- 1 The author would like to thank the two anonymous reviewers for their constructive criticism and the (...)
1The ‘Fifth International Congress of Onomastic Sciences’, dedicated to the study of proper names, gathered in the Castillan city of Salamanca in 1955. A year later, the young historian Halil Inalcık, who had represented the Turkish Historical Society at this congress, published a report in the Society’s Review [Türk Tarih Kurumu Belleten], in which he recommended to Turkish historians the discipline of onomastics as a highly relevant and hitherto neglected field of historical inquiry. Having defined the field of onomastics as the science dealing with toponymy, ‘the archaeology of names’ and anthroponymy ‘the social psychology of personal names’, he outlined the historiographic importance of the field: ‘Place and personal names are living documents, which preserve the ethnic, social and linguistic influences of the ‘longue durée’ (Inalcık 1956: 223). Interestingly, Inalcık’s article appears to be oblivious to the fact that it was published in a country, whose state agencies had already spent considerable effort to change both its toponymy and the personal names of its citizens, and by an institution that had chosen to support this policy with benign neglect. What is more, the little of this ‘archaeology’ that had survived the avid gaze of zealous bureaucrats, was about to be destroyed just a year later with the launch of the ‘Expert Commission for Name Change’, which would, once and for all, do away with the diverse toponymical heritage of the Ottoman lands.
2This paper aims at reconstructing the re-naming of the entirety of a country’s geography and topography as the state-directed policy of ‘toponymical engineering’, discussed here as a policy in close causal relationship with ‘demographic’ and ‘social engineering’. It does so in three steps. First, it refers to two interrelated debates: the current shift from national to post-national historiographies in South East Europe and Turkey that has re-oriented the focus of inquiry from the Kemalist reforms towards the often traumatic workings of the early Republic and the related conceptual opening that has introduced demographic engineering and ethnic cleansing as novel analytical tools. In the second part, it examines the empirical case of toponymical engineering between 1915 and the 1990s by differentiating four waves of administrative measures geared towards the Turkification of place names and by highlighting the semi-autonomous role of bureaucratic elites. Often to the dismay of elected representatives and more forcefully during periods of military intervention and caretaker governments, bureaucrats would advance the toponymic agenda. In the conclusion, the paper addresses the question whether demographic engineering is indeed a meaningful analytical tool for the debate of the issues considered in this volume, a question that will be answered with a qualified yes.
3The empirical evidence of this paper rests on documents seen in the Republican Archives in Ankara [Başbakanlık Cumhuriyet Arşivi] and especially on documents available in the Catalogues of the General Directorate of the Prime Ministry’s Transactions, the Decisions of the Council of Ministers and the Catalogue of the Republican People’s Party. A second and indispensable source consists of the series of place directories, maps and circulars published by the Interior Ministry since 1928.
4Critical studies of nations and nationalism in general and South East Europe in particular have often focused on the social construction of nations, the delimitation of their territories and the emergence of clearly delimited nation-states. More often than not, however, these studies have failed to expose the ‘dark side of nationalism’ and the destructive antecedents of the modern nation-state. This could be deemed surprising insofar as, for the region under study, the historical sequence and the mutual constitution of the two trajectories, i.e. genocide, ethnic cleansing or population exchange as the antecedent to the social construction of nations and nation-states are quite evident. In South East Europe, within which a discussion of Turkey’s nation-state project appears most appropriate, however, the ontological relationship between the two has been obscured by a couple of interrelated factors: National historiographies, until recently dominated by orthodox and nationalist narratives, are by default devised to drive a wedge between the nation’s undesired recent past of foreign domination on one side and the dual time perspective of an ideal present and a mythical distant past on the other.
- 2 ‘Nationalising historiographies’ are ‘mythic narratives, including myths or origin and descent whic (...)
5Consequently, the focus of inquiry used to be on the processes of state and nation-building after the ‘liberation’ from imperial ‘hegemons’ such as the Ottoman or Habsburg empires. In these national narratives, history begins with independence and is posited against two historical time-scales characterised by dichotomic values on the axes of proximity and desirability. The distant past is imagined yet adored, while the immediate past is in the reach of memory, yet abhorred. Nation-builders in Greece created the basis of modern Greek identity by coupling antiquity and to a lesser extent Byzantion with the Otherising of the Ottoman Empire and Turkey (Ozkırımlı & Sofos 2008). Throughout the Balkans, national narratives are based on the dichotomy of positive myths of origin on the one side and belittling accounts of the historical ‘Others’ (Brown 1996; Todorova 1997). Contrary to established wisdom, nationalising historians in Turkey have followed a similarly dichotomic trajectory by vilifying the Ottoman past as religiously backward and ethnically cosmopolitan while imagining ethno-genetic narratives of a central Asian past (Behar 1992; Yıldız 2001).
Such ‘nationalising historiographies’ (Smith 1998: 20),2 by definition, reduce the immediate past preceding the establishment of the nation-state to the struggle of national ‘heroes’ against the hegemonic system of oppression. Groups with competing claims over the same territory are vilified or ignored, never respected and almost always dehumanised. This approach, institutionalised in history institutes and universities, school and textbooks has its own methodology, sources and documents, which all aim at creating and rendering hegemonic a unified version of national history. In this respect, nationalising historiographies, together with nationalising cultural policies such as the naming of a given territory, could be described as the constructive dimension of nation-states. Such strategies, however, would be meaningless without the destructive antecedent that creates the required tabula rasa or at least, subjects and dis-empowers undesired ethno-religious communities.
- 3 A post-nationalist subject position here refers to the position of the scholar operating from a pos (...)
6The recent decentring of national historiographies indeed breaks with this historiographical perspective as it makes possible the shift towards transnational perspectives, and a post-nationalist subject position3 allows for a thorough reconsideration of historiography and identity: Rather than concentrating exclusively on the history of the nation-state and its construction, questions regarding the moments of transition gain importance. What happened in the transition between pre-modern empire and nation-state? Under what conditions were heterogeneous spaces transformed into homogenous territories? The refocusing on the ‘pre-history’ of nations, then, brings to the fore policies of destruction rather than construction.
7A growing body of literature highlights the role of ethnic cleansing in the making of modern Europe, in particular during the transition from empire to nation-state. The authors of this new school have shown convincingly that the modern nation-state is established on the blood of others, be it based on the territories of colonised indigenous people as in the United States, fuelled by stages of ethnic cleansing and forcible assimilation as in Spain and France (Mann 2004; Marx 2003), or by large-scale expulsions and massacres as in the Balkans and Turkey (Mazower 2004) yet also during and in post-world War II central Europe (Liebermann 2006).
