Regulating Weapons: An Aristotelian Account

Abstract

Regulating war has long been a concern of the international community. From the Hague Conventions to the Geneva Conventions and the multiple treaties and related institutions that have emerged in the twentieth and twenty-first centuries, efforts to mitigate the horrors of war have focused on regulating weapons, defining combatants, and ensuring access to the battlefield for humanitarians. But regulation and legal codes alone cannot be the end point of an engaged ethical response to new weapons developments. This short essay reviews some of the existing ethical works on lethal autonomous weapon systems (LAWS), highlighting how rule- and consequence-based accounts fail to provide adequate guidance for how to deal with them. I propose a virtue-based account, which I link up with an Aristotelian framework, for how the international community might better address these weapons systems.

Is it possible to limit the horrors of war by regulating the tools used to wage it? The long history of the just war tradition and the more recent history of international criminal law provide ample evidence of efforts to do so. Multiple declarations, statements, and treaties have sought to regulate the weapons of war. One of the most well-known early examples came from the medieval church's attempt to ban the crossbow. Proclamations by popes Urban II in 1097 and Innocent II in 1139 declared the crossbow to be anathema to God.¹ In more recent history, chemical weapons were banned prior to World War I. Their use during that war only reinforced the taboo associated with them, resulting in a ban that has been surprisingly robust in the many years since.² Efforts to ban or regulate other weapons and technologies of warfare litter the landscape of military history.

Unfortunately, these formal prohibitions have a checkered history of success. Militaries continued to use the crossbow despite the bans from the Catholic Church, and its demise only came about after gunpowder weapons replaced it in European conflicts. The recent use of chemical weapons in the Syrian conflict suggests that the “taboo” against them may not be as strong as we would like.³ To declare a weapon illegal does not necessarily mean it will not be used. Of course, this does not mean we should abandon efforts to ban such weapons, nor does it mean we cannot regulate them. Instead, it suggests that regulating weapons requires something more than formal declarations. One way to address this problem is to look beyond the rules themselves and explore in some depth how rules are and should be made. Thinking about legislation and the ethical and political principles that we ought to consider in rule making leads to some possible alternatives. Because rules are made by people within institutional contexts, we might consider who is involved in rule making and how we might make it more inclusive. The role of experts or affected stakeholders, for instance, can play an important part in how we make rules.

This essay takes this basic point and uses it to critically evaluate how we might regulate lethal autonomous weapon systems (LAWS). I define LAWS as weapon systems that are designed to act without human intervention, or at least to minimize that human intervention. I first introduce some of the existing ethical debates around these weapons. The works on which I focus highlight some of the limitations of simply focusing on the rules. I turn to the virtue ethics tradition to help us think more critically about these systems. One critique of virtue ethics, however, is that by focusing on moral character, we either ignore or discount rules and laws. To address this, I turn to the ancient-Greek philosopher Aristotle, a theorist of virtue who also has much to tell us about how laws are made. I conclude with a set of principles that could inform rule making concerning LAWS in the future.

Ethical Debate around LAWS

Many philosophers (and lawyers, sociologists, and even fiction writers) have proposed ways to normatively evaluate LAWS.⁴ In order to move toward the Aristotelian account I propose here, it is helpful to review some of the existing debates. Admittedly, there is a great deal more out there concerning these matters than I can address in this short piece. The works I highlight here, however, lay out some of the contours of the existing debates and lead toward the position I propose.

Broadly, international ethics falls into three categories. Deontological approaches to ethics focus on rules and rule following and are often closest to international law. Consequentialist approaches focus on the consequences of decisions, evaluating practices based on whether or not they increase the benefits to the largest group possible. Virtue ethics approaches focus more on the character of individual leaders, diplomats, and security officials, proposing that rules and consequences cannot always be determined, so individuals should be evaluated in terms of certain character traits, or virtues. Much of the ethical literature on LAWS draws on deontological and consequentialist assumptions. The Aristotelian approach I propose in this essay draws more from that virtue ethics approach, though it uses this approach less to focus on the automated systems and more on how the laws and regulations around such systems are made. That is, in this article, I aim to combine virtue ethics with deontology.

