Republic of Conspiracies: Cross-Border Plots and the Making of Modern Turkey

Abstract

In August 1935, British authorities tipped off Ankara about a team of assassins who were allegedly headed for Turkey to assassinate its president Mustafa Kemal Atatürk. Within a month, the Turkish authorities arrested a number of suspects in the Turkish-Syrian borderland, and began to pressure London to extradite the Circassian masterminds of the plot who were then living in the British mandate territories of Palestine and Transjordan. This article examines how the British tip-off quickly evolved into an episode fully publicized by the Kemalist regime, exploring the ways in which the alleged conspiracy helped consolidate Ankara’s ideological positions at home and pursue its long sought-after policies abroad. This curious episode illustrates the political and socio-economic relevance of imperial networks that continued to crisscross the post-Ottoman Middle East. On a more analytical level, the conspiracy helps us understand the complex interaction between intelligence and rumors, and in so doing, shows both empirical limits and opportunities in approaching them as a field of historical inquiry.

On 14 August, 1935, British authorities received information ‘from a source whose reliability cannot be guaranteed, but which may be well informed’ that a group of assassins had recently left the port of Beirut with a view to assassinate Mustafa Kemal Atatürk, the President of Turkey. ¹ The British ambassador quickly passed on the intelligence to Şükrü Kaya, the Turkish interior minister. Kaya told the ambassador in somewhat ambiguous terms that ‘he did recollect having seen a report some time in July and to a somewhat similar effect’. ² In hopes of further confirming the British intelligence, the Turkish ambassador then approached the French High Commissioner in Beirut, but the latter was not able to verify the report. ³ That it was the British who initially tipped off Ankara continued to be the major source of authenticity for Turkish authorities. Even though the initial intelligence remained largely generic in content, Ankara soon identified a group of Circassians living under the British and French mandates as complicit in the plot and demanded their arrest. ⁴

Despite their willingness to cooperate, British officials were cautious in the way they communicated with their Turkish counterparts, perhaps in part due to their own concerns about the shaky evidential basis of their intelligence. When the Turkish ambassador to London Fethi Bey expressed his thanks to the British for sharing ‘the memorandum’ with Ankara, for instance, London tellingly felt the need to clarify that this was strictly an oral communication and hence not a written one, advising that it should be referred to as such in future communiques. ⁵ Similarly, when rumors began to circulate that it was the British who initially tipped off the Turks, the British ambassador complained about such ‘rumors’, asking Ankara for discretion and not to leak this ‘friendly gesture’ to the broader public. ⁶ Understandably, they did not want to come across as the responsible party for recycling, let alone planting, what could possibly just have been a rumor of a conspiracy. They certainly did not want to be subjected to ridicule, either, for forwarding a piece of intelligence which made connections that may not have actually existed, which would not be the first time they had done so. ⁷

By tracing and reconstructing what happened after the receipt of this intelligence in Ankara, this article seeks to explore the complex interactions between rumors and intelligence, and the ways in which they could determine the contours of politics at home and shape interstate relations abroad. To be sure, intelligence differs from rumors. The latter reflects the voices of the ordinary people that circulate orally in organic ways without clear origins, and can be understood as a collective response to an ambiguous situation of certain significance. ⁸ Intelligence, on the other hand, is a highly formalized form of communication among bureaucrats in their broader pursuit to map out an increasingly complex world, and make it legible for policy purposes. ⁹ Despite their obvious differences, however, both rumors and intelligence could interact and co-produce one another, specifically when they relayed information on political conspiracies which are, by definition, clandestine. ¹⁰ This was particularly the case in the margins of colonial authority and webs of information, where ‘intelligence material could be little more than a dressing up of rumor or the recapitulation of the flimsiest evidence for political ends’. ¹¹ This article seeks to capture this complex interplay between intelligence and rumors by examining the politics that developed around the alleged assassination plot against Atatürk in 1935. Taking it as a case that illustrates well how intelligence gathering could thrive in the presence of uncertain information, I hope to contribute to the broader methodological debate on how historians should treat rumors and intelligence as a historical object of study. ¹²

In exploring the complex politics of information through the lens of cross-border plots, I build upon a recent literature that makes a case for the connected histories of the post-Ottoman states. Nationalist historiography has long categorized the post-First World War period into distinct national units of analysis. Consolidating this trend was the historiography of Turkish modernization, which, with its emphasis on Turkey’s Western orientation, led to the underappreciation of Ankara’s interactions with its southern neighbors. ¹³ A host of critical works has recently begun to challenge the compartmentalized nature of the field, calling for a connected historiography of the post-Ottoman Middle East. ¹⁴ One way of contributing to this literature is by exploring the persistent patterns of Ottoman mobilities in the interwar years, whether the movements of seasonal workers and pastoral nomads, or those of pilgrims and refugees. The cross-border plots and conspiracies, too, whether real or imagined by newly ascendant national elites, embody a similar analytic potential. Often involving contentious actors that remained outside the fold of nation-states in the post-Ottoman world order, cross-border conspiracies not only illustrate the limits of post-Ottoman sovereignty on the margins of nation-states, but also reveal not-too-distant imperial past that continued to shape politics well into the 1930s.

