltz, Theory of international politics. (Long Grove: Waveland Press, 2010); Kenneth Waltz, Man the State and War (New York: Columbia University Press, 1969); Robert Gilpin. War and change in world politics (Cambridge: Cambridge University Press, 1981); John J. Mearsheimer. The tragedy of great power politics (New York: WW Norton & Company, 2001); Bruce Bueno de Mesquita, War Trap (New Haven: Yale University Press, 1981).

invest more or less resources to military capabilities. This literature, therefore, is unsatisfying in explaining when conflict occurs outside shifts in material balances of power.

Indeed, as many have noted, this simplistic aggregation of military capabilities doesn’t take into account the large role that intervening variables have on subsequent balances of power.15 The ability to adapt and respond to military revolutions shape the overall balance of power between states and the character of the conduct of the war— whether it be in the type of campaign, the weapons, or the impact on society—plays a large role in decisions states make to initiate conflict. Perhaps the most explicit theory that inks the character of war with the potential for conflict initiation stems from offense- defense balance.

Offense-defense balance (ODB) derives from the larger security dilemma theory that explains why security-seeking states may end up in conflict even when strict balance of power theories might expect peace. The crux of the dilemma is that states may inadvertently find themselves in conflict when their attempts to feel secure (particularly by investing in weaponry and campaigns that are offensive in nature) inadvertently make another state feel less secure.16 For security dilemma scholars, it is not simply the

15 Stephen Biddle. Military power: Explaining victory and defeat in modern battle (Princeton: Princeton University Press, 2010); Stephen Biddle and Stephen Long. "Democracy and Military Effectiveness A Deeper Look." Journal of Conflict Resolution 48, no. 4 (2004): 525-546; Stephen Rosen, "Military effectiveness: Why society matters." International Security 19, no. 4 (1995): 5-31; Dan Reiter and Allan C. Stam. "Democracy and battlefield military effectiveness." Journal of Conflict Resolution 42, no. 3 (1998): 259-277; Millett and Murray, eds. Military Effectiveness: Volume 2, The Interwar Period; Caitlin Talmadge, The Dictator's Army: Battlefield Effectiveness in Authoritarian Regimes (Ithaca: Cornell University Press, 2015). 16 Charles Glaser, "The security dilemma revisited." World politics 50 (1997): 171-201; Charles Glaser and Chairn Kaufmann. "What is the offense-defense balance and how can we measure it?." International security 22, no. 4 (1998): 44-82; Robert Jervis, "Cooperation under the security dilemma." World politics 30, no. 02 (1978): 167-214; John Herz, “Idealist internationalism and the security dilemma,” World Politics 2, no. 02 (1950): 157-180; Stephen VanEvera, Causes of War: Power and the Roots of Conflict (Ithaca: Cornell University Press, 1999); George Quester, Offense Defense in the International

aggregation of military capabilities that matters for states when they make conflict decisions. Instead, it is the nature of the weaponry and—in particular—the perception of offensive and defensive balances that impacts many conflict decisions (and especially those that cannot be explained by purely material balances of power). Therefore, for

ODB scholars the offensive or defensive character of military revolutions explains the potential for conflict initiation. If a military revolution benefits the offense, states are more likely to end up in war whereas if a military revolution benefits the defense, we are more likely to see stability.

Unfortunately, ODB has potential flaws that make it an unsatisfying explanation for many scenarios. In particular, critics argue that the difficulty in determining which campaigns and weapons are offensive and which are defensive (offense-defense differentiation) make ODB an unwieldy and useful paradigm.17 Indeed, significant debate exists about the offensive or defensive advantage within many of Krepinevich’s military revolutions. Did the rise of the railroad benefit the defense as Van Evera argues or did it lead to offensive campaigns in colonial and civil wars? In truth, quite often states are unable to effectively differentiate between offense or defense and case studies have suffered from an inability to definitively linking perceptions of offense and defense to cases of conflict initiation.18 In the end, many of the analyses ODB is applied to spend more time debating the state of the offense defense balance (or how to measure it) than in

System (New York: John Wiley Publishers, 1977); Charles L. Glaser, “Realists As Optimists: Cooperation As Self Help,” International Security, Vol. 19, No. 3 (Winter 1994/95), pp. 50–90.

17 Sean M. Lynn-Jones. "Offense-defense theory and its critics." Security Studies 4, no. 4 (1995): 660-691; Keir A. Lieber, "Grasping the technological peace: The offense-defense balance and international security." International Security 25, no. 1 (2000): 71-104. 18 Robert Jervis, "Was the Cold War a Security Dilemma?." Journal of Cold War Studies 3, no. 1 (2001): 36-60.

tying conflict decisions to cases of conflict initiation. To further complicate ODB, studies in the development of doctrine and perceptions of offense/defense show that states often misperceive offense and defense dominance for a variety of reasons completely unrelated to the technological nature of war.19

Perhaps then what ODB is really saying, and what Van Evera argues convincingly, is that states are more likely to initiate conflict when they see an advantage in striking first. ODB scholars link this first strike advantage to perceptions of offensively advantaged campaigns, but in reality the truth is probably simpler (and bypasses much of the complexity of measuring offense and defense). States are more likely to go to war with one another when both a significant vulnerability exists and capabilities exist to exploit the vulnerability. This pattern of capabilities and vulnerabilities is what we see in a lot of the nuclear literature on escalation and inadvertent war. For these nuclear scholars, a dangerous paradox existed in which states could find themselves at war because of asymmetries in vulnerability.20 And while the focus has been on nuclear capabilities, the literature suggests that whenever vulnerabilities are so pervasive that they negate a state’s ability to exert military power, there may be an incentive for first strike. Therefore, the question becomes—when do the capabilities and vulnerabilities created by military revolutions create the incentive for first strike?

19 Stephen Van Evera., "The cult of the offensive and the origins of the First World War.” International security 9, no. 1 (1984): 58-107; Jack Snyder, "Civil-Military Relations and the Cult of the Offensive, 1914 and 1984." International Security 9, no. 1 (1984): 108-146; Barry Posen, The Sources of Military Doctrine (Ithaca: Cornell University Press, 1986); Elizabeth Kier, “Culture and military doctrine: France between the wars,” International Security 19, no. 4 (1995): 65-93. 20 See for example, Barry Posen, Inadvertent Escalation: Conventional War and Nuclear Risk (Ithaca:. Cornell University Press, 1991) and Stephen Cimbala, First Strike Stability: Deterrence after Containment, (New York: Greenwood, 1990).

The Capability/Vulnerability Paradox

In general, in order to maximize military effectiveness, states seek to increase their capabilities while minimizing their vulnerabilities.21 Normally the investment in a military resource creates either an overall increase in capability or, after tit for tat counter weapons development, a null change in capability. However, military revolutions may create a paradox in which vulnerabilities created by a resource outweigh the capabilities, and yet the state still has a rational reason to invest in the resource that enables the military revolution. This is because military revolutions increase capabilities so exponentially that states cannot expect to win a conflict without implementing the military revolution. Though the vulnerabilities created by the resource may outweigh the capabilities, the vulnerabilities are only a concern if an adversary is able to keep the state from accessing the resource required by the military revolution pre-emptively and to a significant enough degree to negate the capabilities created by the military revolution. If the military revolution creates enough of a capability advantage that a revolution-utilizing state can effectively obtain their objectives before the adversary has the ability to attack their vulnerabilities, then investing in a vulnerability-increasing resource makes sense.

There may be uncertainty about whether an adversary has the will or the capability to launch preemptive strikes against the resource-induced vulnerabilities, but there is certainty that without the capability created by the resource the revolution- utilizing state cannot win conflict—whether they are attacked or the attacker. States that choose to invest in the resources of the military revolution may win if they attack first and

21 For more on this understanding of vulnerabilities, see John A. Gentry, How Wars are Won and Lost: Vulnerability and Military Power (Santa Barbara: Praeger, 2012).

they will win against adversaries that don’t have the capability to attack the vulnerability.

Therefore, regardless of the additional vulnerability that the resource introduces, a state choosing whether or not to embrace the military revolution cannot ignore the associated changes in capability conferred by the military revolution and must therefore accept that they are more capable and less secure. Faced with the predicament of losing all conflicts if they don’t invest in the resource required for the military revolution, states have to be willing to accept the potential vulnerability.

When a state is in the capability/vulnerability paradox, it must invest in the resource in order to compete against other states. That resource increases capabilities so much so that a state cannot not invest in the resource and expect to win in conflict— whether it is an attacker or a defender. However, the investment in that resource and the new capability introduces a vulnerability that threatens not only the new resource, but also the entire campaign. This logic introduces a paradox in which there are two incentives for first strike. First, in a potential crisis scenario, a state that is less capable has an incentive to strike first at a more capable state because the less capable state knows it cannot survive unless it is able to cripple the resource-enabled state’s advantage.

The vulnerabilities introduced by the military revolution make a pre-emptive strike in this sort of asymmetric relationship very compelling for a less-capable state. Second, as a more capable military revolution-enabled state moves closer to dependency on the same resource that creates the vulnerability, it is also incentivized to make a first move because it cannot effectively operate without access to the resource and is aware of its vulnerability.22 The more capable state must use its conventional dominance to

22 Steven Metz and James Kievit describe this as the breakthrough point in the adoption of military revolutions in, Strategy and the Revolution in Military Affairs: From Theory to Policy.

preemptively destroy the adversary’s first strike weaponry. Therefore, in order to maximize their chances of military victory, both the more powerful and the less powerful state have an incentive to move first in a crisis.23

Therefore, the capability/vulnerability paradox created by military revolutions is most dangerous when states are both highly dependent on a resource in order to utilize the capabilities of a military revolution and when that state’s access to ability to use the resource is vulnerable to an adversary’s exploitation. This degree of dependence on the resource for military capability as well as the exploitability of the vulnerability interact to predict the overall impact of the military revolution on conflict initiation.24

Table 2. The Capability/Vulnerability Paradox

The Resources of Military Revolutions and the Capability/Vulnerability Paradox

Of course not all military revolutions create the kind of capabilities and vulnerabilities that lead to this paradox. The capability/vulnerability paradox is contingent on two aspects of the military revolution: 1) the degree of dependency on a resource in order to create the capabilities of a military revolution, and 2) the ability of the adversary to keep the state from accessing or using the resource required for the

23 See similar logic as it applies to nuclear first strike in Robert Jervis, The Meaning of Nuclear Revolution, (Ithaca: Cornell University Press, 1990), pg. 145. 24 Schneider. “Digitally-Enabled Warfare: The Capability-Vulnerability Paradox.”

military revolution. It is important at this point to illustrate what these resources are that become pivotal to the capability/vulnerability paradox. Resources are the goods required to implement or utilize a military revolution. They are not necessarily the weapons or technologies that define the revolution—quite often they are the goods necessary to create the weapon or to utilize a new technology or operation. For example, the resource required to utilize the heavy machinery of the mechanization revolution was oil while the resource required for levee en masse was the public support required for the massive armies of the age.

Table 3. Characteristics of Military Revolution Resources

Some revolutions may have clearer pivotal resources that they rely on. It is that dependency on the resource that makes up the first aspect of the capability/vulnerability paradox. What are the characteristics of a military revolution that might induce or exacerbate dependency on a particular resource? First, dependency on the resource is more likely to occur when the resource creates a capability that is integrated in every major part of the conduct of war—across domains and operational levels of combat. This moves a state closer towards dependency on the capabilities created by the revolution and

makes it more likely that the state will double-down on efforts to extend the advantages it receives from the military revolution. Further, in some cases the dependency on the resource will become so pervasive that without that resource the capabilities are not just degraded, but instead are completely negated. For example, imagine a drone that is reliant on satellite link in order to function. If the link to the satellite doesn’t work, the drone is not proportionally less effective but instead is completely ineffective. Finally, in some cases the resource required for the military revolution is also the pivotal resource to enable a societal revolution. In that case, the dependency on the resource becomes existentially dangerous because of its effects not only military capacity but also on societal survival.

Table 4. Characteristics of Most Dangerous Capability/Vulnerability Paradox

Perhaps more importantly, a military revolution is most likely to create the paradox when an adversary is able to keep a state from either accessing or using the resource required for the military revolution. Not all resources can be affected by the adversary. For instance, the military revolutions of fortification and sail introduced novel vulnerabilities for states, however they were largely vulnerabilities that could not be exploited by adversaries. With enough resources, any state could build the fortifications.

If, however, it had been case that the fortification revolution required a specific type of material that was in limited quantities, then the revolution might have led to a

capability/vulnerability paradox.25 Similarly, the sail revolution also introduced significant new vulnerabilities to the operation of navies. However, that vulnerability— weather—was not something that could be affected by the adversary. So while the weather vulnerability impacted the overall capabilities created by the sail revolution, it left adversaries with no means to exploit the vulnerability.

What are the characteristics of resources that can be affected by adversaries?

These characteristics are tied to the ability of an adversary to cut off access or degrade the ability to utilize the resource. For example—how finite is the resource? Can the resource be created? For instance access to saltpeter for the gunpowder revolution was finite, but access could be increased by investing in more mills.26 Saltpeter itself was not a finite good. That’s in contrast to natural resources like uranium or oil that is not generally synthetic and therefore is necessarily finite. Related, is a resource plentiful or scarce?

Coal is finite, but is generally pretty plentiful. On the other hand, the materials to create the English longbow were plentiful but the amount of archers able to wield the longbow

25 In fact, we do see some evidence that this could have occurred even with the fortification revolution. As fortifications grew more complex and technically difficult to construct, there were also fewer craftsmen able to create the fortifications that created the revolutionary defensive advantage. If the adversary had been able to exploit this vulnerability about the lack of skilled workers, then the fortification case would be a candidate for the capability vulnerability paradox. However, despite the lack of skilled workers in some nations, no one nation was able to exclude those workers such that states with less workers would feel the need to conduct a pre-emptive strike to ensure their access to workers. The Germans, in fact, were especially vulnerable to this because they had few native workers able to build and maintain the new fortifications. See Parker, The Military Revolution: Military Innovation and the Rise of the West 1500- 1800. 26 Erna Risch, “Supplying Washington’s Army,” Center of Military History, US Army (Washington, D.C.: US Army, 1981); David L. Salay, ”The Production of Gunpowder in Pennsylvania During the American Revolution,” The Pennsylvania Magazine of History and Biography, Vol. 99, No. 4 (Oct. 1975); Orlando W. Stephenson,”The Supply of Gunpowder in 1776,” The American Historical Review, Vol. 30, No. 2 (Jan. 1925); Jack Kelly, Gunpowder: Alchemy, Bombards, and Pyrotechnics: The History of the Explosive that Changed the World (New York: Basic Books, 2004); Jimmy Dick, “The Gunpowder Shortage,” Journal of the American Revolution, September 9, 2013: https://allthingsliberty.com/2013/09/the-gunpowder- shortage/.

was scarce.27 The more plentiful a resource is, the more difficult it is for an adversary to completely cut off access to the resource. The more scarce, on the other hand, the less potential means for access and therefore the greater vulnerability to exploitation.

Secondly, is the resource available only at certain times? If so, then that also makes it more likely that a state can influence access to the resource because there are limited windows of availability. Additionally, is the resource located within the borders of a state, an ally, or an adversary? How proximate is the resource? Finally, the fungibility of the resource increases the chance that it can be affected by an adversary. This is related to the finiteness of the resource, but in general it is the role that physicality plays in creating logical and geographic chokepoints for exploiting access or use of a resource.

Table 5. Cases of Military Revolution and the Capability/Vulnerability Paradox

How do existing cases of military revolution demonstrate these resource characteristics? Below I examine four cases of military revolutions with various degrees of dependencies and vulnerabilities. Each case demonstrates a different dynamic of high and low vulnerability and capability and therefore helps trace the impact that resources play in creating incentives for instability.

27 John J. Mortimer, Jr., Tactics, Strategy, and Battlefield Formation During the Hundred years war: the Role of the Longbow in the ‘Infantry Revolution.’ Masters Thesis. Indiana University of Pennsylvania, August 2013.

Part 2: Historical Cases

Land War Revolution and Railroads: Low-Medium Dependency, Low-Medium

Vulnerability

The mid to late nineteenth century introduced a series of technologies that changed the course of conflict—from breech loading rifles to the telegraph and finally the railroad. The railroad provides a particularly interesting case of the capability/vulnerability paradox. States were dependent on railroad operations to execute many operational campaigns. However, railroads were often substitutable and so as states moved into more vulnerable territory they were often able to move to railroads or rivers in order to mitigate the vulnerabilities of the roadways. Similarly, until the advent of airpower, railroads inside a state’s controlled territories were difficult targets for adversary attacks. Therefore, while railroads were important resources to the land war revolution, they did not substantially change the propensity for conflict initiation.

The introduction of railroads to warfare began in Prussia but really started to revolutionize warfare in the American Civil War. Faced with transporting large armies across the expansive distances of the United States, both the Union and Confederacy realized the advantage of railroads for mobility. As General Sherman, leader of the

Union forces that conquered much of the South, recounted, his successful campaign into the heart of the Confederacy, “would surely have been impossible without the use of the railroads from Chattanooga to Atlanta . . . that single stem of railroad supplied an army of

100,000 men and 35,000 horses for the period of 196 days . . .”28 By Sherman’s

28 As cited in Nicholas Faith, The World the Railways Made, (New York: Carroll & Graf Publishers, 1990), pg. 315

estimation, his use of the railroad effectively replaced the work of 36,800 wagons and

220,800 mules.29 This allowed large-scale movement of troops into enemy territories without the logistical leash of animals previously used for transport.

Railroads not only solved the problem of taking care of large groups of animals, it also made traversing terrain faster and more reliable.30 Now the huge armies created by levee en masse had the logistical backbone to extend their range and conduct continental conflict. As Showalter explains, “railroads could buy time. They could counterbalance geography. They made possible a new approach to concentration by deploying forces simultaneously to widely separated areas outside the projected theater of operations, then moving them forward into enemy territory.”31

With the lessons from both nascent use by Prussia and the more developed use in the American Civil War, the use of railroads in the land war revolution proliferated across the globe. Russians, Chinese, British, and Austro-Hungarians raced across territories in a race to lay lines of steel and project military power via the steam engine.32

Despite huge expense and logistical requirements to implement the land war revolution, states viewed the use of railroads as existential to their ability to conduct war. As the

29 Christopher R. Gabel, Railroad Generalship: Foundations of Civil War Strategy, (Fort Leavenworth: Combat Studies Institute, 1997). 30 Mark Grimsley, “Surviving Military Revolution: The U.S. Civil War,” The Dynamics of Military Revolution, Eds. Macgregor Knox and Williamson Murray (Cambridge: Cambridge University Press, 2001); George Edgar Turner, Victory Rode the Rails (Westport: Greenwood Press, 1992); Thomas Weber, The Northern Railroads in the Civil War (New York: King’s Crown Press, 1952); Dan Lee, The L&N Railroad in the Civil War: A Vital North-South Link and the Struggle to Control It, (Jefferson: McFarland & Company, 2011); Robert Hodges Jr, American Civil War Railroad Tactics, (Oxford: Osprey Publishing, 2012); Lawrence E. Estaville, Jr., “A Strategic Railroad: The New Orleans, Jackson and Great Northern in the Civil War,” Louisiana History: The Journal of the Louisiana Historical Association, Vol. 14, No. 2 (1973): 117-136. 31 Dennis E. Showalter, “The Prusso-German RMA, 1840-1871,” in Eds. Knox and Murray, The Dynamics of Military Revolution, pg. 105). 32 Martin Van Creveld, Supplying War: Logistics from Wallenstein to Patton, (Cambridge: Cambridge University Press, 2004); Christian Wolmar, Engines of War: How Wars Were Won and Lost on the Railways, (New York: Christian Wolmar, 2010).

Russian War Minister advocated about the development of the Trans-Siberian railroad,

“Railways are now the strongest and most decisive element of war. Therefore regardless even of financial difficulties, it is exceedingly desirable to make our railway network equal to that of our enemies.”33

But with these capabilities also came vulnerabilities. First, the inflexibility of railroads and their slavish adherence to timetables severely constricted maneuver and created campaign vulnerabilities for states reliant on railroad transportation.34 As Gabel recounts “paradoxically, at the level of the individual field armies, railroads actually restricted maneuver. Field armies tended to bunch up around their railheads. One reason for this was the interface at the railhead of two very different modes of transportation.

Up to the railhead, supplies and reinforcements traveled on the industrial-age railroad.

Beyond the railhead, transportation depended upon muscle power . . . Like water behind a dam, armies gathered in large, nearly unassailable masses around their railheads . . . no wonder that so much of the fighting in Virginia occurred within the immediate area of

Fredericksburg!”35

Secondly, the inflexible nature of the railroads made them easy targets for sabotage and attack—especially when used in offensive operations.36 Even Sherman chose rivers over railroads when deep in Southern territory, explaining the vulnerabilities of a , “railroad running through a country where ever house is a nest of secret, bitter enemies’ . . . bridges and watertanks burned, trains fired into, track torn up . . . engines

33 Faith, The World the Railways Made, pg. 314 34 Ibid. 35 Gabel, Railroad Generalship: Foundations of Civil War Strategy, pgs. 6-7. 36 Stephen Van Evera, “Offense, defense, and the causes of war,” International Security 22, no. 4 (1998): 5- 43.

run off and badly damaged.”37 These vulnerabilities led, in the American Civil War, to a series of campaigns aimed at exploiting railroad dependencies. These railroad campaigns severely disadvantaged the Confederacy and had at least an indirect effect on Lee’s final surrender at Appomattox.38 Similar tactics were also influential in both Europe and

South Africa against the British. Faith recounts, “during the winter of 1870-1871 French saboteurs ensured that a whole army corps had to be employed guarding the lines, and the

Prussians had to build a loop line round a tunnel at Nanteuil-sur-Marne when six mines exploded by francs-tireurs filled it with 4,000 cubic yards of sand” while “The Boers naturally exploited the British weakness with a special Irish brigade dedicated to wrecking the line—and ensuring that much of the army’s strength was dissipated in manning hundreds of hot, lonely, often ineffective blockhouses guarding it.”39

The use of railroads was therefore an exploitable vulnerability. However, (until the advent of airpower) railroads’ vulnerability was only a factor when the lines were used in unfriendly territory. Otherwise, railroads within friendly territory were much more difficult targets for adversary attacks. Therefore, colonialists, countries in the midst of civil wars, or states that used railroads that journeyed through enemy territory were most affected by the capability/vulnerability paradox. This was especially true for a state like Russia in the 19th century that was concerned with securing its extensive and heterogeneous territories. And, in fact, the Russians saw the railroad as a significant factor towards instability within Asia. As the Foreign Minister wrote in 1881, “if there

37 As cited in Faith, The World the Railways Made, pg. 321. 38 Gabel, Railroad Generalship: Foundations of Civil War Strategy; John E. Clark, Jr., Railroads in the Civil War: the Impact of Management on Victory and Defeat, (Baton Rouge: Louisiana State University Press, 2001). 39 Faith, The World the Railways Made, pgs. 324-325.

was one thing that symbolized the multifarious Chinese treat to the Russians, it was plans to construct a railroad in Manchuria.”40 And, similarly, leading up to the Russo-Japanese

War, ““next to the absence of a Russian fleet, the most important factor to assist the

Japanese in their offensive strategy and to impede us was the condition of the Siberian and Eastern Chinese railways.”41

Additionally, while railroads were vulnerable to sabotage, they were often only temporarily damaged—better used as a stalling technique to slow a military than a vulnerability that could leave an entire military immobilized for long periods of times.

Further, the capability itself could be substituted—albeit with something less capable— by road and sometimes river travel. States were rarely, therefore, ever completely railroad dependent, even though railroad resources may well have enabled their military capabilities. This therefore limited the danger of the capability/vulnerability paradox especially as the development of capabilities that could attack railroads within friendly territory and inflict large-scale damage did not emerge until the advent of airpower. At this point, the subsequent rise of mechanization made railroads less existential to combat operations. And while the inflexibility of the rail system contributed to the onset of

World War I, the inflexibility did not as much necessitate the precipitation of conflict as it exacerbated other factors.42

The land war revolution, therefore, did not introduce a resource that was excludable in a way that necessitated a first strike. For the vast majority of the time in which railroads were essential to military operations, they operated within friendly

40 As cited in Steven G. Marks, Road to Power (Ithaca: Cornell University Press, 1991, pg. 205). 41 As cited in Marks, Road to Power, pg. 205. 42 David Stevenson, “War by Timetable? The Railway Race before 1914,” Past & Present, No. 162 (1999): 163-194.

territory and were therefore less vulnerable to large-scale preemptive attacks. In the long term, states that were vulnerable to large attacks—especially at logistical junctions— developed mechanized means of transporting people and supplies. By the time airpower was able to hold these capabilities truly vulnerable, states had already innovated away from railroad-dependency and therefore decreased the danger of the land war revolution to crises.

Levee en Masse and Popular Support: Low-Medium Dependency, High

Vulnerability

In 1793, faced with a shortage of military recruits to fight Austrian royalists, the

French government passed a “levee en masse”—or mass rising. Levee en masse institutionalized universal conscription and revolutionized warfare in Europe. By 1789, six years after the institution of the levee en masse, France had gone from a standing military of roughly 150,000 to over a million.43 The massive influx of personnel to the military coupled with revolutionary support (both materially and ideologically) enabled

France to defeat the smaller professional forces of its European rivals while at the same time demonstrating the ability of large armies to make the fortification revolution (and trace italienne) an increasingly irrelevant military strategy.

Levee en masse not only allowed for the large-scale sacrifice of personnel (a huge advantage in the wars of attrition that emerged after levee en masse), but also enabled the creation of large military units able to operate independently. This revolutionized the

43 Parker, The Military Revolution.

scope of objectives now attainable with the levee en masse corps. As Epstein explains, the corps that the levee en masse created “was difficult to destroy. The very size of the corps—twenty to forty thousand troops—along with its own divisional structure and the synergistic effect of its combat arms and coordinating staff, gave the corps a particular resiliency and cohesion in battle that its opponents lacked. The French estimated that a division could fight independently for several hours. A corps could fight independently for a day or more.”44

Further, the integration of populations for the defense of a nation had the potential to create a war fighting revolutionary élan. This revolutionary spirit of the conscripted military would lead to more proficient units more willing to take risk and more willing to die for their country: “the French volunteers made up for what they lacked in formal drill maneuvers through sheer revolutionary zeal, their willingness to press home attacks at the point of a bayonet, combined with an ability to think and act independently.”45 Levee en masse therefore had the potential to revolutionize both the operational and tactical levels of war, making states that adopted the levee en masse revolution exponentially more capable than their non-conscripted adversaries.

The breathtaking and seemingly unstoppable victories of the new French army under levee en masse demonstrated the revolutionary capability unleashed by major societal changes. And levee in masse certainly was a revolutionary capability predicated and dependent on societal change.46 The French revolution not only re-defined what it

44 Robert Epstein, “Patterns of Change and Continuity in Nineteenth Century Warfare,” The Journal of Military History 56 (July 1992): 375-376. 45 S.P. Mackenzie, Revolutionary Armies in the Modern Era, (London: Routledge, 1997), pg. 33. 46 Andrew Liaropoulos, “Revolutions in Warfare: Theoretical Paradigms and Historical Evidence—the Napoleonic and First World War Revolutions in Military Affairs,” The Journal of Military History, Vol. 70, No. 2 (2006): 363-385.

meant to be a citizen, it also placed its legitimacy on the ability to harness public support for the government. This came with its own set of dangers. Levee en masse was as much an attempt to defend the French revolution from external enemies as it was an attempt to focus popular anger away from the revolutionary government.47 As Lytle explains, “the debate of August 1793 over the levee en masse reflected not only a challenge offered by the hungry sections of Paris to a government fearful of explosive insurrection. It reflected also a struggle between the Dantonists, who were attempting to channel the dangerous energies of the sections into military power directed against external enemies, and the

Robespierrists, who were striving to direct these same energies into the political repression of domestic foes.”48 This focus on utilizing levee en masse to not only create militaries for external security, but also to leverage fomenting anger within the revolutionary population led the French government to ask, “What form of army structure would be most acceptable to public opinion, and particularly to volunteers themselves?”49

The answer to this question was a corps army able to conduct wars of attrition. These wars of attrition would require enormous societal support, thus fully occupying the revolution foment so concerning to the Robespierrists. But this focus on domestic inputs and foundation of societal change both created the enormous capability of levee en masse and the vulnerability, thus making this revolution a potentially dangerous case for the capability/vulnerability paradox. The inherent vulnerability of levee en masse is the need

47 Robert Roswell Palmer, The age of the Democratic Revolution: a political history of Europe and America, 1760-1800 (Princeton: Princeton University Press, 2014); R. Claire Snyder, “The Citizen-Soldier and the Tragedy of the Eighteenth Brumaire,” Strategies Vol. 16, No.1 (2003): 23-37; Scott Lytle, “Robespierre, Danton, and the Levee en masse,” The Journal of Modern History, Vol. 30, No.4 (1958): 325-337; John A. Lynn, The Bayonets of the Republic, Motivation and Tactics in the Army of Revolutionary France, 1791-94 (Urbana: University of Illinois Press, 1984). 48 Lytle, “Robespierre, Danton, and the Levee en masse,” pg. 326. 49 Alan Forrest, The Soldiers of the French Revolution, (Durham: Duke University Press, 1990, pg. 51.

for popular support. In order to create the giant armies of the era, a state needed a population that was invested in the cause. It the needed mechanisms to be able to translate that enthusiasm to conscripts that were trustworthy and capable on the battlefield, and it needed a way to control the mass capability and empowerment that resulted from a citizen-staffed military.50

The greatest vulnerability to levee en masse was the need for conscription. As

Woloch finds in his analysis of Napoleonic warfare, “conscription overshadowed every problem of administration in Napoleonic France. In an immediate sense, given its imperial ambitions, the regime's survival depended on it. More generally it was the touchstone of the relationship between state and civil society.”51 While the rhetoric of revolutionary levee en masse implied a common urge to fight for the nation, in reality conscription (especially over time and as the revolutionary fervor waned) was incredibly difficult. In fact, the government and later Napoleon spent great effort to generate the civil mechanisms to enforce the requirements of universal conscription.52 As Bergeron explains, “Resistance to military demands was the form of popular protest or disaffection most characteristic of the period . . . military disobedience was born in the years III and

IV and reached a first climax in year V. A second crest came in the Years VII and VIII, where 37 percent of these two annual classes for the whole of France were refractory, and the shortfall reached 63 percent in the Belgian departments, and even more in the two

Belgian departments . . . where the peasant wars were raging.”53

50 Barry Posen, “Nationalism, the Mass Army, and Military Power,” International Security Vol. 18, No.2 (Fall 1993): 80-124. 51 Isser Woloch, “Napoleonic Conscription: State Power and Civil Society,” Past & Present, No. 111: 101- 129 (May 1986), pg. 102 52 Michael Broers, “The Concept of ‘Total War’ in the Revolutionary-Napoleonic Period, War in History, Vol 15 (2008): pg. 252 53 Louis Bergeron, France Under Napoleon, (Princeton: Princeton University Press, 1981, pg. 104.

Beyond the pure difficulty of energizing and strong-arming a population to man the armies of the revolutionary and Napoleonic era, there was also the inherent vulnerability of providing potential political agitators with weapons. The citizen armies were full of domestic revolutionaries—some not particularly sympathetic to either the government or the often royalist military. The Duke de Biron, commander of the Army of the Rhine, captured this sentiment when he wrote to the government about his conscript soldiers: “All the generals to whom I wish to allocate them are afraid of them rather than eager to have them.”54 In fact, the early integration of conscripts led the

National Convention and the Committee of Public Safety to couple their levee en masse edict with simultaneous measures to propagandize and ensure the loyalty of the generals that would lead the civilian masses to war.55 What made the French so capable—the harnassing of popular support—was also a vulnerability not only to military effectiveness but to the survival of the regime itself.56

But could adversaries keep a nation from leveraging levee en masse by excluding access to popular support? Theoretically, the answer is yes. French adversaries did try and exploit the political vulnerability of levee en masse For example, the Austrians launched a concerted information operations campaign that targeted French soldiers and generals with royalist allegiances.57 However, these efforts were largely unsuccessful and half-hearted at best.58

54 As cited in pg. 41, Mackenzie, quote from G. Bodineier, “L’armee de la Revolution et ses transformations,” Histoire militaire de la France, vol.2, J. Delmas (ed), Paris, 1992, p. 237. 55 Mackenzie, Revolutionary Armies in the Modern Era; Jean-Paul Bertaud, The Army of the French Revolution (Princeton: Princeton University Press, 1988); Wiliam McNeill, The Pursuit of Power: technology, armed force, and society since A.D. 1000, (Chicago: University of Chicago Press, 1982). 56 Snyder, “The Citizen-Soldier and the Tragedy of the Eighteenth Brumaire.” 57 Bertaud, The Army of the French Revolution. 58 Liaropoulos, “Revolutions in Warfare: Theoretical Paradigms and Historical Evidence—the Napoleonic and First World War Revolutions in Military Affairs”; Herman Beukema, “Social and Political Aspects of

The problem with affecting access to popular support from the French levee en masse revolution was a deep-seated fear that such machinations could lead to similar popular uprisings in opponent states. In fact, this fear of the vulnerability of popular support initially led royalist states to resist levee en masse.59 As Beukema explains,

“How complete its reversal from Prussian tradition is evident from Frederick William H's rejection of a proposal to summon the levee en masse in 1794 in answer to France's resort to the "nation in arms." The Prussian king declared that it was "infinitely dangerous

[politically] to assemble such a mass of men." After Prussia's humiliation at Napoleon's hands he had no choice except to take his councilor's advice.”60 Similarly, the Austrian

Empire was slow to adopt the military capabilities of the military revolution because they were worried about the domestic vulnerabilities inherent in arming ethnic populations.61

The larger and more powerful the standing army, the more vulnerable ruling regimes were to coup d-etats and violent removals from power. As opposing states adopted the levee en masse revolution, they became equally (and sometimes more) vulnerable than the French to the exclusion of popular support. This mutual vulnerability made states reticent to act against popular support and so the vulnerabilities inherent with levee on

Conscription,” speech presented at the annual meeting of the American Historical Association, New York, NY, December 30, 1940: pg. 25; Hans Rosenberg, Bureaucracy, Aristocracy, and Autocracy: The Prussian Experience, 1660-1815 (Boston: Beacon, 1958); William O. Shanahan, Prussian Military Reforms 1786- 1813 (New York: Columbia University Press, 1945). 59 Liaropoulos, “Revolutions in Warfare: Theoretical Paradigms and Historical Evidence—the Napoleonic and First World War Revolutions in Military Affairs.” 60 Herman Beukema, “Social and Political Aspects of Conscription,” speech presented at the annual meeting of the American Historical Association, New York, NY, December 30, 1940: pg. 25; Hans Rosenberg, Bureaucracy, Aristocracy, and Autocracy: The Prussian Experience, 1660-1815 (Boston: Beacon, 1958); William O. Shanahan, Prussian Military Reforms 1786-1813 (New York: Columbia University Press, 1945). 61 Liaropoulos, “Revolutions in Warfare: Theoretical Paradigms and Historical Evidence—the Napoleonic and First World War Revolutions in Military Affairs.”

masse were more likely to create domestic limits to the military revolution than to incite adversary action.62

How did the capabilities and vulnerabilities of levee en masse impact conflict initiation? First, the extraordinary capabilities created by large armies made first-movers like France more likely to conduct offensive attacks. Some would argue that those offensive attacks were necessary because leaders needed to stoke nationalism in order to generate the conscription required for levee en masse as well as to keep these armed masses from turning against the domestic regime.63 However, once states adopted the military revolution, wars of attrition and mutual vulnerability meant that there were limited incentives for first strike on the vulnerabilities of levee en masse. Therefore, the inability to exploit levee en masse’s political vulnerability led to stasis between states.

