INTRODUCTION
This article offers a new perspective on how to understand collective violence in the Ottoman Empire through an historical ethnography of how observers, joiners, and victims of the anti-Armenian pogroms in 1895 in Harput, a provincial capital town in the Ottoman East, theorized the atrocities taking place before their eyes. Notwithstanding the wildly conflicting accounts of the massacres, I have discovered that narrators from all different sides agreed on the social mechanism driving the various collectivities into violent action—namely, that violence was the outcome of sinister deception by imposters. Transcending the conventional dichotomy between historical facts and contemporary theory, this historical ethnography seeks to provide, following Marilyn Strathern, “what an indigenous ‘analysis’ might look like.”Footnote 1 The archival material was scrutinized not only to understand the facts but also to deduce theory, in this case a theory of collective action from the point of view of historical actors.
Methodologically, this article is an exercise in “connected history”Footnote 2 on three layers. First of all, by emphasizing the notion of “shared life,” I seek to overcome the teleological supposition that clear and strong divisions existed in societies where collective violence took place. Deception, the central theme of the article, emerges as the key social knot that mediates between violence and shared life. Second, I go beyond the traditional barriers of area studies that tend to dwell on one source or one people and create homogenous and insular images of ethno-religious groups. I use a variety of sources, from official correspondences to letters, memoirs, and petitions in Ottoman Turkish, English, and Armenian, to uncover common conceptual points shared by multiple and otherwise different perspectives. Last, my analysis challenges the distinction between intellectual history and social history by employing the methods of social history to arrive at dominant ideas about crowds in the fin-de-siècle Ottoman Empire.
The article will proceed as follows: The next section briefly details a conceptual sketch of the relationship between deception, collective violence, and shared life. I will then introduce the reader to the city of Harput, the main crime-scene for my analysis. I highlight some of its historically unique characteristics and suggest that Harput presents an ideal case study for understanding the element of deception in collective violence. The third section zooms out from Harput to present the main features of the 1894–1896 Armenian Massacres. In an unorthodox way, I situate the massacres in a context of emerging crowd politics through connected history of the mid-1890s. The paper's last two sections analyze the archival material in a comprehensive way and uncover how crowd behavior was theorized by lay observers of the 1895 Massacres in Harput. I show that the character of the imposter and the performance of disguise were at the center of all narratives of violence and argue that, for the contemporaries, crowds could not be incited to violence by provocation, but only by deception. A concluding section draws the key themes and findings together and calls attention to their relevancy today, in the post-truth era.
DECEPTION AND SHARED LIFE
Deception lay at the intersection of the otherwise competing narratives of the Harput pogroms in 1895. I need at the outset to clarify how deception can serve as an analytical tool that links collective violence to shared life. In his analysis of the anti-Christian violence in Indonesia in 1999, Nils Bubandt reveals “the close connection between empathy, deception, and violence” by focusing on fake letters that carried the signatures of Protestants but were in fact written and disseminated by Muslim provocateurs.Footnote 3 He rejects the widespread understanding of empathy as a remedy for violence, and argues instead that the empathetic sense of the enemy's point of view can actually be deployed for violent purposes. Bubandt's choice of terminology may not be the best fit for his case. Indeed, it is something of a stretch to refer to the dissemination of an image caricaturing Christians, written by and for Muslims, as empathetic.Footnote 4 Moreover, one cannot speak of that case as a proper deception since Bubandt tells us that it was widely known or assumed that the letters were fake. Having said that, I propose to strengthen Bubandt's thought-provoking conclusions by replacing his emphasis on empathy with the notion of the “shared life.” The Muslim authors of the fake letters just needed “an intimate knowledge of Christian rhetoric,”Footnote 5 which, I assert, indexes a shared life, rather than a divided one that needs to be connected by empathy. Empathy presupposes an alterity, as a bridge presupposes two riversides; deception merely requires a shared life.
My emphasis on “shared life” in this article thus aims to forestall the common tendency in analysis of communal violence to fall back onto arguments about otherization and alterity. In her ethnographic work on Rwandan Genocide, Lee Ann Fujii rejected the dehumanization argument and showed that, even in the middle of violence, the joiners did not see their Tutsi targets as distant others inviting extermination. The dominant narrative was not that of ethnic hatred.Footnote 6 Similarly, Stanley J. Tambiah reminded us in his work on ethnonationalist violence that collective violence tends to be unleashed when there are no “firm partitions between the antagonists.” In fact, “the sudden imposition of difference on the basis of previous solidarity and coexistence,” as opposed to being an absolute other, triggers crowd violence.Footnote 7 In the present research as well, the mechanism of deception stands out as an index of a shared life rather than a divided one, since one can only deceive those who are in close proximity. As Bubandt suggests, in order to work deception needs familiarity and intimacy among parties—a distant other can be lied to but cannot be deceived. Thus, we would do well to consider replacing the figure of the Schmittian absolute other (enemy), who is “existentially something different and alien,”Footnote 8 with the Simmelian figure of the stranger, who is instead an “organic member of the group” and whose position in the group is “composed of … nearness and distance.”Footnote 9
The power of deception lies in its ability to change the mind of the addressees (the deceived) by making them believe their action is somehow natural, self-determined, and legitimate, or in Michel Foucault's terms, by subjectification through conduct of conduct.Footnote 10 As scholars have recently emphasized, simple lies need not deceive or even have the intention of deceiving the audience, and deception need not involve any actual lies.Footnote 11 Unlike lying, deception seeks to drive the action of others through the creation of an emotional situation, for which a shared life has to be pre-established. Thus, whereas empathy, provocation, public secrets, and simple lies work by playing on already-deep divisions in the society and reinforcing divisions of “we” and “they,” deception blurs the boundaries and aims to change the mind of the addressee. It counters the otherization arguments widespread in existing studies of collective violence and suggests, instead, the malleability of subjects. In Hanna Arendt's phrasing, the deceptive liar in politics “is an actor by nature; he says what is not so because he wants things to be different from what they are—that is, he wants to change the world.”Footnote 12
HARPUT: AN IDEAL LOCATION FOR IMPOSTERS
Harput province lay right at the invisible border of the Ottoman East, “a region [that] has been characterized by some as (Western) Armenia, by others as (Northern) Kurdistan or as Eastern Anatolia.”Footnote 13 Its demography reflects the characteristics of a border zone; throughout the nineteenth century, Armenians constituted around one-third of the population in the province, the rest being Turks and Kurds (with a meager number of Assyrians).Footnote 14 Thus, Harput province was never associated with a single ethno-religious group, unlike some regions like Diyarbekir (“Kurdish”) or Van (“Armenian”). It was a perfect location for a relatively balanced shared life among communities. Needless to say, shared life did not mean peaceful coexistence and never precluded violence, but it influenced the ways in which violence was unleashed.
