Navigation – Plan du site

Village Evacuation and Reconstruction in Kurdistan (1993-2002)

Joost Jongerden
p. 77-100


Résumé :
Le monde rural a souvent été le théâtre de luttes armées et de contre-insurrections. Pourtant on a prêté peu d’attention à ce phénomène. L’observation des transformations récentes des campagnes du Kurdistan turc permet de discuter des stratégies de contre-insurrection turque, d’un point de vue à la fois militaire et politique ; de l’évacuation et de la destruction des villages lors des conflits ; et des plans de développement relatifs à la zone en guerre. Cette étude révèle que les plans de reconstruction et de développement mis en œuvre par le gouvernement turc prolongent la guerre, mais en usant d’autres moyens.

Haut de page

Texte intégral

1IN THE COURSE OF THE WAR against the PKK in southeastern Turkey during the 1990s, Turkish Armed Forces and paramilitary “village guards” systematically evacuated and destroyed some 3,000 rural settlements. This destruction was enacted as an integral part of warfare following a Turkish revision of its military tactics and general approach to the conflict. In this article I will discuss the destruction of rural settlements as part of a counter-insurgency strategy implemented by the army with the intention of transforming the countryside. The article will argue that the rural space of a huge swathe of territory was reconstructed, from one which provided sustenance for the enemy, the PKK, to one which immobilized the enemy at the expense of millions of rural inhabitants, their livelihoods and their communities.

2Though much has been written about the war between Turkey and the PKK from socio-political perspectives [Imset 1995; Barkey and Fuller 1998; White 2000; Bozarslan 2004], with a few exceptions [Kocher 2004; Etten et al. 2008] the way in which the countryside has been used as a medium for warfare and transformed by the warfare itself has received comparatively little attention. The relative neglect of the rural aspects of the war, in terms of state planning and its human consequences, corresponds to a wider trend in research, at least in Turkish and Kurdish studies, to focus on urban issues, in particular the impact of public policy on the major population centers as the hubs of political and social life. Scholarly attention with respect to the Turkish state in its war with the PKK has focused on the transformation of cities and livelihood issues of forced migrants (internally displaced persons: IDPs) in their new urban environments, especially in some of Turkey’s principal cities, Istanbul, Mersin, and Ankara, as well as Diyarbakır, the main city in the Southeast [Çelik 2005; Yılmaz 2006]. Understandable as it may be that research followed the displaced into the major conurbations, the consequence is that the impact of displacement on the countryside itself and the struggle of those ejected from their homelands has been understudied.

3Internationally, however, the relationship between counter-insurgency and the production of rural space has been a developing field of research. Counter-insurgency campaigns undertaken by the British in Malaya in the early 1950s, the French in Algeria in the second half of the 1950s, and the US in Vietnam at the end of the 1950s and beginning of the 1960s were all concerned with, and have been studied as, a reconstruction of rural space [Zasloff 1962-1963; Donnell 1970; Sutton 1981; Newsinger 2002]; so too has the background to the conflict in Sri Lanka been analyzed in terms of rural space and urban/rural and center/periphery dichotomies [Moore 1984; Woods 1990]; while more recent studies of the Palestinian case have showed how political control is exercised through spatial management [Signoles 2005], or, more particularly, spatially embedded [Weizman 2007].

4Following such analyses of counter-insurgency practices, this article will not only discuss the evacuation and destruction of villages, but extend its enquiry into plans and activities for the region developed by civilian authorities. The aim and activities, it will be argued, was to reconcile military operations with rural development. To this end, a master plan for the reconstruction of the region affected by village evacuation and destruction will be discussed. It will be argued that the scheme for reconstruction, even though opposed by the military, was developed from a similar logic of control and supervision as that underlying the initial evacuation and destruction of rural settlements. In proposing the development of a new settlement structure, this plan envisaged a transformation of rural space on the basis of schematic categories based on control and supervision from outside and above. As such, it will be concluded, the destruction associated with war and the reconstruction associated with peace, far from contradicting each other, have actually been complementary.

5The main actors in the conflict discussed are the Turkish Armed Forces and the PKK. The Turkish Armed Forces (Türk Silahlı Kuvvetleri), the organizational framework for the state military, includes, among other branches, the army and the gendarmerie (jandarma), a quasi-military police force operative in mostly rural areas (i.e. in areas not covered by the regular police). The Turkish Armed Forces assumed a political role of the guardian of the state and took the leadership of the country with periods of coups and military rule [Cizre 1997: 151-166].

6The PKK is the Kurdish acronym for the Kurdistan Workers’ Party (Partiya Karkerên Kurdistan), whose guerrilla led an armed insurgency in the southeast of Turkey from 1984 (it has mostly been observing a self-declared ceasefire during recent years). Formally established in 1978, the PKK emerged from a student and urban environment during the period 1973-1977. The founding members of the party met in Ankara as housemates, in university classes and in left-wing student and youth organizations. Initially the group was organized as a fluid network. The junctions in the network were rented houses, which were changed frequently, and in which meetings, recruitment, and ideological group formation took place [Jongerden and Akkaya 2010]. In 1977, this group – a party had not yet been established – decided to establish itself in the Kurdistan Region in the southeast of Turkey, from which it planned to develop a revolutionary struggle. A military coup in Turkey in 1980 and the severe repression unleashed obstructed PKK plans, resulting in a retreat to outside the borders of Turkey. However, the retreat was temporary. During the course of the 1980s, the PKK sent militants educated in guerrilla warfare in northern Lebanon back to the Kurdistan Region in Turkey, and started to build up forces in the countryside. In this build-up of forces, villages in the countryside became for the organization what the houses had been in the urban environment: the nodes in an extended network. They were both stepping stones for swift movement as well as resources, the villages now providing the guerrilla with shelter, logistical support, recruits and intelligence.

7The PKK, it will be argued, used an insurgency strategy which involved the progressive development of spatial control and counter-institutions. Though the PKK had a well-organized network of militants in the towns and cities in the region, the build-up of forces took place where the state was weakest, i.e. the peripheral countryside. The party aimed at taking over the rural areas first, with the towns and cities to follow only in the final stage of the war. Initially, the Turkish Armed Forces decided just to defend the larger villages and towns. In doing so, they allowed the PKK the space to establish a guerrilla network throughout the Southeast, and unintentionally contributed to the fulfillment of the PKK strategy. However, the PKK’s dependence on this space also was, this article will argue, the guerrilla’s weak point, and spatial deprivation its Achilles heel. The Turkish state’s evacuation and destruction of villages was not simply punishment, nor a mindless response, or collateral damage. On the contrary, it was constitutive of the counter-insurgency strategy of the Turkish military to change the space of combat to its favor [Özdağ 2003; Pamukoğlu 2003].

8Village Evacuations

9As part of its counter-insurgence operations, Turkish Armed Forces evacuated and destroyed rural settlements on a large scale. According to official figures, 833 villages and 2,382 small rural settlements, totalling 3,215 settlements, were evacuated and destroyed in 14 provinces in the East and Southeast, namely Adiyaman, Ağrı, Batman, Bingöl, Bitlis, Diyarbakır, Elazığ, Hakkari, Mardin, Muş, Siirt, Şırnak, Tunceli and Van [Oyan et al. eds. 2001]. In these provinces, the total number of rural settlements (villages and hamlets) had been 12,737 [Doğanay 1993: 6-7]. In other words, around a quarter of all rural settlements in the east-southeast region of Turkey were emptied. Numbers provided by the Human Rights Association (HRA) in Turkey and the Kurdish Human Rights Project (KHRP) suggest that most evacuations occurred in the period 1991-1995, peaking in 1993-1994 (Table 1 p. 80).