- 4 It is by now a truism to state that the debate on the genocidal killings of Ottoman Armenians in 19 (...)
8Reflecting these post-national debates in and on Europe and South East Europe, a progressively revisionist perspective on modern Turkish history and the discriminatory dimension of the Kemalist Republic has flourished alongside conventional, yet increasingly sterile accounts of historical orthodoxy. The vocabulary of the study of modern Turkey has been extended considerably to admit new conceptual perspectives ranging from assimilation, settlement and deportations especially in the debate on Kurdish identity to the contested terminology of massacres, ethnic cleansing and genocide and a renewed, if volatile interest in the analysis of the events of 1915 (Akçam 2004; Akçam 2006; Bloxham 2005).
Probably the most recent addition to the vocabulary is the concept ‘demographic engineering’. The question why yet another concept should be introduced into a scholarly field that already abounds with different theoretical approaches is justified. Does the literature of political mass violence, ethnic cleansing and genocide not explain it all, as many authors specialising on the latter would suggest?4 Or, does this new concept, originating from studies of demography and migration, deflect, once again, from the destruction and suffering of the victims and re-establishes the focus on the constructive efforts of the nation state?
Much of the cited literature on the destructive precedent to the nation-state does indeed stop with the completion of such policies. In the field of genocide studies for instance, the primary concern lies with the planning, unfolding and realisation of genocide, not so much with the question how it ties into larger questions of national identity and nation-building (Levene 1998; Weitz 2003). It is in this context that the conceptual turn to demographic engineering might open up new avenues for research, which might be particularly insightful for South East Europe and the Ottoman/Turkish case (cf. also Table 1).
Table 1: Demographic Engineering as intermediary category
9In this perspective, demographic engineering ‘consists of the full range of policies impinging on a population which states pursue in their own interest’ (Kreager 2002: 323, discussing Weiner & Teitelbaum 2001). These policies are ‘state-directed movement[s] of ethnic groups’ with the goal of ‘the consolidation of the state’s hold over a particular piece of territory’. They have a clear spatial dimension in as much as they recreate the territorial distribution of communities ‘by facilitating control and or assimilation of a minority group or by removing the minority form the territory in question’ (McGarry 1998: 630). In an even more inclusive definition, Seker adds to this all state programmes and policies, ‘which aim to increase the political and economic power of one ethnic group over others’ (Seker 2007: 461).
This inclusive definition based on Weiner, Teitelbaum, McGarry and Seker is characterised by three priority concerns: The primary role of the state, the range from destructive to constructive aspects, and a special concern with space and territory. It is with this inclusive definition that demographic engineering as a conceptual tool and body of scholarship might have the potential of bridging the gap between the studies of the destructive precedent and the constructive phase of the nation-state by insisting on the strong relation between the two. In this perspective, the forcible deportation of entire communities and, if need by their destruction, emerges as an integral condition and constituent part of the national project. The authors of this project, in turn, ignore, relativise or justify the homogenisation of a territory in physical, demographic, ethnic and cultural terms. Taken as an integrative concept, demographic engineering would then allow for two important contributions to the debate. First, it would facilitate the exploration of a continuum between phases of destruction and construction and their mutual dependency. Secondly, it would allow to discuss the destructive and constructive moments independently from a chronological perspective and to treat them as dimensions that can be co-present at any given time and in any nationalising state-policy.
- 5 A recent media debate, initially begun in the Turkish-Armenian newspaper Agos and taken up by the m (...)
10In this reading, demographic engineering synthesizes the destructive core of national projects such as ethnic cleansing and population exchange with its constructive yet discriminatory programme of nation-building. To stay within the context of South East Europe and within the realm of the built environment, where the term destruction becomes most imminent: Contrary to the conventional understanding of nationalist Greek historiography, the reconstruction of Thessaloniki as a Hellenic city after the great fire cannot be analysed without a reference to the destruction of Ottoman, and hence Jewish, Muslim and Slavic Selânik, especially not since Mazower‘s seminal monograph (2004). Ankara is the show-case of the Kemalist mission civilisatrice, and a myriad of studies reiterates the faulty dichotomy between a ramshackle country-town and a modernist project for a new capital (Tankut 1993). An analysis committed to an integrative approach, however, would be incomplete without mentioning the city’s important Armenian heritage on which the Republican nation-builders superimposed their architectural vision and on whose resources they drew.5
- 6 During periods of national consolidation in the 1920s and 30s, such strategies have been at the cor (...)
11We have seen that demographic engineering, in its inclusive definition, deals with the state-directed removal or destruction of certain communities from a given territory in order to consolidate power over that territory and prepare the conditions for the nation state to project its vision of space and time, to ‘Turkify’, ‘Hellenise’ or ‘Bulgarianise’.6 Yet the removal of a population is often not sufficient to create the required tabula rasa, as too many markers of the excised identity remain in space: From the former residents’ architectural heritage to the very names of the cities, villages, quarters and streets they inhabit, artefacts and discursive reminders of the ‘other’ abound. It is not surprising then, that historians of nationalising states should be obsessed with changing the markers of communities, which they wish to forget, i.e. their toponymes, but also their anthroponymes. Probably not surprising either is that very little effort has been made so far to explore these prime examples of nationalising policies, with the notable exception, in the Turkish case, of Samim Akgönül’s extensive work on the politics of place and personal names (Akgönül 2006) and for Bulgaria, Mehmet Hacısalihoğlu’s work on village names in Eastern Rumelia (2008). To this, we can add a few dedicated essays (Aksu 2003; Koraltürk 2003; Tunçel 2000), especially on the 1934 ‘Law on Family Names’ (Ertan 2000) and Esra Danacıoğlu’s (2001) instructive methodological reflections on Ottoman place names and their transformation during the Turkish Republic.
- 7 The Greek, Lazuri- and Armenian-speaking communities of the Black Sea littoral were excluded from t (...)