The town of Harput, the capital of the province, was a classic fortress city that sat atop a hill, commanding a vast plain of villages and small towns below. As part of centralization efforts in the 1830s and 1840s, the imperial army used Harput as a command outpost, and as a rear base when operations advanced to the east during the wars against the Kurdish emirates, the hitherto semi-autonomous ruling power in the region. Though temporary, the government presence transformed the layout of the town; from the 1850s onwards, it expanded onto the plains below to accommodate a larger (mostly Muslim) administrative class. The old upper town was coupled with the new lower town (called Mezre, later Mamuretülaziz), creating a dual-city space reminiscent of colonial urbanism.Footnote 15 At the same time, however, the lower town became an attractive location for the region's newly rising, mostly Armenian commercial bourgeoisie. It became a haven for outsiders coming from smaller towns to embark upon a new life. By the century's turn, more Armenians than Muslims lived in this new part of Harput, even though it was the seat of the local Turkish government. Here, bourgeois suburbanization, combined with state-making in the periphery, forged an elite form of shared life.Footnote 16
For this study, Harput is what Matei Candea has labeled an “arbitrary location,” in that it has “no necessary relation to the wider object of study.”Footnote 17 In other words, I am confident that one could study any of the other places where pogroms took place in the Fall of 1895 and obtain similar findings. That said, Harput is an ideal case through which to study the element of deception in collective violence. First of all, as Sianne Ngai has pointed out for antebellum America, when money capitalism and a modern banking system spread, “confidence” turned into a public feeling with crucial significance because strangers had to do business just by signing papers. A crucial side effect was that fake IDs and imposters proliferated during such periods.Footnote 18
The empire-wide expansion of money capitalism in the nineteenth century was amplified in Harput thanks to its already well-established craft tradition and entrepreneurial classes. Starting in the 1840s, economic power gradually shifted from the landed gentry to tradesmen and manufacturers. For instance, the leading dynastic family of Harput, the Çötelizades, was driven into a corner when bankers and merchants filed a number of debt cases against them, which resulted also in their dismissal from political councils. The new world was one of contracts and signatures rather than genealogical connections and rooted alliances, and contracts and signatures could be faked. Though the Çötelizades tried to adapt to the new system by registering four-year-old family members as tax-farmers, they were doomed.Footnote 19 Moreover, the financialization of the economy not only empowered the commercial groups but also challenged the traditional land boundaries since land itself was rapidly commodified. As archpriest Boğos Natanyan reported in 1878, in the past all land in Harput had belonged to Muslim aghas, but in recent decades Armenians had bought substantial land from Muslims and become landlords.Footnote 20 The spread of money capitalism blurred social distinctions and opened up spaces for transitivity and imitation.
Arjun Appadurai has placed special emphasis on “the language of [the] imposter, the secret agent, and the counterfeit person” that is common in narratives of massacres in places recently hit by transregional effects.Footnote 21 Such effects tend to generate a “distorted relationship between daily, face-to-face relations and the [associated] large-scale identities”; in their daily exchanges with others, people are “animated by a perceived violation of the sense of knowing who the Other was and of rage about who they really turn out to be.”Footnote 22 Harput province, compared to other regions of the Ottoman East, accommodated a high degree of transregional experience. Even in the earlier decades of the nineteenth century there was a tradition of long-distance, temporary labor migration, mostly by men, to large cities like Istanbul, Beirut, Adana, and Tiflis.Footnote 23 Many towns in the province were economically dependent on remittances coming from abroad or otherwise outside.
In the century's second half, Harput's transregional connections were augmented thanks to its distinctive position in the newly emerging Protestant missionary movement. In the 1850s, when imperial reforms were instituted to expand religious freedom, American missionaries expanded their activities from the big coastal cities to the entire inland empire. They targeted Christians who belonged to Eastern Churches, basically Greeks and Armenians. Harput grew rapidly into one of the most successful mission stations in the empire for several reasons, including the extraordinary strength of the missionary cadre in Harput and a hunger for new ideas among Armenian middle classes. The city became a regional hub for students due to its outstanding educational facilities ranging from kindergartens to the Euphrates College and the Theological Seminary.Footnote 24
In addition to the American presence in Harput's daily life, the mission work in the city multiplied its inhabitants’ extra-regional connections in two other ways. First, in reaction to the success of the mission schools, Gregorian Armenians reformed and modernized their own schools, and by virtue of this competition Harput emerged as one of the most important intellectual centers for the Armenian community in the provinces and made itself heard in the capital. For example, the realist genre of provincial literature in Armenian writing, which challenged the elite literature of Istanbul, was created toward the end of the century by native writers of Harput.Footnote 25 Second, Harput served as the regional gate for Armenians (and later, Muslims) traveling to the Americas during the period of mass migration to the New World. As David Gutman concluded, “Between the late 1880s and the onset of the First World War, more than seventy-five thousand Armenians departed the Ottoman Empire for the United States. The majority of these migrants, especially in the years before the 1908 Revolution, originated from in and around the twin cities of Harput/Mezre, the administrative center of Mamuretülaziz province.”Footnote 26
In addition to money capitalism and transregionalism, the urbanization that took place in Harput makes it an ideal location for studying the element of deception. Sheila Fitzpatrick has meticulously shown the connection between new urban society and the rise of imposters and conmen in Soviet Russia in the 1920s and 1930s. The impersonal relations that characterized the urban space and the emergence of a new (non-aristocratic, non-bourgeois) elite brought forth anxieties linked to the inability to be sure of others’ identities and the “shocking revelation that someone close and trusted was not what he or she seemed.”Footnote 27 The dual-city of Harput underwent an urban transformation in the 1880s and 1890s due to population growth and the city's expansion. With reforms in town administration and the creation of new municipalities, the notables of Harput pushed for urban renewal projects. The marketplace was reorganized, the streets and sidewalks were paved, and new grid-style neighborhoods were created.Footnote 28 By the end of the century, Harput had become one of the most densely transregional, urbanized, financialized, cosmopolitan, and multi-ethnic places in Ottoman Anatolia, and a highlight of the eastern provinces. It was a perfect environment for imposters to flourish in.