Further, because opposing nations were largely unable or unwilling to exclude levee en masse nations from generating public support, measures to limit the vulnerability of levee en masse were generally inward focused. The solution to the vulnerability, especially as it became increasingly costly to conduct offensive attack, was to create various civilian and military-led controls over the armed forces. That is not to say that the vulnerabilities of levee en masse have been solved and are no longer a factor in the onset of war. Even today, both autocratic and democratic nations with large standing armies must develop domestic mechanisms that mitigate the inherent vulnerability of armed publics.64 This includes constitutional limits to military power,

62 Indeed Prussia rolled back many of the levee en masse initiatives after 1815. 63 Knox and Murray, The Dynamics of Military Revolution. 64 Talmadge, The Dictator's Army: Battlefield Effectiveness in Authoritarian Regimes; Caitlin Talmadge, "The Puzzle of Personalist Performance: Iraqi Battlefield Effectiveness in the Iran-Iraq War." Security Studies 22, no. 2 (2013): 180-221; Samuel P. Huntington, The soldier and the state: The theory and politics of civil-military relations (Cambridge: Harvard University Press, 1957); Snyder,"Civil-Military Relations

democratic appointment of military officers, but goes all the way up to conflict induced by regimes in order to create rally around the flag domestic support.65

These mechanisms (including nationalist-induced conflict) to threaten public support for the use of force remains a threat to states with large standing armies. Much of the information operations of the Cold War attempted to influence the support regimes needed from their militaries, in fomenting coups, and decreasing public support for international uses of force.66 In fact, the informationized warfare put forth in recent

Russia doctrine reflects a continued emphasis on influencing public support for the use of force.67 Because these threats are generally only affected by information operations

(versus kinetic military operations), they are potentially less dangerous than other military revolution vulnerabilities. Information operations are difficult to conceptualize and to respond to—quite often a state doesn’t even know they are being “struck” against until it is too late. This significantly dampers any potentially de-stabilizing effects that might follow a physical attack on a vulnerability. Further, the protracted time that it takes an effective information operation to occur doesn’t lend itself to the kind of quick decisions that would occur in inadvertent escalation. This makes levee en masse’s vulnerability, however existential it might be to both military and society, less inclined towards tit for tat preemption.

and the Cult of the Offensive, 1914 and 1984."; Peter Feaver and Christopher Gelpi. Choosing your battles: American civil-military relations and the use of force. (Princeton: Princeton University Press, 2004). 65 Posen, “Nationalism, the Mass Army, and Military Power.” 66 John Lewis Gaddis, The Cold War: A New History (New York: Penguin, 2006); Richard H. Immerman, The CIA in Guatemala: The foreign policy of intervention. (Austin: University of Texas Press, 1990); Lubna Z. Qureshi, Nixon, Kissinger, and Allende: US involvement in the 1973 coup in Chile. (New York: Rowman & Littlefield, 2008). 67 Keir Giles, Russia’s ‘New’ Tools for Confronting the West: Continuity and Innovation in Moscow’s Exercise of Power, Chatthamhouse.org, March 2016: https://www.chathamhouse.org/sites/files/chathamhouse/publications/research/2016-03-21-russias-new- tools-giles.pdf.

Nuclear Revolution and C3: Low-Medium Dependency on Resource, High

Vulnerability

Nuclear power introduced revolutionary lethality. The overwhelming ability to destroy made states that possessed the power uniquely able to hold their adversaries at bay. Conflicts could be determined in minutes as opposed to years.68 For states with the ability to create nuclear weapons, the nuclear revolution could fundamentally shift balances of power from state with much larger conventional forces. Despite the extraordinary capability created by the nuclear revolution, nuclear power was largely separate from operational and tactical levels of warfare. In addition, the mechanisms of extended deterrence and norms of non-use limited the need for many states to acquire nuclear weapons. Finally, for many nations the nuclear revolution could not provide the same societal revolution that we saw with oil or even coal. Therefore, while the nuclear revolution created extensive capabilities, I classify the overall dependency on nuclear capabilities as low to medium.

The capabilities that were created by the nuclear revolution were not without their own set of vulnerabilities. First and foremost, the resources required to create nuclear weapons are limited and therefore susceptible to adversary influence. The technical capabilities, access to specific mineral resources, and industrial capacity required to

68 For more on the nuclear revolution see: Jervis, The Meaning of the Nuclear Revolution; Robert Jervis, The Illogic of American Nuclear Strategy, (Ithaca: Cornell University Press, 1984); Michael Mandelbaum, The Nuclear Revolution: international politics before and after Hiroshima, (Cambridge: Cambridge University Press, 1981); Bernard Brodie, Frederick Sherwood Dunn, Arnold Wolfers, Percy Ellwood Corbett, and William Thornton Rickert Fox, The absolute weapon: Atomic power and the world order, (New York: Harcourt, 1946); Charles Glaser, Analyzing Strategic Nuclear Policy, (Princeton: Princeton University Press, 1990); Bernard Brodie, Strategy in the Missile Age, (Princeton: Princeton University Press, 2015); Lawrence Freedman, The Evolution of Nuclear Strategy, Vol. 20 (London: Macmillan, 1989).

develop the bomb not only self-limited states from developing nuclear weapons but also helped other states to exclude potentially interest states from nuclear proliferation. In fact, the development of the International Atomic Energy Agency and the associated nuclear Non-Proliferation treaty and non-proliferation mechanisms institutionalized exclusion from the nuclear club such that many states with the desire to create the capacity for nuclear weapons were effectively deterred.69

There are a few characteristics of this particular type of vulnerability that bear important implications for conflict initiation. First, mechanisms like the IAEA and the

NPT have created non-violent alternatives to conflict that dissuade or punish states for nuclear proliferation short of preemptive strike. Secondly, because of the ability to project large amounts of power, many non-nuclear states have been deterred from proliferating by their existence under a nuclear umbrella. Finally, because nuclear weapons are held by so few states and have been used so infrequently, non-nuclear states without nuclear adversaries may not feel the need to invest in the capability. Therefore, a sizeable amount of countries select themselves out of the nuclear capability/vulnerability paradox even before other conflict-reducing mechanisms like the NPT or nuclear alliances. That is not to say that the excludability of nuclear proliferation is so unique that it no longer creates the propensity for conflict that we would expect under the

69 There is a sizeable literature on the ability to exclude nations from the nuclear proliferation. Some notable works include, Stephen Meyer, The dynamics of nuclear proliferation, (Chicago: University of Chicago Press, 1986); Dong-Joon Jo and Erik Gartzke. "Determinants of nuclear weapons proliferation." Journal of Conflict Resolution 51, no. 1 (2007): 167-194; George H. Quester, The politics of nuclear proliferation (Baltimore: Johns Hopkins University Press, 1973); Bradley A Thayer, "The causes of nuclear proliferation and the utility of the nuclear non-proliferation regime." Security Studies 4, no. 3 (1995): 463-519; Maria Rost Rublee, Nonproliferation norms: why states choose nuclear restraint (Athens: University of Georgia Press, 2009); T.V. Paul, Power versus prudence: Why nations forgo nuclear weapons, (Montreal: McGill-Queen's Press-MQUP, 2000); Etel Solingen, Nuclear logics: contrasting paths in East Asia and the Middle East (Princeton University Press, 2009).

capability/vulnerability paradox. Preemptive strikes in Iraq, Libya, and Syria provide examples of when the capability/vulnerability paradox can provide incentives for first strike on states attempting to acquire the capability to develop nuclear weapons.

The nuclear revolution also creates vulnerabilities for states that obtain nuclear weapons. The platforms from which nuclear weapons are employed (submarines, aircraft, intercontinental ballistic missiles (ICBMs)), the facilities which produce and store nuclear materials, and the command and control systems that direct their use; are all vulnerable at some level to counterforce targeting. Debates raged during the Cold War about the types of weapons, hardening, and employment mechanisms required for a retaliatory second strike.70 The general consensus as the Cold War came to an end was that, while numbers and survivability of some of these weapons platforms and arsenals might affect crisis instability, the dominant variable in crisis instability between nuclear states was the potential to destroy the ability to command or control nuclear weapons. As

Desmond Ball concluded, “the viability of any strategic policy of controlled nuclear warfighting depends, intera alia, upon the ability of C3 systems to survive throughout the nuclear exchange.”71 It didn’t matter whether a state had ten nuclear weapons or thousands—if they were unable to launch those weapons after a first strike, they would be unable to deter a nuclear state. Therefore, the immense capabilities created by the nuclear revolution were contingent on a state’s ability to preserve command and control.

70 For literature on nuclear vulnerabilities, see for example, Albert Carensale and Charles Glaser, “ICBM Vulnerability: The Cures are Worse than the Disease,” International Security, Vol. 7, No. 1 (Summer 1982): pp. 70-85; Paul H. Nitze, "Deterring Our Deterrent," Foreign Policy, No. 25 (Winter 1976-77), pp. 195-210; Richard L. Garwin, "Will Strategic Submarines Be Vulnerable?," International Security, Vol. 8, No. 2 (Fall 1983), pp. 52-67; Charles Glaser, “Do We Want the Missile Defenses we can Build?” International Security, Vol. 10, No. 1 (Summer 1985): 25-57. 71 Desmond Ball, “Can Nuclear War be Controlled?” Adelphi Papers, no. 169 (1981): 9.

What’s worse is that, unlike the nuclear weapons platforms which were numerous, diverse, and sometimes difficult to target, nuclear C3 was vulnerable to a series of threats: nuclear effects on physical components, sabotage, attacks on satellites or early warning radars, and later cyber attacks on networks.72 Ball found that “command and control systems are inherently relatively vulnerable, and concerted attacks on them would very rapidly destroy them or at least render them inoperable . . . the Soviet Union would need to expend thousands of warheads in any comprehensive counterforce attacks against

US ICBM silos, bomber bases, and FBM submarine facilities . . . on the other hand, it would require only about 50-100 warheads to destroy the fixed facilities of the national command system or to effectively impair the communication links between the National

Command Authorities and the strategic forces.”73

Based on these assessments, Ball believed that command and control was the pivotal vulnerability for crisis stability.74 Scholars agreed that nuclear C3, because of its impact on total ability to utilize nuclear power, played a pivotal role in crisis stability. As

Jervis explains, “limited attacks on strategic C2 systems would in all likelihood produce even stronger reasons to escalate. The pressures on the state to relax the limits on its endangered systems while they could still be used would be great, especially if, as is likely to be the case, each side sees the other’s actions as an attempt to gain unilateral advantage.”75

If the U.S. and Russia had perceived each other to be mutually vulnerable to preemptive strikes against nuclear C3, then C3 vulnerability could have been a moot

72 Ibid 73 Ibid, pg.35. 74 Ibid, pg. 2. 75 Jervis, The Meaning of Nuclear Revolution, pg. 93.

issue to overall crisis stability. However, perceptions of asymmetries in vulnerability increase the chance for preemptive strike76 and these perceptions of asymmetry were particularly likely when it came to command and control. Even though the actual technical ability to conduct full exclusion of nuclear C3 was never certain, each side’s fear of the vulnerability of C3 made nuclear C3 both a dangerous target and a potential impetus for preemptive strike.77 This fear was prevalent in nuclear policy discussions and led to the belief that “The Soviet Union has long possessed the capability to destroy the

U.S. command system, and Soviet command vulnerability has probably been comparable.

Inherent in such mutual vulnerability is an appreciable degree of latent instability.

Command deficiencies weaken crisis stability because of the heavy penalty incurred by the side struck first and the tremendous advantage gained by the side that initiates the attack.”78 Further, with the delegation of command and control to allies and the extension of the nuclear umbrella, the C3 vulnerabilities—especially for the United

States—were multiplied. 79

The nuclear revolution generated multiple vulnerabilities—both to the proliferation of nuclear weapons and the ability to utilize them. However, the potential for instability from the nuclear revolution was mitigated by a series of institutional and

76 Bruce Blair, Strategic Command and Control: Redefining the Nuclear Threat (Washington: Brookings Institute, 1985); Barry Posen, Inadvertent Escalation: Conventional War and Nuclear Risks; Bruce Blair, The Logic of Accidental Nuclear War, (Washington: The Brookings Institute, 1993); Kurt Gottfriend and Bruce Blair, Eds., Crisis Stability and Nuclear War, (New York: Oxford University Press, 1988); Stephen J. Cimbala, First Strike Stability: Deterrence After Containment, (New York: Greenwood Press, 1990); John D. Steinbruner, "Nuclear Decapitation," Foreign Policy, No. 45 (Winter 1981-82), pp. 16-28; Caitlin Talmadge, “Preventing Nuclear Escalation in U.S.-China Conflict,” Institute for Security and Conflict Studies Policy Brief, February 2016: https://www.wilsoncenter.org/sites/default/files/china_policy_brief_talmadge_0.pdf. 77 Blair, The Logic of Accidental Nuclear War. 78 Blair, Strategic Command and Control: Redefining the Nuclear Threat, pg. 285. 79 Paul Bracken, The Command and Control of Nuclear Forces, (New Haven: Yale University Press, 1983).

normative developments that decreased states’ dependency on nuclear weapons for survival in the international system. Additionally, this short study reveals how important the characteristics of the vulnerability are for predicting conflict initiation. While nuclear

C2 posed a serious and dangerous vulnerability for nuclear states, uncertainty about the ability to completely cut off nuclear weapons from command and control may have limited first strike incentives. This uncertainty may be tied to the some of the virtual characteristics of command and control. While nodes of C2 are physical, the networks of how information flows are conceptual and therefore induce some level of uncertainty about the effects of attacks on even the physical nodes.80 In contrast, the means by which states keep other states from obtaining nuclear weapons can be incredibly physical and therefore convey potentially greater certainty about the success of physical operations against key nodes in nuclear proliferation.

The difference between the chance for conflict initiation pre-nuclear development and post-nuclear development may also reveal an important distinction in the capability/vulnerability paradox. While in this case, nuclear deterrence helps to explain why there haven’t been first strike attacks against nuclear C3, these two cases may have something to say about non-nuclear revolutions. It may be easier to disrupt a state’s access to a resource than to impact the utilization of that resource. Therefore, resources

80 That said, the potential for danger in influencing access to nuclear C3 has potentially become greater than it was in the pre-digital age. In the heyday of nuclear policy discussion, many scholars and decision- makers proposed norms to not target nuclear C3. However, perceptions of asymmetry in terms of nuclear C3 vulnerability remain a problem and complicate the potential for agreements that completely negate the danger. This excludability has become even more dangerous as nations upgrade their nuclear C3 to digital standards and thus introduce new vulnerabilities via network attack. I will discuss this in more depth below.

that must be obtained from outside the state are necessarily more likely to create incentives for first strike and instability.

The Mechanization Revolution and Oil

Perhaps the greatest example of a resource impacted by access is the case of the mechanization revolution and the need for oil. Internal combustion engines and mechanization opened up remarkable opportunities for weapons development: from tanks to aircraft to destroyers and submarines—combustion engines made nations exponentially more effective on the battlefield.81 But it also made them more dependent on oil and therefore vulnerable to disruptions in the oil supply chain. This paradox creates a scenario in which states without access to oil, but heavily reliant on oil to fuel their military, are induced to seek first mover attacks in order to fend off major vulnerabilities to their oil supply. Charles Glaser identifies this as a major mechanization to war, noting that for states that may be in crisis situations “oil vulnerability influences their assessment of military capabilities and, in turn, the path to war.”82

The phenomenon arose with the technological innovations of World War I. As

Yergin writes in his analysis of oil and power, “It was a war that was fought between men and machines. And these machines were powered by oil . . . for, in the course of the First

World War, oil and the internal combustion engine changed every dimension of warfare.”83 The advent of the internal combustion engine led directly to the rise of mechanized transport, the tank, the diesel-powered submarine, and the aircraft. What

81 Ronald Haycock and Keith Neilson. Men, Machines & War. (New York: Wilfrid Laurier Univ. Press, 1988). 82 Charles Glaser, “How Oil Influences National Security,” International Security 38, no.2 (2013): 118. 83 Daniel Yergin. The prize: The epic quest for oil, money & power. Simon and Schuster, 1991, pg. 168.

started out as a war dependent on steam-locomotion and the limits of human logistics, developed into a war of mechanized machines. To understand the scale of this dependency on oil, it is helpful quantify the vast amount of oil-dependent vehicles created by the end of the war. Not even including their tank inventory, the British had over 110,000 motorized vehicles. The U.S., which had only been in the war for a year and half, deployed over 50,000 oil-dependent vehicles. And a conflict that featured almost no aircraft at its initiation ended with the production of almost 200,000 gas- powered planes by Britain, France, Italy, Germany, and the United States.84 The explosion of gas-dependent machinery and its revolutionary impact on the conflict led

French Prime Minister Clemenceau in December 1917 to conclude that: “a failure in the supply of gasoline would cause the immediate paralysis of our armies [and] compel us to a peace unfavorable to the Allies.”85

The development of the internal combustion engine during the war led to a veritable arms race of oil-dependent technology during the inter-war years. Britain, a coal-rich country that had previously relied on coal-driven locomotion for its naval inventory made the conscious decision to transition completely to oil. Despite the fact that they knew they were vulnerable to attacks on oil supply, Churchill and Admiral John

Fisher argued that the capabilities conveyed by oil-power made it a necessary vulnerability. Faced with an increasingly powerful and oil-fueled German Navy, the

British Navy made a strategic choice to move its weaponry from its domestic source of energy to foreign-supplied oil. As Admiral Fisher argued, "It is a gospel fact...that a fleet

84 Ibid. 85 Ibid, pg. 177.

with oil fuel will have an overwhelming strategic advantage over a coal fleet."86 In the end, the British relied heavily on their relationship with the domestic-oil producing

United States and resources in their colonies to fuel first their navy and then their mechanized Army and gas-powered Air Force. This reliance on foreign sources of oil for combat effectiveness made the German U-boat campaign a strategic source of concern for both Great Britain and the United States. In the end, the German interdiction of these fuel transports led to the enlargement of World War II and the inclusion of the United

States as a primary ally for Britain and France in Europe.

In the Pacific, the converse also led to the engagement of the U.S. in World War

II. Japan, like Great Britain, had virtually no domestic source of oil. Also like Great

Britain, the Japanese had previously made the decision to invest in a navy and air force dependent on oil. At the same time as their naval and air forces required oil in order fend off any potential attack or blockade, Japan also needed more oil in order to support their land conquest of China and Southeast Asia. In 1941, with the United States threatening the Japanese oil supply, both the Japanese Navy and the Japanese Army advocated for a surprise attack on U.S. naval assets at Pearl Harbor. As a Japanese admiral recounted,

“if there were no supply of oil, battleships and any other warships would be nothing more than scarecrows.”87 Japan’s oil vulnerability made it impossible for them to operate an effective military campaign and therefore led the Japanese to decide that a first-mover attack against a potentially more capable United States was their only hope of ever

86 Jeff Reed, “Britain vs. Germany, Coal vs. Oil—The Coming of the Great War,” Oilpro.com, 2014: http://oilpro.com/post/4796/britain-vs-germany-coal-vs-oil--the-coming-of-the-great-war; John H. Maurer, Churchill and the Strategic Dilemmas Before the World Wars: Essays in Honor of Michael I. Handel. (New York: Routledge, 2014). 87 As cited in Yergin. The prize: The epic quest for oil, money & power, pg. 319.

defeating the U.S. in any potential conflict. As Koichi Kido, Lord Privy Seal, recalled:

“the whole problem facing Japan had been reduced to a very simple factor, and that was oil. Japan could not possibly fight a war of certain victory against the United States.”88

Not only did the Japanese military’s need for oil impact its decision to initiate a first- mover attack, but ultimately (and as they had expected) that vulnerability led directly to the decline of their operational effectiveness. As Yergin explains, “the shortage of liquid fuel was Japan’s Achilles heel.”89 Faced with dwindling oil supplies and a force completely dependent on it, the Japanese had to make inefficient combat decisions. By

1944, the lack of oil directly impacted Japan’s ability to achieve victory in the Marianas.

Not only was the Japanese naval battle fleet not able to fight at the Marianas because it couldn’t fuel itself to get to the islands, but the aircraft that was able to arrive had to take a less direct route in order to save fuel.90 In the air, Japan’s lack of oil decimated an otherwise competent air force. Because of the dwindling reserves, the Japanese decided not to train many pilots or to invest in the navigational sorties required for the pilots to complete their mission and return home. In the end, with pilots ill-prepared for missions and suffering from bad fuel or no fuel at all, the Japanese Air Force turned to Kamikaze missions. Without oil, the Japanese military was forced to use its air force and increasingly it’s navy in last-ditch suicide attempts.

Oil-vulnerability was also a determining factor in German strategy and ultimate defeat—to include Hitler’s deleterious decision to invade the Soviet Union.91 Oil was

88 Ibid, pg. 320. 89 Ibid, pg. 357. 90 Ibid. 91 Joel Hayward, “Hitler’s Quest for Oil: The Impact of Economic Considerations on Military Strategy, 1941-42,” Journal of Strategic Studies 18, no.4 (December 1995): 94-135; Anand Toprani, “The First War for Oil: the Caucasus, German Strategy, and the Turning Point of the War on the Eastern Front, 1942.” The Journal of Military History 80 (July 2016): 815-854.

such a large consideration for the Germans and their blitzkrieg style of warfare that

Dwight Eisenhower identified oil as the primary key to German effectiveness and warned that the access to oil resources would give the Axes “an overwhelming tactical superiority.”92 Hitler and his army realized this early on and invested heavily in domestic synthetic fuels. As Yergin notes, “The German army . . . lobbied for an expanded commitment to a domestic synthetic fuels industry, arguing that Germany’s own current supplies would be woefully inadequate to the requirements of the new type of warfare that it was planning.”93 Indeed, it was the inability for Rommel to receive oil in 1942 that led to his ultimate defeat. As Rommel wrote, “The battle is fought and decided by the

Quartermasters before the shooting begins . . . the bravest men can do nothing with guns, the guns nothing without plenty of ammunition, and neither guns nor ammunition are of much use in mobile warfare unless there are vehicles with sufficient petrol to haul them around.” By the end of the war, not only was the German fuel supply unable to support many land and naval operations, but the fuel shortage had decimated the Luftwaffe. In fact in 1944, despite a glut of new airplanes manufactured, the German Luftwaffe was more or less grounded. Without enough fuel to train pilots or conduct sorties, the

Luftwaffe “air operations were thereby made virtually impossible.”94

Tellingly, the reliance on oil had a significant impact on the enlargement of the war to the Pacific, to the inclusion of the United States in the European campaign, and to the German decision to invade Russia. Because oil affected naval, land, and air and

92 Peter Hoffmann, “The Gulf Region in German Strategic Projections, 1940–1942,” Militärgeschichtliche Mitteilungen 44, no. 2 (1988): 67–68, as cited in Toprani 2016. 93 Yergin, The prize: The epic quest for oil, money & power, pg. 343. 94 General Adolph Galland, commander of German fighter forces as cited in Yergin The prize: The epic quest for oil, money & power, pg. 347.

enabled tactics and operations in all these domains, the ability to use the resource made all operations exponentially more capable while the inability to use the resource made states existentially vulnerable. But what makes oil particularly prone to the danger of the capability/vulnerability paradox is the ability to exclude other nations from utilizing oil.

Unlike railroads which were most vulnerable for offensive operations or levee en masse that required an intangible information operations, cutting off another state’s access to oil is ambivalent of offensive or defensive campaigns while also being highly tangible. In fact, the fungibility of oil is so concrete that it can be calculated. States and their adversaries are keenly aware of both the extent of the dependency and the vulnerability.

This heightens the potential for pre-emptive strike by both sides. Further, states without domestic production of oil must transport oil from other oil-producing nations, creating points of vulnerability that can be exploited through the use of force. Similarly, a non-oil producing state can be excluded from access to oil when oil-producing states collude enough to bar the export of oil to the oil-vulnerable state.

The implication for conflict initiation is that the need for oil combined with the ability of adversaries to impact access to oil increases the chance for conflict.95 As

Rosemary Kelanic explains, the paradox for oil involves “the susceptibility of its oil imports to military disruption and its “petroleum deficit,” meaning the difference between the oil it controls and what it needs to meet policy objectives. The greater the susceptibility and deficit, the more likely a country is to adopt an extreme anticipatory

95 Glaser, “How Oil Influences National Security”; Jeff D. Colgan, "Oil, domestic politics, and international conflict." Energy Research & Social Science 1 (2014): 198-205; Llewelyn Hugehes and Austin Long. "Is there an oil weapon?: Security implications of changes in the structure of the international oil market." International Security 39, no. 3 (2015): 152-189; Jeff D. Colgan, "Oil and revolutionary governments: Fuel for international conflict." International Organization 64, no. 04 (2010): 661-694; Ivan L. Pearson, In the Name of Oil: Anglo-American Relations in the Middle East, 1950–1958 (Eastbourne, UK: Sussex Academic Press, 2010).

strategy because it is willing to pay high.”96 This capability/vulnerability paradox is exacerbated by the societal need for oil. Like most other military revolutions that involve both technological and societal changes, states require oil even during peacetime in order to fuel economic prosperity. Therefore, they are more sensitive to the perceptions of vulnerability and are even more likely to conduct preemptive strikes when opposing nations show the will and capability to exclude access to oil.

Kelanic recognizes this existential need that states have for oil and argues that it is the very danger of this situation that has led nations to take strategies to mitigate their vulnerability. While nations are unable to negate their dependency on oil—especially for military capabilities—they are able to decrease the ability of adversaries to impact access to the resource. For oil, this is either through domestic self-sufficiency, direct control of foreign oil resources, or indirect control.97 These decisions about decreasing excludability, while specific to mechanization and oil, have implications for other military revolutions.

Capability/Vulnerability Paradox and Military Revolution Discussion

What do these cases reveal about the capability/vulnerability paradox and military revolution? First, these cases demonstrate that the paradox exists and has impacted the potential for conflict at least since the eighteenth century. Each of the four cases of military revolution demonstrated revolutionary capability that changed the way wars were conducted. In turn, this introduced new vulnerabilities that made states susceptible

96 Rosemary Kelanic, “The Petroleum Paradox: Oil, Coercive Vulnerability, and Great Power Behavior,” Security Studies 25, no.2 (2016): 181-2013. 97 Ibid.

to both attack and defeat in novel ways. However, each case varied the degree of dependency and degree of vulnerability created by the resource required for the military revolution. The variance in these cases reveal important relationships between military revolutions and the potential for conflict initiation.

Perhaps most important to understanding the impact of military revolutions on conflict initiation, variations in the vulnerability of each resource had significant implications for the danger of the capability/vulnerability paradox. In general, the revolutions that created the greatest potential for conflict were also those that created the greatest chance for adversary states to impact access to the use of a resource required for a military revolution. Therefore, highly excludable resources like oil and nuclear command and control are more likely to lead to conflict than the less excludable levee en masse and railroads. Part of the reason that oil and nuclear C3 are more dangerous is because the act of exclusion can occur so quickly. And in fact the immediacy of the vulnerability contributes to the overall danger of the paradox. For revolutions like levee en masse, the intangible nature of information operations and the amount of time it take to make a strike can decrease not only the window of vulnerability but the immediacy of that vulnerability. These are both important considerations when understanding what would induce the more capable revolutionary-enabled state to strike first.

Interestingly, fungibility is not as important to excludability as immediacy might be. At first glance, railroads appear to have many of the same characteristics as oil. They are highly tangible, they have choke points, and they are logistical backbones for conflict.

However, because of different gauges of railroads and the inability to quickly create railroads in enemy territory, railroads were largely insulated from the kind of large scale

attacks to choke points that we have seen with oil. The huge exceptions, of course, are civil and imperial wars—but in terms of inter-state wars, railroad vulnerabilities rarely created extensive first strike opportunities that lead to the most dangerous capability/vulnerability paradoxes. In addition, the ability to substitute other means of transportation decreased overall dependency on railroads and therefore made attacks less dangerous to military capability.

Finally, the character of the military revolution affects the strategies states use to mitigate the danger of the capability/vulnerability paradox. Perhaps the least risky solution, but also sometimes untenable, is to innovate such that the capability created by the military revolution is no longer necessary. We see that happen in the case of the railroads. The rise of mechanization made the vulnerabilities of railroads less existential.

States were less dependent on the resource. But in some cases, the capability that the revolution creates is impossible to innovate away. Levee en masse, though it is less vulnerable than nuclear C3 and oil, is potentially less likely to be innovated away. In fact, the persistence of large standing armies predicated on public support over the last three hundred years demonstrates the vitality of levee en masse’s vulnerability.

Part 3:

Capability/Vulnerability Paradox: The Information Revolution and Cyber

These historical cases of military revolution provide useful illustrations of the capability/vulnerability paradox and the importance of resources for understanding the danger of a military revolution. What do these cases suggest about the information

revolution and the role that computing and cyber might have on the potential for conflict initiation?

What is the information revolution?

In Krepinevich’s 1994 piece, he identifies the Information Revolution as the final possible military revolution in his list of 11. At the time of publication, the debate about whether not the computer would revolutionize warfare was vigorous and drove much of the discussion about revolutions in military affairs within the strategic studies literature.98

Over twenty years later, the debate is largely resolved. Dramatic improvements in processing ability, computing, and interconnectivity have ushered in the information age with significant implications for both societies and conflict.99 Perhaps not since levee en masse have the revolutions in society and military affairs occurred in such step. In fact, the two have become so entwined that it is difficult to disentangle the information age’s military roots from its societal influence. What may have begun as a military project in

98 Steven Metz. Armed conflict in the 21st century: the information revolution and post-modern warfare. (Carlisle: Strategic Studies Institute, 2000); Owens, Lifting the Fog of War; John Arquilla and David Ronfeldt. Networks and netwars: The future of terror, crime, and militancy (Santa Monica: Rand Corporation, 2001); Cohen, "A revolution in warfare."; Cohen, "Change and transformation in military affairs."; Michael O’ Hanlon, Technological Change and the Future of Warfare (Washington, D.C.: Brookings Institute Press, 2000). 99 Robert Keohane and Joseph S. Nye Jr., “Power and Interdependence in the Information Age,” Foreign Affairs 77, no.5 (1998): 81-94; Joseph S. Nye, Jr., and William A. Owens. "America's information edge." Foreign Affairs 75, no.2 (1996): 20-36; David Lyon, The information society: Issues and illusions (New York: John Wiley & Sons, 2013); Peter Drucker, “Beyond the Information Revolution,” The Atlantic, October 1999: http://www.theatlantic.com/magazine/archive/1999/10/beyond-the-information- revolution/304658/; Richard Hundley, Robert H. Anderson, Tora K. Bikson, C.Richard Neu, The Global Course of the Information Revolution: Recurring Themes and Regional Variations (Santa Monica: RAND Corporation, 2003); Gerald Brock, The Second Information Revolution (Cambridge: Harvard University Press, 2003); David Lonsdale, The Nature of War in the Information Age (London: Frank Cass, 2004); Taylor Owen, Disruptive Power (Oxford: Oxford University Press, 2015); David S. Alberts, Understanding Information Age Warfare (Washington, D.C.: CCRP Publication Series, 2001).

communication has morphed into a means of organizing civil society and undergirds the increasingly digital economies of modern nations.

The Information Revolution, therefore, is an astounding example of a military revolution that has changed the conduct of war at every level. Almost every modern weapon contains digital processing capabilities—whether it is an updated computer processing unit on a radar, the digital technology that allows modern jammers to adapt to their electromagnetic environment, or the software that targeteers and combat planners use to allocate weapons and plan campaigns.100 Combat information, previously disseminated via analog and hampered by size and logistical limitations, is now sent via packets of binary code by satellite, fiber-optic cable and radio transmissions. That digital information is stored and processed through data centers filled with computers, routers, processing software, software applications, and accessed by conventional warfighters.

Information has so fully integrated itself into the way modern states fight war that it can’t be separated from tactics and operations. Unlike the infantry or artillery revolution, the information revolution didn’t just create information warriors (probably better understood as “cyber” warriors), it informationized all conventional warriors. The warfighters that utilize these information technologies don’t think of themselves as “information” operators. They are infantry soldiers, munitions experts, forward air controllers, pilots, and war-fighting staffs but they are dependent on cyber to conduct conventional operations. They are the front-line combatants, and they are armed with M-16s, radios, and combat IPads. It is virtually impossible to separate modern warfare from digital

100 For the evolving pervasiveness of computing and modern weapons, see also Rebecca Slayton, Arguments that Count: Physics, Computing, and Missile Defense, 1949-2012 (Cambridge: MIT Press, 2013).

capabilities.101 As former U.S. Secretary of Defense Lynn asserted in 2010, “information technology enables almost everything the military does . . . [it] has evolved from an administrative tool for enhancing office productivity into a national strategic asset in its own right.”102

The information revolution is both the use of information to enable conventional war fighting operations and the use of information as a weapon on its own. Therefore, the information revolution encompasses both digitized warfare and “cyber” and it is the digital information itself that is the primary resource of the information revolution. This helpful to understand what has been a significant divide, institutionally and culturally, between computing and its impact on society and militaries, and cyber—which we generally think of as operations on the networks to affect data or operations in cyberspace. This divide has needlessly kept us from understanding the true impact of the military revolution.

Perhaps most importantly, “cyber” is actually not a noun at all but an adjective to describe the digital or computer-like nature of the noun upon which it is attached (see

Oxford English Dictionary definition of cyber).103 As such, cyber describes the character of an object or concept and delineates that subsequent noun from a less automated, more

101 Robert L. Bateman, Ed., Digital War: A View from the Front Lines (Novato: Presidio Press, 1999); “Joint Vision 2020: America’s Military—Preparing for Tomorrow,” Joint Force Quarterly (Summer 2000): 57-76; Fredrick Thomas Martin, Top Secret Intranet: How U.S. Intelligence Built Intelink—the World’s Largest, Most Secure Network (Upper Saddle River: Prentice Hall, 1999); Nick Cullather, “Bombing at the Speed of Thought: Intelligence in the Coming Age of Cyber-war,” Intelligence and National Security 18, no.4 (2006); Aaron Franklin Brantly, The Decision to Attack: Military and Intelligence Cyber Decision-Making (Athens: University of Georgia Press, 2016). 102 William J. Lynn, “Defending a New Domain: The Pentagon’s Cyber Strategy,” Foreign Affairs 89, no.5 (2010): 98 103 The Oxford English Dictionary defines cyber as: “Chiefly prefixed to nouns. Originally: forming words relating to (the culture of) computers, information technology, and virtual reality, or denoting futuristic concepts. Later also: spec. forming terms relating to the Internet.”

likely to be analog, non-computer dependent term. Because cyber is an adjective, it is equally valid for analyses of “cyber’s” effect on conflict initiation to examine cyber weapons, cyber networks, cyber operators, cyber cultures, etc. For that reason, most of the focus on cyber’s impact on conflict initiation has been as cyber as a weapon. When examined from this narrow interpretation, cyber as a weapon hardly looks like the type of resource that creates a capability/vulnerability paradox. The general consensus appears to be that cyber weapons create unknown capabilities independent from the design of any scheme of maneuver or operational conduct of war.104 They cannot, therefore, possibly create the same effects as resources like oil.

However, if we expand our understanding of cyber beyond cyber-weapons to examine the cyber-enabled digital qualities of warfare, then suddenly the pervasiveness of the information revolution is apparent. This revolution is about how warfare has changed because of digital processing. That includes both the evolution of warfare in the conventional domains because of information, as well as the use of cyberspace operations to defend and attack that information within the cyberspace domain. Cyber attacks matter only because of the terrain created by computing in conventional operations while at the same time many aspects of the vulnerability of that terrain are related to the development of offensive cyber capabilities. Both sides of the information coin are

104 Rid, "Cyber war will not take place"; Gartzke, "The myth of cyberwar: bringing war in cyberspace back down to earth." Timothy Junio, “How Probable is Cyber War? Bringing IR Theory Back in to the Conflict Debate,” Journal of Strategic Studies 36, no. 1; Lucas Kello, “The Meaning of the Cyber Revolution: Perils to Theory and Statecraft,” International Security 38, no. 2 (2013): 7-40; Lindsay "Stuxnet and the limits of cyber warfare”; Jon R. Lindsay, “The Impact of China on Cybersecurity: Fiction and Friction." International Security 39, no. 3 (2015): 7-47.

essential to understanding how the information revolution fits into the capability/vulnerability paradox.

Information Revolution: Capabilities

Closely linked to the emergence of digitally-enabled weaponry and operations is the now pervasive dominance of network-centric warfare. With the advent of digital technologies and the internet, network-centric warfare proponents of the early 90s envisioned a revolution in military affairs that would utilize information technology to execute quick and dominating military operations within increasingly little need to place personnel in danger. Digital technologies could create near real time dissemination of intelligence, establish the means for constant blue force tracking, and enable quick decisions through emails and chat. Advocates of network warfare argued that the digital revolution would fundamentally alter the balance of military victory toward states that were able to achieve information dominance.105 Accordingly, the network centric warfare that developed from these precepts and that has become the standard for modern military capability, is highly precise, integrated, and dynamic. It is extraordinarily capable.

It is also extraordinarily ubiquitous in today’s first world militaries. Even states reliant on Cold War-era machinery are enhancing their analog technology with digital advancements. They have networked air, land, sea, and space capabilities and integrated sensors to disseminate information to weapons, warfighters, and commanders. And as the information revolution progresses and technologies evolve, “big data,” quantum computing, the proliferation of virtual networks, and micro-processing create enormous

105 Cebrowski and John J. Garstka. "Network-centric warfare: Its origin and future”; Jeff Cares, Distributed Networked Operations: The Foundations of Network Centric Warfare (Newport: Alidade Press, 2005).

detection, targeting, and control opportunities for states able to leverage information resources.

