Intercontinental aviation in the inter-war decades was synonymous with imperial aviation. As Europe’s established empires grappled with growing anti-colonial movements, strike waves, and unrest in many of their overseas possessions, government officials and political thinkers in the 1920s and 1930s recognised new aircraft technology as a powerful and novel tool for maintaining or restoring order.Footnote 1 Not only aerial bombing but also commercial aviation was to play an important role in safeguarding imperial rule: regular flights could provide a steady flow of mail and passengers between capitals and distant colonies and dominions. To strengthen imperial unity and governance, governments tasked state-owned or subsidised national airlines with the development and operation of new continent-spanning long-distance air routes. Initially concentrating on their key dependencies in Asia, the major airlines of Britain, France, and the Netherlands worked to link their metropolitan centres with India, French Indochina (present-day Cambodia, Laos, and Vietnam) and the Netherlands East Indies (Indonesia), respectively. Meanwhile, Japanese air networks expanded into the Asian mainland, while the US airline Pan American Airways pushed westward from San Francisco to Hawaii, the Philippine Islands and other US possessions in the Pacific.Footnote 2
As these different networks expanded and interlaced, the air corridor over the Middle East, South Asia, and the Pacific region emerged as what Jeppe Mulich has termed an ‘inter-imperial microregion’ – a zone of dense, overlapping imperial interests and activities.Footnote 3 The technological limitations of the time meant that airplanes could only travel short legs. The Handley Page H.P.42 aircraft, used for flights within the British empire, for instance, had a range of about 800 kilometres, while the Fokker F.XII, used on Dutch imperial air routes, could cover about 1,200 kilometres. With headwinds and a reasonable payload, the actual maximum range was usually shorter. Flying from Amsterdam to Jakarta, therefore, entailed a journey of more than ten days with stopovers at seventeen places. Hence, it was virtually impossible to avoid traversing and landing in each other’s colonial possessions.
At the same time, commercial aircraft were not allowed to freely access foreign airspace. In the aftermath of the First World War, as part of the new international order, delegates to the Paris Peace Conference and its attendant meetings worked out an international framework for aviation: the Convention Relating to the Regulation of Aerial Navigation, also known as Paris Convention of October 1919.Footnote 4 This convention reserved the rights of governments to grant or deny access to foreign aircraft, thus asserting the principle of national sovereignty over the air. That meant that in order to reach their respective colonial destinations, the various airlines needed to obtain flyover and landing rights from all national and imperial governments whose territories they traversed.
The ensuing inter-imperial diplomatic negotiations over airspace access and airport usage in the late 1920s and 1930s are at the centre of this article, which focuses on air routes to the ‘Far East’. The article argues that the global governance of inter-war aviation cannot be fully explained by the frameworks provided by conventions and institutions alone. Aviation indeed saw extensive international regulation in the period under consideration: within a decade of its inception, commercial flight became the subject of over fifty committees across various organisations, including the League of Nations’ Organization for Communications and Transit and the Commission Internationale de Navigation Aérienne (CINA).Footnote 5 However, while these institutions codified the principle of national ownership of the air, they failed to set clear procedures for obtaining flyover and landing rights in foreign skies. Instead, decision-making reverted to the horizontal level of bilateral discussions; and it was there that diplomatic legations, government offices, and airline representatives played a pivotal role in shaping widely accepted practices and principles.Footnote 6
History writing has recently seen growing scholarly interest in the evolution of the international aviation order. Particular attention has been given to the International Civil Aviation Conference held in Chicago in 1944, which sought to set the rules for flying in the post-war world. The central point of disagreement at the conference concerned the rights states should obtain for their airlines abroad, with a British bloc advocating a regulatory framework and a US-led bloc championing freedom of the air. Recent studies have highlighted the effects of this Anglo–American rivalry on network formation during the Cold War. They have also revealed how the Chicago system affected the Soviet world and how marginalised actors challenged this order from its periphery.Footnote 7
While the formation of global air networks since the Second World War is thus receiving increasing attention, there is still a great lack of historical research on their prehistory, specifically the geopolitics of air route development in the pre-‘jet age’ era. The few existing studies that look at bilateral negotiations in the inter-war years have tended to depict them as highly competitive and contentious: most prominently, Marc Dierikx identified the competition for prestige as a key driver in Dutch clashes with the British and US governments, leading him to conclude that ‘aviation was one of the more visible means by which imperialist rivalry was continued in Africa and Asia’.Footnote 8 In a similar vein, Alan Dobson, in his longue durée study of civil aviation as an arena for political conflict, highlighted disputes over airspace and the ‘system of highly politicized bargaining for commercial rights’ in that era, which he described as ‘predatory bilateralism’.Footnote 9
Contemporary observers indeed singled out inter-imperial competition as a hindrance to the expansion of long-distance networks, noting by the early 1930s that intercontinental aviation was ‘turning more and more to the idea of nationalism and guarded frontiers’.Footnote 10 Given their many dispersed overseas territories, sovereignty over both metropolitan and colonial airspaces – and thus the ability to deny entry – gave imperial governments substantial leverage in stalling foreign network growth. And yet, within a single decade commercial aviation came to connect places as distant as London and Brisbane, Amsterdam and Jakarta, or Marseille and Saigon. By the eve of the Second World War, intercontinental air routes spanned over 100,000 kilometres worldwide.Footnote 11 This apparent paradox of rapid network expansion amidst a political climate later described as a worldwide ‘counter-revolution in accessibility in the air’Footnote 12 warrants further investigation.
The key to understanding this paradox is exploring the hitherto overlooked dynamics of inter-imperial air diplomacy. Drawing on archival materials collected in Great Britain, France, the Netherlands, and the United States, this article examines the possibilities and limitations governments faced in connecting their globally fragmented imperial spheres, on the one hand, and exercising their rights to sovereign control over their airspace, on the other.Footnote 13 With this analysis, the article positions itself at the intersection of recent aviation history writing and wider research on the international order of the inter-war period. Waqar H. Zaidi has already shown that the competing visions for organising the emerging air age before and during the Chicago Conference were rooted in inter-war debates on the internationalisation of the world’s airways.Footnote 14 Historians working on post-war aviation have also put the international framework implemented in the second half of the 1940s in relation to its earlier iteration, arguing that the new system ‘tamed the main excesses of its “predatory” predecessor, while still making the strongest aviation nations the primary gatekeepers of the system’.Footnote 15 Still, the evolution of this framework and its underlying dynamics have remained largely unexplored. The following analysis thus advances recent scholarship by providing an important analytical basis for embedding post-war developments in pre-war dynamics.
