Many years ago, when I lived in pre-revolutionary Tehran as an expat teenager, American Forces Radio and Television Service (AFRTS) was the only television channel I could watch and understand. There was a recurring advertisement, which targeted its actual audience to encourage them to move to one or another state in the United States, once their service time had come to an end. Of course, the states advertised were not New York or California, but rather more obscure and admittedly less attractive destinations. The motto was always the same: “Oklahoma – or Alabama, or New Mexico, or Oregon – a nice place to visit, a great place to live.” This is what inspired the title of this epilogue, originally delivered as a keynote address in Amsterdam.Footnote ¹

Obviously, the context is quite different, but for whatever reason this phrase had somehow stuck in my mind, and it immediately popped up when I started to write this piece. Yet my take was at the reverse of the logic behind the American message of the 1970s. Instead of a crescendo from a nice place to visit to a great place to live, my argument was based on the contrast and opposition between the Ottoman lands and Turkey as an attractive destination for visitors and a difficult place to live for its own subjects or citizens.

I believe this contrast sums up one of the dominant contradictions of demographic and human flows between these lands and the rest of the world. A pole of attraction for so many visitors and temporary residents, it remains a place where a prolonged, let alone permanent, settlement is perceived with anxiety and often triggers a feeling of being trapped. Interestingly, the present and very recent past have witnessed a very considerable rise in phenomena that one is likely to link to one form or another of exile. Intuitively, I believe that the proportion of Ottoman subjects and Turkish citizens who may have experienced exile is high. One could musingly adapt the Koranic saying that everyone will taste death and claim that every Ottoman/Turk has tasted or will taste exile. However, I am not one of them. True, my grandmother was exiled for almost thirty years, and my mother was born in exile and remained so for almost fifteen years. Nevertheless, I was spared any feeling that comes even close to the notion of exile, however loosely we may use it. In fact, I am part of a happy minority of people who are mobile and fluid enough to arrogantly brag about the joys of feeling a stranger everywhere, even at home.

Polls in present-day Turkey reveal that a very significant proportion of the young generation dreams of leaving the country, targeting the “West” in its broadest sense. One such poll, conducted by Yeditepe University in 2020 among youths between the ages of eighteen and twenty-nine years, revealed that 76 percent of the sample felt ready to go abroad if given a chance to do so. Europe came on top of their wish list, with 43 percent, followed by the United States (40 percent) and Scandinavia (15 percent) (Radikal 2020); a report published by Konrad-Adenauer-Stiftung (Çağlar and Çağlar Reference Çağlar and Çağlar2022) yielded very similar results for 2021. This is a phenomenon the likes of which has not been seen in decades. Even the flight from oppression and incarceration of leftist and Kurdish activists in the 1970s and after the 1980 military coup could not come close to such a massive potential of human loss, if it were to be realized. It may be likened somewhat to the massive wave of Turks leaving their country to become Gastarbeiters (guest workers) first in Germany and then in other Western European countries, which started exactly sixty years ago, but with a major difference: this labor-based economic exile was predominantly defined by a clear “pull factor,” while today’s intent seems to be motivated by a “push factor,” which cannot be reduced to a precise political or economic nature. A general feeling of hopelessness, a belief that they have no future in their country, seems to be the broadest common denominator behind these intentions, across a very wide political and socio-economic spectrum. “Generational” is probably the best label that one could come up with to characterize the dominant feature of this trend. In a sense, this is what makes it all the more terrifying by its implication of desperation that jeopardizes the very future of the country.