For Ankara, the alleged assassination attempt against Mustafa Kemal Atatürk in 1935 was yet another piece of intelligence in a long list of plots presumably targeting Atatürk, which – as a genre – stretched well into the early 1920s. ¹⁵ Forming a longer register of incoming intelligence briefs, these reports continued to warn about alleged hitmen infiltrating Turkey to assassinate different members of the Turkish leadership. One popular history book counted 11 of them targeting Atatürk alone. ¹⁶ In many of these plots, it was often Armenian revolutionaries who were considered to be the masterminds. ¹⁷ Indeed, Armenian plotters spent the early 1920s in Europe, assassinating exiled Young Turk leaders whom they considered responsible for the genocidal crimes committed against the Ottoman Armenians in 1915. Having already assassinated Talat and Said Halim Pashas, one piece of intelligence warned in 1922 that the Armenian revolutionary organization now identified Mustafa Kemal as their new target in Turkey. ¹⁸ From the late 1920s onwards, similar reports incriminating Armenian conspirators resurfaced time and again, ¹⁹ particularly after the Armenian revolutionaries struck an alliance with the exiled Turkish royalists and the Kurdish nationalists in Syria. ²⁰

The Circassians also claimed a fair share in Ankara’s appraisal of anti-Kemalist conspiracies. In the early 1920s, Circassian irregular forces had played a crucial role in containing not only the Greek advances, but also suppressing the royalist uprisings across Anatolia. Yet, the consolidation of Mustafa Kemal’s authority and the increasing importance of regular armed units soon resulted in irreparable rifts between Ankara and the Circassian leaders who commanded the formerly useful guerilla forces. ²¹ Chief among them were Çerkez Ethem and his two older brothers, Tevfik and Reşit, who ended up defecting to the Greeks in January 1921, together with their 300 fighters, which was a major act of treason in the eyes of Ankara. ²² As the Kemalists gained the upper hand in the upcoming months, the Circassian leaders were forced to seek refuge on Greek islands situated in close proximity to the Anatolian shore, where they continued to hatch plots. On 19 March, 1923, for example, a Turkish intelligence report, gleaned ‘from a person with knowledge of Greek affairs’, stated that Çerkez Ethem had established a revolutionary committee and succeeded in dispatching into Asia Minor a hitman charged with assassinating Mustafa Kemal, Fevzi (Çakmak), İsmet (İnönü), and Refet (Bele) Pashas. ²³ By targeting the emerging leadership of the nationalist movement in Ankara, such strategies sought to create political opportunities for diverse actors who were otherwise denied access to political circles in Ankara.

Whether they involved Armenian or Circassian conspirators, such plots were taken seriously by the Kemalists, in part because they were not strangers to conspiratorial politics. Because the Kemalists hailed from the Young Turk revolutionary tradition, their worldview was deeply shaped by the late Ottoman contentious politics, where conspiracies were indeed commonplace methods through which aspiring elites sought political change. ²⁴ The maneuvers of the dissident Circassians to mobilize their sympathizers through associations and engage in tactics of intimidation were therefore derived from a mutual code of conduct that had defined this last Ottoman generation of revolutionaries-turned-statesmen, which certainly included the Kemalists. ²⁵ Perhaps the most serious reckoning with this mode of contentious politics took place in 1926 in the shape of an elaborate plot that allegedly targeted Mustafa Kemal. Masterminded by the former members of the Young Turks who were estranged from Ankara by the growing power of the Kemalists, the conspiracy involved the assassination of the president during his visit to the port city of Izmir, but the plot was leaked and the would-be assassins were caught red-handed. ²⁶ The trials that followed not only punished those involved in the plot, but also provided, in the words of the influential deputy Falih Rıfkı Bey (Atay), an ‘excellent opportunity’ to purge the broader opposition to Mustafa Kemal. ²⁷

The public prosecutor’s indictment of the 1926 plot – particularly his description of the involvement of the opposition deputies in the conspiracy – illustrates something common in the way bureaucrats frame conspiracies: an attractive tendency to see them as part of larger subversive networks, making connections that may not exist. This inclination should alert us to how ‘fear and confusion were so often the mainsprings of official thought and action’, illustrative as much of governmental mentalities as strategic calculations. ²⁸ In retrospect, each plot and conspiracy not only played into the hands of the Kemalist regime and consolidated it during the 1920s, but also served as a useful tool to distract public opinion, especially when propagated efficiently by the print media. Throughout 1929, for instance, Turkish newspapers reported so many sensational news stories involving plots that when they stopped doing so, rumors abounded that ‘the government has had enough of plots… that no more of them should be unearthed for the time being…, hence the rumored order to the police to be less zealous’ in discovering subversive activities. ²⁹ This is precisely where the significance of studying conspiracies reveals itself as a field of historical inquiry ³⁰ – that is, their capacity to facilitate political action and shape public debate, even though their origins are often shrouded in mystery and the exact extent of broader involvement remains difficult to delineate.

