Open access peer-reviewed chapter - ONLINE FIRST

Are Safety Zones a Threat to the Institution of Asylum?

Written By

M.A. Natalia Gierowska

Submitted: May 22nd, 2022 Reviewed: June 28th, 2022 Published: October 19th, 2022

DOI: 10.5772/intechopen.106148

Human Migration in the Last Three Centuries Edited by Ingrid Muenstermann

From the Edited Volume

Human Migration in the Last Three Centuries [Working Title]

Dr. Ingrid Muenstermann

Chapter metrics overview

31 Chapter Downloads

View Full Metrics


Safe zones emerged as a new form of humanitarian space during the 1990s. Safe zones take various forms; however, they most often assume the shape of a temporal camp in a specified geographical area within the borders of the refugee-producing state. Nevertheless, safe zones are not protected from criticism. The increasingly common practice of States rejecting asylum claims based on the existence of safe zones is in conflict with the notion of asylum as a tool for the international protection of refugees, a human right enshrined in Article 14(1) of the Universal Declaration of Human Rights. To establish the risks associated with safety zones vis-à-vis refugees, attention will be drawn to the scale of permissibility in refugee law in its recognition of safety zones as a ground to restrict access to asylum. Particular attention will be drawn to the notion of Internal Protection Alternatives (IPA), as if the applicant is possibly safely relocating within his or her country of origin, they will not be accorded refugee status. To establish whether safe zones can represent a pragmatic complementary policy to international protection or a threat to the institution of asylum, this chapter will critically assess the performance of three safety zones: Iraq (1991), Bosnia and, more specifically, Srebrenica (1995) and Syria (2020).


  • safety zones
  • internal migration
  • Syrian war
  • involuntary migration
  • human rights abuses

1. Introduction

Safety Zones emerged as a new form of humanitarian space during the 1990 [1]. No single terminology or universal definition has yet been agreed for ‘safety zones’ (the term employed in this thesis). At the most fundamental level, safety zones represent a geographically designated area intended to protect civilian populations during wartime. Safety zones take various forms; however, they most often assume the shape of a temporal camp in a specified geographical area, within the borders of the refugee-producing state and are enforced by a third State or international organisation. Such zones represent the only realistic option for countries at war in the context of closed borders, or for those who do not wish to, or do not have the financial ability to, leave their native country, whilst facilitating the repatriation of those who left. As such, they are painted as a relatively low-cost remedy to the limitations of the contemporary asylum regime. However, safety zones are not free from criticism. The gradual erosion of political appetite to receive refugees has created a context where safety zones are used to curtail refugee influxes, in fact excluding applicants from protection. The existence of safety zones may legitimise repatriation programmes, which would otherwise be considered an abdication from international law obligations, and more specifically, the norm of non-refoulement. This increases safety zones’ chances of delivering sub-optimal results in their efforts to protect civilians.

To establish the risks associated with safety zones vis-à-vis refugees, attention will be drawn to the scale of permissibility in refugee law in its recognition of safety zones as a ground to restrict access to asylum. Particular attention will be drawn to the notion of Internal Protection Alternatives (IPA), as if there is a possibility of the applicant safely relocating within his or her country of origin, they will not be accorded refugee status. During status-determination hearings minimum conditions must be met to invoke a legal IPA which, based on the legal scholarship, this paper discerns as: (a) accessibility (both legal and physical) to the region proposed as an IPA, (b) protection from the original risk of persecution in the proposed IPA, (c) no new risks of persecution and indirect refoulement in the proposed IPA. The assessment of an IPA permits an open-ended humanitarian evaluation of the relevant factors in a cumulative manner, of a kind not occasioned by the Convention Relating to the Status of Refugees, also known as the Geneva Convention or the 1951 Refugee Convention [2]. That said, the increasingly frequent invocation of an IPA during status determination hearings raises fears of lowering the threshold for invoking an IPA in order to revoke refugee status. The IPA analysis has incorrectly become tied to concept of ‘well-founded fear’, central to the Convention’s definition of refugee. The increasingly common practice of States rejecting asylum claims based on the existence of safety zones is in conflict with the notion of asylum as a tool for the international protection of refugees, a human right enshrined in Article 14(1) of the Universal Declaration of Human Rights (UDHR): ‘Everyone has the right to seek and enjoy asylum’ [3]. Furthermore, this legal development carries the risk of concealing the degree of protection offered in the established safety zones, enabling States to abrogate from the principle of asylum by distancing themselves from accountability for the conditions on the ground in such areas.

In order to assess whether safety zones can represent a pragmatic complementary policy to international protection or a threat to the institution of asylum, this paper will critically assess the performance of three safety zones: Iraq (1991), Bosnia and more specifically, Srebrenica (1995) and Syria (2020). The safety zones’ effectiveness will be evaluated against their ultimate ambition, that is, meaningful protection. ‘Protection’ is approached from the legal perspective and is understood as the entitlement of individuals under Refugee Law. The establishment and the legality of these safety zones has been made possible by a retrospective application of the IPA standard to these three case studies. Whereas the legality of the former two case studies has been assessed by engaging with secondary and primary sources, the examination of the newly carved Syrian safety zones has been supported by primary research. This paper argues that while is it theoretically possible to establish safety zones based on international humanitarian law, in practice their implementation is often impacted by geopolitical considerations of nation-states, as can be seen in all three case studies- Bosnia, Iraq and especially in Syria.