12Reflecting the longue durée of ethnic, religious and linguistic influences of pre-1915 Ottoman lands, place names in the territory that was to become Turkey in 1923 displayed a high degree of diversity. Place names of Greek and Slavic origin were widespread in the Aegean and Thrace. Greek, Georgian, Lazuri and Armenian7 toponymes abounded at the Black Sea littoral and in the North East. Aramaic, Syriac, Arabic, Kurdish and Armenian dominated in the East and South East. In addition, a great number of toponymes included religious connotations, either of Muslim provenance (references to religious titles like Şeyh, Molla or Hacı) or of non-Muslim background (such as Church, Kilise or Monastery, Keşişlik, as a composite part of the place name), neither of which had a place in the time and space vision of the Republic.
13The state policies targeting this diverse toponymical heritage, I shall call ‘toponymical engineering’ due to its conceptual proximity to demographic engineering, with which it often, if not always goes hand in hand. Its main objective is the destruction of the interwoven layers of historical and linguistic meaning, i.e. of the ‘archaeology’ of place names and its replacement with an alternative toponymical order that conforms with the time and space vision of the nation-state.
- 8 Toponymical engineering began in earnest in 1915. Earlier, a cartographic codification of intercha (...)
14Turkey’s toponymical destruction/construction is closely related to the emptying of its territory and its re-population. Yet, with its hybrid identity as a destructive/constructive policy, state agencies have not ceased to enforce it throughout the entire republican history of Turkey. I shall look into four waves of name changes and two significant overlaps of instances of demographic and toponymical engineering.8 The first wave overlaps with the decade of demographic engineering preceding and leading to the foundation of the Turkish Republic, the second covers the period of one-party rule, while the third resonates with the operations of the Expert Commission for name change’ (1950-1980s). Finally, the last section discusses the second overlap of demographic and toponymical engineering: the ‘Symposium of Turkish Toponymes’ in the 1980s and the eviction of hundreds of thousands of Kurds in the country’s South East.
15When the Committee of Union and Progress (CUP) declared the deportation law for ‘those opposing the government in times of war’ on 27 May 1915, more than a million Armenians, Syriac Christians, and some Kurdish communities were forced into exile and destruction. In only a few weeks, the government initiated the name change of evacuated villages (Dündar 2001: 65). At the same time, some of these villages were swiftly resettled with Muslim refugees, pouring into the country from the Balkans and the easternmost provinces under Russian occupation. In a directive, the Chief of the General Staff and one of the three leaders of the Committee of Union and Progress (CUP), Enver Pasha, declared that:
It has been decided that provinces, districts, towns, villages, mountains and rivers, which are named in languages belonging to non-Muslim nations such as Armenian, Greek or Bulgarian, will be transformed into Turkish. [...] In order to benefit from this suitable moment, the goal of name change should be achieved as soon as possible (translated from Dündar 2001: 82).
16This initial attempt of re-naming proved to be short-lasted, as the re-naming during times of warfare threatened to impede military communications. Yet, local military commanders continued a policy of fait accompli in most provinces regained from the Greek army and replaced toponymes that had a Greek or Slavic etymology. A body of correspondence between ministries, the Ankara government and the Office of the General Command from 1921 exposes the tension between hardened commanders on the ground who wanted to efface immediately the memory of ’the enemy’ from the territory they had just liberated and government representatives who were thinking in terms of a long-term ‘scientific solution’.
After the expulsion of Greek troops from Western Anatolia, the Chief of the General Staff replied to a circular from the Interior Ministry, accepting the need for changing place names in principle, yet advising against imprudent decisions:
It is made known that the change of some village and town names, which are named with foreign names will be delayed until the demobilisation of troops, because a part of the population of these villages and towns is now mobilised, and the change would cause disruptions to communications and intelligence. Especially the national feelings of the population of the western provinces, which has suffered from the atrocities and aggressions of the Greeks and the local Rum is now too vehement to name and refer to their places of origin with foreign names. Therefore, Sir, I submit and suggest forcefully that it will be attempted immediately and swiftly to change these (translated from BCA 030.10/66.439.21 25/6/1923 [correspondence mostly from 1921]).
17The Interior Minister agreed, yet insisted on a ‘scientific examination’:
As the details of changing of village and town names is a matter, which has to be examined scientifically, taking into account the historical circumstances and the geographical works, and the change of geographical names would require the historical registers […] the following registers, which contain the village and town names in the western provinces, which the Glorious President of the General Staff deemed necessary to be changed […] have been submitted to the [Education?] ministry. I submit that in the choice of the examination, the necessary steps will be taken with great speed […] (translated from BCA 030.10/66.439.21 25/6/1923 [correspondence mostly from 1921]).
- 9 The two representatives Ahmet Saki and Fuat of Antalya and Kırklareli brought forward a motion in 1 (...)
- 10 Cf. (Akgönül 2006: 94; Koraltürk 2003: 99) ; BCA 030.10.0.0/ 66.440.10 (31/10/1925)
18Although the quest for a systematic renaming remained inconclusive after the Minister of Education, Ziya Gökalp, stalled the process, quite a few members of the nascent Parliament in Ankara urged a complete and immediate change of place names (Koraltürk 2003: 98). Many place names were nevertheless changed after long and emotive debates in the Parliament,9 mostly in line with the pragmatic position of the Interior Ministry ‘to change the names of provincial and district centres gradually, while refraining from re-naming villages for the time being’. Apparently, the debate became most heated during the debate of a motion in December 1924 that led to the change of the city of Kırkkilise in Western Thrace – the Ottoman translation of the old Byzantine name ‘Seranda Ekklesies’ [40 Churches] (Aksu 2003) – to Kırklareli [Land of the Forties].10
19When the first place-name directory of the Republic, ‘Our Villages’ was compiled in 1928, many of the Greek names in Aegean Turkey and many Kurdish and Armenian toponymes in the Southeast had already been replaced with Turkish substitutes, especially on the level of district centres (DV 1928; Kreiser 1975: XX). Yet, although place names were changed in this period, a more emotive and less rigorous approach to the issue seems to have prevailed within the government of the time. This was not yet the high-tide of toponymic engineering, but rather a spontaneous initiative by military commanders, local administrators and Parliamentarians, competing to outdo each other in proving their nationalist credentials.
- 11 Dahiliye Vekaleti, Mahalli İdareler Müdüriyet-i Umumiyesi.