Moreover, by focusing on the practical application of air sovereignty, rather than its legal and intellectual evolution, this article foregrounds an approach that, beyond aviation, also has broader implications for the study of international relations prior to the Second World War. Historiography has commonly characterised the inter-war period as a time of growing internationalism amidst mounting nationalism: observing a nascent ‘global community’, scholars have highlighted the role of new international organisations, chief among them the League of Nations and its many subdivisions, which served as both sites of ‘imperial internationalism’, enabling empires to uphold their power, and forums for non-state actors and pressure groups.Footnote 16 International cooperation, according to Daniel Gorman, shifted ‘from a pattern of interlocking bilateral relationships to one of organized multilateralism’,Footnote 17 as states increasingly came to regard intergovernmental institutions as a means of managing cross-border flows. However, as many recent studies adopted either above-state-level or below-state-level perspectives, they risk overlooking the state level itself. Critics have noted a tendency to ‘write the state out of history’,Footnote 18 that is, to obscure the power that state governments and (even more so) empire-state governments retained in shaping the international order.
The little-explored ‘inter-imperial microregion’ of the air corridor to Asia provides an important case study for reevaluating the influence of both institutions and state-to-state interactions on the formation of global norms. Inter-imperial air diplomacy was crucially shaped by two factors: the rudimentary framework established by international conventions, which – while leaving key questions open – nonetheless created conceptual path dependencies for subsequent bilateral negotiations; and the geography of imperial globalisation with its distinct dynamics. Bound to navigate these forces in the absence of comprehensive regulation, imperial brokers gradually developed their own code of conduct for acquiring and granting air access, effectively supplementing the legal codification of airspace.
This article is divided into three sections, each of which is devoted to a separate (yet interrelated) case study. The sections reveal the various exchange logics that emerged on the horizontal level of deal-making, thereby demonstrating that the concept of air sovereignty did not function as an isolated doctrine but rather operated in conjunction with a key counterpart: the concept of reciprocity, which can be defined as ‘exchanges of roughly equivalent values in which the actions of each party are contingent on the prior actions of the others in such a way that good is returned for good, and bad for bad’.Footnote 19
The first section begins by outlining the conceptual evolution of sovereignty. It then goes on to discuss the British approach to air diplomacy. Although the British government was eager to exploit its own geographical position against perceived competitors, its leverage was constrained by Britain’s own ambition to develop global air links. As will be seen, there was extensive geographical overlap when it came to air interests, and this compelled state actors to remain accommodating to each other. The second section underscores the notion of interdependence within the sovereignty–reciprocity nexus by exploring the extent to which French officials – in an attempt to bolster French aviation in instances where it would otherwise be outpaced – were able to enforce cooperation by leveraging control over their airspace. The third section looks at the Pacific terminus of different imperial routes. Tracing debates about connecting American and European networks in the mid- to late 1930s, it assesses how firmly the principles that had by then become standard among Europe’s air-going empires held in the lead-up to the Second World War.
Navigating Hostility: Aerial Relations between Sovereignty and Reciprocity
Before the twentieth century, the airspace was empty and no law was required to govern it. When the inventions of the Wright Brothers and Ferdinand von Zeppelin began to make motorised air travel a realistic mode of future transportation, they also prompted new questions about how to regulate the third dimension. Debates among legal scholars and politicians swirled around the opposing pillars of freedom and sovereignty: could governments restrict air navigation, or were pilots free to fly wherever they pleased? What kind of space was the air: part of the national territory on the surface, or a free zone analogous to the high seas? The ocean provided the most obvious point of reference, having long been governed by the freedom of the seas and the right of innocent passage through territorial waters. However, ship traffic along maritime borders appeared less dangerous than foreign aircraft crisscrossing a country’s interior and potentially landing within it.Footnote 20
States in Europe and worldwide thus opted for national sovereignty over the air. After an international conference to codify airspace, held in Paris in 1910, had failed to produce a joint agreement, Great Britain became the first country to unilaterally proclaim full air sovereignty over both its mainland and its overseas territories in 1911 and 1913, respectively.Footnote 21 Also in 1913, France and Germany agreed on a set of rules governing the operation of aircraft over each other’s territory, following previous overflight incidents.Footnote 22 With the First World War and its air raids, the principle of air sovereignty gained wide acceptance, in not only Europe but also the Americas and Asia. The Paris Convention of 1919 stipulated in Article 1 that ‘every Power has complete and exclusive sovereignty over the air space above its territory [. . .], both that of the mother country and of the colonies, and the territorial waters adjacent thereto’.Footnote 23 Even though important states like the Soviet Union, Germany, Brazil, and the United States did not sign or ratify the Paris Convention, this core principle gained global traction: as Sean Seyer has recently demonstrated, even non-signatories such as the United States soon aligned their national air laws with it.Footnote 24
Immediately after the First World War, airline companies in different parts of the world began developing regional and transregional networks.Footnote 25 To harmonise the different national aviation laws, the signatories to the Paris Convention established a permanent regulatory body, the Commission Internationale de Navigation Aérienne (CINA). This organisation addressed practical matters related to trans-border aviation, including standards for determining the airworthiness of aircraft, regulations on lighting and signage and the standardisation of maps and meteorological data.Footnote 26 By contrast, the pragmatic hands-on approach of CINA’s many subcommittees meant that the group failed to engage with the more delicate political side of air accessibility. Even though the Paris Convention generally permitted innocent passage over a state’s territory, Article 15 of the convention defined special rules for commercial aviation, stipulating that ‘the establishment of [foreign] international airways shall be subject to the consent of the states flown over’.Footnote 27 The contracting states revised this article at the CINA meeting of June 1929 when they replaced ‘consent’ with ‘prior authorization’, implying a more formal process.Footnote 28 However, CINA’s members never specified the conditions and criteria for such authorisation, leaving it to diplomatic legations to negotiate them on a case-by-case basis.