1.1 Literature review

1.1.1 Safety zones: History

To understand the contemporary issues this paper explores; it is important to take note of the genesis of safety zones. Whilst historically civilians in countries engulfed in war sought refuge in places of worship where sanctity was a deterrent, ‘safety zone’ as a concept was developed during the Franco-Prussian war (1870–1871), by Henri Dunant, the founding father of the International Committee of the Red Cross (ICRC) [4]. Dunant’s advocacy for safety zones bore fruit in the twentieth century- in 1937 general Franco designated Madrid a safety zone amid the brutalities of the Spanish Civil War [5], In 1937, the Sino-Japanese War welcomed a Shanghai Safety Zone [6], but efforts to establish safety zones during World War II were futile [5]. In 1948, however, populations found temporary protection from armed attacked in ICRC-administered safety zones in Jerusalem, leading to the codification of ‘neutralised zones’ in the Fourth Geneva Convention. Long [7] argues that these early examples were all provisional measures inaugurated with the consent of all of the parties to the conflict, within the framework of war laws, intended to offer temporary havens from hostile acts, rather than an alternative for national protection.

This framework mapped out the silhouette of safety zones as seen in the 1949 Geneva Conventions, and became a component of International Humanitarian Law (IHL) [8]. Under IHL and its treaties, safety zones are traditionally be referred to as ‘protected zones’ and are consensually established demilitarised areas in military conflicts [5]. Article 23 in the First Geneva Convention instructs to ‘protect the wounded and sick’, Article 14 in the Fourth Geneva Convention extends the protection in ‘hospitals and safety zones’ to ‘wounded, sick and aged persons, children under fifteen, expectant mothers and mothers of children under seven’, whereas the succeeding Article 15 foresees the establishment of consensually recognised neutralised zones where its protected status is subject to the agreement between the Parties to the conflict [9]. Since the endorsement of the Geneva Conventions, six such zones have been established: Bangladesh (1971), Cyprus (1974), Vietnam (1975), Cambodia (1975), Nicaragua (1979), Croatia (1992) and Brinie et al. [10]. These provisions represented the only explicit protections from violence to be enjoyed by civilians under IHL [8]. However, these protections were not codified until the adoption of the Additional Protocol in 1977, which further expanded the concept of safety zones to cover the entire civilian population, as opposed to carving out specific categories of beneficiaries [8].

1.1.2 Can safety zones offer an alternative to asylum?

In 1994, Sandoz, the Director for International Law and Communications at ICRC, indicated conditions, which are likely to contribute to the success of safety zones. A model safety zone was envisioned as ‘carefully defined geographically, fully demilitarised and with concrete assurances that they cannot be used as a base from which to pose risks to any belligerent party’ [11]. However, lessons drawn from history show that a safety zone’s recognition by all sides of the conflict is ‘indispensable’ [5] to its success [1, 4, 12, 13]. Consent may take two forms- tactical consent by combatants, or an official consent granted by the State itself agreeing to an international presence. Acknowledging the importance of recognised safety zones from the operational angle, Brinie et al. [10] draw further attention to the normative significance of consent-based zones. Their argument is drawn from IHL and international normative theory, where consent minimises the forced imposition of values by aliens and instead becomes a manifestation of sovereignty and self-determination [10]. However, Outram [14] underscores, that even this fundamental condition of consent under IHL is rarely satisfied and respected.

Due to different forms of warfare in the twentieth century, protecting civilians became increasingly difficult. Wars drew both state and non-state actors where most violence directly targeted civilians- a reality, which was conceptualised by Koldor as ‘New Wars’ ([15]: 8). In the post-Cold War era, conflicts became increasingly internal, aiming to eliminate communal or competing ethnic groups [5]. Paradoxically, it was against this backdrop that response to humanitarian crises were respatialised in the 1990s [12] with a sudden rhetorical shift by both UNHCR and host states away from ‘the right to leave’ towards the ‘right to remain’ [1].

During the twentieth century, safety zones became the manifestation of this new approach, as they are intended to bring ‘safety to people rather than people to safety’ ([12], p. 169). Posen [16] explained that this resulted in the intensification of coercive strategies by international actors to erect safety zones. Birnie et al. [10] show that the post-Cold War period brought a proliferation of enforced safety zones, citing the examples of designated protected areas in Northern Iraq (1991), Bosnia (1993–1995), Rwanda (1994), Kosovo (1999) and Libya (2011), which Orchard [1] refers to as ‘credible presence safety zones’. Interventions by politically and military robust States prompted Landgren [5] to remark that such action further obfuscates the need for humanitarian efforts with the protection of international peace and security.