20With the inception of the Republic, the foundation of scholarly institutions such as the ‘Turkish Linguistic Society’ and the ‘Society for the Study of Turkish History’ set the frame within which the state would attempt to realise a ‘scientific’ consideration of the matter. With the promulgation of the Republic’s first constitution [Teşkilat-i Esasiye] in 1924, the local government was defined as a three-tier system of provinces [Vilayet], districts [Kaza] and sub-districts [Nahiye], largely in continuation of earlier Ottoman provincial reform acts based on the French model of local governance. The provincial government, understood as a division of home rule from Ankara, was placed under the supervision of a General Directorate of provincial administration.11 The Directorate published, in 1928, the first in a series of directories called ‘Names of our villages according to the new territorial division’ [Yeni teşkilat-i mülkiyede köylerimizin adları] in Arabic/Ottoman script. This directory was the last one to be written in the ‘old script’, and hence, the last to allow for the non-standardised – and hence more accurate- rendering, even of place names in Turkish.
21The ‘Law on the Adoption and Application of Turkish letters’ [Türk Harflerinin Kabul ve Tatbiki Kanunu] promulgated by the Parliament in Ankara in the same year outlawed Arabic script in all official and private documents (Heyd 1954: 22). It opened the way for a complete revision of all media, including maps and place-name directories, which were to be published in Latin script in the following years. In 1929, the General Map of Turkey was printed by the Office of the General Staff in Latin script, and four years later in 1933, the General Directorate for Provincial Administration sent a revised version of the directory ‘Our Villages’ to all its local branches (DV 1933b). The ‘Map of Turkey’ [Türkiye Haritası] followed in 1934, to be accompanied by a directory of place names edited by the Turkish Geography Association (TCK 1946).
- 12 The ‘Society for the study of Turkish history’ [Türk Tarihi Tetkik Cemiyeti] was later re-named as (...)
- 13 The general nationalist atmosphere of the time and the fervour with which the ‘Turkification of Tu (...)
22With the foundation of the ‘Society for the Study of Turkish History’ (TTTK)12 in 1931 and the ‘Turkish Linguistic Society’ in 1932 on orders of Mustafa Kemal (Heyd 1954: 25), both the Language Reform and the ‘Turkish History Thesis’, which depicted the history of Turkey as that of the advance of Turkic tribes from Central Asia to Anatolia, reached their revolutionary pinnacle. The first ‘Turkish History Congress’ in July 1932, brought together members of the Society, the Istanbul University and history teachers from institutions of secondary education (TTTC 1933), with the aim of propagating the ‘Turkish History Thesis’ to a wider public.13
- 14 The petitions came from primary school children, citizens and civil servants. Among them were sugg (...)
23Renaming efforts continued throughout these years, stretching from the urban toponymy to the meta-level of regional names. In 1927, all street and square names in Istanbul, which were not of Turkish origin, were replaced (Okutan 2004: 182). During the heyday of Kemalist one-party rule, citizens petitioned the General Secretary of the Republican People’s Party, now intertwined with the Interior Ministry, to rename the capital Ankara as Gaziyuva [Nest of the Ghazi].14
- 15 In the statement of reasons, the General Directorate for Provincial Administration stated that ‘it (...)
24A number of district centres in the Southeast, that by then had become largely Kurdish, were renamed during the administrative re-organisation of the eastern provinces in 1936.15 It was also during this phase that another topos of a specific Republican vision was established, namely the effort to extend its vision of time and space beyond its borders. The reference to historical regions such as Armenia, Kurdistan or Lazistan – the official name of the eastern Black Sea province of Rize until 1921 – was forbidden and a ban imposed on the importation of maps containing these terms (BCA 030.18.01.02/88.83.20 31/8/1939).
- 16 Geoffrey Lewis, the doyen of the history of Turkish language at Oxford gives an impressive account (...)
25Despite the great numbers of official maps and directories published in these years, and the aforementioned renaming efforts, the toponymical strategy remained limited, when compared to the ‘success’ of the ‘purification’ of the language. Above all, the pressing issue appears to have been the excision of Arabic script and Arabic/Persian vocabulary from daily communication,16 and hence the deletion of the Ottoman/Arabic typeface from official documents. This might explain why, in 1940 and facing WW II, the Interior Ministry felt obliged to declare that a large-scale campaign of renaming place and geographical names had become necessary because of the ‘vital importance of the issue to our national existence’ (Gökçeer 1984: 1), an attempt, which was to be largely inconclusive. The management of place names, however, would become top priority. With the end of WW II, the General Directorate of provincial administration issued a circular requesting the provincial governors to ‘ascertain place names with roots in foreign languages and to send the respective files to the [Interior] Ministry’ (Gökçeer 1984: 1). The Provincial Administration Act [İl İdaresi Kanunu] in 1949 provided the legal basis for the future renaming. The act explicitly referred to the renaming of villages and assigned the responsibility of name changes to the Provincial Councils, under the supervision of the Interior Ministry. Provincial governments could now initiate the change of place names, yet it seems that the process was delayed by limited compliance on the local level.
- 17 Settlements below village status were hamlets, pastures, alms, a group of houses owned by a family (...)
26The directories published by the General Directorate in 1946 ‘Populated Settlements in Turkey’ [Türkiye’de meskun yerler kılavuzu] (IIB 1946) and ‘Divisions of Home Rule’ [Mülki İdare Taksimati], still abounded with village names of non-Turkish origin, although a standardised orthography had now been established. More importantly, the directory of 1946 established the ground for the systematic renaming, which was to begin a few years later. It contained more than 67,000 entries, including settlements below village status,17 and served as the key reference book that facilitated the renaming in the 1950s. The years of the early Republic, then, saw the preparation of the ‘scientific policy’ promised in the founding moments of modern Turkey and the emergence of the bureaucratic and legal infrastructure that would make this policy possible. Its execution, however had to wait, ironically, for the advent of democracy.
- 18 The members of the Commission were appointed with a ‘Decree of the Council of Ministers dated 19/1 (...)
27In the wake of multi-party politics in 1950 and the ascent to power of the culturally conservative Democrat Party, the most systematic phase of the renaming of villages and topography began. In 1957, the General Directorate for Provincial Administration initiated an ‘Expert Commission for name change’ [Ad Değiştirme İhtisas Kurulu].18
The Commission met three times a week in the Directorate with the mission to examine Turkey‘s toponymes and suggest Turkish alternatives where possible:
The Commission examined natural place names on a number of maps with different scales. It examined village names and related names, names of train stations, gendarmerie posts, lighthouses, capes and bays. It suggested Turkish names to the responsible provincial councils. According to the Provincial Administration Law (No. 5442), the necessary decrees were passed and these place names were Turkified’ (IIB 1977: 3).
