Legal technologies: Conceptualizing the legacy of the 1923 Hague Rules of Aerial Warfare

Abstract

Many contemporary armed conflicts are shaped by the reliance on airstrikes using traditional fighter planes or remotely piloted drones. As accounts of civilian casualties from airstrikes abound, the ethics and legality of individual airstrikes and broader targeting practices remain contested. Yet these concerns and debates are not new. In fact, a key attempt to regulate aerial warfare was made 100 years ago. In this article, we approach the regulation of aerial warfare through an examination of the 1923 Hague Draft Rules of Aerial Warfare and the contemporary scholarly discussion of these rules. While the Draft Rules have never been converted into a treaty, they embody logics of thinking about civilians, technologies of aerial warfare, and targeting that are still resonating in contemporary discussions of aerial warfare. This article argues for a contextualized understanding of the Draft Rules as an attempt to adapt International Humanitarian Law (IHL) to the new technological realities while maintaining distinctions between different kinds of spaces and non-combatants. We argue that the Draft Rules prefigure later debates about the legality of aerial bombing by tacitly operating with a narrow understanding of the civilian and by offering a range of excuses and justifications for bombing civilians.

1. Introduction

The recent and ongoing armed conflicts in Afghanistan, Iraq, and Syria have been marked by a strong reliance on aerial warfare by some of the parties to the conflict: US and NATO forces as well as Russia (in Syria) and Turkey (in Syria and Iraq). In Afghanistan, US and coalition forces used at least 13,072 airstrikes between 2015 and 2021 alone. ¹ Between 2008 and 2020, coalition airstrikes have killed at least 4,000 Afghan civilians. ² Afghan protests against the US and Coalition military presence in the country had often focused on airstrikes and the lack of accountability for them. For example, in 2008, the Afghanistan Independent Human Rights Commission published a report entitled From Hope to Fear: An Afghan Perspective on the Operation of Pro-Government Forces in Afghanistan, which identified US and Coalition airstrikes, a lack of acknowledgment of civilian casualties, and a lack of accountability for civilian harm as key areas of concern for Afghan publics. ³ On 29 August 2021, hours before the withdrawal of the US troops, one last widely publicized drone strike was aimed at an ISIS operative but killed ten civilians, including seven children and an aid worker. ⁴ This and other airstrikes raised the question about the precision of aerial warfare and about accountability for civilian harm arising from attacks that kill civilians.

While remotely piloted armed drones are recent technological innovations, aerial warfare has been practiced, discussed, and regulated for more than a century. In the early twentieth century, aerial warfare had transformed the geography of war’s violence: planes not only added a vertical dimension to warfare, but they also extended the spaces that were vulnerable to bombardment because they could fly and strike far beyond sovereign borders and the front lines on the ground that had previously marked the territory of warfare. International lawyers’ responses to the possibility and the subsequent practice of aerial bombardment showed a range of regulatory approaches and understandings of the relationship between law and technology. Before aerial warfare was even a possibility, the 1899 Hague IV Declaration prohibited the ‘the launching of projectiles and explosives from balloons, or by other new methods of similar nature’ for a period of five years. ⁵ This ban, which was limited to conflicts between state parties, was renewed in 1907. ⁶ Yet, as aerial bombing became a technical possibility, the global support for such a ban waned.

In 1911, an Italian aviator dropped two incendiary bombs on targets near Tripoli (Libya). ⁷ In subsequent years, the French and British militaries used aerial bombardment in imperial wars in Morocco, Afghanistan, Syria, and Iraq. Yet when early twentieth century international lawyers in the US and the UK discussed the legality and regulation of aerial bombardment in the aftermath of the First World War, they referred almost exclusively to the use of aerial warfare within Europe during the First World War. Allied bombing raids killed 746 German civilians, while the German military killed at least 1,284 people in air raids on Britain. In Paris, 275 civilians were killed by German air raids and 1,600 by artillery shelling. ⁸ In the early twentieth century, leading up to the end of the First World War, aerial bombing was imagined, banned, practiced first in colonial warfare, and finally brought into European spaces of war. In the aftermath of the First World War, the negotiations about the future of international law and war included deliberations about the regulation of warfare, which culminated in the 1923 Hague Draft Rules of Aerial Warfare. The Draft Rules articulated new rules on legitimate targets and the distribution of responsibilities for the use of this novel technology of warfare. In the process of doing so, the Draft Rules also defined and conceptualized the status of the civilian. ⁹ A century after their articulation, the logic of the Draft Rules still shapes the adjudication of aerial warfare and the civilian status. Our inquiry into the Draft Rules one hundred years after their publication is prompted by concerns about the present state of regulating aerial bombing. In the 2020s as well as in the 1920s, aerial warfare is frequently used by economically powerful states with access to advanced weapons technology on the territories and peoples of states that lack access to such technology. The fact that militaries from the industrialized Global North have bombed cities in Afghanistan, Iraq, and Syria in the 1920s as well as in the early twenty-first century is not an indication that nothing changed, but it should prompt us to ask how international law, global hierarchies, and new technologies have interacted and are shaping current conflicts. ¹⁰

This article embeds a discussion of the 1923 Draft Rules in a systematic analysis of discussions of aerial warfare and civilian harm in English language international law and international relations journals published in the US and the UK from 1908, three years before the first aerial bombardment from a plane, to the early 1930s. Early scholarly conversations called for the regulations of aerial warfare, culminating in the 1923 Hague Rules. Subsequent debates in these journals discussed and interpreted these rules, providing further analysis into the logics, fears, and insights that animated the regulation of aerial warfare a century ago. In systematically tracing the discourses and logics of regulating aerial warfare by scholars and practitioners in these journals, we aim to contribute to scholarship on the imperial pasts and present of international law as well as to debates about the relationship between law, race, the protection of civilians, and new military technologies. Our analysis leads to four connected arguments.

First, the relationships between the early practitioners, regulators, and commentators of aerial warfare were complex. Some lawyers called for a restriction of aerial warfare before it was even practically possible, and others were more cautiously optimistically embracing the new weapons technologies. Participants in the 1922–1923 Hague Conference included jurists as well as military officers acting in ‘advisory’ capacity. A number of these delegates and advisors subsequently published analyses, recollections, and commentary on the Draft Rules in international law and international relations journals. ¹¹ The communities of practitioners and regulators overlapped considerably. By collaborating in different forums, they produced a body of regulations and commentaries that legitimized the use of aerial warfare and created excuses for civilian harm that could be used by belligerents.

Second, most participants in the discussions understood aerial warfare as a ‘disruptive technology’ in that it significantly changed the geography of war and rendered older forms of regulation impracticable. They sought to establish a novel legal vocabulary that could respond to these technological changes. For example, while the 1899 Hague Convention prohibited the bombardment of ‘undefended’ places in land and naval warfare, the category of ‘undefended’ was found to be increasingly meaningless in relationship to airplanes against which no meaningful defence was available. ¹² The 1923 Hague Rules solidified the shift from the focus on ‘defended’ and ‘undefended’ places as a criterion for targeting to a rule that only allowed the targeting of ‘military objectives’. This new rule necessitated defining the boundaries of military and civilian objects, persons, and infrastructure. States that were the dominant users of aerial warfare or had the economic ability to build up an air force played a leading role in shaping the rules designating legitimate targets, and the interests of the participating states were clearly reflected in the negotiations and the resulting Draft Rules. Through these processes, global technological asymmetries – marked with racial and spatial undertones – shaped the architecture of the legal norms regulating these same technologies.

