Paul Scharre
When Jon Snow comes face-to-face with the Night King and his army at Hardhome, it’s clear they are an enemy like nothing he has seen before. Defeating the White Walkers will require new weapons, such as dragonglass, and new tactics, like using fire against the wights, the White Walkers’ army of undead foot soldiers. But even more than that, the enemy’s inhumanity means that the fundamental nature of the conflict Jon Snow is engaged in is different from other wars against humans. The wights show no fatigue or fear, and at least at the start of season 8, the White Walkers’ aims are unknown. They seem uninterested in negotiation and immune to the political intrigue, subterfuge, and shifting alliances that are so common among the human wars of Westeros. The war Jon Snow and his allies are fighting against the White Walkers is unlike any other war they have fought in the past, if it is a war at all.
In Western military circles, it is common to speak of the ever-changing character of warfare and the immutable and unchanging nature of war.1 Warfare refers to the means and methods of fighting and is constantly in flux as militaries develop new tactics, technologies, and strategies. The essential nature of war, however, is timeless. War is a violent struggle among competing groups of people for political power.2 Or as the Prussian strategist Carl von Clausewitz wrote, “War therefore is an act of violence to compel our opponent to fulfil our will.”3 Even while the means and methods of fighting (the character of warfare) change over time, all wars have common characteristics that stem from its unchanging, underlying nature. One of the U.S. Marine Corps’ fundamental doctrine publications, Warfighting, states in its chapter on the nature of war, “The essence of war is a violent struggle between two hostile, independent, and irreconcilable wills, each trying to impose itself on the other. War is fundamentally an interactive social process.”4
It may seem strange to describe war as an “interactive social process,” but as practitioners and observers know, war is a human endeavor.5 It is a struggle between people, even if the tools they use evolve over time. Because of the human nature of war, some elements of war are constant. Wars are motivated by, as Thucydides wrote some 2,400 years ago, some combination of “fear, honor, and interest.”6 Once underway, violence, uncertainty, and chance rule the battlefield. The life of the warfighter across all ages has been marked by danger, physical exertion, exhaustion, and suffering. And battlefields are home to what Clausewitz described as friction, which is best summed up in his statement, “Everything is very simple in War, but the simplest thing is difficult.”7 Friction is the message that didn’t make it through to a unit, the supply train that got lost in the night, the vehicle that got stuck in the mud, and the battle that was lost as a result.
There is great value in differentiating between continuity and change in war. War is the most extreme of human endeavors, taxing individuals and even entire nations to their maximum limit with the gravest of consequences on the line. Yet militaries, in truth, rarely fight wars, or at least not big ones. Most of the time—thankfully—nations are at peace and militaries occupy their time with training and preparing for war. Knowing how best to prepare, however, is an immense intellectual challenge, akin to training a professional sports team for a game that is played only once a generation, where the rules are unknown and constantly changing and where losing a play could mean death. Military scholars have quite reasonably sought to distill from the millennia-long history of war some lessons to guide their planning.
There is danger, however, in leaping from the well-founded observation that wars in the past have always had certain commonalities to a future-looking statement that the nature of war can never change. If the nature of war is to mean anything, it should be possible to define conditions under which it might change, even if those conditions seem unlikely. To assert simply that it cannot change, no matter what, is not a scientific statement—it’s a faith-based statement. Then military thinkers are no longer studying war, but war becomes a religion and Clausewitz a prophet. If scholars of war want to truly understand it, then they must open their minds to the possibility that the nature of war might change.
James Mattis, retired general of the U.S. Marines and secretary of defense under President Donald Trump, has raised precisely this intriguing possibility. During a conversation with reporters about artificial intelligence, he remarked, “I’m certainly questioning my original premise that the fundamental nature of war will not change. You’ve got to question that now. I just don’t have the answers yet.”8 Mattis’s statement is a bold proposition in Western military circles and one that challenges a tightly held dogma about the unchanging nature of war. Yet there are a number of disruptive technologies, including artificial intelligence and synthetic biology, that are certain to change the character of warfare and quite possibly raise intriguing questions about the nature of war. It’s a bold assertion, but one worth considering. What would it mean for the nature of war to change?
