Published by The Lawfare Institute
in Cooperation With
In the wake of George Floyd’s death, a firestorm of protests has swept the United States. Some protests have been violent, though most have been peaceful. In cities around the country, curfews have been implemented as heavily armed police have squared off against cornered crowds, many of them nonviolent. These unorthodox crowd-control tactics have often ended badly. Reporters and bystanders have been caught in the crossfire. Of particular concern to many observers is law enforcement’s heavy reliance on tear gas, a nonlethal chemical irritant designed to disperse crowds. According to the New York Times, tear gas has been used in 98 cities since the protests began, constituting “the most widespread domestic use of tear gas against demonstrators since the [1970s]” and potentially exacerbating the spread of the novel coronavirus.
One event in particular has grabbed headlines. On June 1, police fired tear gas without warning at peaceful protesters in order to clear space for an impromptu Trump photo-op outside a Washington, D.C., church. The action drew widespread criticism from elected officials and the public, including that church’s own bishop. Guides on do-it-yourself protective suits to mitigate the asphyxiating and lachrymatory effects of tear gas were circulated widely on the internet. Ten days later, the chairman of the Joint Chiefs of Staff, who had been present for the photo-op, took the extraordinary step of making a video apology for his role in what happened. A Seattle judge ordered local police to temporarily suspend tear gas use in reaction, and some members of Congress are mobilizing legislation to ban it nationwide.
In the wake of the St. John’s Church incident, the claim that tear gas is a weapon of war under the Geneva Conventions has become a clarion call online, garnering thousands of likes on social media. Rep. Alexandria Ocasio-Cortez of New York, attempting to marshal support for a ban, described tear gas to other lawmakers as “a chemical weapon that is used in war.” Rapper Ice Cube, who has been outspoken about the protests, garnered more than 142,000 likes for his claim on Twitter that tear gas is “actually illegal.”
Tear gas is clearly a chemical that has been weaponized by law enforcement, but it is not always clear-cut how tear gas and other incapacitants fit into the international legal framework regarding chemical weapons. In fact, the truth is stranger than fiction: Most governments consider tear gas a weapon of war yet routinely use it against their own populations during periods of internal unrest. Scholars have written extensively about the emergence of a chemical weapons taboo after the terrifying experiences of World War I. Notional links are now being drawn between chemical weapons and tear gas. So why is there no apparent taboo on tear gas where we would most expect it—in use against one’s own fellow citizens?
Tear gas is part of a broad class of gas incapacitants known as riot-control agents (RCAs). While technically nonlethal, these incapacitants can cause pain, difficulty breathing and temporary blindness upon exposure. Although highly unpleasant and ethically dubious, RCAs are a preferred tool of law enforcement for riot control because they are “safe” and effective relative to lethal alternatives. There have been tear-gas-related deaths, but not many. For example, a Palestinian man was killed in a tear gas attack in 2011—but from being hit with the canister. And 17 high school students in Venezuela were stampeded to death during an attack in 2018. One Ohio woman died after being exposed last month.
Gas warfare in World War I—which spurred rapid developments in opinio juris against chemical weapons—began with RCAs, when the French Army used a rudimentary nonlethal agent (probably a precursor to xylyl bromide) to disperse German defenders from the trenches. Militaries quickly learned that a more potent agent cuts out the middleman—instead of forcing combatants out from cover, lethal alternatives like blood gases and blistering agents (and later nerve gases) can kill them outright. After the war, the traumatized victors sat down to negotiate a more comprehensive ban than the anticipatory restrictions on chemical weapons first set out in 1899 and 1907. The resultant 1925 Geneva Protocol banned “the use in war of asphyxiating, poisonous or other gases, and of all analogous liquids, materials or devices” (emphasis added). As one might imagine, the words “asphyxiating” and “analogous” later caused some interpretive problems. Although U.S. delegates were among the protocol’s most vocal proponents during the discussions in Geneva, the U.S. Senate failed to ratify the protocol under intense industry pressure, and extrabellum uses—like law enforcement—were left unregulated.
Later, by the mid-1960s, interpretation of the normative prohibition on gas warfare had become an increasingly “thorny problem,” according to State Department history. News broke that U.S. military forces were using RCAs in Vietnam—powerful versions of tear gas and mace—ostensibly for clearing suspected Viet Cong tunnels and caves or subduing rioting prisoners of war. The Geneva Protocol regulates poisonous and asphyxiating gases analogous to those its interwar framers had experienced. The protocol is silent on whether nonlethal chemicals (including nongaseous variants) whose effects are temporary technically count, and the spotty negotiating record, through which we can sometimes glean the architects’ intent, makes it difficult to know with certainty whether the protocol was meant to include tear gases by design. A more detailed Chemical Weapons Convention, designed to minimize these inherent ambiguities, would not be negotiated until 1993.
Though the U.S.’s conservative reading of the Geneva Protocol may have meant that tear gas was technically exempted from international humanitarian law, the controversy resonated with voters, who had come to know tear gas well as it was used by police extensively during the tumultuous 1960s. The North Vietnamese and Viet Cong also allegedly used tear gas during the war, but the news that U.S. forces were doing it was a major blow to the Johnson administration, not least because many insurgents were reportedly asphyxiated by tear gas rather than being dispersed. Gassing enemy troops struck many observers as blatantly unethical—one step away from outright chemical warfare and too close for comfort at an analogical level. This may have precipitated growing American public sentiment that the war in Vietnam was immoral, and it motivated renewed international interest in clarifying and gaining further accessions to the Geneva Protocol.