Third, the regulation of aerial warfare drew on specific moral geographies of international law. As evidenced in the debates we analyse, the regulation was prompted by concerns about intra-European uses of aerial warfare while preserving legal space to use ‘air control’ and airstrikes in imperial and colonial warfare outside of Europe. The regulation of aerial warfare drew on and contributed to the legal distinction between public war within Europe, where civilian protection was at least rhetorically important, and colonial wars outside of Europe, where the local population was imagined as collectively hostile and not deserving of legal protections. ¹³

Finally, drafters and commentators understood the 1923 Hague Rules as ‘facilitating’ rather than simply restricting aerial bombing. ¹⁴ While the Draft Rules limited the targets and purposes of aerial bombardment, their regulatory techniques provided space for legitimizing aerial bombardment. Two aspects stand out in this regard: first, the Rules operated with a vague but narrow understanding of the civilian. The tacitly envisioned ‘civilian’ was white and European, not part of a non-European nation under European rule. In addition, the category of the civilian was narrowed by commentators who were eager to exclude, for example, munitions workers from the protective cover of the civilian status. Second, the Hague Rules focused on the asserted intent of the perpetrators of airstrikes over the effects on the ground and offered excuses and justification for the bombing of civilians. The high level of tolerance for ‘collateral damage’ is a feature of this regulatory logic. We show that contemporary analysts understood the rules to be purposefully permissive of air attacks while retaining a veneer of humanitarianism.

The article proceeds with a short description of our systematic content analysis, followed by an account of the 1922–1923 Hague Conference that highlights the participants’ approach to the role of international law in regulating technological innovations in warfare. In Section 4 we focus on the conceptualization of the civilian in the 1923 Draft Rules as well as the surrounding scholarly discussion, showing the implied limits of the civilian status. Section 5 demonstrates that the mechanisms by which the 1923 Draft Rules sought to restrict aerial bombardment were understood to pose weak constraints on the practice of aerial warfare. Section 6 discusses the contemporary evaluations as well as the long-range legacies of the Draft Rules.

2. Overview: Sources, methods, networks of authors

In order to trace the scholarly discussion about the regulation of aerial warfare before and after the 1923 Draft Rules, we conducted a qualitative content analysis of articles and notes published in US and British international law and International Affairs/International Relations journals between 1908 and 1935. ¹⁵ In comparison to other literature that makes arguments about the Hague Rules, our approach focuses on the English language literature in two countries (as exemplified by journals), and it is committed to a systematic socio-legal analysis. Instead of approaching the early twentieth century discussion by inadvertently picking out the most surprising, shrillest, or best amplified texts, our project treats the searchable databases of international law and international relations journals as the archive from which we glean trends in scholarly concerns and terminologies, trace debates, and understand the context of the articles we end up analysing in more detail. ¹⁶

Using the JSTOR database and combinations of the keywords ‘civilian’, ‘air’, ‘bomb’, and ‘aerial warfare’ for the years 1900 (well before the first aerial bombing) to 1935 (to focus on the discussions before the Second World War), we compiled a comprehensive list of primary sources. We then added two secondary searches: one search was conducted with the same search terms but open academic disciplines; this search yielded a small number of additional sources from Engineering and Science journals. The second secondary search used different terms to refer to possible civilians. The literature on the history of the civilian status suggested that early twentieth century Western lawyers frequently referred to non-combatants outside of Europe not as civilians but as ‘natives’ and to colonial warfare as insurgencies and ‘small wars’. ¹⁷ In order to account for this racialized history of the civilian status, we conducted separate keyword searches that include the term ‘native’, ‘insurgency’, and ‘small wars’ instead of ‘civilian’. We then selected all the articles that treated questions of aerial bombardment and civilian status in more than a passing reference. This list was supplemented with a small number of articles that had not come up in our search (because the relevant journals were not available to be searched in this time frame) but were referenced frequently, indicating a central role of the authors and the pieces in the discussions. Within the body of journals available in the database, the first scholarly discussion of possible aerial bombardment was published in 1908. Since our goal is to capture the scholarly discussion before and after the 1923 Hague Rules, we chose to end the time frame in the mid-1930s before debates about Japanese aerial bombing of China, the Second World War in Europe, and the 1935 Italian bombing of Ethiopia would become subjects of discussion and reflection.

The articles we analyse do not represent the entirety of the professional discourses about bombing in international law and international relations. Rather, they are an accessible and searchable ‘sample’ of larger English language discursive formations that also include books, conferences and non-academic writing. ¹⁸ The lively discussions about aerial warfare and law that simultaneously took place in France, Italy, Germany, and other states are beyond the scope of this article. ¹⁹ We read all the articles on the shorter list for their understanding of international law, technologies of aerial warfare, and geographies of warfare. ²⁰ After reading the primary sources for the first time, we established a range of ‘themes’ and thematic statements that had been raised by the authors. These themes included ‘aerial warfare should be banned’, ‘aerial warfare cannot be banned’, ‘the existing rules on bombardment have become obsolete with the new technology’, ‘German warfare in the First World War is an important reason for regulating aerial warfare’, and ‘the distinction between civilians and combatants is important for civilization’. After establishing a range of relevant themes, we conducted the ‘coding’ of the primary sources by re-read the primary sources by identifying specific thematic statements in each source that fit under the umbrella of these themes and marking them as such. In order to render our interpretations more robust, both authors conducted this coding process in parallel and independently from each other. We then compared and synthesized our findings and connected these themes with discussions in the theoretical literature on IHL, technology, and the racialization of warfare. These steps in the research process were designed to ensure that we capture the available range of scholarly voices and systematically identify convergences in themes across a range of articles.

The articles chosen for analysis were published between 1908 and 1935 in a range of journals. The following journals were represented with more than one article: the American Journal of International Law, the (British) Transactions of the Grotius Society, the (British) Journal of the Royal Institute of International Affairs, and Foreign Affairs (US), The Military Engineer, and Scientific American (US). The authors included six academics and lawyers, nine military officers, as well as three businessmen, engineers, and politicians. Of the seven articles published in international law journals such as the American Journal of International Law, four were written by serving or retired military officers, two by academics, and one by a lawyer.

All authors were white men affiliated with centres of political, military, intellectual, and financial power. They frequently wrote with specific emphasis on the position and interests of their country. In some cases, the confluence of the author’s main occupation and their writing was obvious. For example, in 1932, James E. Mills, former Technical Director of the US military’s ‘Chemical Warfare Service’ opined that gas was ‘the most humane weapon which exists today for use in actual warfare’ because it killed only two percent of those who were exposed to it. ²¹ In 1927, Anthony Fokker, the Dutch builder of airplanes who had been instrumental in arming Germany with war planes in First World War, proclaimed ‘[i]f the war to end all wars ever comes it will be a war of aerial development’ and added that he was ‘deeply invested in American aviation, for I have become a citizen of this great country’. ²² These overtly self-interested accounts of aerial warfare should not hide the fact that scholars, lawyers, and military officers were joint participants in the discussions and reflections on the regulation of aerial warfare – even in ostensibly academic spaces.

Throughout the period we observed, the beginning, the practice, ethics, and regulation of aerial warfare was discussed jointly by military officers, diplomats, and academic lawyers and political scientists. For example, within two years of the formulation of the Draft Rules, the American Journal of International Law, the flagship journal of the American Society of International Law, published three articles that centrally discussed these Draft Rules. The first of these appeared in 1923 and was written by William L. Rodgers, then Rear Admiral and a ‘technical advisor’ on the US delegation at the Hague conference. ²³ James Garner, the author of the second article published in 1924, was an historian and political scientist whose research focus had shifted from US post-civil war Reconstruction to questions of international law raised by the First World War. ²⁴ In 1925, the journal published an article by Elbridge Colby, who was a military officer and later became a professor of English and journalism. ²⁵ The prominent role of military officers and military lawyers in developing international law concerning aerial targeting is not an idiosyncratic trait of the 1920s. Craig Jones has shown that US and Israeli ‘war lawyers’ have reshaped their governments’ claims about IHL and the legality of specific targeting practices. ²⁶ The relationship between international lawyers and military officers has transformed over the course of a century of practicing, regulating, and legitimizing aerial warfare, but the close collaboration between practitioners and regulators of aerial bombardment has been a consistent pattern. In the next section, we provide background and context on the Hague Rules. The two subsequent sections explore the effects of these overlapping networks and collaborative practices by focusing on the conceptualization of civilians and the space the regulations provided for excusing civilian casualties.