Game of Thrones gives us a good vehicle to consider this question. We can examine a conflict—albeit a fictional and fantastical one—in which the nature of war might be different. So what is the nature of the conflict Jon Snow is fighting against the White Walkers?
There are many elements of the fight against the White Walkers that illustrate the changing character of warfare. The wights are already dead, and so they are unbothered by wounds that would kill a human, like an arrow to the head. Wights move with an unnatural speed. White Walkers’ weapons will shatter a traditional sword. And the White Walkers can be harmed only with dragonglass or Valyrian steel (and possibly dragon fire). Jon Snow takes these lessons learned at Hardhome and other battles to refine his army’s tactics and weapons. These are examples of the changing character of warfare in action.
But the White Walkers and their wight soldiers present a much deeper and more fundamental challenge to the nature of war—they are no longer human. Again, war—at least war as we conceive of it now—is a human endeavor. This is not merely a minor definitional issue, arguing that all wars have in the past been between humans. The fact that war is fought by humans against other humans is foundational to its nature. War is political—“a continuation of policy by other means,” as Clausewitz described it.9 But what are the White Walkers’ aims? Are they also motivated by some combination of “fear, honor, and interest”? Do they seek power in a sense that the Westerosi lords would understand? At the outset of season 8, we simply do not know. It is possible that they have humanlike aims and motivations. It is also possible that they have no more desires than a virus has.
Even if their motivations could be understood in human terms, the inhuman nature of the White Walkers and the wights means that many commonalities in wars against people do not apply. Strategy often involves anticipating an enemy’s move and using deception to trick an enemy. The ancient Chinese strategist Sun Tzu stated, “All warfare is based on deception.”10 Anticipation and deception require a working theory of mind in order to predict how an enemy thinks. While White Walkers are clearly sentient, it isn’t clear how similar their thought processes are to humans. Some fan theories speculate that they can even see the future, which would fundamentally change essential elements of warfare, like deception.
In battle, the undead wights are immune to many of the challenges that human warfighters face. The U.S. Marine Corps’ Warfighting manual observes that “individual conscience, emotion, fear, courage, morale, leadership, or esprit” are important, intangible “moral forces” in war that exert an even greater influence on the outcome of war than physical forces.11 On fear in particular, the manual states, “Since war is a human phenomenon, fear, the human reaction to danger, has a significant impact on the conduct of war. Everybody feels fear. Fear contributes to the corrosion of will. Leaders must foster the courage to overcome fear, both individually and within the unit. Courage is not the absence of fear; rather, it is the strength to overcome fear.”12
The wights do not feel fear, fatigue, conscience, or emotion. Their morale cannot be sapped; their esprit de corps cannot be broken. They are an unthinking and unfeeling wave of partially decayed flesh, more like a natural force than a human opponent. When Jon and his band of raiders are trapped beyond the Wall, there is never a sense that the wights might tire of their siege or succumb to fatigue or exhaustion. The Marine Corps manual points out that (for humans) “it is physically impossible to sustain a high tempo of activity indefinitely. . . . The tempo of war will fluctuate from periods of intense combat to periods in which activity is limited to information gathering, replenishment, or redeployment.”13 The wights do not need to rest, however. They are unburdened from supply lines or the need to tend to their wounded and bury their dead. They are already dead. These moral forces of war still weigh on Jon and his comrades and are an aspect of war for them, but not for their enemy. The nature of this war is different.
The war against the White Walkers is a kind of war, but it is different than any of the wars among the humans of Westeros. The inhuman nature of the enemy means that some elements of the nature of war have changed. Definitionally, this presents something of a challenge. If it does not meet all the criteria that define war, then what is it? The most sensible approach would be to define it as another kind of war, broadening the horizon for what war is into two categories—human wars and wars against nonhumans.