This was also the situation in summer 1935 when the British forwarded their intelligence to Ankara, indicating that a coalition of Kurds, Circassians, and Armenians had reportedly acted to assassinate the Turkish president. Atatürk received the news when he was in his Yalova residence, his favorite spot to spend the summer. This was where he assessed the situation with Tevfik Rüştü Aras and Şükrü Kaya, the foreign and interior ministers – i.e. two of his most-trusted cabinet members. ³¹ After their meeting, local authorities were asked to be extra vigilant against illicit border crossings by suspect individuals, while Atatürk was rushed out to Ankara, ‘a place where it is easier to secure his safety’. ³² By late August, the measures by the Syrian border seemed to have borne fruit. A rich landowner named Üzeyir from a Circassian village (Kümbetir – today’s Yeşiltepe in Kahramanmaraş) informed the authorities of the suspicious presence in his village of a person called Yahya. Yahya was also of Circassian descent, but he was from Çarşamba, a town situated in the northern Anatolian, while his parents hailed from Damascus. ³³ In past years, he had been imprisoned a few times for crimes ranging from rustling livestock to assisting robbers and evading the military draft. After deserting the army for a second time in Mersin, Yahya had crossed into Syria illegally and was arrested after his return back to Turkey, when Üzeyir tipped off the authorities. ³⁴

As a result of Yayha’s testimony, many other borderlanders were also arrested, including Üzeyir himself, as well as Ali Saip Ursavaş, the deputy for Urfa, an important provincial center by the Turkish-Syrian border. ³⁵ Üzeyir and Ali Saip were connected to one another. For one, the local authorities had long suspected Üzeyir of playing important roles in the regional smuggling networks. He particularly did so through his brother Arif, a driver, who helped Üzeyir cultivate useful cross-border relations, with places as far south as Amman. By 1935, Üzeyir was already a middle-aged man, but under the Ottomans, he had served in the gendarmerie in Aleppo, where Ali Saip was his commander. The relationship between the two further consolidated in the early 1920s, when both Üzeyir and Ali Saip aligned themselves with the ascendant Kemalist movement in Ankara. ³⁶ At the time of his arrest in 1935, Ali Saip was a deputy but, similar to Üzeyir and many other deputies of the time, he was a landowner. With several homesteads located close to the Syrian border, he was constantly rumored to have tapped into expanding circuits of smuggling. Unlike Üzeyir, however, Ali Saip had direct access to Mustafa Kemal as a frequent guest on the president’s famous dinner tables. This friendship in fact dated back to 1918 when Mustafa Kemal was appointed as the head of the Seventh Army in Aleppo, where Ali Saip was the commander of the gendarmerie in the district of Deir-ez-Zor. ³⁷ Despite his Kurdish descent, Ali Saip was deemed so reliable that he was appointed as the head of the tribunals in the aftermath of the Shaikh Said Rebellion (1925) throughout which he continued to maintain a private line of communication with Mustafa Kemal. In late 1920s, however, Turkish intelligence suspected of Ali Saip of cultivating personal links with exiled Kurdish leaders in northern Syria. ³⁸

These are the facts about Yahya, Üzeyir and Ali Saip, loosely attaching them to one another and to Mustafa Kemal, but it was the rest of Yahya’s story that in fact connects the dots. ³⁹ According to Yahya’s first testimony, he started working odd jobs by the Syrian border after he had deserted the barracks in Mersin. Soon, he fell in love with a girl from the village of Karamezar in Ceyhan, but he did not have enough money to marry her. He worked at Üzeyir’s farm for six months, among other places, but could not save much. Eventually he decided to seek his fortune in Syria, where he visited not only Aleppo and Damascus, but also Quneitra, a town with a significant Circassian population. There, he met the head of the Circassian Association, Emin Semko, who advised him to go to Amman and work for a certain Manol. ⁴⁰ The latter had close relations with Çerkez Ethem, whom Yahya actually met one day in a cafe. Ethem ended up offering a better job for Yahya in Haifa, where he worked next to a certain David for some time. One day, David received a letter from Ethem, who said that it was time for Yahya to head back to Turkey, together with four other friends, to assassinate Atatürk in a plot Üzeyir and Ali Saip would help facilitate. Upon entering Turkey, Yahya split ways with his friends as he wanted to go and seek the girl he loved in Ceyhan, but agreed to meet up in Üzeyir’s farm. Because Yahya ran late, however, Üzeyir had already panicked and sent the four assassins back to Syria, and forced to turn in Yahya to the authorities the moment he showed up.

In the first session of the public hearing, however, Yahya dismissed this version of events, which he claimed to have been secured under torture. He then described in detail how he was forced to mention the names of Üzeyir, Çerkez Ethem and Ali Saip in hopes of stopping the torture. According to his revised testimony in court, Yahya said that he had in fact worked for seven years in the village of Karamezar in Ceyhan after deserting the army. There, he had fallen in love with a girl named Nadide but failed to secure loans to ask for her hand. Hence, he decided to go to Syria to make money but had to cross the border without a proper passport, because he was a draft dodger. Yahya then described having difficulty finding a job, and how he asked the Circassians he came across in Damascus for help, albeit without much success. Yahya admitted that he met Çerkez Ethem and his brother Reşit in an Amman café where he was looking for a job, and recalled the brothers as having been inquisitive about politics in Turkey. He then described how he eventually ended up spending two years in Palestine working for individuals such as David among others. Yet, he claimed that when he returned to Turkey, he directly went to Karamezar to see Nadide, only to find out that she had already married someone else. Yahya then went to Üzeyir, hoping that he could help him re-claim the monies he had already spent to secure Nadide’s hand, but Üzeyir turned him in to the authorities.