28Among its members were representatives of the Office of the General Staff, the Ministry of Defence, the Ministry of Education, the Faculty of Letters, History and Geography of Ankara University [Dil ve Tarih Coğrafya], the Turkish Language Society and the General Directorate of Cartography.
29Despite the systematic work of the Commission, however, local resistance in the Provincial Councils seems to have slowed down the process, as the name changes had to be confirmed by elected Councils rather than by Ankara-appointed governors. In order to accelerate the process, the Commission prompted the General Directorate to initialise an amendment of the Provincial Administration Act. Passed in 1959, the amendment transferred the responsibility of the ‘Turkification of place names’ from the Provincial Councils with elected members to the Provincial Administration, which was composed of the Governor and civil servants appointed by the Interior Ministry. Through this amendment, the Commission succeeded in imposing changes without having to wait for the approval by locally elected bodies, stressing that this was now a project of the bureaucratic elites that would be continued irrespective of the political party in governments.
30Yet, the process was decelerated further by a lack of support for the name-change strategy on the side of the government. While the Democrat Party (DP) made no overt effort to forestall the work of the Commission, it could be suggested that its conservative elites, known for their desire to revert the language reform, were not as fervently committed to the Turkification of toponymes, and certainly all but enthusiastic of its secularist tendencies. Soon, however, the Commission was to find a more favourable environment for its proceedings: After the military coup of 27 May 1960, and despite the swift re-introduction of multi-party politics in the following year, the principles of the language reform and the renaming policy were reinforced by the military-appointed care-taker government. A circular of this government was communicated to all ministries in January 1961, outlawing the ‘use of any foreign word for which a Turkish equivalent existed’ (Lewis 1999: 157).
31The Commission assembled throughout the 1960s, and presented the outcome in the new edition of ‘Our Villages’ in 1968 [Köylerimiz] (IIB 1968). The directory introduced more than 12,000 toponymes, replacing approximately thirty percent of the 45,000 village names in Turkey (cf. Table 2 for an exemplary change of district names in the province of Urfa).
Table 2: Changed district names in the province of Urfa
|
District (Ilçe)
|
Sub-district (Nahiye, Bucak)
Name in 1967
|
Translation
|
Sub-district (Nahiye, Bucak)
Name prior to 1967
|
Remarks/ Translation
|
|
Merkez
|
Merkez (Urfa)
|
--
|
--
|
|
|
Akziyaret
|
White Shrine
|
Cülmen
|
|
|
Kalecik
|
Tiny Castle
|
Kabahaydar
|
Turkish, restored later
|
|
Çamlıdere
|
Pine Brook
|
Mecrihan
|
Kurdish
|
|
Dolunay
|
Full Moon
|
Payamlı
|
Later reverted to Payamli
|
|
Yardımcı
|
Helper
|
Sumatar
|
Semitic
|
|
Akçakale
|
Merkez (Akçakale)
|
Whitish Castle
|
Tell El-Abyad
|
Arabic, White Mound
|
|
Altınbaşak
|
White ear
|
Harran
|
Semitic
|
|
Şehit Nusret Bey
|
Martyr Nusret
|
Nusratiye
|
Arabic
|
|
Birecik
|
Merkez (Birecik)
|
--
|
--
|
--
|
|
Böğürtlen
|
Blackberry
|
Bugurtalan
|
Turkish
|
|
Bozova
|
Merkez (Bozova)
|
Grey Plain
|
Bozabad
|
Turkish composite, Grey city
|
|
Yaylak
|
Mountain Pasture
|
Baziki
|
Kurdish tribe name
|
|
Kanlıavşar
|
--
|
--
|
|
|
Halfeti
|
Merkez (Halfeti)
|
--
|
Rumkale
|
Turkish, Greek Castle
|
|
Hilvan
|
Merkez (Hilvan)
|
--
|
--
|
|
|
Gölcük
|
Pond
|
Tilfelis, Hamdun
|
Arabic/ Semitic
|
|
Ovacik
|
Meadow
|
Hasanik
|
Armenian/ Kurdish
|
|
Siverek
|
Merkez (Siverek)
|
--
|
--
|
|
|
Bucak
|
Corner
|
Fak
|
Kurdish
|
|
Çaylarbaşı
|
Brooks point
|
Herheri
|
Kurdish
|
|
Dağbası
|
Mountain point
|
Karahan
|
Turkish, Black House
|
|
Karacadağ
|
Black Mountain
|
Kaynak
|
Turkish, Source
|
|
Karakeçi
|
Black Goat
|
Mizar
|
Kurdish
|
|
Şekerli
|
Sweet
|
Karacaviran
|
Turkish, Blackish ruins
|
|
Suruç
|
Merkez (Suruç)
|
--
|
--
|
|
|
Viransehir
|
Merkez (Viransehir)
|
--
|
--
|
|
|
Ceylanpınar
|
Source of Gazelles
|
Ra’s El-‘Ayn (Resülayn)
|
Arabic, Spring Point
|
|
Demirci
|
Ironmonger
|
Sergirti
|
Kurdish
|
Based on the Divisions of Home Rule 1933, 1940 and 1948 (DV 1933a; DV 1940; DV 1948), the Provincial Yearbook of Urfa (Urfa Valiligi [Urfa Governorate] 1967), (Rousseau 1825) and (Kiepert 1884).
The term merkez [centre] signifies the administrative centre of the province, the district or the sub-district, as each administrative unit is named after the administrative centre.
32A general reform of the state administration after the coup d’Etat in 1971 instigated a re-organisation of the General Directory for Provincial Administration. A new office, the ‘fifth branch’ [Beşinci Şube], was established for
[T]he printing and dissemination of maps, the incorporation of changes in city limits into existing maps; the ongoing reporting of the Expert Commission for Name Changes; the preparation of annual province reports [İl Yıllığı] and the implementation of tasks pertaining to administrative geography; the publication of reports; and the execution of the Development Plan, the annual programs and the implementation plans (translated from http://www.icisleri.gov.tr).
33Although the Expert Commission had become defunct after the completion of the 1968 directory, it appears that the ministry’s bureaucrats did not regard the task of renaming as completed. In 1973, the Commission was reconvened and commenced work on smaller-scale maps, focussing on geographical names and settlements below village status. In addition to more than 2,000 village names, the names of 12,884 hamlets (out of a total of approximately 39,000) were changed (Gökçeer 1984: 4).