This self-exile in the making is not the ruling Justice and Development Party’s (Adalet ve Kalkınma Partisi; AKP) and President Recep Tayyip Erdoğan’s only contribution to Turkey’s reconnecting with long-forgotten notions of exile and banishment. The attempted and failed coup of July 2016 by their former Gülenist allies has triggered a backlash of unprecedented proportions which targeted tens, if not hundreds of thousands of citizens suspected of collusion and connections with, or simply sympathy for, the movement. Thousands were arrested, incarcerated, or simply deprived of their jobs and means of livelihood through a series of arbitrary measures, thus sending those who could escape into exile, much like their alleged leader, Fethullah Gülen, who has been living in self-imposed exile and seclusion in the United States since 1999. The rise of a personalized and authoritarian regime in the past decade has also contributed to the emergence of a new type of political exile, that of the “disappearance” of bureaucrats and politicians discarded by the system. One striking example is that of Erdoğan’s own son-in-law, Berat Albayrak, who, after a disastrous term at the helm of the economy, simply vanished into thin air, a bit like dismissed Soviet officials in the past. We have no reason to worry about his fate; he just chose – probably following strong familial advice – to withdraw from the political scene, perhaps in the hope of making a comeback if circumstances were to change in his favor. A much more pathetic example is that of Melih Bulu, the apparatchik who was appointed leader of the attack on Boğaziçi University in early 2021. Rejected by the university he was supposed to conquer, and accused of plagiarism, he turned out to be a liability and was simply discarded six months later, only to be replaced by an insider who was willing to carry out the government’s plan of dismantlement and conquest. Nobody knows what happened to the unfortunate Bulu, who simply vanished, but with much less panache than Albayrak; rumor has it that he is teaching in a Macedonian university, living a double exile outside the country and in total oblivion.

Interestingly, Turkey today is not just losing, or risking to lose, what is probably the most precious component of its population. It has also become a land of refuge for millions of individuals fleeing war and instability in the region. The Syrian refugees constitute the most conspicuous layer of this incoming wave of exiles, recently topped by a flow of men – the stress on gender is meaningful – from Afghanistan, following the Taliban takeover. It has been a very long time since this part of the globe has become a haven for a sizeable population fleeing war and oppression. I am excluding the waves of immigration caused by the dismantling of the Ottoman Empire or its aftershocks, where the incoming populations were in one way or another viewed as being religiously, ethnically, or culturally close to the core population. This was the case with the Muslim populations fleeing the lands lost in the Balkans after 1878 and Crete at the turn of the century (see, in this issue, Peçe Reference Peçe2024), of those exchanged against Anatolian Greeks in 1924, or of the “Turks” who were allowed to leave Yugoslavia and Bulgaria during the Cold War. The Syrian refugees are somewhere in the middle, not as distant as the Polish or Hungarians who fled the repressive backlash of 1849, but certainly not as close as the Balkan Turks driven from their lands by the Russian advance in the wake of the crushing defeat at the end of the 1877–1878 Russo-Ottoman war. Perhaps a reasonable comparison would be with the massive exodus of tribes from the Caucasus, fleeing Russian expansion after the Crimean War (1853–1856).

Syrian immigration has revealed all the ambiguities inherent in Turkish politics and culture with respect to incoming aliens. On the one hand, one can easily claim that Turkey has fared better than Europe – particularly some European Union (EU) member countries – in accepting huge numbers of refugees, more than half of the entire displaced population from Syria. The conditions for their settlement were far from ideal, but the natural porosity and inefficiency of the system of control and settlement soon compensated for that by enabling a great majority to reach the major urban centers, thus starting a process of integration, albeit of a precarious nature. True, behind this humanitarian action lie a number of rather cynical calculations, starting with the EU’s desire to use Turkey as a buffer to manage the risk of direct immigration. Turkey, on its side, has shown great enthusiasm for the financial scheme put forward by the EU as a compensation for this role, but the government has also had no scruples about using this matter for political leverage, especially by threatening to push – and occasionally actually push – refugees into the EU. The darker side of immigration could also be witnessed in society, where Syrians were often exploited as underpaid laborers by businesses that took advantage of their vulnerable status. More importantly, however, the severe economic recession of the past few years has triggered numerous acts and reactions of racism and exclusion across the political spectrum, revealing to what extent socio-economic discrimination and xenophobia run deep in Turkish society and culture, providing a disturbing sense of cohesion in an otherwise highly polarized environment.

If we are to talk about exile to or from the Empire and Turkey in a historical context, it seems that we are faced with two dominant and opposite trends, based on the most traditional identitarian category one can think of, religion: Muslims in, non-Muslims out. The steady inflow of Muslims parallels the erosion of the Empire throughout the long nineteenth century. From the Morean refugees in the 1820s to the Thracian Muslims a century later, hundreds of thousands of Muslims took refuge from the lost provinces of the Empire, mostly in Anatolia, with particularly dramatic peaks after the defeats of 1878 and 1913. To these one should add those populations, which, although not formally subjected to Ottoman rule, chose to migrate to the Ottoman lands due to the annexation of their lands by rival powers, especially Russia, such as the Crimean Tatars in the 1780s or the aforementioned tribes of the Caucasus in the 1860s. The Republic of Turkey inherited a much smaller-scale version of this phenomenon when some remaining groups of Balkan Muslims chose to migrate from Communist-ruled Albania, Yugoslavia, or Bulgaria to Turkey.