  • 1 Human Rights Watch, “Displaced and Disregarded. Turkey’s Failing Village Return Program”. Washingto (...)
  • 2 Kurdish Human Rights Project, “Internally Displaced Persons. The Kurds in Turkey”. London, 2002. Av (...)

10The approximate number of settlements evacuated and destroyed is not really in dispute, but the number of people affected has been a subject of great controversy. Government sources are extraordinarily precise. They report that 384,793 people were evacuated during the 1990s. Human Rights Organizations, however, claim that Turkey deliberately presents low numbers to camouflage the magnitude of the displacement,1 and have estimated the number of displaced at as high as 3 to 4 million, i.e. ten times the government figure.2 Other calculations tend more towards 1.5 million [Aker et al. 2005: 8] or put the figure between 950,000 and 1,200,000 [Tezcan and Koç 2006]. Since reliable statistics are not available, the number of displaced persons is necessarily a rough estimate.

Table 1. Evacuated and Destroyed Villages, 1991-2001






















Source: Kurdish Human Rights Project, “Internally Displaced Persons: The Kurds in Turkey”, 2002.

11The resettlement of the rural population did not take the form of a scheme, in the sense of an elaborate and systematic plan of action encompassing the provision of shelter and the reconstruction of livelihood and/or granting of compensation, and for the execution of which specific personnel and resources were allocated. Rather, the evacuation of villages was organized in the form of what I term a “rural-to-urban resettlement tracks”, defined as (multiple) routes from rural to urban settlement entities along which people were forced to move, without support or assistance from authorities [Jongerden 2007]. In contrast to a scheme, tracked resettlement is little more than a collection of multiple routes from hamlet and village to town and city.

  • 3 The village guards are a militia, legally established in 1985 and which became operational from 198 (...)

12The evacuation and destruction of villages was haphazard but orchestrated. During my fieldwork in the region, forced migrants in Diyarbakır explained how the process of evacuation and destruction had taken place in their village. At first, the army put pressure on the villagers to become village guards.3 They refused. Some of the villagers I spoke to asked rhetorically, and rather shrewdly:

The state is our state, and those in the mountains are our brothers – how can we choose between the state and our brothers?

13As a consequence of the villagers’ refusal to become part of the militia, the military evacuated the village. After that, village guards from a nearby village moved in and plundered the place. Wheat stocks and animals were stolen, window frames and doorposts broken out of the houses and taken away [Jongerden 2007: 253-254]. There are many narratives like this, and, taken together, they reveal a pattern. A typical village evacuation would proceed as follows. Villagers would be put under pressure to join the so-called village guards. If they refused and did not leave their village after warnings and pressure, then regular soldiers and/or special teams would one day enter the village and order the village chief (the muhtar) to gather the inhabitants and evacuate the settlement. Sometimes villagers were given the opportunity to collect their belongings, but often the village was plundered or soldiers would begin firing at the houses and set fire to them together with their contents. Livestock would be stolen or shot, orchards and crops burned. The villagers would take refuge in a nearby town, later moving to a main city in the region (such as Van or Diyarbakır), from where many would continue their journeys to cities in the West (such as Adana, Mersin, and Istanbul).

  • 4 Human Rights Watch, “Displaced and Disregarded. Turkey’s Failing Village Return Program”. Washingto (...)

14The concern of the military was to clear the villages. They were not concerned with what happened to the people after eviction, and in the towns and cities the hundreds of thousands (or millions) of displaced were left to their own devices.4 Most found a temporary place to stay first and long-term shelter after that, mainly through chain-migration and self-help. The chain-migration mechanism implied that the evacuated selected urban centers which had already been established as settlement destinations by their relatives or hemşehri (people from the same place of origin). The self-help method implied that they would re-establish themselves in the urban entities through informal support networks.

A Rural Insurgency

  • 5 In Dersim/Tunceli, for example, the Teacher’s Training School was a recruitment focus of the moveme (...)

15In the years preceding the evacuation and destruction of villages, the PKK had established itself in the countryside. It was at a meeting in Ankara at the beginning of 1976 that the cadres of the organization decided to start a process of party-building in the Kurdish Region. This process was initiated in urban areas elsewhere, in particular at centers of higher education.5 Then, in 1977, a decision was taken to prepare for a prolonged people’s war, although the execution of this strategy only followed after the military coup d’État of 1980. Following the strategy developed by Mao, the PKK envisaged a three-stage struggle: strategic defense (armed propaganda, small scale attacks, mobile warfare), followed by strategic balance, and finally a strategic offensive [İmset 1995; Özdağ 2003].

16The three-stage model involved a move from guerrilla to conventional warfare. The shift from guerrilla to mobile warfare was considered necessary to annihilate the enemy’s manpower and to liberate land. In combination with a popular uprising, the guerrilla army was supposed to force the Turkish army to leave all the Kurdish land it was occupying, i.e. Kurdistan in Turkey (as the PKK saw it) [Jongerden 2007]. According to the theory of asymmetric warfare, such a final battle is only to be launched once the enemy has become bogged down and demoralized, psychologically prepared for defeat [Beckett and Pimlott eds. 1985: 59].

  • 6 Theoretically, a unit was composed of a commander, a vice-commander, a political commissary, someon (...)

17In the early 1980s, small armed units composed of 3 to 5 militants6 had moved through the region, spreading propaganda; by the beginning of the 1990s, roughly from 1990 to 1992, the PKK established control over much of the countryside in the (primarily) Kurdish region of Turkey. In PKK terms, these were considered “semi-liberated zones”. In these areas – large parts of the provinces of Hakkari, Van, Şırnak, Siirt, Batman, Diyarbakır, Bingöl – the PKK established a permanent presence. The region was essentially controlled by a network of guerrilla units. These had their own local bases but were also in regular (near daily) contact with other local guerrilla forces, and sometimes even lodged them in the many small rural settlements in the area. These hamlets and villages provided not only shelters, but also intelligence, recruits, and food supplies. These were the material and immaterial components of the network, allowing swift and informed movement throughout the region.

18The decision of the PKK to build up forces in the countryside was based on the party’s analysis of the state as being weak in rural areas, which was certainly the case. Although Turkey’s state system was marked by centralization and authoritarianism, it was in practice rather detached from large parts of the countryside which were outside its control [Jongerden 2009]. Indeed, many villagers only knew the state from occasional patrol missions by units of the gendarmerie. The relative absence of the state in rural areas, in particular the more remote and mountainous ones, and the antagonistic social relations between the state and the Kurdish population, made the countryside a good “site” for guerrilla warfare [Jongerden and Akkaya 2010].

  • 7 Devlet Planlama TeŞkilatı (State Planning Organization), “Türkiye’de Yerlesme Merkezlerinin Kademel (...)

19In 1982, at a time when armed guerrilla units of the PKK moved through the mountainous border region with Iraq to mobilize support, avoiding direct confrontations with the state, a study by the State Planning Organization put squarely on record the weakness of the state’s spatial-administrative system. The study showed that many of the rural settlements in Turkey, and not only in the Kurdish Region, were effectively outside the formal administrative system, and simply self-governing.7 This was an important reason why the PKK was initially able to move around rather freely. The state was weak in the rural areas, and had no control over smaller settlements.

20The countryside provided good cover and concealment (typically mountainous and forested areas) as well as practical support for the PKK from the local population; it afforded a physical and social environment that operated as a force-multiplier for those conducting a war of movement. For the state forces, on the other hand, the population being dispersed, the physical terrain rugged, and the communication and transport infrastructure poor combined as factors of a force-diffuser. The military were “diluted” in the territory [Kocher 2004]. This was clearly a serious disadvantage for the supervision and control of the population and effective combat against the guerrilla.