3. The Hague Rules of Aerial Warfare

In 1922–1923, diplomatic, legal, and military experts from the US, Great Britain, France, Italy, the Netherlands, and Japan met in The Hague in order to negotiate Draft Rules of Aerial Warfare. ²⁷ These rules were widely reprinted, excerpted, and discussed in international law and international relations journals in 1923–1924. They occupy a curious position in IHL: on the one hand, the Rules are non-binding and have never been converted into a treaty that would have been open to ratification. On the other hand, specialists in historical approaches to aerial warfare and the civilian status regard these Draft Rules as an important starting point for IHL’s understanding of the civilian and a turning point in the rules on aerial warfare and targeting. ²⁸ Within the broader discussion of aerial warfare in the interwar years that is the focus of this article, the Draft Rules provided an important focal point: while some earlier articles called for the regulation of aerial warfare, ²⁹ later articles interpreted, discussed, and dissected the Draft Rules. ³⁰ Drawing on the contemporary discussion of the Draft Rules ³¹ as well as on the subsequent literature contextualizing this event, ³² this section introduces the Draft Rules and makes arguments about the process and contents that will be developed further in the later sections of the article.

3.1 Drafting the Draft Rules

In the aftermath of the First World War, a series of diplomatic conferences aimed to clarify urgent questions of international law that had arisen over the course of the war. ³³ The 1921–1922 Washington Conference on Disarmament explored mechanisms for limiting the naval arms race and made decisions on the status of contested territories in East Asia. The Conference appointed a ‘Commission of Jurists’ to explore the questions:

(1) Do the existing rules of international law adequately cover new methods of attack or defense resulting from the introduction or development, since the Hague Conference of 1907, of new agencies of warfare; and (2) if not so, what changes in the existing rules ought to be adopted in consequence thereof, as part of the law of nations. ³⁴

The US Government, acting as the convener, chose radio communications and aviation as the two new technologies of warfare that required further study and regulation. ³⁵ In December 1922, the Commission formally convened in The Hague in order to discuss these questions. From the beginning, aerial warfare was framed as a technology that had made previous modes of legal regulation obsolete. As Committee Chair John Bassett Moore reported, while other recent innovations such as tanks ‘were considered merely as inventions extending or intensifying the operations of well-known methods of attack and offence’, the use of airplanes in war ‘seemed to be on a somewhat different footing’. ³⁶

The participants in the conference at The Hague came from the five national delegations that had taken the lead in the Washington Conference – United States, France, Italy, United Kingdom, and Japan – as well as the Netherlands, which had been invited to join as the host country. Each delegation was supposed to be led by two jurists, though participants recall that some delegates’ status as jurists as opposed to diplomats might have been contestable. ³⁷ The delegates were accompanied by a number of ‘technical advisors’, most of whom were military experts. ³⁸

While the delegations were headed by lawyers, the large number of military personnel acting in ‘technical advisor’ capacity hints at the robust participation of military practitioners in the formulation of the rules on aerial warfare. While the Netherlands had remained neutral during First World War, all other states that negotiated the Draft Rules had all been on the winning side of First World War. In addition, most states that were involved in formulating the rules of aerial warfare had experience using aerial bombardment: French forces had bombed rebels in Morocco in the 1912–1914 war, the US used bombs in ‘interventions in Central American and the Caribbean’ during the 1920s, ³⁹ the UK Royal Air Force had bombed several cities in Afghanistan and in Somaliland in 1919 and 1920, and the Italian military had first used bombs dropped from planes in Libya in 1911. ⁴⁰ The UK Royal Air Force developed a doctrine and practice of ‘air control’ over Iraq that was formally put in place in October 1922, two months before the negotiations on the Hague Rules of Aerial Warfare commenced. ⁴¹ While a number of the states that negotiated the 1923 Draft Rules had resorted to aerial bombardments not only in the First World War in Europe, but also in colonial and imperial aggressions, the discussions of the proposed rules were based on experiences with the use of aerial warfare within Europe.

When the Commission convened in December 1922, the US international lawyer John Bassett Moore was elected as chair. ⁴² According to participants, ‘he immediately presented the American draft codes on aviation and radio’. ⁴³ The UK delegation submitted a draft code on aviation, which was perceived as being similar to the US draft in logic and wording. ⁴⁴ The Commission worked on the basis of the US draft ‘and the British draft as supplementary thereto’. ⁴⁵ The composition of the delegations and the respective influence of different delegations matters because, as participants acknowledged, ‘each nation seemed chiefly guided by the principle of promoting its own national policies and its position in the world’. ⁴⁶ For example, Moore noted that the delegations from the Netherlands and Japan, the two states least committed to the use of aerial warfare at the time of the Conference, ‘were in favour of the utmost restriction possible and advocated a general prohibition of aerial bombardment of towns and villages outside the immediate area of military operations’. ⁴⁷ The Italian delegation, in turn, noting that Italian historical buildings and monuments had frequently been under air attack during First World War, prompted the adoption of a series of regulations treating ‘historic monuments’ as well as ‘buildings dedicated to public worship, art, science, or charitable purposes’ differently than other public or private buildings, enshrined in Articles 25 and 26. ⁴⁸ While the Italian delegation gave impulses for Articles 25 and 26, the draft for Article 24 – which defined legitimate targets of aerial bombardment – was provided by John Bassett Moore, the chairperson of the Commission. The structural features of the Commission – the six national delegations, the mix of lawyers, diplomats, and military specialists – shaped the content of the 1923 Hague Draft Rules.

The 1923 Hague Draft Rules responded to concerns about the use of aerial warfare during First World War, but they were not the first attempt to regulate the use of airplanes for warfare. In 1899, The parties to the Hague IV Declaration decided ‘to prohibit, for a term of five years, the launching of projectiles and explosives from balloons, or by other new methods of similar nature’ in armed conflicts among the signatories. ⁴⁹ At that time, no single bomb had been dropped from an airplane yet, and the parties were committed to a complete ban on using airplanes and hot air balloons to release munitions and explosives. As Bogliolo writes, ‘aerial bombing was seen as unnecessary and cruel’ by most participants in the conference. ⁵⁰ The 1899 ban was initially envisioned as a perpetual ban, but a US representative re-opened the discussion, leading to a temporary five-year ban. ⁵¹ Even the temporary ban was not comprehensive: it only applied to conflicts in which all belligerents were parties to the declaration. In practice, this would make the ban inapplicable to most imperial and colonial conflicts since the theorists and practitioners of international law were not willing to recognize most non-European nations as members of the ‘family of nations’ or as ‘civilized’. ⁵² Given the low number of signatories, the prohibition ‘became a dead letter’ in August 1914, the start of the First World War. ⁵³ As Chris af Jochnick and Roger Normand argue, the waning enthusiasm for banning the use of planes in warfare is part of a larger pattern in which international treaties ‘banned only those means and methods of combat that had no military utility while permitting new and destructive technologies’. ⁵⁴

Despite its limited practical effect, the 1899 Declaration formed an important reference point and precursor to the 1923 Draft Rules because it suggested the possibility of a complete ban as a possible approach to regulating aerial bombardment. The legal technique of a ban was also used in the 1920s to address chemical warfare. The 1925 Protocol for the Prohibition of the Use in War of Asphyxiating, Poisonous or Other Gases reiterates that ‘the use in war of asphyxiating, poisonous or other gases, and of all analogous liquids, materials or devices, has been justly condemned by the general opinion of the civilised world’ and prohibited in specific treaties, rendering the complete ban of these substances as agents of warfare a matter of relative consensus. ⁵⁵ When the regulation of aerial warfare was taken up in the 1920s, only few commentators called for a complete ban. Yet the theoretical possibility of banning the use of technological innovations shaped the discussion about aerial warfare indirectly. Commentators frequently stressed that aerial warfare could not or should not be banned, proposing regulation as the milder and more desirable alternative. ⁵⁶

Luís Bogliolo has argued that international lawyers experienced the rise of aerial warfare as a ‘technological crisis event’ and that international law itself was transformed by the encounter with aerial warfare. ⁵⁷ Amanda Alexander has made the case that the 1923 Draft Rules mark the ‘entry of the civilian into the legal lexicon’. ⁵⁸ These claims rest on careful analyses of the choices made by the drafters of the 1923 Rules. Since the Draft Rules have had an enduring impact on the legal concepts we use to talk about aerial warfare and harm to civilians, it is worth highlighting how much the Rules departed from previous techniques of regulating armed conflict. Three aspects stand out: first, the Draft Rules completed the shift from the prohibition on targeting ‘undefended’ places to the prohibition on targeting civilian persons and objects. Second, the Draft Rules defined ‘civilian’ and ‘non-civilian’ in relationship to persons and objects. Third, the Draft Rules make civilians on the ground partially responsible for making themselves and specific civilian objects visible to the eyes in the sky. The next sections outline these shifts.