The question of how to obtain flyover and landing rights gained urgency towards the end of the 1920s, when imperial airline companies began developing links between continents. British efforts originally focused on an air route to India. The British initially hoped that their state airline, Imperial Airways, could inaugurate this service by 1927, with plans to later expand it to Australia. After some delays, the route to Karachi, in present-day Pakistan, became operational in April 1929.Footnote 29 Meanwhile, France and the Netherlands were engaged in establishing their own intercontinental air connections. In 1927, the French government provided the first subsidies for an air route from Marseille to Beirut, which was to be gradually extended to Saigon in French Indochina.Footnote 30 In the same year, the Dutch started actively developing their plans for air services to Indonesia.Footnote 31
There were a few instances in which independent, ‘non-imperial’ governments withheld the necessary permissions for these routes, most prominently Persia and Siam (present-day Iran and Thailand), both signatories to the Paris Convention.Footnote 32 By and large, however, the governments of comparatively small states rarely turned down imperial requests, as many wanted their countries to be integrated into budding transport networks. By 1929 more than 500 bilateral treaties had been concluded worldwide.Footnote 33 Typically, it was not smaller states but rather empires that represented the greater political obstacle to the expansion of intercontinental routes. Imperial governments carefully weighed whom to grant access to their overseas possessions.
On their way to the colonial world, imperial airlines had to fly over and land in each other’s territories: the first foreign country on Imperial Airways’ planned route to India was France, and the extension of that route to Australia went through the Netherlands East Indies. On the way to their own Asian colonies, the Koninklijke Luchtvaart Maatschappij voor Nederland en Koloniën (Royal Dutch Airlines for the Netherlands and Colonies, called KLM) and the French airline Air Orient (which later merged with other French carriers to become Air France) had to cross mostly British territory east of the Mediterranean. Wherever the routes of these airlines overlapped, tensions were likely. Economic rivalry arose over the scarce payload, particularly when it came to the main source of income, air mail. Economic conflicts were intensified by geostrategic considerations and the fact that air routes were regarded as a symbol of imperial power. Being the first to establish regular flights to imperial outposts was a way for an empire to demonstrate its strength both externally and internally.Footnote 34
For these reasons, British authorities deemed it wise not to admit foreign airlines into their colonial realm. Deutsche Lufthansa, founded in 1926, encountered this prohibitive attitude firsthand while planning a Berlin–Beijing connection. Although Germany had lost its colonial empire in the First World War, the airline was conceived as a promoter of national ingenuity and technological prowess, encouraging German trade and investment to follow its wing, and accordingly developed an air route to China, where the German–Chinese airline Eurasia had established a domestic network.Footnote 35 Seeing the planned transcontinental link as an instrument of Nazi soft power, the British authorities prohibited the Berlin–Beijing route from traversing British territory and rejected an application for access to Burma in 1935.Footnote 36
British opposition to granting flyover and landing rights to foreign aircraft en route to Asia was also aimed at the Dutch, as Marc Dierikx has shown. In mid-1929, shortly after the British air service to Karachi, located in the western part of the Indian subcontinent, had been inaugurated, KLM applied for authorisation for a regular Dutch service across India to its own colony of Indonesia. The British Air Ministry warned that, if permitted, the Dutch would be the first to fly across the expanse of British territory on the Indian subcontinent, because a British service eastward of Karachi was still in the planning stage.Footnote 37 In order to preserve His Majesty’s prestige, the ministry recommended that ‘all possible steps should be taken to delay the inception of the Dutch service over Indian and Malayan territory’.Footnote 38 The fact that Britain controlled numerous regions along the planned Dutch air route gave it a significant stronghold against competition. Because CINA had not laid out the conditions of commercial access, Article 15 of the Paris Convention put Britain – as well as any other signatory – in a position to prevent foreign airlines from establishing scheduled air services over their territories, without even needing to negotiate or provide any reason for the refusal.Footnote 39
And yet, although the British could theoretically cite the convention to block Dutch competition, their subsequent handling of the Dutch request reveals the limitations of the written word and thus of purely obstructive tactics: unlike Lufthansa, whose application was denied until the late 1930s, KLM was permitted to fly through Indian airspace after some diplomatic back-and-forth. One important reason for this divergent outcome was certainly Germany’s increasingly aggressive behaviour after the Nazi rise to power. However, as will become clear in the following pages, KLM’s authorisation also stemmed from the fact that the Netherlands, unlike Germany, maintained a territorial empire.