THE ERA OF CROWDS AND THE 1895 MASSACRES
The year 1895 was a turning point in the history of mass politics in the Ottoman Empire. In that year, a diplomatic crisis electrified politics, a mass demonstration shook the capital, widespread massacres broke out in the provinces, and the leading Turkish and Armenian political parties underwent irreversible transformations. Historians across different literatures—such as political history, intellectual history, and Armenian history—have largely interpreted these events through the lens of high politics. However, I contend that developments in 1895 also marked the beginning of the era of crowds for the Ottomans. In Ottoman historiography the few studies of the politics of the masses have been mainly confined to the 1908 Constitutional Revolution and its aftermath.Footnote 29 This section situates the year 1895 in the history of mass politics, and summarizes the main events that led to the massacres.
The regime of Sultan Abdülhamid II (r. 1876–1909) gave birth to mass politics, and it was mass politics that ultimately did the regime in. The distinctive feature of the regime was its concentration of power in the sultan's palace at the expense of the larger bureaucracy, concurrent with the invention of the masses and mass media as the new sources of legitimacy.Footnote 30 Thus, counterintuitively, it was an autocratic regime “with a vibrant public sphere and an expanding civil society.”Footnote 31 The constitutional movement that burgeoned among the ranks of the bureaucratic cadres (or with their support) in the Age of Reform (Tanzimat, 1839–1876) was now pushed into political opposition, given Abdülhamid's suspension of the first Ottoman parliament at the end of its first year, in 1878. Toward the end of the 1880s, Ottoman reformists and revolutionaries began founding political parties to fight Hamidian despotism, among which the Committee of Union and Progress (CUP) and the Armenian Hnchak Party stood out in the first half of the 1890s.
That period was marked by a crescendo in violence and aggressiveness against Armenian peasants in the eastern provinces. The origins of the oppression specific to this region lay in the Ottoman government's aforementioned centralization efforts in the 1830s and 1840s. The military expeditions against the Kurdish emirates (commanded first from Harput) succeeded in removing the local dynastic leaders yet failed to replace them with new governmental organizations. A power vacuum emerged, and the remaining Kurdish tribes and confederations competed for power and leadership with unchecked aggressiveness. Sedentary peasant communities in general and non-Muslim villages in particular were the main victims of this process since they had to pay escalating extortion payments and suffered arbitrary coercion.Footnote 32 During the reign of Abdulhamid II, two interrelated developments further complicated the already worsening situation of Armenian peasants: First, at the Congress of Berlin in 1878, European powers acknowledged the Armenians’ complaints and asked the sultan to institute reforms, thus internationalizing the “Armenian issue.” Second, partly as a response, the Turkish government's armed forces began taking a more active role on the side of chosen, collaborative tribes at the expense of Armenian communities. This collaboration reached its apex in the 1890s when Hamidiye Regiments were formed as locally organized crime units with official backing.Footnote 33
To resist the routinization of violence, in the 1890s Armenian peasants took up arms to defend themselves, and during the same period the newly founded Armenian revolutionary groups intensified their activities. As Toygun Altıntaş's meticulous work has detailed, “armed resistance [in some localities] was not the result of revolutionary organization … [but rather] the nascent revolutionary networks in the region sought to incorporate existing practices of resistance and redirect Armenian grievances to the Ottoman state.”Footnote 34 When the government used such activities as a pretext for increasingly reinforcing its practices of cooptation with Kurdish pastoralist tribes, everyday violence turned into openly conducted massacres. The Sasun Massacre in summer of 1894 was followed by pogroms in many towns and villages in the eastern provinces in fall of 1895 and again in 1896—a series of events known today as “the 1894–1896 Armenian Massacres” or “the Hamidian Massacres.” Each wave of violence though they shared common features, had a different immediate context, and in what follows I will concentrate on the 1895 Massacres.
International politics was decisive for the chronology of the events in 1895. Until the 1890s, the European powers were not particularly ardent in compelling the sultan to implement the reforms proposed at the Congress of Berlin, a reticence that came in for criticism from the new Armenian political parties. However, when Armenians in Sasun were attacked and thousands were killed by Kurdish tribes with the support of government forces in August 1894, international actors were once again forced to confront the Armenian issue head-on.Footnote 35 In December 1894, a multi-nation investigative committee was formed, and the following May the European powers proposed a detailed reform plan.Footnote 36 It addressed several practical measures such as pardoning Armenian political prisoners, securing reparations in the massacre area, returning Armenian refugees, making new appointments in local administrations, and establishing institutions to oversee the implementation of reforms.Footnote 37 Over the next five months, until the sultan finally accepted a pared-down version of the European plan in October, the government faced intense pressure from foreign powers and the threat of a rapidly ripening political opposition at home. As an example of the latter, the police began a thorough investigation among royal academy students. Dozens were interrogated, and a raft of secret CUP networks was ultimately exposed.Footnote 38
On 30 September 1895, the Hnchak Party organized a protest rally in Istanbul to denounce the government's reluctance in initiating reforms. Protesters also called for justice for the Sasun Massacre victims and regarding the plight of Ottoman Armenians in general, but when thousands marched from the patriarchate to the seat of government the police intervened and violent clashes ensued. Over the following days, pogroms took place in residential neighborhoods across the entire city and against all Armenian inhabitants.Footnote 39 Three weeks after the Istanbul rally, news reached the provinces that the sultan had accepted the European reform plan, which immediately instigated armed attacks against Armenians there. Across a few short weeks through October and November, militias, tribesmen, soldiers, and even ordinary Muslims killed around one hundred thousand Armenians in the eastern provinces of the Ottoman Empire, mostly in a brutal and wild fashion. These killings marked the apogee of anti-Armenian violence in the mid-1890s.