3.2 New rules for new technologies

The rise of aerial warfare challenged the existing rules on targeting. The 1907 Hague Convention on Land Warfare summed up the standard test that ‘the attack or bombardment, by whatever means, of towns, villages, dwellings, or buildings which are undefended is prohibited’ (Article 25). The key legal and moral distinction was between defended and undefended localities. The use of airplanes in war changed the geographical contours of war in two important ways: airplanes added a vertical dimension to warfare, and airplanes could move many times faster than ground-based vehicles of warfare. As a consequence, airplanes could attack cities far from the frontlines, catching inhabitants unaware. The traditional distinction between ‘defended’ and ‘undefended’ places seemed inapplicable in relationship to aerial attacks: any infrastructure of military defence that might offer protection against a land or naval attack would still not protect inhabitants from aerial bombardment. ⁵⁹ In the context of aerial warfare, the distinction between defended and undefended localities was widely understood to be ‘illogical’ and ‘lead[ing] to absurd results’. ⁶⁰ During the First World War, European belligerents who had used aerial warfare had already partially justified their targeting using the concept of a ‘military objective’: the destruction of targets of a military character would give the attacker a direct advantage. ⁶¹ In their view, the new technology had made the old legal distinction obsolete. The new legal vocabulary was developed by military practitioners and lawyers during the First World War and in its aftermath: it was crafted by the users of the new technology, and not by those who faced the greatest dangers arising from its use. The Hague Rules listed prohibited purposes of bombing in Articles 22 and 23, provided a list of permitted targets in Article 24, and established zones of special consideration in Articles 25 and 26.

Article 22 prohibited aerial bombardment ‘for the purpose of terrorizing the civil population or destroying or damaging private property without military character or injuring non-combatants’. While Article 22 excluded some possible purposes of bombing from the scope of legality, Article 24, drafted by John Bassett Moore, defined legitimate targets: ‘An air bombardment is legitimate only when is directed against a military objective, i.e., an objective whereof the total or partial destruction would constitute an obvious military advantage for the belligerent’ (Section 1). Section 2 defined military facilities that could be targeted: ‘military forces, military works, military establishments or depots, manufacturing plants constituting important and well-known centres for the production of arms, ammunition or characterized military supplies, lines of communication or of transport which are used for military purposes’. Section 3 introduces a distinction between areas near the front and areas remote from (land) fighting. The bombardment of ‘cities, towns, villages, habitations and building which are not situated in the immediate vicinity of the operations of the land forces, is forbidden’. In these areas, even legitimate targets that are ‘situated that they could not be bombed but that an undiscriminating bombardment of the civil population would result therefrom’, may not be bombed. The balance between military goals and civilian protections is struck differently for legitimate targets located in the ‘immediate vicinity of the operations of the land forces’. There, ‘the bombardment of cities, towns, villages, habitations and buildings is legitimate, provided there is a reasonable presumption that the military concentration is important enough to justify the bombardment, taking into account the danger to which the civil population will thus be exposed’. According to participants in the Conference, the negotiations surrounding Article 24 had been particularly tense and time consuming. The permission to attack military targets outside of the direct combat zone was particularly controversial. While the Dutch and Japanese delegations proposed to ban all ‘aerial bombardment of towns and villages outside the immediate area of military operations’, the Italian delegation put forward a draft that allowed strikes on military objectives outside the combat zone only ‘on condition that no hurt is suffered by the civilian population’. ⁶² The US proposed the least restrictive option, allowing attacks on any legitimate objective that ‘is sufficiently important to justify such bombardment, having regard to the danger thus caused to the civilian population’. ⁶³ In the end, Article 24 paragraph 3 prohibited bombardment of military targets ‘not in the immediate neighborhood of the operations of land forces’ unless such attacks can be achieved without ‘indiscriminate’ bombardments of the civilian population.

The Hague Rules no longer divided the landscape into ‘defended’ and ‘undefended’ places, but into military and civilian objects and persons. In making this new distinction, the Hague Rules not only defined ‘military objectives,’ but also for the first time introduced the ‘civilian’ as a specific non-combatant into the vocabulary of international law. ⁶⁴ The civilian population is named in Article 22 that prohibits bombing ‘for the purpose of terrorizing the civilian population’. The civilian population is recognized as both vulnerable – hence they should not be bombed – and an attractive target – hence the prohibition on deliberately bombing civilians was necessary. As Amanda Alexander has shown, discourses of civilian-ness in the years prior to the Hague Rules understood both civilian vulnerability and the essential role of civilians in supporting, legitimizing, and feeding the war machine. ⁶⁵ As we show in later sections, the civilian status envisioned by the Hague Rules and its interpreters was more circumscribed and fragile than might appear at first glance.

Aside from grounding the legality of aerial bombardment in the new – but now familiar – military/civilian distinction and defining the relevant terms, the Hague Rules also distributed the responsibility for the precision of aerial bombardment between the aircrews in the sky and the civilians on the ground. Articles 25 and 26, which established special protection for medical, religious, educational, and historical buildings, command the aircrews to take ‘all necessary steps’ to ‘spare as far as possible’ specific categories of buildings. Yet these same articles also mandate that such buildings ‘must by day be indicated by marks visible to aircraft’ and ‘a belligerent who desires to secure by night the protection for the hospitals and other privileged buildings above mentioned must take the necessary measures to render the special signs referred to sufficiently visible’ (Article 25). These regulations were crafted against the backdrop of knowing that bombs would often miss their intended targets and that crews frequently had trouble visually distinguishing features in a landscape. ⁶⁶ The obligations for hospital staff and others working in protected buildings to secure the inviolability of their buildings with ‘sufficiently’ visible signage opens the door to make civilians on the ground co-responsible for being bombed. As we will show later, these avenues of blaming civilian victims for targeting errors have persisted in the adjudication of airstrikes.

The 1923 Hague Rules responded to a relatively new technology of warfare by adopting new modes of regulation. The Rules departed from the pre-existing model of a complete ban because, as the Commission put it, ‘the aircraft is a potent engine of war’ and no military would ‘take the risk of fettering its own liberty of action’ by voluntarily agreeing to a ban that might not be agreed to by all parties to the conflict. ⁶⁷ The Hague Rules also formalized the switch from allowing the bombardment of ‘defended’ places to permitting the bombardment of specific ‘military objectives’ while taking risks to civilians into account.

4. Conceptualizing civilians

The Hague Rules are the first document in international law to name ‘civilians’ as a category of non-combatants and to demand that they are protected from the effects of war. ⁶⁸ Yet the Rules as well as their interpreters worked with a much narrower understanding of the category of the civilian than the surface of the document suggests: at least in the US and UK understandings of the term, the concept of the civilian did not include non-European non-combatants and also failed to include a range of European non-combatants whose paid labour directly contributed to the war effort. While different authors affirmed the theoretical importance of the distinction between civilians and combatants, they chipped away at the practical scope of the category of the civilian.