Diffuse Warfare

21Unintentionally, the army’s combat strategy contributed to the success of the PKK [Özdağ 2003]. The Turkish military took up defensive and static positions, especially at night, when soldiers were thought to be safe in their enclosures. Garrisons were built and fortified, and army units confined themselves to these garrisons. Operations were carried out, but units returned to their barracks before dusk. The Turkish Armed Forces had decided to concentrate on the defense of larger settlements and refrain from night-time operations, and thus did not hinder the PKK from establishing and even tightening control over the smaller settlements, or from moving by night. Every now and then the army would organize large sweeps, sending tens of thousands troops into an area, but these actions were not very effective as the guerrilla escaped into hiding while the troops were being massed and stayed away during the operation only to return after its completion and the troops’ retreat. In spite of its numerical superiority – in 1993, a state force of 185,000 (excluding the gendarmerie and village guards) was pitted against some 15,000 to 20,000 guerrilla fighters – the Turkish army rapidly lost control [Özdağ 2003; Pamukoğlu 2003].

  • 8 In fact, the milis constituted a major part of the total PKK force. In his study of the PKK, P. Whi (...)

22American military sources ascribed the poor performance of the Turkish army during the 1986-1993 period to the lack of an integrated strategic counter-insurgency doctrine. The military strategy of the Turkish army was based on the doctrine of static territorial defense by a standing army against a coherent mega-threat (i.e. an invading Soviet army), centrally organized and clearly identifiable. The war with the PKK did not fit these doctrinal remnants of the Cold War. On the contrary, the PKK guerrilla specifically avoided classic, pitched battles. Instead, the PKK’s war tactics were characterized by what may be termed call “poly-centricity”, “multi-directionality” and “multi-dimensionality” [Hardt and Negri 2004: 69-92]. They were “poly-centric” in that the “people’s army” did not have one single center, but was composed of innumerable relatively independent units, clustering with and separating from other units whenever deemed necessary; “multi-directional” because the guerrilla army was difficult to locate, its militants permanently on the move and giving the impression of being able to attack at any time and any place; and “multi-dimensional” since not all party militants were full-time guerrillas, but were active as milis, hidden militia in both rural settlements and urban centers that could shift swiftly between the roles of fighter and civilian (the milis had responsibilities for the provision of logistical support and intelligence, and were often active in civil society organizations, such as trade unions).8

  • 9 The concept of “dromopolitics” comes from P. Virilio [2006], who emphasizes the importance of the m (...)

23The army learned that a counter-insurgency war could not be fought from a defensive and static position. Therefore the Turkish Armed Forces developed a new strategy, the field domination doctrine, and new tactics on the ground. The field domination doctrine had been laid down in 1991, but not put into full practice until after the reorganization of the army, initiated in 1992 and completed in 1993. In this new conceptualization of warfare, the army did not adapt itself to the rural boundaries as determined by the rural geometry, but reorganized the rural syntax in accordance to its own ideas. It started to consider the countryside not as a site where the fighting took place, but as a medium for warfare, to be shaped and changed by “micro-tactics”. A central objective of the field domination doctrine was to bring about a contraction of rural space. This contraction, annulling remoteness, inaccessibility and distance, was realized through a combination of geopolitics and dromopolitics.9 While the latter took the form of a focus on military mobility and rapid response, the former was based on the evacuation and destruction of rural settlements as a means to “smooth” space.


24The Armed Forces were suspicious of the villagers in so-called “PKK populated areas” [Özdağ 2003: 33], and enforced firm measures against the rural populations. In addition to curfews, villages were put under food embargoes, allowed only to bring limited amounts of flour, rice and other food products into their settlements. A man from Siirt described the situation thus:

Economic sanctions are applied against our village. [...] For example, a family of 4 to 5 persons will be allowed to buy 50 kilograms of flour. If you want to buy food, you first have to go to the gendarmerie. After you have identified yourself and they have checked the number of people in the household, you get a form which states how much you can buy of what products. You take the form with you when you go to the city, and when you return they check if the amounts are in accordance with what is written on the form. If you had more with you than written on the form, the goods were confiscated and you were taken for interrogation [Jongerden et al. 1997: 44].

  • 10 The evacuation of rural settlements was made possible in 1987 by Decree No. 285 concerning the stat (...)

25The Armed Forces started to systematically evacuate and destroy villages considered supportive of the PKK at the beginning of the 1990s.10 This came increasingly to be regarded as a productive strategy. The army inverted the rationale of guerrilla warfare as formulated by Mao Tse Tung:

  • 11 Selected Works of Mao Tse-Tung. Vol. IX (1937). Available at (...)

Because guerrilla warfare basically derives from the masses and is supported by them, it can neither exist nor flourish if it separates itself from their sympathies and cooperation.11

26The logic employed by the military in Turkey was, conversely, to force this separation. The evacuation of the population was intended to isolate the guerrilla forces from the population and deprive them of finance, intelligence, food, shelter and recruits. The evacuation of rural settlements reduced opportunities to cooperate with the guerrilla.

27The military openly hinted at a large-scale evacuation of rural settlements and the concentration of the population in a few large settlements. In the words of General Osman Pamukoğlu, who headed the fight against the PKK in Hakkari:

Where there is sea, there are pirates. In this province [Hakkari] [there] are 674 villages and hamlets. These settlements form the spider’s web in which the PKK feeds itself. [...] Why don’t we concentrate all [the villagers] in two or three main settlements? [Pamukoğlu 2003: 59]

28With no major resettlement schemes organized and the rural population forced to leave their houses and flee to urban settlements, rural depopulation took place on a massive scale. In the Kurdish region administered under the state of emergency regulation, a huge rural-to-urban migration took place. According to Matthew Kocher:

The region of emergency rule urbanized at nearly a three times the rate of the rest of Turkey [while] the rural population declined by 12% in absolute terms, at a time of torrid population growth [2002: 41].

  • 12 “Tunceli” is the modern, official (Turkish replacement) name for the province of Dersim, the old (Z (...)

29If we break down the numbers and focus on provincial and district levels, we see that the rural population in some provinces and districts with relatively high numbers of mountain villages, such as the districts Lice, Kulp, Hazro, Kocaköy and Dicle (in the province of Diyarbakır), dropped by tens of percent [Kocher 2002: 8]. Large areas of the countryside were completely depopulated. Using satellite images, a case study for Dersim/Tunceli12 suggests that village evacuation and destruction was often accompanied by widespread forest-burning [Etten et al. 2008: 1795] in order to smooth out space and destroy potential guerrilla hideouts.

30The evacuation of villages should not be considered a side-effect of the counter-insurgency of the Turkish Armed Forces – notwithstanding the gradual and undeclared way in which it was introduced and extended – but one of its primary constituents, intended to contribute directly to the “environmental deprivation” of the guerrilla. It was a means of destroying the guerrilla’s physical and social environment, intended to force the guerrilla either into isolated retreat (high in the mountains, across the border with Iraq) or else into undesired, hastily planned combat in “urban” environments. In other words, space was not a background for the actions of the Turkish Armed Forces, as it had been in the initial phase of the war (up until 1991), not an abstract grid on which events occurred, but rather, as Eyal Weizman [2007: 7] puts it, “The medium that each of their actions seeks to challenge, transform or appropriate”. By reorganizing rural space, the army intended to establish control, and suppress the insurgency.

31The smoothing of rural space by means of village evacuation and destruction was a necessary step for the army to achieve the desired spatial contraction, but not enough. Permanent, partly-manned checkpoints as well as temporary, mobile checkpoints were established to divide the space and create a matrix of control. Though checkpoints may initially emerge as a series of tactical responses of military officers, they can assume an overall strategic lay-out [Weizman 2007: 146]. The matrix of control enabled the supervision and regulation of movements of people and goods. Young men and women whose identity cards showed that they were from another area were questioned, the types of goods taken into (or from) an area were checked, and the quantities subject to limitations in order to cut off supplies to the PKK [Marcus 2007: 222].