People naturally tend to demand congruence between their laws and their ethos, and substandard laws are a poor excuse for bad behavior. The U.S. government’s all-or-nothing position had a certain logical consistency. But by doubling down on the position that tear gas is justified both on the battlefield and on its own streets, this consistency provoked public backlash. Defenders of U.S policy struggled to make their case. As one contemporary commentator penned, “Would it have been less evil for Southern police to machine-gun Freedom Marchers than to tear-gas them?”—eliding the fact that two things can be evil at once.
In response, the issue was taken up for formal discussion by the U.N. General Assembly and the Conference of the Committee on Disarmament. Under pressure to make the U.S. seem broadly compliant with normative standards, President Johnson suspended the offensive use of tear gas, though after intense lobbying by the Army Chemical Corps and Joint Chiefs of Staff, he declined to permanently forswear the option. Congress passed several resolutions from 1966 to 1968 that called for “strict observance” of the Geneva Protocol (despite not having ratified it). When President Nixon came into office, he instituted a sweeping review of U.S. chemical weapons policy but nonetheless favored resumed RCA operations in Vietnam. Finally, President Ford renounced “first use” in warfare in Executive Order 11850. By 1974-1975, after four decades of inaction, the U.S. finally acceded to the protocol, and two decades later a more expansive Chemical Weapons Convention, which governs peacetime situations as well, was negotiated as a follow-up.
But though it renounced “first use” of tear gas in wartime, the U.S. government has been reluctant to back down from its long-standing insistence that RCAs like tear gas are not chemical weapons. It is telling that the Chemical Weapons Convention provides an exception under Article II(9)(d) for “certain law enforcement purposes” even while specifying that RCAs are an outlawed “method of warfare” that “may not be used.” Because neither “warfare” nor “law enforcement” is defined in the convention, the U.S. found this rule permissive enough to accept. Officially, the U.S. “considers that the prohibition [against tear gas] applies in international as well as internal armed conflict but that it does not apply in normal peacekeeping operations, law enforcement operations, humanitarian and disaster relief operations, counter-terrorist and hostage rescue operations,” and many other situations, according to the Naval Handbook (emphasis added). Other U.S. government sources corroborate this view.
It seems absurd to claim that enemy combatants should be protected from tear gas but that one’s own compatriots should not. Chemical weapons were outlawed because they cause unnecessary suffering, and RCAs are widely considered to be a kind of chemical weapon. How can a weapon framed in fundamentally humanitarian terms be said to cause unnecessary suffering when used by the military in combat but not when used by police against civilians, irrespective of scale and intensity? Yet nearly every government in the world upholds this double-standard, spotting advantages in being able to turn to tear gas when quelling domestic unrest, while preventing more powerful countries from turning it back on them.
U.S. leadership has long justified its position on tear gas on the basis that, as a temporary, nonlethal agent, tear gas is more humane than lethal conventional alternatives in any context. McGeorge Bundy, Johnson’s national security adviser, described public outrage over the use of tear gas as a “stupid fuss” while he and the president worked to formulate an initial media response in 1965. Johnson expressed frustration during a press hearing, calling it “a mountain out of a molehill.” Nixon maintained that tear-gassing tactics in Vietnam were “sound and basically humane.” Even as late as 1975, on the eve of ratification, Ford was prepared to veto the Geneva Protocol entirely if the Senate insisted on a reading that tied the U.S. military’s hands on RCAs. A compromise was reached, and Ford agreed to some self-imposed restrictions—as a matter of policy, albeit not law. Under these rules, military use of tear gas is now controlled more strictly than police use, but neither is prohibited.
But depictions of tear gas as temporary and nonlethal and thus “humane” are determined by strength and dosage, not by who uses it in what context. And although there are many RCAs that range in toxicity, one of the White House’s main talking points from 1965 to 1975 was that battlefield tear gas was quite literally “no different” than the kinds used by law enforcement around the world and thus should be allowed in all contexts. Moreover, because the U.S. government’s position is primarily a matter of national policy, not international legal doctrine, there are fewer institutional barriers to unilateral reversals. For example, a “Restatement of Policy” meant to encourage U.S. troops to use more tear gas passed the Senate 98-1 in 2006. Its provisions would have gone further if not for congressional opposition to language that specifically promoted “[use] in combat.”
Now the controversy has resurfaced. To be clear, tear gas is a more humane crowd suppressant than, say, live ammunition—especially given that other nonlethal alternatives like rubber bullets can still cause permanent damage. And whether or not one considers tear gas to be ethical, its use in crowd-control contexts is, strictly speaking, legal. But such use is questionable under both human rights law and the laws of war, and it should be used in law enforcement scenarios only where crowds constitute a clear threat, and never abused.
Political leaders in the good graces of the White House—namely Sen. Tom Cotton and Secretary of Defense Mark Esper—have variously described the protests as a “battle space” (Esper) in which American citizens who take to the street should be given “no quarter” (Cotton). Such inflammatory remarks encourage law enforcement to use every means at its disposal to quash demonstrations, including tear gas, which is of course the “default response” to controlling crowds. Esper has since walked back his comments; Cotton has not.
Ironically, those who cast civil unrest—even in its most extreme forms—as analogous to war only undermine their own calls-to-arms and expose themselves to opprobrium. By reminding us that the legality of tear gas rests on flexible definitions, Cotton and others like him underscore the double-edged logic of the Article II(9)(d) exception. If police are fighting a “war,” then tear gas is no longer a legitimate peacetime instrument but instead is transplanted to the sole context—the battlefield—where its use is subject to near-universal condemnation.