When the concept of the civilian started to be included in diplomatic documents and treaties, international lawyers relied on different strands of discourses about the civilian. As Amanda Alexander has shown, the (now familiar) discourse of civilians as innocent, vulnerable, and removed from conflict was an important component of professional conversations and journalistic accounts. ⁶⁹ Writing almost a century after the drafting of the Hague Rules, Rebecca Sutton has identified ‘innocence, harmlessness, and non-participation’ as attributes that humanitarian actors working in South Sudan in the early 2000s understood as embodiments of ‘civilianness’. ⁷⁰ This figure of the civilian as a vulnerable non-combatant elicits sympathy and calls for protection – in the 1920s as well as in the 2020s. Yet not all civilians are ‘perfect civilians’. In early twenty-first century conflicts, we see militaries suspecting ‘local nationals’ in Iraq and Afghanistan of supporting ‘insurgents’. ⁷¹ In the early twentieth century, some writers were concerned that civilians in countries at war would primarily identify as patriotic citizens and support the war instead of passively awaiting its end. German authors in particular shared – largely fictional – stories of the invading German armies having been attacked by duplicitous Belgian civilians. ⁷² Yet authors were not only worried that civilians would directly participate in hostilities, but also that their labour power and their political support would make them an indirect part of the war effort. The understanding of civilians as political and economic actors critical for the war effort was in tension with the idealized representation of civilians as innocent and vulnerable.

4.1 Civilians and civilization

In the broader literature on the civilian status, some case studies suggest a strong gendering of the civilian as female (and concomitantly more innocent and vulnerable). ⁷³ Yet, as Helen Kinsella argues, the discourses of gender and ‘civilization’ are frequently interwoven in different configurations. ⁷⁴ As a result, not all women are presumed to be civilians, and not all civilians are feminized. In the texts we examined, authors discussed aerial bombardment and the fate of civilians as legal problems exclusively by mobilizing examples from intra-European warfare. The civilians they mentioned, or imagined, as being in need of protection were white, ‘civilized’, and European. This pattern tracks with other findings about the explicit and indirect boundaries of the civilian status: Helen Kinsella demonstrates that the 1863 ‘Lieber Code’ imposed obligations of civilian protection on US troops fighting the secessionist Confederacy, but not on troops fighting Indigenous nations within North America. ⁷⁵ Within this group of non-combatants, women were not always presumed to be civilians: a range of authors specifically mentioned that men and women who work in the production of munitions could not claim civilian status. ⁷⁶

A contextual reading of the Hague Rules and the connected discussions suggests that ‘civilian’ was a status restricted to non-combatants from Western countries. In the early twentieth century, international lawyers used the term ‘civilian’ almost exclusively to refer to either European non-combatants within Europe or to white colonial administrators with a non-military portfolio. ⁷⁷ When US and British international lawyers mentioned aerial warfare against non-European non-combatants, they did not use the term ‘civilian’ and rather referred to the local population as ‘natives’. ⁷⁸ Within the texts we have found and analysed, the regulation of aerial bombing and concerns about the fate of civilians were exclusively voiced in relationship to armed conflicts within Europe. The German bombardment of British and French cities was discussed as relevant practice calling for regulation, but the British bombing of Jalalabad and Kabul was not. ⁷⁹

The difference between the permissive rules for bombing within Europe and the dominant insistence that law should not apply to the bombardment of peoples in the imperial peripheries is reflected in Elbridge Colby’s writings. His 1925 article on aerial warfare focuses on wars among states of the Global North, while his 1927 article ‘How to Fight Savage Tribes’ exclusively focuses on the use of aerial warfare in colonial and imperial conflicts. Disagreeing with AJIL editor Quincy Wright, Colby opines that international law does not require the application of the laws of war to ‘people of a different civilization’. ⁸⁰ Instead, he proposes,

the real crux of the matter of warfare between civilized and uncivilized peoples almost invariably turns out to be a difference in fact as well as a difference in law. In fact, among savages, war includes everyone. There is no distinction between combatants and non-combatants. Whole tribes go on campaign. ⁸¹

Colby is careful to stress that the lawlessness of colonial warfare is not due to decisions taken by Western militaries, but a result of the character of the opposing side:

[i]n small wars against uncivilized nations, the form of warfare to be adopted must tone with the shade of culture existing in the land, by which I mean that, against peoples possessing a low civilization, war must be more brutal in type. ⁸²

Colby’s writing exemplifies a specific – and not universally shared – approach to the role of alleged ‘civilization’ in international law and the identification of civilians. While his strident justifications of colonial violence stand out for their violent rhetoric, articles in US and UK international law and international relations journals in the 1920s generally posited ‘civilized’ countries as the sole authors, adjudicators, and beneficiaries of international law. For example, untrammeled aerial warfare was understood to be a threat ‘to the lives and property of every civilized country’ ⁸³ and should not be ‘countenanced’ by ‘any civilized community in the world’. ⁸⁴ The Hague Rules are presented as having been designed for warfare among Western countries only: ‘The rules of war as codified are therefore most likely to be best observed when wars are between civilized nations and on a small scale.’ ⁸⁵

In short, the Hague Rules and the subsequent debates on civilians under aerial warfare were based on the assumption – which is sometimes spelled out explicitly – that the scope of IHL would be confined to armed conflicts among ‘civilized’ countries. As a consequence, the ‘civilian’ was modeled on non-combatants living in Western societies. In a second step, the civilian status as a protected category was further narrowed through a careful consideration of the political and economic support work that made armed conflict possible.

4.2. Targeting civilians

While non-European non-combatants had been tacitly left out of the conceptualization of the new civilian status, military strategists were conceptualizing civilians within Europe as the indirect – and occasionally direct – targets of aerial bombardment. The Hague Rules’ prohibition on indiscriminate bombing for ‘terrorizing’ civilians did not emerge from an uncontested consensus, but against the backdrop of the repeatedly articulated temptation to target ‘civilian morale’ through bombing campaigns. In addition, from the perspective of the civilians on the ground, the distinction between being ‘terrorized’ and having their ‘morale’ attacked seemed spurious at best. ⁸⁶

The First World War within Europe was a war within industrialized nations. The political discourse and propaganda in many belligerent states emphasized that all citizens can and should contribute to the war effort. Far from imagining non-combatants as innocent, apolitical and passive, as certain strands of the discourse on civilians do, the rhetoric of industrialized war addressed non-combatants as citizens whose patriotic duty was to support the war effort through their political, social, and economic activities.

As Amanda Alexander has argued, legal and cultural discourses in the post-First World War period suggested ‘that a war against civilians could be an appropriate way of waging war’. ⁸⁷ After all, civilians had ‘made the policy and sent the soldiers out to their fate’. ⁸⁸ In an imaginary where ‘everyone was involved and implicated’ in war, ‘conflating the people, the nation and the state’, there was little space for the figure of the innocent civilian. ⁸⁹ Within the scholarly debates on aerial warfare, civilians alternately appeared as unjustly terrorized and legitimate indirect targets of the horrors of war.

During and after the First World War, many international lawyers condemned the use of aerial bombardment ‘with the deliberate intention of terrorising the population as a whole’, as G.G. Phillimore phrased it in 1915. ⁹⁰ In 1921, Herbert Manisty declared that ‘the principal object of any convention or rules laid down to regulate the conduct of aerial warfare’ should be to ‘make unlawful’ the bombing of civilian objects ‘for the purpose of terrorising the civilian population and thereby weakening the morale of the whole population’. ⁹¹ In Manisty’s interpretation, the German air raids on London had been illegitimately designed ‘to terrorize and demoralize the civil population’. ⁹² His language prefigured Article 22 of the 1923 Draft Rules, and it also gestured towards the reasoning that air power theorists had employed to target civilian populations: ‘civilian morale’.