  • 13 Actually, the southeast of Turkey had been ruled under martial law and emergency regulations since (...)

32It is important to note that such measures as village evacuation and destruction, the burning of belongings and the environment, and restrictions on the movement of people and goods only are possible in a context of weak citizenship. Beyond the already weak sense of state (see above) – in the sense of a relationship to higher (civil) authorities to which recourse was even conceivable (let alone practically accessible) in the minds of villagers – the implementation of the state of emergency (Olağanüstü Hal: OHAL) in the Southeast created what Weizman [2007] names a zone of weak citizenship, i.e. a region where civil rights are restricted.13

  • 14 The spatially-grounded state of exception in its extreme form was the kill-zone. In a kill-zone, hu (...)

33A raft of measures was employed to restrict political activities in the OHAL region during this period. In addition to routine military and police activities (including the system of checkpoints), freedom of the press, freedom of expression, and freedom to organize were practically abolished in the Southeast. Daily newspapers available outside the OHAL region were banned inside it, events and demonstrations allowed elsewhere were forbidden, and many associations forcibly disbanded. Politically active public sector employees were subject to exile outside the region. And it was in this political environment of weak citizenship that the Regional Governor was granted strong powers, among them the authority to evacuate rural settlements if he was advised that this would render the region more secure. In other words, the OHAL was a spatially-grounded state of exception.14

Organization of Mobility

  • 15 Ministry of Defense, “Defense White Paper”, Ankara, 2000.

34Following the new doctrine, the General Staff decided to reorganize the army, shifting from a cumbrous divisional and regimental structure (designed in the period of the Cold War) to a relatively flexible corps and brigade structure (to fight a multi-directional and multi-dimensional internal enemy), which was supposed to contribute to a more rapid response and higher mobility.15 Prior to the army reorganization, the principal tactical units had consisted of 16 infantry divisions and 1 armored division, plus 23 independent brigades, of which 6 were armored and 4 mechanized. Basically, the Turkish land force was a large but badly equipped infantry force. Under the reorganization, all except 3 of the 17 infantry divisions were dismantled. The existing 9 corps were retained, with brigades directly responsible to the corps commanders and more flexible in their modus operandi.

35The Armed Forces abandoned the approach of garrison line-of-defense as its primary military tactic. As General Osman Pamukoğlu, commander of the Hakkari Mountain Warfare and Commando Brigade between 1993 and 1995 wrote:

Garrisons do not provide protection [...] but because of their static disposition are targets for the enemy [2003: 137].

36As part of the new strategy of spatial control, Commando brigades and Special Forces became key elements in the war against the PKK. Their soldiers were supposed to stay in the field day and night, searching for PKK units and camps. In addition, Special Action Teams (Özel Haraket Timleri), under the authority of the police, and Special Teams (Özel Tim), under the authority of the gendarmerie (these with a combined total of 15 to 20,000 men), were created as rapid response units.

37The old approach of positional warfare was not entirely abandoned, insofar as the job of the garrison was farmed out to civilian militia. Crucially, however, these were located in small settlements, in the villages as opposed to towns, thus extending the depth of spatial control. The system of village guards, Korucular, was established in conjunction with the abandonment of the old garrison system, instituted as the Temporary and Voluntary Village Guards (Geçici ve Gönüllü Köy Korucuları) – somewhat of a euphemistic misnomer – and incorporated into the organization of the Turkish Armed Forces.

38Numerical superiority was considered a key element in the establishment of spatial control. Between 1993 and 1995, the number of Turkish troops and paramilitaries in the region was almost doubled, from 185,000 to 360,000 men (including as many as 70,000 village guards). In Hakkari, for example, the land forces in 1995 comprised fourteen battalions amounting to 56,000 soldiers, with five battalions employed in a war of movement. These five battalions were organized into special units of “go-getters” who lived in the mountains for weeks, hunting the guerrilla. The other nine battalions had regular tasks to fulfill, generally involved in patrolling an assigned area. When PKK guerrillas were spotted in the matrix, the mission of regular units was not to engage into contact, but to surround the guerrilla. Special units were transported to the area by helicopter and took up the pursuit. The field domination strategy of the Turkish Armed Forces did not mean that the army had always to be everywhere, but rather that it could be anywhere at any time. The smoothing of space and rapid response facility were the necessary conditions for this.

39Rapid response is dependent on reliable intelligence, good communication and infrastructure, and means for the transport and movement of troops. In order to secure timely and accurate intelligence, it was emphasized that the army had to show it was in control. The Armed Forces improved the network of roads and communication facilities, and modernized its armament and transportation for a more effective employment of troops over land and by air [Özdağ 2003: 53]. This included the purchase and use of helicopters, enabling a speedy deployment of commando and infantry troops across difficult terrain, and new armored infantry fighting vehicles and armored personnel carrier vehicles.

A New Reality on the Ground

40The Turkish military’s new geopolitical tactics (the intersection of space and power) and dromopolitical tactics (the intersection of pace and power) profoundly changed the rural space of war, from a network of support for the guerrilla into a matrix of control by the army. That the situation on the ground had changed considerable was illustrated by the words of a former PKK commander:

The evacuation of the villages really helped the state. [...] The villagers provided everything for us, supplied materials and information. When the villages were emptied, all this was taken away from us [Marcus 2007: 222].

41The PKK units were henceforth seriously hampered by logistical problems. The countryside was emptied, and fighters could not move easily without being spotted. It was not only that resources had been lost, but also that the state’s mobile units were in hot pursuit. The army shook off its cumbrous inertia. Its forces continuously tried to engage in fighting, entering areas they had never entered and which the guerrilla considered a safe sanctuary. Ironically, however, just as the army had contributed to the success of the PKK strategy in the initial stage of the war, so did the unfolding of the PKK strategy contribute to the army’s success in the period after 1993.

42At its height, the PKK had established hegemony in large parts of the Southeast, in particular in Serhat (covering parts of Hakkari, Van and some territories in Iraq and Iran), Garzan (covering parts of Batman, Siirt, Van, Hizan and Gevaş), Botan (covering parts of Şırnak, Hakkari, Van, Siirt, Eruh and parts of Northern Iraq), and Amed (covering parts of Diyarbakır, Bingöl, Genç, and Muş). Although the PKK did not establish permanent control, it succeeded in preventing the security forces from entering or remaining in these areas for long periods of time. These areas were considered to be one step away from the establishment of so-called “liberated zones” and had the status of “semi-liberated zones” (marking a transition from the phase of “strategic defense” to that of “strategic balance”).

43In June 1993, PKK leader Murat Karayılan confidently announced that a so-called Parliament of Kurdistan would be established in the liberated area of Botan on the Turkish side of the border with Iraq. In line with the objective of creating and defending liberated areas, the PKK started to move to conventional warfare tactics. Larger units were created which more often moved in combination, in groups of up to several hundred guerrilla fighters. At times, the PKK resisted with positional warfare. In other words, by 1993 the army had started to apply guerrilla tactics, while the PKK was preparing for conventional warfare [Özdağ 2003]. Guerrilla commanders, confronted with heavy losses, wanted to break up back into smaller units and dilute forces, but the PKK leadership, keeping to its overall war plan and most likely not properly appreciative of the change in its enemy’s, insisted on a regular army strategy [Marcus 2007: 223].