Until recently scant academic work had been done on the 1894–1896 massacres, and all other references to this event were incorrect projections based on Armenian Genocide (1915) studies.Footnote 40 In the last few years, a renewed interest in the topic has generated new micro-historical studies that have debunked the teleological perspective and revealed distinctive aspects of these event.Footnote 41 Elsewhere, I have presented a detailed chronology of the Harput pogroms and argued that the core of these events was not Armenian “mischief” or government conspiracy, but rather the invasion of the city by pastoralist Kurdish tribesmen.Footnote 42 Yet, so far, most available studies have restricted themselves to political explanations (blaming anti-Christian policies, the Armenian revolutionary movement, the Hamidian regime of despotism, etc.) or economic ones (pointing to land conflicts, class-based resentment, and so forth). They have left out the central question of this article, one focused squarely on historical ethnography: How did people theorize what they and others had done?
The old literature had an answer to this question: the provocation thesis. It was grounded in the idea that the Armenian political groups intentionally acted in a provocative manner to instigate the Muslim crowds to attack Armenians. It theorized the massacre as an instance of automated crowd response to defiance by a weaker social group. The provocation thesis was prevalent in the past, but critical scholarship, especially in the last two decades, has debunked it on the grounds that it serves to justify massacres, places blame on victims’ shoulders, and wrongly assumes that the state's murderous response would always be predicted by revolutionaries.Footnote 43 I have also shown for the Harput case that there was no provocation whatsoever.Footnote 44 Nonetheless, although it is factually groundless and politically incorrect, the provocation thesis has the merit that it conceptualizes the massacres from the perspective of mass politics and crowd behavior. Though, as I will show, the thesis is wrong, we need to do justice and criticize it not only factually or politically, but also theoretically. I will argue that local people at the time of the massacres believed that the masses had been steered by deception, not provocation.
Though this article is mainly interested in the ways local people with first-hand experience theorized the massacres, political leaders and intellectuals also tried to make sense of the mass violence. In fact, the transformation the CUP and the Hnchak Party went through after the massacres, and the ever-increasing influence Gustave Le Bon's gained in Young Turk ideology, were both directly related to the massacres. Regarding the former, generally forgotten is that the Hnchak rally in Istanbul also gave rise to the first openly political act of the Young Turk movement. In the following days, some of CUP's founding members prepared an appeal and distributed a thousand copies in Istanbul streets, especially in the mosques, calling upon the masses to rise up against the government. The appeal criticized the Armenian revolutionary strategies but also called for cooperation with Armenians against the despotic regime. The authorities responded with mass arrests and prosecutions of the CUP leadership, and then exiled almost all of them, including the movement's founders. As a result, in 1895 a new central committee was formed, transforming “the CUP in the capital … from a student organization into a committee of high-ranking bureaucrats and ulema.”Footnote 45
The massacres also transformed the Armenian revolutionary movement. The Hnchak Party never recovered from the calamity, having split into factions due to debates over the causes of the massacres. Subsequently, the party ceded leadership of the movement to the Armenian Revolutionary Federation.Footnote 46 As we have seen, the histories of the CUP and the Hnchak Party were intertwined. The rally the Hnchaks organized in Istanbul led to the CUP's first-ever popular action and its subsequent transformation due to exiles. The massacres, in turn, affected the Armenian political movement since it sidelined socialism and the issue of autonomy within the Hnchak programs. In addition, in December Ahmet Rıza, the now undisputed leader of the CUP, began publishing his Meşveret (Consultation) in Paris, which for months was dedicated to discussions of the massacres and the Armenian cause.Footnote 47 After the exiles of 1895, both political movements continued their activities abroad, often in collaboration. The leaders clearly saw the power of the masses and they concentrated on propaganda activity using the mass media that was by then freely accessible in Europe and elsewhere.