The imperial globality of the inter-war period significantly hampered empire-states’ ability to exercise full sovereignty over their airspace in practice. Observing London’s adversarial stance to KLM’s plans in mid-1929, the Government of India pointed out that ‘placing obstacles in the way of the Dutch seemed doubtful having regard to the fact that their co-operation would be essential for the extension of [British] imperial services beyond Singapore’.Footnote 40 Unilateral air network expansion was impossible: just as KLM needed access to India, Imperial Airways would eventually need permission to fly over and land in the Netherlands East Indies once the airline was ready to expand its service to Australia, the ultimate terminus of Britain’s planned empire route. At that point, if KLM’s request for access to India had not been met, then it would be easy enough for the Dutch government to simply refuse the concession required by Imperial Airways. KLM director Albert Plesman indeed urged his government to leverage this fact in their negotiations.Footnote 41
However, intensified Dutch pressure was not necessary as British officials pre-emptively avoided overly harsh actions. The Government of India faced a difficult dilemma over the future of the subcontinent’s air network. On the one hand, it had a strong interest in developing connections with England and distant outposts, with the goal of reinforcing imperial cohesion. On the other hand, it also insisted on building its own domestic network but was unable to implement a workable plan for the time being, which is why Imperial Airways’ routes initially extended no further than Karachi. Accordingly, the Government of India chose a compromise approach, aiming not to block the Dutch application but to delay it, and this approach was agreed upon with the metropolitan government.Footnote 42 To prevent Dutch retaliation, the British Foreign Office, the India Office, and the Colonial Office jointly informed the Governor-General (Viceroy) of India that ‘it is most important to avoid any appearance of indefinite obstruction of the establishment of a regular Dutch service linking up mother country and her colonies. His Majesty’s Government would therefore like to soften the blow to Dutch aspirations.’Footnote 43
To appease the Dutch, the Foreign Office recommended giving them a definite date for their passage through India: October 1931 at the latest. An apologetic explanation for the delay was to be delivered, even though the Paris Convention permitted refusal of access without the need to justify it.Footnote 44 Of course, the true motive – to give British aircraft a head start on the route across the subcontinent – was not to be disclosed to the competitors. Instead, the British authorities offered the lack of airport facilities as a justification: the purportedly inadequate ground infrastructure in India made a regular service unfeasible at that time. Once additional facilities were available, they promised, foreign airplanes could fly.Footnote 45
Because of Anglo–Dutch interdependence, all Britain could do was delay the start of the foreign air service, not fully prevent it. This allowed the Dutch to press their claims. In December 1929, the Dutch envoy in London, René de Marees van Swinderen, renewed his government’s request for permanent air access. He observed that pilots had recently found the Indian ground infrastructure to be perfectly suitable and concluded that the earlier justification was now invalid.Footnote 46 A few days later, the British Foreign Office and the Air Ministry agreed that retaliation on the future air route to Australia was likely if they refused to be accommodating. To ensure cooperation, they opened up India’s airspace.Footnote 47 The terms of this permission, however, were mutual: Britain offered a deal to the Dutch government, granting them air access to India from October 1930, so that KLM could finally inaugurate its service to the Netherlands East Indies. This was on the condition that Dutch authorities would later allow a British service to fly to Australia through Dutch territory – reciprocity, that is.Footnote 48
Reciprocity functioned under the maxim do ut des, meaning ‘I give so that you may give’. This transaction allowed both sides to operate flights to their far-flung colonial outposts. Yet, even years after air rights had been exchanged, the British government remained cautious in its dealings with its Dutch counterpart. They recognised that their fellow imperial power was still in a position to block air links to Australia: the reciprocal agreement of 1930 clearly stated that the Dutch authorities would allow a British airline to fly over Indonesia on the projected England–Australia route. In the meantime, however, Britain had decided to extend Imperial Airways’ service only as far as Singapore; a connection was eventually established in December 1933. Singapore was envisioned as a meeting point for Imperial Airways and the Australian airline Qantas, with the Australians flying the Singapore–Brisbane leg of the Empire route. Thus, the passage over the Netherlands East Indies would technically be conducted by an Australian airline rather than a British one.Footnote 49
Because this fact gave Dutch negotiators some leverage, British officials continued to tread carefully to avoid diplomatic conflicts. This became particularly evident in 1932, when they received a preliminary Dutch inquiry about the possibility of KLM flying across the Trucial States, a collection of sheikdoms in the Arabian Peninsula under informal British control.Footnote 50 Even though they were most unwilling to grant this permission, British officials agreed internally that ‘we should show ourselves reasonably forthcoming to the Netherlands government. [. . .] The Dutch might conceivably seek to draw a distinction between British and Australian aircraft if they felt that they had been shabbily treated.’Footnote 51 According to Dierikx, British concerns over a potential Dutch blockade grew so acute in the following year – when the Dutch were also attempting to extend their own air route from Indonesia to Australia – that the British and Australian airlines formed a joint venture in which Imperial Airways held 51 per cent of the shares. This majority stake ensured that the service could be safely considered ‘British’, effectively nullifying Dutch objections.Footnote 52
The case study introduced in this section confirms that aviation did indeed become a site of inter-imperial tension between Britain and the Netherlands. Concerned for its reputation, the British government sought to use air sovereignty as an obstructive tool. Yet, as has been shown, while it worked against non-imperial Germany, this strategy’s effectiveness vis-à-vis other empire-states proved limited. Involuntary interdependence, rather than competition, was at the heart of inter-imperial transactions. Flying to the same world region, both empires found themselves in a conundrum: they were rivals who still relied on mutual access to airspace. Both governments were in fundamental agreement that territorial sovereignty encompassed the skies. This realisation, in combination with the global geography of empires, imposed reciprocity as a basic practice. British and Dutch negotiators thus resorted to a transactional approach, granting each other’s airline mutual flyover and landing rights.
Still, British officials described these reciprocal transactions in rather sour terms as a system of ‘bartering’.Footnote 53 Given that businessmen and travellers within the British empire increasingly complained about the speed and reliability of Imperial Airways compared to KLM’s new service, the resulting compromise was far from what they had desired. This feeling grew even stronger by the mid-1930s when KLM supplemented its fleet with state-of-the-art Douglas DC-2 aircraft that could reach Singapore from Europe in six days (compared to eight days with Imperial Airways). The British officials struggled with the pervasive fear of being disadvantaged, despite the fact that Britain’s vast territorial holdings theoretically made it the supreme power. Accordingly, the Secretary of State for Air cautioned his offices to always attach ‘the utmost importance to our securing a “quid” whenever we possibly can in return for our “quo”’.Footnote 54 As the next section will show, this attitude also influenced the British government’s difficult negotiations with the French over air access.
Negotiating Collaboration: Bargaining Power and Mutual Dependency
At the same time that KLM inquired about flyover rights in the Trucial States in 1932, the French government officially requested use of an airport there. British officials were adamant in their opposition. Still, echoing their response to the Dutch inquiry, George William Rendel of the Foreign Office and Francis Bertram of the Air Ministry agreed that ‘our reply should be as helpful as possible in the case of the French [. . .]. Situated as we are in the north-west corner of Europe, we are entirely in the hands of our eastern neighbours.’Footnote 55 Even though the French request was refused, it was not done so on the grounds of Article 15 and Britain’s conventional rights but under the ‘pretext of the lack of [Imperial Airways’ own] operating experience’.Footnote 56 The British officials’ cautionary remark underscores the importance of French territory – particularly metropolitan France – for their operations. Although the French were unsuccessful in this case, sovereignty over airspace was a crucial bargaining asset for them: being the first stop on Britain’s empire routes, France served as a launching pad to southeastern Europe and the Mediterranean, and thus to the colonies. Fully aware of the value of this strategic position, the French government and its flag carrier Air Orient (and from 1933, its successor Air France) focused on using their air rights to secure cooperation.