Warfare by Means of Reconstruction

44A top-secret letter written in 1993 by President Turgut Özal revealed some of the basic characteristics of the counter-insurgency strategy that began to emerge in Turkey. Turgut Özal had been Prime Minister of the Republic of Turkey between 1983 and 1989, the period in which the PKK had developed into a force capable not just of causing the Turkish state problems, but of actually winning the struggle for independence from Ankara and achieving the establishment of Kurdistan as a state. In 1989, Özal ascended to the presidency, even as the PKK was rising to the height of its power, effectively controlling great swathes of supposedly Turkish territory. He knew as well as anyone how grave the situation had become. Shortly before his sudden death on the 17th of April 1993, the president wrote a secret letter to the then Prime Minister, Süleyman Demirel, proposing a solution to the PKK insurgency. In this letter, he emphasized the severity of the threat posed by the PKK:

  • 16 Kurdish Human Rights Project, “Internally Displaced Persons. The Kurds in Turkey”. London, 2002, pp (...)

In the Southeast [...] we are faced with perhaps the most significant problem in the Republic’s history.16

45Turgut Özal outlined not only the need, but also the main lines of the new war strategy. Part of that strategy was, indeed, village evacuations – but a second stage also was concerned with what should be done after these villages had been evacuated and destroyed. Özal made the following suggestion:

Starting with the most troubled zones, villages and hamlets in the mountains of the region should be gradually evacuated. [...] With the evacuation of mountain settlements, the terrorist organization (PKK) will have been isolated. Security forces should immediately move in and establish complete control in such areas. To prevent the locals’ return to the region, the building of a large number of dams in appropriate places is an alternative.

46More than fifteen years since Özal wrote this letter, in Dersim/Tunceli and Hakkari, two mountainous areas that continue to be hotspots of the conflict, where PKK guerrillas are even now able to find shelter and move, preparations for the construction of two series of dams are being made. Eight dams have been planned for the Tunceli region and eleven for Hakkari, which will turn large parts of the countryside into artificial lakes. The construction of dams in Dersim/Tunceli is purportedly intended to contribute to development, but the local population is convinced that the dams are to be built to evict them [Ronayne 2004: 47]. In Hakkari the dams are to be constructed in an area largely evacuated in the 1990s. The dams will be built wall-like, with the express purpose of making it difficult for PKK guerrilla-fighters to penetrate Turkey’s borders [Jongerden 2010]. According to the authorities, the mountainous region, with its many caves, contains many cross-border trails. Elongated dams are to be constructed on such sites, in ribbon formation. With preparations for their construction already underway, this project will replace earlier ideas for a five-meter high concrete wall along this border. The dams are strikingly visible markers of the transformation of rural space taking place in the southeast of Turkey.

47In his letter, Turgut Özal also referred to the development of cities into centers for population growth in the region. In this respect, he made the following suggestion:

Such cities as Adiyaman, Diyarbakır, Urfa, Mardin, Siirt, Elazığ, Malatya, Erzincan, Erzurum, Kars, Ardahan, and Iğdır [...] must be turned into centers of attraction for the local population currently settled in the countryside.

  • 17 Meclis Araştirmasi Komisyonu, “Doğu ve Güneydoğu Anadolu’da Boşaltilan Yerleşim Birimleri Nedeniyle (...)

48Actually several plans were made to scheme the urban resettlement of rural populations. Two examples are the 1994 toplu kondu and toplu çiftlik plans. Developed by the Housing Development Administration of Turkey (Toplu Konut İdaresi Başkanlığı), the toplu kondu (collective shelter) project was designed as a hybrid of the shantytown type dwellings known as gecekondu (literally “night shelters”, implying that they are put up overnight, under cover of darkness), and the government housing program toplu konut (collective housing). In the toplu kondu blocks, the government was to provide basic, one-floor shelters of 50 square meters on pieces of land varying in size between 200 and 400 square meters. Inhabitants were supposed to construct additional floors, extra rooms or workplaces according to their own needs and means. Several of these toplu kondu blocks, with a total of 8,000 shelters, were planned near economic growth centers of the cities Adana, Urfa, Diyarbakır and Gaziantep, and another 2,000 shelters in the proximity of several district towns in the provinces of Adana, Urfa, Diyarbakır and Gaziantep.17

  • 18 Turkey had requested a loan of 50 million dollars from the World Bank for the toplu kondu plan and (...)

49A similar plan proposed by then Prime Minister Tansu Çiller envisioned the construction of large settlements with an average population of 1,000 inhabitants each. These settlements were referred to as toplu çiftlik modeli projesi (collective farm model projects), or simply toplu çiftlik (collective farms). Despite its name, the toplu çiftlik plan was meant to resettle people in the vicinity of urban areas. These two schemes were not implemented, but a few projects incorporating ideas embodied in the toplu kondu and toplu çiftlik projects have been realized, as out-of-town developments near urban centers. Examples are the construction of Beşyüzevler, near the city of Diyarbakır; Doğankent, near the provincial center of Hakkari; and Kandolar Mahallesi, Afet Evleri (literally “Kandolar Quarter”, “Disaster Houses”), also known as “80 Evler”, near the district town of Ovacik, in the province of Tunceli. Neither of these schemes – toplu kondu or toplu çiftlik – were implemented on a large scale, since Turkey failed to find the funds needed.18

50A year later, in 1995, the issue of village return entered the political agenda, when a coalition government of the True Path Party (Doğru Yol Partisi: DYP) and the Republican People’s Party (Cumhuriyet Halk Partisi: CHP) proposed a return-to-village program in the context of the Southeast Restoration Project (Güneydoğu Onarım Projesi: GOP). Not much information is available about this program, although it was reportedly a blueprint for the organization of a gradual return to those evacuated villages where security could be provided. Politically significant was its acknowledgement that people from villages which had been evacuated and destroyed (by the army) might be eligible for return. The idea was opposed by the military and governors in the region, who claimed that a re-habitation of evacuated villages would create a security risk. In 1997, Mesut Yılmaz, leader of the Motherland Party (Anavatan Partisi: ANAP) and the then Prime Minister, also announced that his government would support the return of evacuated populations to their villages. Although return was again made conditional on the ability of the army to provide security, the implicit, and fundamental, purport was that the evacuation of the countryside was an anomaly, and return inescapable.

  • 19 At the time of the study, Prof. Dr. Oğuz Oyan was president of the Turkish Social Sciences Associat (...)
  • 20 GAP has now expanded to become “a multi-sector and integrated regional development effort approache (...)

51A comprehensive master-plan for return was drafted in 2001 under Prime Minister Bülent Ecevit, leader of the nationalist Democratic Left Party (Demokratik Sol Partisi: DSP). Research for the plan was conducted under Professors Oyan and Ersoy.19 It was coordinated by the Southeast Anatolia Project (Güneydoğu Anadolu Projesi: GAP), an organization originally established to coordinate the construction of dams and related large infrastructural works.20 As a regional administrative body, GAP had experience in the preparation and implementation of large-scale infrastructural programs at the regional level. Moreover, GAP administration had experience with resettlement projects. For the construction of eight dams in the region, 336 settlements had been or were to be evacuated, affecting, according to government statistics, about 180,000 people living in approximately 25,000 households. Finally, in response to the rise of the PKK, GAP had been given the responsibility to, in effect (but, I argue, quite intentionally), contribute to the combat through rural development projects.

52A close look at the Village Return and Rehabilitation Development Plan shows that it was organized as a hierachization of space in order to facilitate control. While the name of the plan might suggest that it was intended to provide for the return of villagers to their settlements, this was not actually the case. The only “return”, in fact, was to geographical location, to the general vicinity. Rather, the ultimate objective – and the only realistic framework in which return might take place – was a new geo-social (spatial) structure, one that would facilitate the establishment of effective control over groups of settlements.

53It was acknowledged in the plan that the evacuation of villages and the displacement of people had inflicted great pains and sorrows on those involved, but that the forced evacuation had to be considered an opportunity for the creation of something new, i.e. the creation of the conditions in which the “forced migrants” could become more productive, both for themselves and for “their country”. The evacuation of the rural settlements, therefore, was treated as an opportunity for the development of a new settlement structure that was more “rational” and “vital”. The old dispersed and unproductive settlement units were to be exchanged for new settlement units of sufficient size and potential. At the core of the plan, therefore, was the concept of a transformed rural space [Oyan et al. eds. 2001: 1].