The chronology of the migration of non-Muslim populations from the Empire overlaps to a large extent with the influx of Muslims I have just described. Of course, one could always go back to the earliest centuries of the Empire and talk, for example, of the Greek exodus from Constantinople after its fall, which Republican historiography, pointing at scholars taking refuge in Italy, went so far as presenting as the Turkish contribution to the Renaissance. The existence, throughout the early modern period, of Ottoman Greek communities in Rome, Venice, Vienna, Marseille, or Amsterdam attest to a flow, or rather a trickling, of Orthodox subjects from the Empire to certain parts of Europe (Grenet Reference Grenet2012; Çolak Reference Çolak2018). The same could be said of Armenians, whose trading networks extending from India to Europe were particularly conducive to migration, as was the oppression faced by the Armenian Catholics who took refuge in Venice and Vienna in the eighteenth century. Nevertheless, it was in the nineteenth century, particularly in its second half, that the movement took on truly massive proportions. The repression of the Greek insurrection resulted in minor flights, such as those of some families from Chios, and similar examples can be found during the following decades. However, it is really during the Hamidian era, in the 1890s and 1900s, that massive movements were witnessed among two major groups, Armenians and Christian Arabs, fleeing disastrous economic conditions, and, in the case of the former, outright persecution and massacres (Baycar Reference Baycar2015, Reference Baycar2020; Gutman Reference Gutman2019). As in the case of Muslim immigration, the Republic inherited to a large extent this pattern of emigration, targeting non-Muslim populations, which did not fit in the state’s and the majority’s vision of the Turkish nation. Together with Greece, it organized the first population exchange, which set a precedent for many tragedies that were to follow throughout the world. The pogrom of 1955, followed by expulsions and tensions over Cyprus have reduced the remaining Greek population in the country to a mere 1,500 to 2,000 (Anastassiadou and Dumont Reference Anastassiadou and Dumont2003; Wikipedia 2024). Jews, targeted by a wealth tax in 1942 left en masse after the creation of the state of Israel in 1948; the community stands today at about 15,000 (Levine Reference Levine2022; World Jewish Congress 2024). A century after its demise, practically nothing of the Empire’s religious diversity remains in Turkey today, in blatant contrast to the material remains, often in a sad state of disrepair, of the presence of once thriving communities. This is all the more ironic, when one considers that the country which prides itself with being the only secular state in the Middle East is at the same time the most homogeneously Muslim country of the region.

Of course, one will always find exceptions to the rule, starting with the settlement of Iberian and later Italian Jews in the Ottoman lands after the Spanish Reconquista and the fall of Granada in 1492, or the settlement of European Jews in Palestine at the turn of the twentieth century. One could also mention Charles XII of Sweden, who sought refuge in Ottoman lands after being defeated by the Russians in 1709, or more famously the influx of post-1848 refugees from the Habsburg and Russian lands, provided we remember that most of these exiles never intended to settle but had sought asylum until they could reach a safe haven in Western Europe, like the famous example of Lajos Kossuth (1802–1894) and his companions. There are all those renegades, converts, engineers, officers, artists, scholars – like the German Jewish scientists in the early 1930s – who, for one reason or another, ended up in the Ottoman lands or Turkey for relatively long periods of time, sometimes for the rest of their lives.

In the other direction, Muslims leaving or fleeing their land to settle abroad may be rare, but they exist. Mehmed II’s son Cem (1459–1495), half exile, half hostage, throughout Europe and the Mediterranean, is an obvious early example that comes to mind. Much later, Young Ottomans and Young Turks, from Namık Kemal (1840–1888) to Prince Sabahaddin (1879–1948) and from Ali Suavi (1839–1878) to Ahmed Rıza (1858–1930), constitute a small but consistent sample of political exiles who broke the dominant pattern to seek refuge in Western capitals. And then, of course, there were the 145 statesmen exiled to Malta in 1919–1920, and the 155 members of the Ottoman dynasty and 150 collaborators of the ancien régime who were exiled in 1924. Yet on both sides, these cases are consistent with what I wrote earlier, namely that they constitute exceptions to the general rule of Muslims in, non-Muslims out. It is only in the last decades that this rule may have been slightly altered, not by a sudden rush of non-Muslims to Turkey, but by significant departures of Turks from their homeland, especially as economic migrants.