At the time of the arrests, the authorities were not able to collect any collaborating evidence, such as the guns and bombs to which Yahya referred in his initial testimony. The only evidence law enforcement managed to get their hands on was an empty bottle of rakı which Yahya claimed in his revised version of events to have consumed on his way to Üzeyir’s farm. In fact, the interrogating magistrate publicly declared that had this been an ordinary case, he would have dismissed it for lack of evidence from the very start. Yet, because it involved Atatürk, ‘he felt it should come up for trial even though there might be only a very small chance that the persons accused were actually guilty’. ⁴¹ There were indeed historical precedents that must have inspired caution for the Turkish leadership and the court’s approach to the case. After all, Çerkez Ethem and his two brothers had a troubled personal relationship with Atatürk, with a proven track record of conspiracies that were designed to derail the Kemalists. As former members of the Ottoman Special Organization, they were well-versed in subversive tactics, too, and certainly capable of hatching yet another plot. ⁴² Finally, the Circassian networks had been conspicuous across the board: Yahya was himself a Circassian and picked up employment from Üzeyir who was an important landlord in an area dotted with Circassian villages. As Yahya continued to seek employment while navigating Greater Syria, he continued to tap into Circassian networks and was noted to have met Çerkez Ethem in Amman, a city largely shaped by Circassian migrations. ⁴³

Scholars have crucially noted this nexus between labor migration and ethno-religious ties of origins to be an important factor in enabling the creation of new ties of loyalties and formation of trust among revolutionary groups. ⁴⁴ Yahya’s Circassian networks surely matched these historical precedents, but obviously illicit labor circuits alone did not make one criminal. ⁴⁵ In its verdict of 17 February, 1936, the Ankara court ended up dismissing the case altogether, which was astonishing ‘because modern Turkey provided no precedent for a wholesale acquittal on such a charge… [and] the dice were heavily loaded against the prisoners from the beginning’. ⁴⁶ The positions developed by the public prosecutor were more ideological than empirical, failing to substantiate a premediated Circassian conspiracy. Perhaps the conspiracy of 1935 did not result in charges as expected, but it certainly served a broader political function to galvanize the public opinion across Turkey.

In fact, from the moment Turkish newspapers began to report on the Circassian conspiracy until the dismissal of the case by the court, the assassination attempt had been an event on full public display. A massive rally was quickly organized in Ankara only three days after the initial news of the assassination attempt. 50,000 people reportedly attended the event, forming a crowd that was ‘the largest ever seen in the capital’. ⁴⁷ This was indeed the power of the single-party regime in mobilizing the masses as it saw it fit. The same newspapers which had reported only two days previously how deserted Ankara’s squares had become due to the national census were now reporting that nearly half of Ankara’s population turned up to attend the meeting to decry the conspirators. ⁴⁸ Without doubt, this brand of mass politics was in line with the spirit of the mid-1930s – i.e. the age of massive public rallies where the cult of leadership was constructed and the discursive parameters of nation-states were reset. ⁴⁹ Turkey was no exception to this rule, with the Circassian conspiracy providing the necessary stimulus for Ankara’s populist moment.

Turkish populism dates back to the late Ottoman times when a range of populist initiatives were put into practice from boycotts to mass mobilizations. ⁵⁰ The particular Kemalist brand of populism (halkçılık), however, largely evolved in response to the challenges brought about by the Great Depression (1929). The latter laid bare the lack of legitimacy of the regime among the masses, forcing Ankara to start taking populist politics more seriously. ⁵¹ By 1931, the convention of the Republican People’s Party not only declared the six principles of Kemalist ideology – thus the party’s emblem of six arrows – but also announced the establishment of People’s Houses, which were to bridge the gap between the elites and the masses. ⁵² Helping to put to test some of these evolving practices were the decennary celebrations of the Turkish Republic in October 1933. In preparing for the anniversary, every village was asked to construct a republican square that could host celebrations, while each household across the cities was expected to fly the national flag. Out of many banners that dotted the capital, the one that stood on top of the unfinished building of the Ministry of Interior read that ‘the Gazi is one of us: he is the greatest of us: he is the sum total of us all’. ⁵³ At the time, Mustafa Kemal was still referred to with his honorific status of war hero, gazi, and was yet to receive the surname Atatürk (i.e. the father of Turks). Therefore, the slogans, symbolisms, and public festivities all suggested an ideological work-in-progress, illustrating the regime’s desire for populist appeal since the early 1930s. ⁵⁴

The mass gatherings organized to denounce the plot of 1935 largely built upon this background of evolving set of ideological and symbolic practices. Soon after the rally in Ankara, another one was organized in İstanbul’s Beyazıt Square, which reportedly drew close to 100,000 people. ⁵⁵ Such public uproar in major cities certainly pressured all other provincial authorities to reciprocate by organizing their own events if they had not already done so, and thus illustrate their commitment to the broader republican project. Accordingly, every town of significant proportion throughout Turkey began to host public gatherings, although the number of participants was significantly lower compared to Istanbul and Ankara. Only 3000 people were reported to have attended the rally in Izmir, for instance, where ‘the general public was conspicuously absent’ and the participants were largely ‘confined to school children, scouts, guides, and members of corporations and other organized institutions’. ⁵⁶ Reflective of the growingly official character of the events, the newspapers not only jotted down the news of thousands of telegrams sent to Atatürk on behalf of labor organizations, city councils, or religious congregations, but also continued to report on a daily basis the meetings of similar size being held in town squares from east to west, and north to south. ⁵⁷ Needless to say, in a political environment dominated by the single party-rule of the Kemalists, these efforts were visibly coordinated by the People’s Houses. Badges featuring the party emblem of six arrows were conspicuous across the rallies, worn by many of the speakers decorating their fiery remarks. The profiles of the participants were also telling. On the day of the rally in Ankara, all ministry buildings were emptied off their staff, who were asked to attend the event. ⁵⁸