54The master-planners endeavored to reorganize the scattered and therefore difficult-to-supervise settlement structure and to create a more transparent rural structure. And the creation of transparency was to be created through rehabilitation, which did not refer to some kind of restitution of the modes of livelihood and social systems of the forced migrants, wrecked by evacuation and village destruction, but to the creation of the necessary framework for the establishment and effective operation of state institutions in the region [Oyan et al. eds. 2001: 5]. Apart from the construction of a road and highway network providing easy access to the region, the scatter of settlements had to be reorganized in such a way that they would become “legible” for outsiders, i.e. state representatives. This is the prime logic behind the application of a model approach, as it enables the reduction of complex realities to schematic categories [Scott 1998: 77]. To accomplish a schematic simplification in the settlement structure, the master-planners introduced two concepts: the concept of sub-region (alt-bölge) and center-village (merkez-köy) [Oyan et al. eds. 2001].

Table 2. Center-Village Pilot Projects and Their History of Designation





Number of dependent settlements




13 villages  4 hamlets




 3 villages 22 hamlets




 6 villages




 9 villages  3 hamlets




 2 villages  6 hamlets




 6 villages  4 hamlets




 7 villages  1 hamlet




 9 villages 18 hamlets




 8 villages




 4 villages  9 hamlets




 4 villages  2 hamlets




 6 villages 36 hamlets

Source: J. Jongerden [2007].

55A sub-region was defined as a cluster of settlements distinguished from other clusters by economic, cultural, administrative, or social characteristics. Supposed affinity and coherence between people and villages were to be used as characteristics to delineate the sub-regions. This type of concept is, in fact, what Scott refers to as a “summary-description”, that is, designed to allow officials to create (formal) spatial categories and identify their citizenry [1998: 53-83]. The center-village was defined as the settlement within a sub-region, which, by its characteristics – size, location and infrastructure – could become a junction or hub for other settlements and developed into an intermediate entity between the district town and small villages and hamlets. The center-village concept did not refer to the spatial concentration of people, but rather to the spatial concentration of services and the articulation of surrounding settlements with this center. The concept of center-village was twinned to that of sub-region, with the center-village as a nucleus and the sub-region as its surroundings. The key characteristic of the center-village concept was that it created a system of formal hierarchical relations, in which a cluster of settlements was administratively subordinated to a center-village, which was subordinated to a district town, which was subordinated to a provincial city, which was subordinated to the capital. As such, the reconstruction of rural space was an attempt to organize legibility and administrative control.

56The Village Return and Rehabilitation Development Plan proposed twelve pilot projects, one per province (Table 2 p. 94). The pilot projects were aimed at the (zonal) development of twelve sub-regions, each including one center-village and a number of dependent villages and hamlets (varying between 6 and 42). In addition to the twelve center-villages, a total of 77 villages and 105 hamlets was planned for the pilot zones, encompassing 194 settlements all told.

57That the objective of this plan was the organization of control was also evidenced by the fact that the master-planners included non-evacuated rural settlements in their development plans as well as evacuated ones. In fact, the Diyarbakır pilot zone comprised only inhabited, non-evacuated rural settlements. As the master-planners explained:

Village evacuation in East and Southeast Anatolia occurred on a wide scale. [...] The solution to the problem [of an unproductive settlement structure] is not to stay within the boundaries of a return to the evacuated villages. [...] Villages that have not been evacuated at all have to be included in the alternative models that are part of the framework of the sub-regional development plan [Oyan et al. eds. 2001: 7].

  • 21 Another issue is that settlements inhabited by village guards recruited during the original scorche (...)

58Also, the master-planners did not consider all the evacuated and destroyed settlements as qualified for return. The Bingöl pilot plan showed that some of the evacuated settlements (Küçükbayırlı and Bayırlı) were not to be reconstructed: their populations were to be resettled at locations more suitable to the state. The organization of a new settlement structure was paramount, and return subordinated to that.21

59The military and governors opposed such a return, however, even in the context of the master-plan. First, they did not favor an organization of space based on clustering, but preferred the concentration of the population in a few large settlements. Second, they wanted to classify the settlements into two types: those settlements which were appropriate for reconstruction because of the feasibility of turning them into centers for the spatial concentration of populations, and those that were not. Most likely it would not have been politically possible for the plan to have been forced through against this opposition by a relatively weak coalition government (as existed at the time and had been for two decades). Anyway, it mattered little, as fate transpired. When the Justice and Development Party (Adalet ve Kalkınma Partisi: AKP) swept to power in 2002 (with large support from the region’s Kurdish population), the master-plan was withdrawn and the pilot projects abandoned.


60The evacuation and destruction of rural settlements in Turkey’s Southeast was part of an attempt to organize a space that would make it possible for the Turkish Armed Forces to act at will, unrestricted by social and environmental obstacles. In combat zones, the existence of rural settlements whose inhabitants had not been recruited into the village guard system presented the army with a double problem. First, the settlements provided various and vital resources for the enemy; and second, the state forces had to differentiate in military operations between combatants and non-combatants, with an enemy force whose existence was partially dependent on a blurring of that distinction, i.e. the so-called milis. The evacuation and destruction of rural settlements eliminated these problems. The fact that this contravened the Geneva Conventions did not mean that it had to be abandoned, but just that it was not officially announced or openly acknowledged. The true numbers of displaced remain unknown.

61The large-scale evacuation and destruction of settlements as part of the counter-insurgency strategy of the Turkish military against the PKK guerrilla radically transformed the countryside. A contraction of space in which the guerrilla could operate was effected through geopolitical and dromopolitical microtactics. Insurgent territory was first “smoothed” by the population clearance, and then aggressively reduced by mobile, well-armed attack forces. Field domination was enabled by a combination of these along with the sheer scale of numbers (in current parlance, a troop “surge”), including the addition of the casual forces of a newly instituted system of civil militia. Given that the population not co-opted into the village guard system was seen to be sympathetic to the PKK cause, any preexisting rift between the Turkish state army and this local Kurdish population was further widened by the evacuation, which only served to enable the forces of the state to mistreat its own people. Their land was effectively regarded as state property, and the surrounding commons (forests, mountains) usurped in the “national interest”.

62The state may position itself as the ultimate guarantor of the individual liberty of its citizens, but the price it exacts for this is the right to determine the nature and scope of that liberty. In times of conflict, that leads it to define its own space. In times of direct challenge to the central power, indeed, a civil war of secession, the state may go so far as to invoke the ultimate territorial sanction, to denude and destroy its own land, as if it were “cutting off its gangrened left hand with its right” [Akın 1993: 41].

63With the success of the Turkish military strategy – in combination with the PKK’s own fatal escalation of the war towards a strategy of conventional warfare – the die was cast. The illegality of the state’s actions meant that the villagers could not be properly resettled and compensated (even had this been desired), while the logic of this success implied that people would not be allowed to simply return to their homes when the fighting in their area was over. Thus, the evacuation and destruction of rural settlements has been discussed here as a constitutive part of spatial contraction, but also put in perspective through discussion of the ensuing peace-based transformation.

64Plans for return and a reconstruction of rural space advanced by state agencies did not envisage that people would go back to their homes once their communities were rebuilt. The conception of “return”, it is argued here, employed a rather Orwellian doublespeak. The examination of the main rehabilitation plan developed by the state shows it to have clearly been based on the implementation of a categorization and hierarchization of space that set administrative supervision and control as a prime objective. If the people could not be trusted to support the state, then the state would have to enter the lives of the people more directly, through the organization of settlements and closely controlling the relationship of the center to its periphery. As such, the master-plan was drafted from the logic of counter-insurgency, even though the military did not embrace it. We can conclude that the post-war development plan was actually the continuation of war by different means.