In this post-massacre period, the Young Turks’ ideas about mass politics and public opinion ripened. One of the most important ideologues for the Young Turks, and later for the Kemalists, was Dr. Abdullah Cevdet (1869–1932), one of four founders of the CUP and a dedicated translator of Gustave Le Bon. He was arrested and exiled in 1895 and later became one of the core editors of Osmanlı (Ottoman) in Geneva, the new official organ of the party. Over the following years, however, he gradually separated from the CUP. In 1905, he moved to Cairo with his own infant periodical, İçtihad (Opinion), which he published for twenty-eight years until his last day (1904–1932). There were personal and political reasons for his retreat from the CUP, but at issue also was an ideological difference in the conception of mass politics. That Osmanlı, with its highly sophisticated style, struggled to reach a wider audience led Cevdet to turn his attentions to mass psychology.Footnote 48
It seems that the 1895 Massacres revealed to Cevdet the dark power of the masses, as the Paris Commune had done for Le Bon. Cevdet diverged from the elitist understanding of reform from above toward a more Foucauldian approach of directing the conduct of the masses with an eye to long-term success. Scholars working on the Young Turks have all pointed to elitism as a key characteristic, if not the primary one, of their ideology and practice.Footnote 49 Yet, the Platonic elitism of the later CUP (and Kemalism) was different from the elitism in Abdullah Cevdet and Le Bon's work: in the former, the elites dictate the rules, whereas in the latter they manipulate the masses. For the same reason, we must question the prevalent view that Abdullah Cevdet's (and the Young Turks’) intense interest in Le Bon (instead of Durkheim or Tarde) was as a sign of theoretical weakness.Footnote 50 My historical human sources, most of them educated but none of them a philosopher, conceptualized crowd behavior in practice in the same manner Le Bon did: crowds could not be mobilized by either simple orders from their own side or open provocation from the enemy's side. Instead, they needed to be patiently worked on: behave like their friend, implant ideas, convince them of pseudo-facts about the enemy, and manipulate their behavior. Thus, Le Bon's theories prevailed in the thought of Young Turks, not because they could not understand sophisticated theories, but because his ideas concerning crowd behavior were shared by lay people in the provinces who had never read The Crowd, which he also published in 1895.Footnote 51
DISGUISE AS A WAY OF DECEIVING THE OTHER
I turn now to how observers in Harput conceptualized what happened on 11 November 1895. My sources are correspondence between the local government and the center, letters of the American missionaries to their headquarters in Boston, and the reports published in Armenian periodicals and books. Shortly after the massacres began erupting in the eastern provinces in late October, the central government in Istanbul sent a circular to the provinces to alert the governors and order them to take necessary measures. Though at first sight the circular appears a classic example of the sort of thing highlighted by the provocation thesis, in essence it was a deception:
Due to the Muslim uprising that the Armenians’ scattered revolutionary movement produced in places, the two groups sprang into mutual slaughter and pillage. Some of the Armenian mischief-makers—who stand out as the chief cause of the emergence of this worrying situation and who committed the major crime of disturbing the public order—dressed up as Muslim clergy by wrapping a turban on their heads. The latter put about hypocritical and diabolical words among Muslim villagers who are ignorant of the real situation, including the idea that Armenians are rising against Muslims, and that it is a sacred duty to immediately take revenge. This stirred up the communal bigotry of those who are not able to distinguish the real meaning of the content of such mischievous language. And by spreading inappropriate statements [knowing well] that they [Muslims] were bound to repeat them—so that strangers would be led to believe that Muslims were responsible for the mutual slaughter and plunder—this fraudulent plot is being prosecuted to the advantage of evil-wishers.Footnote 52
In line with the classic provocation thesis, here a group of Armenian “evil-wishers” is said to have provoked the Muslim community, and particularly the Kurds, with the perlocutionary force of words and rumors, inevitably causing Muslims to collectively rise up.
Nevertheless, the middle part of the circular contains something redundant for the straightforward provocation argument. There is something, at least to today's eyes, excessive in this paragraph: namely, the case of dressing up. For provocation, it should be sufficient to state that infidel Armenians were spreading malefic rumors and thus provoking Muslims to act against themselves. Why did a serious circular, then, make mention of Armenian “mischief-makers” dressed as Muslim clergy, something bizarre and almost comical? Shall we take these words simply as rhetoric or narrative strategy, or as an invented pretext? If the latter, we must ask why a pretext was needed, given that the provocation thesis already criminalized Armenians.
The circular was not unique; the representatives of the Assyrian community in Harput also blamed Armenians for assuming fake identities. They appealed to the local governor and later sent a petition to the acting patriarch of the Assyrian Church in Istanbul demanding that Harput's Assyrian community be distinguished (tefrîk) from Armenians in every aspect. The petition was framed in a rather defensive tone, as an explanation of why some of their members were killed and their properties were plundered during the attacks despite their being perfectly loyal (implying that Armenians were not). They were also attacked because the Armenians who dressed as Muslim clergy with turbans wrapped on their heads approached the Kurds and showed them Assyrian dwellings: “These are the houses of infidel Armenians; hurry, attack!” Assyrians had always lived with their Muslim neighbors as a family, the petition continues, and it was the Armenians who consistently disturbed the peace; they, too, wished to steadily ruin the Assyrian community.Footnote 53
Armenians’ ability to alternate between identities was indicative of shared life among communities. Contemporary accounts were not short on such instances of disguise, although the reasons and the forms of disguise varied. In Tokad, for instance, it was reported that many Muslims were attacked by Armenians in Georgian dress with Martini rifles.Footnote 54 In Zeki Pasha's Sasun report, revolutionary leader Hampartsum Boyajian disguised himself as a European, this time to convince the peasants to attack the pastoralist Kurds.Footnote 55 Armenians dressed up not only to deceive, but also, naturally, to protect themselves from imprisonment or murder. Zabel Boyajian's 1901 novel Yestere, an autobiographical account of the 1895 Massacres, describes a scene where the Armenian protagonist, who was jailed for revolutionary activity in Diyarbekir, escaped the city in Kurdish dress right before the massacres, and when he came back during the massacres his life was spared thanks to his disguise.Footnote 56
The cases of disguise that aimed to deceive and instigate violence were treated by historians from two contrasting perspectives. Turkish nationalist accounts like to enumerate such cases as hard evidence of Armenian complicity,Footnote 57 while Armenian-friendly accounts tend to dismiss them as fabrications.Footnote 58 Jelle Verheij has admitted that “This subject of dressing up to fake an identity does occasionally appear in the accounts of the massacre period,” yet he has concluded that these were put into the reports deliberately to “manipulate evidence” and “to impress on us the idea that Armenians were the culprits in all these instances.”Footnote 59 I decline to take such moments as either fabrications or hard evidence; I treat them here as a building block to reconstruct how contemporaries conceptualized collective action. All historians are bound to interrogate their documents, but “critical reading” might on occasion lead them to project their own judgments onto their sources.Footnote 60 What makes sense to us might preclude our noticing alternative mentalities. Thus, putting words into the documents’ mouth in the form of uncovering “motives” or providing “context” can be as detrimental to historical research as taking its every word at face value. Some of the petitions I have drawn upon for this article might have been written under duress, or out of enmity toward Armenians. The defense that ethnographic methodology has typically suggested against such cases has been to sideline singularities and to look for systematic trends in the totality of the event. Here, as I will show, all accounts of the massacres from all perspectives mention moments of deception being used to instigate crowds to act collectively and violently.