The capabilities created by the information revolution mean that no longer do weapons create effects additively. Instead, what gives information revolution states the ability to target over long-ranges, project power over the horizon, and conduct precision attacks is rapid computer-enabled fusion of a multitude of sensors. Some of these sensors are traditional radars that collect via analog techniques. However, with the information revolution has also come the proliferation of digital radars, digital signal collectors, and computer network exploitation.106 Sensor fusion—made possible by advances in computing—allows states to negate the inherent limitations of one sensor in order to generate targeting solutions from a variety of sensors.

A useful way to illustrate the capabilities created by the information revolution on war is to think about how ships used to be targeted versus how they are targeted in a state that embraces the information revolution. In a paper on U.S. vs. China naval capabilities, authors Biddle and Oelrich argue that the physical limitations of radar waveforms limit the range of conflict between the two states and drive the competitive balance between the adversaries.107 Therefore, the states that can win in this conflict are those that can bypass radar detection (stealth). The expense and difficulty of creating stealth technology keeps the balance of competition between these two states relatively static.

106 See for example the extensive analysis on sensor fusion and over the horizon targeting conducted by the Department of Defense Systems Engineering Cohort 21/Team Alpha, “Organic Over-the-Horizon Targeting for the 2025 Surface Fleet,” Naval Postgraduate School, June 2015: http://www.dtic.mil/dtic/tr/fulltext/u2/a632418.pdf and an even earlier analysis by Norman Friedman, Seapower and Space: From the Dawn of the Missile Age to Net-Centric Warfare (Annapolis: Naval Institute Press, 2000). 107 Stephen Biddle and Ivan Oelrich. "Future Warfare in the Western Pacific: Chinese Antiaccess/Area Denial, US AirSea Battle, and Command of the Commons in East Asia." International Security 41, no. 1 (2016): 7-48.

However, in a state that has embraced the innovations of the information revolution, a long-range early warning radar is just one input to a big data targeting solution. A 2015

Aviation Week and Space Technology report on emerging sensors found that the traditionally low resolution targeting of long-range radars “can be mitigated by networking multiple radars, and by using OTH-R to cue high-resolution sensors.”108

Similarly, a recent study by the Center for Strategic and Budgetary Assessments found that the use of big data and computing advancements may allow for revolutionary advancements in undersea detection.109 Meanwhile, examples of computer network exploitation (CNE) demonstrate how targeteers sitting well behind homeland borders of states may hack into adversary networks to “find” weapons and civilian vessels without the use of any active emitters.110

Our very concrete example of targeting a naval surface vessel demonstrates the huge capabilities created by the information revolution. A modern vessel needs access to satellite relays for accurate navigation, weather updates, and communication. These exchanges of information via satellite relays are both interceptable by passive antenna arrays and hackable111 by computer network operators. This vulnerability allows a state

108 Bill Sweetman, “New Radars, IRST Strengthen Stealth-Detection Claims,” Aviation Week and Space Technology, March 16, 2015: http://aviationweek.com/technology/new-radars-irst-strengthen-stealth- detection-claims. 109 Bryan Clark, “The Emerging Era in Undersea Warfare,” Center for Strategic and Budgetary Assessments, January 22, 2015: http://csbaonline.org/research/publications/undersea-warfare/. 110 Aditya K Sood and Richard Enbody, “U.S. Military Defense Systems: The Anatomy of Cyber Espionage by Chinese Hackers,” Georgetown Journal of International Affairs, December 19, 2014: http://journal.georgetown.edu/u-s-military-defense-systems-the-anatomy-of-cyber-espionage-by-chinese- hackers/; Kelly Jackson Higgins, “Chinese Hackers Target Logistics and Shipping Firms with Poisoned Inventory Scanners,” Information Week Dark Reading, July 10, 2014: http://www.darkreading.com/attacks- breaches/chinese-hackers-target-logistics-and-shipping-firms-with-poisoned-inventory-scanners/d/d- id/1297182; Pierluigi Paganini, “Hacking Ships: Maritime Shipping Industry at Risk,” Securityaffairs.co, March 31, 2015: http://securityaffairs.co/wordpress/35504/hacking/hacking-maritime-shipping- industry.html. 111 Jason Fritz, “Satellite Hacking: A Guide for the Perplexed,” Culture Mandala: Bulletin of the Centre for East-West Cultural and Economic Studies, Vol. 10, No. 1, (2012): 21-50.

to generate targeting information from navigation data and potentially to affect the operation of the vessel.112 In addition to vulnerabilities from satellite communications, most modern maritime shipping is controlled remotely in shore-based centers.

Increasingly, maritime pirates/hacktivists have demonstrated the ability to hack into these centers and associated vessels in order to view real time navigation data on-line, to impact the sea-worthiness of ships, and even to re-route cargo.113 Finally, whereas in the past targeting and detection via imagery was significantly limited by the inability to image and analyze large areas, the proliferation of both commercial and military satellites

(as well as the potential to combine imaging resources from manned and unmanned aerial assets) may provide states with the opportunity to image wide swaths of the ocean with much greater fidelity. Meanwhile, increased bandwidth and cloud storage technology allow previously prohibitive amounts of imagery to be stored while new image-searching algorithms and advances in computer processing allow these databases to be analyzed near-real time without the limitation of human manpower.114 These are all emerging technologies that digital innovation has introduced to targeting, but big data sensor fusion also includes the use of air traffic control, coastal defense sites, and coast guard radars as part of the targeting solution. These vignettes help show not only that the information revolution exists, but that the success of modern warfare in the information revolution

112 For instance, a professor at the University of Texas spent just $3,000 to build an apparatus from which he was able to take control of a digital navigation system on a yacht in the Mediterranean. Pierluigi Paganini, “GPS Vulnerability Exploitable to Control the Route of a Vessel,” Securityaffairs.co, July 27, 2013: http://securityaffairs.co/wordpress/16698/hacking/gps-vulnerability-exploitable-to-control-the-route- of-a-vessel.html. 113 Paganini, “Hacking Ships: Maritime Shipping Industry at Risk.” 114 Marcus Weisgerber, “The Increasingly Automated Hunt for Mobile Missile Launchers,” Defenseone, April 28, 2016: http://www.defenseone.com/technology/2016/04/increasingly-automated-hunt-mobile- missile-launchers/127864/.

may be based more on a state’s ability to collect, process, disseminate, and protect digital data than the previous physical limits of war.

Information Revolution: Vulnerabilities

These vignettes compelling tell a story of immense capabilities, but these capabilities come with extraordinary new vulnerabilities.115 The proliferation of digital technology has created a new terrain of war fighting. Cyberspace, or “a global domain within the information environment whose distinctive and unique character is framed by the use of electronics and the electromagnetic spectrum to create, store, modify, exchange, and exploit information via interdependent and interconnected networks using information-communication technologies”116 is a contested field of military power.

Indeed, what makes offensive cyber weaponry (or any other cross-domain warfare that targets digital capabilities) a potential game changer for modern conflict is the connection that states have built between digital capabilities and conventional warfare.117

These connections create lucrative cyber targets that impact the ability to achieve mission objectives across every domain and every spectrum of warfare. In 1935, Hitler worried that the German way of war would collapse without oil. In 1976, U.S. Army doctrine writers wrote about their concern that “as modern armies strive for more efficient

115 Brantly, The Decision to Attack: Military and Intelligence Cyber Decision-Making; O’Hanlon, Technological Change and the Future of War, Schneider, “Digitally-Enabled Warfare: The Capability- Vulnerability Paradox”; Robert J. Bunker, Five-Dimensional (Cyber) Warfighting: Can the Army after next be Defeated through Complex Concepts and Technologies? (Carlisle: U.S. Army War College Strategic Studies Institute, 1998); Zalmay M. Khalilzad, A.H. Marshall, and John P. White, Eds., The Changing Role of Information in warfare (Santa Monica: RAND Corporation, 1999). 116 This is Daniel Kuehl’s definition as found in Franklin D. Kramer, Stuart H. Starr, and Larry K. Wentz, Eds., Cyberpower and National Security (Washington, D.C.: national Defense University Press, 2009). 117 Edward F. Halpin, Cyberwar, Netwar, and the Revolution in Military Affairs (New York: Palgrave Macmillan, 2006); Brantly, The Decision to Attack: Military and Intelligence Cyber Decision-Making; Chris C. Demchak, Wars of Disruption and Resilience (Athens: University of Georgia Press, 2011).

command and control, better battlefield surveillance, and increased target acquisition capabilities, they have both opened new opportunities for battle success and created new vulnerabilities.”118 Thirty years later, the 2012 Joint Operational Access Concept iterates this concern, noting that control of the digital domain is not only vital for the success of

U.S. access operations (for example against Anti-Access/Area Denial operations like those against China or Iran) but is also increasingly contested.119

This inability to function without cyber dominance (vice cyber superiority) is because the vulnerabilities to the information revolution are exponential. The networked nature not only of digital information, but network-centric warfare, mean that vulnerabilities in cyber key terrain create cascading effects across weapon platforms and campaign capabilities.120 It is exactly because of the fast-paced, information driven, centralized control operations that digital capabilities have created that the information vulnerabilities are so existential to a digitally-dependent military. As the 2012 U.S.

Defense Strategic Guidance asserts, “modern armed forces cannot conduct high-tempo, effective operations without reliable information and communication networks and assured access to cyberspace and space. The information revolution makes states that embrace it extremely capable and yet makes network-dependent operations vulnerable to cross-domain cyber threats.”121

Even at the onset of this information revolution, there have already been examples of how these vulnerabilities might impact military operations. Syria’s digitally-upgraded

118 Department of the Army, Field Manual 100-5: Operations, 1976 (pg. 2-26). 119 U.S. Department of Defense, Joint Operational Access Concept (17 January 2012): http://www.defense.gov/pubs/pdfs/joac_jan%202012_signed.pdf. 120 Schneider, “Digitally-Enabled Warfare: The Capability-Vulnerability Paradox.” 121 U.S. Department of Defense, Sustaining US Global Leadership: Priorities for 21st Century Defense, 2012, http://www.defense.gov/news/defense_strategic_guidance.pdf, pg. 6.

Integrated Air Defense System was brought to its knees before an Israeli attack.122 The reason this networked system with modern digital processing was unable to detect the attack? Because the same digital updates that made the IADs a competent modern system also meant that it was vulnerable attack on its digital networks. According to reports, Israel was able to use a cyber-attack to shut down Syria’s Surface to Air Missile

Inventory and therefore fly over the weapons without threat.123 Similarly, a vulnerability with the digital maintenance system on the newest U.S. fighter, the F-35, would have grounded the system from combat operations.124 Despite the fact that this digital maintenance system has nothing to do with the avionics or weapons of this aircraft, its tie in with onboard systems has become so critical that it is “the core to making sure that the

F-35 functions . . . It’s so dependent on this technology, and so having all of these workarounds really limits operational capability.”125 A 2016 report from the Government

Accountability Office lambasted the reliance on this digital capability, citing concerns that even non-adversary interference such as weather or internal connectivity issues could essentially make the cutting-edge F-35 a useless weapon in war.126 This is especially dangerous to U.S. capabilities because the F-35 has been touted as the combat

122 Chandler Atwood and Jeffrey White. “Syrian Air-Defense Capabilities and the Threat to Potential U.S. Air Operations,” The Washington Institute, May 23, 2014: http://www.washingtoninstitute.org/policy- analysis/view/syrian-air-defense-capabilities-and-the-threat-to-potential-u.s.-air-operat. 123 Ward Caroll, “Israel’s Cyber Shot at Syria,” Defensetech.org, November 26, 2007: http://www.defensetech.org/2007/11/26/israels-cyber-shot-at-syria/. 124 Angus Batey, “F-35 Logistics System May be Vulnerable to Cyberattack,” Aviation Week, March 3, 2016: http://aviationweek.com/defense/f-35-logistics-system-may-be-vulnerable-cyberattack. 125 Insinna, Valerie. “Troubled Logistics System Critical to F-35s Future,” National Defense Magazine. April 2015: http://www.nationaldefensemagazine.org/archive/2015/April/pages/TroubledLogisticsSystemCriticaltoF35 sFuture.aspx. 126 Bill Carey. “GAO Questions Deployability, Redundancy, of F-35 ALIS System,” AIN Online, April 21, 2016: http://www.ainonline.com/aviation-news/defense/2016-04-21/gao-questions-deployability- redundancy-f-35-alis-system.

quarterback of American campaigns, a flying data-cloud that can coordinate intelligence and direct attacks essential to the United States’ network-centric operations.127

Events within the civilian world help demonstrate the prominence of information vulnerabilities. The extraordinary vulnerability of the information age was particularly evident on July 8, 2015, when a router malfunction grounded all United Airlines flights, a software configuration issue ceased trading on the New York Stock Exchange for half a day, and a server overload took down the Wall Street Journal website.128 These were not even due to malicious attacks, but instead to the inherent difficulty of maintaining the various critical nodes necessary for digitally-dependent operations. As Richard Danzig points out, “digital technologies ... are a security paradox: even as they grant unprecedented powers, they also make users less secure ... their concentration of data and manipulative power vastly improves the efficiency and scale of operations, but this concentration in turn exponentially increases the amount that can be stolen or subverted by a successful attack. The complexity of their hardware and software creates great capability, but this complexity spawns vulnerabilities and lowers the visibility of intrusions ... in sum, cyber systems nourish us, but at the same time they weaken and poison us.”129

Information Revolution: Can Adversaries Limit the Use or Access of Information?

127 Jeff Harrigian and Max Marosko. “Fifth Generation Air Combat: Maintaining the Joint Force Advantage,” The Mitchell Forum, no.6 (2016): http://media.wix.com/ugd/a2dd91_bd906e69631146079c4d082d0eda1d68.pdf. 128 Schneider, “Digitally-Enabled Warfare: The Capability-Vulnerability Paradox.” 129 Richard Danzig, “Surviving on a Diet of Poisoned Fruit: Reducing the National Security Risks of America’s Cyber Dependencies.” Center for a New American Security, July 21, 2014: https://www.cnas.org/publications/reports/surviving-on-a-diet-of-poisoned-fruit-reducing-the-national- security-risks-of-americas-cyber-dependencies.

The information revolution certainly creates a capability/vulnerability paradox.

However, the previous historical cases of military revolution demonstrate the importance that the adversary and the characteristics of resources plays in the how dangerous that paradox becomes. Can adversaries limit the use or access of information in order to negate or degrade the capabilities of the information revolution?

Many who are cynical of the importance of cyber to conflict doubt that these vulnerabilities can actually be exploited in a meaningful way.130 However, the information revolution has created a variety of vulnerabilities—not all within cyberspace—that have varying degrees of exploitability. If you conceptualize the actual components of the information revolution you realize that the digital information that characterizes this new military resides in both virtual and physical components. It is created (quite often) by human beings on physical components like computers that require hardware: microchips, memory cards, video cards, and batteries. It is transmitted via sea cables, satellites, and ground based antennas. It is processed through routers and saved in servers. To access and use the data, machines and human beings require software. And that software in turn requires personnel with the skills to create it.

The point is not that all cyber vulnerabilities can be attacked at once or that there is a magic weapon—cyber or conventional—that will “turn of the internet” in modern combat. Even oil, the most fungible resource explored in this analysis, generally has more than one point of vulnerability. The danger for the information revolution and the vulnerability of information is that the new way of warfare (and increasingly society as well) is so tied into these digital capabilities that it creates millions and billions of

130 Rid, Cyber war will not take place.

vulnerabilities.131 As Brantly argues, “the vulnerability largely originates out of the sheer complexity of the systems on which the national security infrastructure depends . . . there is no silver bullet vulnerability within cyber infrastructure. Rather, vulnerabilities expose a dependent party to the potential for a death by a thousand cuts. These cuts can range in size and scale, and if unchecked, many can fester and become lethal.”132 The cyber key terrain for defense is so large and dynamic that it is almost impossible to completely defend. And what is potentially more deleterious for these vulnerabilities is that the networks have become so complex and pervasive that some of these vulnerabilities are central nodes that that have cascading effects that exclude information beyond the initial vulnerability.133 Therefore, it may not be possible to exploit all vulnerabilities, but it may still be possible to exclude nations from leveraging information to conduct war.

Brantly identifies five types of vulnerabilities in cyberspace: 1) software vulnerabilities, such as zero-days that can be exploited through access to the coding, 2) interactive vulnerabilities which occur at the network level and involve access to key network nodes, 3) resource vulnerabilities, which is the physical supply of components such as routers, cables, memory cards; 4) environmental vulnerabilities, or the effect of weather and geography, and 5) capacity vulnerabilities—or how people and the network are able to operate this equipment, software, and networks.

131 Brantly, The Decision to Attack: Military and Intelligence Cyber Decision-Making; Nir Kshetri, “Information and Communications Technologies, Strategic Asymmetry and national Security,” Journal of Information Management 11, no. 4 (2005): 563-580; John Markoff, “Ideas & Trends: Blown to Bits; Cyberwarfare Breaks the Rules of Military,” New York Times, October 17 1999. 132 Brantly, The Decision to Attack: Military and Intelligence Cyber Decision-Making, pg. 39. 133 Richard Little, "Controlling cascading failure: Understanding the vulnerabilities of interconnected infrastructures." Journal of Urban Technology 9, no. 1 (2002): 109-123; Tomas Hellström, "Critical infrastructure and systemic vulnerability: Towards a planning framework." Safety science 45, no. 3 (2007): 415-430; Chee-Wooi Ten, Chen-Ching Liu, and Govindarasu Manimaran. "Vulnerability assessment of cybersecurity for SCADA systems." IEEE Transactions on Power Systems 23, no. 4 (2008): 1836-1846.

Each of these vulnerabilities provides a point of potential exploitation. Software vulnerabilities are generally only exploitable through cyber attacks. In order to conduct those cyber attacks—for instance through zero-day vulnerabilities—states must know not only where the zero day exists, but also how to access the adversary’s network in order to exploit the vulnerability. This effect is more likely temporary and the act itself is complicated to perform. However, the explosion of reliance on software to conduct day to day military operations increases the potential that even these temporary exclusions of information can provide the window of opportunity necessary to seize the advantage in digital vs. digital warfare.

Interactive vulnerabilities also bear the potential for exploitation. These vulnerabilities involve excluding access to networks—not just software. They are, therefore, more expansive and potentially can be physical and permanent. Access to networks in order to conduct the attack can be virtual—for instance through spear-fishing or DDOS, but it also can be physical.134 States may also try and exploit the vulnerability by accessing the physical components of the networks for instance through electromagnetic jamming,135 tapping into sea-bed cables,136 jamming or destroying satellites,137 conducting SOF or covert operations that physically inserts components into the networks.

134 Peter W. Singer and Allan Friedman, Cybersecurity: What Everyone Needs to Know (Oxford: Oxford University Press, 2014). 135 Michael Senft, “Convergence of Cyberspace Operations and Electronic Warfare Effects,” The Cyber Defense Review, January 4, 2016: http://www.cyberdefensereview.org/2016/01/04/convergence-of- cyberspace-operations-and-electronic-warfare-effects/. 136 Olga Khazan, “The creepy, long-standing practice of undersea cable tapping,” The Atlantic, July 16, 2013: http://www.theatlantic.com/international/archive/2013/07/the-creepy-long-standing-practice-of- undersea-cable-tapping/277855/; Greg Miller, “Undersea Internet Cables are Surprisingly Vulnerable,” Wired, October 29, 2015: https://www.wired.com/2015/10/undersea-cable-maps/. 137 John Klein, Space warfare: Strategy, principles and policy (New York: Routledge, 2012).

The resources that make up the networks are also vulnerable. Access to the resources can be limited, the resources themselves can be degraded or manipulated, and they can be destroyed. Supply-chain vulnerability for the components that enable the information revolution can create excludability in two ways. First, it may be that some of the components required to transmit information are made only in a few locations and so the excludability of these types of information resources can actually be very similar to oil. For example, the lithium required for high-tech batteries is increasingly becoming an excludable resource. Secondly, states can insert malware in components and therefore the supply chain provides the potential for access and excludability of information at a later date.138 Both of these resource vulnerabilities have been a concern of China and

Russia, who have voiced alarm at the preponderance of information technology resources within the United States.

Additionally, resources may also be vulnerable to physical attack. Ballistic missile attacks on satellite,139 cut seabed cables (see for example Russian cable cutting exercises),140 SOF operations that physically destroy routing centers, even physical bombs that destroy key communications or electronic facilities—the potential for

138 Craig Martin and Kip Shepard, “Supply Chain Risk Management,” Fire Eye: Best Practices in Cyber Supply Chain Risk Management, https://www.fireeye.com/content/dam/fireeye-www/global/en/current- threats/pdfs/rpt-best-practices-in-cyber-supply-chain-risk-management.pdf; G.E Smith, K. J. Watson, W. H. Baker, and J. A. Pokorski Ii. "A critical balance: collaboration and security in the IT-enabled supply chain." International journal of production research 45, no. 11 (2007): 2595-2613; Joseph S. Nye, Jr., Nuclear lessons for cyber security (Maxwell: Air University Press, 2011). 139 Sydney J Freedberg, “Pentagon Reports on China’s Satellite Killers,” Breakingdefense.com, May 11, 2015: http://breakingdefense.com/2015/05/pentagon-reports-on-chinas-satellite-killers/. 140 Bruce Dorminey. “How Bad Would it Be if the Russians Started Cutting Underwater Cables? Try Trillions in Damage,” Forbes.com, November 2, 2015: http://www.forbes.com/sites/brucedorminey/2015/11/02/russian-navy-probing-u-s-undersea- communications-cables-in-new-global-threat/#520cbaf866b1; David Sanger and Eric Schmitt, “Russian Ships Near Data Cables are too Close for U.S. Comfort,” New York Times, October 25, 2015: http://www.nytimes.com/2015/10/26/world/europe/russian-presence-near-undersea-cables-concerns- us.html?_r=0.

physical destruction of information networks is incredibly vulnerable. The United States demonstrated compellingly in both Gulf Wars that physical attacks on command and control could decimate a nation’s military in the first few hours or days of conflict.

Imagine an example from current network centric warfare. The United States increasingly turns to aggregated data centers to process vast quantities of intelligence.

These facilities, which serve as the primary processing, exploitation, and distribution nodes for all war fighting intelligence, are centralized and vulnerable to physical attacks.

How would countries that remotely process intelligence be able to function if they could only collect but no longer had the means to disseminate any of that intelligence to the warfighter?141

Brantly’s last two vulnerabilities: environment and capacity are generally not excludable. However, they compound the potential for losing the capabilities created by the information revolution. They highlight the fragility of the information revolution as states become more and more dependent on information systems to conduct war.

Cyber and the Capability-Vulnerability Paradox: Implications for Conflict

Initiation

Information is an exploitable resource and the information revolution has created not only the dependency on this resource, but also increasingly the capabilities required to undermine access or use of the resource itself. Cyber attacks, taken in conjunction with physical acts to exclude the use of information, make the capability/vulnerability

141 Schneider, “Digitally-Enabled Warfare: The Capability-Vulnerability Paradox.”

paradox for the information revolution potentially dangerous. What can lessons from past military revolutions suggest?

First, the capability/vulnerability paradox is most dangerous when resources can be affected by adversaries and everyone knows it. The fear of the vulnerability is almost more important than the ability to exploit that vulnerability. It is the fear of losing the capability that drives the incentive towards first strike for a state that utilizes the military revolution. And it is the dependency on that capability that incentivizes asymmetric states to attack when they would otherwise be deterred by an adversary’s greater conventional capability. In the case of the information revolution, states are highly aware of their own vulnerabilities. They may be less certain about the actual effects that attacks on those vulnerabilities might have and they may be much less certain about the ability to create similar effects on adversary information vulnerabilities. As long as states are uncertain about their ability to conduct effective first strikes against adversary information resources, there will be less incentive to conduct attacks. However, as states become more dependent on digital capabilities to conduct modern warfare, adversaries will become more certain that they will be able to conduct attacks with significant effects.

The obvious solution to this potential danger is for states to decrease their digital dependence. However, this requires either: 1) innovating cyber defenses and resiliencies that don’t currently exist, or 2) in the short term trading off military effectiveness for digital independence.

Because this trade-off is difficult to make and because states will probably continue to double-down on digital capabilities that make them conventionally superior, there is a good chance that the capability/vulnerability paradox will make contentious

dyads more unstable. This is particularly true when asymmetries in conventional capabilities mean that a less capable state must strike first at a digitally-dependent and superior state in order to survive in any conflict. Additionally, when a digitally- dependent state is aware of these vulnerabilities and fears a first attack, they may use their own superior digitally-dependent capabilities in what they view as a preemptive attack. Together, these dynamics made cyber in contentious dyads a dangerous destabilizer.142

It is also important to realize that the information revolution didn’t just co-opt other revolutions—the information revolution made other military revolutions more capable and vulnerable. For example, information is the means, increasingly, by which public opinion is excluded. This is still a vital vulnerability for states with large standing armies a la levee en masse143 and recent information operations conducted by the

Russians during the U.S. election demonstrate the immense capabilities that the information revolution has given to states to try and exploit public opinion for the use of force. Similarly, the information revolution made the operation of logistical networks like railroads and oil much more effective. Leading up to World War I, one of the largest vulnerabilities of railroads was the difficulty in managing them. With the resources of the information revolution, railroads and oil are no longer vulnerable to rigidity and mismanagement due to human inefficiencies. However, the digitization of these capabilities introduced new vulnerabilities to cyber attack. Cyber intruders can

142 David Gompert and Martin Libicki. "Cyber warfare and Sino-American crisis instability." Survival 56, no. 4 (2014): 7-22. 143 Andrew A. Weisburd, Clint Watts, and J.M. Berger, ”Trolling for Trump: How Russia is Trying to Destroy our Democracy,” War on the Rocks, November 6, 2016: http://warontherocks.com/2016/11/trolling-for-trump-how-russia-is-trying-to-destroy-our-democracy/.

manipulate rail switching and automated systems that manage the rails can be both hacked and controlled remotely.144 In 2016, a San Francisco commuter rail’s ticketing system was hacked into and held for ransom145 and in 2012 rail travel in the Pacific

Northwest was delayed as hackers interfered with railway signals.146 That same year an attack on the world’s largest oil company, Saudi Aramco, led to the destruction of thousands of corporate computers and left the company scrambling to maintain oil production.147 Since then cyber attacks on oil companies and infrastructure have continued to grow in sophistication and number. A report that came out in late 2015 indicated that over a year’s time period attacks had increased 50 to 100%.148

Finally, analysis of the nuclear revolution reveals that the most excludable element of nuclear capabilities is C3.149 Unfortunately, the information revolution has not only not solved this problem but likely exacerbated both the potential and the fear of preemptive strikes on nuclear C3. And as states continue to upgrade and digitize their nuclear command and control, the danger of preemptive attacks will grow. This is especially true for cyber attacks on nuclear C3 because they can be conducted by asymmetric states or even non-state actors. Additionally, problems with attribution for cyber operations complicate retaliatory policies for attacks on C3. States can use that

144 Pierluigi Paganini, “Modern railroad systems are vulnerable to cyber attacks,” Securityaffairs.co, January 2, 2016: http://securityaffairs.co/wordpress/43196/hacking/railroad-systems-vulnerabilities.html. 145 Joe Fitzgerald Rodriguez, “’You Hacked’ appears at Muni stations as fare payment system crashes,” San Francisco Examiner, November 26, 2016: http://www.sfexaminer.com/hacked-appears-muni-stations-fare- payment-system-crashes/. 146 Kim Zetter, “Hackers Breached Railway Network, Disrupted Service,” Wired, January 24, 2012: http://www.sfexaminer.com/hacked-appears-muni-stations-fare-payment-system-crashes/. 147 Nicole Perlroth, “In Cyberattack on Saudi Firm, U.S. sees Iran Firing Back,” The New York Times, October 23, 2012: http://www.nytimes.com/2012/10/24/business/global/cyberattack-on-saudi-oil-firm- disquiets-us.html. 148 “Oil and Gas Industry Increasingly Hit by Cyber-Attacks,” Security Week News, January 14, 2016: http://www.nytimes.com/2012/10/24/business/global/cyberattack-on-saudi-oil-firm-disquiets-us.html. 149 Gartzke, Erik and Lindsay, Jon R., Thermonuclear Cyberwar (June 15, 2016). Available at SSRN: https://ssrn.com/abstract=2836208

uncertainty to test retaliation policies and non-state actors (or other states) can use that uncertainty to hope for mis-attribution.150

Finally, as with all technologies, there are technical advancements that may mitigate or negate the capability/vulnerability paradox. In particular, the current consensus is that the offense-defense balance in cyber is with the offense. If there are technological advancements that make defense a more viable investment, then the dangers of the paradox will diminish. Regardless of the future technological developments, this analysis makes a significant contribution to our understanding of the implications of cyber for conflict initiation. Additionally, by identifying the importance of the relationship between capabilities and vulnerabilities, this research helps us understand other variables that can explain when states go to war.

150 Andrew Futter, “The Dangers of Using Cyberattacks to Counter Nuclear Threats,” Arms Control Today, July/August 2016: https://www.armscontrol.org/ACT/2016_07/Features/The-Dangers-of-Using- Cyberattacks-to-Counter-Nuclear-Threats; Gartzke and Lindsay, Thermonuclear Cyber War; Andrew Futter, “The double-edged sword: US nuclear command and control modernization,” The Bulletin of Atomic Scientists, June 29, 2016: http://thebulletin.org/double-edged-sword-us-nuclear-command-and- control-modernization9593.

Chapter 2.

Offense-Dominance in the Information Age: Exploring the

Determinants of U.S. Doctrine Over the Last 40 Years

The information age has ushered in a revolution in technologies and operations for military campaigns. With the advent of micro-processing, states were able to develop precision-guided munitions, enable automation, and exponentially increase the volume and speed of information on the battlefield. What began in the late 1970s and 80s as an enabling technology for existing operations led to the advent of the network-centric warfare of the 1990s and finally the cyberspace domain of conflict today. With these

advances in technology have also come changes in U.S. operational doctrine. That doctrine, which vacillated between offense and defense in modern warfare, swung decisively towards offense with the advent of Air Land Battle in 1982 and codified offense-dominance in today’s Air-Sea Battle and Joint Concept for Access Maneuver in the Global Commons (JAM-GC).

What drives these perceptions of offense-dominance in U.S. information age doctrine? The U.S. military narrative explicitly links technology to offensive-dominance.

In the ‘80s it was the role of precision-guided munitions that shaped an offensive counter-

Russian doctrine, at the beginning of the new millenium it was the computer and digital information that drove offensively-dominated network-centric warfare, and finally in the current day and age it is big data that enables long-range global strike. Further, the

United States’ acquisition strategies like the Third Offset, seem to cement the pre- dominant role that technology takes in dictating offensive U.S. doctrine. For the United

States, the narrative of the information age seems to be that technology creates the balance between offensive and defensive campaigns.

Despite the explicit links between technology and offensive doctrine in U.S narrative, academic literature on the sources of offensive doctrine suggest that militaries often construct perceptions of offense dominance—despite of or without regard to the true technological balance. For instance, Kier Lieber argued at the height of emerging information age campaigns that the core assumption of ODB, that technology drove perceptions of ODB to influence international stability, was flawed. The crux of Lieber’s argument was that politics would always trump technology, asserting that “military doctrines and strategies are often more responsive to military organizational biases and

statesmen’s perceptions of international politics than to the implications of prevailing weapons technologies . . . leaders want to know how given technologies can aid their strategy, not how their strategy should change in light of these technologies.”151

Almost twenty years later, the information age continues to dominate discussions about international stability. Whether it is the internet of things, the interconnected nature of a digital economy, the wired and networked weapons of modern warfare, or the ability to conduct offensive cyber operations—computing power and network capability have revolutionized the way states understand national security. Technology, more than ever, is king. And these technologies of the information revolution appear to have created a strong perception within U.S. military decision-makers of offensive advantage.152 This certainly is the narrative from practitioners who warn about imminent instability stemming from the inherently offensive-dominance of computing and cyberspace operations. However, as strong as this narrative has been from practitioners, scholars question how much of this perceived offense dominance is based on the attributes of the technology and how much instead is a manifestation of cultural,

151 Keir Lieber, War and the Engineers: The Primacy of Politics Over Technology (Ithaca: Cornell University Press, 2005): pg. 153. 152 Leon Panetta, “Remarks by Secretary Panetta on Cybersecurity to the Business Executives for National Security, New York City,” October 11, 2012: http://www.defense.gov/transcripts/transcript.aspx?transcriptid=5136; Vago Muradian, “Adm. Michael Mullen,” DefenseNews, June 10, 2011: http://www.defensenews.com/article/20110710/DEFFEAT03/107100301/Adm-Michael-Mullen; Bob Violino, “Unseen, all-out cyberwar on the U.S. has begun,” Infoworld.com, January 28, 2013, http://www.infoworld.com/article/2612825/hacking/unseen--all-out-cyber-war-on-the-u-s--has-begun.html; Michael Joseph Gross, “Silent War,” Vanity Fair, July 2013, http://www.vanityfair.com/culture/2013/07/new-cyberwar-victims-american-business#; Anna Mulrine, “CIA chief Leon Panetta: the next Pearl Harbor could be a cyberattack,” The Christian Science Monitor, June 9, 2011, http://www.csmonitor.com/USA/Military/2011/0609/CIA-chief-Leon-Panetta-The-next- Pearl-Harbor-could-be-a-cyberattack; Richard A. Clarke, Cyber War: The Next Threat to National Security and What to Do About It, (New York: Harper Collins Publishers, 2010).

organizational, and individual biases towards offensive doctrines.153

In this article, I examine this question to understand what drives U.S. offense- dominant doctrine in the information age. Is it the offensive tilt of information age technologies? Is it organizational incentives? Cultural biases toward offense? Or a manifestation of politics and geopolitical context? In the past, these hypotheses have generally been an either-or . . . it is either technology or politics. My analysis shows that the truth is more intertwined and that critical junctures in politics can codify organizational offensive biases to create a path dependency for technologies that then necessitate offensive operations.

I focus this analysis on offense and defense within U.S. doctrine. By examining the change of U.S. doctrine over time, I am able to focus on in-case variation while holding constant a series of variables related to domestic capacity, type of governance, and institutional design. In particular, I examine two cases of U.S. doctrine—FM 100-5 and the shift between Active Defense to AirLand Battle and the 2012 Air Sea Battle/Joint

Operational Access Concept. Based on analysis of these two cases, I find that Lieber is partially right, but so are the technology advocates of more traditional ODB analysis.

Politics combined with organizational biases towards the offense trump technology in the

153 Jerry Brito and Tate Watkins, “Loving the Cyber Bomb? The Dangers of Threat Inflation in Cybersecurity Policy,” Harvard National Security Journal, April 10, 2012, as found at http://mercatus.org/publication/loving-cyber-bomb-dangers-threat-inflation-cybersecurity-policy-0; Sean Lawson, “Beyond Cyber Doom: Cyberattack Scenarios and the Evidence of History,” Mercatus Center 11, no.1 (2011): 1-37; Stephen Walt, “Is the cyber threat overblown?” Foreignpolicy.com March 30, 2010, http://foreignpolicy.com/2010/03/30/is-the-cyber-threat-overblown/; Ryan Singel, “Richard Clarke’s Cyberwar: file under fiction,” Wired, April 22, 2010, http://www.wired.com/2010/04/cyberwar-richard- clarke/; Thomas Rid, “Cyber war will not take place,” Journal of Strategic Studies 45, no. 1 (2012); Eric Gartzke, “The myth of cyberwar: Bringing war in cyberspace back down to earth,” International Security, 38, no. 2 (2013): 41-73; Jon R Lindsay. "Stuxnet and the limits of cyber warfare," Security Studies 22, no. 3 (2013): 365-404; Jon R. Lindsay. "The impact of China on cybersecurity: fiction and friction," International Security 39, no. 3 (2015): 7-47.

design of AirLand Battle, but that the doctrine was a critical juncture for the United

States and created a path dependency in which the acquisition of offensive-driven information revolution technology created a technological offense dominance. This technological offense dominance drove the offensive considerations of the Air Sea Battle concept. However, technological changes in cyber—not dependent on U.S. defense acquisitions—are impacting some elements of offense-defense within U.S. doctrine such that a shift may occur in the future towards defensive doctrines driven more by society and politics than by technology.

Below, I examine this question in four parts. Part one discusses the literature on offense and defense in military doctrine. Part two reviews the information revolution and the offense-defense balance. Next part three introduces the methodology and the cases.

Part four traces the role of the information revolution through the two doctrinal cases and finally I conclude with implications for our understanding of the origin of offensive doctrine as well as policy implications for the United States.