  • 22 Human Rights Watch, “Displaced and Disregarded. Turkey’s Failing Village Return Program”. Washingto (...)

65Finally, notwithstanding some drift back, the land vacated continues to be largely empty. Those villagers trying to return have been hampered by the authorities in many ways, from the refusal of permission to resettle their villages to lack of local service provision and various forms of bias towards village guards [Jongerden 2007: 252-277].22 Some villages seem to be permanently destroyed, while others have changed character, for example operating more as summer holiday locations than living communities. Realpolitik sides with the strong; the state designs its space, even if this sometimes means returning to a blank canvass, a rural minimalism with barely any design at all.

Haut de page


Agamben, Giorgio — 1998, Homo Sacer. Sovereign Power and Bare Life. Stanford, Stanford University Press.

Aker, Tamer, Betül Çelik, Dilek Kurban, Turgay Ünalan and Deniz Yükseker — 2005, Türkiye’de Ülke İçinde Yerinden Edilme Sorunu. Tespitler ve Çözüm Önerileri. Istanbul, TESEV.

Akın, Cihangir — 1993, “Güneydogu Anadolu bölgesinde kalkınma: Güvenlik ve istihdamın temini için cazibe merkezi modeli”, Kooperatifçilik (Sept.-Dec.): 39-49.

Armitage, John — 2000, Beyond Postmodernism. Paul Virilio’s Hypermodern Cultural Theory. Available at articles.aspx?id=133.

Barkey, Henri and Graham E. Fuller — 1998, Turkey’s Kurdish Question. Boston, Rowman & Littlefield Publishers.

Beckett, Ian F.W. and John Pimlott eds. — 1985, Armed Forces and Modern Counter-Insurgency. London-Sydney, Croom Helm.

Bozarslan, Hamit — 2004, Violence in the Middle East: From Political Struggle to Self-Sacrifice. Princeton, Markus Wiener Publishers.

Çelik, Ayşe Betul — 2005, “’I Miss My Village’: Forced Kurdish Migrants in Istanbul and Their Representation in Associations”, New Perspectives on Turkey 32: 137-163.

Cizre Sakallıoglu, Ümit — 2005, “The Anatomy of the Turkish Military’s Political Autonomy”, Comparative Politics 29 (2): 151-166.

Doğanay, Filiz — 1993, Merkez Köyler. Ankara, Devlet Planlama Teskilati.

Donnell, John C. — 1970, “Expanding Political Participation. The Long Haul from Villagism to Nationalism”, Asian Survey 10: 688-704.

Ergil, Dogu — 2009, “Kürt Raporu, güvenlik politikalarindan kimlik siyasetine”. Istanbul, Timas Yayinlari.

Etten, Jacob van, Joost Jongerden, Hugo J. de Vos, Annemarie Klaasse and Esther C.E. van Hoeve — 2008, “Environmental Destruction as a Counter-Insurgency Strategy in the Kurdistan Region of Turkey”, Geoforum 39: 1786-1797.

Hardt, Michael and Antonio Negri — 2004, Multitude, War and Democracy in the Age of Empire. New York, The Penguin Press.

Imset, Ismet G. — 1995, The PKK: Freedom Fighters or Terrorists? Available at

Jongerden, Joost — 2007, The Settlement Issue in Turkey and the Kurds. Leiden-Boston, Brill. — 2009, “Crafting Space, Making People. The Spatial Design of Nation in Modern Turkey”, European Journal of Turkish Studies 10: 2-22. — 2010, “Dams and Politics in Turkey. Utilizing Water, Developing Conflict”, Middle East Policy 17 (1): 137-143.

Jongerden, Joost and Ahmet Hamdi Akkaya — 2010, “Born from the Left. The Making of the PKK”, in C. Marlies and J. Jongerden eds., Nationalisms and Politics in Turkey. Political Islam, Kemalism and the Kurdish Issue. Oxon-New York, Routledge.

Jongerden, Joost, Rene Oudshoorn and Henk Laloli — 1997, Het verwoeste Land, berichten van de oorlog in Turks-Koerdistan. Brede, uitgeverij papieren Tijger.

Kellner, Douglas — 2000, “Virilio, War and Technology. Some Critical Reflections”, in J. Armitage and P. Virilio eds., From Modernism to Hypermodernism and beyond. London-New Delhi, Thousand Oaks-Sage.

Kocher, Matthew — 2002, “The Decline of PKK and the Viability of a One-State Solution in Turkey”, International Journal on Multicultural Studies 4 (1): 1-20. — 2004, “Human Ecology and Civil War”. PhD thesis in political sciences. University of Chicago.

Marcus, Aliza — 2007, Blood and Belief. The PKK, Turkey and the Kurdish War. New York, New York University Press.

Moore, Mick — 1984, “Categorizing Space. Urban-Rural or Core-Periphery in Sri Lanka”, Journal of Development Studies 20 (3): 102-122.

Newsinger, John — 2002, British Counter-Insurgency, from Palestine to Northern Ireland. New York, Palgrave.

Oyan, Oğuz, Melih Ersoy, H. Çağatay Keskinok, H. Tarık Şengül, Galip Yalman, Remzi Sönmez and Erdal Kurttaş eds. — 2001, Doğu ve Güneydoğu Anadolu Bölgesi Köye Dönüş ve Rehabilitasyon Projesi Alt Bölge Gelişme Planı (12 volumes). Ankara, Bölge Kalkınma İdaresi (BKI)-Türk Sosyal Bilimler Derneği.

Özdağ, Ümit — 2003, The PKK and Low Intensity Conflict in Turkey. Ankara, Frank Cass.

Pamukoğlu, Osman — 2003, Unutulanlar Dışında Yeni Bir şey Yok, Hakkari ve Kuzey Irak Dağlarındaki Askerler. Ankara, Harmony.

Ronayne, Maggie — 2004, The Cultural and Environmental Impact of Large Dams in Southeast Turkey. Galway-London, National University of Ireland-Kurdish Human Rights Project.

Scott, James — 1998, Seeing Like a State. How Certain Schemes to Improve the Human Condition Have Failed. New Haven, Yale University Press.

Signoles, Aude — 2005, “Gestion des espaces et contrôle politique. Israël au cœur du processus décisionnel local (1993-2000)”, Études rurales 173-174: 87-108.

Sutton, Keith — 1981, “The Influence of Military Policy on Algerian Rural Development”, Geographical Review 71: 379-394.

Tezcan, Sabahat and Koç, İsmet — 2006, Turkey Migration and Internally Displaced Population Survey (TMIDPS). Ankara, Hacettepe University Institute of Population Studies.

Virilio, Paul — 2006, Speed and Politics. An Essay on Dromology. New York, Semiotext(e).

Weizman, Eyal — 2007, Hollow Land, Israel’s Architecture of Occupation. London, Verso.

White, Paul — 2000, Primitive Rebels or Revolutionary Modernizers? The Kurdish National Movement in Turkey. London-New York, Zed.

Woods, Michael D. — 1990, “Rural Awakenings. Grassroots Development and the Cultivation of a National Past in Rural Sri Lanka”, in J. Spencer ed., Sri Lanka: History and the Roots of Conflict. Abingdon-Oxon, Routledge: 164-183.

Yılmaz, Bediz — 2006, “Far Away, So Close: Social Exclusion and Spatial Relegation in an Inner-City Slum of Istanbul, Tarlabaşı”, in F. Adaman and Ç. Keyder eds., Poverty and Social Exclusion in the Suburbs of Large Cities in Turkey. Brussels, European Commission, Employment, Social Affairs and Equal Opportunities: 26-40.