It was not only Armenians who dressed up as a way to deceive the other; Turks reportedly did the same. Take the long report penned the afternoon after the massacres occurred by Orson P. Allen (1827–1918), who had come to Harput thirty-eight years earlier and was a founding father of the American mission there. He wrote of the invaders: “How many of them were Koords is doubtful, but probably not many. We have reason to believe that many of them were Redif [reserves] in disguise.”Footnote 61 When the invaders threatened the gorgeous missionary compound of Harput, the army officer in command said he could not protect the Americans against the huge Kurdish invasion (“What could I do against fifteen thousand Koords?”) and proposed to move them to the government building. Herman N. Barnum (1826–1910 in Harput), another leading missionary with by then thirty-six years of service, fully refused the offer on the grounds that, “The raiders were for the most part Turks from Harpoot city and surrounding villages, aided by Redif soldiers, disguised as Koords. There were Koords among the raiders, but they were used as a mere cover to conceal the real authors of this diabolical outrage.”Footnote 62
Missionaries would relate further evidence of posing in the following months. For example, twenty bullet holes made by Martini rifles on the walls of the house of missionary Caleb F. Gates were cited as partial evidence of soldiers’ complicity in the attack. Moreover, Mihran [Ma]dzadoorian, an Armenian from a nearby town, testified that he was hiding in Gates’ house when “a man in Koordish dress with an axe—probably a soldier in disguise—broke it [the safe] open, and … some fifteen soldiers divided the spoil[s].”Footnote 63 In their later reports to Boston headquarters, missionaries in Harput were even more assertive. Allen wrote that “pseudo-Koords” from the city and close-by towns joined the assailants,Footnote 64 whereas Barnum thought that “those ‘Koords’ are imaginary beings.”Footnote 65 The news that, during the Sasun Massacres a year before, in 1894, “some of the troops assumed Koordish dress and helped them in the fight with more success” should have consolidated their belief.Footnote 66 The missionaries were not alone in furnishing evidence of disguise. In the most detailed account of the Harput massacres, written in Armenian, Hagop Parejamyan stated that there were Turkish townsmen in the crowd who turned to killers when disguised as Kurds.Footnote 67
In sum, in the narratives of the massacres, in three languages, we encounter people dressing up to fake an ethnic identity as a viable method to provoke crowds and/or mislead observers. For each example, historians might come up with educated explanations and declare them “fabricated” by Armenians’ and missionaries’ wanting to implicate the Turkish government, Turks’ wishing to put the blame on Armenians, Assyrians’ desire to pledge alliance to the sultan, and so on. However, it requires a good deal of projection to uncover the desires or motives of historical witnesses, and those same desires can always be expressed in many different ways. Instead, I evaluate the recurrence of disguise in the narratives as an index of shared life, in which, for instance, an Armenian could theoretically be mistaken for a Kurdish notable.
Ottoman society was sensitive to attire as one of the most important signs of social position, yet it was an unreliable sign during the age of the money economy simply because the relevant dress could be acquired without having the necessary social standing. In fact, Mahmud II lifted restrictive dress codes in 1829 due to the pervasive practice of imitation—especially of official regalia—and the degraded symbolic value of clothing.Footnote 68 At the turn of the century, with further transregionalization and financialization, faking another ethnic identity by dressing up was a productive way to violate “the sense of knowing who the Other was,” to reiterate Appadurai's words. What other than a turban could differentiate a Muslim scholar from an educated Armenian in a mixed society without physical or even cultural marks to distinguish people? Thus, we need to evaluate the scenes of disguise as they are part of a more ubiquitous form of social interaction: deceiving the other amidst a shared life.
NO VIOLENCE WITHOUT AN IMPOSTER
Disguise had limitations since the relations of power within Ottoman society were reflected in the narratives of disguise. After all, “shared life” did not mean equality. In the massacre accounts, almost everyone dressed up as Kurds, not as Armenians or Turks, simply because it was the Kurds’ collective attacks that needed to be made sense of. Deception, however, taken here as a parent category to disguise, was relatively free from physical and social constraints. There was no account of the 1895 massacres in Harput that did not contain a moment of deception. Nor could any social group avoid being labeled as deceptive. In this section, I will classify the narratives according to those who were implicated in deluding the masses: American missionaries, Armenian revolutionaries, and Turks, respectively.
To begin with, a week after the massacres, sixty-five Armenian notables signed the following petition sent to the Sublime Porte:
Having come to Harput 34 years ago, in their schools seemingly established to serve humanity, American missionaries caused deviation in the ideas and deeds of the Armenian child and a situation was brought about by some outcasts, in which the 600-year sublime protection was cursed with ingratitude. Since their presence in our country will disturb the public wellbeing and security; and since in His Majesty's land every community has its own schools; and since their help is not needed even if new schools are required, we petition, please, that they be removed from here as immediately as possible.Footnote 69
In one of his reports to the central government, the acting governor, Ali Rıza Pasha, referred to this petition and claimed that the American missionaries had long inculcated noxious ideas among Armenians, were consistently exploiting the Armenians’ “light-brained” (ermeni sebükmeğazâtı, read: naïve/silly) characteristics in order to bring about trouble, and had also “poisoned their minds.”Footnote 70 Interestingly enough, the Armenians’ petition comports with classic provocation theory in that it lays the blame for the attacks against them on Armenians. Yet, the real malefactors were the missionaries because they dissuaded and enticed the Armenian youth, particularly through education and indoctrination à la Le Bon.