Short of a White Walker invasion, it is possible to envision scenarios in the real world where evolutions in technology do indeed truly lead to changes in the nature of war itself. Five possibilities are outlined below. None of these possibilities are necessarily likely, either in the short term or the long term. They are most certainly not predictions. Rather, they are merely thought experiments to help conceive of developments that might require a rethinking of the nature of war.
The most obvious candidate for a nonhuman adversary in war would be advances in artificial intelligence that could lead to the creation of an artificial entity that would pursue its own aims in a conflict. These aims need not be aligned with human desires or even appear sensible from a human standpoint. In science fiction, AI is often depicted as “waking up” and turning against humans, but that need not be the case. AI threats could come from nonsentient but intelligent AI systems, like an adaptive cyberweapon that is capable of resource acquisition, planning, and causing harm and is pursuing its own aims. While such an AI system would originally be designed by humans, it could slip out of control due to an accident. In fact, designing goal-driven intelligent systems that stay within the bounds of intended human design is an unsolved problem in machine learning and artificial intelligence.14
Synthetic biology offers the potential to alter humans directly, leading to the prospect of wars in the future that are still fought by humans, but humans who are unlike those who fought wars in the past. While still science fiction today, it is possible to imagine a world in which biological advances empower humans to fight without fatigue, exhaustion, conscience, fear, or mental weakness. Such a development would change the nature of war, humanity’s relationship with it, and even humanity itself.
War is not merely violence. It is organized violence among competing groups for political aims. Social media is a new tool of organizing humans, and its introduction is as profound as that of the printing press. Already, social media has demonstrated a radically democratized information landscape where ideas and memes can spread virally without any hierarchical human organization to promote them. This has, among other things, led to a transformation in the nature of terrorist threats. The Islamic State is both an organization and a decentralized global movement that can inspire people around the globe through social media to carry out lone-wolf attacks on their own. The organization can be destroyed. Defeating the ideas is much harder. In the space of ideas (but not yet violence), competing ideologies and memes spread and evolve on social media in cycles of reaction and counterreaction. While organizations and groups may participate in these cycles of ideas, generating hashtags and arguments, they do not control the spread of ideas. No one does.
Changes in the nature of war are unlikely at the current state of social media, but we are only at the beginning of a new age of human communication. It would be unwise to anticipate that social media in its current incarnation will remain static. As digital technology evolves, the depth and breadth of human communication will continue to evolve as well. It is possible to conceive of future forms of democratized human communication that could lead to viral outbreaks of violence by individuals radicalized by social media. If this violence caused reactions and counterreactions, it is conceivable that it might occur on the scale of what we think of as war, particularly if other advances in future technology increase the destructive potential of individuals (for example, if advances in synthetic biology made it possible for individuals to create virulent pathogens in a garage that could target certain racial or ethnic groups). Such an activity would have many characteristics of war—violence, uncertainty, suffering, friction, political motivations, and human aims—but not the organization that has characterized wars in the past. This could make it uniquely difficult to control or end such wars, since it would not be a war among any groups or entities as they have traditionally been conceived—such as nation-states, tribes, religions, or paramilitary organizations. Rather, it would be a violent struggle for political supremacy among competing ideas. Individual humans would be both the authors and agents of those ideas, but there would be no organizations to sit at a negotiating table or declare an end to the war. The nature of such a war would be different than that of wars in the past, driven by an unprecedented change in human organization and communication.
The twentieth century saw the advent of nuclear weapons whose destructive power eclipsed human understanding, and the twenty-first century has the potential for even more transformative weapons in artificial intelligence and synthetic biology. It is possible to imagine weapons so destructive that they render politics irrelevant, since everyone would be dead. A hypothesized doomsday device that destroyed humanity would be such a weapon. There could be no rational policy aims for actually using such a weapon, since there would be no political outcome to achieve after its use. There might be reasons to threaten to use such a weapon and even to brandish one, as a threat of mutual suicide, but none for its use. Many wars have been fought in which both sides came out worse than before, but a war that was certain to obliterate both sides would be a war unlike any other. It would shift conflict to coercion and brinkmanship beforehand, a contest not in violence but in risk-taking. The nature of this contest would have some similarities to war but would be different in other ways. And if such a war were actually fought, it would be unprecedented—a war without any purpose.