Other publicists built a case for targeting civilians directly or indirectly by emphasizing the crucial role of civilians in sustaining the war. For example, British air strategist P. R. C. Groves, writing in the British Journal of the Royal Institute of International Affairs, stated that ‘the defeat of the enemy’s forces is only a means to an end. The real objective is, and has always been, the will of the enemy people’. ⁹³ Even if the bodies of civilians are not the officially intended targets of bombs, their will and ability to support the war as voters and workers had become a target of ‘morale bombing’. Echoing these sentiments, British law professor E. G. de Montmorency argued that the German plan ‘to destroy or vitally cripple London was … a perfectly legitimate object in war’. ⁹⁴ This reasoning is remarkable given that outrage about the German bombing of British and French civilians had been a staple of British communications and propaganda about the war effort. ⁹⁵ Although the General Report of the Hague Conference commented on Article 22 proclaiming that ‘no difficulty was found in reaching an agreement that there are certain purposes for which aerial bombardment is inadmissible’, ⁹⁶ the discussions in professional journals suggest military strategists including Conference participants such as James Moloney Spaight (UK) were tempted to target European civilians either directly – by focusing on their workplaces and homes – or indirectly – by destroying the infrastructures that would sustain civilian lives. The Hague Rules’ prohibition on bombing civilians for the purpose of ‘terrorizing’ them was not a product of a universal agreement, but rather a compromise designed to limit the reach of war. The fine-grained discussions of the place of munitions workers within the civilian/combatant binary provide additional evidence that many European civilians were conceptualized as potentially targetable.

4.3 Jam and guns: Negotiating the civilian status of workers

During and after the First World War, the idea of the passive and vulnerable civilian competed with the discourse of the ‘wilfully dangerous’ civilian who identified as a citizen and patriot. ⁹⁷ Military strategists of the time stressed both the role of ‘the people behind the lines who made the policy and sent the soldiers out to their fate’ ⁹⁸ and the ‘home front’ in which ‘female workers were as essential as male soldiers’. ⁹⁹ On the basis of a functional understanding in which the political support and economic contributions of non-combatants were essential to the war effort, some international lawyers as well as military officers framed non-combatants as ‘a key military target’. ¹⁰⁰

The Draft Rules proposed that most non-combatants would be seen through the lens of the ‘innocent and vulnerable’ rather than ‘active and hostile’ civilian. Yet this framing was not applied to all non-combatants. Workers in munitions factories became understood as essential contributors to the war and therefore legitimate targets. Article 24 of the Draft Rules specified ‘military forces’ as well as ‘factories constituting important and well-known centres engaged in the manufacture of arms, ammunition, or distinctively military supplies’ as legitimate targets. According to James Garner’s interpretation published in the American Journal of International Law, this clause specified that munitions workers would not be legitimate targets per se, but that ‘the munitions factories … in which they work maybe bombarded, so that in fact the immunity covers them only while they are in their homes’. ¹⁰¹ At the same time, Garner speculated that ‘those of the civil population who are engaged in the production of materials of war’ might be considered ‘enemy combatants’. ¹⁰² In the Draft Rules as well as the surrounding discussions, the figure of the European munitions worker hovered at the boundary of the newly forged civilian status. It is worth paying attention to the competing logics of vulnerability and contribution to the war effort that pulled munitions workers into and out of the civilian status.

The writers in US and UK international law and international affairs journals generally showed an understanding of armed conflict that recognized the functional implication of most of the population in the war effort. They did not necessarily advocate for a ‘total’ or ‘totalitarian’ war, ¹⁰³ but they noted the social, economic, and technological changes undergirding warfare in the early twentieth century. Writing in 1921, Montmorency observed that ‘the distinction between the civilian who produces the sinews and weapons of war and the soldier who is in direct action is no longer possible’. ¹⁰⁴ Writing in the British Yearbook of International Law, British military air strategist James Moloney Spaight aimed for a similar line dividing ‘the ordinary population’ from combatants and those who aid the war effort. Munitions workers, he proposed ‘cannot be considered to be entitled to the immunity which otherwise they can claim’ while they are ‘actually at work in the munitions factories’. ¹⁰⁵

The focus on munitions workers as a borderline figure of the civilian/combatant distinction also shows the limits of gendered discourses of innocent civilianhood. Some writers specifically recognized that the munitions workers they were pushing out of the status of fully protected civilians – at least while at work – included significant numbers of women. For example, James W. Garner cautioned that in future wars ‘women who work in munitions factories and in other ways contribute towards the winning of the war, cannot expect to enjoy the immunities of non-combatants’. ¹⁰⁶ Once munitions workers were excluded from bona fide civilian status, it was possible to extend this logic to other industries. Commenting on the phrase ‘distinctively military supplies’ from Article 24 section 2 of the Hague Rules, Paul Whitcomb Williams argued that ‘so much of the output of almost every factory producing anything from jam to steel goes to assist the conduct of military operations’. ¹⁰⁷ ‘No doubt’, he added, ‘this provision was designed to exclude jam factories and include woolen mills making army clothing’, but ‘who is to say that meat is less important to an army in the field than raiment?’ ¹⁰⁸ Williams’ train of thought suggests that there is no clear dividing line between civilians and combatants if the criterion is whether a person functionally contributes to the war effort. It is noteworthy that all the authors discussed the exclusion from different groups of workers from the civilian status on account of their waged labour, but there was no discussion of the economic, intellectual, or political contributions of non-working-class people to the war effort and the implication of such forms of support for their status as civilians.

While many authors advocated removing categories of workers from the category of protected civilians, they also ritually reaffirmed the importance of distinguishing between civilians and combatants. For example, Garner expressed concern that ‘the wars of the future’ might ‘degenerate into struggles of reciprocals reprisals and barbarism, in which no distinction will be made between combatants and non-combatants’. ¹⁰⁹ The distinction between civilians and combatants, he wrote, ‘is fundamental and eternal; it is founded upon considerations of humanity’. ¹¹⁰ At the same time, he had no qualms about moving the line between civilians and combatants and therefore denying legal protections to munitions workers. Similarly, John Bassett Moore, who had chaired the Committee drafting the Rules, wrote in a 1924 essay that abandoning the distinction between combatants and non-combatants would ‘necessarily signify a reversion to conditions abhorrent to every man who cares for law, or for those elementary considerations of humanity the observance of which law is intended to assure’. ¹¹¹ In a speech at the Conference he emphasized that:

the principle most fundamental in character, the observance of which the detailed regulations have largely been designed to assure, is the distinction between combatants and non-combatants, and the protection of non-combatants against injuries not incidental to military operations against combatants. ¹¹²

Even as the principle of distinguishing between civilians and combatants allegedly secures the existence of law and humanity, the Hague Rules and their interpretations in professional journals carefully circumscribe who counts as a civilian under which circumstances.

Taken together, the silent exclusion of non-European non-combatants from the civilian category and the open support for treating European munitions workers (and potentially workers from other industries) as functional quasi-combatants prefigures current debates about the boundaries of the civilian status. The US military in particular has aggressively expanded the categories of people who may be targeted in armed conflict against non-state parties: ‘Military-Aged Males’, ‘hostiles’, and more recently, the owners and workers of drug labs in Afghanistan who had been taxed by the Taliban in return for security. ¹¹³ When the category of the civilian first appeared in an international legal document, it did not include all non-combatants and it allowed combatants to withdraw civilian protections from people whose paid labour or political support contributed to the war effort. While the legal category of the civilian has been heralded as fundamental to the laws of war, it has never had universally protective effects for all non-combatants.