This strategy first became apparent in 1929, when the French Undersecretary for Air, Laurent Eynac, and his British colleague Christopher Thomson signed an agreement. The Thomson–Eynac Concordat – touted as not a treaty but rather a gentlemen’s agreement – outlined the future conditions for flying within each other’s colonial realm. In it, the two empires committed to establishing ‘a spirit of wide mutual co-operation’. They made reciprocal provisions for aviation in Africa and Central and South America (with an eye towards future transatlantic services) and for routes to Asia. The agreement stipulated that French commercial aircraft would be allowed to fly over and land in India and continue to Saigon in French Indochina. In return, Imperial Airways gained access to French facilities in Indochina for its planned route extension to Hong Kong.Footnote 57
Full permission for a regular French service across India came only in late 1930, after a British–Indian air service had inaugurated regular flights between Karachi and Delhi and was making provisions for an extension to Kolkata. The first French aircraft embarked on its flight in January 1931. As a result of the Thomson–Eynac Concordat, the British and French airlines coordinated their timetables to minimise commercial competition. Still, protectionist tendencies prevailed: when granting permits for foreign trans-India air services, British authorities prohibited French and Dutch airplanes from delivering any mail in India.Footnote 58
Tensions were not confined to imperial skies but also extended to the European sector of the different air networks. In June 1934, officials at the British Air Ministry contacted their colleagues in Paris, asking for permission to reroute the European segment of Imperial Airways’ India route via Marseille. This request arose from the fact that the British route was still not fully airborne: because Italy had refused Great Britain air access, Imperial Airways passengers flew from London to Paris, then took a train ride to Brindisi in southern Italy, from which point they continued by air. To get the India route fully off the ground, the airline planned to add a stopover in Marseille, which was already the terminus of the French air route to Indochina. This additional stop would bring Brindisi within range, enabling Imperial Airways’ aircraft to reach it in a single hop.Footnote 59
The French Air Ministry was extremely disinclined to grant the requested permission for three reasons. First, Britain’s obstructive stance regarding the delivery of air mail in India had led to considerable frustration.Footnote 60 Second, French authorities were cautious about commercial competitors. For instance, even though KLM was allowed to make regular stops at the Marseille airport, it was prohibited from embarking or disembarking passengers, freight, or mail there.Footnote 61 A third reason was that allowing an additional direct service from London into the country would exacerbate the existing imbalance between British and French carriers on intra-European routes.
Air traffic between Great Britain and France was an anomaly within Europe’s network of air routes. By the early 1930s, intercompany arrangements had become common in continental air traffic: airlines operating between the same destinations shared revenues and resources in so-called pools to avoid competition and balance income. Almost all air routes through Western Europe were either operated by one single company or governed by pooling agreements. The only major route still operated in competition was London–Paris.Footnote 62 Shortly after its founding in 1933, Air France had proposed a pool for this route, but the request had been denied.Footnote 63 The French carrier was unable to compete in transportation between the two countries. According to company statistics, British aircraft carried 47,751 passengers on that line through 1933, compared to the 17,680 carried by Air France, and 85 tonnes of air mail compared to a mere 13.Footnote 64
This economic disadvantage notwithstanding, France’s location on the map afforded it considerable leverage in bargaining for better conditions in Europe. The country held the key to immediate British access to the Mediterranean. Therefore, when asked for a landing permit for Marseille, the French Air Ministry swiftly requested pool arrangements between Air France and Imperial Airways in exchange: for flights across the English Channel both airlines would have to agree to share all costs and income based on a fixed percentage division, regardless of the actual individual revenues. Despite Britain’s market power, France demanded a share between the French and British airlines of 60:40 for Marseille–London and 40:60 for the Paris–London service.Footnote 65
These demands came as a shock to British officials. While keen to secure economic advantages for Imperial Airways and to disadvantage its competitors within Britain’s own colonial realm, they were surprised by a similar approach from their French neighbours, which left them ‘most gravely concerned at the unexpectedly obstructive attitude’.Footnote 66 Only now did they fully realise that the Thomson–Eynac Concordat of 1929 had overlooked an important segment of their empire route: Europe itself. Although pooling seemed to be an economic issue confined to the Channel, it nevertheless affected Britain’s intercontinental route as a whole. Following reception of the French demands, Christopher Bullock, the second-in-command at the British Air Ministry, thus personally called Imperial Airways director George Woods-Humphrey to urge him to accept the proposed pool.Footnote 67 However, after the airline had deemed the terms unacceptable, the Air Ministry concurred and strongly advised that ‘we must resist any attempt to blackmail Imperial Airways into “pooling” agreements with Air France’,Footnote 68 warning that an imposed pool there could soon turn into imposed pools for extra-European sections of the India and Africa services as well.
Mirroring the Dutch–British interactions discussed in the previous section, the acceptance of aerial sovereignty combined with the prevalence of reciprocity were key to preventing an escalation of the dispute. British geopolitical dominance allowed them to reject the French offer. Through the numerous permits it had already issued to Air France for British territories, the British government had gained some leverage to pressure France into a deal regarding access to Marseille. Officials across the British government agreed that they could threaten France with the withdrawal of permissions, including for a recently requested French service through southeast Africa to Madagascar and for the Indochina service.Footnote 69 During meetings with French representatives in October 1934, Director General of Civil Aviation Francis Shelmerdine made it clear that this action was indeed on the table.Footnote 70
Neither the British nor the French went to these extremes as mutual dependency again posed an obstacle to overly harsh demands. A tentative agreement regarding the rerouting of Britain’s India service was achieved two months later, even though it failed to satisfy either party. For the time being, the French granted Imperial Airways only an ‘escale technique’ for Marseille, that is, landing rights solely for refuelling and maintenance purposes. Starting in 1936, British aircraft were also permitted to disembark passengers and mail arriving from overseas, excluding places in the French empire. Still, ticket sales between London and Marseille remained prohibited.Footnote 71
Over the next years the issue of securing full authorisation for British commercial use of Marseille remained unresolved. British fear grew that these tensions around Marseille could eventually spill over to the colonies and affect aviation over French territory everywhere.Footnote 72 On 16 June 1939, the Director General of Civil Aviation thus invited representatives of the French Air Ministry and of both airlines to London to finally pursue a compromise. Imperial Airways was henceforth allowed to make a full commercial stop in Marseille. In turn, the British Air Ministry and Imperial Airways reluctantly approved an Anglo–French pool that offered slightly better terms than those originally proposed in 1934. It divided the revenues from passengers, freight, and mail equally between London and Marseille, and on a 40:60 Air France–Imperial Airways ratio between London and Paris.Footnote 73
The focus on Anglo–French–Dutch interactions in this and the preceding section has shown that through the late 1920s and 1930s, the representatives of these countries developed a mode of deal-making among themselves. Articles 1 and 15 of the Paris Convention had provided the framework, but little else. Diplomatic transactions thus emerged as a key factor in the development of intercontinental air routes. Whether the different ministries and airlines wanted it or not, negotiations concerning the colonies frequently resulted in mutual concessions. The negotiating parties were forced to find common ground, which in the case of foreign access to Marseille allowed the French empire to hold its own against Britain: in the end, the French position largely prevailed, and Air France secured the means to economically compete on routes between England and France, even though the outbreak of the Second World War prevented the launch of regular pooled services.