If some accounts focused on how the missionaries manipulated the Armenian youth/revolutionaries, others stressed how the latter had deceived their own people. It is well known that the official narrative about the massacres, plotted mainly by Field Marshal (müşir) Mehmed Zeki Pasha, criminalized Armenians by taking revolutionary provocation to have been the underlying reason for the attacks. Yet, scholars generally missed the deception pillar embedded in the provocation thesis. Zeki Pasha not only accused Armenians of attacking mosques and government buildings to provoke Muslims, but he also stated that the plotter groups—namely the revolutionaries—had seduced the entire Armenian community with the idea of achieving autonomy, and for that he charged them with inviting foreign intervention.Footnote 71
The missionaries also agreed that the Armenian revolutionary committees deceived their own people. Already in 1891, at the dawn of the revolutionary movement, Barnum despised the “silly talk of a few Armenians about ‘autonomy’”Footnote 72 and condemned them for being as “equally unreasoning nationalists”Footnote 73 as were the Turkish fanatics. He was especially furious since they constituted “a very dangerous & easily excited element throughout the country.”Footnote 74 Right before the massacres in 1895, the tensions between the missionaries and the revolutionary groups escalated. During the graduation ceremony of the Euphrates College, most of the senior students refused to read a piece in Turkish expressing gratitude to the sultan (a customary formality) and, in the presence of Turkish officials, began singing an Armenian national song when the sultan's name was first uttered. Fortunately, Barnum explained, “these rowdies” were made to leave the room without further ado, but that night and the following day they attacked the houses of the community leader and some others.Footnote 75 In contrast with Zeki Pasha's narrative, the missionaries strongly disagreed with the provocation arguments. Barnum believed that “the revolutionary sentiment never gained any foothold” in Harput.Footnote 76 However, they agreed that a handful of Armenian adventurers had deceived their gullible countrymen with seductive promises and thrown them to the lions for the sake of their political ambitions. As Orson Allen's wife had put it following the February 1891 protests in the city of Van, Armenians were “easily excited people”Footnote 77 and the revolutionaries deceived and manipulated them just as easily.
Along with the government and the missionaries, the Armenian notables and middle classes also condemned the revolutionaries for deluding their own people. They separated themselves both from the revolutionary groups and the general public, condemning the former for deceiving the latter. For example, after the massacres, forty prominent Armenians of Peri town in the Çarsancak region sent a letter to the sultan expressing gratitude for the fact that, during the attacks of the tribes, government officials and soldiers had protected the town and the Armenians. Moreover, they stated, the Çarsancak community was completely free of “groundless ideas” of Armenian mischief.Footnote 78
Even though he was not in Harput in 1895 and his memoir was published much later, in 1925, Hagop Bogigian's ideas are a good example of the stance Armenian businessmen took toward the revolutionaries. Bogigian was born in a village of Harput and had immigrated to the United States in 1876, where he became a very successful and prominent tradesman. He was firm in his anti-revolutionary ideas, so much so that in the 1890s he helped the Ottoman ambassador to secretly monitor Armenian revolutionary activities in America.Footnote 79 In his memoir, he mentioned the “revolutionary propaganda,” which “proved to be suicidal to the nation,” and recalled a scene about the massacres that exemplifies the imposter role attributed to the revolutionaries: “I remember watching the bulletin board of a newspaper in Boston; I noticed four of those [revolutionary] leaders who were in front of me, but did not see me. The bulletin was announcing the massacre in Diarbekir [sic]. These men seemed to be overjoyed at the occurrence, saying in Armenian, ‘We caused the massacres in Diarbekir, too.’ Arm in arm they marched into a saloon, to drink to their success.”Footnote 80
Finally, after the American missionaries and the Armenian revolutionaries, the third and the last imposter character appearing in the massacre narratives were the Turks, mostly officials but also ordinary people. Missionaries and revolutionaries agreed on one thing only: that a merciless conspiracy was designed by the Turkish government. We have seen that the missionaries in Harput thought that Turks disguised themselves as Kurds to mislead observers into thinking the Turkish government remained distant from the massacres orchestrated by wild mountain Kurds. Yet physical disguise was just a small part of the greater performance. On the day of the massacre, the missionaries were able to watch the crowds entering the city thanks to the commanding location of the mission campus on the slope. They saw soldiers deploy two cannons on the hill and that the attacking crowd grouped before them. Some prominent men on horses approached the soldiers and then headed to the assailants, possibly seeking to persuade them to retreat. However, the anxious observers were perplexed when they saw the cannon being turned away from the crowd. The consequent march of “the marauding rabble” into the city was followed by gunfire and cannon shots but, Orson Allen wrote, almost no damage was done to the invaders, indicating that they were not really aiming at them. This was “the farce of defending the city” and this “pretended defense of the city was a very thin sham, too thin to cover the deliberate plan….”Footnote 81 The deliberate plan was a plan of deception: “I am credibly informed that the orders for the attack were issued from the house of Mustafa Pasha himself, the general commanding the military force in this region and he of course acted under orders from his superiors. The Koords themselves affirm that they were given license to despoil and kill the Christians.”Footnote 82
As a result, after Monday's events, Allen gave his final word: “After forty years acquaintance I have a [nice] idea of the perfidy, treachery and crue[lty] of which some Turks are capable. [However,] I did not believe that all would condescend to such a depth of perfidy and treachery.”Footnote 83 His bewilderment did not arise from a sudden vanishing of naïve confidence in the government; like every inhabitant of the empire, missionaries were fully aware the state was more a performance than a sincere friend. Were it a matter only of sophistic words, empty promises, unfulfilled assurances, or theatrical shows of power, no one would have been puzzled. States could kill people and lie about it, as they did about other things, but would they deceive their own people? Just a day before the massacres, following repeated assurances from the local government, the missionaries and the Armenians had even surrendered their weapons to quell the tension in the city. Thus, on Monday, while the security forces looked the other way while the assault was happening, all Christians of Harput were astonished and felt they had simply been fooled. This collective feeling of having been deceived was also captured by Johannes Lepsius, a German missionary who traveled in the region in May 1896. In every town he visited he heard people talk about the “deceptive promises” of the officials and “Turkish fabrications.”Footnote 84