5. Accidental terror: Prohibited and permitted purposes of bombing

While Articles 22 and 23 of the Hague Rules prohibited the use of aerial bombardment for the purposes of ‘terrorizing the civil population’, ‘destroying or damaging private property without military character’ or enforcing ‘requisitions’ or debts, Articles 24 to 26 spelled out what kinds of targets could be bombed under which circumstances by enumerating legitimate military targets, mandating a balance between civilian harm and military advantage, and making some civilians co-responsible for the recognizability of civilian objects from the sky. The logic of these articles focused legal attention on specific questions about airstrikes: the intention behind the bombing, the specific target, the measure of proportionality, and the distribution of blame for target misidentification. A century later, these legal questions still largely shape the adjudication of the legality of airstrikes. Yet this legal framework also leaves the purveyors of violence ample argumentative space to legitimize airstrikes that result in harm to civilians. As early commentators of the Hague Rules pointed out, the regulatory framework left militaries with space for justifying and excusing civilian harm. ¹¹⁴ One form of defending targeting decisions was to argue that the persons targeted were not actually civilians. We have addressed exclusions from the civilian category in the preceding section. Here we focus on two additional connected excuses that have been built into the architecture of the Hague Rules: disavowing intent and blaming damage on the unpredictability of technology.

5.1 Good intentions

In late twentieth century and early twenty-first century aerial warfare, civilian deaths have frequently been described as ‘accidents’, ‘collateral damage’, ‘tragedies’ and ‘errors’ – rendering them events for which the aircrews cannot be fully held responsible. ¹¹⁵ Neta Crawford has critiqued these discourses, arguing that many forms of ‘collateral damage’ are not unforeseen, but systemically predictable results of dropping bombs on densely populated areas. ¹¹⁶ This analysis treats human intention and the known limitations of technologies as connected in the process of decision-making. For Crawford, predictable but not specifically intended damage falls within the moral responsibility of the purveyors of violence.

The high-tech weaponry of ‘later modern war’ features much more technological precision than the weapons used in the First World War and the interwar years. ¹¹⁷ Whereas in twenty-first century warfare, the bombs typically fall on their targets, this was not the case in the early twentieth century. ¹¹⁸ In the 1920s, aerial warfare was widely understood as a new complex technology that was powerful but imprecise. Distinguishing legitimate targets from other objects was hard, and pilots repeatedly struggled with basic navigational challenges, especially in unfavourable weather. Bombs regularly fell far away from the intended target, which means that they destroyed non-targets. In British ‘air control’ of Iraq, for example, ‘it was not unusual for aircraft to bomb the wrong town’, but, as Priya Satia argues, the imagination of Iraq ‘as an essentially deceptive place made such errors tolerable and acceptable’. ¹¹⁹ British forces tolerated and defended this lack of precision because of their Orientalist vision of Iraq as ‘a place in which indiscriminate violence did not matter’ because the population was ‘congenitally insurgent’. ¹²⁰ In Iraq, Satia writes, ‘there were no civilians’. ¹²¹ Within Europe, however, the recognition of the distinction between civilians and combatants had been declared a matter of ‘civilization’. Yet the limits of the technology also shaped the limits of the combatants’ responsibility towards civilians.

Early twentieth century international lawyers and military strategists knew that fine-grained legal distinctions between legitimate and illegitimate targets would not be realistically observable and achievable in the practice of aerial warfare – whether in Iraq or in Europe. Writing in the early years of the First World War, Phillimore suggested that ‘it seems important to insist on the duty of aircraft to use the utmost practicable care in this respect to prevent damages from extending beyond the proper objects of attack’. ¹²² Yet he cautioned that considering ‘the height in the air from which an air attack is launched, it is manifestly difficult to fix responsibility on the attacker for damage done to other objects than the proper ones’. ¹²³ When the Draft Rules were written, authors operated with the assumption that precision in aerial warfare was largely impossible. This knowledge relativized their commitment to civilian protection. For example, James Garner noted that ‘[t]o require aviators to single out one class of persons and things from the other and to confine their attacks “exclusively” to one of them will in many cases be tantamount to an absolute prohibition of all bombardment’. ¹²⁴ Since the option of banning aerial bombardment was no longer seen as tenable in the 1920s, these authors tacitly recognized that the conceptual architecture of the Hague Rules would not be directly translated into the practice of warfare. Due to technological and human errors (or claims of technological or human errors), civilians would inevitably be harmed. The Hague Rules would supply the legal concepts to debate, justify, and legitimize harm to civilians. Militaries would be able to use a range of mechanisms for disavowing responsibility for civilian harm, including arguing that the victims were not civilians, insisting that they were not targeted intentionally, and claiming a number of civilian casualties as incidental harm not out of proportion to the military objective.

For early twentieth century aircrews, accurate target identification remained a persistent challenge. Writing in 1921, Herbert Manisty argued for making aviators responsible for the limitations of the technologies they used: ‘if aviators are unable in night raids to distinguish Red Cross signs and lights, they must refrain from night raids’. ¹²⁵ The Draft Rules did not endorse this perspective. Instead, they only demanded that aviators take ‘all necessary steps’ ‘to spare as far as possible’ protected buildings (Article 25). The aircrews were not to be held responsible for risky decisions that made them operate at the boundaries of what was technologically possible and controllable. Instead, civilians on the ground were conscripted in creating the conditions of visibility that would enable air crews to not target historic buildings or hospitals: they were obliged to mark these buildings ‘by marks visible to aircraft’ and ‘must take the necessary measures to render the special signs referred to sufficiently visible’ from the air (Article 25). The co-responsibility of civilians on the ground for making protected buildings visible from the air can easily be turned against them in case hospitals are bombed. This line of reasoning that makes victims on the ground partially responsible for target misidentifications has an enduring legacy. For example, in 2015 the US military suggested that it had mis-identified the MSF Trauma Center in Kunduz as a target and bombed it in part because the building had not been marked by the red cross or crescent symbol. ¹²⁶ The regulatory structure of the Hague Rules transfers not only the physical risks, but also some legal responsibilities for aerial warfare onto civilians on the ground.

The writings by Elbridge Colby illustrate another logic of disavowing responsibility for civilian harm that is based on the language of the Hague Rules. Writing in the American Journal of International Law, the US Army Captain described the practical work of targeting:

Sometimes the town containing the military objectives of his flight is revealed to him only by a river line or some very distinct natural landmark of some other kind, from which he must calculate his distances and estimate the location of his objective. When he attacks in this fashion, innocent people are bound to be struck. How can the man across the street from the General Post Office in London be safe, when sixty yards is laid down as the average striking distance from a thirty-foot target which is attained only at an advanced stage of bombing training? Because a tobacconist or a haberdasher has a little shop over there, shall the enemies of England be compelled to refrain from using their aerial power to strike at the center of postal and telegraphic communications of the Empire? Such deadly accuracy as would be needed to demolish the government building with one or more bombs and have none fall anywhere else, is not probable. It is not reasonable to expect such accuracy under the conditions under which aviators have to work. ¹²⁷

In this passage, Colby combines different logics: on the one hand, he points to the complications of navigation to suggest that aviators will inevitably misidentify targets. On the other hand, he uses the logic of ‘collateral damage’ to argue that civilians within the vicinity of a legitimate target – particularly in an urban area – would not be safe because it was impossible to confine the impact of the bomb to the direct target. (Colby’s assumption that the General Post Office would be a legitimate target does not seem to be in conformity with Hague Rules, which underlines to the capacity of militaries to argue for expansive definitions of ‘military objectives’). In this text, Colby suggests that the effects of bombing would regularly exceed the stated and actual intentions of the aircrews, and he comes close to arguing for a right to cause extensive collateral damage:

No belligerent should be required to forfeit the normal percentage of hits which might be expected on his target, simply because there will be a percentage of ‘misses.’ … By his effective five per cent. he may destroy his ‘military objective’ wherever the other ninety-five per cent. may go. It is not a question of intent to hit civilians instead of military depots, or of an intention to terrorize generally. Like the actuary figuring expectant mortality for a life insurance company, he cannot foretell what will happen in any individual case, but he can tell what his average will be. His intent is to place ‘the maximum number of hits’ on his target according to his average accuracy. ¹²⁸

In Colby’s view, the five percent ‘accurate’ hits are a sufficient reason for allowing a technology that predictably will result in 95 percent of bombs landing elsewhere. With the ever increasing technological precision of bombs, the rate of bombs that do not fall on their intended target diminished over the course of the twentieth and early twenty-first century. While twenty-first century bombs might technically be capable of destroying the post office without damaging the shop across the street, the recently released files of US military investigations into civilian casualty allegations arising from airstrikes in Iraq and Syria suggest that while the promise of technological precision leads to the increasing use of bombs in urban warfare, it is frequently decoupled from the human ability to correctly assess the status of buildings as legitimate targets or civilian residences.