Offense and Defense in Military Doctrine: Why it Matters and Why it Happens

Why does it matter whether states develop offensive or defensive doctrine?

Offense-defense theorists trace the relevance of offense and defense in international stability to John Herz’ security dilemma.154 The general premise behind the security

154 John H. Herz, “Idealist Internationalism and the Security Dilemma,” World Politics 2, no . 2 (January 1950): 157-180; Herbert Butterfield, History and Human Relations (London: Collins, 1950); Robert Jervis, “Cooperation under the Security Dilemma,” World Politics 30, no. 2 (1978): 167-214; and Charles Glaser, “The Security Dilemma Revisited,” World Politics 50, no. 1 (1997): 171-201. Yoav Gortzak, Yoram Haftel, and Kevin Sweeny, “Offense-Defense Theory: An Empirical Assessment,” Journal of Conflict Resolution 49 (2005): 67-89; Stephen Van Evera, Causes of War (Ithaca: Cornell University Press, 1998); George H. Quester, Offense and defense in the international system (New York: Wiley, 1977); Kier A. Lieber, “Grasping the technological peace: The offense-defense balance and international security,”

dilemma is that as security-driven states seek to safeguard their own survival, their acquisition of power drives others to also feel more insecure. Therefore, as one state seeks to ensure its own security, it inadvertently creates a security dilemma for other states so that, “since none can ever feel entirely secure in such a world of competing units, power competition ensues, and the vicious circle of security and power accumulation is on.”155 The tragedy of the security dilemma is that the attempt to escape insecurity fuels increases in arms and generates conflict volatility.

Subsequent literature on the security dilemma identified two main variables that affected the uncertainty and danger of the security dilemma: offense-defense differentiation (or the ability to determine whether operations/weapons are offensive or defensive) and offense-defense balance (ODB) (or the advantage between offensive and defensive operations). The vast majority of attention within the literature has been on the latter variable and subsequent studies on the security dilemma link the danger of the dilemma with ODB. As Jervis explains, the concept of ODB is that “when we say that the offense has the advantage, we simply mean that it is easier to destroy the other's army and take its territory than it is to defend one's own. When the defense has the advantage, it is easier to protect and to hold than it is to move forward, destroy, and take.”156

The difficulty, of course, becomes measurement. This has been a consistent problem with the security dilemma as scholars have struggled to provide empirical analysis of ODB.157 The accepted understanding of offense-defense balance is that it

International Security 25 (2000): 71-104; and Karen Ruth Adams, “Attack and Conquer: International Anarchy and the Offense Defense Deterrence,” International Security 28, no. 3 (2004): 45-83. 155 Herz, “Idealist Internationalism and the Security Dilemma,” pg. 157. 156 Jervis, “Cooperation Under the Security Dilemma,” pg. 188. 157 Glaser and Kaufmann, “What is the Offense-Defense Balance and can we Measure it?”; Jack S. Levy, "The Offensive/Defensive Balance of Military Technology: A Theoretical and Historical Analysis," International Studies Quarterly, Vol. 38, No. 2 (June 1984), pp. 219-238; Sean Lynn-Jones, “Offense-

“should be defined as the ratio of the cost of the forces that the attacker requires to take territory to the cost of the defender’s forces.”158 This understanding of offense-defense assumes that there is some true technological balance between offensive and defensive campaigns. However, in reality scholars have struggled both to measure an objective

ODB and to trace states’ perceptions of ODB to conflict decisions.159

The problem has been that states, much less scholars, seem to have difficulty determining the cost to attack vs. to defend. Faced with the difficulty in determining the objective ODB, a series of scholars have hypothesized that technology and material capabilities are less important to the development of offensive campaigns than non- material factors such as politics, organizational biases, cultural predilections, or civil- military relations.

For example, Keir Lieber argues that policymakers make decisions about conflict independent of the technological ability to create offensive or defensive campaigns.

Instead, they rely on political context—both domestic and international—to design offensive or defensive campaigns and strategies.160 Similar arguments about the development of offensive and defensive doctrine also critique ODB’s focus on technology.161 Van Evera argues that a “cult of the offensive” in which militaries reified

Defense Theory and its Critics,” Security Studies Vol.4, No. 4 (1995): 660-691; Stephen Biddle, “Rebuilding the Foundations of Offense-Defense Theory.” The Journal of Politics Vol.63, No.3 (2001): 741-774; Stephen VanEvera, Causes of War: Power and the Roots of Conflict (Ithaca: Cornell University Press, 1999); George Quester, Offense Defense in the International System (New York: John Wiley Publishers, 1977). 158 Glaser and Kaufmann, “What is the Offense-Defense Balance and can we Measure it?” pg. 44. 159 Robert Jervis, "Was the Cold War a Security Dilemma?." Journal of Cold War Studies 3, no. 1 (2001): 36-60. 160 Lieber, War and the Engineers; Keir Lieber, “Grasping the technological peace: The offense-defense balance and international security." International Security 25, no. 1 (2000): 71-104. 161 Scott Sagan, "1914 Revisited: Allies, Offense, and Instability." International Security 11, no. 2 (1986): 151-175; Deborah D. Avant "The institutional sources of military doctrine: Hegemons in peripheral wars." International Studies Quarterly 37, no. 4 (1993): 409-430.

offensive doctrine despite an objective technological defensive advantage drove states to choose sub-optimal offensive doctrines prior to World War I.162 Snyder further elaborated on this thesis, arguing that civil-military relations may exacerbate and create this cult of offensive. As he explains, “on balance, offense tends to suit the needs of military organizations better than defense does” and that civil-military relations can exacerbate that bias when “the lack of civilian control allowed it to grow unchecked or because an abnormal degree of civil-military conflict heightened the need for self- protective ideology.”163 Meanwhile, Posen points to similar organizational and structural determinants of offensive doctrine and asserts that militaries prefer offensive doctrine because they would rather become “specialists in victory” than “specialists in attrition.”164 Finally, other arguments question the agency militaries, governments, or technologies have in determining offensive or defensive doctrine and instead argue that national culture impacts the development of offensive doctrine.165 Cumulatively, these authors suggest that while offense and defense may matter for stability, states’ ability to determine the objective balance and then build campaigns accordingly is questionable.

Therefore, non-technological factors drive the development of offensive or defensive technology instead of an objective material ODB creating offensive or defensive doctrine.166

162 Stephen Van Evera, "The cult of the offensive and the origins of the First World War." International security 9, no. 1 (1984): 58-107. 163 Jack Snyder, "Civil-Military Relations and the Cult of the Offensive, 1914 and 1984." International Security 9, no. 1 (1984): 121. See also, Jack Snyder, The ideology of the offensive: Military decision making and the disasters of 1914. Vol. 2. (Ithaca: Cornell University Press, 1989). 164 Barry Posen, The Sources of Military Doctrine,” (Ithaca: Cornell University Press, 1984), pg. 49. 165 Elizabeth Kier, Imagining war: French and British military doctrine between the wars (Princeton: Princeton University Press, 1997). 166 Jonathan Shimshoni, “Technology, Military Advantage, and World War I,” International Security 15, no.3 (1990): 187-215.

Central to these arguments is the role that doctrine plays as a manifestation of states’ offensive or defensive strategies. It guides the training, equipment, disposition, and utilization of a state’s military forces and captures the way in which military leaders view the future of warfare, including whether militaries think offensive or defensive operations are more likely to achieve victory. As Brig Gen Wass de Czege, drafter of

AirLand Battle 1982 explains, “a doctrine is actually a sum total of the thought models that commanders in the field, their staffs, and their subordinates share . . . they address, for instance, . . . how successful defenses or offenses are composed.”167 Additionally, doctrine serves as the outward-facing signal of how a state intends to use its military forces. As such, they signal to the international community not only whether or not a state will use its forces defensively or offensively but also the state’s perception of the prevailing offense-defense balance within the international system. As Posen compellingly argues, “by their offensive, defensive, or deterrent character, doctrines affect the probability and intensity of arms races and of wars.”168 What then does U.S. offensive doctrine in the information revolution reveal about how the U.S. developed its preference for offensive campaigns? Did the technology drive the development of offensive or defensive doctrines? Or are there other factors that trumped technology in the development of these doctrines?

The Information Revolution, Cyber, and Offense-Defense: The Debate

The offense-defense debate about the information revolution is generally

167 Huba Wass de Czege, “Lessons from the Past: Making the Army’s Doctrine ‘Right Enough’ Today,” Landpower Essay, Vol. 6, No. 2 (2006): 13; for the importance of doctrine also see Donn A. Starry, “To Change an Army,” Military Review, Vol. 63(1983): 24. 168 Barry Posen, The Sources of Military Doctrine (Ithaca: Cornell University Press, 1984), pg. 15.

separated into two lines of reasoning. The first evaluates the role of the information revolution—or the rise in computing power and networks—on conflict. These analyses generally occurred between the late 80s and the early 2000s. They discuss how computing and networks change conventional operations.169 The second line of reasoning, which is oddly separate of the first, looks specifically at cyber and the offense- defense balance.170 A very small but increasingly vocal literature combines the both, arguing that offensive or defensive advantages in one side of these technologies has a relationship with how the other affects overall offense-defense balances.171

The literature on how the information revolution affected conventional operations is heavy with assumptions of offense dominance. Digital capabilities decrease the fog of war, increase precision and long range strike. These digitized capabilities would allow states to seize the initiative and create information dominance while cutting off decision space for adversaries. These offense-dominant predictions were both drawn from and

169 Cebrowski and John J. Garstka. "Network-centric warfare: Its origin and future”; Jeff Cares, Distributed Networked Operations: The Foundations of Network Centric Warfare (Newport: Alidade Press, 2005); Robert L. Bateman, Ed., Digital War: A View from the Front Lines (Novato: Presidio Press, 1999); “Joint Vision 2020: America’s Military—Preparing for Tomorrow,” Joint Force Quarterly (Summer 2000): 57- 76; William A. Owens and Ed Offley. Lifting the fog of war. (Baltimore: Johns Hopkins University Press, 2001); Eliot A. Cohen, "A revolution in warfare." Foreign Affairs (1996): 37-54; Eliot A. Cohen, "Change and transformation in military affairs." Journal of Strategic Studies 27, no. 3 (2004): 395-407; Erik Dahl, “Network Centric Warfare and Operational Art,” Defence Studies 2/1(Spring 2002): 15-35. 170 Rebecca Slayton, "What Is the Cyber Offense-Defense Balance? Conceptions, Causes, and Assessment." International Security 41, no.3 (2017): 72-109. Rid, “Cyber war will not take place,” Journal of Strategic Studies 45, no. 1 (2012); Gartzke, “The myth of cyberwar: Bringing war in cyberspace back down to earth,”; Lindsay. "Stuxnet and the limits of cyber warfare"; Lindsay. "The impact of China on cybersecurity: fiction and friction," International Security 39, no. 3 (2015): 7-47; Martin Libicki, Crisis and Escalation in Cyberspace (Santa Monica: RAND Corporation, 2012); Herbert Lin. "Escalation Dynamics and Conflict Termination in Cyberspace." Air University (2012). 171 David Gompert and Martin Libicki, “Cyber Warfare and Sino-American Instability,” Survival 56, no.4 (2014): 7-22; Jacquelyn Schneider, “Digitally Enabled Warfare: the Capability-Vulnerability Paradox,” Center for a New American Security, August 29, 2016: https://www.cnas.org/publications/reports/digitally- enabled-warfare-the-capability-vulnerability-paradox; Edward F. Halpin, Cyberwar, Netwar, and the Revolution in Military Affairs (New York: Palgrave Macmillan, 2006); Chris C. Demchak, Wars of Disruption and Resilience (Athens: University of Georgia Press, 2011); Aaron Franklin Brantly, The Decision to Attack: Military and Intelligence Cyber Decision-Making (Athens: University of Georgia Press, 2016).

characterized by the heavily offensive (and successful) campaigns led by the United

States into Iraq in both 1991 and later in 2005. There is little debate in this literature about defense dominance and also notably missing is the role that this offense dominance might have in causing states to misperceive each other and inadvertently end up at war.

This was driven by the fact that the United States dominated this portion of the information revolution—there were no real competitors and therefore little worry about incentives for adversaries to conduct first strike.

The debate about cyber and offense-defense balance is much more robust.172 The current debate generally divides into three sets of arguments, with the vast preponderance of scholarly conclusions within the two latter arguments: 1) cyber as an offense-dominant domain and cyber capabilities as a crisis destabilizer, 2) cyber as a defense-dominant domain and a crisis stabilizer, and 3) cyber as an offense-defense neutral domain with no significant effect on crisis stability.

Advocates of offense-dominance within cyberspace argue that cyber fundamentally changes the nature of war.173 Because cyber is a relatively cheap weapon system to develop and because great power nation states are often dependent on cyber

172 Much of the following conception of cyber relies heavily on U.S. doctrine—China’s translation of cyberspace means virtual host. The word the Chinese typically use is informatization. Despite the fact that the word is not literally cyber, informatization captures a vital part of cyber as it is understood in this paper—the focus on information within a virtual domain and the struggle to achieve information superiority between two nations. See Keir Giles and William Hagestad II, “Divided by a Common Language: Cyber Definitions in Chinese, Russian, and English,” paper presented at 5th International Conference on Cyber Conflict, 2013; Vincent Joubert, “Getting the Essence of Cyberspace: A Theoretical Framework to Face Cyber Issues,” paper presented at Conference on Cyber Conflict, Estonia, Tallinn, 2010. 173 Amit Sharma, "Cyber Wars: A Paradigm Shift from Means to Ends," Strategic Analysis 34, no.1 (2010): 62-73; Clarke and Knake, Cyber War; Lynn, “Defending a New Domain-The Pentagon's Cyberstrategy;” Clarke, “War from Cyberspace;” Denning, “Barriers to Entry: Are They Lower for Cyber Warfare?”; Kello, “The Meaning of the Cyber Revolution: Perils to Theory and Statecraft”; Patrick Morgan, "Applicability of Traditional Deterrence Concepts and Theory to the Cyber Realm," Proceedings of a Workshop on Deterring Cyber Attacks: Informing Strategies and Developing Options for US Policy.

infrastructure for both their military and economy to thrive, cyber provides a new tool for less powerful states to disturb power balances.174 This, in turn creates crisis instability.

Additionally, because even great power states have difficulties defending their cyber terrain, there is an incentive for “pro-active” defense measures,175 which might be easily mistaken for offensive cyber strikes and also destabilize crises.

Defense-dominant theorists argue that cyber may actually play a stabilizing effect.176 For these scholars, the uncertainty inherent in both cyber effects and cyber rules of engagement can lead states to limit both cyber and conventional operations.

Also, because cyber networks are often interdependent with commercial infrastructure, states will be less likely to take first strikes at cyber targets, thus greatly limiting the potential opportunities for conflict escalation through non-kinetic means.

Finally, offense-defense neutral advocates argue that, regardless of the offense- defense balance within cyberspace, cyber effects generally neutralize themselves.177

Instead of cyber exacerbating or mitigating a crisis, it instead takes a back seat to more conventional concerns about nuclear weapons, alliances, and balance of power. These

174 Audrey Kurth Cronin, "Cyber-Mobilization: The New Levée en Masse," Parameters 36, no.2 (2006): 77-87; Stephen Korns, Cyber operations: the new balance (DC: National Defense University Institute for National Strategic Studies, 2009); Ross Rustici, "‘Cyberweapons: Leveling the International Playing Field," Parameters (2011): 36-47. 175 For instance, the U.S. Department of Defense classifies these actions as defensive counter operations responsive actions (DCO-RA) in their Joint Publication 3-12, Cyberspace Operations (2013: GL-4): “deliberate, authorized defensive measures or activities taken outside of the defended network to protect and defend Department of Defense cyberspace capabilities or other designated systems.” 176 Lindsay, “Cybersecurity and International Relations: Evaluating the Threat from China,”; Crosston, “World Gone Cyber MAD: How ‘Mutually Assured Destruction’ is the Best Hope for Cyber Deterrence”; Lieber, “The Offense-Defense Balance and Cyber Warfare.” 177 Betz, "Cyberpower in Strategic Affairs: Neither Unthinkable nor Blessed;” Libicki, “Don’t Buy the Cyber Hype;” Liff, "Cyberwar: A New 'Absolute Weapon'? The Proliferation of Cyberwarfare Capabilities and Interstate War;” Lindsay, "Stuxnet and the Limits of Cyber Warfare;” Valeriano and Maness, “The Fog of Cyberwar: Why the Threat doesn’t Live up to the Hype;” Junio, “How Probable is Cyber War? Bringing IR Theory Back in to the Conflict Debate;” Barzashk, “Are Cyber-Weapons Effective?”; Rid, “Cyber war will not take place;” Lindsay, “Stuxnet and the Limits of Cyber Warfare;” Liff, “Cyberwar: A New ‘Absolute Weapon’? The Proliferation of Cyberwarfare Capabilities and Interstate War;” and Gartzke, “The myth of cyberwar: Bringing war in cyberspace back down to earth.”

scholars stress the limitation of cyber weaponry and argue that cyber capabilities are unable to cause significant enough effects to induce states to cross the threshold for war.

For the scholars that combine both the information revolution computing power and conventional operations with cyber capabilities, the verdict about offense-defense is more nuanced. These scholars argue that the capabilities created by computing also create vulnerabilities that interact to influence stability.178 It is the net balance between the capabilities and vulnerabilities that matters for the information revolution.

Like almost all literature that uses the offense-defense balance to predict and explain stability, these existing analyses offer conflicting conclusions about the true offensive or defensive advantage of the technologies of the information revolution. If anything, the overwhelming conclusion from the literature is that there may be no objective truth about the standing offense/defense balance in the midst of the information revolution. What then has driven offense-dominance in American doctrine during the information age?

Case Selection and Hypotheses

In order to answer these questions, I focus on three pivotal doctrinal developments in the United States military that occurred in conjunction with the information revolution. I chose to focus on U.S. doctrine for a series of reasons. First, by examining the evolution of doctrine within the United States and not across nations, I control for different types of governance, geography, and capacity. This allows me to trace the role of information age technology as an input in doctrinal development and,

178 Schneider, “Digitally Enabled Warfare: the Capability-Vulnerability Paradox”; Brantly, The Decision to Attack: Military and Intelligence Cyber Decision-Making.

specifically, to evaluate whether the introduction of these technologies impacts the decision to write offensive or defensive doctrine in the United States.

Secondly, there is a strong argument to suggest that the United States—as the hegemon in the international system—is a norm entrepreneur for conflict.179 The U.S., as a leader in NATO as well as a series of alliances in the Pacific, dictates how its allies prepare for conflict.180 In fact, one of the primary criticisms of the U.S. AirLand Battle

Doctrine in 1982 was that its focus on the offensive (particularly in regard to tactical nuclear use) would, by association, force other NATO members to fight offensive campaigns. In this sense, the U.S. system of alliances creates a critical mass in which the norm of U.S.-style warfare creates a cascading effect among a multitude of nations. The sheer number of countries that participate in these alliances with the United States creates a momentum in which non-aligned states, or even adversaries, may also adopt the U.S. style of warfare. This may have been especially true over the last twenty years since the demise of the Soviet Union. In particular, since the resounding success of the United

States against the Saddam regime in both 1991 and 2003, adversaries such as China and

Russia have modified their doctrine to adopt U.S. network-centric warfare. Therefore, studying U.S. doctrinal evolution during the information revolution may potentially shed light on other states’ subsequent development of information-driven doctrines.

In order to trace the influence of information revolution technologies, I examine

179 For a discussion of norms and national security see, Nina Tannenwald, The nuclear taboo: the United States and the non-use of nuclear weapons since 1945, (Cambridge: Cambridge University Press, 2007); Peter Katzenstein, The culture of national security: Norms and identity in world politics (New York: Columbia University Press, 1996); Martha Finnemore and Kathryn Sikkink, "International norm dynamics and political change," International organization Vol. 52, No. 04 (1998): 887-917; Martha Finnemore, The purpose of intervention: changing beliefs about the use of force (Ithaca: Cornell University Press, 2004). 180 See for example, Terry Terriff, “U.S. Ideas and Military Change in NATO, 1989-1994,” Theo Farrell and Terry Terriff, Eds., The Sources of Military Change (Boulder: Lynne Reiner, 2002): pgs. 91-118.

two pivotal doctrinal developments in the U.S. military, starting in 1976 and culminating in 2012. This time period captures the advent of computing technology as well as the codification of U.S. cyber weaponization with the creation of U.S. Cyber Command in

2011. I focus on two doctrines: AirLand Battle from 1976-1986 and the Joint Operational

Access Concept/Air Sea Battle Concept of 2012. I chose these doctrines because they were seen as pivotal periods in U.S. doctrine, they focus on state on state conventional conflict (as opposed to COIN doctrine), were influential in multiple services, and led to new weapons acquisitions programs.

Within this time frame are major advances in computing, network technology, and cyber weaponization. The antecedents of computing can be traced to the late 19th century, but modern computing really made its debut in World War II and the immediate years after. In that time period, the U.S. Department of Defense was heavily involved in the development of computer technology and the integration of that technology within weapons and campaigns. It deployed computers in its air defenses as early as the late

1950s and by the 1960s, it had introduced computer guidance to the Minuteman 1 missile system, computer data integration to its premier naval vessels, and had provided the impetus for the development of the Internet. Warfare in the 1960s and 70s saw some effects of these developments, but it was in the late 1970s and early 1980s that computing really took off as processing technologies became exponentially smaller and more powerful. In 1983, Time Magazine heralded the computer revolution as they announced the man of the year was not a man, but a machine—the computer. The 80s and 90s saw the rapid development of computing and network capabilities—so much so that in 1997

The Economist published an article entitled, “Select enemy. Delete” in which it claimed

that computers had revolutionized warfare and only the U.S. was prepared to harness that capability181. So ushered in the information revolution182 and with it the ubiquitous nature of everything cyber. From instant messaging to email to smart phones and smart weapons, computers and the networks they operated on became central to societies. In the meantime, as the U.S. Department of Defense integrated these cyber capabilities into its conventional war fighting arsenal, it also began to realize the opportunities for computer network attack through the cyberspace that had been created by the integration of computers into modern warfare.183 In 2010, it made the ultimate institutionalization of cyber network attack as a weapon in its own right through the creation of U.S. Cyber

Command.

This narrative of the development of computing and cyber technologies highlights the increased capabilities and importance that the information revolution has played in

U.S. warfare over the last 50 years. Arguably, these information revolution technologies have been the primary technological advances in modern war fighting. This case, therefore, provides an opportunity to evaluate the role that technology plays in the development of offensive and defensive doctrines. How has the advancement of computer processing and networks, and later the weaponization of cyber impacted the development of U.S. doctrine? Has the technology played a pivotal role in choices about

181 “Select Enemy. Delete,” The Economist, March 6, 1997. Of note, this article was also the impetus for Lieber’s 2000 critique of offense-defense balance and its ability to explain the implications of the information revolution for international stability. 182 Paul Levinson, The soft edge: A natural history and future of the information revolution (Psychology Press, 1998); Robert O. Keohane and Joseph S. Nye Jr. "Power and interdependence in the information age." Foreign affairs (1998): 81-94; Joseph S. Nye Jr. and William A. Owens. "America's information edge." Foreign affairs (1996): 20-36; Steven Metz, Armed conflict in the 21st century: the information revolution and post-modern warfare (Carlisle:Strategic Studies Institute, 2000). 183 Fred Kaplan, Dark Territory: The Secret History of Cyber War (New York: Simon and Schuster, 2016); Jason Healey, A Fierce Domain: Conflict in Cyberspace, 1986 to 2012 (The Atlantic Council: 2013); Richard A Clarke and Robert K. Knake. Cyber war. (New York: HarperCollins, 2011).

offense and defense in the doctrine? Or have other factors, like organizational biases, structural politics, or civil-military relations trumped the role of these technologies? To answer these questions, I examine the development of U.S. doctrine in two periods. First,

I look at the evolution of Army Doctrine in the late 1970s and early 1980s to trace the impact of microprocessing and supercomputers on offense and defense in doctrine.

Secondly, I examine the period of 2009-2016 to understand how the 2010 weaponization of cyber through the creation of U.S. Cyber Command as well as the revolutionary advances in information and communication technologies affected offense and defense decisions in U.S. doctrine. This analysis explores the following competing hypotheses:

H1: Information revolution technologies drove the development of offensive or defensive

U.S. doctrine during the information revolution.

- The development of computer processing and network capabilities drove the

offensive or defensive of nature of U.S. doctrine in the early 1980s.

- The development of cyber network attacks drove the offensive or defensive

nature of U.S. doctrine in the late 2000s.

H2: Military organizational biases drove the development of offensive or defensive U.S. doctrine during the information revolution.

H3: Civil-military relations drove the development of offensive or defensive U.S. doctrine during the information revolution.

In order to examine these competing hypotheses, I first determine whether the doctrine is primarily offensive, defensive, or neutral. I do this in a variety of ways. First,

I examine the doctrinal text and analyze whether the objectives use offensive or defensive language. For example, words like seize, attack, maneuver, exploit, and disrupt imply offensive operations while words like protect, defend, attrite, fortify, shield, safeguard all imply defensive operations. I perform both an objective and subjective analysis of this language. Objectively I ask, is the preponderance of verbs offensive or defensive? Does the doctrine devote more space to offense than defense? Subjectively, does the doctrine argue for an offensive or defensive advantage? I then turn to contemporary and post- doctrinal writings within professional military publications to understand whether the practitioner community interpreted the doctrine as offensive or defensive. Finally, when possible, I also conducted interviews and examined oral histories and memorandums to piece together whether the creators of the doctrine believed there was an offensive or defensive bent to their creation.

Once I determine the offensive or defensive nature of these doctrines, I then turn to the military and scholarly analysis of the doctrinal development as well as interviews and primary source documents to trace the motivations behind this offensive or defensive bent. If information revolution technologies influenced the offensive or defensive nature of these doctrines, then these sources will point to specific technologies that influence the

ODB. However, if technology is not the dominant factor, then we will see heavy influence from within military organizations, domestic politics, culture, or international structures.

Case 1: AirLand Battle and the Rise of Modern Computing

U.S. Army Field Manual 100-5 was the primary land-based operational war fighting doctrine from 1939 to 2001. As Michael Cannon explains in his analysis of the doctrine, “it is the manual from which all other Army field manuals flow and so assumes an importance above that of all the others.”184 It is the product of the Army Training and

Doctrine Command (TRADOC) and is used as the primary authority for planning operational land campaigns and, in some instances, all served a strategic guidepost for the allocation of forces and capabilities. From its inception prior to World War II to the final iteration in 2001, there was variation in offense and defense dominance. Offense dominated from 1949-1958. However, in 1962, with the advent of flexible response, the doctrine became more neutral between offense and defense and stayed largely the same throughout the Vietnam conflict.

However, in 1976 a major re-write of the doctrine, led by General William

DePuy, made a drastic shift to favor defense. Active Defense, as the doctrine was colloquially known, asserted that the balance between offense and defense had shifted so that “the defender has ever advantage but one—he does not have the initiative.”185

Active Defense writers were strongly influenced about the dominance defense by the

1973 Yom Kippur War. 186 The war featured the advent of precision-guided munitions

184 Michael Cannon, The Development of Field Manual 100-5 from 1945 Until 1976 (Alexandria: Army Military Personnel Center, 1984), pg. 1. 185 Department of the Army, FM 100-5: Operations (Washington: 1976), pg. 5-2. 186 John L. Romjue, From Active Defense to AirLand Battle: The Development of Army Doctrine 1973- 1982 (Fort Monroe: United States Army Training and Doctrine Command, 1984); Jeffrey W. Long, The Evolution of Army Doctrine: Active Defense to Airland Battle and Beyond (Carlisle: U.S. Army Command and General Staff College, 1991); Wass de Czege, “Lessons from the Past: Making the Army’s Doctrine ‘Right Enough’ Today”; Richard Lock-Pullan, “How to Rethink War: Conceptual Innovation and AirLand Battle Doctrine,” The Journal of Strategic Studies Vol. 28, No. 4 (2005): 679-702; Paul H. Herbert, Deciding What Has to be Done: General William E. DePuy and the 1976 Edition of FM 100-5 Operations, (Carlisle: U.S. Army Command and General Staff College, 1998); Joyce, “The Micropolitics of ‘the Army

and suggested to the doctrine designers that for states to win in conflicts with these technologies, they would have to be able to take initial losses and defend against increasingly precise and long-range strikes. For the small group of DePuy’s doctrine- writers, the nature of war had fundamentally changed to advantage the defense primarily because of the advent of new defense-dominant technology. As Joyce writes, “DePuy seems to have been driven by . . . a belief that the 1973 Arab-Israeli War showed a modern battle to be too lethal to risk offensive action.”187 Active Defense doctrine creators were especially taken with the lethality of tank guns with increased muzzle velocity, antitank weapons with precision guidance and laser range finders, and air defense missiles that coupled new infra-red and radar guidance with improved early

warning systems. 188 Their doctrine is replete with these technologies, filled with

Lanchester equations of attrition and technology-driven analyses of offense-defense balance.189 As the doctrine explains:

The war in the Middle East in 1973 may well portend the nature of modern battle. Arab and Israelis were armed with the latest weapons, and the conflict approached a destructiveness once attributed only to nuclear arms. Use of aircraft for close support . . . was greatly reduced by advancing surface-to-air missiles and air defense guns.190 . The doctrine makes it clear that technology drove a shift in U.S. war planners’

you Have’: Explaining the Development of U.S. Military Doctrine After Vietnam”; William E. Depuy, "The Implications of the Middle East war on US Army Tactics, Doctrine and Systems." Keynote Address,(Washington, DC: Training and Doctrine Files (1975); Saul Bronfeld, “Fighting Outnumbered: The Impact of the Yom Kippur War on the U.S. Army,” The Journal of Military History 71(April 2007): 465-498. 187 Joyce, “The Micropolitics of ‘the Army you Have’: Explaining the Development of U.S. Military Doctrine After Vietnam,” Studies in American Political Development Vol. 26 (2012): 193. 188 Nadav Safran, “Trial by Ordeal,” International Security Vol.2, No.2 (1977): 133-170; Robert Bolia, “The Over-reliance on Technology in Warfare: the Yom Kippur War as a Case Study,” U.S. Army command and General Staff College, 2004. 189 Department of the Army, FM 100-5: Operations (Washington: 1976). 190 Ibid, pg. 2-2.

perceptions of the ODB from offense-defense neutrality to an advantage for the defense.

DePuy confirmed the pivotal role of technology in the doctrinal shift, writing in 1976 to the Chief of Staff of the Army that the Army was at “an historic turning point . . .we must organize the Army to employ and maintain the modern weapons which can drive the outcome on the battlefield. Thus, we started FM 100-5 with a rather long discussion of weapons.”191

The shift to defense in 1976 represents an attempt to quantify offense-defense balance that, in recent history, probably most closely mirrors technological understandings of ODB. The guidance to develop the doctrine was explicit that technologies and measurement of balances of capabilities were to be prioritized over intangible factors. In the 1974 guidance by the TRADOC Chief of Staff, the doctrine development team was told “tactics had to be based on hard, cold facts and had to be taken out of the abstract. TRADOC had to examine the most recent military experience and employ the best weapons data—such as . . . hit probabilities of Soviet weapons as a function of range.”192 Accordingly, TRADOC set out on an ambitious attempt to simulate, model, and quantify the ease of attack versus the ease of defense.193

Clearly in 1976, the shift to defense in U.S. doctrine was driven by technology.

But was this ODB shift related to advances in information revolution technology? The early 1970s was certainly an active period of development for computer processing.

Since the last doctrinal update in 1968, UNIX operating system was developed, floppy

191 Letter from William E DePuy to Fred C. Weyand, February 18, 1976. 192 Romjue, From Active Defense to AirLand Battle: The Development of Army Doctrine 1973-1982, pg. 20. 193 Benjamin Jensen, Forging the Sword: Doctrinal Change in the U.S. Army (Palo Alto: Stanford University Press, 2016).

disks made their advent at IBM, Intel introduced the first microprocessors, and the first personal computers began to emerge. These advances had significant impact on air defenses, which enabled better cueing for surface to air missiles like the SA-2 and SA-3.

However, it was not the SA-2 and SA-3, or even early warning radar facilities, that pre- occupied the Active Defense drafters. Instead, it was the guidance of hand-held, electro- optical and infra-red anti-tank and air defense missiles coupled with advances in firepower that influenced the perception of a defensive advantage. These were largely non-digital upgrades. For example, the doctrine speaks extensively about the new lethality of Anti-Tank (TOW) missiles, focusing on improvements in penetration power, range, and laser range finding capabilities. The clear implication is that improvements in lethality—firepower, kinematics, and guidance—advantage defensive campaigns. While computing plays a role to a limited extent in various weapons’ guidance, the physical kinematics of firepower drove the doctrine authors’ belief in defensive dominance.

This is not to say that the doctrine fails to discuss the implications of information technologies. In fact, in the doctrine’s analysis of war in the electromagnetic spectrum

(which is arguably the predecessor to modern cyberspace domains) the doctrine argues that “any commander is prone to defeat, whatever his strength in numbers and weapons, if EW denies him the means to convey orders, provide for fire support, or to arrange for logistics and administration. EW is now a form of combat power [italics in doctrine].”194

Perhaps interesting to later doctrinal development and significant to tracing the impact of information technologies on ODB, the authors in the 1976 FM 100-5 conclude that EW sparring may not advantage either the offense or the defense. As they write, “ in the

194 Ibid, pg. 2-27.

future, an invisible yet crucial struggle with be waged to attack or defend the means of command . . . countermeasures will be met by countermeasures.”195 In fact, their analysis of EW within the Yom Kippur War concludes ambivalently that both sides attacked and defended without any decisive winners on the offensive or defensive side.196

The 1976 “Active Defense” doctrine had a relatively short life. It was revised in

1982 with a much more offensively-advantaged AirLand Battle. As AirLand Battle asserts about the ODB, “the offense is the decisive form of war, the commander’s only means of attaining a positive goal or of completely destroying an opposing force.”197 In fact, even though the authors devote a chapter to defense (half the size of the offense chapter), they conclude their remarks on defense by proclaiming that, “to win, one must attack.”198 As AirLand Battle doctrine explains about the offensive advantage, “this doctrine is based on securing or retaining the initiative and exercising it aggressively to defeat the enemy. Destruction of the opposing force is achieved by throwing the enemy off balance with powerful initial blows from unexpected directions and then following up rapidly to prevent his recovery. The best results are obtained when initial blows are struck against critical units and areas whose loss will degrade the coherence of enemy operations.”199

The shift from the defense to the offense was not the only significant change from the 1976 to 1982 doctrine. Missing from the 1982 doctrine were the technical weapons

195 Ibid, pg. 2-26. 196 See also, William E. DePuy, “High Technology and Land Warfare,” Asa Clark and John Lilly, Eds., Defense Technology (New York: Praeger, 1989), pgs. 119-128. In the article DePuy re-emphasizes the importance of EW and C3 on conventional balances, writing that “a new dimension of war is emerging—a war of detection and counterdetection” (pg. 124). 197 Department of the Army, FM100-5: Operations, 1982, pg. 8-1. 198 Ibid, pg. 10-1. 199 Ibid, pg. 2-1.

analyses of the 1976 Active Defense. Instead of graphs depicting predicted probabilities of kill, the 1982 AirLand Battle drew largely from history, narrating their case for offensive advantage.200 As Long describes, AirLand Battle “stressed the human dimensions of war instead of the technical; and it favored the offense over the defense.”201 In fact, when listing the three fundamentals to victory, the doctrine first listed humans, then technology, and finally doctrine.

Indeed, the consensus on the 1982 AirLand Battle was that it was not technology that drove the development of offensive doctrine. Instead, a series of pressures led to an offensive doctrine that subsequently drove the development of technologies needed to conduct those offensive strategies.202 As Wass de Czege, one of the primary authors of the doctrine, explains, AirLand Battle “provides the parameters within which technology should be pursued.”203 Long provides evidence for this assertion when he traces doctrinal change between 1976 and 1982, arguing that AirLand Battle “called for advances that surpassed existing civilian technology.”204 According to his analysis, the technology needed to implement the 1982 doctrine, and later the 1986 follow-on doctrine, was not yet in development at the time the authors created the 1982 AirLand Battle. Instead,

Long argues that institutional incentives led to the development of the offensive doctrine and offensive technology developed subsequently from those institutional incentives.