Where the meetings were held were also significant. In many of the Turkish cities which had suffered greatly from the ravages of warfare since the 1910s, Ankara had been busy promoting an urban construction program in line with the principles of Turkish revolution, which often manifested itself in the building of a government mansion ‘facing a formal government square with an Atatürk statue in the middle (bust, equestrian, standing figure)’. ⁵⁹ Many of the demonstrations in late October 1935 were in fact held in these republican squares, transforming them into discursive spaces where crowds circled the statue of Atatürk and chanted nationalist slogans. Just like in other countries of the time where biological metaphors were freely used to construct the nationalist body politic, these rallies across Turkey promoted the singularity of the nation and the unified purpose of those who gathered. ‘Fifty thousand in Ankara had the same heartbeat and spoke the same language’, Ulus declared the day after the rally in Ankara. ⁶⁰ This discourse persisted in Istanbul, where Hasene Ilgaz, a teacher, similarly announced to the crowds that ‘our grand chief is in the hearts of 18 million Turks, and in front of the gazes of 36 million eyes’. ⁶¹ Repeated time and again was the readiness of every Turkish citizen to sacrifice their lives to protect that of Atatürk, demonstrated with the popular chants of ‘Long live Atatürk!’

Simply put, these rallies became an important medium through which the cult of leadership around Atatürk could be further constructed and disseminated. ⁶² The deputy Naşid Uluğ who spoke in the Ankara rally, for instance, framed Atatürk not only as a savior of the Turkish nation, but also its creator, noting how the love of all Turks for their president – from that of a shepherd to that of a professor – is what unites the country. ⁶³ Such phrasing, falling just short of attributing divine powers to the Turkish president, was ubiquitous throughout the print media. The famed Turkish nationalist Ahmed Ağaoğlu, for instance, called the alleged assassins ‘fools for daring to hurt such a spiritual being’, because Atatürk’s watch miraculously spared him a bullet during the Gallipoli campaign many years ago. Here was a leader entrusted with a historical mission for the Turkish nation. ⁶⁴ Indeed, the assassination attempt was quickly framed not as a simple plot against the person of Atatürk, but targeting, above all, his modernizing reforms in Turkey. In his indictment, Baha Arıkan, the public prosecutor of the conspiracy trial, struck a similar tone as he declared that it was ‘the republican regime that was on trial’, while he framed Atatürk to be an unworldly being, i.e. ‘the ultimate ego of the material and spiritual existence of 17 million Turks’. ⁶⁵

Such statements were not just limited to the courtrooms. Nor were they solely chanted across the mass rallies. From the very start, newspapers gave the assassination and the ensuing trial the most detailed coverage and played a key role in the consolidation of state discourses by communicating each episode to the broader public through sensational headlines. This was certainly less of a surprise than a confirmation of the tight control that the single-party regime exercised upon the Turkish press, both national and provincial. ⁶⁶ In this particular episode involving the assassination plot, however, the use of the print media to shape the public opinion was too blunt to gloss over: the summaries of court proceedings were printed page after page, and the final indictment was reproduced word by word. Public interest in the outcome of the trial is emphasized time and again, which was illustrated with pictures taken inside the packed courtroom or showing those outside who failed to get in. ⁶⁷ Even if the republican regime was not on trial, it certainly looked like it was.

The Circassian conspiracy of 1935 was constitutive in ways that went well beyond the realm of domestic politics, however. As we have already seen, the construction of the cult of leadership around Atatürk was indicative of how Turkish discourses at home were in conversation with the emerging political trends abroad. This interstate context was not limited to discursive borrowings. After all, political assassinations – due to the cross-border nature of their planning and implementation – could not only strain relations between neighboring states, but also possibly provide opportunities to facilitate interstate rapprochement. In 1935, too, this seems to be part of the calculations from the very start. For the British, tipping off Ankara certainly helped secure Turkey’s goodwill at a time of alliance-building in Europe. To be sure, doing so was also a way of deflecting any responsibility particularly if ‘an attempt, let alone a successful one’ were to be made by Turkish political refugees living in the British mandates of Palestine and Transjordan. ⁶⁸ Therefore, political assassinations, whether real or imagined, were opportune moments to initiate a range of productive interstate exchanges.

This was particularly the case from the early 1930s onwards when the political landscape across Europe was turning into a hotbed of controversies that gradually brought states into conflict with one another, characterized by an onset of high-profile assassinations across Europe. The assassination of the Austrian Chancellor Engelbert Dollfuss in July 1934 and that of King Alexander I of Yugoslavia in November of the same year illustrated the explosive potential of such acts of violence to strain interstate relations in Europe. The latter episode in particular resulted in nearly a year-long debate in the League of Nations on the suppression of international conspiracies and on state responsibilities, bringing to the forth deeply embedded tensions among state actors and ideologies, while also facilitating broader international cooperation. ⁶⁹ After all, such cross-border plots – similar to late nineteenth-century patterns – became prominent ways in which interstate actors sought influence by hosting, if not tacitly encouraging, groups seeking irredentist or revolutionary agendas. The corresponding increase in the number of assassinations in Europe did not go unnoticed by Turkish authorities who, by the summer 1932, issued clear directives to tighten the border controls of individual crossings through the Turkish customs, particularly those who might carry fake passports or IDs. ⁷⁰