The Hague Rules focused on prohibiting the intentional bombing of civilian targets. This mode of regulation suggested a very limited responsibility for bombs that would either hit something else than the target or destroy lives and infrastructure beyond the immediate target area. As Elbridge Colby commented in 1925: ‘the draft articles do not say that the bombs must fall exclusively on military objectives, only that they must be directed exclusively at such’. ¹²⁹ This technique for regulating bombing rendered predictable collateral damage legally unproblematic.

6. Judging the Hague Rules

The Hague Rules were written not as a draft treaty, but as a basis for possible future treaty negotiations. ¹³⁰ Yet no treaty governing aerial warfare was ever negotiated. Instead, the post-Second World War Geneva Conventions and the 1977 Additional Protocols that addressed the different modalities of warfare largely without distinguishing between aerial, naval, and land warfare. ¹³¹ While the Hague Rules were not directly converted into a treaty, they have introduced key concepts and forms of regulation into the repertoire of international lawyers. For example, the Hague Rules prohibited direct attacks on civilian populations, defined legitimate military targets, and introduced different forms of balancing of civilian harm and military advantage. Heinz Marcus Hanke argues that these regulatory techniques have been ‘reworded, considerably expanded, made more specific, and modified’ in the 1977 Additional Protocols. ¹³² Given this influence of the 1923 Hague Rules, it is worth recalling the contemporary responses to these new forms of regulation. How did commentators react to the newly proposed regulatory infrastructure? Did they think that the chosen approaches would meaningfully constrain aerial warfare against civilians?

Upon reviewing the Hague Rules in the 1920s, different authors recognized that these rules gave a lot of responsibility and discretion to air crews: ‘To a much larger degree than in land and naval warfare they are made the judges of the legitimacy of their attacks,’ ¹³³ James Garner argued in AJIL. He worried that ‘aviators will take large chances’ and ‘broadly interpret their rights and consider whatever damage may result to the civil population from their bombarding operations as merely incidental to the accomplishment of a military advantage, and therefore justifiable’. ¹³⁴ He added that ‘for this reason the rules proposed may not prove to be a very effective limitation upon their conduct’. ¹³⁵

Garner’s legal analysis was confirmed and radicalized by the commentary of Elbridge Colby, a key US advocate of aerial warfare. Colby approached the Hague Rules from the perspective of a potential commander or crew member. The specific forms of regulation embodied in the Hague Rules would allow officers to obscure their intentions as well as errors:

The bombing will ostensibly be at military objectives. If the man-power of the nation is reduced, if the manufacturing efficiency of the nation is hurt, if the morale of the nation is lowered, so much the better; but of course the strategic statesman and the commander who orders his planes out will speak only of military objectives and will wave the document as his justification. ¹³⁶

Colby’s perspective is that of a potential user of the technology and its regulation. Rather than asking what the law is, he is asking how he can use it and what it can do for him: what are the key legal concepts, how can relevant facts be ascertained, and which excuses and justifications are potentially available? It is important to pay close attention to these publicly disclosed engagements with newly proposed regulations of warfare. As Eliav Lieblich reminds us, law can have a ‘facilitative function’ by legitimating conduct ‘irrespective of its formal position’. ¹³⁷ For example, rules that prohibit specific actions might implicitly condone other actions or establish exceptions under which the prohibited conduct can nonetheless be justified. ¹³⁸ The legal rules get invoked, used, and become ‘an integral part of the war discourse’ that legitimates both the armed conflict and specific conduct of belligerents. ¹³⁹ After the Hague Rules were printed and circulated, military officers like Colby freely discussed the possibilities for justifying or excusing civilian harm with reference to the regulatory logic of the Hague Rules themselves. The Rules were understood to be facilitative of aerial bombardment.

International lawyers were aware that the vague legal terms used in the Draft Rules would leave space for claims of technological and human errors that would render civilian harm legally unobjectionable. James Garner called the rules ‘an earnest endeavour to reconcile in a just manner the legitimate rights and interests of belligerents with those of the non-combatant population in particular, and the rights of humanity in general’. ¹⁴⁰ He concluded that ‘whatever may be their defects’, these rules were ‘better than no rules at all’ because ‘there is at least a chance that they may serve to deter belligerents from illegal conduct’. ¹⁴¹ Paul Whitcomb Williams held an even more pessimistic view. He called the Hague Rules ‘a halfway measure, worse than useless, for it permits a belligerent to do under the guise of legality what he might otherwise scruple to do’. ¹⁴² Williams’ response is based on the understanding that the regulation of warfare is negotiated less frequently in courts of law than in the forum of public opinion. The Hague Rules offered belligerents that used aerial warfare a range of excuses and justifications for civilian harm that would legitimate such conduct in the eyes of the (western) publics to whom the militaries would be accountable. While the Hague Rules had never been converted into a treaty specific to aerial warfare, the regulatory logic focused on the intent behind targeting, the taking of feasible precautions, and the balance between military objectives and civilian harm has been incorporated into IHL in the decades that followed. In our discussions about the ‘facilitative’ and ‘legitimizing’ potential of IHL, it is worth remembering that a century ago the Hague Rules that introduced some key IHL concepts were met with scepticism about their ability to meaningfully reign in aerial bombardment.

7. Connections and conclusions

The contemporary discussion of the 1923 Hague Rules shows that the first attempt to regulate aerial warfare involving planes was widely understood as an effort to manage and facilitate rather than restrict aerial warfare. The Hague Rules limited the circumstances under which civilian targets could be under attack and prohibited the bombing of civilians for the purpose of terrorizing them. The frequently emphasized commitment to the distinction between civilians and combatants – as well as civilian and military targets – went hand in hand with attempts to move and blur the lines between civilians and combatants. In the collective process of interpretation, the impact of the Hague Rules was limited to specific kinds of civilians. The regulatory framework and the subsequent discussions by British and US legal and military professionals tacitly excluded non-European non-combatants. The ink on the Hague Rules had not yet dried when commentators started to make the case that munitions workers – and maybe also jam factory and steel mill workers – cannot enjoy civilian protections while at work because their labour contributed to the war effort. In addition, the Hague Rules offered belligerents a range of excuses for civilian harm on account of technological errors and imprecision.

The 1923 Hague Rules, we argue, inaugurated patterns of legal arguments about warfare that allowed belligerents from the Global North to disclaim responsibility for the effects of aerial warfare insofar as they were ‘unintended’ or ‘unforeseen’. ¹⁴³ Aerial bombing is used by states that have the economic and technological resources to buy and maintain the planes and drones and to train and employ the crew members. Aerial warfare is offensive, not defensive. It is therefore largely used by industrialized states with imperial ambitions. The legal privileging of aerial warfare therefore affects a range of armed conflicts that are highly asymmetrical. The Hague Rules are also part of a history of international law that restricted the civilian status to white non-combatants and was quick to question the ‘civilianness’ of non-combatants whose paid labour or political engagement could impact the outcome of the war. ¹⁴⁴ These legal regulations of aerial warfare were, as the discussion in the 1920s showed, jointly produced by the users and regulators of these new technologies in a process that made (proposed) international law fit the needs and perspectives of Western militaries. While the peoples of the Global South had been the dominant targets of aerial warfare at the time the regulations were developed, their perspectives were not represented at the negotiation table. The legal grammar of the Hague Hague Rules of Aerial Warfare converts the technological asymmetry between militaries of the Global North and peoples on the peripheries of empire into a legal asymmetry that renders the purveyors of aerial violence rarely responsible for the effects of high-tech warfare.