Armenians were subjected to the “deceptive promises” of officials not only before the event but while it was happening. The traumatic memory of having been deceived by more or less familiar local authorities was discernable even decades later. In his 1934 book on the town of Arapgir, Sargis Bakhtikian reconstructed the 1895 events from a letter he had received on 15 March 1933 from K. K. Yegavyani, who had been among the Armenian resisters imprisoned in Arapgir. According to Yegavyani, the Armenians of Arapgir were made to stop resisting and resume normal activity by sinister news, circulated by town criers, to the effect that the massacres had stopped and security was guaranteed. As soon as they stepped out of their hiding places, however, 105 people were arrested, some of whom remained jailed in Harput until 1899, including Yegavyani himself. The Armenians, he concluded, had been deceived.Footnote 85
A last example from the year that followed will show that ordinary Turkish friends might also dupe Armenians. According to official documents, Kasap [butcher] Manug from Eğin, in Harput province, was the leader of the Armenian guerillas in the city. He and his 340 men rebelled against army forces in September of 1896, a year after the widespread massacres, and on 21 September the commander-in-chief announced that he had finally been killed during armed conflict.Footnote 86 However, Droshak, the official organ of the Armenian Revolutionary Federation, published a different story, reporting that he was working with a fellow Turkish artisan in their friendly workshop. One day, this Turkish fellow warned Manug that nearby Kurds would soon attack the city in order to massacre Armenians, and he suggested that Manug should take his friends and relatives and hide in the mountains outside the city. Since Manug was a “simple-minded and credulous” (barzamid yev türahavad) person, he complied with the suggestion. But it turned out that the local government had created this plan, and when the group hid in the mountains the governor declared them a revolutionary militia and easily got permission to pursue them as a dangerous mob. This was the signal the Turks had been waiting for, and the massacre began in Eğin on 3 September 1896.Footnote 87
CONCLUSION
Recent years have seen an increasing scholarly interest in the politics of lies and deception due to the unprecedented level of dishonesty in populist-authoritarian regimes like those of Donald Trump, Vladimir Putin, and R. Tayyip Erdoğan. The relentless devaluation of objective truth in the face of emotional convictions indicates that we may be living in an age of “post-truth,” Oxford Dictionaries’ word of the year for 2016. Intellectuals and ordinary people, mainly using social media, have been fighting a fierce battle against the invasion of untruths by tirelessly refuting inventions, fabrications, and simple lies. However, in the battle over facts we should not miss the opportunity to decode the theoretical integrity and public relevance of post-truth thinking. As Carole McGranahan succinctly puts it, “An anthropology of lies must ask not how to correct them but how to understand lies and liars in cultural, historical, and political context so that we see clearly the work of lies, including their meaningful and sometimes violent consequences.”Footnote 88
The 1894–1896 massacres took place in another populist-authoritarian regime, that of Abdülhamid II. My research on the massacres in Harput in November 1895 reveals that all observers who left a written narrative of the event—mostly in the form of private correspondence not intended for a wider audience—explained the violent crowd actions as a result of sinister deception. In this historical ethnography I have analyzed these deception narratives to understand “their meaningful and sometimes violent consequences,” rather than simply trying to refute or correct them. The most important goal of this research has been to reconstruct the thought system that my historical informants spoke from when they talked about the events in Harput on Monday, 11 November 1895. What Jonathan Mair has proposed for post-truth studies is also valid for historical ethnographies of deception: “Post-truth anthropology will only make its distinctive contribution if anthropologists are able to … study the phenomenon ethnographically. That means paying attention to people's theories of truth, evidence, belief, and ignorance—taking thought about thought seriously.”Footnote 89
My analysis of the logic of deception in the massacre narratives has showed that contemporary observers conceived collective violent action only as a result of sinister deception by trusted people. The deceivers used different techniques, such as dressing up to fake ethnic identities, using gentle language to convincing listeners to acquiesce, or implanting mischievous thoughts through education. The crowds were especially susceptible to the words of powerholders like religious leaders and government officials. We find here a theory of crowd behavior—in fact, almost an exact replica of Gustave Le Bon's ideas—that is deduced from practices and narratives of lay people. Thus, we need to flatten the barriers between intellectual history and social history on behalf of an historical ethnography of people's theories about society. After all, any talk about social action, regardless of who does the talking, contains a theory of social action, apparent or tacit.
My findings also necessitate a critical intervention in scholarly discussions of the “provocation theses.” Much as the factual grounds and political intentions of the provocation thesis have been debunked or corrected, the scholarship so far has failed to take the thesis seriously as a conceptual framework. The provocation thesis, among other things, assumes that crowds can be put into action by open provocation. As Margaret Anderson has noted, “Implicit in the provocation thesis is the premise that the society concerned is able to be provoked.”Footnote 90 Thus, the thesis presupposes that masses act like automatons and have no real agency, and behave like a herd when provoked. Critical studies of the case investigated here have reproduced this primitive image of the crowd because discussion has been confined to the question of whether or not there was Armenian provocation. Yet the deception narratives show that people at the time did not consider provocation a working tool. In their view, the masses had to be deceived into taking violent action. Not only was the provocation thesis factually wrong and politically incorrect; it did not exist within the conceptual frameworks of the historical actors.
Finally, I want to revisit my emphasis throughout this article on the “shared life.” Scholars talking about collective violence almost instinctively refer to deep societal divisions. The provocation thesis reinforces this tendency because it presupposes two separated parties (provoker and provoked). Other perspectives, like those that highlight empathy, also rely on presumed absolute others. Deception relies on an opposite situation—to function, it needs a shared social environment, an intimate familiarity among parties. A focus on deception also reminds us that people killing one another is not normal. All accounts from Harput agreed that it took a great deal of manipulation, deception, and deluding to compel one social group to attack and kill others. Processes grounded in provocation or empathy reinforce existing divisions, whereas deception aims to change the other's mind; it is a technique for conducting conduct. Deception is a way to govern the behavior of proximate others.