The doctrine itself argues that the offensive bent of AirLand Battle and the need to attack

200 Long, The Evolution of Army Doctrine: Active Defense to Airland Battle and Beyond 201 Long, The Evolution of Army Doctrine: Active Defense to Airland Battle and Beyond, pg. 48. 202 Long, The Evolution of Army Doctrine: Active Defense to Airland Battle and Beyond; Wass de Czege, “Army Doctrinal Reform,” The Defense Reform Debate: Issues and Analysis, Asa Clark, Peter Chiarelli, Jeffrey McKitrick, and James Reed, Eds. (Baltimore: Johns Hopkins Press, 1984); John L. Romjue, From Active Defense to AirLand Battle: The Development of Army Doctrine 1973-1982; Lock-Pullan, “How to Rethink War: Conceptual Innovation and AirLand Battle Doctrine,” The Journal of Strategic Studies. 203 Huba Wass de Czege, “Army Doctrinal Reform,” The Defense Reform Debate: Issues and Analysis. 204 Long, The Evolution of Army Doctrine: Active Defense to Airland Battle and Beyond, pg. 71.

in depth was a product of terrain and numerical inferiority.205 According to the doctrine, technology did not inherently create the offensive advantage. Instead, the U.S. needed to create the technology that would provide the means to conduct the offensive operations the Army believed it needed in order to succeed against the Soviet Union.

So if, as the authors argue, there was no existing technological offensive dominance to drive the move towards offensive doctrine, then what might have driven the offensive tilt in AirLand Battle?

Organizational Incentives

One of the strongest arguments against technology driving the offensive bias in

AirLand Battle is the role of organizational influences and the Army desire for an offensive culture. The previous doctrine, Active Defense, lasted only a few years and generated significant dissent from the Army. Professional Army publications were strident in their critique of the defensive and technologically-focused bent of Active

Defense.206 Arguments against Active Defense from the professional Army corps found a receptive audience in General Starry, who led the development of AirLand Battle. As

Starry reminisces, when he took over from General DePuy, “I found an Army in the field that looked upon itself as just a bunch of speed bumps on the way to the Rhine as far as the Soviets were concerned. They didn’t think they could win. There is nothing more frightening that an army, the American Army particularly, that thinks it can’t win. We had to do something . . .”207

205 Department of the Army, TRADOC Pam 525-5: The Airland Battle and Corps 86, 1981: pg. 6. 206 Joyce, “The Micropolitics of ‘the Army you Have’: Explaining the Development of U.S. Military Doctrine After Vietnam” 207 Lewis Sorley, Ed., Press On! Vol.2 (Fort Leavenworth: Combat Studies Institute Press, 2009), pg. 1265.

For Starry, AirLand Battle had to transcend current technology and Lancaster equations and instead harken back to what he believed to be fundamental characteristics of war. In this perception of historical concepts of war, Starry saw the initiative and the offensive as necessary characteristics of victory—regardless of context, technology, or time. As he recounts about his differences with DePuy, “I do believe if he and I had a difference of opinion about doctrine it lay in his conviction that, because of the lethality of modern weapons, improvements of fire control, the lethality of the weapons themselves, that whole combination of things, defense was the best way to fight. You see that in Active Defense. Basically I do not agree with that. On the other hand, the history of battle instructs that, somewhere in the battle, the guy who seizes the initiative is the guy who wins . . . AirLand Battle is about taking the initiative.”208

The new Army doctrine under Starry would codify these historical precepts and then subsequently drive the development of doctrine that supported the initial perceptions of offensive dominance. Therefore, Starry set about to create the acquisition and organizational requirements necessary for turning the offensive doctrine of AirLand

Battle into weapons and units that could execute offensive campaigns. As he explained,

“It is part of the American psyche to presume that technology is going to win everything .

. . technology wins nothing unless it serves some doctrinal purpose.”209

Civil-military Relationships

The offensive bias of the Army under Starry found its ally within the civilian sectors of the Reagan administration. In a particularly influential piece by William Lind,

208 Ibid, pg. 1268. 209 Ibid, pg. 1271.

the Senate aid argued that not only was the defensive approach of Active Defense incorrect, but it was incorrect because the doctrine had failed to account for a historical precedent that advantaged the offense.210 It was a narrative that resounded with a

Presidential administration that sought to take back the initiative from the Soviets and infuse the Department of Defense with a large weapons acquisition and modernization program.211 Politics would reinforce the Army impetus for offensive doctrine and codify the offensive bias of AirLand Battle in the largest weapons acquisition efforts the United

States would see for twenty years.

We can follow the logic of how offensive bias and politics drove technology for

AirLand battle. The Army perception of offensive dominance, combined with a civilian administration sympathetic to forward-leaning defense weapons acquisitions programs, combined to create a national security strategy and military that would drive the technological development of the information age to the offensive. To understand this process, let us trace a few of the technologies that developed from one of the core offensive operations envisioned by AirLand Battle. A significant component of the offensive shift of 1982 was the focus on “deep attack.” AirLand Battle envisions deep battle that “prevents the enemy from massing and creates windows of opportunity for offensive actions.”212 In order to conduct these deep attacks, the doctrine advocated for technologies with long ranges and high mobility. In turn, these deep interdiction

210 William S. Lind, “Some Doctrinal Questions for the United States Army,” Military Review 57 (March 1977): 58 211 Jeffrey Richelson, "PD‐59, NSDD‐13 and the Reagan strategic modernization program," The Journal of Strategic Studies 6, no. 2 (1983): 125-146; Barry Posen and Stephen Van Evera. "Defense policy and the Reagan administration: Departure from containment." International Security 8, no. 1 (1983): 3-45; David Hendrickson, Reforming defense: The state of American civil-military relations. (Baltimore.: Johns Hopkins University Press, 1988); Richard Stubbing, "The Defense Program: Buildup or Binge?." Foreign Affairs 63, no. 4 (1985): 848-872. 212 Department of the Army, FM100-5: Operations, 1982, pg. 7-13.

weapons called for new intelligence capabilities and advances in command and control.

As Long describes, the doctrine’s “emphasis on command and control and deep attack justified a new radio, Mobile Subscriber Equipment, the Maneuver Control System, the

Army Tactile Missile System, and the Joint Surveillance and Target Attack System . . . analysts credit AirLand Battle with restoring the Army’s aggressive, warrior spirit.”213

These technologies were part of the Corps 86 acquisition program.214 Corps 86 and the development of acquisition processes explicitly tied to doctrinal development was a bureaucratic initiative tied to AirLand Battle. General Starry, who led the development of AirLand Battle, wanted to make sure that doctrinal development was not constrained by technology and instead would be a priori and drive the subsequent development of technology. As Benjamin Jensen concludes in his analysis of U.S. Army doctrine, “the intent was to construct an operational concept that would ‘drive the labs’ and get a handle on material developments by ‘getting ahead of the technology.’”215 This is essentially what happened with Corps 86, which created requirements concurrent with doctrine and drove the development of technologies that did not see operational deployment until the late 80s and potentially early 90s. In fact, JSTARS, an airborne system of providing long- range situational awareness and target-cueing was developed from the Corps 86 program and traces its initial design to AirLand Battle 1982. The system was first deployed in the field in Desert Storm in 1991.216 The figure below charts these developments from the

Corps 86 program to the 1986 version of the AirLand Battle doctrine, with the 1982

213 Long, The Evolution of Army Doctrine: Active Defense to Airland Battle and Beyond, pg. 3. 214 Ibid, pg. 72. 215 Jensen, Forging the Sword, pg. 82. 216 Long, The Evolution of Army Doctrine: Active Defense to Airland Battle and Beyond, pg. 75.

AirLand Battle Concept as the primary driver of these technological developments.217

Figure 1. Corps 86 Acquisitions Program

Computing power, while not identified singularly in the figure above, is integral to the successful creation of the technologies and operations envisioned in 1982’s offensive doctrinal shift. Within the Corps 86 Operational Concept, computing is highlighted multiple times as a necessary technology for the operational concept to

217 While the Corps 86 acquisition program and operational concept was published in 1981 under the title TRADOC Pam 525-5, AirLand Battle and Corps Operations—1986, it was created in concert with the FM 100-5 revision that was formally published in 1982. These documents should be thought of as contemporaneous to one another.

succeed. “Micro-computers” were necessary for joint targeting218 and computer simulations would be essential to test the new operation. Word processing and databases would store the extensive information required to conduct the offensive campaigns of

AirLand Battle while advances in supercomputing and the miniaturization of processing would allow decentralized command.219

But where the AirLand Battle concept saw the impact of computing was in the future. The operational concept, which in 1981 is at the cusp of the computer processing revolution, predicts that the ability to develop offensive operations to win a competition with the Soviet Union is reliant on who can harness computing power best. The concept explains that there is, “promise of a major increase in combat effectiveness with current means. There also exists an enhanced capability to exploit new sensors, weapons, and command control systems as they are fielded. This enhanced capability is even more evident in the field of micro-processors and computers. As a nation, we have a considerable advantage over our potential adversaries in this technological field. If we strive to put that advantage to work for us, it could become a significant multiplier.”220

Center for Naval Analysis review conducted in 1988 by Douglas Skinner goes further even than the Operational Concept and identifies computing as the central capability required to generate AirLand Battle offensive operations. He writes:

“These requirements create the need for increasingly more sophisticated C3I systems, which will give the ground commander the necessary intelligence and reconnaissance so as to be able to assess the disposition of his forces and those of the enemy. The battlefield will be complex and

218 Department of the Army, TRADOC Pam 525-5: The Airland Battle and Corps 86, 1981: pg. 22. 219 General Don Starry, “Letter to Major General Albert N. Stubblebine Jr., U.S. Army Intelligence and Security Command,” May 12, 1981, as published in Lewis Sorley, Ed., Press On! Vol.2 (Fort Leavenworth: Combat Studies Institute Press, 2009), pg. 477; General Don Starry, “Computer Systems Needs: Message to Major General Clay Buckingham,” January 29, 1979 as published in Sorley, Press On! Vol. 2, pg. 661. 220 Ibid, pg. 24.

rapidly changing. High-speed computers that can process, analyze, and sometimes make decisions on the masses of information produced will be a central feature of such battles. Areas of influence/interest require the ability to ‘see’ beyond the limited horizon defined by the traditional front, further increasing the need for timely and accurate intelligence over a large area. Thus there is a need for sophisticated sensor technology, and, again, computers to process the information thus gained . . . Advanced computer technology, both in software and hardware figures centrally in the AirLand Battle. The development of very-high speed integrated circuits (VHSIC) is well underway, allowing high-speed processing of large amounts of command and control information. VHSIC provide for the processing and analysis of intelligence data, make possible ‘natural language’ recognition, give ‘vision’ ability to target acquisition and guidance, and perform tasks requiring artificial intelligence capabilities. Much of the technological development in this area is concentrated on small, highly reliable, high-speed computers for use at the tactical level.”221

According to this line of reasoning, AirLand Battle drove the acquisition of computing technologies to advantage the offense. The acquisition decisions made in the late 1980s, with the advent of Reagan and inflated defense budgets, were a critical juncture that created a path dependency for the use of cyber computing power to advantage offensive operations. The doctrine could have driven the advances in computing towards defensive operations—for example investing in long-range detection technologies tied to air defenses—but instead the doctrine focused on using computing processing power to create offensive technologies that would allow the United States to utilize offensive operations.

The alternate hypothesis is that the computer advances occurring at the same time

AirLand Battle was developed were actually the chicken that laid the egg, and not vice- versa. It certainly is difficult to disentangle the influence of technology on how doctrine developers imagined technology could impact the future battlefield. As Skinner argues,

221 Douglas W. Skinner, Airland Battle Doctrine (Alexandria: Center for Naval Analyses, 1988), pgs. 22 and 25.

one of the most important factors in the development of AirLand Battle doctrine was

“rapid developments in technology.”222 It is also impossible to completely disentangle the doctrinal development of AirLand Battle with what is now being called the Second

Offset, or Under Secretary of Defense Perry’s focus on technology to overcome Soviet quantitative advantage.223 Perry’s initiatives speak very explicitly about the role that computer processing will play in U.S. offensive advantage, though the application is more specifically tied to U.S. air capabilities: “microelectronics technology will play a critical role in maintaining air superiority and in providing the next generation of electronics.”224

Despite the explicit linkage between ongoing developments in information technologies and the Second Offset, the discussion about computing within both the doctrine and the AirLand Battle concept are rife with uncertainty. That, compounded with the relative sparseness of “computing” versus the focus on lasers or electronic countermeasures suggests that the leaders of doctrine development didn’t completely comprehend the advances in computing that were occurring. And they certainly didn’t have a debate about whether those technologies were more or less likely to advantage the offense or the defense. Instead, it appears the Army believed it needed to fight offensively to defeat the Soviets and sought out technology that would create the offensive advantage they needed for their attack in depth operations to succeed.

The FM 100-5, Active Defense to AirLand Battle case shows a shift from defensive to offensive doctrine. In the 1976 case, that shift to defensive balance is driven

222 Ibid, pg. 26. 223 William Perry, "Defense Investment Strategy." Foreign Affairs 68, no. 2 (1989): 72-92; William J. Perry, The Department of Defense Statement on Technology and Military Manpower, December 4, 1980. 224 Ibid, pg. 4.

by technology but it is not related to cyber computing power. The move to offensive advantage in 1982, unlike in 1976, does not appear to be driven by technology.

However, the offensive operations that the doctrine suggests create a need for offensive weaponry enabled by cyber computing power. This in turn creates a family of offensive weapons enabled by computers that advent in the early 1990s.

Case 2: Air Sea Battle, Joint Operational Access Campaign, JAM-GC and Cyber

Weapons

From AirLand Battle to the 1990s until Air Sea Battle in 2012, technology-driven doctrine dominated U.S. operations and acquisition strategies. From effects-based operations to network-centric warfare to the revolution in military affairs, the planners of the U.S Department of Defense saw emerging technology, and in particular computing, as pivotal to increasingly offensive operations.225 Over twenty years separate AirLand

Battle and Air Sea Battle. Air Sea Battle is written in a new threat environment, with new international dynamics, and new technology. However, it still traces its offensive roots to AirLand Battle, making it an apropos case to trace the effect of cyber on perceptions of conventional ODB. As Air Sea Battle explains:

AirLand Battle was developed in the 1970s and 1980s to counter a Soviet backed combined arms attack in Europe. A key component of AirLand Battle was the degradation of rear echelon forces before they could engage allied forces . . . The ASB Concept is similarly designed to attack-

225 For a review of the literature of the time see, William Owens and Ed Offley. Lifting the fog of war (Baltimore: JHU Press, 2001); David Deptula,. Effects-Based Operations: Changes in the Nature of War (Arlington: Aerospace Education Foundation, 2001); Michael Mazarr, The revolution in military affairs: A framework for defense planning (Carlisle: Army War College Strategic Studies Institute, 1994); Arthur Cebrowski and John J. Garstka. "Network-centric warfare: Its origin and future." US Naval Institute Proceedings, vol. 124, no. 1: 1998, pp. 28-35; Eliot A. Cohen, "A revolution in warfare." Foreign Affairs (1996): 37-54; Joseph S. Nye, Jr., and William A. Owens. "America's information edge." Foreign affairs (1996): 20-36; Andrew Krepinevich. "Cavalry to computer: The pattern of military revolutions." The National Interest 37 (1994): 30-42.

100

in-depth, but instead of focusing on the land domain from the air, the Concept describes integrated operations across five domains (air, land, sea, space, and cyberspace) to create advantage. The ASB Concept further differentiates itself from its predecessor in that the ASB Concept also strives to protect our rear echelon across the same domains. The defensive aspect of ASB helps the Joint Force reduce risk in the face of increasingly longer range and more precise weapons which could affect our space-based platforms, land forces, capital ships, and network infrastructure.226

Air Sea Battle (ASB) began to surface within the Washington beltway in 2009.227

While it would take years for ASB to coalesce into a published and releasable war fighting concept, it proved to be the guiding concept for U.S. war fighting principles (or at least the only ones that managed to make it to the forefront of the DoD’s public face)

st in near-peer conflict during the 21 century.228 Shortly after ASB was officially released to the public, the Joint Operational Access Concept doctrine was published. The doctrine, under which Air Sea Battle technically falls as a sub-concept, provides the roadmap for U.S. offensive and defensive operations in antiaccess and area-denial scenarios in order to “project military force into an operational area with sufficient freedom of action to accomplish the mission.”229 ASB has since evolved into the re- named Joint Concept for Access and Maneuver in the Global Commons (JAM-GC).

226 Air Sea Battle Office, Air Sea Battle: Service Collaboration to Address Anti-Access and Area Denial Challenges, May 2013, pg. i. 227 Sam Lagrone, “Pentagon’s Air-Sea Battle Plan Explained. Finally.” Wired, August 6, 2012: https://www.wired.com/2012/08/air-sea-battle-2/; Milan Vego, "AirSea Battle Must Not Work Alone." Naval Proceedings, vol. 137: 2011. 228 Interestingly, the Information Warfare, network-centric warfare, Effects-based operations, and revolution in military affairs of the late 90s and early 2000s never codified in an actual doctrine or concept and so Air Sea Battle (and later the Joint Operational Access Concept and JAM-GC) can be seen as the first codification within an operational concept of the computer-based operations that came to the forefront of tactics and operational planning after AirLand Battle. For more information on the pervasiveness of ASB within U.S. doctrine and operations post 2009, see Terry S. Morris, Martha VanDriel, Bill Dries, Jason C. Perdew, Richard H. Schulz, and Kristin E. Jacobsen, “Securing the Operational Access: Evolving the Air- Sea Battle Concept,” National Interest, February 11, 2015: http://nationalinterest.org/feature/securing- operational-access-evolving-the-air-sea-battle-12219?page=2. 229 U.S. Department of Defense, Joint Operational Access Concept, January 2012.

101

Significant changes between ASB and JAM-GC include a greater emphasis on land operations as well as increased focus on operations in digitally-contested environments.

As the authors of the new concept explain, “current doctrinal command and control methodologies will likely be inadequate to address A2/AD environments where beyond- line-of-sight communications and other connectivity between units can be disrupted or denied by an adversary.”230 Because JAM-GC has not been released to the public, in this analysis I will focus on JOAC and ASB.231 The figure below depicts the relationship between strategy, JOAC, and ASB.

Figure 2. JOAC/ASB Family of Plans. Source: “Securing the Operational Access: Evolving the Air-Sea Battle Concept”232

ASB, as the ideational precursor to the JOAC, is an appropriate place to begin tracing perceptions of offense and defense. The concept, which the authors claim to be more defensive than AirLand Battle, is unabashedly offensive. In fact, the entire aim of

230 Terry S. Morris et al, “Securing the Operational Access: Evolving the Air-Sea Battle Concept.” 231 Despite the final version of JAM-GC still circulating through the Department of Defense, the initial discussion about the change to ASB indicates that there will be even greater emphasis on the role of cyber operations and, in particular, the offensive advantage of adversary operations with cyberspace. For more information, see the above-mentioned Morris et al article, “Securing the Operational Access: Evolving the Air-Sea Battle Concept.” 232 Terry S. Morris et al, “Securing the Operational Access: Evolving the Air-Sea Battle Concept.”

102

the concept is to ensure “effective power projection,”233 whether that be in ensuring freedom of the seas or defending U.S. allies from the threat of maritime or air aggression

(enemy A2/AD capabilities). It aims to first disrupt the enemy’s ability to locate or target

U.S. forces, destroy enemy capabilities to engage those forces, and then defeat any employed enemy weapons.234 Like AirLand Battle, ASB calls for increasingly long- distance attacks in depth. The operational concept calls first for striking intelligence, surveillance, and reconnaissance platforms and capabilities; once the enemy is sufficiently blinded, the U.S. conducts a “ ‘missile suppression campaign’ to disrupt the enemy’s air defense networks, using stealthy long-range platforms and supported by submarine launched weapons and sensors.”235 This allows the U.S. to gain dominance of the air and maritime domains in order to conduct attacks on the enemy’s offensive capabilities. The U.S. then focuses its combat efforts on denying the adversary the ability to launch both air defense and surface-to-surface missiles, as well as the infrastructure necessary to launch a large scale amphibious operation. Finally, the U.S. conducts cleanup operations, defeating any missiles, air, or maritime weaponry that the enemy managed to launch as the U.S. was destroying their operational centers of gravity.236 A figure of ASB is included below. Notice the verbs used: disrupt, destroy, defeat. These verbs are all offensive in nature while the end result of the actions is to “sustain offensive operations.” Indeed, where defense (or protection) is discussed, the focus is not on defending in order to win a war of attrition. Instead, it is in the “defense of offensive

233 Philip Dupree and Jordan Thomas, “Air-Sea Battle—Clearing the Fog,” Armed Forces Journal, June 1, 2012: http://armedforcesjournal.com/air-sea-battle-clearing-the-fog-2/. 234 “Air Sea Battle Implementation.” 235 Richard Bitzinger and Michael Raska. “The Air Sea Battle Debate and the Future of Conflict in East Asia,” RSIS Policy Brief, accessed 27 September 2013. http://www.rsis.edu.sg/publications/policy_papers/RSIS_Air%20Sea%20Battle_190213%20v1_Print.pdf 236 Ibid.

103

capabilities.”237

Figure 3. U.S. Air Sea Battle Strategy. Source: U.S. Air Sea Battle Office “Concept Implementation Summary May 2013”

The focus on offensive operations within ASB is a continued theme that traces back to the shift to offensive doctrine in AirLand Battle 1982. In 1982, doctrine writers insisted that this shift towards the offensive was not because of technology, but instead because the nature of war emphasized the inherent advantage of offensive operations. In contrast, ASB links offensive advantage to the rise in technology—and in particular a shift in the perceived technological advantage between the United States projecting power and adversaries seeking to deny that power projection. As the concept explains,

237 E-mail with Col Jordan Thomas, Ret., October 18, 2016.

104

“While A2/AD ideas or not new . . . technological advances and proliferation threaten stability by empowering potentially aggressive actors with previously unattainable military capabilities.”238

The doctrinal successor and overarching concept over ASB is the Joint

Operational Access Concept (JOAC). Like ASB, JOAC emphasizes the advantage of offensive operations. The figure below, which summarizes the operational precepts of the JOAC, depicts the heavy perception of offensive advantage in modern battle.

Highlighted in red are the precept verbs that entail offensive operations, while highlighted in green are the defensive operations. Of the 11 precepts, eight are entirely offensive, two are neutral, and two are offense-defense balanced. Unlike AirLand Battle which devoted an entire section to defense, JOAC (like ASB) refers only to defense as the protection of two vulnerabilities: air bases and space/cyber assets. With the exception of the precepts concerning these two vulnerabilities, the JOAC charges the U.S.

Department of Defense to organize and fight for a war that “seizes the initiative,”

“attacks in depth,” “maneuvers,” “penetrates,” and “exploits advantages.” The microprocessing and computing capability advocated by AirLand Battle and Corps 86 is vital to the C3I required to conduct this offensive, highly maneuverable, highly dynamic, long-range operation. Clearly, the computing power and network capability created by the information revolution continues to enable offensive operations in U.S. doctrine.

238 Air Sea Battle Office, “Air-Sea Battle: Service Collaboration to Address Anti-Access and Area Denial Challenges,” September 2012, pg. 2.

105

Offense Defense

Figure 4. Joint Operational Access Concept. Source: Joint Operational Access Concept, pg. 17

Not only have there been significant advances in computing since AirLand Battle, the information revolution has also introduced the ability to conduct operations within cyberspace, to include cyber network defense, cyber network attack, and cyber network exploitation (intelligence). What role do these new cyber operations play in the offensive doctrines with ASB and JOAC? As operations became increasingly more dependent on computers from 1982 to 2012, so also did new capabilities to attack these cyber nodes begin to emerge. By the mid 1990s, both the U.S. intelligence community (led by the

106

National Security Agency) and elements of the Department of Defense were organizing the capabilities and manpower to attack these new cyber vulnerabilities.239 In 1998,

President Clinton released the first national cyber strategy and recognized the increasingly prolific dependencies within the U.S. society and military on computers, warning that the U.S. was “increasingly reliant upon certain critical infrastructures and upon cyber-based information systems.”240 Shortly after, in response to a series of hacks into DoD systems, the military created the Joint Task Force for Computer Network

Defense. The organization would become the predecessor for the current U.S. Cyber

Command. Formed in 2009, U.S. Cyber Command represents the initial operating capability of cyber weapons. While offensive operations occurred sporadically in the late

90s and early 2000s as well as under the guise of the National Security Agency prior to

2009, the creation of a formal military command to conduct defensive and offensive operations represented the use of the cyberspace domain and offensive cyber actions as a viable tool of both American military power and adversary coercive tactics. In 2012, rumors of a joint U.S.-Israeli cyber attack on Iranian nuclear facilities had introduced the international system to weaponized conflict in cyberspace.241 Cyber was now a tool of military power.

Would the advent of cyber as an offensive weapon of military power significantly affect the offensive or defensive design of campaigns? The analysis has already shown a

239 Kaplan, Dark Territory: The Secret History of Cyber War; Jason Healey, A Fierce Domain: Conflict in Cyberspace, 1986 to 2012; Richard A Clarke and Robert K. Knake. Cyber war. 240 Presidential Decision Directive 63 on Critical Infrastructure Protection, August 5, 1998: https://www.gpo.gov/fdsys/granule/FR-1998-08-05/98-20865. 241 David E. Sanger, “Obama Order Sped up Wave of Cyberattacks,” New York Times, June 1, 2012: http://www.nytimes.com/2012/06/01/world/middleeast/obama-ordered-wave-of-cyberattacks-against- iran.html; Kaplan, Dark Territory: The Secret History of Cyber War; Jason Healey, A Fierce Domain: Conflict in Cyberspace, 1986 to 2012; Richard A Clarke and Robert K. Knake. Cyber war.

107

strong perception of offensive advantage within U.S. doctrine—especially with technologies that utilize computer processing advances to conduct long-range offensive operations. Did the addition of cyber weapons impact the already heavily offensively biased doctrine? Was there a perception of offensive dominance within cyberspace?

And would that perception of offensive dominance within cyberspace have changed the overall ODB for the JOAC or ASB?

First, it is clear from ASB and, to an even greater extent, JOAC, that advancements in cyberspace played a large role in the creation of the doctrine. JOAC identifies technological advances within cyberspace as one of three primary motivators for the doctrine, asserting that “the third trend is the emergence of space and cyberspace as increasingly important and contested domains with critical importance for the projection of military force. Arguably, this emergence is the most important and fundamental change in the opposed access challenge over the past several decades.”242

To put in perspective, in the same time frame as ASB and JOAC were being developed, national security leaders were testifying to Congress and among foreign policy outlets that cyber was the next big threat to U.S. power. Secretary of Defense Panetta warned in

2012 of a “cyber Pearl Harbor,”243 and only a year prior Admiral Mullen warned congress that cyber was an existential threat to the United States.244 Col Jordan Thomas, one of the heads of the Air Sea Battle Office, recalls that during the time of the

ASB/JOAC doctrinal development, there was a keen sense that the kill chains of the U.S.

242 Department of Defense, “Joint Operational Access Concept,” 2012: pg. 12. 243 Elisabeth Bumiller and Thom Shanker, “Panetta Warns of Dire Threat of Cyberattack on U.S.,” New York Times, October 11, 2012: http://www.nytimes.com/2012/10/12/world/panetta-warns-of-dire-threat-of- cyberattack.html. 244 Vago Muradian, “Adm. Michael McMullen,” DefenseNews, June 10, 2011, http://www.defensenews.com/article/20110710/DEFFEAT03/107100301/Adm-Michael-Mullen.

108

military were significantly threatened by offensive cyber operations.245 But did this sense of vulnerability translate to offensive doctrine?

The doctrine and concept reveals a tension between the sense of cyber vulnerability on the part of the United States and an uncertainty about the ability (or will) to conduct similar offensive operations against an A2/AD adversary. While ASB discusses the ability to use cyber as an offensive weapon for attacks in depth,246 both

JOAC and ASB appear ambivalent or undecided about whether the attacker or the defender has the advantage in cyberspace. As JOAC explains about cyberspace technology, “advancements apply not only to the attacker but also to the defender, who can exploit those capabilities to engage the enemy at greater ranges.”247 To add to this equanimity about ODB in cyberspace, a look at the verbs used within both ASB and

JOAC reveals that the greatest use of defensive verbs is in conjunction with cyber (C3 nodes).248 Perhaps not coincidentally, in the precept that discusses cyber operations, the verb “protect” is used before “attack.” It is the only precept in which defense leads offense in the precept.

This disconnect between defensive use of cyber and offensive use of conventional digitally-enabled capabilities is probably due to a series of factors. First, institutionally,

Col Thomas points out that, “lack of authorities to use offensive cyber at the operational level of war” was a major influencer on the defensive focus of cyber operations within

ASB and JOAC. He explains, “instead of the JTF/CC or the JFACC having the authority

245 E-mail with Col Jordan Thomas, Ret., October 18, 2016. 246 “Air Sea Battle Concept Implementation.” 247 Department of Defense, “Joint Operational Access Concept,” 2012: pg. 7. 248 The only other aspect of the plans in which we see a defensive discussion is in the disposition and survivability of airbases.

109

to use offensive cyber, the authorities were still held at the Title 50 or the

STRATCOM/CC level . . . the authority to use cyber against an adversary in the operational level were not defined, therefore this led to a focus on defensive cyber operations.”249 Secondly, the United States, and particularly the military, is more networked and dependent on cyber dominance than other nations, making them also more vulnerable to attack.250 This is a common theme that is discussed in the JOAC, in which the writers recognize a difficult tension between the capabilities of information-enabled warfare and the vulnerabilities this entails to computer network attack.

These dynamics suggest that defense-dominant cyber operations may have influenced the ASB/JOAC doctrine to find offensive capabilities in other domains. As

JOAC explains, “gaining space and cyberspace superiority when and where needed is not necessarily a symmetrical effort—that is, cyberspace operations to gain cyberspace superiority . . . but often can be achieved more effectively, like superiority in other domains, through the cross-domain application of combat power.”251 Col. Thomas further elaborates, “The fact DoD, USAF/USN, and Intel Community were being contested in cyber had a lot to do with describing the operational art of ASB operation which was "Cross-Domain Operations" . . . Just as AirLand Battle shifted the paradigm from fighting at the Fulda Gap to taking the fight to the enemy's rear echelons, the ASB

249 E-mail with Col Jordan Thomas, Ret., October 18, 2016. Some of these concerns about authority, especially within conventional campaigns, have since been institutionally resolved (See Department of Defense Cyberspace Strategy, April 2015: http://www.defense.gov/Portals/1/features/2015/0415_cyber- strategy/Final_2015_DoD_CYBER_STRATEGY_for_web.pdf). As such, it will be interesting to see whether the new evolution of ASB (JAM-GC) devotes greater attention to U.S. use of offensive cyber operations. 250 Jacquelyn Schneider, “Digitally-Enabled Warfare: The Capability-Vulnerability Paradox,” Center for a New American Security, August 29, 2016: https://www.cnas.org/publications/reports/digitally-enabled- warfare-the-capability-vulnerability-paradox. 251 Department of Defense, “Joint Operational Access Concept,” 2012: pg. 27.

110

and JOAC concepts were a recognition that the US did not solely have freedom of action in the cyber domains.”252 Therefore, cyber operations within doctrine may generally defense dominant but create incentives for offensive campaigns enabled by the computer- driven technologies first generated by AirLand Battle in order to create American offensive advantage.

This analysis has focused almost exclusively on the role that technology played in the offensive-dominant doctrine of the U.S. military under Air Sea Battle and the Joint

Operational Access Concept. As alluded to in the previous analysis of AirLand Battle, decisions about offense dominance made in the large-spending inputs of the Reagan

Administration appear to have cemented a perception of technological offense dominance for U.S. warplanners. In both the professional military publications and most of the foreign policy analysis, there is almost no debate about the dominance of offense vs. defense in both JOAC and ASB. Further, the resounding success of U.S. offensive campaigns in Iraq and to some extent Libya and Bosnia has been tied to the dominance of offensive technology like long-range strike, distributed information distribution, and persistent long-range surveillance.

Conclusion

Did the information revolution technology drive the development of the highly offensive characteristics U.S. AirLand Battle and Air Sea Battle? The answer to the question is less binary than a simple division between technology and politics. For

AirLand Battle, offensive technology was the output of a doctrine inspired by an

252 E-mail with Col Jordan Thomas, Ret., October 18, 2016.

111

organizational and civil-military incentive for offensive advantage. When Active

Defense was published in 1976, rank and file Army was immensely dissatisfied by what they saw as a doctrine of defeat. Much of the pressure to revise and re-write that 1976 doctrine stemmed from a perception that the Army needed to be a service of victory.

And this desire for offense was hugely successful coming out of the militarily lackluster

Carter administration and moving into a more aggressive and idealistic Reagan administration. In the case of the move from Active Defense, which was clearly driven by technology, to AirLand Battle, technology followed non-material factors.

But here’s where the division between politics and technology becomes murky.

Because AirLand Battle was tied to the Corps 86 acquisition program that emerged in the heyday of Reagan spending, the decision to utilize advances in computing for offensive technologies (versus investing in the defensive operations advocated by the 1976 Active

Defense) became a critical juncture for offensive U.S. doctrine. Indeed, the technology that cascaded from the acquisition programs of AirLand Battle would create a path dependency towards the heavy tendency for offensive operations in Air Sea Battle. As

Long foreshadowed in the late 1980s, “The technologies that will be fielded in the nineties are the long-awaited capabilities programmed in Corps 86, that inspired the 1982 edition of FM 100-5. The 1982 edition designed the deep battle concept so that it could incorporate then current and Corps 86 technologies.”253

There is no shift from offense to defense after AirLand Battle. The United States, clearly a hegemon in the early 90s, could have chosen to take the information revolution and invest in defensive technologies. Instead, programs like missile defense or upgrades

253 Long, The Evolution of Army Doctrine: Active Defense to Airland Battle and Beyond, pg. 136.

112

to integrated air defense systems remained relatively stagnant while the vast majority of technological advances favored long range strikes, target acquisition, and C3I for maneuver and attack. The cult of the offensive that ultimately drove the offensive-nature of AirLand Battle ended up creating a path dependency for information revolution technologies. In turn, these technologies were so incredibly offensively dominant that they necessitated subsequent offensive doctrine.

By the 2012 ASB/JOAC doctrines, technology was the driving force in determinants of offensive and defensive campaigns. Computing and processing power became such a tool for the offense by the time ASB/JOAC came to fore that U.S. planners grew even more confident in the overwhelming advantage of offensive operations. In the thirty year period between AirLand Battle and ASB/JOAC almost no weapons innovations occurred outside of computer processing. Kinematics and lethality remained relatively stagnant254 while computer processing power revolutionized the ability for an array of sensors to find, fix, and target in near real time. Missiles were able to fire at longer distances not because the missile was more aerodynamic or faster, but because computer processing allowed adversaries to see each other from further and further away. Computing enabled AirLand Battle’s offensive doctrines and then locked in the offensive nature of future doctrines because of the dependency on computer-driven offensive weapons. Politics created the technology, but the technology became so powerful that it then necessitated offensive doctrines.

But the information revolution is far from complete and the computer enabled offensive technologies created by the doctrines of the early information revolution are

254 An exception to this could be the advent of stealth technology, though the ability to engineer stealth machines was entirely contingent on advances in computer processing power.

113

now under threat by advances in cyber operations. By 2012, the doctrine creators of

ASB/JOAC clearly saw an adversarial offensive advantage in cyberspace. Worried about the computer-dependent operations of U.S. campaigns and institutional limitations to U.S cyber operations, the ASB/JOAC crafters envisioned a doctrine that leveraged the offensive advantage of conventional digital technologies to offset defensive operations within cyberspace.

With these burgeoning cyber technologies and potentially large shifts in the role of information technologies in warfare, can information revolution technologies influence the development of overall defensive doctrines? Or are the path dependencies in offensive acquisition strategies so strong that offense will continue to dominate doctrine?

ASB and JOAC’s concern about contested cyber dominance and reliance on computing power for successful operations indicates that cyber may be able to influence the development of defensive doctrines. As states move closer and closer to dependence on

C3I capabilities to conduct offensive operations, an offensive advantage in cyber attack could influence more attrition-based doctrine. Like Active Defense of 1976, future doctrinaires may perceive that in order to win campaigns, states will have to be prepared to survive after the first devastating cyber on cyber blows. This could lead to doctrines that focus on resilience, protection, and depth instead of speed, maneuver, and attack.

What does this mean for systemic ODB in a cybered world? The disconnect between U.S. perceptions of adversary cyber offensive dominance and perceived U.S. vulnerabilities may be idiosyncratic to the United States. In addition, the acquisition strategies that locked in U.S. development of computer-enabled offensive technologies are certainly unique to the U.S. However, the United States’ role as a norm entrepreneur

114

in conflict may lead states to adopt similar strategies for different reasons. Further research should be conducted on how similar doctrine in states like China and Russia perceive cyber ODB and how that affects conventional operations. If these perceptions do prove to be systemic than cyber—both as computing power and as a weapon—may influence ODB towards the offensive and increase the chance that states in crises end up in inadvertent conflict.