Right after the receipt of the British intelligence in mid-August 1935, too, Ankara responded by taking extra security measures along its borders, which were duly announced to the borderlanders through the local press. ⁷¹ Yet, these measures were not limited to Turkey’s southern borders. The road connecting Ankara to the south via Konya was placed under strict surveillance; so was the motor traffic coming from Bolu, a region dotted by Circassian villages. ⁷² ‘Exceptional precautions’ were similarly taken to circumvent any ‘clandestine arrivals and departures’ in the province of Izmir in Western Turkey. Foreign sailors were accordingly subjected to heightened searches; yachts sailing in the Turkish waters were reported to have been intrusively monitored by the coastal patrols; and villagers by the coast were given arms and asked to be on the lookout for strangers. ⁷³ Such security practices by the coastal regions seem to have reflected a genuine concern about a possible infiltration of assassins into Turkish territories. After all, the Western coast had witnessed earlier incursions by the Circassian irregulars who crossed into Turkey from the Greek islands as recently as 1927. ⁷⁴

While such historical precedents conditioned the types of measures introduced on the Western coast, Turkish newspapers quickly embarked upon a discursive campaign on ‘the dangers lurking in the Middle East’, highlighting the alleged complicity of the French authorities in the assassination plot. ⁷⁵ Crucial in setting the tone of these discourses was the person of Yunus Nadi, who was both the deputy for Muğla and the chief editor of Cumhuriyet, the mouthpiece of the single-party government. As a member of parliament, Nadi had already played a key role by heading the committee that had stripped Ali Saip of his immunity and therefore had direct access to the ongoing policy discussions. ⁷⁶ Soon after news of the assassination attempt broke out, he quickly began to publicize an anti-French stance in his daily columns in Cumhuriyet, framing the mandatory authorities in Syria as responsible for the incursions across Turkey’s southern borders, including but not limited to those of the assassins. He questioned the anti-Turkish sentiments rampant among the French officers and highlighted the hypocrisy of the mandatory regime in Syria and Lebanon. In particular, Nadi pointed out how French authorities established Armenian villages along Turkey’s southern border, turned a blind eye to the Khoybun, the Kurdish revolutionary organization, and employed the Circassians in the gendarmerie as leverage against the local Arabs. ⁷⁷

To be sure, the promotion of such discursive frameworks was part and parcel of a larger effort by Ankara to mount pressure on the French in hopes to secure concrete concessions it had long sought, particularly the removal of Armenians, Kurds and Circassians away from the Turkish-Syrian border. ⁷⁸ When Şükrü Sökmensüer, Turkey’s director of General Security, left for Beirut upon the invitation of the French High Commissioner to co-investigate the plot, ⁷⁹ his primary goal was accordingly to map out ‘the seditious networks behind the assassination plot’, particularly those extending into Syria, ⁸⁰ In the two week-long meetings in Beirut, Sökmensüer presented the outlines of a complex plot that involved individuals such as Haco Agha, Kamuran Bedirhan and Eşref Kuşçubaşı, among many others. They presumably operated in a network that crisscrossed Eastern Mediterranean, connecting French Syria to the Greek islands, Egypt, Haifa and Beirut, ⁸¹ but the evidence presented by the Turkish authorities failed to convince them of the organization of a plot within Syria. ⁸² While the French were largely collaborative with Ankara in general terms, the initial propaganda mounted by the Turkish press remained to be a sore point, occasioning French protests through diplomatic channels. ⁸³

British authorities, on the other hand, were spared the criticism that was easily directed against the French, and were instead publicly framed as the most cooperative of the Great Powers. This was all the more interesting, because the alleged masterminds actually hailed from the British controlled mandates of Palestine and Transjordan. Fourteen out of the 16 plotters even lived in Amman. ⁸⁴ Yet, the broader geo-political considerations in fact dictated a rapprochement of relations between Turkey and Britain, and the conspiracy in some sense facilitated it. As one British official put it, ‘it would be an excellent thing if we could tell the Turks that the houses of these people (Circassian suspects) had been raided and if we could at the same time pass on to them some concrete results of the raids’. ⁸⁵ This they did on October 22, for instance, with raids in Amman and Haifa that resulted in the detentions of the alleged masterminds. ⁸⁶ Even though they were soon released due to lack of corroborating evidence, the British honored further Turkish requests in the upcoming days. Because Ankara was eager to map out the subversive networks Çerkez Ethem cultivated in Turkey, British authorities even provided the evidence to the Turkish consul, who took ‘copious notes’ trying to discover Ethem’s contacts inside in Turkey. ⁸⁷

Such a rapprochement between Turkey and England, marked by the collaboration in investigating the Circassian conspiracy, was in fact facilitated largely because of the new geopolitical exigencies in the Eastern Mediterranean brought about by the start of the Italian invasion of Ethiopia in October 1935. ⁸⁸ Being next to the Italian controlled Dodecanese islands, Turkey had long entertained a deeply-rooted fear of Italian expansionism that could target Asia Minor. ⁸⁹ Yet, Turkey was one among many that was threatened by Mussolini’s invasion of Ethiopia. Italian revisionism also illustrated the lack of British military preparations against such military aggression. ⁹⁰ Precisely at the moment that London sought to cultivate the goodwill of Turkey as a bulwark against Italy in the Eastern Mediterranean, the Circassian conspiracy of 1935 began to play into the broader improvement of bilateral relations between the two countries. As the interests of both countries overlapped, this context of ‘the further rapid growth and strengthening of Anglo-Turkish friendship and mutual confidence’ soon led King Edward VIII to visit Turkey in September 1936. ⁹¹