- 19 It is not clear which date Emiroğlu sets in order to establish that place names have been changed. (...)
34By 1968, and even more clearly in the ‘Divisions of Home Rule’ of 1978, not only the scale of the name change became apparent but also its regional differentiation. Around 36% of all village names in Turkey, excluding settlements below village status, had received new names19 (cf. Table 3). While the proportion of changed village names in provinces in western and central Turkey was below 30%, the same figure ranged between 44 – 91% for the provinces in the Southeast and East of the country. In all provinces shown in the table, either significant non-Turkish communities existed in 1978, or large Armenian or Syriac populations had lived there before 1915. In the province of Mardin, where virtually all place names (91%) were Turkified, Arabic, Kurdish and Syriac-speaking communities accounted for the majority. The provinces of Bitlis, Siirt, Bingöl, Hakkari, Muş and Van, all with a proportion of over 75% of new place names, had substantial Armenian communities before 1915 and Kurdish majority populations thereafter. The emerging image seems to have been largely coherent with the ethnic policy of the Republic: The Southeast, and to a lesser extent the East, with a prevalent Kurdish population and a strong Armenian heritage were the Commission’s priority target, followed by the Black Sea region with its significant communities of Armenian- [Hemşin], Lazuri- and Greek-speaking communities. The two exceptions can be explained with a historical reference: Central Anatolian Konya hosts sizeable enclaves of resettled Kurds, while the Arabic-speaking province of Hatay was part of the French Mandate on Syria until 1942, and hence had escaped earlier efforts of Turkification.
Table 3: Proportion of changed village names per province and region
|
Province
|
Geographical Region
|
Significant ethnic communities
|
Name change in %
|
|
Mardin
|
Southeast
|
Armenian, Arab, Syriac, Kurdish
|
91
|
|
Bitlis
|
Southeast
|
Armenian, Kurdish
|
86
|
|
Siirt
|
Southeast
|
Armenian, Arab, Kurdish
|
84
|
|
Bingöl
|
Southeast
|
Armenian, Kurdish
|
80
|
|
Hakkari
|
Southeast
|
Syriac, Kurdish
|
80
|
|
Mus
|
Southeast
|
Armenian, Kurdish
|
77
|
|
Van
|
Southeast
|
Armenian, Kurdish
|
75
|
|
Trabzon
|
Black Sea
|
Greek, Lazuri
|
72
|
|
Diyarbakir
|
Southeast
|
Armenian, Kurdish
|
68
|
|
Elazig
|
Southeast
|
Armenian, Kurdish
|
68
|
|
Gümüshane
|
East
|
Armenian
|
68
|
|
Adiyaman
|
East
|
Armenian, Kurdish, Arab
|
65
|
|
Agri
|
East
|
Armenian, Kurdish
|
65
|
|
Erzincan
|
East
|
Armenian, Kurdish
|
63
|
|
Erzurum
|
East
|
Armenian, Kurdish
|
63
|
|
Urfa
|
Southeast
|
Armenian, Arab, Kurdish
|
57
|
|
Kars
|
East
|
Armenian, Georgian, Kurdish
|
52
|
|
Gaziantep
|
Southeast
|
Armenian, Kurdish, Arab
|
48
|
|
Malatya
|
Southeast
|
Armenian, Kurdish
|
44
|
|
Artvin
|
Black Sea
|
Georgian, Lazuri
|
39
|
|
Hatay
|
Çukurova
|
Arab
|
37
|
|
Giresun
|
Black Sea
|
Greek, Lazuri
|
34
|
|
Rize
|
Black Sea
|
Greek, Lazuri, Hemsin
|
33
|
|
Sivas
|
East
|
Armenian, Kurdish
|
33
|
|
Konya
|
Central
|
Kurdish
|
32
|
Table 4: Proportion of changed village names per region
|
Geographical Region
|
Range of changes
|
Median
|
|
Southeast
|
44 – 91
|
76
|
|
East
|
33 – 68
|
63
|
|
Black Sea
|
33 – 75
|
37
|
|
Central
|
32
|
-
|
|
Hatay
|
37
|
-
|
Based on the alphabetical list of provinces in Emiroglu (Emiroglu 1984: 198f.), and the directory ‘Divisions of Home Rule 1978’. Only provinces with a proportion of new village names over 30% are shown.
35Around the same time, the Directorate distributed the directory ‘New natural place names’ [Yeni Tabii Yer Adları], edited by the director of the fifth branch, Remzi Atman. It listed 1,819 new geographical names that had been changed under the two Commissions (IIB 1977: 4). The Commission was once again dissolved in 1978, apparently by intervention of the Prime Ministry, which objected to the Commission’s zeal in ‘changing historical place names’ (Gökçeer 1984: 3), a practice allegedly causing confusion especially in the tourism industry, as Roman and ancient Greek cities received new names. Thanks to the concerted efforts of the two successive Expert Commissions and the modified legal framework that had transferred the authority of naming places to the Commission, maps of Turkey by 1978 displayed a largely uniform and mostly Turkish toponymy.
36On the scholarly front, interest in the study of place names was reignited among historians and linguists, when a group of Turkish linguists attended the eleventh congress of Onomastic Sciences in Sofia in May 1972. A member of the Turkish delegation, Doğan Aksan, presented a survey on the categorisation of place names in Turkey and their etymology and established a continuity of central Asian and Turkish toponymes:
We are convinced that it has been more or less established which criteria the Turkish people applied when naming places, where it settled down, which concepts it drew from and what its general approach to the matter was […] (translated from Aksan 1974: 186).
- 20 Soylu demonstrated, on the example of a village in the province of İçel, how the name change had l (...)
37Aksan himself was referring to data published in the ‘Bulletin for Turkish Language Studies’ (Baskan 1970), which drew on the 1968 edition of ‘Our Villages’. As I have shown earlier, this directory had listed more than 12,000 new village names. Based on these directories – produced by civil servants with the intent to replace ostensibly non-Turkish place names with Turkish substitutes evoking a Central Asian heritage – the self-declared ‘onomasts’ now proved that Turkish place names appeared to be coherent with toponymes in Central Asia. Despite this highly unethical research agenda, however, little criticism seems to have been expressed against the name change campaign in the 1960s and 70s, be it in academic publications or in the general media. The few critical voices mourned the loss of originally Turkish place names, which the members of the Commission had mistaken for ‘foreign’ toponymes, and hence had changed (Soylu 1972: 6239).20
38To recapitulate, the period between the 1950s and 1980s hosts the most momentous changes to Turkey’s toponymy, with the grip of the Commission getting ever tighter and reaching out ever further, into hamlets, alms, pastures, mountains and rivers. A new pattern also emerges: Democratically elected governments even if they do not always stop the practice of renaming, are remarkably less inclined to support and facilitate the Commission’s work. Considered in this light, the Turkification of Turkey’s time and space emerges as a policy of bureaucratic elites that lingered on during democratic periods and was imbued with renewed vehemence during the interludes of military rule.