Ronald Arkin set the stage for many of the ethical debates in favor of LAWS in his 2010 article “The Case for Ethical Autonomy in Unmanned Systems.” He highlights the fact that human beings seem incapable of moderating excesses on the battlefield, due to the pressures of war, the emotional toll of battle, and a range of other issues. He focuses on the failure of militaries to respect the laws of war. In contrast to these human failings, Arkin argues, LAWS can better achieve military objectives while also protecting against atrocities and the psychological damage that combatants suffer as a result of using force. He details six reasons why LAWS will result in better consequences than humans conducting warfare.⁵ These include the fact that robots will not act to protect themselves so they are more conservative in a crisis; that they will be able to sense and monitor a wider field; that judgements will not be clouded by emotion; that they will have an avoidance of scenario fulfillment; that robots have the ability to integrate more information than their human counterparts; and that their ability to report ethical violations by human agents is high. This article has been the focus of much debate among ethicists and international lawyers, as it makes a strong argument in favor of LAWS.⁶

One important critique of Arkin and approaches like his is that they narrowly focus on consequences. This critique highlights Arkin's problematic emphasis on a narrow set of consequences as being the only way to evaluate the ethics of LAWS, instead turning to the character of the human person and how that character is shaped by and shapes the political and social activities in which they are engaged; that is, it turns to virtue ethics as an alternative. For instance, Robert Sparrow argues that only through the socially shared meanings that constitute our lives can we begin to ethically evaluate our actions.⁷ Rather than focus on consequences alone or the internal processes by which LAWS operate, he proposes “respect” as a social meaning that should structure how we understand and evaluate weapon systems of any type, seeing this as an important character trait, or virtue, that needs to be at the forefront in the use of force or the employment of any weapons system. He argues that if a weapons system does not inculcate respect for fellow human persons, it is morally flawed. In the same way that we have a shared moral sensibility that desecrating corpses on a battlefield is morally wrong, so too can we come to some shared moral understanding of how to evaluate these weapon systems.⁸

Continuing with the theme of ethical standards arising from meaning, Daniele Amoroso and Guglielmo Tamburrini compare the definitions of LAWS from the U.K. Ministry of Defence (MOD) and the U.S. Department of Defense (DOD). The British military defines LAWS as systems having the capacity for “higher level intent and direction.”⁹ Defining them as such makes for a permissive environment in which the British military can classify many systems as not being automated, so that they need not be subject to moral or legal evaluation.¹⁰ This demonstrates one of the problems with rule-based approaches; a rule that is open ended can easily be interpreted in ways that agents wish rather than binding them to standards. The authors argue that the U.S. definition is preferable—systems that can “select and engage targets without further intervention by a human operator.”¹¹ They propose three principles by which we should evaluate LAWS:

1. 1. Humans must always act as a “fail-safe” in any use of LAWS, ensuring that if there is any morally complex situation, a human becomes the key agent.

2. 2. Humans must be “accountability attractors”; that is, there must be systems designed to ensure that humans can be held accountable at some level for the use of LAWS.

3. 3. Humans must serve as “moral agency enactors” to protect the dignity of those subject to the use of such weapon systems.¹²

These principles put the human person at the center of how we evaluate these weapon systems, ensuring that there is space for accountability and agency in order to be morally justified.

One recent work that turns to a virtue ethics approach comes from Shannon Vallor. She explores not just LAWS but a whole range of technological issues in her book Technology and the Virtues. Vallor turns to the virtue ethics tradition to explore what kinds of virtues are necessary to navigate, use, and benefit from new technologies. She points out that much of the literature on the ethics of robotics and AI is either deontological (rule based) or consequentialist. Yet these two moral approaches “struggle to accommodate the constant flux, contextual variety, and increasingly opaque horizons of emerging technologies and their applications.”¹³ She argues that a virtue-oriented perspective, one that focuses on the moral character of the individual human rather than rules and/or consequences, can better adapt to such changes. More importantly, by cultivating particular virtues—such as courage, humanity, humility, and the desire to know—humans using technology can be better positioned to act in morally appropriate ways.