There still is a major dimension of the question that is missing, that of internal exile. In fact, I would even claim that this is probably its most dramatic expression, and one that is almost inherent to the Empire and, to a certain extent, to Turkey. Indeed, right from the very start, and much like so many other contemporary polities, exile, deportation, banishment, and relocation have been at the core of the Ottoman state’s policy regarding subject populations. We know how deportations, known as sürgün, were used to drain certain regions of unwanted elements, but most of all to repopulate others, as in the famous case of Constantinople after the conquest. We also know that recruitment for the army and bureaucracy relied greatly on the extraction of children from conquered populations, through a system that was called devşirme, or “collection.” The slave trade fed an incessant flow of men and women who were uprooted from their homelands in Africa or in the Caucasus to end in the imperial palace and in the mansions and homes of individuals who could afford this sinister commodity. Banishment, perhaps the form of exile that comes closest to the original Roman notion of exsilium, was practiced on a regular basis as a punishment for individuals who were accused of wrongdoing but whose crime fell short of deserving execution, or simply as a form of retirement for certain high-ranking officials. The latter measure was particularly used in the case of palace eunuchs who were sent to Mecca and Medina at the end of their career, or of the women of a former sultan, who were sent out from the new seraglio to the old one, to end their lives in an exile of a mere two kilometers that must have felt like thousands of leagues away.

The list could be extended, even to include the “Great Flight” (Büyük Kaçgun), characterizing one particular aspect of the seventeenth-century crisis consisting of the flight of peasants from villages to inhospitable and inaccessible regions to escape a number of threats, from banditry to tax collectors. At any rate, it is clear that displacement, at an individual or communal level, seems to have been an endemic feature of the Empire well into modernity. One could point to the reign of Abdülhamid II, from 1876 to 1909, as a peculiarly fertile period for political exile, with such striking examples ranging from Midhat Pasha (1822–1884), whose second banishment in 1881 was a commuted death sentence, which ended up being carried out unofficially in the prison of Taif, to scores of political opponents who were sent out to the islands or to other remote provinces of the Empire, sometimes with a local bureaucratic post, as in the case of Namık Kemal in Cyprus and the Archipelago, of Tunalı Hilmi (1871–1928) in Madrid, or of the Young Turks who ended up in the Libyan desert, nicknamed Ottoman Siberia. After all, following his deposition in 1909, Abdülhamid himself was exiled to Thessaloniki, and if he returned to İstanbul when the city fell to Greek forces in 1913, it was only to be relegated to the palace of Beylerbeyi. Yet, if there is one event that tragically resonates with some of the practices of the earlier centuries of the Empire, it is certainly the deportation of the Armenian population during the Great War. Indeed, we should not forget that what is known as the Armenian genocide from the perspective of its underlying intent and final results was initially planned and implemented as a deportation, whose unprecedented violence found its roots in the deadly combination of an ancient practice with a modern ideology that viewed the very existence of this community, even deported, as a threat to the survival of the state and its newly discovered “dominant/ruling nation” (millet-i hakime).

Before the enormous diversity of the phenomena and events I have referred to in this long historical panorama, one feels compelled to ask the question of whether they can really all be lumped under one broad category of exile. If we are to stick to a literal, “orthodox,” understanding of the word, the answer is obviously, no. After all, a true exile, ostracism for the Greeks, is a punishment, a capital one for the Romans, decreed by the authority or voted by the community, whereby a member is banished from his/her land. Yet, obviously our present understanding of exile, while inclusive of the initial definition, is much broader and flexible. Our focus is not so much on the factor conducive to estrangement, as it is on the fact that one has had to leave one’s home/land and is left in the situation of living in a foreign land or environment. In fact, even the notion of a forceful departure, of an obligation to leave, is sometimes tempered by the inclusion of self-exile as a choice, often triggered by the estimation of the cost and risks of staying, but at times even independently of such “push” factors. In short then, to use a cliché, we might be justified in saying that exile is in the eyes of the beholder, or rather in the feelings and thoughts of the exiled.