The US State Department, looking at these transactions from afar, was thus correct in its assessment that ‘the European countries bargain very closely on the question of permitting foreign air transport lines to operate in their territory, such operation being permitted on a strict basis of reciprocity’.Footnote 74 The United States was neither a CINA member nor a signatory to the Paris Convention of 1919 and therefore was not subject to its principles. Yet, in Asia the United States’ emerging imperial air network touched those hailing from Europe. The third and final section of this article examines how the US government engaged with diplomatic patterns shaped by European powers.
Protecting Territories: Reciprocity as a Threat to Imperial Defence
Initially, the developing US air network had been separated from those of imperial Europe by vast oceans that aircraft in the early inter-war years could barely cross. Therefore, the issue of reciprocity in imperial aviation only became relevant to the United States a few years after it had emerged in Europe. It first appeared in connection with commercial aviation to the Philippine Islands, which had been a US territory since 1898. In March 1934, the US Congress passed the so-called Tydings-McDuffie Act, which provided for Philippine independence after a ten-year transition period. During this period, the Philippines would remain part of a US Commonwealth and the US government retained full control over their foreign affairs and defence. Around the same time as the passage of the Tydings-McDuffie Act, Pan American Airways, the US primary flag carrier, developed an air route from San Francisco across the Pacific to the US island empire, with stops in Hawaii, the islands of Midway, Wake, Guam, and Manila, the Philippine capital. Its inaugural flight, carrying only air mail, took place in November 1935, and passenger services commenced in October of the following year.Footnote 75
Both Pan American Airways and the US Navy had a strong interest in this route. From a military point of view, an air service to Manila would help maintain US influence in the region and would allow for rapid deployments even after Philippine independence, which was anticipated for 1946. This was particularly important because another empire, Japan, was increasingly exhibiting its expansionist ambitions in the Pacific region. Since the late nineteenth century, US policy towards Japan had rested on two pillars: containing its expansion and guaranteeing China’s territorial integrity. Tensions between the two powers first intensified when Japan expanded its territorial presence in the Pacific in the immediate aftermath of the First World War and then further with its invasion of the Chinese province of Manchuria in 1931. When Roosevelt took office in 1933, he identified Japanese militarism as a major threat. Reaffirming the two doctrines of his predecessors, he supported China’s Nanking government and increased US naval spending to deter further aggressions.Footnote 76 Air politics were closely tied to this broader geopolitical rivalry: US officials believed that Pan American Airways’ air infrastructure could serve as an important stepping stone for US military aircraft in case of warfare.Footnote 77
From a commercial point of view, the expansion of Pan American Airways’ impending Pacific route beyond its Philippine terminus in Manila appeared imperative. While the route to Manila was still under development, plans were already underway to link the future service with Europe’s transcontinental air systems: company president Juan T. Trippe wanted Pan American Airways to connect with the major imperial airlines, utilise their facilities in Asia, and potentially establish its own round-the-world service. Two access points for linking the hemispheric networks existed – Hong Kong and the Netherlands East Indies – and Trippe sought landing rights for both in 1934.
In November 1934, Pan American Airways applied to the British Air Ministry for a Manila–Hong Kong service.Footnote 78 In response, Trippe learned that only the US authorities could make such a request and that, if they did, the Air Ministry would require reciprocal rights for Imperial Airways in Manila.Footnote 79 Until that point, however, no foreign airline had ever been allowed to fly into an American overseas territory, and for reasons to be explored below, the US government was extremely disinclined to change that policy. Fortunately for them, Pan American Airways pre-empted any official request and managed to secure a favourable outcome through its own coercive diplomatic efforts: ostensibly preparing to by-pass Hong Kong if necessary, the company identified a potential alternative destination in nearby Macau, a Portuguese territory whose administration was willing to grant unilateral air access. Suddenly fearing exclusion from the evolving international network, the Hong Kong administration urged the London government not to insist on immediate reciprocal rights.Footnote 80 As a consequence, permission was granted ‘without demanding any quid pro quo other than an assurance of general reciprocal treatment in principle’, to be invoked at some point in the future.Footnote 81 Once Imperial Airways had inaugurated its Penang–Hong Kong service in 1936, Pan American Airways was allowed to follow, establishing the first fully airborne connection between the United States and Europe.Footnote 82
The US flag carrier was able to sidestep the reciprocity issue in the case of Hong Kong, primarily because the negotiations occurred in an informal arena where the airline took it upon itself to deal with foreign authorities. By contrast, the same question proved more complicated for the second potential extension of the trans-Pacific route, the link with the Netherlands East Indies. The colony lay along two of the world’s major trunk routes: KLM’s Jakarta–Amsterdam service and the Imperial Airways–Qantas service between London and Brisbane. By establishing a regular service from Manila to the nearby Dutch territory, Pan American Airways could thus position itself at the very heart of European imperial air networks.