What does this mean for U.S. policy? The follow-on concept to Air Sea Battle,

JAM-GC, as well as the follow on doctrine to AirLand Battle (MultiDomain Battle) is currently under review. Doctrine writers should challenge themselves to envision what defensive-dominant campaigns and technologies could look like in an increasingly cyber- contested world. Is offensive dominance in campaigns a determined outcome? Or will the rise of cyber network attacks re-invigorate the potential for defense?

115

Chapter 3.

Cyber and Crisis Escalation: Insights from Wargaming

Does the advent of cyber operations make the international system more unstable?

The overwhelming scholarly consensus is no,255 but practitioners paint a far more dangerous picture. Time and time again, U.S. policy makers have testified about the de- stabilizing nature of cyber, warning of “a cyber Pearl Harbor,”256 “the single biggest existential threat,”257 and “the no.1 threat facing the nation.”258 Neither the scholars nor the practitioners have empirical precedent for their assumptions. Indeed, many of the characteristics of cyberspace—the rapid speed of actions, lack of geographical buffers,

255 Jerry Brito and Tate Watkins, “Loving the Cyber Bomb? The Dangers of Threat Inflation in Cybersecurity Policy,” Harvard National Security Journal, April 10, 2012: http://mercatus.org/publication/loving-cyber-bomb-dangers-threat-inflation-cybersecurity-policy-0; Stephen Walt, “Is the cyber threat overblown?” Foreignpolicy.com March 30, 2010, http://foreignpolicy.com/2010/03/30/is-the-cyber-threat-overblown/; Erik Gartzke, "The myth of cyberwar: bringing war in cyberspace back down to Earth." International Security 38, no. 2 (2013): 41-73; Jon R. Lindsay, "Stuxnet and the limits of cyber warfare." Security Studies 22.3 (2013): 365-404; Jon R. Lindsay, "The Impact of China on Cybersecurity: Fiction and Friction." International Security 39, no. 3 (2015): 7- 47; Thomas Rid, "Cyber war will not take place." Journal of Strategic Studies 35, no. 1 (2012): 5-32. 256 Leon Panetta, “Remarks by Secretary Panetta on Cybersecurity to the Business Executives for National Security, New York City,” October 11, 2012, jklhttp://www.defense.gov/transcripts/transcript.aspx?transcriptid=5136. 257 Micah Zenko, “Admiral McMullen: Thank You and Farewell,” Council on Foreign Relations, September 29, 2011: http://blogs.cfr.org/zenko/2011/09/29/admiral-michael-mullen-farewell-and-thank- you// 258 Aaron Boyd, “DNI Clapper: Cyber bigger threat than terrorism,” Federaltimes, February 4, 2016: http://www.federaltimes.com/story/government/cybersecurity/2016/02/04/cyber-biggerthreat- terrorism/79816482/.

116

proliferation of actors, dependency on digital capabilities, and overall extreme uncertainty—suggest that policymakers are correct to be wary of the escalatory effects of cyberspace operations.

Interestingly, despite the cries of danger from U.S. policymakers, there is very little evidence linking cyber attacks—whether it be Stuxnet, Sony, or Russian attacks in

Ukraine—to crisis escalation. In fact, what has been remarkably consistent is how non- escalatory responses to cyber attacks have been. In the Stuxnet case, there was no discernable escalation against the U.S. or Israel from Iran.259 In the case of Sony, the

Obama administration framed their response in terms of “proportionality” and developed an economic sanctions tool to respond to similar cyber attacks. Finally, the Russian cyber attacks on infrastructure within Ukraine seem largely removed from the conventional dimensions of the crises and have not made a noticeable impact on either horizontal or vertical escalation of the low-intensity conflict.

What could explain this seemingly dissonance between the danger of cyberspace operations and the limited precedent for danger and cyber attacks? The body of work on cyber and stability has so far articulated a series of competing hypotheses about the impact of cyber operations on conflict initiation as well as a series of unsatisfying analogies with little ability to test any the validity of most of the assertions.260 Empirical

259 There is potentially a link between the 2012 Saudi Aramco cyber attack and the Stuxnet attack on Iranian nuclear capabilities. However, it is important to note that—despite the extensive Aramco attack —Saudi Arabia did not escalate to significant conventional reprisal attacks on Iran (Bronk and Tikk-Ringas 2013). 260 Emily Goldman and John Arquilla, Cyber Analogies. No. NPS-DA-14-001. NAVAL POSTGRADUATE SCHOOL MONTEREY CA, 2014; Martin C. Libicki, Crisis and Escalation in Cyberspace (Santa Monica: Rand Corporation, 2012); Herbert Lin. "Escalation Dynamics and Conflict Termination in Cyberspace." Air University (2012).

117

analysis and hypothesis testing about “cyber”261 and stability are difficult for a series of reasons: the virtual nature of the domain, the technical difficulty to understand balance of capabilities, the covert nature of cyber operations, as well as the general infancy of its use. In some ways, these difficulties may be hurdles that are impossible to overcome, not only for researchers but also for foreign-policy decision makers faced with extreme uncertainty when making decisions about how to respond and use cyber operations in a crisis.262 The interesting question, therefore, may not be whether scholars or practitioners are right about cyber and stability, but instead how states perceive the impact of cyber operations. We can, therefore, make significant strides toward our understanding of the impact of cyber on crisis stability by shifting from an analysis of capabilities to an exploration of states’ perceptions about the impact of cyber operations on escalation. By bypassing technical questions of capabilities, we can focus instead on how decision- makers process the uncertainties of cyber operations, with implications not only for potential behaviors during crisis situations but also for understanding the variables that shape foreign policy decision-makers’ understandings of the cyber domain.

The focus on perceptions vice capabilities is not entirely novel; the study of nuclear escalation was also plagued by a lack of empirical precedent. Similarly, research conducted during the Cold War focused not only on the pure capability balance between the major powers, but also on how those states perceived their capabilities and how institutional, cognitive, and strategic variables influenced nuclear decision-making.263 It

261 In this work, when I use cyber as a noun, I am doing it in an intentional way in order to capture an overall characteristic of the digital, networked, or computer-based set of technologies, capabilities, and operations. 262 Slayton, "What Is the Cyber Offense-Defense Balance? Conceptions, Causes, and Assessment." 263 Robert Jervis, "Deterrence and perception." International security 7, no. 3 (1982): 3-30; Robert Jervis, Perception and misperception in international politics (Princeton: Princeton University Press, 1979);

118

was during this time that wargaming emerged as a prominent mechanism to understand nuclear dynamics. From Schelling to Robert Mandel and games played not only at the

Pentagon but also at Yale and MIT, wargaming provided much-needed insights into the motivations and perceptions of nuclear decision-making.264

Following the precedent set by those who explored wargaming to explain nuclear decision-making, in this article I examine data from a crisis wargame conducted at the

U.S. Naval War College from 2011 to 2016. By conducting the analysis over six years, I am able to explore the evolution of the use of cyber and perceptions about the escalatory nature of cyber’s use while examining how changes in context, capabilities, and sample influence those perceptions. My research shows that these decision-makers view cyber operations as highly escalatory and are therefore cautious about using offensive and cyber operations and cyber network exploitation, even after conventional conflict has begun.

Additionally, I find that these decision-makers—despite their concern about escalation— never once responded to cyber attacks by the adversary. These findings suggest that U.S. policymakers may believe cyber is escalatory, but that perception may induce risk- adverse cyber policies that keep cyber from influencing escalation dynamics.

The exploration proceeds as follows—in the first section I explore existing cyber and conflict escalation literature. Next I introduce the method and explain the benefits and limitations of the wargaming data to explore cyber and crisis escalation. I then

Robert Jervis, Robert, Richard Ned Lebow, and Janice Gross Stein. Psychology and deterrence (Baltimore: Johns Hopkins University Press, 1989); Graham Allison and Phillip Zeilkow. Essence of Decision (New York: Longman, 1999). 264 Robert Mandel, "Political gaming and foreign policy making during crises." World Politics 29.04 (1977): 610-625; Robert Mandel, Garry D. Brewer, Martin Shubik, Alfred H. Hausrath, and Stuart A. Bremer. "Policy-making perspectives on war simulations." (1980): 359-375; Thomas C. Schelling, Arms and Influence: With a New Preface and Afterword. (New Haven: Yale University Press, 2008).

119

generate a series of hypotheses about cyber and escalation and introduce the data and discuss findings. Finally, I provide conclusions and implications for our understandings of escalation and state policies on the use of cyber operations.

Cyber and Crisis Escalation

From rifles to tanks, to aircraft and nuclear weapons, there is an expansive set of empirical and theoretical analyses that explore the impact of technologies on crisis escalation.265 These examinations of conflict and technological development present a knot of conflicting explanations for when and why crises escalate to conventional or nuclear war. From the structural and material,266 norms of behavior,267 behavioral perceptions,268 and organizational dynamics269—the diverse variables impacting crisis

265 Richard Smoke. War: Controlling Escalation (Cambridge: Harvard University Press, 1977); Barry Posen, “Crisis Stability and Conventional Arms Control,” Daedalus 120, no.1 (1991): 217-232; Barry Posen, Inadvertent Escalation: Conventional War and Nuclear Risks (Ithaca: Cornell University Press, 1991); Robert Powell, “Crisis Stability in the Nuclear Age.” American Political Science Review 83, no.1 (1989): 61-76; Herman Kahn, On Escalation: Metaphors and Scenarios (New Brunswick: Transaction Publishers, 2012); Caitlin Talmadge, “Assessing the Risk of Chinese Nuclear Escalation in a Conventional War with the United States,” International Security, Forthcoming; Barry Nalebluff, Brinkmanship and nuclear deterrence: the neutrality of escalation (Princeton: Princeton University Press, 1987). 266 J.D. Fearon, “Domestic political audiences and the escalation of international disputes.” American Political Science Review (1994): 577-592; J.D. Fearon, “Rationalist explanations for war.” International Organization, 49 (1995): 379-399; John Mearsheimer, The tragedy of great power politics. (New York: WW Norton & Company, 2001); Stephen Van Evera, Causes of war: Power and the roots of conflict (Ithaca: Cornell University Press, 1999); Robert Jervis, “Cooperation under the security dilemma,” World Politics Vol 30, no.2 (1978): 167-214. 267 Martha Finnemore, The purpose of intervention: changing beliefs about the use of force (Ithaca: Cornell University Press, 2004); C.Gelpi, The power of legitimacy: Assessing the role of norms in crisis bargaining (Princeton: Princeton University Press, 2003); I. Hurd, “Breaking and making norms: American revisionism and crises of legitimacy,” International Politics, vol. 44, no.2 (2007): 194-213. 268 Robert Jervis, Perception and Misperception in International Politics (Princeton: Princeton University Press, 1976); Richard Lebow, Between Peace and War (Baltimore: Johns Hopkins University Press, 1984). 269 Jeffrey Legro, “Military culture and inadvertent escalation in World War II,” International Security (1994): 108-142; Barry Posen, The sources of military doctrine: France, Britain, and Germany between the world wars (Ithaca: Cornell University Press, 1986); Jack Snyder, “Civil-Military Relations and the Cult of the Offensive, 1914 and 1984,” International Security (1984): 108-146; Graham Allison and P. Zelikow, Essence of decision: Explaining the Cuban missile crisis (Vol. 2) (New York: Longman, 1999).

120

escalation are often so intertwined that it is difficult to generalize the effects revolutionary technologies have (if any) on the potential for conflict erupting from crises.

Perhaps because of the extraordinary complexity and breadth of literature on crisis stability, cyber creates a particularly puzzling case in which conflicting explanations predict starkly different effects of cyber on crises.270 For some, the complexity and interdependence of cyber operations and technology decreases the probability for conflict eruption.271 For these scholars, the technical difficulty of creating physical effects through cyberspace operations make the impact of cyber on crisis stability negligible if not inclined towards peace. In addition, the civilian nature of digital capabilities and the ways in which these capabilities undergird national economies make significant cyber attacks in crises unlikely to occur. These scholars argue that states will restrain their use of significant cyber attacks and therefore mitigate the potential for inadvertent escalation through the use of cyber operations.

However, for others the extreme uncertainty, speed of evolving capabilities, and perception of offense dominance increases the potential that cyber will induce conflict.272

For these scholars, the ubiquitous nature of the digital economy and digitized nature of conventional warfare create de-stabilizing incentives for cyber first strikes that could

270 Libicki, Crisis and Escalation in Cyberspace (Santa Monica: Rand Corporation, 2012); Lin, "Escalation Dynamics and Conflict Termination in Cyberspace”; John Stone, "Cyber War Will Take Place!." Journal of Strategic Studies 36, no. 1 (2013): 101-108; Gartzke, "The myth of cyberwar: bringing war in cyberspace back down to Earth”; Rid, "Cyber war will not take place." 271 Lindsay, "Stuxnet and the limits of cyber warfare." Lindsay, "The Impact of China on Cybersecurity: Fiction and Friction." 272 David Gompert and Martin Libicki, “Cyber Warfare and Sino-American Instability,” Survival 56, no.4 (2014): 7-22; Adam Liff, “Cyberwar: A New ‘Absolute Weapon’? The Proliferation of Cyberwarfare Capabilities and Interstate War,” Journal of Strategic Studies 35, no.3 (2012).

121

escalate to armed conflict.273 Fundamentally, these arguments posit that the extreme uncertainty of cyber effects and dependencies increases the risk for inadvertent escalation due to cyber operation.

Methodology

In order to examine these conflicting hypotheses about cyber and crisis stability, I use data from a U.S. Department of Defense wargame. The game, called the Deterrence and Escalation Game in Review (DEGRE), is conducted annually with a sample of U.S. foreign policy decision-makers at the United States Naval War College. It is sponsored by U.S. Strategic Command and takes place over four days. For this research, I have drawn data from wargames conducted from 2011 to 2016. The use of wargaming in social science is experiencing a period of relative resurgence (this was a somewhat common practice in the early part of the twentieth century) and so a discussion about the merits and limitations of wargaming for international relations research is timely and important.

What is a wargame? Peter Perla, a lifetime wargamer at the Center for Naval

Analysis and a trained social scientist presents the most accepted definition which is that,

“a wargame is a warfare model or simulation whose operation does not involve the activities of actual military forces, and whose sequence of events affects and is, in turn, affected by the decisions made by players representing the opposing sides. In the end, a

273 Jacquelyn Schneider, “Digitally Enabled Warfare: the Capability-Vulnerability Paradox,” Center for a New American Security, August 29, 2016: https://www.cnas.org/publications/reports/digitally-enabled- warfare-the-capability-vulnerability-paradox.

122

wargame is an exercise in human interaction . . . its forte is the exploration of the role and the potential effects of human decisions.”274 Wargames, as opposed to simulations or probabilistic models of operational research, examine the processes of warfare and do not present quantitative analyses of military effectiveness. As Perla explains, “wargames are tools for gaining insights into the dynamics of warfare. They can help players come to a more complete understanding of the sources and motivations underlying the decisions made . . . wargames are best used to investigate processes.”275

Wargames are explorations of human behavior when placed in a notional crisis or war scenario. The focus on human behavior makes the analogy of wargames and experimental research intuitive. Wargaming, however, is not an experiment, though it does have some experimental qualities. Like an experiment, wargames are tools to examine behaviors. They are also designed to test behavior when placed in certain constraints. However, experiments are designed specifically to deduce a causal effect.

As such, special emphasis is placed in experimental design on controlling for confounding variables. Subjects of experiments are carefully randomized and alternatively provided with either control or treatment scenarios. Unlike experiments, wargames are usually one-off iterations (the wargame in this research is an exception to this rule). Because wargames are generally designed to exercise decision-making in complexity, they are rarely built to control for a specific variable. Finally, samples are generally not randomized and also are not divided into both a control and treatment

274 Peter Perla, The Art of Wargaming (Annapolis: Naval Institute Press, 1990, pg. 164). 275 Ibid, pg. 9.

123

group. These differences between wargaming and experiments have important implications for social science.

Most importantly, the non-experimental design of most wargames creates significant limitations both for determining causal effects and generalizing inferences beyond the single wargame iteration. This is for a number reasons. First, wargames are generally conducted with a small sample of individuals and are not usually repeatable or reproducible. For instance, they are usually played with specific countries in mind and therefore are also highly contingent on contextual biases. Secondly, wargames tradeoff between simplicity and accuracy. Simplicity helps wargame designers better understand the variables that influence players’ decisions and wargame outcomes. However, with simplicity comes a loss of accuracy, particularly when playing the very complex game of war. Not only are players less invested in the outcomes of the games when there is less accuracy, but also the outcomes and behaviors are less indicative of a true decision in war.

Despite these limitations, wargaming provides a host of benefits to international researchers. And, in fact, many of both the limitations and the benefits of wargaming are similar to field experimentation in which scientists trade control and randomization for realism in both the environment and the participants.276 Perhaps the most obvious benefit of wargaming is the sample that is generally used to play the games. One of the fundamental problems with using experiments in international relations is the difficulty in generating an appropriate sample that may replicate the kind of decision-making that

276 Alex Mintz, Yi Yang, and Rose McDermott. "Experimental approaches to international relations." International Studies Quarterly 55, no. 2 (2011): 493-501.

124

occurs between nation states.277 Critics have rightly pointed at the inability for undergraduates, the primary sample of experiments in international relations, to simulate an experienced foreign policy decision-maker. Additionally, experiments conducted in a lab have difficulty reproducing the stress and high stakes of foreign policy decisions.

Wargames provide at least a partial solution for both those problems. Frances McHugh, former Department Chair of the U.S. Naval War College gaming department notes, “by explicitly allowing human decisions that are made under the press of time and on the basis of imperfect or incomplete information to influence the course of events, and by incorporating the capricious effects of randomness and ‘luck,’ wargaming comes closer than any other form of intellectual exercise to illuminating the dynamics of warfare . . . the ‘unquantifiable’ factors.”278

By using the same decision-makers that are vested with decisions in actual crises and conflicts, the behaviors of wargame participants are inherently more valid than student-generated experiments. Also, unlike undergraduates who are asked to play games for monetary compensation, wargame participants are generally significantly invested in the outcomes of the games (whether for institutional, personal, or professional reasons) and are placed under similar decision-making constraints as actual foreign policy decision-makers. As Perla and McGrady argue, “games give players active responsibility for their decisions, similar to what they would experience in the real world,

277 Alex Mintz, Steven B. Redd, and Arnold Vedlitz. "Can we generalize from student experiments to the real world in political science, military affairs, and international relations?." Journal of Conflict Resolution 50, no. 5 (2006): 757-776. 278 Francis McHugh. Fundamentals of Wargaming (Newport: United States Naval War College, 1966, pg. 9).

125

and force them to bear many of the same consequences of those decisions, both positive and negative.”279

Finally, examining wargames played over time may provide insight into patterns of behaviors that have generalizable implications for conflict onset or outcomes.

Schelling, for example, famously drew much of the insight for his works on conflict dynamics from his experience with wargaming. For him, playing wargames over time and across contexts revealed consistent biases and interactions that he then validated through real world examples. As he writes, “finding something in a game does not necessarily mean that the same thing happens in real life, but at least one know what to look for in real life, and knowing what to look for, one may find it” (Schelling 1987,

441).280 It was just this kind of analytic process and iterative wargaming that made the interwar wargames conducted at the Naval War College so useful to predicting the integral role of the aircraft carrier in a future fight against the Japanese.

For this research, I use the Naval War College’s Deterrence and Escalation Game and Review (DEGRE) to examine cyber and crisis stability. DEGRE is conducted once a year and is designed to examine the “strategic impact of nuclear weapons proliferation, deterrence, employment, and escalation and its implications on U.S. plans, policy, and force structure.” Its players come from U.S. think tanks, government departments, and academia and include both current and former high-level policy makers. The vast majority of players (approximately 25-30) are placed on a “blue” team in which they play the role of U.S. government positions from Commander in Chief to Department of

279 Peter P. Perla and E. D. McGrady. "Why wargaming works." Naval War College Review 64, no. 3 (2011): 113. 280 Thomas C. Schelling, “Designing War Games.” In Ashton B. Carter, John D. Steinbruner, Charles A. Zraket, and Sherman Frankel, Managing nuclear operations (1988): 94-96.

126

Treasury, Secretary of State, Secretary of Defense, and subordinate combatant commands. A secondary set of players (approximately 15-20) with subject matter expertise is placed on the “red” team to simulate a notional adversary. The game takes place over a week, starting the players in a crisis scenario and then letting the play evolve with decisions made by each team in six rounds of play. The wargame is considered a free-form game in which umpires serve only to arbitrate the net effects of decisions made between red and blue teams and not to induce an effect.

DEGRE is a particularly useful wargame for studying cyber and crisis stability because of some unique qualities of the game. First, DEGRE is a crisis game in which cyber may be played as a tool but is not the subject of the game. Therefore, the wargame shows behaviors towards cyber embedded within larger conventional military operations and other whole of government tools (diplomacy, sanctions, trade, etc.). The integration of cyber within this larger scenario is a more accurate gauge of decision-makers’ perceptions of cyber than a cyber-only game because it introduces trade-offs between the usage of cyber and other conventional operations. This enables the analysis of the interaction between the use of cyber and subsequent or preceding uses of conventional or non-military options. Secondly, DEGRE does not dictate the play of the game. Instead, arbitration (i.e. the umpires of the game) serves only to adjudicate the result of the interaction of each side’s move and not to decide outcomes. Therefore, the outcomes of the game are tied more closely to the behaviors of the players than the design of the game, which makes DEGRE less likely to introduce systematic bias through game design. Additionally, DEGRE uses an elite sample for its game from top current and former government officials. And finally, DEGRE has been played for six years using

127

the same play format. Even though the notional crisis scenarios as well as the players vary in many of the years, the iterative play of the game allows for longitudinal analysis, increases the sample size, and can help control for scenario-based intervening variables.

Further, in all six years of analysis the same individual played the cyber subject matter expert; this individual’s play was remarkably consistent throughout the games and therefore provides a useful control throughout the wargames.281

There are two large limitations to using DEGRE for this analysis. First, with the exception of two years that ran the same scenarios, the scenarios change each year.

Therefore, significant effort must be devoted in the analysis of the data to determine what role scenario variables play in any changes over the years. Perhaps a larger limitation is that the game is played at a Top Secret classification level. Many of the details about the scenarios and the players are masked, which impact the transparency of the research method. I have tried to work around this by providing generalizable characteristics, vice specific scenarios or capabilities.

In terms of sample, the institutional affiliation of most of the players has been masked. However, in general, the allocation of blue team players is dominated by the

Department of Defense or Defense-related think tanks (approximately 12-15),282 with one to two players from the State Department, one to three players from the National

Intelligence Community, two to three players from national laboratories, and one player

281 In all of the games, the cyber player advocated an offensively-focused use of cyber operations and consistently advocated for the early use of these offensive cyber operations. 282 Defense players represented all of the functional commands (Strategic Command, Cyber Command, Space Command, Transportation Command) and relevant component Commands (for example either Pacific Command, Northern Command, European Command, etc.). Because of Strategic Command’s involvement, there is a slightly greater proportional representation from Strategic Command than might be expected in a real world scenario.

128

from Treasury in the 2015 and 2016 wargames. The only significant aberration in this player make-up was the inclusion of Treasury in the last two wargames. As might be expected, these games also saw the greatest use of economic sanctions.

In the data section below, I detail when contextual variables were similar and when they varied as well as significant changes in play structure, sample, and blue team lead. The variance of these variables, while affecting the ability to generalize patterns over time, also provides a mechanism to better understand motivating factors that determine cyber behavior. If cyber play vis-à-vis escalation changes with the context, the sample or the blue team lead, then that provides insight into how balance of power, personality traits, and organizational biases may impact U.S. use of cyber in a crisis situation. However, if we see consistencies in play despite changes in context, sample, etc., then that provides evidence for enduring behaviors that may transcend changes in administration, etc. It is important to note that I will not be analyzing red play for insights on cyber and crisis stability. While red team members are generally subject matter experts in an adversary, they are at best a representation of U.S. perception of the adversary and so cannot provide insights into true “red” perceptions about cyber and crisis stability.

Data from DEGRE is generated in a variety of ways. First, an end of game report is created that details major moves of the game, lessons learned, and outcomes.

Secondly, transcripts of player conversations are taken in both the red and blue cells during move discussions. These transcripts detail the decision-making process and capture conversations between players about the adversary, their potential courses of

129

action, and their decision-making about future actions.283 In addition, in two of the games, post-move surveys ask players specifically about perceptions of escalation. This survey data was analyzed in concert with the transcripts of conversations. Finally, a move summary provides the list of actions taken by both blue and red, at what point in the game the actions were taken, and to what effect.

Hypotheses

For the U.S. foreign-policy decision-makers that participate in DEGRE, the wargaming data allows me to examine a series of questions: do perceptions about the effects of cyber operations on crisis escalation impact decisions to utilize cyber operations at different phases of crises? At what point in crises do decision-makers choose to use cyber operations? What are their perceptions of the escalatory effects of cyber operations (especially vis-à-vis other potential actions)? Conversely, how does blue respond to red cyber operations? Does red cyber play lead to escalation on the part of blue? These questions lead to two categories of hypotheses: 1) “blue” hypotheses about how blue perceives itself and the actions it takes against red and 2) “red” hypotheses about how blue perceives red cyber actions against blue. Evidence for these hypotheses is derived both from the sequence of moves as well as transcripts and surveys about the decisions in the wargame.

“Blue” Hypotheses: Perceptions of the Effect of Blue’s Actions

283 These transcripts are generated by two to three “ethnographers” (military officers provided with some training on data acquisition methods) positioned on opposite sides of the room. Because these are not trained social scientists and because there are physical limitations towards what conversations these individuals are able to hear, there is some variance in the reporting. I have compiled the cyber-relevant data from the transcripts and cross-referenced between the ethnographer reports. Any potential inconsistencies were discarded.

130

The first set of hypotheses explores how blue believes its own use of cyber operations will impact the adversary’s escalation calculus. Evidence for these hypotheses are derived from discussions about action, and not necessarily from the move itself. For instance, if blue decided to conduct a cyber attack did they think it would lead to escalation? Were there conversations about not using particular cyber operations because of fears of escalation?

Hypothesis 1: Blue perceives blue cyber operations will lead to crisis escalation.

Hypothesis 2: Blue perceives blue cyber operations lead to crisis de-escalation.

Hypothesis 3: Blue perceives cyber operations have no effect on crisis escalation.

“Red” Hypotheses: Effect of Red’s Actions on Blue

The second set of hypotheses explores action—how did blue respond to the adversary’s cyber operations? Did the blue players take any actions specifically to respond to red cyber activity? What kind of actions and were there any conversations about the bleu team’s perceptions of escalation in these retaliatory strategies?

Hypothesis 4: Blue takes cyber action in response to red cyber operation.

Hypothesis 5: Blue takes conventional kinetic action in response to red cyber operation.

Hypothesis 6: Blue takes nuclear action in response to red cyber operation.

Hypothesis 7: Blue takes diplomatic or economic action in response to red cyber operation.

131

Hypothesis 8: Blues takes no action in response to red cyber operation.

These hypotheses examine how cyber is utilized in the game, but the games also provide insight into the motivation behind these cyber behaviors. In other words, cyber play within the first set of hypotheses act as a dependent variable on escalation and crisis stability. However, the variance of games over time provides insight into cyber operations as a dependent variable—what drives the cyber play of U.S. decision-makers in the DEGRE wargame?

This leads to a second set of hypotheses which test five “schools” of explanations for cyber behavior in crises: organizational, capability, situational context, individual personality, and cognitive. If cyber play varies according to the departmental distribution within the gaming sample (i.e. department of defense or geographic command players vs. department of state), then there may be evidence that organizational identities will be important drivers of U.S. cyber operations in crises. If, on the other hand, cyber operations within the game vary based on the evolution of capabilities over time, then the increased institutional capacity of U.S. cyber tools within its military would impact the future use of cyber operations. Alternatively, if cyber play varies based on the wargame’s context than that would provide evidence that cyber’s role in crisis stability will be based primarily on situational variables such as balance of power or geography.

Additionally, play that is dominated by the personality of the blue team lead would provide evidence of the importance of Presidential personalities in determining cyber operations in crises. And, finally, analysis of the text and patterns of behavior that occur across all these scenarios might suggest that there are patterns of cognitive or cultural biases that impact the use of cyber operations in crises.

132

Motivation Hypotheses:

Hypothesis 9: Organizational variables explain blue cyber operations.

Hypothesis 10: Capability variables explain blue cyber operations.

Hypothesis 11: Situational context variables explain blue cyber operations.

Hypothesis 12: Individual personality variables explain blue cyber operations.

Hypothesis 12: Cognitive variables explain blue cyber operations

Data

133

Table 6. Summary of Wargames and Cyber Activity

Below I summarize the cyber play from the DEGRE wargames conducted from

2011 to 2016. A few notes about the terminology are required. First, I will use cyber attack, cyber network attack, and offensive cyber operations interchangeably. Over the course of these wargames, the official Department of Defense terminology on cyber attack vs. offensive cyber operations changed. In this analysis, I understand them to be the same and will be specific about the type of cyber operations conducted if there could be a confusion between cyber network attack and offensive cyber operations. For

134

instance, I use the term cyber-led information operations as separate from “cyber attack.”

Additionally, in order to protect sensitive information, I have generalized all armed conflict with weapons in the air, space, naval, and land domains as conventional conflict and further lump all potential nuclear activity into nuclear conflict. Finally, I will generally refer to players and the teams as “blue” (sample of U.S. decision-makers),

“red” (adversaries), and “green” (allies). In these wargames, U.S. experts play the role of red and green. As such, I will not focus on their actions and instead on the reaction of blue to red and green activities.

Wargame 2011:

Wargame 2011 involved a land war scenario with a near-peer adversary. A female Department of State representative led the blue team. This was one of the first times that cyber was integrated in DEGRE and was the most rudimentary cyber play of the six wargames that I examined. Cyber was modeled very similarly to conventional capabilities. Unlike later games where there was extensive discussion about accesses, authorities, and current capabilities, wargames in 2011 and 2012 did not replicate institutional and capacity variables. One cyber token was generally equal to one conventional token. Without questions about capacity and institutional authorities, one would expect that cyber play would be very similar to conventional capabilities.

However, that was not how the blue team played cyber in the wargame. Instead, cyber capabilities were viewed as special and qualitatively different than their conventional counterparts. Throughout the game, players discussed the escalatory nature of cyber and these perceptions of escalation caused them to curtail almost all cyber

135

operations—to include network defense and network exploitation—until after conventional conflict had broken out and nuclear forces were placed on alert. At this point, once they no longer viewed cyber as more escalatory than other conventional or nuclear actions, the players became frustrated that the remaining cyber options could not substitute effectively for what was now deemed more escalatory and damaging physical attacks.

In post-move surveys, the players were explicit about how this fear of escalation from operations in cyberspace led to decisions to curtail cyber operations. In numerous survey responses and in discussions during the wargame, blue players repeatedly questioned whether the use of cyber—to include both cyber network exploitation and cyber attacks—would lead to a nuclear response by the adversary. This concern went beyond escalation from cyber attacks and included the use of cyber network exploitation to obtain accesses into adversary networks. As the blue team ethnographer noted, “Blue

Lead argued that the adversary would know about this [network exploitation] . . . many questions grew including deep discussions into exploitation. Rail lines could be affected and viewed as hostile vice de-escalating.” Debates about escalation continued after conventional conflict begun as the team discussed targeting strategic command and control with cyber attacks. Once again, concerns about escalation curtailed the use of cyber operations and all on the team decided that attacks against strategic command and control would necessitate full-scale nuclear war.

This discussion displays a few assumptions about cyber and escalation. First, despite the difficulty of attribution, players assumed that adversaries would have knowledge not only of blue’s cyber attacks, but also of the team’s cyber network

136

exploitation. Secondly, the players created an equivalency between cyber and nuclear attack—any cyber attack would necessarily lead to a nuclear response. The solution to these assumptions was not to use cyber operations until much later in the conflict—at which point, the capabilities were not able to create the same sort of kinetic effect that the players were able to produce with much more certainty from conventional capabilities.

In addition, only after conventional conflict was underway were cyber information operations authorized, but even then the concern for escalation limited the scope and tactics of these information operations. Cyber players had to demonstrate to the blue team lead that any information operations taken via cyber means would be reversible so that there would be no potential permanent damage against civilian networks and capabilities.

This concern about escalation appears to have played some part in the blue players’ response to red cyber attacks on conventional military capabilities. While the red team took offensive cyber operations prior to the blue team and before they launched conventional military strikes, the blue team did not view believe these red cyber attacks warranted a response—in the cyber of physical domains. This was partly because the blue team viewed a cyber tit for tat as potentially dangerous. As the defense lead player explained, “I did not feel any of the cyber attacks raised to the level where retaliation was needed and/or warranted! It was not risking nuclear war!” Note the cognitive dissonance in this explanation. The cyber attacks were not worrisome enough to warrant response.

This was an observation that was held throughout the game by most of the players. As one argued in discussion, “cyber-attacks although annoying do not appear crippling.”

Despite the fact that these attacks were considered annoying and below the level of

137

retaliation, the players were concerned that any response would necessarily go nuclear.

Therefore, the players were both deterred from responding because of a belief that any cyber act would go nuclear, but also didn’t need to be deterred because they didn’t feel cyber warranted any response.

Wargame 2012:

Wargame 2012 involved a naval war scenario with a near-peer adversary. A male former military official led the blue team. The design of cyber play within the game was more robust than in 2011 in that the type of targets and effects were more explicit.

However, as in 2011, cyber was generally not played with significant institutional or capacity limitations.

Wargame 2012 saw the largest cyber play than in any other DEGRE game before or since. It was also the most escalatory game that has been played in DEGRE in the recent history of the wargame. Cyber operations were a primary line of effort from move one and cyber attacks on missile command and control were authorized prior to conventional force employment. In fact, the use of conventional power was placed in reserve to be contingent on the success of cyber attacks to degrade the enemy’s ability to conventionally respond to U.S. operations. As the blue team lead directed, “don’t get the air expeditionary and tanking assets in place too early—until cyber effort has been effective.” Those initial cyber moves were not just designed as a combined arms operation to maximize military effectiveness. Instead, cyber attacks were envisioned as a signal to the adversary of U.S. will and capability, that “you don’t want to go to war with

138

us.” The blue team lead believed that by demonstrating cyber capability, the blue team would be able to convince the red team to de-escalate the crisis. The perception by the blue team was that blue was just as vulnerable (if not more) as the red team to early cyber attacks on command and control. Therefore, the blue team needed to take the initiative and conduct the first strike in order to ensure that blue was able to maintain its advantage in cyberspace while also cutting off the opponent’s ability to later mass and control conventional forces.

After pursuing these cyber attacks on the adversary’s military command and control, the blue leader then pursued horizontal cyber escalation and attacked red’s primary-civilian use cyber infrastructure in order to decrease the opponent’s economic ability to support war. Finally, as the crisis escalated to a naval blockade, the blue team lead advocated the use of cyber attacks against opponent nuclear weapons as well as concentrated cyber-led information operation campaigns. The greatest debate—in terms of escalation—about the use of cyber was actually in reference to the information operation campaign and the use of cyber to impact domestic populations.

In wargame 2012, red was unable to conduct any cyber attacks and so there was no response by blue to red operations. However, discussion by the blue team indicated that they would be willing to escalate cyber attacks and conventional attacks to red cyber attacks on blue command and control. In fact, blue considered their cyber vulnerabilities as so existential that they could not wait to respond to the cyber attack and therefore had to pre-emptively strike red’s ability to conduct both cyber and kinetic attacks against blue command and control.

139

Wargame 2013:

Wargame 2013 involved a naval war scenario with a near-peer adversary. A male former Department of State Representative led the blue team. The design of cyber play within the game included greater flexibility in target choice as well as greater fidelity of the cyber planning/targeting process than in the 2011 and 2012 games.

The 2013 wargame saw a large emphasis on deterring adversary cyber operations as well as the use of cyber operations as a signaling tool. Combined, these concerns about deterring adversaries while also signaling from cyberspace led to cautious use of offensive cyber operations and computer network exploitation. The focus on deterrence seemed to stem from the perception that the blue team was uniquely vulnerable both to attacks and to effects on its national economy based on these attacks. As the blue team lead commented at the onset of the game, “consider the economic impact of cyber ops to blue and global economy. Deterrence is key . . . be noisy in defense of homeland and computer network defense/computer network exploitation.” As with 2011, the players were more comfortable committing conventional forces and physical effects than they were with conducting cyber attacks and consequently offensive cyber was not played in the game until after sizeable conventional escalation, to include deployment of large forces to the area, air to air engagement, and a naval blockade. These cyber attacks that blue played were against military targets and were limited (by direction of the blue team lead) to reversible and virtual attacks.