Zasloff, Joseph J. — 1962-1963, “Rural Resettlement in South Vietnam: The Agroville Program”, Pacific Affairs 35: 327-340.

Haut de page


1 Human Rights Watch, “Displaced and Disregarded. Turkey’s Failing Village Return Program”. Washington DC, 2002, p. 25. Available at turkey/Turkey1002.pdf

2 Kurdish Human Rights Project, “Internally Displaced Persons. The Kurds in Turkey”. London, 2002. Available at

3 The village guards are a militia, legally established in 1985 and which became operational from 1987 onwards. About 5,000 men joined this paramilitary force in its first year, a number which had increased to 67,000 by 1995. Village guards were expected to take defensive positions against the PKK and also to participate in military operations, some of which involved cross-border incursions into northern Iraq. See HRW, “Weapons Transfer and Violations of the Laws of War in Turkey”. Washington DC, 1995. Available at reports/1995/Turkey.htm

4 Human Rights Watch, “Displaced and Disregarded. Turkey’s Failing Village Return Program”. Washington DC, 2002, p. 25. Available at turkey/Turkey1002.pdf

5 In Dersim/Tunceli, for example, the Teacher’s Training School was a recruitment focus of the movement [Jongerden and Akkaya 2010].

6 Theoretically, a unit was composed of a commander, a vice-commander, a political commissary, someone operating a machine-gun and someone operating a RPG rocket launcher. In PKK terminology, this was named a manga, or squad. Three squads formed a takım, or platoon, and three platoons a company or bölük.

7 Devlet Planlama TeŞkilatı (State Planning Organization), “Türkiye’de Yerlesme Merkezlerinin Kademelenmesi. Ülke Yerlesme Merkezleri Sistem”, Cilt 1-2. Ankara, Yayini, 1982.

8 In fact, the milis constituted a major part of the total PKK force. In his study of the PKK, P. White [2000: 143] presents the figures of the US State Department, according to which the PKK in 1995 had a guerrilla force of about 15,000 supported by a part-time militia of 75,000. According to D. Ergil [2009: 382] the PKK had a guerrilla force of 20,000 men, some 10,000 armed militants (milis) in the country and 500,000 active supporters.

9 The concept of “dromopolitics” comes from P. Virilio [2006], who emphasizes the importance of the modern technologies of motion and acceleration. The subtitle of this work, “Essays on dromology”, refers to the need for a knowledge or science of speed, which, according to Virilio, lays at the foundation of modern society. In respect to armed conflict, the disappearance of traditional defenders of spaces or territories, the castles and fortified cities of the Middle Ages and the bunkers and trenches of more recent times, Virilio argues, is the consequence of new weapons systems and military strategies based on high speed and mobility, which have transformed positional warfare into wars of movement. The need for fortification, important in positional warfare, was surpassed by the need for mobility [Armitage 2000; Kellner 2000].

10 The evacuation of rural settlements was made possible in 1987 by Decree No. 285 concerning the state of emergency, which granted formal authority for this to the Regional Governor (a kind of “super-governor” appointed by Ankara for all the provinces of the state of emergency area, i.e. the whole of Southeast Turkey). The decree enabled the governor to order the temporary or permanent evacuation of settlements in the interest of “public security” (in reality, military commanders decided about the evacuation of rural settlements). The second Protocol Additional to the Geneva Conventions (1977) (ratified by Turkey) had banned this practice, but the resettlement of the population was never declared openly as official (civil or military) policy.

11 Selected Works of Mao Tse-Tung. Vol. IX (1937). Available at 1937/guerrilla-warfare/ ch01.htm

12 “Tunceli” is the modern, official (Turkish replacement) name for the province of Dersim, the old (Zazaki/Kurdish) name. The twentieth century program of Turkification in the new republic included an attempt to claim the space through name changes to thousands of settlements [Jongerden 2009].

13 Actually, the southeast of Turkey had been ruled under martial law and emergency regulations since 1927, and thus been a zone of weak citizenship since long before the period here considered. Until 1952, most of the area (Bitlis, Diyarbakır, Elazığ, Hakkari, Mardin, Siirt, Urfa, and Van) was administered by an Inspector General, an office established in 1927 to bring “order and discipline”. In 1935, two further Inspector Generals were appointed to administer Kurdish populated areas, one for the “Murat and Munzur” region, covering Dersim/Tunceli, and the other for the northern part of the East, covering Ağri, Çoruh, Erzincan, Erzurum, Gümüshane, Kars, and Trabzon. The Southeast was closed to foreigners until 1965, and the region subsequently ruled under martial law or state of emergency from 1980. The OHAL state of emergency area was created in 1983, before the PKK started its military campaign, and the office of the Regional Governor established in 1987, maintaining military control in the Southeast after executive power had elsewhere been transferred back to the civilian authorities following the 1980 coup d’État. Civilian rule was later returned to the area, slowly and piecemeal, until the final lifting of OHAL law from the remaining four provinces to which it applied in 2002.

14 The spatially-grounded state of exception in its extreme form was the kill-zone. In a kill-zone, human existence is reduced to what G. Agamben [1998] calls “bare life”, that is to say, deprived of rights and only included in the political order in the form of its exclusion, i.e. qualified to be killed.

15 Ministry of Defense, “Defense White Paper”, Ankara, 2000.

16 Kurdish Human Rights Project, “Internally Displaced Persons. The Kurds in Turkey”. London, 2002, pp. 118-120. Available at ulkeincinde.pdf

17 Meclis Araştirmasi Komisyonu, “Doğu ve Güneydoğu Anadolu’da Boşaltilan Yerleşim Birimleri Nedeniyle Göç Eden Yurttaşlarimizin Sorunlarin Araştirlarak Alinmasi Gereken Tedbilerin Tespit Edilmesi Amaciyla Kurulan”. Report, December 1st 1997, p. 44. Ankara, Türkiye Büyük Millet Meclisi.

18 Turkey had requested a loan of 50 million dollars from the World Bank for the toplu kondu plan and 278 million dollars from the European Community for the toplu çiftlik plan, but both requests were denied, the former for reasons undisclosed and the latter because Turkey was seen to be shifting the economic cost of the village evacuation onto Europe [Jongerden 2007].

19 At the time of the study, Prof. Dr. Oğuz Oyan was president of the Turkish Social Sciences Association (Türk Sosyal Bilimler Derneği: TSBD) and vice-president of the CHP; Dr. Melih Ersoy was professor at the Department of Urban and Regional Planning, Faculty of Architecture, Middle East Technical University in Ankara.

20 GAP has now expanded to become “a multi-sector and integrated regional development effort approached in the context of sustainable development” covering the entire southeast region. See gap_eng.php?sayfa=English/ Ggbilgi/gnedir.html

21 Another issue is that settlements inhabited by village guards recruited during the original scorched-earth campaign were clustered with evacuated and destroyed settlements in the pilot projects. It would appear that prospective returnees were to be settled alongside village guards, as far as information is available – leastways, this was the case with the projects in the provinces of Batman, Bingöl, Bitlis, Hakkari, Muş, Şırnak and Van.

22 Human Rights Watch, “Displaced and Disregarded. Turkey’s Failing Village Return Program”. Washington DC, 2002. Available at Turkey1002.pdf

Haut de page

Pour citer cet article

Référence papier

Joost Jongerden, « Village Evacuation and Reconstruction in Kurdistan (1993-2002) », Études rurales, 186 | 2010, 77-100.

Référence électronique

Joost Jongerden, « Village Evacuation and Reconstruction in Kurdistan (1993-2002) », Études rurales [En ligne], 186 | 2010, mis en ligne le 11 mars 2013, consulté le 25 février 2018. URL :

Haut de page

Droits d’auteur

© Tous droits réservés

Haut de page
  • Logo Éditions de l’EHESS
  • OpenEdition Journals