- 21 The Commission was re-established according to a directive published in the ‘Official Gazette’ on (...)
39By February 1983, shortly before the election of the Motherland Party under its leader Turgut Özal, and in the shadow of the Kurdish insurgency, the toponymical campaign was resumed, hence beginning a second phase of overlapping demographic and toponymic engineering.21 As we have seen, the overwhelming majority of toponymes in the Southeast and Eastern provinces had already been changed by 1967. This left the toponymes of settlements below village status such as hamlets, pastures and clan-based or tribal settlements. These were the central sites of daily life practices of nomadic Kurds in the mountainous regions of historical Kurdistan, and hence, highly symbolical resources for Kurdish identity politics.
40Against the backdrop of the aftermath of the 1980 military coup and the rising violence in the Kurdish provinces, in September 1984 the ‘Office for the research of National Folklore’ and the Ministry of Culture and Tourism organised a ‘Symposium on Turkish Toponymes’. The everyday fascism of the military junta and the hysteria of the participants left its mark on the conference proceedings, which resembled in content and language the 1933 First National History Congress. Fikri Gökçeer, President of the General Directorate for Provincial Administration, in his opening speech, defined the aims and working goals for a re-convened Commission:
a) Names that are not Turkish, whose pronunciation and structure is incompatible with the vocal harmony of Turkish, which might be confused due to similar pronunciation and which do not have a pleasant meaning and are contrary to the common sense of the people shall be changed.
b) Names, which are Turkish, yet corrupted by local dialects, shall be restored according to the correct orthography.
c) Foreign place names shall not be substituted by their translation into Turkish. However, if the old name refers to a natural or topographic characteristic of the village, a translation may be considered.
d) When foreign place names are replaced, no names shall be given, whose pronunciation may evoke the old name (Gökçeer 1984: 4).
41The president introduced two new principles: The first was the beautification of ‘unpleasant names’, already part of earlier practices, and the second a clear reminder that the translation of toponymes would help reconstituting the history of a place and should therefore be shunned. The latter implies a hardening of the toponymic policy, which had tolerated translations in earlier phases.
42The director of the Cartographic Office of the Ordnance Command [Harita Genel Komutanlığı, Kartografya Şubesi] relayed the current view of the country as beleaguered by external and internal enemies, when he outlined a further goal for the Commission unmentioned by Gökçeer:
To intervene in international institutions, in order to correct dangerous statements in the foreign press, which claim the existence of foreign place names in Turkey[…] To correct internal publications, which do not comply with the Turkish [History] Thesis and to ensure their convergence (Orcan 1984: 60).
43If the director’s military logic re-enacted the simple, yet coercive mindset of the early Turkish Republic, the representative of the organising body, the ‘Office for the Study of national culture’, Nail Tan, went a step further. He suggested that the tenets of the ‘Turkish-Islamic’ synthesis, an amalgam of nationalist rhetoric superimposed on a state-controlled version of Islam should guide future name change policies. Resounding with Enver Pasha’s directive of 1915, he urged the audience of historians, linguists and civil servants that ‘the efforts towards the changing of toponymes of non-Turkish or non-Turkified forms should be completed swiftly’, in order to forestall those ‘who argue that toponymes stemming from the Anatolian civilisation should not be altered’ (Tan 1984: 279). His justification of renaming was a prime example of the ‘Turkish-Islamic synthesis’ in progress, which was to shape Turkey’s ideological landscape for almost two decades:
The state of the Republic of Turkey is a state founded by the Turkish nation under her great leader Atatürk. It is inconceivable that the lands considered as homeland by this state founded with the blood of hundreds and thousands of martyrs should abound with foreign place names (translated from Tan 1984: 280).
44Tan’s concession that it is not the people but the state that considers the lands of Turkey as homeland is truly remarkable and helpful in revealing the underlying understanding of state and citizenship that has been at the root of the renaming policies. Even more momentous is his remark about the ‘foreign place names’, which abound on those lands, if one takes into consideration that up to 90 % of place names in the provinces of the Southeast had been changed during earlier name change campaigns. The Symposium’s major target then was the renaming of hitherto unchanged hamlets and other settlements below village status. Tan also argued for a more effective implementation of the policy, as he believed that the new names were not fully internalised by local populations:
- 22 Tan’s concern that the name change was not taking hold even in government offices might have been (...)
Above all, it should be communicated to the local people that Armenian and Greek toponymes were changed because they were not Turkish, maps complying with the new names should be swiftly drafted and distributed to schools and state agencies.22 […] Local village, mountain, hill, plain, quarter and street names should be taught in the geography and social sciences courses at primary and secondary schools, without mentioning the old place names (Tan 1984: 282).
45One of the few critical voices at the Symposium was an academic from the University of the Euphrates in Elazığ in Southeast Turkey. He did not, as many others, justify his critique with the undesirable fact that toponymes of Turkish origin had also been subjected to the pedantry of the Commission. He rather drew attention to the problems, which the escalating process would cause for studies in the field of historical geography, and delivered a remarkable plea against this final renaming frenzy:
The Commission has continuously and repeatedly changed the names of villages, train stations, gendarmerie posts, lighthouses, capes and bays. Yet it seems to be unimaginable that names of locales [Mevkii adları] could be changed with such ease, as these names have become a property of the people and a reflection of their material culture (translated from Karaboran 1984: 144).
46Was Karaboran aware of the fact that Nail Tan was alluding precisely to these hamlets and locales of Kurdish and other linguistic enclaves, when he recommended that ‘the efforts towards the changing of toponymes of non-Turkish or non-Turkified form should be completed swiftly’?
- 23 In the meantime, legal changes to facilitate the campaign were introduced. The ‘Law on the organiz (...)
- 24 In the following decade of military conflict and emergency rule, during which the provinces of the (...)