In another account that draws on virtue ethics, Cappuccio and colleagues, responding directly to Arkin, argue that while there is some truth to Arkin's argument, it is less true if morality is understood as the cultivation of a particular character.¹⁴ An important part of forming that character for a combatant is having to face danger and make decisions that embody virtues such as courage and wisdom. If robots, rather than humans, conduct warfare, the humans designing those robots and ultimately controlling them will lose the capacity to develop as fully human moral persons. That is, they will become more and more immune to the dangers and stresses of war, making it easier to create weapons that will reflect that more casual approach to death and destruction.

What can we draw from these different ethical accounts, which, as I note above, lay out some of the contours of the existing debates? First, many of the strongest criticisms of LAWS arise from a virtue-oriented or relational perspective rather than a deontological or consequentialist perspective. This may be because LAWS can more easily be defended in deontological and consequentialist categories. Deontologist advocates argue that LAWS can be programmed with rules that will ensure compliance with existing normative standards,¹⁵ and consequentialists can argue that AWS may well lead to fewer civilian deaths and avoid moral injury for combatants. These conclusions, however, leave out the specifically human dimension of warfare. This is not to say that warfare humanizes or makes us better people; rather, that if we try to take the human out of warfare through increased reliance on technology, those who command and control decisions to go to war can more easily wage it. In the end, a consequentialist account is ambiguous in outcome, because it may just shift the casualties onto vulnerable populations, only reducing “our” deaths at the expense of less dangerous forms of warfare that would be politically unpalatable.

Aristotle, Virtue Ethics, and LAWS

The turn to virtue ethics and the associated idea of human dignity provides an alternative, a different way of seeing the moral dilemmas that arise from LAWS. However, while I support this turn and find in virtue ethics a great deal to admire, I also recognize that the virtue ethics tradition can lead to conclusions that move us away from formal regulatory structures. If our focus is on the individual human commander or combatant and on retaining a human role in decisions to use force, we can be left in a situation where rules and consequences become a stumbling block. That is, in efforts to overcome deontological and consequentialist forms of reasoning, virtue ethicists can be led to conclude that the morally virtuous individual should not be bound by rules or consequences but should act in ways that reflect their understanding of what it means to be a good person.¹⁶

Can the virtue ethics tradition be deployed in a way that contributes to the regulation of these weapons rather than detracts from it? One place to turn to accomplish this is the philosophy of Aristotle, the ancient-Greek philosopher to whom many in the virtue ethics tradition look for inspiration. Admittedly, Aristotle is not normally associated with any kind of global orientation, which makes virtue ethics seem initially less plausible as the basis for a global regulatory framework on LAWS. Those who draw on his work tend to be oriented toward more communitarian or state-based approaches. As will become clear below, my argument does not assume a shared set of virtues for all humanity. Rather, I look to a particular global institution to find the existence of shared virtues and a potential for how virtues can develop in a deliberative process of lawmaking. Like the realist thinker Hans J. Morgenthau, I find inspiration in Aristotle's approach to linking politics and ethics rather than drawing directly on his views on the specific virtues that bring together peoples in particular communities.¹⁷

Virtues, for Aristotle, are character traits that enable the human person to be the best that he or she can possibly be. The ancient-Greek word that Aristotle uses for virtue, arete, can also be translated as “excellence,” or being the best that one can be at some activity. So, a virtuous teacher is one whose students learn, a virtuous carpenter is one who makes the best wooden artifacts, and so on. Aristotle takes this idea of excellence and applies it to the human condition; that is, he asks, what is it to be an excellent human being? For Aristotle, to be such a person is to be one who thinks and lives in community, as these are the characteristics that differentiate humans from all other animals. From these two assumptions about the human person derive the wide range of virtues that Aristotle, and many others in the virtue ethics tradition who followed, emphasized.¹⁸

