Originally published on WAR ON THE ROCKS
The decision to invade Iraq was among the most controversial (and arguably worst) foreign policy decisions in modern American history. For many, the term “weapons of mass destruction” is inseparable from the war, motivated or justified as it was by fears that al-Qaeda would acquire Iraq’s weapons of mass destruction (WMD) — the chemical and biological weapons the United States accused Iraq of hiding.
The war also demonstrated the weaknesses of the term “WMD.” Critics argue the term equated the risk of Iraq’s supposed chemical and biological weapons arsenal with far more dangerous nuclear weapons, which Iraq was not accused of possessing. The public and policymakers were more afraid than they otherwise would be. “Weapons of mass destruction” just sounds scary.
While the term has clear flaws, it is still relevant. Getting the terminology right has real-world consequences: The applicability of the term to drone swarms and other future weapon systems has direct consequences for weapons deployment, weapons acquisition, decisions on the use of force, strategic planning, and the character of future battlefields. Alternatives such as “chemical, biological, radiological, and nuclear weapons” (often shortened to “CBRN weapons”) resolve some problems posed by the term weapons of mass destruction. Other alternatives (e.g., “CBRN”; nuclear, biological, and chemical (NBC); and chemical, biological, radiological, nuclear, and explosive (CBRNE) weapons) also worsen other problems and create new ones. Weapons of mass destruction is the only analytical framework useful for assessing whether the risks of drone swarms or any other emerging technology should be addressed the way chemical, biological, and nuclear weapons are.
Why ‘WMD’ Matters
At first blush, debates over terminology look like scholars yelling about navel lint. However, the debate over weapons of mass destruction impacts the real world in significant ways, particularly when it comes to emerging technologies. Chemical, biological, and nuclear weapons are treated differently from conventional weapons across a broad range of policies. A host of bilateral and multilateral treaties have been developed to control their development, export, deployment, and use. At least two, the Outer Space Treaty and Seabed Treaty, restrict the use of weapons of mass destruction without specifically defining what they are. While these treaties might not always constrain use completely, they influence where the weapons can be deployed and the international norms surrounding them. The use of such weapons (or belief that they will be used) also drastically changes public and policymaker support for military intervention, even when the usage is relatively limited in effect or quantity. Emerging drone swarms technology may merit the same policies and consideration.
The framework to limit and control nuclear, chemical, and biological weapons is among the most successful arms control efforts ever, the often-justified criticisms about state commitment to disarmament notwithstanding. While other weapons have drawn international concern, none have generated the same widespread disarmament as weapons of mass destruction. Consider the landmine treaty: 32 states did not sign on, including most of the great powers. Non-signatories include China, Russia, and the United States (though the United States largely adheres to the principle of the treaty). By contrast, the United States, Russia, China, and the rest of the United Nations Security Council have signed onto the Chemical Weapons Convention, the Biological and Toxin Weapons Convention, and the Treaty on the Non-proliferation of Nuclear Weapons. (Of course, in the last case, Security Council members do gain special status as nuclear weapons states). The world’s great powers have also signed on to bilateral arms reduction agreements like the Strategic Arms Limitation Talks and Strategic Arms Reduction Treaties. They also signed multilateral agreements like U.N. Resolution 1540, to combat nuclear, chemical, biological, and radiological weapon terrorism, and the Outer Space Treaty, which bars the placement of weapons of mass destruction in space.
Of course, a critic may agree with all of the arguments regarding the success of the arms control framework, yet disagree with the term “weapons of mass destruction” itself. Some have argued that the term is vague, that it obfuscates real differences between the different types of weapons, and, as a consequence, policymakers and the broader public incorrectly equate the elements. While a variety of new terms resolve the first concern, they worsen the second, and polling data suggests the third is unfounded. Moreover, alternatives to the term “weapons of mass destruction” introduce a significant new problem: How should a new weapon be handled?
What constitutes mass destruction is inherently vague, and more precise alternatives resolve this issue. Alternatives like CBRN, NBC and CBRNE weapons are not vague at all. Does the weapon use a chemical weapon agent to inflict harm? Then it is clearly one of these agents. No ambiguity. That is a definite value of those alternatives. Keen observers will note I use CBRN frequently in my own writings for that reason. While this lack of ambiguity is a clear advantage, these terms also create other problems.
