https://orcid.org/0000-0002-7689-6988
Damir Kovačević, Afrimadona, and Martin Claar’s (KAC) statistical adjudication between the potential role of norms versus strategic deterrence in preventing interstate chemical-weapons use is a welcome contribution (Vol. 26, Nos. 3–4, pp. 251–66). The authors conclude that a powerful non-use norm is the most convincing explanation for downtrending chemical warfare. However, closer examination reveals several analytical issues, including inaccurate presentation of data and alternative explanations, that may undermine their results. Further evidence is required to evaluate whether the effect identified by KAC is driven by normative prohibitions or other factors. Use of chemical weapons (CW) also continues with disturbing frequency in contexts not discussed by the authors, where norms against these activities may be more fragile.
KAC enter the debate over norms and deterrence noting a sharp reduction in interstate chemical warfare during the twentieth century. Drawing on James D. Morrow and Hyeran Jo’s dataset on the laws of war, they count “222 cases of CW use in interstate wars from 1900 to 2000.”Footnote1 Inspection of the data suggests this figure is misleading for two reasons. First, the dataset encompasses 1900–90, precluding any analysis of major conflicts during the 1990s between CW-capable states, such as the Gulf War. Second, and more importantly, KAC appear to inadvertently overstate the number of CW uses in interstate war.
Rather than cataloging 222 cases of CW use, as KAC indicate, Morrow and Jo analyze 222 dyad-pairs of countries at war to see if they used either chemical or biological weapons at any point during conflict. Only twenty-nine of the 222 meet KAC’s stated threshold of confirmed violations.Footnote2 We examined each of these twenty-nine dyads and found that six referenced biological warfare or use of other banned weaponry aside from CW.Footnote3 The remaining twenty-three cases reflect only four conflicts: World War I (1914–18), the Italo-Ethiopian War (1935–36), the second Sino-Japanese War (1937–45), and the Iran-Iraq War (1980–88). Of these twenty-three, eighteen occurred during World War I, and only two happened after World War II: Iraq against Iran in 1983 and Iran against Iraq in 1987.Footnote4 By this yardstick, interstate CW use appears to be a relic of a previous era of warfare.
It is heartening news that—excepting World War I—the use of CW in interstate war has been rare. Yet, this makes us wary of any statistical claim that additional agreements prohibiting CW are normatively responsible for their declining use. Rather, the data suggest that they were essentially obsolete as a weapon of interstate war decades before the Chemical Weapons Convention (CWC) entered into force in 1997. KAC do not appear to consider why states might decide to use CW and whether the nature of interstate warfare has reduced their relative military utility. Their analysis assumes that, in the absence of a deterrent or a normative taboo, all capable states would use CW in war. We do not believe this is the case.
When deciding to use CW, states have almost always considered military utility alongside deterrence and normative considerations.Footnote5 When the anticipated benefits are low, states may simply prefer other forms of warfare.Footnote6 We should be cautious about attributing non-use either to norms or to international law in the absence of compelling evidence controlling for the effects of military utility.
There are legitimate questions about military utility of CW in interstate conflict. The polar opposite of modern precision-guided munitions, CW are wholly indiscriminate; their battlefield use may affect enemy combatants, friendly soldiers, and civilians alike. Weather patterns—including wind, temperature, humidity, and precipitation—and terrain contours all have considerable implications for dispersal patterns and efficacy.Footnote7 States have also developed countermeasures such as portable detectors, personal protective equipment, decontamination capabilities, and antidotes.Footnote8
For CW to be most effective, targets must be concentrated, fixed in place, and relatively unprotected. This occurs infrequently in modern interstate warfare, as the lessons of World War I resulted in mechanized militaries that practice dispersion of forces and small unit maneuver.Footnote9 Soldiers in vehicles are less exposed to the environmentFootnote10 and can quickly move to avoid contaminated areas.Footnote11 Further, because CW are indiscriminate and can have lingering effects, their use may may limit friendly freedom of maneuver in affected areas.Footnote12 Consequently, we should only expect to see CW used when battle lines are static and the prevailing strategy is one of attrition—i.e., trench warfare—or against unprotected civilian populations.