While I am less interested in the specific virtues that Aristotle chooses,¹⁹ the two overarching ideas of thinking and living in community can guide us in how we might use this tradition to think about creating rules and laws to govern LAWS. Virtues arise, according to Aristotle, from living in a community that is itself well ordered; in other words, a virtuous political order. As he notes at the end of the Nicomachean Ethics,

But if one has not been reared under the right laws it is difficult to obtain from one’s earliest years the correct upbringing for virtue, because the masses, especially the young, do not find it pleasant to live temperately and with endurance. For this reason, their upbringing and pursuits should be regulated by laws, because they will not find them painful once they have become accustomed to them.²⁰

The book that follows from this one is The Politics, in which he sets out what kinds of citizens and political arrangements can best inculcate the virtues. Importantly, The Politics ends by discussing education, because it is only through education that we can create good citizens.²¹

Aristotle provides us a link between the virtue ethics tradition and politics. What we need is a political system with legislators who can create good laws that will make us better people. In looking at contemporary politics, both domestically and internationally, this certainly seems like a distant dream. Legislative bodies tend to display the worst of political life, with self-interested and ideological debates overwhelmingly the norm. Yet, despite this, legislative bodies remain perhaps the most important in any political system; it is no accident that Article I of the United States Constitution describes the powers of congress. In addition, a legislature usually is the most representative body in a political system. As Jeremy Waldron has argued, there is great dignity in legislation and legislative systems.²²

One other insight from Aristotle is relevant here. In one of his works on logic, Aristotle describes what it means to be a thinking animal. He proposes two types of knowledge, the scientific and the dialectic. The former is knowledge that comes from precise logical deductions. The other type of knowledge, which is equally important, comes from “opinions that are generally held.”²³ Commonly held does not mean uninformed but rather knowledge that derives from real life experiences of those living in community. Aristotle suggests that this type of knowledge comes from the back and forth of critique and deliberation, making it an important form of knowledge in political and social life.

Aristotle's philosophy points us to the importance of lawmaking, but it also points us to broad principles on which lawmakers should undertake their task. Laws should be framed with those two overarching human characteristics in mind—that humans are thinking animals and that humans live in purposefully designed communities. In addition, the knowledge that is most relevant for political deliberation and lawmaking is a combination of the scientific and the dialectic. These two points might seem distant from how we might regulate weapons, but they give us a hint as to the type of legislative bodies that can ensure the best for the human community. In the final section of this essay, I suggest how we might draw on Aristotle's ideas to develop some principles for making laws to regulate LAWS.

Regulating LAWS

In the previous section, I argued that legislative bodies deserve more attention and that the Aristotelian framework I have developed here demonstrates how ethics can make its way into lawmaking. One important objection can be raised at this point. While there are many different legislative bodies in domestic and even regional settings, there does not exist a single global legislative body. The United Nations General Assembly is perhaps the closest we have to one, but its resolutions are nonbinding and the fact that each state gets one vote means there is a radical misrepresentation in terms of population numbers (for instance, China is equal to Luxembourg in voting power). Though this particular body is problematic, some have argued that various other international organizations can play the role of quasi-legislative bodies.²⁴ Those who see constitutional qualities at the global level have also made a case that legislation can be found in global politics.²⁵ There exist numerous bodies that make laws and regulations of different sorts at the global level, many of which are devoted to specific issue areas.