One needs only to conduct a lexicographic search of synonyms to get a sense of the wide range of meanings and categories associated with the notion of exile. This is true of English, of course, but I shall limit myself to Turkish for purposes of consistency with our topic. Only one of these numerous words is really Turkish: sürgün, the one that has best survived in the language today as a synonym for exile. Derived from the verb sürmek, to drive, with a connotation of herding, it reflects best the sense of a forced displacement, and was used to describe the deportations of the early period. The rest of the vocabulary is of Arabic origin, and some words, such as istifaz, sebr, or icla, are oddities that one rarely sees outside of dictionaries. This leaves us with two major lexical groups, one derived from nefy (to banish), and the other from hicr (to leave). An essential difference between the two is obvious. Nefy and its derivatives – menfi (exiled person), menfa (place of exile) – denote the punitive nature of the act and was commonly used in the nineteenth century to describe state-ordered political exile. Hicr, on the contrary, as suggested by the hijra (journey) of the Prophet Muhammad in 622, focuses on the displacement itself, with a very strong connotation of migration. Just a cursory glance at the Ottoman archives reveals how its main derivative, muhacir (migrant) emerged. First used in a neutral sense covering any migrant, it started taking up a new and collective meaning in the eighteenth century with a wave of migrants from Shirvan, in Azerbaijan. By the end of the century, it acquired the sense of refugee with the Crimean muhacir, followed by a deepened sense of despair once the Greek rebellion created a flow of refugees from the Morea and other lost territories. By the mid-nineteenth century, it was the turn of the Caucasian tribes to acquire this status; a peak was reached after the Russo-Ottoman War of 1877–1878, so much so that the geographic reference that had until then accompanied every occurrence of the word was dropped, leaving the term muhacir alone to describe the hundreds of thousands of refugees who flocked the capital. Ironically, the same Arabic root was at the origin of another word that is today associated with the massive displacement of the Armenians after 1915, tehcir, literally meaning “making leave,” but used with a much stronger connotation of deportation.

I would like to mention one last term, the noun and adjective of garib, stranger, remarkable for its emotional connotation, which makes it interchangeable with poor and forsaken, and for its much more common substantive form, gurbet, best translated as homesickness, to the point that the popular Turkish term used for Gastarbeiter is gurbetçi, “the one practicing/living homesickness.” The reason I have added this last term is to introduce the sentimental and pathetic dimension of the matter, which is central to the very notion of separation so deeply embedded in our understanding of exile. While it is undeniable that there are happy exiles, and that the logic behind many of the forms of migration rests on the desire to leave misery and suffering behind and move into a better world, the idea of loss is practically inseparable from any account of exile, be it at a purely rhetorical level. This in turn brings me to what may indeed be the most challenging aspect of the central subject that frames this special issue: narrating exile.

At one level, nothing could be simpler or more obvious. For the reasons I have just listed, exile has always been one of the most attractive, not to say popular, themes in literature and the arts. From Ovid and Cicero to the German Exilliteratur, poets and authors of all kinds have found in exile a source of inspiration, if not the urge to reveal their deepest feelings in the face of estrangement and loneliness. It would be an exaggeration to say that the Ottomans were not different, given the difference of scale in their literary production. Nevertheless, a number of texts can be shown to belong to a literature of exile: the Vakıat (Adventures) of Cem Sultan during his exile in Europe in the fifteenth century (Vatin Reference Vatin1997); the writings of the seventeenth-century Sufi Niyazi-i Misri (1618–1694) during his exile on the island of Lemnos (Terzioğlu Reference Terzioğlu2002); and İzzet Molla’s (1786–1829) Mihnetkeşan (Those who suffer), written during his exile in Keşan in the 1820s (Keçecizade Reference Keçecizade1269/1852). Yet it is with modernity that the phenomenon really took off. Under Abdülaziz and Abdülhamid, two generations of “Young Turks” went through some form of exile, and left memoirs and accounts of their experience: Namık Kemal, Ahmed Midhat (1844–1912), Ebüzziya Tevfik (1849–1913), İsmail Safa (1867–1901), and Hüseyin Siret (1872–1959). The Young Turk Revolution of 1908 unleashed an avalanche of such memoirs, betraying an open competition for recognition by the new regime. They were followed by those who were sent into exile by the Young Turks, then those who returned from Malta, and, finally, those who were banished by the Kemalist regime. Some were unlucky enough to experience exile under two consecutive regimes: Refik Halid Karay (1888–1965) was exiled to Anatolia by the Young Turks in 1913 and returned when they were expelled in 1918, but then paid the price of opposing the Kemalist movement by exiling himself to Syria and Lebanon, where he stayed until 1938; each of these was the occasion to write a series of short stories focusing on his places of exile (Karay Reference Karay1335/1919, Reference Karay1940; Philliou Reference Philliou2021).