The military-utility explanation accords with the Iran-Iraq War, the one modern interstate war involving CW use. Frequently compared to World War I,Footnote13 the conflict devolved into a stalemate with a long, static front and mass “human-wave” offensives—even bayonet charges—launched against forces dug-in behind mounted machine guns and barbed wire. These are the conditions where CW are most potent. It is therefore unsurprising that both Iraq and Iran resorted to their use.Footnote14 Considering military-utility calculations, we agree with KAC (pp. 263–64) that future investigations should involve in-depth qualitative case studies complementing statistical analysis.
Reinforcing this point is a bevy of historical evidence indicating that the deterrence logic the authors discount played an important role in several cases of CW forbearance, such as: British restraint against Nazi Germany, German non-use at Normandy, and the Iraqi decision to fight a conventional campaign in the Gulf War.Footnote15 For example, the commanding general of the German Chemical Troops recalled non-use on D-Day: “Enemy superiority in the air had become so overwhelming in the meantime that—quite apart from the moral and political aspects—the initiation of gas warfare by us might have had incalculable consequences for our homeland.”Footnote16 Indeed, non-use explanations likely differ on a case-by-case basis and are rarely monocausal. It would be useful for researchers to study this variation, as searching for a single driver of declining interstate use risks overshadowing historical details that could inform policy making.
Because interstate use has not occurred in over three decades, we wanted to see the full scope of CW incidents. As KAC note, a strong normative taboo would extend to intrastate and non-state warfare, and the authors plan later study of these uses (p. 264). Toward that end, we present a new table of every verified intentional first-use of CW by a political actor from 1900–2019. Because the outcome of interest in is the decision to use CW, we code each first-use by a belligerent in a specific conflict. This avoids treating multiple dyads within the same conflict as individual observations, as a decision to use CW against one member of an alliance is not independent of prior choices targeting other members of that alliance.
Table 1. Users of chemical weapons, 1900–2019Footnote20
shows that while interstate use was quite rare even prior to the CWC, intrastate use remains prevalent in the twenty-first century. CW may be perceived as more effective in these contexts, since rebels and civilians are unlikely to have the defensive countermeasures available to state militaries, and civilian populations are mostly fixed in place.Footnote17 The indiscriminate nature of CW may even be seen as advantageous by groups seeking to victimize civiliansFootnote18 or maximize chaos and disruption.Footnote19 States and non-state actors also continue to employ chemical agents in peacetime operations, including assassinations, where the difficulty of attributing covert uses of CW creates plausible deniability.Footnote20 Footnote21
The data suggest that trying to limit interstate CW use may be misguided—the problem lies elsewhere. Studying CW in intrastate conflicts should be central to any study of CW, as these are the most frequent use cases. Most uses in the last eighty years involved attacks on civilians. Norms against targeting civilians are considered among the most robust and sacrosanct constructs in international politics,Footnote22 Footnote23 Footnote24 raising concerns that potential CW users may be those least concerned with normative compliance. It is telling that, despite deciding not to use CW in combat, Nazi Germany employed gas as a tool of genocide during the Holocaust.
Ultimately, we are unconvinced by the evidence put forward to substantiate KAC’s claim that norms are responsible for declining interstate chemical warfare. Even if, as we hope, there is a growing consensus against CW use, we urge restraint in overstating the strength of the norm. History suggests that use of these weapons is likely to persist wherever it is perceived to convey a strategic advantage or when deterrence is lacking. Barring changes in the anticipated utility of CW, we see no reason to believe that additional international agreements mean the use of these weapons will become less likely in intrastate conflicts, by non-state actors, or when attribution will be difficult or impossible. Accordingly, we look forward to KAC following their intuitions about a promising line of research: the systematic quantitative study of intrastate and non-state uses of CW alongside qualitative case studies. Policy makers and activists interested in ending the abominations of chemical warfare should not rest on their laurels simply because one (already rare) type of CW use has decreased.
Notes
1 James D. Morrow and Hyeran Jo, “Compliance with the Laws of War: Dataset and Coding Rules,” Conflict Management and Peace Science, Vol. 23, No. 1 (2006), p. 7.