140

The players started the game with an initial objective of deterring adversary cyber attacks through the use of declaratory deterrence policies. This focus on deterrence also led the players to question whether or not blue cyber network exploitation would create incentives for the adversary to conduct a preemptive cyber strike. Consequently, strict rules of engagement—to include no network exploitation of strategic command and control and limited military command and control—were placed on computer network exploitation with the assumption that these activities would be detected and would be interpreted as signals of the United States’ desire to escalate the crisis. At one point in the game, in the midst of a naval blockade, the blue team’s cyber operations were detected as it conducted exploitation of networks related to the adversary’s conventional maritime operations. Despite the fact that shots had already been fired between vessels, the gameplayers were very concerned that the detection of their cyber exploit (not attack) would inadvertently escalate the conflict. In that same move, 20 adversary aircraft were shot down. However, exit surveys suggested that the majority of the blue players believed that the discovery of the cyber exploit was potentially the most escalatory action that occurred in the move (the red team did not make any statement or conduct any operation in response to this cyber action).

This concern with escalation even after the exchange of conventional fire was especially prevalent in discussions about cyber attacks that might impact civilians. At one point after the exchange of fire and loss of life, a conversation occurred between the blue team lead and the cyber lead. The cyber lead was trying to advocate for the use of cyber attacks against targets that affected both military and civilian logistics. The blue lead was emphatic that cyber options, especially those against trains and power grids,

141

were off the table. In frustration, the blue lead told cyber to look into a lower range of options and voiced a concern that cyber was “outpacing the kinetic efforts.”

In general, the blue team was more comfortable conducting cyber deception that affected the adversary’s ability to control and influence its population, than strictly offensive cyber operations. They saw the deception operations as comfortably deniable and reversible, and therefore a less escalatory use of cyber operations than cyber attacks.

However, to complicate decision making about these deception operations, the blue team thought of these deception operations more as an effort to signal than as an asymmetric influence operation. Therefore, the cyber operations were designed to signal potential capability and will while not at the same time signaling aggression or the willingness to escalate. As can be expected, the red team failed to understand this elegant distinction.

In terms of the blue team’s response to red cyber actions, there was limited if any response. In the 2013 wargame, red undertook significant cyber play including attacks on blue command and control nodes as well as attacks on allied economic markets. Both of these actions were taken before conventional fire in the initial moves of the game. In both of these cases, as well as the case of cyber attacks on conventional blue forces after the naval blockade, the blue team viewed the cyber attacks as less escalatory than other kinetic options and therefore believed it was not worth response. The ethnographer captured some of this dynamics on day one in a conversation between the cyber lead and the blue team lead “cyber briefs that the adversary has conducted ‘very escalatory’ destruction of blue homeland nodes. Blue lead says, ‘we need to have a discussion about how we treat cyber attacks vice kinetic attacks.’ Cyber feels this is nearly kinetic, like bombing a command and control tower. Blue lead says it is different psychologically.”

142

Blue lead’s distinction about the psychological difference between cyber and kinetic was evident later in the game as well when blue vessels came under attack. As the message came in, the navy lead reported that “blue combined surface groups are under attack.”

The room appeared ready to escalate and then the navy lead corrected, “according to the commander in the region, it was only a cyber attack.” The team chose not to respond.

Wargame 2014

Wargame 2014 involved a land war scenario with an asymmetric adversary. A male policy leader led the blue team. The design of cyber play was consistent with the game played in 2013. Like the wargames in 2011 and 2013, the 2014 iteration showed cautious use of offensive cyber operations. There was a focus on deterring cyber operations consistent with behavior in 2013 and a concern about escalation to the nuclear realm by offensive cyber operations taken both by blue and by allies. Similarly, the blue team was concerned about the signaling effects of computer network exploitation. This concern was so strong that it affected not only the use of cyber exploitation to achieve access but also the use of cyber exploitation for traditional intelligence operations. At one point in the beginning of the game, a blue player cautioned the team, “they’re ready to shoot. Without provoking them we need to use cyber ability to locate them.” The concern was that even being in the adversary’s networks would be enough to start a conflict. A comment later by the defense lead in blue showed this assumption that cyber attacks could lead to nuclear escalation when he noted, “not a lot of strategic force mobilization [by red]. That will depend on whether red sees cyber and bomber

143

movement by blue.” The blue defense lead was—probably subconsciously—creating an equivalency between cyber operations and the use of airborne nuclear assets. He believed that either could instigate the movement of the adversary’s nuclear forces. Note also that he didn’t define what kind of cyber operations would be equivalent with nuclear bombers, but that broadly “cyber operations” of all flavors could have the same tailored effect as the movement of a nuclear bomber.

Due to these concerns, the blue team focused on cyber defense until after conventional military operations had commenced and blue nuclear forces had been put on alert. Cyber attacks taken by blue at that time were limited in scope to tit for tat attacks against the adversary’s cyber offensive capabilities. These were taken with great reticence. As one team member cajoled, “maybe we can take away some of their capabilities by taking out some of their cyber in a defensive way.” Even the perception of being offensive in cyber—even within a conventional conflict—was very concerning for the blue team.

The wargame in 2014 also demonstrated responses to red cyber activity that were consistent with play from all previous wargames. Despite significant attacks—to include a cyber attack that created physical effects on an allied nuclear facility—the blue team did not believe that the red cyber attacks warranted escalation. Even after significant allied pressure in response to a red cyber attack on the allied nuclear facility, the blue team warned of caution and the potential for escalation. As the blue team lead concluded, “we have to strike something soon. But . . . tell country Y we are preparing option to respond to country X’s provocation but we are also delaying these responses in order to allow ceasefire discussions to continue. We have continued to show caution,

144

doing forensic on cyber attack.” Interestingly, during this same period in the game conventional forces remained in place and continued low scale violence. In addition, this commentary reveals a dissonance in which the blue team was concerned that their exploitation activities would be caught and attributed and escalates to nuclear war, but when nuclear plants were attacked, the difficulty in attribution caused blue to take no response.

Wargame 2015

Wargame 2015 involved a land war scenario with a near-peer adversary. A female policy leader led the blue team. The design of cyber play was consistent with the game played in 2013 and 2014. The cyber play in the 2015 wargame was once again characterized by escalation concerns about cyberspace activity. Not only was there significant discussion about the use of cyber network exploitation, cyber-led information operations, and cyber attacks before conventional conflict, but even the deployment of a defensive cyber protection team to a foreign ally was questioned. Additionally, as the conflict escalated and shots were fired, significant debate still occurred about the escalatory nature of cyber operations taken in conjunction with other military operations.

And while the ability to conduct offensive operations against military targets became less likely to succeed (due to the loss of access during the wargame), the blue team was still concerned with the signaling created by the use of the cyber than the ability to create military effects. They were willing to sacrifice military effectiveness in cyberspace for the perception of greater stability. Consequently, throughout the wargame the blue team

145

placed significant restrictions on cyber operations and required that any attacks or information operations conducted through cyberspace would be non-attributable and reversible. This was consistent both pre and post conflict initiation.

The wargame in 2015 continued many of the nuclear equivalencies of earlier wargames, but the discussion about escalation to nuclear use was rich and articulated some of the logics of how cyber might lead to inadvertent escalation. In particular, there was a discussion that occurred early in the wargame about conducting a variety of cyber network exploitations. At one point, cyber asked the blue team lead if they would be authorized to conduct cyber network exploitation of the adversary’s strategic command and control. She responded, “prep for the environment is fine,” but iterated the concern that any move to action on these accesses would be centrally controlled at the presidential level. The defense lead echoed concerns about the authorization to achieve accesses, warning: “there are certain things there we will have to watch very carefully. We will have to be very careful about their nuclear redlines . . . wouldn’t want to worry them too much. Can we limit it to accesses and not worry Country x that their nuclear deterrent is held at risk?” The blue team lead affirmed these concerns and ended the conversation with, “I think it would hit our redlines. We would have to talk through these logics.”

The debate in the 2015 wargame also highlighted the players’ perception that cyber attacks that affected domestic populations would lead to nuclear war. This concern was not just about cyber attacks that created virtual or physical effects, but also cyber- driven information operations. In fact, the concern that social influence operations conducted via cyber would go nuclear led information operations to be conducted after kinetic action (air strikes and special operations forces). In a conversation about whether

146

or not to use cyber operations in conjunction with conventional operations, concerns about effects on American citizens curtailed the use of cyber attacks to aid the operation:

Cyber: We should look at cyber attacks on conventional military targets . . . Blue lead: What are the pros and cons of doing an operationally significant cyber cut? And doing a demonstration of that capability? Cyber: We could do a demonstration of that capability for instance on a dual-use system in the adversary’s homeland. Intel lead: Those are good ideas but you need to communicate to American that there could be dead Americans Cyber: If we demonstrate, they could demonstrate on American targets. Curtailing cyber operations due to civilian impact seems to be a product of two competing concerns. One is the belief that the American homeland is more vulnerable to cyber attacks than other nations. The other concern is that the high premium placed on not affecting civilians biases decision-makers towards operations with known effects.

Decision-makers are more confident that they can limit war with a 500 lb. bomb than a cyber-led information operation. Taken together, these concern create a strong perception that cyber operations that affect domestic populations are both dangerous for

Americans and dangerous for nuclear war. A revealing conversation at the end of the wargame in 2015 sheds light on these perceptions:

Defense lead: I’m looking for categories of cyber and when in the timeline you would implement them. Social networks would be very early in the conflict. Cyber: We wouldn’t do that in phase 0; very loathsome to do that in peacetime. Defense lead: Social early but not before phase 1. At the other end is the power, financial, and transport. Cyber: We need to do that early in phase 2. We are loathe to do that in cyberspace because it affects civilians . . .

147

Defense lead: I liken this to the use of WMD. Is there another category between social and these large political targets?284

Finally, as in all the previous wargames, the blue team in 2015 did not view red cyber attacks as grounds for escalation. In fact, after a successful red attack against an allied economic system, allies specifically requested a tit for tat cyber operation. The blue team demurred and instead supported continued sanctions against red.

Wargame 2016

Wargame 2016 played the same land war scenario with a near-peer adversary as in 2015. As such, changes in cyber play between the two are of particular interest. A male policy leader led the blue team. The design of cyber play was consistent with the game played in 2013, 2014, and 2015.

The 2016 wargame saw increased focus on proactive defensive measures, hardening, resiliency, and the use of cyber protection teams pre-emptively to mitigate vulnerabilities. Additionally, leaders were more comfortable with computer network exploitation for access than in previous years; however, reluctance to use cyber attacks

284284 The phases of conflict that the players refer to in this discussion is a planning framework used by the U.S. Department of Defense. The phase construct separates warfare into six phases. Phase 0- shaping, Phase 1: deterring, Phase 2- seizing the initiative, Phase 3-dominating, Phase 4- stabilizing, and Phase V- enabling civil authority. Though not explicitly tied to the law of armed conflict, the general assumption is that war begins in phase 2 and therefore the authority to conduct most armed operations is delegated to the geographic combatant command at that point. Prior to that time, hostilities will be highly controlled by peacetime or crisis rules of engagement. For a brief overview of the phasing construct, see Lauren Fish, “Painting by Numbers: A history of the U.S. Military’s Phasing Construct,” War on the Rocks, November 1, 2016: http://warontherocks.com/2016/11/painting-by-numbers-a-history-of-the-u-s-militarys-phasing- construct/.

148

remained with extreme emphasis placed on reversibility and “scoping” cyber attacks away from civilians and adversary leadership (this was despite the use of economic sanctions that directly targeted adversary leadership and had implications for civilians).

Once conventional operations were under way, there was a general support for cyber attacks that affected military capabilities, though there were still concerns that some cyber operations even after conventional force would lead to escalation. Therefore, rules of engagement about non-attribution curtailed many cyber operations. For instance, at one point cyber lead suggested an operation that would gain access to the supply chain and distribution of materials for nuclear and conventional war fighting. The blue lead, who was otherwise risk acceptant with his use of conventional force, declined to use the cyber operation arguing that “the risk of attribution is too high to move forward with this option.” Finally, the wargame in 2016 saw a continued adherence to an unofficial norm of non-attack against strategic command and control, to include exploits that could be misinterpreted at as an imminent attack.

Like the previous years’ wargames, blue did not escalate after any red cyber activity to include attacks on the blue homeland. On day three, after the employment of conventional forces and loss of life on both sides, a cyber attack was conducted on mainland blue that led to the loss of power that affected large blue civilian populations.

A conversation about the response led to the decision not to escalate, including not using cyber operations in response.

149

Discussion

Table 7. Cyber Escalation Ladders

Patterns of Play

With the exception of 2012 (which I will discuss further in the motivations section below), all the other wargames showed a strong belief that cyber operations would escalate crises, potentially even to nuclear war. These perceptions of escalation were so strong that they significantly curtailed the blue use of cyber attacks, cyber network exploitation, and often cyber-led information operations. Additionally, fears about

150

escalation led to tight rules of engagement for the cyber attacks that were utilized, including requirements for non-attribution and reversibility. These beliefs about escalation were unique from their conventional counterparts and across all the wargames players commented about the special nature of cyber. While cyber operations were often conflated with nuclear or space, they were never conflated with tanks, aircraft, or ships.

There are concrete policy implications for these escalation fears, all of which manifested during the five wargames in a surprisingly consistent way. First, intelligence in cyberspace is viewed qualitatively different than other more traditional means of military intelligence collection. Because obtaining access to a network can provide both situational awareness of the enemy’s activity and act as a launching point for an attack, blue players were wary about the effects of extensive cyber network exploitation. This fear was present in discussions about network exploitation of civilian or dual-use infrastructure, such as railroads or energy. However, the fear of escalation was especially pronounced when debating whether or not to seek accesses within adversary strategic command and control. Further, because many of these networks serve dual purpose with conventional military command and control, fears of escalation due to nuclear pressures also curtailed significant network exploitation into military command and control.

Interestingly, despite the murky nature of cyber espionage, discussions about the escalatory nature of these activities assumed that the U.S. activities would be attributed.

Therefore, any spying the U.S. was doing within cyberspace would necessarily become a signal of intentions to adversaries. This belief that cyber network exploitation would be a credible signal to adversaries created an interesting phenomenon in which the blue teams were both deterred from conducting exploitation because it would signal aggressive

151

intent and also considered cyber espionage a credible and discernible signal of U.S. capabilities that could be used to de-escalate in other domains. Though red team actions in wargames are generally not useful predictors of actual adversary behaviors, it may be worth noting that the red team never understood these activities as signals and were neither deterred nor driven to escalation based on cyber network exploitation.

Concerns about escalation had significant effects not only on whether to conduct cyber attacks, but also on the character of these attacks: what kinds of targets were most escalatory? What kind of effects? When in the crisis were they considered appropriate?

And what impact did these beliefs have on the decisions to utilize cyber-led information operations?

There are four overarching categories of cyber targets that were debated in these wargames: military targets, dual-use targets, civilian targets, and nuclear targets. Based on the fear of escalation, the first target of choice in the wargames was military capabilities—even better if those military capabilities were the adversaries’ offensive cyber operations. When the debate came to dual-use targets (i.e. energy, transportation, or communication), there was extreme reticence to conduct cyber attacks and pure civilian as well as nuclear targets were completely off the table. These decisions were driven by escalation and not by capability because the commentary suggests that military targets were not necessarily the easiest to attack or the ones that would make the greatest effect. Dual-use and civilian targets that use SCADA systems and are not controlled by the government are generally the easiest kind of cyber targets; reticence to target these reflects concerns not about capability but about escalation.

152

Additionally, there was significant discussion about the types of effects that cyber could create that would lead to escalation. With the exception of 2012, every single blue lead asked cyber to create effects that were virtual and reversible—all while retaining non-attribution. In fact, these leads were quite often willing to trade off the ability to achieve effect in order to create the perception of escalation control. This could be because cyber provides a flexibility in effects that you don’t get with a lot of physical weapons. You can’t choose whether a bomb physically or virtually destroys its target; even the anti-radiation missiles that are considered non-kill weapons physically destroy a radar. Cyber allows some potentially flexibility that might be appealing for decision makers. Unfortunately, these choices for virtuality and reversibility were often made after conventional force had already been committed so in some ways, while it provided the decision makers more flexibility, they were already committed to more escalatory actions in other domains.

This brings me to the third consideration for cyber attacks: the point in the crisis in which cyber attacks were considered to be least escalatory. Barring 2012, all cyber attacks were conducted after conventional force on force conflict had occurred. In U.S.

Department of Defense parlance, the crisis was firmly in Phase 2 (the phase generally associated with armed conflict) before cyber operations were considered non-escalatory.

Even virtual attacks were not condoned prior to those conventional actions. This has significant implications for U.S. responses to increasingly persistent use of cyber operations before armed conflict.

Finally, the wargames demonstrated perhaps a uniquely American concern about cyber-led information operations and escalation control. For many of the wargames,

153

these information operations were not used until after cyber attacks on military targets and were used in conjunction with nuclear alert and even nuclear demonstration.

Conversation debating the use of the information operations often equated their effect with nuclear capabilities and implied an existential threat to adversaries. Because of this existential threat, not only were these not used quite often until the end of the crises, but they were not targeted at regime overhaul but instead at decreasing public support for the use of force. These wargames were all conducted prior to the U.S. election in 2016; perceptions about escalation may have changed after the increased focus on Russian led information operations in peacetime.

In general, the fear of escalation due to cyber operations seems to be based on three factors: 1) a perception that the U.S. is more vulnerable to cyber attacks than its adversaries, 2) concerns about the relationship between cyber and nuclear capabilities, and 3) concerns about the domestic implications of cyber attacks. All of these concerns are magnified by the uncertainty ubiquitous in cyber operations. That uncertainty caused individuals to look for analogies in the nuclear realm and exacerbated concerns about collateral damage and escalation. The bounds of possibility with cyber effects are so expansive that it may become easier cognitively for decision-makers to drop a 1000lb bomb than a virtual cyber attack on the same target. Though the chance of a catastrophic effect in cyber is incredibly low, the high uncertainty of the attack means that it can’t be ruled out. In many ways, physics and standard intelligence can bound the uncertainty of the 1000lb bomb so that while the net effects are potentially greater than cyber, there is no potential effect at the extremes.

154

These factors contribute to the lack of response by blue to red cyber actions. Once in

2016 blue responded to a cyber attack with a slight increase in economic sanctions, but otherwise red cyber attacks either didn’t reach the threshold of concern or blue couldn’t find a response that they thought was proportional. Additionally, while the blue teams assumed that their cyber operations would be attributed, confusion about attribution seemed to decrease the chance that blue would escalate to red cyber attacks.

Motivations behind cyber play

Minus 2012, the cyber play in all of the wargames was remarkably consistent.

Blue was concerned about the escalatory effects of cyber operations so they were generally cautious in their use of cyber network exploitation, cyber attack, and cyber-led information operations. Also consistent was their lack of response to red cyber activity.

In no wargame did blue choose to escalate because of a red cyber attack. This is an interesting divergence in behavior. What can explain the motivations behind this seemingly contradictory logic?

Previously, I identified five potential hypotheses to explain the motivations for perceptions of escalation due to cyber operations. The consistency of play for five of the six scenarios presents problems for two of the hypotheses. Both capability variables and the situational context varied over the six years the game was played.

In terms of capabilities, we see a significant evolution of U.S. capabilities over 2011 to 2016. The United States brought to operational capacity Cyber Command and built 120+ cyber mission teams with skills in defense and offensive operations. The design of the game closely followed the evolution of these capabilities by integrating

155

offensive and defensive cyber teams as cyber capabilities (vice more general cyber tokens associated with effects as was used in the first two games). Consequently, cyber defensive play became much more robust over time; cyber protection teams were used as a capability and were forward deployed in 2016 as a part of an overall deterrence package. That represents a substantial change for how computer network defense was conceptualized in both real life and the game from 2011 to 2016.

However, we didn’t see a difference in offensive play over those six years as capabilities changed. This is despite the same cyber player in all of the games who consistently advocated for the greater use of offensive cyber operations across all games.

This would suggest that cyber capabilities are not the primary motivator for decisions to utilize offensive cyber operations. In contrast to the variance in cyber capabilities, what has been consistent over this time period is U.S. conventional dominance. Therefore, the non-use of cyber operations may not be tied to U.S. cyber capabilities, but it may instead be tied to the fact that the U.S. has so many conventional options in crises. For states that are much more capable than their peers in other domains, the decision to not respond to cyber attacks while at the same time not utilizing cyber attacks may be a gift of power.

You can be concerned about the escalatory effects of cyber when you have conventional dominance to fall back on.

Second, over the six years the wargame was played, there were five different scenarios with multiple adversaries in multiple contexts. And yet there was consistent use of cyber operations. The consistency suggests that the situational context was not a driving factor in these cyber decisions. The one year that we did see cyber played

156

differently, 2012, was run the next year with a similar scenario and returned to the play that we had seen previously and that we saw in 2014, 2015, 2016.

That aberration year does, however, lend some credence to the power individual personalities play in the use of cyber operations. Especially because there are few existing or solidified norms of behavior in cyberspace, limited U.S. policies, and almost no empirical precedent, cyber operations are particularly malleable and prone to leadership motivations. This is exacerbated by the fact that authority to conduct cyber operations in the United States is centrally controlled and quite often at the highest levels.

That means that whoever is the President will have a large role in how cyber is utilized in crises. Previous research suggests that individual personality of Presidents matter for the way states fight wars and that this especially important in emerging technologies.285

Therefore, while 2012 was an “aberration” in cyber play, it does point to the extreme importance of the U.S. President in the role that cyber plays in crises.

The remaining two variables—organizations and cognition can help explain some of the dissonance in the cyber behaviors of the blue teams in five of the six years of wargaming play. First, in terms of organizational influences, the United States’ delegation of cyber responsibilities within the Department of Defense has created some institutional legacies for cyber operations. Cyber Command falls under Strategic

Command as a sub-unified command. Strategic Command is traditionally the organization that deals with nuclear capabilities and space. They focus on strategic effects and worry about deterrence and escalation control. Because of Cyber Command’s

285 Julia Macdonald and Jacquelyn Schneider, "Presidential Risk Orientation and Force Employment Decisions The Case of Unmanned Weaponry." Journal of Conflict Resolution (2015): 0022002715590874.

157

position as a sub-unified command of Strategic Command, there may be a false equivalency between the strategic assets and effects of the predominantly nuclear

Strategic Command and cyber operations.286 Throughout the wargames the players associated cyber with nuclear and often claimed that the use of cyber operations would lead to nuclear war. Note that very few of the blue players were experts in the red doctrine and very few used evidence from red statements or behavior to support that assumption. If cyber is a strategic resource like nuclear weapons, then it logically should be used sparingly, late in conflict, and with clear rules of engagement that allow for escalation control.

But if cyber is an operational resource than these limitations to its use are no longer as applicable. The question is how much of the equivalency with nuclear weapons is false. Blue responses to red cyber attacks would suggest that cyber operations do not have the strategic effect that the blue players were concerned about in their debates about escalation. Part of this may be tied intimately with the way we as human beings process uncertainty and the high amount of uncertainty tied to cyber operations. While that uncertainty may create many potential deleterious outcomes from cyber operations

(which limits our use), when they are actually conducted the virtual nature of those operations doesn’t lend itself to the same fear-inducing crisis behaviors created when physical effects occur. And indeed even when physical effects occur from cyber operations, their second-order nature tempers the fear generated from a cyber attack.

286 It is also significant to note that this is a Strategic Command exercise and is staffed with many (but not the majority) Strategic Command employees. This may decrease some of the generalizability of the findings. However, statements by the Obama administration as well as the scope of cyber operations under his administration indicate that these views are prevalent in many parts of the Washington decision-making apparatus.

158

In one of the wargames, the blue team lead commented that he didn’t need to respond to a cyber attack because it was psychologically different. This could be a fundamental truth that goes beyond the wargaming players, beyond U.S. foreign-policy decision makers and explains how humans react to technological threats and particularly cyber. Research on fear suggests that human beings process fear in very similar ways across cultures because of hundreds of thousands of years of evolutionary conditioning.

However, cyber is a new threat and is not conditioned for response. We are not primed by fear to respond to cyber operations. Therefore, cyber operations are more likely to create another emotional response: anxiety, which in turn tempers reactions to cyber operations and makes us choose risk-averse cyber strategies.287 As Libicki aptly describes it, “cyberwar engenders worry.”288 Worry is very different from fear and may explain the lack of reaction to red cyber attacks and the reluctance to use blue cyber attacks.

Conclusion

These wargames provide some potential insight into the future use of cyber operations in crises and their impact on escalation for the United States. For five of the six wargames, the players were reticent to use cyber operations due to their fears of escalation while at the same time choosing not to respond to red cyber attacks. These five wargames suggest that cyber operations, if conducted similarly to these wargames, would not lead to escalation. But these uses were highly bound—and bound primarily to

287 Jacquelyn Schneider, “Beyond Strictly Material Assessments of Weapons’ Effectiveness: Examining the Effect of Fear in Threat Assessment,” paper presented at International Studies Association Annual Meeting, February 2014, New Orleans, Louisiana. 288 Martin Libicki, Crisis and Escalation in Cyberspace (Santa Monica: RAND Corporation, 2012, pg. 21).

159

control escalation. In the one wargame where the blue team did utilize offensive cyber operations early and with great effect, it had significant impact on crisis escalation.

But there are some significant caveats to the overall generalizability of these findings. First, the game is framed as a deterrence and escalation game and is sponsored by Strategic Command. Care is taken within the game to remain above the level of operational warfare. This will necessarily lead individuals to focus on strategic concerns over operational objectives and therefore we likely see less inadvertent escalation from conventional operations to war than we might see in other more operationally-focused games. Secondly, because of the nature of the sample, most of the explanations for cyber behavior are specific to the United States. Therefore, conclusions from these wargames should not be mirrored across cultures or states.

What then can this research say about the potential for cyber operations to lead to escalation? This research provides evidence that many within the U.S. decision-maker community are worried that U.S. cyber operations will lead to escalation. Because of that concern, they have over the last six years built policies that centralize control and limit the use of cyber operations prior to armed conflict. This could significantly decrease the chance that cyber operations used against the United States would escalate to conflict, while also decreasing the chance that the U.S. would use cyber operations in a way that might create inadvertent escalation from adversaries. But these policies are at their infancy and the centralized control of these operations mean that Presidential risk proclivities will have a significant impact on how these cyber operations are used in the future. The early use of offensive cyber operations in the 2012 game coincided with the most escalatory wargame in the six year time span. Therefore, these findings cannot

160

preclude the possibility that risk acceptant Presidents may use offensive cyber operations to advertently escalate conflict to achieve objectives.

Additionally, in none of the wargames did the gameplayers feel a need to respond to cyber attacks—even when these attacks affected civilians on the homeland and even when they caused nuclear fall out in an ally. While these findings were found within

U.S. players, the pattern of low-level responses that we have seen across countries to known cyber attacks suggest that these behaviors may be generalizable across states.

Future research should be conducted to better understand what motivates the lack of response. How generalizable is it? Is it something that is unique to American decision- makers, or is it something more pervasive that can explain cyber behaviors of adversary states? And if it is more pervasive, then should American decision-makers be more open to using cyber operations earlier in crises?

161

Bibliography Adams, Karen Ruth. 2003. "Attack and Conquer? International Anarchy and the Offense- Defense-Deterrence Balance." International Security 28 (3): 45-83. Air-Sea Battle: Service Collaboration to Address Anti-Access & Area Denial Challenges. 2013a. http://www.dtic.mil/docs/citations/ADA584067. Alberts, David S. 2001. Understanding Information Age Warfare. Washington: CCRP Publication Series. Arquilla, John and David F. Ronfeldt. 2001. Networks and Netwars. Santa Monica, CA: Rand. Attwood, Chandler and White, Jeffrey. "Syrian Air-Defense Capabilities and the Threat to Potential U.S. Air Operation." The Washington Institute., last modified May 23, http://www.washingtoninstitute.org/policy-analysis/view/syrian-air-defense- capabilities-and-the-threat-to-potential-u.s.-air-operat. Avant, Deborah D. 1993. "The Institutional Sources of Military Doctrine: Hegemons in Peripheral Wars." International Studies Quarterly 37 (4): 409-430. Ball, Desmond. 1981. "Can Nuclear War be Controlled?" Adelphi Papers 169: 1-35. . Bateman, Robert L. 1999. Digital War. Novato: Presidio. Batey, Angus. "F-35 Logistics System may be Vulnerable to Cyberattack." Aviation Week., last modified March 3, accessed March 13, 2017, http://aviationweek.com/defense/f-35-logistics-system-may-be-vulnerable- cyberattack. Bergeron, Louis. 1981. France Under Napoleon. Princeton, N.J: Princeton Univ. Pr. Bertaud, Jean-Paul. 1988. The Army of the French Revolution. Princeton, N.J: Princeton Univ. Pr. Betz, David. 2012. "Cyberpower in Strategic Affairs: Neither Unthinkable nor Blessed." Journal of Strategic Studies 35 (5): 689. Beukema, Herman. 1941. "Social and Political Aspects of Conscription: Europe's Experience." Military Affairs 5 (1): 21-31. Biddle, Stephen. 2010. Military Power. Princeton: Princeton University Press ———. 2001. "Rebuilding the Foundations of Offense-Defense Theory." The Journal of Politics 63 (3): 741-774. Biddle, Stephen and Ivan Oelrich. 2016. "Future Warfare in the Western Pacific: Chinese Antiaccess/Area Denial, U.S. AirSea Battle, and Command of the Commons in East Asia." International Security 41 (1): 7-48. Biddle, Stephen and Stephen Long. 2004. "Democracy and Military Effectiveness: A Deeper Look." The Journal of Conflict Resolution 48 (4): 525-546.

162

Bitzinger, Richard and Raska, Michael. "The Air Sea Battle Debate and the Future of Conflict in East Asia." RSIS Policy Brief., accessed September 27, 2013, http://www.rsis.edu.sg/publications/policy_papers/RSIS_Air%20Sea%20Battl e_190213%20v1_Print.pdf. Blair, Bruce . 1985. Strategic Command and Control. Washington, D.C: Brookings Institution. ———.1993. The Logic of Accidental Nuclear War. Washington: Brookings Institute. Bolio, Robert. 2004. The Over-Reliance on Technology in Warfare: The Yom Kippur War as a Case Study. Carlisle Barracks: U.S. Army Command and General Staff College. Boot, Max. 2006. War made New. New York: Gotham Books. Boyd, Aaron. "DNI Clapper: Cyber Bigger Threat than Terrorism." Federaltimes.com., last modified February 4, accessed March 13, 2017, http://www.federaltimes.com/story/government/cybersecurity/2016/02/04/cyb er-biggerthreat- terrorism/79816482/. Bracken, Paul J. 1983. Theœ Command and Control of Nuclear Forces. New Haven: Yale University Press. Brantly, Aaron Franklin. 2016. The Decision to Attack : Military and Intelligence Cyber Decision-Making. Athens, GA: University of Georgia Press. Brewer, Garry D., Martin Shubik, Alfred Hausrath, and Stuart Bremer. 1980. Policy- Making Perspectives on War Simulations. Vol. 24. Ann Arbor, Mich: Sage Publications. Brito, Jerry and Watkins, Tate. "Loving the Cyber Bomb? the Dangers of Threat Inflation in Cybersecurity Policy.", last modified April 10, accessed March 13, 2017, http://mercatus.org/publication/loving-cyber-bomb-dangers-threat-inflation- cybersecurity-policy-0. Brock, Gerald. 2003. The Second Information Revolution.. Cambridge: Harvard University Press. Brodie, Bernard. 1971. Strategy in the Missile Age. Princeton, NJ: Princeton University Press. Brodie, Bernard, Frederick Sherwood Dunn, Arnold Wolfers, Percy Ellwood Corbett, William Thornton, and Rickert Fox. 1946. The Absolute Weapon. New York: Harcourt. Broers, Michael. 2008. "The Concept of `Total War' in the Revolutionary—Napoleonic Period." War in History 15 (3): 247-268. Bronfeld, Saul. 2007. "Fighting Outnumbered: The Impact of the Yom Kippur War on the U.S. Army." The Journal of Military History 71 (2): 465-498. Bronk, Christopher and Eneken Tikk-Ringas. 2013. "The Cyber Attack on Saudi

163

Aramco." Survival 55 (2): 81-96. Bueno de Mesquita, Bruce. 1981. The War Trap. New Haven: Yale Univ. Press. Bunker, Robert J. 1998. Five-Dimensional (CYBER) Warfighting: Can the Army After Next be Defeated through Complex Concepts and Technologies. Carlisle: U.S. Army War College Strategic Studies Institute. Butterfield, Herbert. 1950. History and Human Relations. London: Collins. Cannon, M. W. 1984. The Development of FM (Field Manual) 100-5 from 1945 Until 1976. Cares, Jeff. 2005. Distributed Networked Operations: The Foundations of Network Centric Warfare. Newport: Alidade Press. Carey, Bill. "GAO Questions Deployability, Redundancy, of F-35 ALIS System." AIN Online., last modified April 21, accessed 13 March, 2017, http://www.ainonline.com/aviation-news/defense/2016-04-21/gao-questions- deployability-redundancy-f-35-alis-system. Carlier, Claude, Philippe Contamine, André Corvisier, and Jean Delmas. 1994. Histoire Militaire De La France. Paris: Presses Universitaires de France. Carter, Ashton, John D. Steinbruner, Charles A. Zraket, and Sherman Frankel. 1987. Managing Nuclear Operations. Washington, DC: Brookings Inst. Carnesale, Albert and Charles Glaser. 1982. "ICBM Vulnerability: The Cures are Worse than the Disease." International Security 7 (1): 70-85. Cebrowski, Arthur K. and John J Garstka. 1998. "Network-Centric Warfare: Its Origin and Future." United States Naval Institute. Proceedings, Jan 1, 28. Chee-Wooi Ten, Chen-Ching Liu, and G. Manimaran. 2008. "Vulnerability Assessment of Cybersecurity for SCADA Systems." IEEE Transactions on Power Systems 23 (4): 1836-1846. Cimbala, Stephen J. 1990. First Strike Stability. Contributions in Military Studies. Vol. 101. Clark, Asa, Peter Chiarelli, Jeffrey McKitrick, and James Reed, eds. 1984. The Defense Reform Debate. Baltimore: Johns Hopkins Press. Clark, Asa and John Lilly, eds. 1989. Defense Technology. New York: Praeger. Clark, Bryan. 2015. The Emerging Era in Undersea Warfare: Center for Strategic and Budgetary Assessment. Clark, John E. 2001. Railroads in the Civil War. Baton Rouge: Louisiana State Univ. Press. Clarke, Richard. 2010. Cyber War: The Next Threat to National Security and what to do about It. New York: Harper Collins.