47With the legitimacy of the ‘scholarly’ Symposium in Ankara and a streamlined legal framework,23 the resuscitated Commission embarked for the last and final campaign of name changes. This time, it focused largely on the hamlets and settlements below village-status in the Southeast. With a cynical turn of events, many of these places were destroyed during the Kurdish insurgency, as their renaming continued.24 Arguably, the 1980s wave of name change was particularly brutal and to a large extent facilitated by the re-emergence of zealous bureaucratic elites under military tutelage, this time with a Turkish-Islamic rather than Turkish-secularist vision.
48Riding on the high-tide of ethnic purism, Balkan governments changed the names of their towns, villages and districts, particularly in the interwar years (Hart 1999; Mahon 1999). Turkey with its recurrent waves and campaigns stretching from 1915 to the 1990s is particularly instructive: Spontaneous efforts at renaming in the aftermath of WW I and the Greco-Turkish War; preparation of the ‘scientific’ infrastructure for large-scale renaming during the early Republic; systematic work of the Name Change Commission in the 1950s and 60s; and finally the ‘toponymical cleansing’ of the surviving pockets of linguistic diversity, framed by the witch-hunts in the aftermath of the 1980 coup. Twice, in 1915 and in the 1980s, the renaming of place names coincided with larger attempts at demographic engineering through ethnic cleansing and forced migration.
49The most striking insight, which we can gain from this history, is the role of the bureaucratic apparatus in the execution of the toponymical policy. Debates in the Turkish Grand National Assembly on the issue have been emotional at times and often contested. And indeed, early republican governments and administrators appeared overhasty in their zeal to purge the country of every reminder of the expelled Christian communities and thereby to prove their nationalist credentials. Yet, more often, notably during the rule of Menderes and Ozal, governments were reluctant to comply with the Turkification strategy and embarrassed by its excesses. It was the bureaucracy and the technocratic elites that took it on themselves to elevate the toponymic strategy to the level of state policy. Acting in a semi-autonomous space, bureaucrats would return to the task of Turkifying the time and space of Turkey the day after any military intervention. The resilience of the bureaucracy and its repeated empowerment by military interventions also explain the seeming paradox that the most aggressive episodes of the Turkification of place names occurred during the years of transition to democracy following interludes of military rule.
50The recurring waves of toponymical engineering were exceptionally destructive, as they successfully submerged what İnalcık called the ‘archaeology’ of the longue durée. In its constructive dimension, the policy led to the emergence of a Turkified toponymical order, which, however, is bereft of historical depth, symbolic meaning or emotional appeal. In fact, it would be fair to state that Turkey’s contemporary toponymical order with the exception of larger towns and cities is devoid of meaning per se, characterised by ‘dead documents that obfuscate the ethnic, social and linguistic influences of the longue durée, to rephrase Inalcik. Hence, even in its constructive dimension, it is the lacklustre imagination of bureaucrats that has acted as godfather for the new toponymical order. The consequence is a sterile, factually misleading and intellectually disappointing repetition of a limited number of beautified place names that do not correspond to the topographical, historical or linguistic structures they denominate. In this sense, the Republic’s virtually clandestine toponymical project did not redeem the older, imperial web of meanings with a new one. It destroyed the meanings of the former, obfuscated historical connections and ethno-religious patterns, but failed to replace it with an alternative sense of meaning.
- 25 Renee Hirschon discusses this loss of the common ground between the peoples of the Ottoman Empire (...)
51Indeed, there is now an official and almost exclusively Turkish order of place names in maps, newspapers, school books and public correspondence that has replaced and obfuscated the archaeology of the ‘ethnic, social and linguistic influences of the longue durée, which Halil İnalcık so passionately praised after his visit to the Congress of Onomastic Sciences in Salamanca. More than fifty years later, the knowledge of the old place names has largely dissipated. And so has the sense of societal awareness of diversity and multicultural sociability.25 To some extent, reminders of the pre-national order survive: Residents know the old name of their village and maybe those of a few neighbouring hamlets. However, their knowledge is less likely to stretch further afield. Some old place names also resurface in electoral rolls for the sake of accuracy. The thousands of Green Valleys, Happy Brooks and Pretty Mountains that indwell modern Turkey’s maps and registers are not trusted by the state that created them in the first place.
52Thanks to an ironic twist, however, the old toponymic order lives on in the archives of the Republic. As the naming of places was primarily a bureaucratic endeavour, the old toponymes are well documented and available as official documents with the seal of the Interior Ministry. This wealth of available documents, hence, also stakes out future research agendas. Re-establishing the etymology of human settlements and the historical, linguistic, religious and ethnic legacies they engender will become an indispensable aspect of the already growing field of local histories in Turkey. The recovery of old place names will also help to reconnect these local histories to Greek, Armenian and Syriac memories, hence re-establish the destroyed continuity the late-Ottoman notion of time and space. Beyond the sphere of local history, I would hope that some of the issues raised in this paper, especially the question of local and national actors and the specific workings of the Expert Commission for Name Change will be revisited in the framework of fresh research. Who were the members of the Commission, how did they work and what did they think about their mission? When, if not now should we take the scholarly content of Onomastic research seriously and use its possibilities for a post-national reconsideration of Turkey’s multiple, if denied legacies.
53On the conceptual side, I have tried to show that the analytical tool of demographic engineering can be used to push the debate further towards acknowledging and exploring the ontological relationship between destruction ex ante and discriminatory construction from a social-constructivist and integrative perspective.
54Despite the new insights, which the demographic engineering approach may bring, however, it should not be taken as an easy way out of the contested and polarised debate on the destructive content of transitional periods between imperial dissolution and national ‘re-emergence’. A whole set of constraints needs to be taken into account. The approach is, by definition, state-centred and hence focuses on policies and actions of government agencies and the bureaucracy. The primary site of research is the state, its departments and archives, and the large-scale impact of their actions. Individual suffering is hard to accommodate within this macro-perspective, and so is the experience of the communities that have been written out of the official narrative. I therefore believe that demographic engineering is not an alternative to the contested debates of ethnic cleansing or political mass violence, to whose usage much higher opportunity costs are attached. It is, however, a complementary tool in the study of contemporary Turkish history that highlights the paradoxical binary relation between policies of destruction –here the state-directed movement and excision of ethnic groups – and strategies of construction –such as the invention of a new toponymical order – which have shaped the time and space of the Turkish nation-state.
RSS feed