There are, however, two sides to every coin. While there is no doubt that these narratives of exile provide us with an often-fascinating insight into the lives and minds of these victims of exile, we should not forget that they also present a consistently skewed profile: almost all of these authors are male, politically active and motivated, inevitably self-conscious about what they chose to write – or not – in view of publication. Given the very loaded context of political exile – which was the case for the overwhelming majority of such authors – it is clear that several forms of self-fashioning and self-censorship were likely to influence their narrative, starting with an inevitable temptation to exaggerate their political role or their sufferings, or to suppress or marginalize aspects of their experience that they did not deem worthy of, or compatible with, the image they might have wanted to give of themselves. True, there are circumstances and formats that help reduce this drawback, starting with unpublished sources, such as diaries or correspondence. Unfortunately, diaries are rare, but some of these men are known to have kept a sustained correspondence with their friends and family, part of which has survived to this day. One needs only to consult Fevziye Abdullah Tansel’s (Reference Tansel1967) remarkable work on Namık Kemal’s correspondence to get an idea of how informative such a documentation can be when treating such a subject.

Nevertheless, however detailed the documentation, however intimate the writings, these authors still remain men, and in the overwhelming majority of cases, men of a certain standing, often self-important, endowed with the educational and intellectual skills needed to engage in a self-consciously modern act of narration, or self-narration. The biases caused by gender, socio-economic standing, and political power are simply immense. What do we know about women, peasants, soldiers, slaves, landless refugees, convicts – in short, about the mass of ordinary individuals who made up the vast majority of people who were displaced against their will or out of fear and desperation? How many firsthand personal accounts do we have and can we hope to find of Muslim refugees flooding the capital after the debacles of 1878 or 1913? How many women who were dragged into exile as wives, daughters, or as slaves were able to tell their story in a way that might have left a trace for posterity?

It is undeniable that there is a strong correlation between the socio-economic standing of individuals and their likelihood or ability to narrate their life story, for better or for worse. Can one not widen this observation to cover even the orally transmitted memory of past events? One immediately thinks of the extraordinary archive of oral testimonies and recollections of Greeks displaced from Anatolia constituted by Melpo and Octave Merlier at the Centre for Asia Minor Studies (Kéntro Mikrasiatikón Spoudón) in Athens, which Artemis Papatheodorou puts to a very specific use in this issue (Papatheodorou Reference Papatheodorou2024).

However, it has always struck me to what extent memories of a lost homeland acquire different levels of detail and precision between the descendants of Greeks who left their home in Anatolia to settle in Greece, and those of Turks, or rather Muslims, who were forced to make the opposite journey from the Balkans to Anatolia? The former generally cherish the memory of their ancestral home down to the village, the street, and the house where their grand- or great grandparents lived. The latter, as far as I have seen, can rarely go beyond the name of a province or a nearby city, or even the vaguest reference to Rumelia. I think it is significant that the exceptions on the Turkish side generally have to do with those areas, such as Crete or Ioannina, where the Muslim population enjoyed a wealthier status, generally due to landownership.

If we apply all these observations to the question of exile and its narratives, it becomes clear that it is practically impossible to retrieve the “voices” of most of the individuals concerned, unless they belong to a privileged minority possessing the material and intellectual means and motivation to leave a written trace of their ordeal, or if their plight ends up finding a strong moral and political mediation, as may have been the case with the survivors of the Armenian genocide or the Greek refugees from Anatolia, in ways that were never made available to Muslim refugees in 1878, 1913, or 1924.