Chemical, biological, and nuclear weapons have different scales of harm and rely on different physical principles, but “weapons of mass destruction” puts them all together. Alternative terms and categories make the problem worse. Radiological weapons pose little risk of prompt harm and are best thought of as weapons of mass disruption, but they are included in CBRN and CBRNE. Likewise, focusing only on physical properties equates relatively less harmful nonstate actor acid attacks with state-level chemical weapons programs, and city-destroying nuclear weapons. All are concerning, but they are hardly equivalent risks.
Scholars have suggested that congressional and public ignorance of the differences between chemical, biological, and nuclear weapons encouraged the Iraq War (and administration officials may have “systematically misrepresented” these differences). Because the idea of “mass destruction” combines these weapons into a single category, the public equates typically less harmful weapons like chemical and biological weapons with nuclear weapons. Nuclear weapons might justify intervention, but not the less harmful chemical and biological agents. However, subsequent evidence suggests chemical weapons usage alone changes public and policymaker views on military intervention in its favor.
Debate has raged over whether the United States should intervene militarily in the ongoing Syrian civil war. In the early years, most of the public was strongly against intervention, especially after the debacle of the Iraq War. However, in the event Syria used chemical weapons, pollsters found the public’s views reversed. In one 2012 Washington Post poll, only 17 percent of respondents favored military intervention (perhaps due in part to a public fear of repeating the Iraq War). However, if Syria used chemical weapons — and specifically chemical weapons — against its civilians, then support for intervention jumped to 63 percent. If Syria lost control of its stockpile, support jumped again to 70 percent. The Syria conflict also shows that chemical weapons change policymaker views too: Chemical weapons concerns prompted former President Barack Obama to issue his controversial red line on military intervention in 2012. While ultimately Obama did not carry out strikes, President Donald Trump did in 2017 and 2018.
Different ways of describing weapons of mass destruction also introduce a new problem: Under what conditions should these categories be expanded to include new weapons? While these weapons could be subsumed under a category like “unconventional weapons,” that just rewords the problem. When should a new weapon be considered unconventional? No clear rationale exists on what binds the weapons variously referred to as NBC, CBRN, and CBRNE together, except for the strategic goals of policymakers and the weapons’ traditional inclusion as weapons of mass destruction. That is particularly a challenge when turning to emerging weapon systems like drone swarms.
The Problem of Swarms
In 1946, the United Nations defined weapons of mass destruction as, “atomic explosive weapons, radioactive material weapons, lethal chemical and biological weapons, and any weapons developed in the future which have characteristics comparable in destructive effect to those of the atomic bomb or other weapons mentioned above” [emphasis added]. When taking a broad look at human warfare, little reason exists to believe humans will stop inventing novel ways to kill one another. Likewise, little reason exists to believe at least some of those novel killing methods might merit serious international concern akin to weapons of mass destruction. Armed, fully autonomous drone swarms represent one such possibility, because they are capable of bypassing any arbitrary threshold for “mass destruction” and cannot adequately differentiate between civilian and military targets. For the purposes of this article, “fully autonomous” means self-mobile and self-targeting, and drone swarms are multiple unmanned platforms that communicate with one another to achieve common objectives.
Armed drone swarms can bypass any arbitrary threshold for “mass destruction.” A single swarm could consist of 1,000, 10,000, or potentially even 100,000 drones. Each drone may be armed with small explosives, large missiles, electronic warfare equipment to jam adversary defenses, sensors to identify targets, or even anti-tank weaponry. Of course, whether a particular swarm can inflict mass destruction is a separate issue. As a heuristic, I use the baseline of 1,000 drones each equipped with a single M67 hand grenade’s worth of explosive. However, additional modeling of swarm size, armament, and overall capability is necessary to develop a less arbitrary threshold.