164

Clarke, Richard A. and Robert K. Knake. 2010. Cyber War. New York: Ecco. Cohen, Eliot A. 2004. "Change and Transformation in Military Affairs." Journal of Strategic Studies 27 (3): 395-407. ———. 1996. "A Revolution in Warfare." Foreign Affairs: 37-54. Colgan, Jeff D. 2010. "Oil and Revolutionary Governments: Fuel for International Conflict." International Organization 64 (4): 661-694. ———. 2014. "Oil, Domestic Politics, and International Conflict." Energy Research & Social Science 1: 198-205. doi:10.1016/j.erss.2014.03.005. Cronin, Audrey Kurth. 2006. "Cyber-Mobilization: The New Levee En Masse." Parameters 36 (2): 77-87. Cross, Matthew D. 2011. World Gone Cyber MAD: How "Mutually Assured Debilitation" is the Best Hope for Cyber Deterrence. Vol. 5. Maxwell Air Force Base: Air Force Research Institute. Cullather, Nick. 2006. "Bombing at the Speed of Thought: Intelligence in the Coming Age of Cyber-War." Intelligence and National Security 18 (4). Dahl, Erik J. 2002. "Network Centric Warfare and the Death of Operational Art." Defence Studies 2 (1): 1. Danzig, Richard. 2014. Surviving on a Diet of Poisoned Fruit: Reducing the National Security Risks of America's Cyber Dependencies: Center for a New American Security. Demchak, Chris C. 2011. Wars of Disruption and Resilience : Cybered Conflict, Power, and National Security. Studies in Security and International Affairs. Athens: University of Georgia Press. Department of Defense. 2015. Department of Defense Cyberspace Strategy. ———. 2012a. Joint Operational Access Concept. ———. 2013a. Joint Publication 3-12, Cyberspace Operations. ———. 2013b. Resilient Military Systems and the Advanced Cyber Threat. ———. 2012b. Sustaining U.S. Global Leadership: Priorities for 21st Century Defense. Department of the Army. 1976. Field Manual 100-5: Operations ———.1982. Field Manual 100-5: Operations. ———. 1981. TRADOC Pam 525-5: The Airland Battle and Corps 86. Deptula, David. 2001. Effects-Based Operations: Changes in the Nature of War. Arlington: Aerospace Education Foundation. Depuy, William. 1975. "The Implications of the Middle East War on US Army Tactics,

165

Doctrine, and Systems.". DePuy, William E. 1976. Letter from William E DePuy to Fred C. Weyand, edited by Fred C. Weyand. Dick, Jimmy. "The Gunpowder Shortage." Journal of the American Revolution., last modified September 9, accessed March 13, 2017, https://allthingsliberty.com/2013/09/the-gunpowder-shortage/. DoD Operational Test and Evaluation. "FY2014 Annual Report;" http://www.dote.osd.mil/pub/reports/fy2014/pdf/other/2014DOTEAnnualR eport.pdf., last modified January. Dong-Joon Jo and Erik Gartzke. 2007. "Determinants of Nuclear Weapons Proliferation." The Journal of Conflict Resolution 51 (1): 167-194. Dorminey, Bruce. "How Bad would it be if the Russians Started Cutting Underwater Cables? Try Trillions in Damage." Forbes.com., last modified November 2, accessed March 13, 2017, http://www.forbes.com/sites/brucedorminey/2015/11/02/russian- navy-probing-u-s-undersea-communications-cables-in-new-global- threat/#520cbaf866b1. Drucker, Peter F. 1999. "Beyond the Information Revolution." The Atlantic, Oct 1, 47. Dupree, Philip and Jordan Thomas. 2012. "Air-Sea Battle: Clearing the Fog." Armed Forces Journal, Jun 1, 10. Epstein, Robert M. 1992. "Patterns of Change and Continuity in Nineteenth-Century Warfare." The Journal of Military History 56 (3): 375-388. Estaville, Lawrence E. 1973. "A Strategic Railroad: The New Orleans, Jackson and Great Northern in the Civil War." Louisiana History: The Journal of the Louisiana Historical Association 14 (2): 117-136. Faith, Nicholas. 1990. The World the Railways Made. 1. ed. ed. New York: Carroll & Graf. Farrell, Theo and Terry Terriff. 2001. Sources of Military Change : Culture, Politics, Technology. Boulder: Lynne Rienner Publishers. Finnemore, Martha. 2004. The Purpose of Intervention: Changing Beliefs about the use of Force. Ithaca: Cornell University Press. Finnemore, Martha and Kathryn Sikkink. 1998. "International Norm Dynamics and Political Change." International Organization 52 (4): 887-917. Fish, Lauren. "Painting by Numbers: A History of the U.S. Military's Phasing Construct." War on the Rocks., last modified November 1, accessed March 13, 2017, http://warontherocks.com/2016/11/painting-by-numbers-a-history-of-the-u-s- militarys-phasing-construct/. Forrest, Alan. 1990. Theoe Soldiers of the French Revolution. Durham: Duke Univ.

166

Press. Freedberg, Sydney J. "Pentagon Reports on China’s Satellite Killers." Breakingdefense., last modified May 11, accessed March 13, 2017, http://breakingdefense.com/2015/05/pentagon-reports-on-chinas-satellite- killers/. Freedman, Lawrence. 2003. The Evolution of Nuclear Strategy. Studies in International Security. 3. ed. ed. Vol. 20. Basingstoke: Palgrave Macmillan. Friedman, Norman. 2000. Seapower and Space. London: Chatham Publ. Fritz, Jason. 2012. "Satellite Hacking: A Guide for the Perplexed." Bulletin of the Centre for East-West Cultural and Economic Studies 10 (1): 21-50. Fulghum, David A., Robert Wall, and Amy Butler. 2007. "Cyber-Combat's First Shot: Israel shows Electronic Prowess." Aviation Week & Space Technology, Nov 26, 28. Futter, Andrew. 2016. "The Dangers of using Cyberattacks to Counter Nuclear Threats." Arms Control Today 46 (6): 8. Gabel, Christopher R. 1997. Railroad Generalship. Fort Leavenworth, Kan: U.S. Army Command and General Staff College, Combat Studies Inst. Gaddis, John Lewis. 2006. The Cold War: A New History. New York: Penguin. Garwin, Richard L. 1983. "Will Strategic Submarines be Vulnerable?" International Security 8 (2): 52-67. Gartzke, Erik. 2013. "The Myth of Cyberwar: Bringing War in Cyberspace Back Down to Earth." International Security 38 (2): 41-73. Gartzke, Erik and Jon Lindsay. 2017. "Thermonuclear Cyberwar." Journal of Cybersecurity: 1-12. Gelpi, Christopher. 2003. The Power of Legitimacy. Princeton, NJ: Princeton Univ. Press. Gelpi, Christopher and Peter D. Feaver. 2004. Choosing Your Battles: American Civil- Military Relations and the use of Force : With a New Afterword by the Authors Princeton, NJL Princeton University Press. Gentry, John A. 2012. How Wars are Won and Lost. Santa Barbara, CA: Praeger. Giles, Keir. 2016. Russia's 'New' Tools for Confronting the West: Continuity and Innovation in Moscow's Exercise of Power: The Royal Institute of International Affairs. Giles, Keir and William Hagestad II. 2013. "Divided by a Common Language: Cyber Definitions in Chinese, Russian, and English." Gilpin, Robert. 1981. War and Change in World Politics. Cambridge: Cambridge Univ. Press.

167

Glaser, Charles L. 1990. Analyzing Strategic Nuclear Policy. Princeton, NJ: Princeton Univ. Press. ———. 1985. "Do we Want the Missile Defenses we can Build?" International Security 10 (1): 25-57. ———. 2013. "How Oil Influences U.S. National Security." International Security 38 (2): 112-146. ———. 1994. "Realists as Optimists: Cooperation as Self-Help." International Security 19 (3): 50-90. ———. 1997. "The Security Dilemma Revisited." World Politics 50 (1): 171-201. Glaser, Charles L. and Chairn Kaufmann. 1998. "What is the Offense-Defense Balance and how can we Measure it?" International Security 22 (4): 44-82. Gompert, David C. and Martin Libicki. 2014. "Cyber Warfare and Sino-American Crisis Instability." Survival 56 (4): 7-22. Gortzak,Yoav Yoram Z. Haftel, and Kevin Sweeney. 2005. "Offense-Defense Theory: An Empirical Assessment." The Journal of Conflict Resolution 49 (1): 67-89. Gottfried, Kurt and Bruce Blair, eds. 1988. Crisis Stability and Nuclear War. New York: Oxford Univ. Pr. Gray, Colin S. 2002. Strategy for Chaos. London: Cass. Grimsley, Mark. 2001. "Surviving Military Revolution: The U.S. Civil War." In The Dynamics of Military Revolution 1300-2050, 74-91. Cambridge: Cambridge University Press. Gross, Michael Joseph. 2013. "Silent War." Vanity Fair, July. Halpin, Edward F. 2006. Cyberwar, Netwar, and the Revolution in Military Affairs. New York: Palgrave Macmillan. Harrigian, Jeff and Marosko, Max. "Fifth Generation Air Combat: Maintaining the Joint Force Advantage." The Mitchell Forum., accessed 13 March, 2017, http://media.wix.com/ugd/a2dd91_bd906e69631146079c4d082d0eda1d68.pdf . Haycock, Ronald and Keith Neilson. 1987. Men, Machines, and War. New York: Wilfrid Laurier University Press. Hayward, Joel. 1995. "Hitler's Quest for Oil: The Impact of Economic Considerations on Military Strategy, 1941-42." Journal of Strategic Studies 18 (4): 94-135. Healey, Jason. 2013. A Fierce Domain : Conflict in Cyberspace, 1986 to 2012. Washington: The Atlantic Council. Hellström, Tomas. 2007. "Critical Infrastructure and Systemic Vulnerability: Towards a Planning Framework." Safety Science 45 (3): 415-430.

168

Hendrickson, David C. 1988. Reforming Defense. Baltimore: Johns Hopkins Univ. Pr. Herbert, Paul H (Paul Hardy). 1988. Deciding what has to be done: General William E. DePuy and the 1976 Edition of FM 100--5, Operations. Carlisle Barracks: U.S. Army Command and General Staff College. Herz, John H. 1950. "Idealist Internationalism and the Security Dilemma." World Politics 2 (2): 157-180. Higgins, Kelly J. "Chinese Hackers Target Logistics and Shipping Firms with Poisoned Inventory Scanners " Information Week Dark Reading., last modified July 10, accessed March 13, 2017, http://www.darkreading.com/attacks-breaches/chinese- hackers-target-logistics-and-shipping-firms-with-poisoned-inventory-scanners/d/d- id/1297182. Hodges, Robert. 2012. American Civil War Railroad Tactics. Elite. Vol. 171 Osprey Publishing. Hughes, Llewelyn and Austin Long. 2015. "Is there an Oil Weapon?: Security Implications of Changes in the Structure of the International Oil Market." International Security 39 (3): 152-189. Hundley, Richard O., Robert H. Anderson, and Tora K. Bikson. 2003. Global Course of the Information Revolution : Recurring Themes and Regional Variations. 1st ed. Santa Monica: RAND. Huntington, Samuel P. 1957. The Soldier and the State. Vol. 514 : Caravelle edition. New York: The Vintage Press. Hurd, Ian. 2007. "Breaking and Making Norms: American Revisionism and Crises of Legitimacy." International Politics 44 (2-3): 194-213. Iasiello, Emilio 2015. "Are Cyber Weapons Effective Military Tools." Military and Strategic Affairs 7 (1): 23-40. Insinna, Valerie. 2015. "Troubled Logistics System Critical to F-35's Future." National Defense, Apr 1, 28. Jensen, Benjamin. 2016. Forging the Sword : Doctrinal Change in the U.S. Army. Redwood City: Stanford Security Studies. Jervis, Robert. 1978. "Cooperation Under the Security Dilemma." World Politics 30 (2): 167-214. ———. 1982. "Deterrence and Perception." International Security 7 (3): 3-30. ———. 1976. Perception and Misperception in International Politics. Princeton, NJ: Princeton Univ. Press. ———. 1984. The Illogic of American Nuclear Strategy. Ithaca: Cornell Univ. Press. ———. 1990. The Meaning of the Nuclear Revolution. Ithaca: Cornell Univ. Press.

169

———. 2001. "Was the Cold War a Security Dilemma?" Journal of Cold War Studies 3 (1): 36-60. Jervis, Robert, Richard Ned Lebow, and Janice Gross Stein. 1985. Psychology and Deterrence. Baltimore: Johns Hopkins Univ. Pr. Joint Vision 2020: America's Military - Preparing for Tomorrow. 2000: Joint Force Quarterly. Joubert, Vincent. "Getting the Essence of Cyberspace: A Theoretical Framework to Face Cyber Issues."2010. Joyce, Adam. 2012. "The Micropolitics of "the Army You have": Explaining the Development of U.S. Military Doctrine After Vietnam." Studies in American Political Development 26 (2): 180. Junio, Timothy. 2013. "How Probable is Cyber War? Bringing IR Theory Back in to the Cyber Conflict Debate." Journal of Strategic Studies 36 (1): 125. Kahn, Herman. 2012. On Escalation. New Brunswick: Transaction Publications. Kaplan, Fred M. 2016. Dark Territory. First Simon & Schuster hardcover edition ed. New York: Simon & Schuster. Katzenstein, Peter. 1996. The Culture of National Security: Norms and Identity in World Politics. Cambridge: Cambridge University Press. Kelanic, Rosemary A. 2016. "The Petroleum Paradox: Oil, Coercive Vulnerability, and Great Power Behavior." Security Studies 25 (2): 181-213. Kello, Lucas. 2013. "The Meaning of the Cyber Revolution: Perils to Theory and Statecraft." International Security 38 (2): 7-40. Kelly, Jack. 2009. Gunpowder : Alchemy, Bombards, and Pyrotechnics: The History of the Explosive that Changed the World. New York: Basic Books. Keohane, Robert O. and Joseph S. Nye Jr. 1998. "Power and Interdependence in the Information Age." Foreign Affairs: 81- 94. http://www.foreignaffairs.com/articles/54395/robert-o-keohane-and-joseph-s- nye-jr/power-and-interdependence-in-the-information-age. Khalilzad, Zalmay, A. H. Marshall, and John P. White, eds. 1999. The Changing Role of Information in Warfare. Santa Monica, CA: Rand. Khazan, Olga. "The Creepy, Long-Standing Practice of Undersea Cable Tapping." The Atlantic., last modified July 16, accessed March 13, 2017, http://www.theatlantic.com/international/archive/2013/07/the-creepy-long- standing-practice-of-undersea-cable-tapping/277855/. Kier, Elizabeth. 1995. "Culture and Military Doctrine: France between the Wars." International Security 19 (4): 65-93.

170

———. 1997. Imagining War. Princeton, NJ: Princeton Univ. Press. Klein, John J. 2012. Space Warfare: Strategy, Principles, and Policy. New York: Routledge. Knox, Macgregor. 2001. The Dynamics of Military Revolution, 1300-2050. Cambridge: Cambridge University Press. Korns, Stephen. 2009. Cyber Operations: The New Balance. Washington: National Defense University. Kramer, Franklin D., Stuart H. Starr, and Larry K. Wentz. 2009. Cyberpower and National Security. 1. ed. ed. Washington, D.C: Potomac Books. Krepinevich, Andrew F. 1994. "Cavalry to Computer: The Pattern of Military Revolutions." The National Interest, Oct 1, 30-42. Kshetri, Nir. 2005. "Information and Communications Technologies, Strategic Asymmetry and National Security." Journal of International Management 11 (4): 563-580. Lawson, Sean. 2011. "Beyond Cyber Doom: Cyberattack Scenarios and the Evidence of History." Mercatus Center 11 (1): 1-37. Lebow, Richard Ned. 1984. Between Peace and War. Baltimore: Johns Hopkins University Press. Lee, Dan. 2011. The L&N Railroad in the Civil War. Jefferson: McFarland & Company Inc, Publishers. Levinson, Paul. 1998. The Soft Edge: A Natural History and Future of the Information Revolution. New York: Psychology Press. Levy, Jack S. 1984. "The Offensive/Defensive Balance of Military Technology: A Theoretical and Historical Analysis." International Studies Quarterly 28 (2): 219- 238. Liaropoulos, Andrew N. 2006. "Revolutions in Warfare: Theoretical Paradigms and Historical Evidence: The Napoleonic and First World War Revolutions in Military Affairs." The Journal of Military History 70 (2): 363-384. Libicki, Martin. 2013. "Don'T Buy the Cyber Hype." Foreign Affairs. https://www.foreignaffairs.com/articles/united-states/2013-08-14/dont-buy- cyberhype. Libicki, Martin C. 2012. Crisis and Escalation in Cyberspace. Santa Monica, Calif: Rand Corp. Lieber, Keir A. 2000. "Grasping the Technological Peace." International Security 25 (1): 71. ———.2005. War and the Engineers. Ithaca, N.Y: Cornell Univ. Press.

171

———.2014. "The Offense-Defense Balance and Cyber Warfare." In Cyber Analogies, edited by John Arquilla and Emily Goldman, 96-107. Monterrey: Naval Postgraduate School. Liff, Adam P. 2012. "Cyberwar: A New 'Absolute Weapon'? the Proliferation of Cyberwarfare Capabilities and Interstate War." Journal of Strategic Studies 35 (3): 401. Lin, Herbert. 2012. "Escalation Dynamics and Conflict Termination in Cyberspace." Strategic Studies Quarterly 6 (3): 46. Lind, William S. 1997. "Some Doctrinal Questions for the United States Army." Military Review 77 (1): 55-78. Lindsay, Jon R. 2014. "The Impact of China on Cybersecurity: Fiction and Friction." International Security 39 (3): 7. ———. 2013. "Stuxnet and the Limits of Cyber Warfare." Security Studies 22 (3): 365- 404. Little, Richard G. 2002. "Controlling Cascading Failure: Understanding the Vulnerabilities of Interconnected Infrastructures." Journal of Urban Technology 9 (1): 109-123. Lock-Pullan, Richard. 2005. "How to Rethink War: Conceptual Innovation and AirLand Battle Doctrine1." Journal of Strategic Studies 28 (4): 679-702. Long, Jeffrey W. 1991. The Evolution of Army Doctrine: Active Defense to AirLand Battle and Beyond. Carlisle Barracks: U.S. Army Command and General Staff College. Lonsdale, David J. 2004. Nature of War in the Information Age. New York: Routledge. Lynn, John A. 1984. The Bayonets of the Republic. Urbana u.a: Univ. of Illinois Press. Lynn, William. 2010. "Defending a New Domain: The Pentagon’s Cyber Strategy." Foreign Affairs 89 (5): 97-113. Lynn-Jones, Sean M. 1995. "Offense-Defense Theory and its Critics." Security Studies 4 (4): 660-691. Lyon, David. 1988. The Information Society. New York: John Wiley and Sons. Lytle, Scott. 1958. "Robespierre, Danton, and the Levee En Masse." The Journal of Modern History 30 (1): 325-337. Macdonald, Julia and Jacquelyn Schneider. 2017. "Presidential Risk Orientation and Force Employment Decisions." Journal of Conflict Resolution 61 (3): 511-536. Mackenzie, S. P. 1997. Revolutionary Armies in the Modern Era. London: Routledge. Mandel, Robert. 1977. "Political Gaming and Foreign Policy Making during Crises." World Politics 29 (4): 610-625..

172

Mandelbaum, Michael. 1981. The Nuclear Revolution. Cambridge: Cambridge Univ. Press. Markoff, John. 1999. "Ideas & Trends: Blown to Bits; Cyberwarfare Breaks the Rules of Military Engagement." The New York Times, Oct 17,. Marks, Steven. 1991. Road to Power. Ithaca: Cornell University Press. Martin, Craig and Shepard, Kip. "Supply Chain Risk Management." Fire Eye., accessed March 13, 2017, https://www.fireeye.com/content/dam/fireeye- www/global/en/current-threats/pdfs/rpt-best-practices-in-cyber-supply-chain-risk- management.pdf. Martin, Frederick Thomas. 1999. Top Secret Intranet: How U.S. Intelligence Built Intelink—the World’s Largest, most Secure Network. Upper Saddle River: Prentice Hall. Maurer, John H. 2014. Churchill and the Strategic Dilemmas before the World Wars. New York: Taylor and Francis. Mazarr, Michael J. 1994. The Revolution in Military Affairs: A Framework for Defense Planning. McGraw, Gary. 2013. "Cyber War is Inevitable (Unless we Build Security in)." Journal of Strategic Studies 36 (1): 109. McHugh, Francis. 1966. Fundamentals of Wargaming. Newport: United States Naval War College. McNeill, William Hardy. 1982. The Pursuit of Power. Oxford: Blackwell. Mearsheimer, John J. 2001. The Tragedy of Great Power Politics. New York: Norton. Metz, Steven. 2000. Armed Conflict in the 21st Century. Carlisle: Strategic Studies Inst. Metz, Steven K. and James O. Kievit. 1995. Strategy and the Revolution in Military Affairs: From Theory to Policy. Meyer, Stephen M. 1986. The Dynamics of Nuclear Proliferation. Chicago: Univ. of Chicago Press. Millett, Allan R. and Williamson Murray. 2010. Military Effectiveness: Volume 2, the Interwar Period. 2nd ed. New York: Cambridge University Press. Mintz, Alex, Steven B. Redd, and Arnold Vedlitz. 2006. "Can we Generalize from Student Experiments to the Real World in Political Science, Military Affairs, and International Relations?" The Journal of Conflict Resolution 50 (5): 757-776. Mintz, Alex, Yi Yang, and Rose McDermott. 2011. "Experimental Approaches to International Relations." International Studies Quarterly 55 (2): 493-501. Mitchell, Paul T. and Jeffrey Bardzell. 2008. Network Centric Warfare : Coalition Operations in the Age of US Military Primacy. Adelphi Series. Vol. 385. Florence:

173

Routledge.. Morgan, Patrick M. 2010. "Applicability of Traditional Deterrence Concepts and Theory to the Cyber Realm--Patrick M. Morgan." In Proceedings of a Workshop on Deterring Cyberattacks, 55-76: National Academies Press. Morris, Terry S., Martha VanDriel, Bill Dries, Jason Perdew, Richard Schulz, and Kristin Jacobson. 2015. "Securing the Operational Access: Evolving the Air-Sea Battle Concept." National Interest, February 11,. Mortimer, John J., Jr. 2013. "Tactics, Strategy, and Battlefield Formation during the Hundred Years War: The Role of the Longbow in the "Infantry Revolution"."ProQuest Dissertations Publishing. Mulrine, Anna. 2011. "CIA Chief Leon Panetta: The Next Pearl Harbor could be a Cyberattack." The Christian Science Monitor, Jun 9,. Muradian, Vago. 2011. "Adm. Michael Mullen." Defense News, Jul 11, 30. Murray, Williamson. 1996. Military Innovation in the Interwar Period. Cambridge: Cambridge Univ. Press. Nalebluff, Barry. 1987. Brinkmanship and Nuclear Deterrence. Princeton: Princeton University Press. Nitze, Paul J. 1976. "Deterring our Deterrent." Foreign Policy (25): 195-210. Nye, Joseph Jr and William Owens. 1996. "America's Information Edge." Foreign Affairs 7 (52): 20-36. Nye, Joseph S. 2011. Nuclear Lessons for Cyber Security?. Maxwell: Air University Press. Nye, Joseph S. 2017. “Deterrence and Dissuasion in Cyberspace.” International Security 41 (3): 44-71. O'Hanlon, Michael E. 2000. Technological Change and the Future of Warfare. Washington, DC: Brookings Inst. Press. "Oil and Gas Industry Increasingly Hit by Cyber-Attacks." Security Week News., last modified January 14, accessed March 13, 2017, http://www.nytimes.com/2012/10/24/business/global/cyberattack-on-saudi-oil- firm-disquiets-us.html. Organic Over-the-Horizon Targeting for the 2025 Surface Fleet. 2015: Naval Postgraduate School. Owen, Taylor. 2015. Disruptive Power. Oxford Studies in Digital Politics. New York: Oxford University Press. Owens, William A. and Edward Offley. 2000. Lifting the Fog of War. New York: Farrar, Straus, Giroux.

174

Paganini, Pierluigi. "GPS Vulnerability Exploitable to Control the Route of a Vessel." Securityaffairs.co., last modified July 27, accessed 13 March, 2017, http://securityaffairs.co/wordpress/16698/hacking/gps-vulnerability- exploitable-to-control-the-route-of-a-vessel.html. ———. "Hacking Ships: Maritime Shipping Industry at Risk." Securityaffairs.com., last modified March 31, accessed March 13, 2017, http://securityaffairs.co/wordpress/35504/hacking/hacking-maritime-shipping- industry.html. ———. "Modern Railroad Systems are Vulnerable to Cyber Attacks." Securityaffairs.com., last modified January 2, accessed March 13, 2017, http://securityaffairs.co/wordpress/43196/hacking/railroad-systems- vulnerabilities.html. Palmer, Robert R. 2014. The Age of the Democratic Revolution. Princeton: Princeton University Press. Parker, Geoffrey. 1996. The Military Revolution: Military Innovation and the Rise of the West, 1500-1800. Cambridge: Cambridge University Press. Paul, T. V. 2000. Power Versus Prudence : Why Nations Forgo Nuclear Weapons. Foreign Policy, Security and Strategic Studies. Montréal: McGill-Queen's University Press. http://lib.myilibrary.com?ID=285883. Pearson, Ivan Lester Glen. 2010. In the Name of Oil. Brighton [u.a.]: Sussex Acad. Press. "Pentagon's Air-Sea Battle Plan Explained.;Finally." 2012b.Wired, August 6,. Periroth, Nicole. 2012. "In Cyberattack on Saudi Firm, U.S. Sees Iran Firing Back." New York Times, October 23,. http://www.nytimes.com/2012/10/24/business/global/cyberattack-on-saudi-oil- firm-disquiets-us.html. Perla, Peter. 1990. The Art of Wargaming. Annapolis: Naval Institute Press. Perla, Peter and E. D. McGrady. 2011. "Why Wargaming Works." Naval War College Review 64 (3): 111. Perry, William. 1989. "Defense Investment Strategy." Foreign Affairs 68 (2): 72-92. ———.1980. The Department of Defense Statement on Technology and Military Manpower. Posen, Barry R. 1991a. "Crisis Stability and Conventional Arms Control." Daedalus 120 (1): 217-232. ———. 1991b. Inadvertent Escalation. Ithaca: Cornell Univ. Press. ———. 1993. "Nationalism, the Mass Army, and Military Power." International Security 18 (2): 80.

175

———. 1984. The Sources of Military Doctrine. 1. publ. ed. Ithaca: Cornell Univ. Press. Posen, Barry R. and Stephen Van Evera. 1983. "Defense Policy and the Reagan Administration: Departure from Containment." International Security 8 (1): 3-45. Powell, Robert. 1989. "Crisis Stability in the Nuclear Age." The American Political Science Review 83 (1): 61-76. Quester, George H. 1977. Offense and Defense in the International System. Ithaca: Cornell University Press. Quester, George Herman. 1973. The Politics of Nuclear Proliferation. Baltimore: Johns Hopkins University Press. Qureshi, Lubna Z. 2008. Nixon, Kissinger, and Allende. New York: Rowman and Littlefield. Reed, Jeff. "Britain Vs. Germany, Coal Vs. Oil—The Coming of the Great War." Oilpro.com., accessed March 13, 2017, http://oilpro.com/post/4796/britain-vs- germany-coal-vs-oil--the-coming-of-the-great-war. Reiter, Dan and Allan C. Stam. 1998. "Democracy and Battlefield Military Effectiveness." The Journal of Conflict Resolution 42 (3): 259-277. Remarks by Deputy Secretary Work and General Jean-Paul Palomeros on "the Third U.S. Offset Strategy and its Implications for Partners and Allies" at a Forum Hosted by the Center for a New American Security, Washington, D.C. 2015c. Lanham: Federal Information & News Dispatch, Inc. Remarks by Secretary Panetta on Cybersecurity to the Business Executives for National Security, New York City. 2012c. Lanham: Federal Information & News Dispatch, Inc. Richelson, Jeffrey. 1983. "PD‐59, NSDD‐13 and the Reagan Strategic Modernization Program." Journal of Strategic Studies 6 (2): 125-146. Rid, Thomas. 2012. "Cyber War Will Not Take Place." Journal of Strategic Studies 35 (1): 5-32. Risch, Erna. 1981. Supplying Washington's Army. Washington, D.C.: U.S. Army. Rodriguez, Joe Fitzgerald. 2016. "’You Hacked’ Appears at Muni Stations as Fare Payment System Crashes." San Francisco Examiner, November 26,. http://www.sfexaminer.com/hacked-appears-muni-stations-fare-payment- system-crashes/. Romjue, John L. 1984. From Active Defense to AirLand Battle : The Development of Army Doctrine, 1973-1982. Fort Monroe: United States Army Training and Doctrine Command. http://catalog.hathitrust.org/Record/002603826. Rosen, Stephen Peter. 1995. "Military Effectiveness: Why Society Matters." International Security 19 (4): 5-31.

176

———. 1991. Winning the Next War. Ithaca: Cornell Univ. Press. Rosenberg, Hans. 1968. Bureaucracy, Aristocracy and Autocracy. Harvard Historical Monographs. Vol. 34. Cambridge, MA: Harvard Univ. Press. Rublee, Maria Rost. 2009. Nonproliferation Norms. Athens: Univ. of Georgia Press. Rustici, Ross M. 2011. "Cyberweapons: Leveling the International Playing Field." Parameters 41 (3): 36-47. Safran, Nadav. 1977. "Trial by Ordeal." International Security 2 (2): 133-170. Sagan, Scott D. 1986. "1914 Revisited: Allies, Offense, and Instability." International Security 11 (2): 151-175. Salay, David. 1975. "The Production of Gunpowder in Pennsylvania during the American Revolution." The Pennsylvania Magazine of History and Biography 99 (4). Sanger, David E. 2012. "Obama Order Sped Up Wave of Cyberattacks Against Iran." New York Times, Jun 1. Sanger, David and Schmitt, Eric. "Russian Ships Near Data Cables are Too Close for U.S. Comfort." New York Times (Online). New York Times Company, last modified Oct 25. Schelling, Thomas C. 2008. Arms and Influence. New Haven: Yale Univ. Press. Schneider, Jacquelyn. "Beyond Strictly Material Assessments of Weapons' Effectiveness: Examining the Effect of Fear in Threat Assessment." February 2014. Paper presented at International Studies Association Annual Meeting. New Orleans. Schneider, Jacquelyn. 2016. "Digitally-Enabled Warfare: The Capability/Vulnerability Paradox." Center for a New American Security Reports. "Select Enemy. Delete." 1997.The Economist 342 (8007): 21. Senft, Michael. "Convergence of Cyberspace Operations and Electronic Warfare Effects." The Cyber Defense Review., last modified January 4, accessed March 13, 2017, http://www.cyberdefensereview.org/2016/01/04/convergence-of-cyberspace- operations-and-electronic-warfare-effects/. Shanahan, William O. 1945. Prussian Military Reforms, 1786-1813. Studies in History, Economics and Public Law. no. 520. New York: Columbia University Press. Sharma, Amit. 2010. "Cyber Wars: A Paradigm Shift from Means to Ends." Strategic Analysis 34 (1): 62-73. Shimshoni, Jonathan. 1990. "Technology, Military Advantage, and World War I." International Security 15 (3): 187-215. Showalter, Dennis E. 2001. "The Prusso-German RMA, 1840-1871." In The Dynamics of Military Revolution 1300-2050, 92-113. Cambridge: Cambridge University Press.

177

Singel, Ryan. 2010. "Richard Clarke's Cyberwar: File Under Fiction." Wired, April 22,. Singer, Peter W. and Allan Friedman. 2014. Cybersecurity and Cyberwar. Oxford: Oxford Univ. Press. Skinner, Douglas W. 1988. Airland Battle Doctrine. Alexandria: Center for Naval Analyses. Slayton, Rebecca. 2013. Arguments that Count. Inside Technology Series. Cambridge: Mit Press. Slayton, Rebecca. 2017. "What is the Cyber Offense-Defense Balance? Conceptions, Causes, and Assessment." International Security 41 (3): 72-109. Smit, Wim A. 1992. Military Technological Innovation and Stability in a Changing World. Amsterdam: VU Univ. Press. Smith, Edward A. 2003. Effects Based Operations: Applying Network Centric Warfare in Peace, Crisis, and War. Smith, G. E., K. J. Watson, W. H. Baker, and J. A. Pokorski II. 2007. "A Critical Balance: Collaboration and Security in the IT-Enabled Supply Chain." International Journal of Production Research 45 (11): 2595-2613. Smoke, Richard. 1977. War. Cambridge, Mass: Harvard Univ. Pr. Snyder, Jack. 1984. "Civil-Military Relations and the Cult of the Offensive, 1914 and 1984." International Security 9 (1): 108-146. ———.The Ideology of the Offensive. Ithaca: Cornell Univ. Press. Snyder, R. Claire. 2003. "The Citizen-Soldier and the Tragedy of the Eighteenth Brumaire." Strategies: Journal of Theory, Culture & Politics 16 (1): 23-37. Solingen, Etel. 2007. Nuclear Logics. Princeton [u.a.]: Princeton Univ. Press. Sood, Aditya K. and Richard Enbody. 2014. "U.S. Military Defense Systems: The Anatomy of Cyber Espionage by Chinese Hackers." Georgetown Journal of International Affairs. Sorley, Lewis, ed. 2009. Press on!. Vol. 2. Fort Leavenworth: Combat Studies Institute Press. Starry, Donn A. 1983. "To Change an Army." Military Review 63: 1-25. Steinbrunner, John D. 1982. "Nuclear Decapitation." Foreign Policy: 16-28. Stephenson, Orlando W. 1925. "The Supply of Gunpowder in 1776." The American Historical Review 30 (2): 271-281. Stevenson, David. 1999. "War by Timetable? the Railway Race before 1914." Past & Present (162): 163-194. Stubbing, Richard. 1985. The Defense Program: Buildup Or Binge?.

178

Sweetman, Bill. 2015. "New Radars, IRST Strengthen Stealth-Detection Claims." Aviation Week and Space Technology, March 16,. Talmadge, Caitlin. "Assessing the Risk of Chinese Nuclear Escalation in a Conventional War with the United States." International Security. ———. 2015. The Dictator's Army : Battlefield Effectiveness in Authoritarian Regimes. Cornell Studies in Security Affairs. Ithaca: Cornell University Press. ———. 2016. Preventing Nuclear Escalation in U.S.-China Conflict. ———. 2013. "The Puzzle of Personalist Performance: Iraqi Battlefield Effectiveness in the Iran-Iraq War." Security Studies 22 (2): 180-221. Tannenwald, Nina. 2007. The Nuclear Taboo. Cambridge Studies in International Relations. 1. publ. ed. Vol. 87. Cambridge: Cambridge Univ. Press. Thayer, Bradley A. 1995. "The Causes of Nuclear Proliferation and the Utility of the Nuclear Non-Proliferation Regime." Security Studies 4 (3): 463-519. The Clinton Administration's Policy on Critical Infrastructure Protection: Presidential Decision Directive 63. 1998: Electronic Privacy Information Center. Thomas, Jordan. 2016. Email with Col Jordan Thomas, Ret., edited by Jacquelyn Schneider. Toprani, Anand. 2016. "The First War for Oil: The Caucasus, German Strategy, and the Turning Point of the War on the Eastern Front, 1942." The Journal of Military History 80 (3): 815. Turner, George Edgar. 1992. Victory Rode the Rails. Lincoln: Univ. of Nebraska Press. "Undersea Internet Cables are Surprisingly Vulnerable." WIRED. ACI Information Group, last modified Oct 29, accessed March 13, 2017, http://scholar.aci.info/view/1476b1d538930fb01bd/150b6672fbc00010007. Valeriano, Brandon and Ryan Maness. 2012. "The Fog of Cyber War: Why the Threat Doesn'T Live Up to the Hype." Foreign Affairs. https://www.foreignaffairs.com/articles/2012-11-21/fog-cyberwar. Van Creveld, Martin L. 2004. Supplying War. 2. ed. ed. New York: Cambridge Univ. Press. Van Evera, Stephen. 1999. Causes of War. 1. publ. ed. Ithaca: Cornell Univ. Press. ———.1998. "Offense, Defense, and the Causes of War." International Security 22 (4): 5-43. ———.1984. "The Cult of the Offensive and the Origins of the First World War." International Security 9 (1): 58-107. Vego, Milan. 2011. "AirSea Battle must Not Work Alone." United States Naval Institute. Proceedings, Jul 1, 64.

179

Violino, Bob. "Unseen, all-Out Cyberwar on the U.S. has Begun." Infoworld., last modified January 28, accessed March 13, 2017, http://www.infoworld.com/article/2612825/hacking/unseen--all-out-cyberwar-on-the-u-s--has-begun.html. Walt, Stephen. 2010. "Is the Cyber Threat Overblown?" Foreign Policy, March 30,. Waltz, Kenneth Neal. 1969. Man, the State and War. New York: Columbia Univ. Press. ———. 2010. Theory of International Politics. Reissued ed. Long Grove, Ill: Waveland Press. Wass de Czege, Huba. 2006. "Lessons from the Past: Making the Army's Doctrine 'Right enough' Today." Landpower Essay 6 (2): 1-24. Weber, Thomas. 1952. The Northern Railroads in the Civil War, 1861-1865. New York: King's Crown Press. Weisburd, Andrew, Watts, Clint and Berger, J. M. "Trolling for Trump: How Russia is Trying to Destroy our Democracy." War on the Rocks., last modified November 6, accessed March 13, 2017, http://warontherocks.com/2016/11/trolling-for-trump- how-russia-is-trying-to-destroy-our-democracy/. Weisgerber, Marcus. "The Increasingly Automated Hunt for Mobile Missile Launchers." Defenseone., last modified April 28, accessed 13 March, 2017, http://www.defenseone.com/technology/2016/04/increasingly-automated- hunt-mobile-missile-launchers/127864/. Wolmar, Christian. 2010. Engines of War. GB: Atlantic Books. Woloch, Iser. 1986. "Napoleonic Conscription: State Power and Civil Society." Past & Present (111): 101-129. Yergin, Daniel. 1991. The Prize. New York [u.a.]: Free Press. Zenko, Micah. "Admiral McMullen: Thank You and Farewell." Cfr.org., last modified September 29, accessed March 13, 2017, http://blogs.cfr.org/zenko/2011/09/29/admiral-michael-mullen-farewell-and- thank-you/. Zetter, Kim. 2012. "Hackers Breached Railway Network, Disrupted Service." Wired, January 24,.

180