Nuclear weapons flirt with this threshold of destructiveness, with some defense intellectuals arguing that the only purpose of nuclear weapons is to prevent the other side from using them.15 Certainly, no rational policy aims could be achieved in a mass exchange of nuclear weapons among superpowers. Even more destructive weapons are conceivable. During the Cold War one conceptualized weapon was a salted nuke, which would leave radioactive fallout that could render an area uninhabitable for generations and an arsenal of which could extinguish humanity. Other potential doomsday devices include bioengineered plagues or AI weapons.
Another conceivable development in weapons could be automation that led to automatic war without political aims—that is to say, a set of machine-enabled reactions and counterreactions that led to a war without human intent. Humans would have designed the system, but once in place, the system would spring into action of its own accord. If everything works according to plan, then the automation would be the embodiment of human intent. But if things were to go awry—if this automated network of reactions were to interact with the enemy or the real world in some unexpected way leading to a reaction that was not intended by humans—then the result would be something else entirely: a war without political aims.
While mishaps from malfunctioning machines are a part of everyday life in the form of software bugs and frozen computer screens and occasionally lead to fatal accidents, such as from self-driving cars, large-scale mishaps are rare. Few domains are so automated that humans cannot step in and correct errors. Stock trading is a rare exception, and machine-driven accidents like flash crashes have come hand in hand with the rise of algorithmic trading.16 Should militaries automate their forces to a similar degree, perhaps driven by an arms race in speed as was the case in financial markets, then a comparable accident could result—a flash war.
Once underway, humans may be able to regain control of their systems and shut them down, but the damage could already be too great. An opponent may not be willing to ignore an incident simply because “the machines did it.” With passions and tensions inflamed, machine-generated war could give way to a more traditional human war. Or a machine-initiated war could simply subside, a spasm of apolitical violence. Such a concept is not entirely new. In his escalation ladder of levels of war, nuclear strategist Herman Kahn put at the top a “spasm or insensate war,” a nuclear strike delivered by virtue of automatic protocols put in place ahead of time by leadership but devoid of considered intent or political purpose. Kahn described such a war as “blind and irrational.”17 War without political purpose is not war as it is traditionally understood, and an insensate war would be of a fundamentally different nature than wars in the past.
None of these five visions presented above is necessarily likely, but defense thinkers should keep their minds open to the possibility that future technological developments might change the nature of war. To believe otherwise is to willfully blind oneself to possible risks. The future is, as always, uncertain. In his essay “War—Continuity in Change, and Change in Continuity,” the strategist Colin Gray remarked, “We know everything there is to know about war, unsurprisingly, since we have variable access to at least 2,500 years of bloody history. But we know nothing, literally zero, for certain about the wars of the future, even in the near-term.”18
Ironically, Gray goes on in his essay to reiterate the claim that the nature of war is immutable. But we cannot know that. Scholars would be well served to keep in mind Gray’s observation of our profound uncertainty about the future. For thousands of years, the nature of war—a violent clash of wills between opposing groups of people for political power—has not changed. Might it change in the future? It is possible. Jon Snow’s war against the White Walkers is a vivid illustration of the possibility of a war of a fundamentally different nature than wars in the past.
1. For example, see Colin S. Gray, “War—Continuity in Change, and Change in Continuity,” Parameters 40, no. 2 (Summer 2010): 5–13; U.S. Marine Corps, Warfighting, Marine Corps Doctrinal Publication 1 (Washington DC: U.S. Government Printing Office, 1997); and Christopher Mewitt, “Understanding War’s Enduring Nature alongside Its Changing Character,” War on the Rocks, January 21, 2014, https://warontherocks.com/2014/01/understanding-wars-enduring-nature-alongside-its-changing-character/.