Like the weapons generally classified as weapons of mass destruction, fully autonomous drone swarms are inherently incapable of differentiating between civilian and military targets. An autonomous targeting system must be able to tell a farmer holding a rake from a soldier with a rifle, and a convoy of refugees from a convoy of soldiers. While autonomous target recognition has made tremendous progress, it is still unable to account for these and other battlefield complexities. A soldier might be sick or injured. Smoke, debris, and other obstructions may lead to false positive identifications of civilian targets as military. Traditional chemical warfare agents can cause mass harm, but an errant wind may blow the agent into a civilian area. Likewise, contagious diseases may spread to civilians and aerosolized biological agents such as bacillus anthracis (anthrax) may also blow into unintended areas. And the destructive power of nuclear weapons is so high that major civilian casualties are inevitable: The nuclear attack on Nagasaki killed between 39,000 and over 50,000 people, most of whom were civilians. While drone swarms are unlikely to inflict the truly catastrophic harm of a city-destroying Tsar Bomba, they fit well alongside chemical weapons, some biological weapons, and potentially low-yield nuclear weapons.
The nature of drone swarm technology means mistakes in one drone can create system-wide mistakes. By definition, drone swarms communicate to accomplish shared objectives. Communication can lead to emergent collective behaviors. Typically, this might mean a swarm adjusting its behavior in the event of one member being disabled or destroyed. More dangerously, one drone may incorrectly identify civilian vehicles as military transports and communicate that the vehicles are an acceptable target to the rest of the swarm.
Given these two properties — an ability to cause mass destruction and a lack of ability to differentiate among targets — armed fully autonomous drone swarms are plausibly weapons of mass destruction.
A skeptic might argue that words are nice, but bullets decide wars. Sure, “WMD” might offer analytical advantages over alternative terms. Sure, a lawyer might classify a drone swarm as a weapon of mass destruction. But what does this matter for the ability to wage and win war?
Weapons Deployment and International Treaties
The Seabed Treaty and the Outer Space Treaty prevent the deployment of weapons of mass destruction on the seabed and outer space without narrowly defining the term to traditional weapons of mass destruction. If armed, fully autonomous drone swarms or any future weapon system are weapons of mass destruction, these treaties apply and those drone swarms could not be used in those domains. Of course, states may elect to abandon or ignore these treaties: Armed, autonomous swarms could be quite useful in space or stationed on the seabed, as they do not require life-sustaining equipment. But abandoning the treaty as a whole may increase the risks of states placing nuclear weapons in space and the seabed, drastically complicating nuclear deterrence. Accepting deployment limitations on drone swarms may be necessary to avoid more serious consequences.
Weapons Acquisition and the Future of Arms Control
The arms control framework applied to chemical, biological, and nuclear weapons offers a model for future weapons that create the same concerns. The approach to countering these weapons includes international norms and treaties against the weapon’s usage, coupled with actions to defeat acquisition pathways and deter usage with threats of military force. Drone swarms and other future technologies may require the same approach. But to know for sure requires serious consideration of “weapons of mass destruction” as a category: What is it about these weapons that merits such a comprehensive response? It also requires serious debate over whether a country’s use of such weapons merits the use of military force in a conflict. And if so, under what conditions?
Shaping the Battlefield
Applying the arms control framework to future weapons will also influence the information and physical battlefields. International norms and treaties to counter traditional and emerging weapons of mass destruction shape the types and quantity of capabilities that adversaries and friendly forces deploy. A treaty restricting swarms with more than 1,000 drones would simplify U.S. planning requirements, to the degree the treaty is adopted, of course. An adversary might just develop a 999-drone swarm, but the risks of a 10,000-drone swarm are reduced. That simplifies requirements, acquisition, and deployment of counter-drone systems. Likewise, international norms and treaties shape the information battlefield. The public may latch onto the belief of a given weapon as being capable of mass destruction, influencing public opinion regarding the weapon system and state behavior. Of course, this sort of information-shaping is more likely to affect democratic regimes where public opinion plays a larger role in shaping defense policy.
Applying experiences from previous arms control efforts can help counter emerging risks. Whether the United States and global governments elect to address drone swarms or other emerging technologies as weapons of mass destruction has a significant impact for national defense and international security. “Weapons of mass destruction” versus alternative terms may be a war of words, but the consequences are real.