One place where such bodies exist is around international legal treaties. For instance, the Conference of the States Parties, which meets every year around the United Nations Framework Convention on Climate Change, has provided an important yearly meeting structure in which states and other organizations come together to move debate forward around climate change. Although when it comes to the question of LAWS, there does not exist a treaty, one might consider an existing treaty and its associated monitoring body as a framework within which legislation might be possible. The Convention on Certain Conventional Weapons (CCW) was adopted in 1980 and came into force in 1983. It was designed to prevent unnecessary harm resulting from certain types of conventional weapons. Later protocols added different types of weapons, such as lasers and unexploded ordnance. As with all major treaties, an administrative structure creates space for continual meetings of the signatories along with discussions and debates about new protocols. In 2013, the signatories proposed the creation of a group of experts to explore the question of LAWS. This group issued the “Final Report” in December 2019, which sets out eleven key principles to be followed in the development and use of LAWS.²⁶ In 2021, the Sixth Review Conference around the CCW agreed that a further meeting of the experts group was warranted.²⁷

This structure is not ideal in many ways. The representatives who monitor and discuss the treaty and any additional protocols are state representatives. States are, of course, the primary lawmakers of international law through treaties. This means that their interests will be shaped by domestic priorities. It also means that the most powerful states will determine much of the debate and dialogue around the issue. But the very fact that such a structure exists and that the states that are part of this structure agreed, twice, to a meeting of experts to provide them guidance is a positive sign.

How can an institutional framework like this foster more “virtuous” deliberation and lawmaking in the Aristotelian sense of that word? The creation of the experts group was one step in that direction. Sometimes diplomats are seen as simply parroting the views of their masters. But diplomacy is a nuanced art, and it can be seen through the lens of Aristotle's ideas around deliberation and knowledge. As noted in the previous section, Morgenthau drew on an Aristotelian ideal as a way to understand how politics might function. He translated this to the global level when he concluded his famous textbook Politics among Nations by highlighting how diplomacy could lead to international peace. I have argued elsewhere that Morgenthau's emphasis on diplomacy draws in many ways on an Aristotelian framing.²⁸ So, perhaps we need to look to the regular meeting of state parties as a potential opportunity to create new legislative and regulatory structures. And, if Morgenthau is correct, to view these institutions not as a utopian ideal but rather as a sensible and even “realistic” locus for deliberative lawmaking.

Second, building on Aristotle's idea of dialectic knowledge, including more voices in the conversations around these systems would be beneficial.²⁹ Including NGOs like Article 36, which is dedicated to reducing the harm from weapons of all kinds, would be an important first step.³⁰ A second would be to include voices from those who have been victims of the use of these weapon systems, such as Pakistanis and Yemenis subject to drone strikes. These voices should not be restricted to state representatives but should rather include individuals from communities who have seen their loved ones and ways of life destroyed by such weapons. A third would be to include testimonies from individuals, both civilian and military, who are tasked with using these systems. There is evidence to suggest that individuals who control drones, sometimes civilians, are subject to as much if not more moral injury than combatants who use force in person. The tracing and tracking of individuals who are then killed with drones creates empathies with those who one eventually must kill. Even though the killing is done via an automated system of sorts, it nevertheless creates distress and harm to those utilizing this system.³¹

These are just suggestions and, admittedly, ones that are unlikely to be adopted soon. However, they demonstrate an important part of lawmaking, virtue ethics, and LAWS. Virtue ethics does not require those involved in lawmaking to be perfect; the term “virtue” can sometimes incorrectly lead us to that conclusion. Rather, together, a group of people with different perspectives and different experiences can bring about a set of laws arising from virtuous deliberation in order to promote virtuous actions. So, while the inclusion of international legal experts and state representatives is important in formulating good laws and regulations, voices from those with other experiences and areas of excellence can only enhance this process. This “common knowledge,” as Aristotle calls it, might bring new critical insights into the process of formulating rules around LAWS.

In the end, these weapons will continue to be developed, and they will often be defended in terms of rules and consequences. Attention to how we might think differently about the ways in which laws and regulations are made, and who makes them, could benefit us all. This essay is an attempt to move us forward in thinking about alternatives to how we make laws and regulations about these weapon systems. Creating a more peaceful order requires us to think imaginatively about how we move toward that order.