In 1934, while the Pacific route was still under development, Pan American Airways and the Dutch colonial airline Koninklijke Nederlandsch-Indische Luchtvaart Maatschappij (KNILM) – an independent airline affiliated with KLM that provided domestic and regional services in the East Indies – launched an initiative to secure government approval for such a service.Footnote 83 In October, Trippe, after consulting with the Dutch, asked the US State Department to request permission for flyover and landing rights from the authorities of the Netherlands East Indies. The Pan American Airways president also instructed American diplomats to submit an additional document declaring that ‘the American Government would be prepared to favourably consider an application by KNILM to operate a reciprocal schedule to Manila on the same frequency’.Footnote 84 Similar to their British colleagues, the Dutch had made it clear to Trippe that a US application would be approved only if such reciprocity were granted. They pressed for mutual air access, and the US officials concerned with the matter were indeed correct in observing that ‘this is in accordance with the practice of the European countries’.Footnote 85
Uncertain about how to proceed, the State Department turned to the War Department for guidance. The War Department immediately sensed a potential threat to US defence interests in the Pacific. Internal communication following Trippe’s request shows that military officials were afraid of what the concept of reciprocity could potentially entail for US air sovereignty: a reciprocal arrangement, they believed, would set a precedent.Footnote 86 On 17 December 1934, Secretary of War George Dern wrote to Secretary of State Cordell Hull, explaining that ‘the interests of the national defence must be safeguarded. [. . .] The real danger of the present proposal lies in the not improbable request of other nations for similar concessions’, for not only the Philippines but also the United States’ other Pacific territories, especially Hawaii.Footnote 87 The War Department was opposed to granting KNILM the necessary permission. Its main concern was not the Dutch but the Japanese Empire. Once the Dutch airline was granted access, officials believed, the Japanese government would make similar requests for their state-owned flag carrier, Nihon Kōkū Yusō Kabushiki Kaisha (or Japan Air Transport Company).Footnote 88
In many ways, the development of Japan’s international air transport resembled that of its European counterparts. Having negotiated and signed the Paris Convention in 1919, the Japanese Empire had been ‘airminded’ from early on in the inter-war era. Companies such as Mitsubishi began collaborating with German aircraft designers and engineers, and the government participated in CINA meetings before withdrawing from the League of Nations in 1933.Footnote 89 The Japan Air Transport Company was founded in 1928 under the initiative of Prime Minister Tanaka Giichi. Very similar to the simultaneous formation of flag carriers in Europe and the United States, the airline’s establishment ‘reflects the expansionist and colonialist ambitions of the new government, which sought to extend its influence far beyond the Japanese archipelago’,Footnote 90 as Cécile Bronn recently observed in her study of Japanese aviation before the Second World War. By the early 1930s, Japan had launched its first imperial service to the Korean colony and was planning air routes to China, Taiwan (then Formosa) and its South Pacific dependencies, including Palau and the Northern Mariana Islands. Airline managers were also making plans to connect their network with the ‘Far East’ routes of Europe’s air-going powers, investigating, on the one hand, a route to Hong Kong and, on the other, an air link to Bangkok. Later, these plans were pursued in conjunction with Lufthansa.Footnote 91
In light of Japan’s expansionist policies in the Pacific, US officials grew increasingly concerned about this network expansion. There was no doubt a general concern among American authorities that US territories might fall into the Japanese sphere of influence and that airline facilities and ground crews could function as door openers. But even in the absence of an immediate military threat, commercial airplanes flying over the Pacific Fleet’s supreme naval bases would likely facilitate espionage through aerial photography.
The War Department based its belief that troubles could arise from a concession to KNILM on the most-favoured-nation principle. In international trade, this principle specifies that the treatment of trade partners from one nation should not be worse than that of trade partners from another. Article 2 of the Paris Convention indeed prohibited making distinctions between contracting states based on nationality. However, the United States had never joined the convention. Thus, the War Department’s judgement was inaccurate: the granting of the requested air link to the Dutch colony imposed no legal obligation on the US government to later allow the Japan Air Transport Company into US territory. For this reason, officials at the State Department vehemently disagreed with their colleagues and spoke in favour of a Dutch air link.Footnote 92
Because air diplomacy fell within the State Department’s remit, its staff foresaw what an obstructive approach to reciprocity would entail: the officials recognised from recent observations, especially regarding Europe’s imperial routes, that national sovereignty over the air was a powerful tool and that intercontinental air traffic was impossible without a readiness to compromise and accommodate: ‘Unless the authorities of this Government follow the same course the extension of American transport lines into foreign territory will be greatly handicapped if not prevented altogether’,Footnote 93 the State Department officials warned in late 1934. In light of the War Department’s reservations, they were seriously worried that if security concerns prevailed, such a protectionist attitude would impede further network growth. This was all the more concerning since Pan American Airways was simultaneously preparing to go global. Parallel to the Pacific route, the airline had formally teamed up with Imperial Airways to explore the possibility of commercial aviation across the North Atlantic.Footnote 94 Even though a regular service on this complicated route was delayed until 1939, it was to be feared that Pan American Airways would not be able to find overseas partners if the US government failed to adhere to the now common principles of air diplomacy.