The fears of Italian expansionism had other consequences, too. In neighboring Greece, for instance, another country that was wary of Italian aggression, newspapers were quick to report that Italian currency was allegedly found on the detained Circassian suspects. This was a claim which presumably linked the plot to the machinations of the Italian intelligence services. ⁹² Yet, this was certainly neither the first rumor, nor the last one that the Circassian conspiracy of 1935 came to produce. In fact, conspiracies, which are clandestine by design, embody an unmatched potential to result in a constant cycle of rumors. This was particularly the case in places like Turkey, where the suppression of the press led to the expansion of unofficial channels of information, thereby paving the way for rumors to spread rapidly. ⁹³

In the case of the Circassian conspiracy, too, Turkish newspapers quickly found themselves in a position to report on the plot in the absence of reliable channels of information, whereby, operating like a rumor mill, they were extra vulnerable to, and responsible for, the spread of misinformation. In its 27 October issue, the Istanbul daily Cumhuriyet, for instance, quoted a Greek newspaper Eleftheros Anthropos which read: ‘reporting from Istanbul: even though important documents were found on the detained individuals, Ankara continues to keep its silence, but it is understood that the assassins were in contact with foreign agents’. ⁹⁴ Even though the Greek newspaper openly relied upon the Istanbul press for its own coverage in the first place, the editorial at Cumhuriyet still went ahead and published it. In quoting one another from Istanbul to Athens and then vice versa, newspapers added an aura of authenticity to their reporting by transmitting unverifiable oral information into the generically reliable format of the print.

Of course, the newspapers had their own vested interest in doing so, as sensational news – often printed in large headlines – helped them sell well. It also mattered little whether they were recycling rumors or not. They often did not hide it when they did. One headline announced that ‘the traitors who plotted the assassination apparently numbered more than hundred’. ⁹⁵ In reporting as such, the newspapers did not shy away from using reported past tense in their coverage, even though its use was grammatically limited to instances where the actions are heard, but not seen, let alone verified. This process of reporting at will was not devoid of bureaucratic intervention, however, because states naturally had direct interest in keeping narratives under their control. As we have already seen, Ankara immediately sought to benefit from the populist momentum the press coverage came to create. French authorities in Syria, on the other hand, suffered from the negative publicity created by Syrian coverage of the plot. Hoping to set the narrative straight, French authorities began to brief journalists officially about the developments, thereby trying to prevent the further circulation of misinformation. ⁹⁶

The politics of what constituted the truth involved the output of other state actors as well. Foreign diplomatic corps, for one, also picked up and recycled the rumors circulating in Ankara, albeit certainly in politically savvy ways. ⁹⁷ Even though in their reports the diplomats clearly pointed out that the conspiracy helped advance Ankara’s domestic agenda, they were also eager to report on the factionalism the plot came to expose in the Turkish capital. One such report from the American Embassy in Ankara claimed that the conspiracy was blown out of its proportions by the interior minister Şükrü Kaya who ‘was in very serious political trouble’ with the premier İsmet İnönü. The timing of the plot could not be better for him, since the report suggested that the rumors were already circulating that İnönü was about to name a replacement for Kaya, whom he held responsible for what he had seen in a recent trip to Anatolia. Yet, ‘it was surely unwise to change the head of the Ministry’ at a time when it was the most necessary for it to function seamlessly. ⁹⁸ No matter how plausible such rumors may have sounded, it was never easy for diplomats to verify them. ⁹⁹ Nor did it get any easier for later generations of historians.

Rumors did not cease to exist, either, even years after the acquittal of the suspects. In fact, with the expansion of the Turkish public sphere after the Second World War – ushered in by the multiparty politics and the growing number of independent newspapers – the Circassian conspiracy quickly began to experience an interesting afterlife. After all, the acrimonious public quarrels, so characteristic of this period, did not seek to remove the specter of doubt and reveal the truth for good. Instead, they often unfolded by digging up the past to serve the present political purposes, and in doing so came to recycle a whole new set of rumors. One such debate took place between Şükrü Sökmensüer and the journalist Cihad Baban in late 1948. Sökmensüer was the director of General Security and the head of the Turkish police at the time of the Circassian conspiracy, and he later served as the minister of interior in 1946–7. In the otherwise convoluted debate that took place on the pages of Tasvir, which exposed all the contours of the postwar discourses on communism, free press, and democracy, the Circassian conspiracy quickly resurfaced as an episode from Sökmensüer’s dark past – one that presumably illustrated his abuse of power in pursuit of personal agendas.