2 KAC set their threshold for inclusion at Level 3, or “some major violations.” But of the 222 dyads, Morrow and Jo code 173 as Level 1, or “no violations at all.” Three are Level 2, or “minor violations.” Nine are Level 3, and twenty are Level 4, or “many violations such that compliance doesn’t matter.” Seventeen were undetermined due to missing data.
3 These cases are: Italy’s exploding-bullet use in the Italo-Turkish War (1911), Austria-Hungary’s dum-dum-bullet use against Serbia in World War I (1914), Bolivia’s attempted biological-weapons use against Paraguay in the Chaco War (1932), Ethiopia’s dum-dum-bullet use in the Italo-Ethiopian War (1935), Japan’s biological-weapons use against the Soviet Union in the Nomonhan Incident (1939), and Nazi Germany’s uncorroborated use of CW against the Soviet Union in World War II (1945). Morrow and Jo’s source for this last incident states unequivocally that Germany did not use CW. See: Gerhard L. Weinberg, A World at Arms. Cambridge: Cambridge University Press (1994), pp. 558–60.
4 KAC’s figure, “Frequency of CW use, 1900–2000,” (p. 262) also appears to include two Level 2 observations in 1939 and one in 1965—below their stated threshold for inclusion.
5 See e.g., US Department of the Army, “Armed Forces Doctrine for Chemical Warfare and Biological Defense,” FM 101-40 (Washington DC: Department of the Army, 1976), <www.bits.de/NRANEU/others/amd-us-archive/FM101-40%2876%29.pdf>, pp. 3-1–3-3.
6 Rebecca K.C. Hersman and William Pittinos, Restoring Restraint: Enforcing Accountability for Users of Chemical Weapons, (Washington, DC: Center for Strategic and International Studies, 2018), p. 5.
7 US Department of the Army, Field Behavior of NBC Agents (Including Smoke and Incendiaries), FM 3-6 (Washington, DC: Department of the Army, 1986), <https://fas.org/irp/doddir/army/fm3-6.pdf>, pp. 1-1–1-16.
8 Dana A. Shea, US Congressional Research Service, “Chemical Weapons: A Summary Report of Characteristics and Effects,” CRS Report R42862, September 13, 2013, <https://fas.org/sgp/crs/nuke/R42862.pdf>, pp. 2–7.
9 Stephen Biddle, Military Power: Explaining Victory and Defeat in Modern Battle (Princeton, NJ: Princeton University Press, 2006), pp. 28–51.
10 US Department of the Army, NBC Field Handbook, FM 3-7 (Washington, DC: Department of the Army, 1994), <https://fas.org/irp/doddir/army/fm3-7.pdf>, p. 2-13.
11 See e.g., US Department of the Army, Infantry Rifle Company, ATP 3-21.10 (Washington, DC: Department of the Army, 2018), <https://armypubs.army.mil/epubs/DR_pubs/DR_a/pdf/web/ARN8519_ATP%203-21x10%20Final%20Web.pdf>, pp. H-1–H-8.
12 On CW persistency, see US Department of the Army, NBC Field Handbook, p. 2-2.
13 See e.g., Will D. Swearingen, “Geopolitical Origins of the Iran-Iraq War," Geographical Review, Vol. 78, No. 4 (1988), p. 408; Thomas L. McNaugher, “Ballistic Missiles and Chemical Weapons: The Legacy of the Iran-Iraq War,” International Security Vol. 15, No. 2 (1990), pp. 5–34; Ray Takeyh, “The Iran–Iraq War: A Reassessment,” Middle East Journal, Vol. 64, No. 3 (2010), p. 373.
14 Iranian CW use was significantly less widespread than Iraqi CW use and came late in the conflict. See: David Segal, “The Iran-Iraq War: A Military Assessment,” Foreign Affairs, Vol. 66, No. 5 (1998), p. 956; Gregory F. Giles, “The Islamic Republic of Iran and Nuclear, Biological and Chemical Weapons” in Peter R. Lavoy, Scott D. Sagan, and James J. Wirtz, eds., Planning the Unthinkable: How New Powers Will Use Nuclear, Chemical, and Biological Weapons (Ithaca, NY: Cornell University Press, 2000), pp. 82–83.