The Dutch legation in Washington, DC, was likely unaware of these discussions within the US government. When they did not receive an update on the planned air link, Dutch diplomats took the initiative in June 1935 and submitted an official request for air access to the Philippines.Footnote 95 Even though the War Department again voiced its opposition to the Dutch application, it gradually softened its stance after several months of consultations. Observing that Pan American Airways’ new Pacific service would benefit from a feeder service to the Dutch colony, it withdrew its obstructive attitude in July 1936, provided that any future foreign, that is, Japanese, air service to the Philippines would still require military approval.Footnote 96 In theory, the time had come for bilateral negotiations over mutual access. However, due to the complicated nature of the US Commonwealth, it was not entirely clear which authority – Philippine or American – was actually responsible for granting air access during this transitional phase.Footnote 97
While legal consultations were still underway, on 5 April 1937, KNILM submitted a memorandum to the US government urging it to finally grant the concession.Footnote 98 Only one day later, the Dutch legation in Washington, DC, received the official response: access denied.Footnote 99 While they were kept in the dark about the reason, internal State Department communication points again to the issue of national security: KNILM was no longer the only foreign airline with an interest in flying to the Philippines. In March, just three weeks prior, the Japanese ambassador to Washington, DC, had informally inquired about establishing a Japanese air service between Manila and the Japanese colony of Taiwan.Footnote 100
While the previous debate on Japanese aircraft in US skies had been largely hypothetical, this inquiry brought the issue into practical consideration. This time the War Department did not need to voice its apprehension since State Department officials already shared the same concerns: that the Japan Air Transport Company posed a serious security risk given the airline’s entanglement with the Japanese government, and that commercial airplanes could open skies over US naval bases to espionage.Footnote 101 Under no circumstances could a Japanese air service be allowed to fly over the Philippines. For Dutch aviation, the timing was most unfortunate. The State Department was forced to shelve its previous considerations with respect to global network expansion and reciprocity. Because both requests would have to be addressed almost simultaneously, the department deemed it ‘unwise to offend the Japanese by disapproving their application and at the same time approving the Netherland application’.Footnote 102 Both governments were informed in April 1937 that their requests had been denied. Pan American Airways’ planned connection to Indonesia was consequently put on hold.Footnote 103
A few weeks later, Japan’s aggressions in China escalated into the full-scale Second Sino–Japanese War. While these events were likely seen by the Americans as a confirmation of their reservations, the Dutch continued, without success, to urge the State Department to finally make a reciprocal air service happen.Footnote 104 An international air link became vital for the Dutch colony when the Second World War broke out in Europe and even more so after Germany invaded the Netherlands in May 1940. The Netherlands East Indies were cut off from communication with the mother country and looked to the US military for defence.Footnote 105 KNILM made a number of charter flights to the Philippines, but no permanent service ever came about.Footnote 106 Debates ensued within Dutch colonial circles about whether they should give up their demand for reciprocity and secure a lifeline by unilaterally allowing Pan American Airways’ service into the colony. Ironically, this idea was quickly discarded out of fear that a foreign air service might prompt the Japanese to demand landing rights in Indonesia as well.Footnote 107 A scheduled air connection never materialised because Japan attacked Pearl Harbor on 8 December 1941 and then went on to conquer the Philippines and the Netherlands East Indies within the first few months of 1942.
Conclusion
The three featured case studies speak to the full spectrum of diplomatic transactions over air access in the 1920s and 1930s. Weighing both legal claims and political realities, the preceding analysis has demonstrated that the massive expansion of intercontinental air routes in the period under consideration can only be fully understood by considering the interplay of multilateral norm setting and state-to-state interaction. An international consensus had emerged immediately after the First World War that the existing definition of territory required expansion to include the air. Originally connected to the Paris Convention of 1919, air sovereignty soon developed into an independent doctrine governing global air infrastructure. Indeed, all four states studied in this article – Britain, France, the Netherlands, and the United States – respected one another’s sovereignty over national and imperial skies with regard to commercial aviation. At a time when novel technologies such as aircraft or wireless telegraphy permeated boundaries, the wide acknowledgement of air sovereignty reaffirmed centuries-old ideas of Westphalian territoriality.Footnote 108
By contrast, the abstract provisions of the Paris Convention failed to achieve widespread implementation, primarily because of their vague and patchy nature. Hence, as the US–Dutch interactions show, the question of whether the negotiating partners were signatories to the convention at all seemed largely irrelevant. Moreover, officials never relied solely on the legal text. According to Article 15 of the Paris Convention, the establishment of permanent and regular air routes required the consent of the states concerned, and this consent could be denied without justification. And yet, imperial governments often used evasive explanations and pretexts to lend legitimacy to their actions, instead of simply invoking their conventional sovereign rights in a straightforward manner. As the analysis of bargaining dynamics in the first two case studies has made clear, the relevant governments could never resort to permanently closing their skies or to fully escalating conflicts. Any inclination towards cut-throat behaviour was significantly constrained by the fear that other states might respond in kind.
Overt rivalries notwithstanding, the geopolitical – and above all, geographical – realities of a world divided into empires necessitated compromises whenever and wherever intercontinental air route planning gained momentum. Involuntary interdependence caused diplomatic missions to mutually craft a modus operandi of granting access rights. Careful analysis of the activities within and between government offices underscores the persistent significance of bilateral interactions within a rudimentary framework established by multilateralism. These bilateral transactions soon gave rise to a prevailing standard, at least among Europe’s air-going empires. Reciprocity complemented sovereignty as the second guiding principle of aviation, and these two principles acted in combination whenever two countries sought to operate air routes into or through each other’s airspace. Sovereignty and reciprocity together emerged as the two pillars of inter-war air diplomacy on a global level, and it was only through their interplay that route expansion became feasible in an increasingly interconnected world.
Importantly, by the eve of the Second World War, the European system for awarding flyover rights for the colonial world was eroding. War anxieties and military concerns had not played a significant role between old-established European empires, but they did become a central factor informing US air policies when Japan adopted a more aggressive posture in the Pacific region. The United States’ obstruction of an agreement with the Dutch authorities thus fails to conform to the bargaining rationale described in the first two case studies and marks a clear departure from the exchange logics that informed the European empires’ interactions in the previous decade.
Bilateralism, however, continued to shape global aviation during and after the Second World War. As the war drew to a close, the desire for a new framework for air access grew among the allied states. To define a ruleset for post-war aviation, in late 1944, the United States invited delegates from fifty-two nations to the International Civil Aviation Conference in Chicago. There, the US delegation’s advocacy for an ‘open skies’ policy met with fierce British resistance.Footnote 109 Because this conflict could not be resolved, the Chicago Convention of 1944 eventually went the way of the Paris Convention of 1919, stipulating that ‘[n]o scheduled international service may be operated over or into the territory of a contracting State, except with the special permission or other authorization of that State’.Footnote 110 Accordingly, the United States reverted to bilateral negotiations with several countries and, in 1946, reached an agreement with Britain.Footnote 111 It was this Bermuda Agreement and not the Chicago Convention – the bilateral compromise rather than the multilateral accord – that became the blueprint for aviation in the ‘jet age’.Footnote 112 Today, the global transport system still rests on an estimated 3,000 bilateral agreements of this sort.Footnote 113
Acknowledgements
I wish to thank Heidi Tworek, Marc L.J. Dierikx, Mario Peters, Kelly McCullough as well as Contemporary European History’s editors and anonymous reviewers for their many valuable suggestions that significantly improved the article.