Of course, one cannot dismiss all hopes of finding alternative sources or alternative ways of filling the gaps in our documentation. After all, feelings of exile end up popping up in all sorts of sources, especially anonymous and popular ones. Everyone in Turkey knows the song of Yemen (Yemen türküsü), a lament that speaks of the sorrow and sufferings of a conscript posted in Yemen. If anything, this is a meaningful reminder that compulsory military service, when it meant such distances and such long periods, amounted pretty much to an exile from which one lost hope of ever returning: “Those who leave never return, I wonder why” (Giden gelmiyor, acep ne iştir) said the famous song.

Let me conclude by referring to somewhat different and unexpected sources of exile from my ongoing research (Eldem Reference Eldem2019, Reference Eldem2021) on Selahaddin Efendi’s memoirs and diaries. The son of Sultan Murad V, who was deposed on grounds of insanity in 1876 and replaced on the throne by Abdülhamid, Selahaddin Efendi (1861–1915) bore witness to his and his father’s tragic fate. Indeed, the deposed sultan and his entire household – about 100 persons, almost all women – were kept under custody in the palaces of Çırağan and Feriye for almost thirty years, until Murad’s death in 1904. This captivity amounted to an exile and a form of social death, which apparently triggered in young Selahaddin – he was sixteen years old at the beginning of their captivity – the urge to do something extraordinary by Ottoman standards, that is, to keep a diary, to write his memoirs, and to collect personal and family mementos in scrap books, thus creating a set of ego-documents of unprecedented and unequalled wealth and originality.

Understandably, the young prince’s writings are strongly imbued with feelings of frustration and despair. If it does not come to his mind to call this an exile, it is probably because the theme of being subjected to injustice (mazlumiyet) conveyed much more powerfully the image he wanted to give of their treatment at the hands of a tyrant (zalim), and also because “true” exile was already a subpart of their ordeal, when many of their slaves and servants were actually banished immediately after the deposition. On at least two occasions, the sight of the imperial yachts İzzeddin and Fuad in the Bosporus made Selahaddin comment that “there was an exile in the making,” a reminder of the frequent banishments that were carried out to distant provinces of the Empire (Selahaddin Efendi 1901–1904).

Yet the truly striking discovery I made while studying these documents was the way they revealed some aspects of one of the least known forms of uprooting and exile still in practice at the time, that of female slavery (for a discussion on representations of female slavery in late Ottoman literature, see, in this issue, Gürsel Reference Gürsel2024). With the exception of himself and his son, Selahaddin, the fallen sultan’s household consisted almost exclusively of women, and of half a dozen eunuchs. All these women, apart from Murad’s and Selahaddin’s daughters, were of slave origin; some were of high standing, such as Murad’s mother, as well as his and Selahaddin’s wives, while the others constituted the workforce of the palace. Selahaddin’s writings were predictably highly focused on himself and on his father, but they inevitably contained a mass of information on some of these women. Much of this is difficult to use, or even to comprehend, given the degree of intimate knowledge that would be necessary to penetrate the logic of such a complex and hierarchized structure. Nevertheless, some writings are clear enough to allow for a straightforward analysis of their contents. I have selected two passages, both concerning Selahaddin Efendi’s wives, and which, to my opinion shed light on the harsh realities of the life and career of a slave woman in an Ottoman palace. In one of these, he gives a short life story of Dilaviz Hanım, his first consort and mother of his first child, who had just died of tuberculosis:

Another passage consisted of the transcription of a letter he had sent to the parents of another of his wives, Zatıgül, apparently to try to reassure them as to her wellbeing:

I believe that both these texts give a fleeting, but poignant, sense of the tragedies that lie behind each of these fragmented and scattered lives. Handed over from one household to another, following their masters throughout the Empire, cut off from any form of contact with their family, their livelihood and fate depended on the will of men who could not be bothered with the details of their past life and of their true identity. Perhaps the most striking document I have ever encountered in this respect, which I dug out from the Topkapı Palace archives, is a letter written by Murad’s first wife to his mother – from a slave wife to a slave mother – about her name. To grasp the meaning of this short letter, one needs to know that when a slave entered the harem, she was given a new name, generally of poetic resonance, to replace the proper Muslim name she had been given at birth. The letter was about this woman having managed to convince her master, former sultan Murad, to abandon her slave name, Elaru, and to revert to her true name, Mevhibe:

I believe that no words could better describe the frustration that this common practice caused to scores of enslaved women who were suddenly deprived of their identity, of their last remaining tie to their childhood and family, and found themselves thrown into an exile of a kind none of the proud political victims of banishment could ever imagine.