2. While the political nature of war is taken as an unquestioned assumption in contemporary Western defense circles, it is worth noting that there are dissenting points of view. Historian John Keegan opens his sweeping tome A History of Warfare by stating, “War is not the continuation of policy by other means. The world would be a simpler place to understand if this dictum of Clausewitz’s were true. . . . War embraces much more than politics: . . . it is always an expression of culture, often a determinant of cultural forms, in some societies the culture itself.” John Keegan, A History of Warfare (New York: Alfred A. Knopf, 1993), 3, 12.
3. Carl von Clausewitz, On War, trans. James John Graham (London: N. Trübner, 1873), bk. 1, chap. 1, available online at the Clausewitz Homepage, https://www.clausewitz.com/readings/OnWar1873/BK1ch01.html#a.
4. U.S. Marine Corps, Warfighting, 3.
5. For example, see H. R. McMaster, “The Pipe Dream of Easy War,” New York Times, July 20, 2013, https://www.nytimes.com/2013/07/21/opinion/sunday/the-pipe-dream-of-easy-war.html.
6. Thucydides, The Peloponnesian War, trans. Richard Crawley (London: J. M. Dent; New York: E. P. Dutton, 1910), bk. 1, chap. 76, online at Perseus, http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%3A1999.01.0200%3Abook%3D1%3Achapter%3D76%3Asection%3D2.
7. Clausewitz, On War, bk. 1, chap. 7.
8. Aaron Mehta, “AI Makes Mattis Question ‘Fundamental’ Beliefs about War,” C4ISRNET, February 17, 2018, https://www.c4isrnet.com/intel-geoint/2018/02/17/ai-makes-mattis-question-fundamental-beliefs-about-war/.
9. Clausewitz, On War, bk. 1, chap. 1.
10. Sun Tzu, The Art of War, trans. Lionel Giles, chap. 1, online at http://classics.mit.edu/Tzu/artwar.html.
11. U.S. Marine Corps, Warfighting, 16.
12. U.S. Marine Corps, Warfighting, 15.
13. U.S. Marine Corps, Warfighting, 10.
14. Dario Amodei et al., “Concrete Problems in AI Safety,” preprint, submitted July 25, 2016, 4, https://arxiv.org/pdf/1606.06565.pdf; Stephen M. Omohundro, “The Basic AI Drives,” Self-Aware Systems, 9, https://selfawaresystems.files.wordpress.com/2008/01/ai_drives_final.pdf; Stuart Russell, “Of Myths and Moonshine,” Edge, November 14, 2014, https://www.edge.org/conversation/the-myth-of-ai#26015; and Nick Bostrom, Superintelligence: Paths, Dangers, Strategies (Oxford: Oxford University Press, 2014).
15. Associated Press, “McNamara Calls on NATO to Renounce Nuclear Arms,” New York Times, September 15, 1983, https://www.nytimes.com/1983/09/15/world/mcnamara-calls-on-nato-to-renounce-nuclear-arms.html.
16. U.S. Commodity Futures Trading Commission and U.S. Securities and Exchange Commission, Findings Regarding the Market Events of May 6, 2010 (Washington DC, September 30, 2010), 2, http://www.sec.gov/news/studies/2010/marketevents-report.pdf; Maureen Farrell, “Mini Flash Crashes: A Dozen a Day,” CNNMoney, March 20, 2013, http://money.cnn.com/2013/03/20/investing/mini-flash-crash/index.html; Matt Egan, “Trading Was Halted 1,200 Times Monday,” CNNMoney, August 24, 2015, http://money.cnn.com/2015/08/24/investing/stocks-markets-selloff-circuit-breakers-1200-times/index.html; Todd C. Frankel, “Mini Flash Crash? Trading Anomalies on Manic Monday Hit Small Investors,” Washington Post, August 26, 2015, https://www.washingtonpost.com/business/economy/mini-flash-crash-trading-anomalies-on-manic-monday-hit-small-investors/2015/08/26/6bdc57b0-4c22-11e5-bfb9-9736d04fc8e4_story.html?utm_term=.c12bdb13b4e2.
17. Herman Kahn, On Escalation: Metaphors and Scenarios (London: Pall Mall, 1965), 194.
18. Gray, “War—Continuity in Change, and Change in Continuity,” 5.