15 Lawrence Freedman and Efraim Karsh, “How Kuwait Was Won: Strategy in the Gulf War,” International Security, Vol. 16, No. 2 (1991), pp. 5–41; Jeffrey Legro, Cooperation Under Fire: Anglo-German Restraint During World War II (Ithaca, NY: Cornell University Press, 1995); Daniel Byman, Kenneth Pollack, and Matthew Waxman, “Coercing Saddam Hussein: Lessons from the Past,” Survival, Vol. 40, No. 3 (1998), pp. 127–51; Jonathan B. Tucker, War of Nerves: Chemical Warfare from World War I to Al-Qaeda (New York: Pantheon, 2006).
16 Herman Ochsner, History of German Chemical Warfare in World War II: Part I (The Military Aspect), US Army Chemical Corps Historical Study No. 2 (Washington, DC: Department of the Army, 1949), <www.ibiblio.org/hyperwar/NHC/NewPDFs/USArmy/USArmy%20German%20Chemical%20Warfare%20WWII%20Pt.%201.pdf>, p. 23.
20 Year references the first verified intentional use by a political actor in a conflict. Use by one recognized state against another is coded as interstate. Use in civil war, counterinsurgency, or by insurgents against a state or occupier is coded as civil. Other types of attacks are coded on a by-case basis. Unattributed attacks, lone-wolf attacks, and attacks using primitive CW (e.g., poisoned arrows) are not included. However, it should be noted that applying a more permissive evidentiary standard and including unverified allegations provides even further support for the notion that CW use has become an intrastate phenomenon. (See, e.g.: Julian Perry Robinson, “The Negotiations on the Chemical Weapons Convention: A Historical Overview,” in Michael Bothe, Natalino Ronzitti, and Allan Rosas, eds., The New Chemical Weapons Convention: Implementation and Prospects (Dodrecht, Netherlands: Kluwer Law International, 1998), pp. 33–35.) Data from Morrow and Jo, “Compliance with the Laws of War”; Susan B. Martin, “The Battlefield Use of Chemical, Biological and Nuclear Weapons from 1945 to 2008: Structural Realist Versus Normative Explanations,” paper presented to the 2010 Annual Meeting of the American Political Science Association, <https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1643305>, p. 28; William R. Johnston, “Summary of historical attacks using chemical or biological weapons,” Johnston’s Archive, December 5, 2017, <www.johnstonsarchive.net/terrorism/chembioattacks.html>; John D. Sislin, “Chemical warfare in the interwar period: insights for the present?,” Nonproliferation Review, Vol. 28, No. 3–4 (2018), pp. 185–202. Special thanks to Cyrus Jabbari, Amanda Moodie, Richard Pilch, and Joshua H. Pollack for comments and additional suggestions. Any errors are our own.
21 Nakam, meaning “Revenge,” was a group of Holocaust survivors who sought to kill Germans and Nazis.
22 The People's Movement for the Liberation of Angola (MPLA) is the party that has ruled Angola since 1975.
23 The Kurdistan Workers’ Party (PKK) is a far-left militant organization in Turkey and Iraq.
17 While CW may be more deadly in these contexts, military utility is not guaranteed. See Geoffrey Chapman, Hassan Elbahtimy, and Susan B. Martin, “The Future of Chemical Weapons: Implications from the Syrian Civil War,” Security Studies, Vol. 27, No. 4 (2018), pp. 714–25.
18 See e.g., Alexander B. Downes, Targeting Civilians in War (Ithaca, NY: Cornell University Press, 2007).
19 For an overview of literature on non-state use, see Gary A. Ackerman, Jeffrey M. Bale, Victor Asal, R. Karl Rethemeyer, Amanda Murdie, Mila Johns, and Markus K. Binder, Anatomizing Chemical and Biological Non-State Adversaries: Identifying the Adversary (College Park, MD: National Consortium for the Study of Terrorism and Responses to Terrorism, 2014).
24 See Benjamin Valentino, Paul Huth, and Sarah Croco. “Covenants Without the Sword: International Law and the Protection of Civilians in Times of War,” World Politics, Vol. 58, No. 3 (2011), pp. 339–77.