For Baban, the Circassian conspiracy was a façade put together by Sökmensüer, so as to achieve higher office and greater personal influence by getting on the good side of Atatürk. Baban insisted that Sökmensüer had in fact embodied ‘a sinister police psyche’, which he defined as ‘a constant state of doubting everyone’. This ‘condition’ from which he suffered, Baban claimed, had turned Sökmensüer into an individual who always saw the hands of agitators behind what might otherwise be a seemingly ordinary event. ¹⁰⁰ It was this mentality that was responsible for torturing innocent people like Yahya under custody and bringing an end to the political careers of individuals such as the Urfa deputy Ali Saib. For Baban, Sökmensüer had quit the military and decided to pursue a career in the police for a good reason. In the end, Sökmensüer was attracted by nature to ‘professions defined by mysterious and clandestine adventures, informed by top secret reports’. The police work was an occupation Sökmensüer excelled at, Baban claimed, with a level of ‘efficiency that made [Heinrich] Himmler pale in comparison’. ¹⁰¹

In spring 1975, many years later after this public quarrel, Sökmensüer sat down in an interview with Arı İnan in the province of Hatay, not very far from where Yahya and Ali Saib were first detained back in 1935. Among many other things, Sökmensüer talked about that day was also the Circassian conspiracy that took place 40 years earlier. At the age of 85, his memory was still surprisingly lucid. Yet, at the end of his account of the conspiracy, there was a particular moment when Sökmensüer began to mix up dates, individuals, and events. This is exactly where he brought up the later accusations towards him in the press, how he was publicly blamed for masterminding the plot himself. In hopes to set the historical record straight, ‘it all started with the note of the British ambassador … how could I be the ringleader?’ Sökmensüer asked. ‘Did I dictate the note to the British government myself? … I am a man who simply carried out the orders of Atatürk.’ ¹⁰²

Atatürk’s voice had been conspicuously absent throughout this entire episode, even if it was his life that was presumably at stake. According to his secretary-general, ‘he hardly paid any attention to the different phases of the legal proceedings and did not make a fuss about its outcome’. ¹⁰³ The public prosecutor Baha Arıkan similarly stressed that the episode illustrated Atatürk’s full confidence in Turkish judges. So did the accounts of his contemporaries such as Kazım Özalp who painted a picture of a coolheaded leader respectful of the rule of law. ¹⁰⁴ Yet, Atatürk was in fact much more involved in the process than one may initially assume. Arıkan himself noted that he was asked to visit Atatürk after each session to personally inform him about the proceedings, and how, at least in one instance, Atatürk snapped at him after asking when the trial was going to conclude. ¹⁰⁵ Tellingly, one of the two things that the historian Yusuf Hikmet Bayur could remember many years later about the trial was Atatürk’s frustration with the inexperience of Arıkan: ‘How could such a case be entrusted to a boy?’ ¹⁰⁶

Atatürk’s involvement in the investigation reached a new high after the dismissal of the case in February 1936. ‘Enveloped by an ever-expanding shadow of uncertainty’, he prepared and forwarded a list of nine questions about the trial, demanding answers to his doubts. Unsatisfied with the written answers, Atatürk called İnönü, Aras, Bayar, Sökmensüer, and Arıkan for a personal audience in Istanbul. ¹⁰⁷ Their meeting first started as private conversations on Atatürk’s yacht and, by the evening, continued at the rakı table in Dolmabahçe Palace, lasting – as it usually did – until the next morning. ¹⁰⁸ As Atatürk drank his rakı followed by roasted chickpeas, he began to ask for more evidence to be telegraphed from Ankara and Istanbul, in particular a range of private correspondence related to Ali Saip that he carefully examined. Atatürk’s own conclusion at the rakı table that night was that the Circassian conspiracy was all real, but that the deputy Ali Saib’s name was mixed up in the event, only because he was a well-known patron of smugglers across the region, with an extensive network of illicit cross-border contacts. ¹⁰⁹

‘Evidence of such clandestine activity tends to be circumstantial and hard to corroborate’, noted Brian Jenkins. ‘In the end interpretation is inevitably influenced by the political perspective of the observer.’ ¹¹⁰ Was this the case with Atatürk? Here was an astute military officer who was seasoned in the intricacies of contentious politics and the violent know-how of revolutionary conduct, surviving the tumultuous transition from the collapse of empires to the making of new nation-states. Did his past experiences shape his later judgments? Commenting on his heavy drinking habits at such rakı tables, one diplomatic report from many years prior reported that ‘the Ghazi’s mind is already becoming affected…. One rumor, which I have the honor to report with all reserve, is that the President is already suffering from a mania of persecution… he is now more afraid of attempts on his life.’ ¹¹¹ Should a historian reproduce rumors just as this diplomat did, ‘with all reserve’? Indeed, how to study the circumstantial evidence borne out of the curious nexus between conspiracies and rumors?

I have suggested in the beginning that historians should in fact resist the temptation to authenticate conspiracies. After all, rumors are oral affairs, and conspiracies unfold without written directives. The documents historians work with, on the other hand, are imperfect bureaucratic filters, particularly more so when they report on rumors and conspiracies. ¹¹² For a historian then, there is no perfect rumor that fully reflects the popular mentalities; no conspiracy is a purely non-state affair, either. What there is, I argue, is a co-productive process between state elites and flows of unverifiable information, a process which historians could try to capture. As it was case with the Circassian conspiracy of 1935, it mattered little if the initial intelligence actually drew from a pool of circulating rumors or produced new ones, for it had very real implications in sharpening the tone of domestic politics, while also setting the background for broader alignments in international politics. Illustrated by the set of events that followed the receipt of the British intelligence in Ankara, the curious interaction between rumors and conspiracies ultimately helped shore up the legitimacy of the Kemalist regime and propagate further the personality cult of Atatürk. They also prepared the groundwork for growing cooperation among states that favored regional status quo, which was crucial at a time when territorial revisionism increasingly tested waters across the Eastern Mediterranean.