Arguments in favour of in-country protection point to inadequate reception conditions for asylum seekers, which do not meet the standards spelt out in human rights law and thus further complicate the process of seeking asylum [17]. Those who are awarded refugee status receive very little state assistance, while those who await status determination are frequently placed in detention centres, which inhibit their right to freedom of movement [18]. Most importantly, however, Edwards [19] points out that asylum is unattainable to those who are unable or unwilling to leave their country of origin to seek refuge abroad. Further, the right to stay and the right to return is not protected under the institution of asylum [19]. As a consequence, those who remain in their country of origin are seen as unworthy of asylum, even though they often suffer the type of persecution which would qualify for political asylum [20]. Responsibility for the protection of the Internally Displaced Persons (IDPs) rests with national authorities [21]. Since the Geneva Convention’s mandate does not extend to IPDs, and the Guiding Principles on Internal Displacement published by the Secretary-General of the United Nations in 1998 is a non-binding instrument, the internally uprooted populations habitually find themselves in a situation whereby their own State is unable or unwilling to avail them with protection against persecution [10].

Scholars such as Gilbert and Rüsch [22], Hyndman [12], Recchia [13] or Orchard [1] are more cautious in their advocacy for safety zones, Birnie et al. [10] recapitulate that in the light of the inflexible legal framework of asylum and the changing nature of warfare, limited forms of intervention like safety zones, are the most suitable policy option to contemporary refugee emergencies. While safe zones address the procedural shortcomings of status determination, a significant body of literature views such areas with scepticism. Birnie et al. [10] point out that the invocation of safe zones allows for states to abrogate from their international obligations to refugees. While safety zones may offer a degree of safety and protection (given that access to humanitarian relief is secured), legal scholars such as Edwards [19, 23] or Clark [24] agree that such restrictive refugee policies result in the ‘denial of some of the important social, economic and cultural rights guaranteed by the 1951 Convention’ [23]. Long [7] argues that while providing assistance and protection to the civilian population is an ineliminable component of humanitarian relief, the inception of safety zones reduces the right to be a refugee simply to finding the minimum ‘safety’ [7], which prompts Landgren [5] to appeal to Agamben’s terminology of ‘bare life’. It is crucial, as noted by Birnie et al. [10], not to accept this refugee policy as a surrogate for asylum.

Chimni [25], however, argues that safety zones do precisely that - provide a potential alternative to asylum. The concern is shared by Ghrainne [26], who recognises a pattern of States applying an increasingly narrow interpretation of the 1951 Geneva Convention to accommodate their national interests. States can prematurely repatriate refugees and asylum seekers, as well as exclude applicants from protection by invoking safety zones as an IPA [5, 25, 27]. Under Refugee Law, such actions amount to refoulement, however, Ghrainne [26] correctly notes that the creation of safety zones allow States to argue that applicants are not refugees and are therefore excluded from the principle of non-refoulement. Birnie et al. [10] show that this is not simply a theoretical probability, as the scholars cite historical examples of safety zones, which were established as an alternative to asylum and a refugee containment strategy: Rwanda, Northern Iraq and Bosnia.

Ghrainne [26] raises an important point: refugee policy is motivated by two conflicting objectives, compassion and State interest. When analysing a safety zones suitability for delivering protection, it is crucial to embark on a wide-ranging analysis, which incorporates broader political considerations of ‘refugee containment, deterrence, and deflection’ [26]. Safety zones should be treated favourably if seen as a complementary mechanism of protection provision but should not be used as a replacement and thus a potential threat to asylum. To adequately address this issue, this paper has implemented a research framework with several research methods to adequately address the previously identified issues.


2. Methodology

A qualitative research framework has been applied to the question, following Frederick’s [28] argument, that this method is most suited for refugee-related research, due to its close relationship with sociology, ethnology and anthropology. The depth of the information could not be reproduced by quantitative research where participants are expected to fit their responses into a set of limited options supplied by the researcher. The majority of the participants are Internally Displaced Persons, and some are former refugees who have experienced the brutality of war and forced migration as its by-product.

2.1 Methods

The interviewees are Syrian nationals currently in Turkish safety zones, researchers, non-governmental organisations and journalists and Syrian nationals outside of Turkish safe zones. The questions address issues raised during a status determination hearing and IPA tests, which would invoke the ‘Reasonableness’ analytical framework for the IPA test. For the Turkish safety zone to be a relevant relocation alternative, the proposed IPA must satisfy four baseline criteria’s: accessibility of the IPA to the applicant, protection from the original risk of persecution, absence of new risk of persecution and indirect refoulement. Aside from surveys, the paper also analysed the existing sources, particularly articles from peer-reviewed journals, as well as relevant documents such as UN resolutions. The paper will synthesise information from secondary and primary sources to draw relevant conclusions on the research question.

2.2 Surveys

Data from Syrian nationals currently residing in Turkish controlled Safety Zones was collected from semi-structured interviews, which permit for the incorporation of the respondent’ opinion about what they think is ad rem. The interviewer was instructed to skip sensitive questions and themes if necessary, allowing to adapt the interview to the respondent’s readiness to reveal particular information, preserving the underlying ethical dimension of this research. Clear communication was ensured by the interviewer’s fluency in the Syrian interviewees’ native language, Arabic, which also facilitated the establishment of a trustworthy relationship with the interviewee.

Fixed response interviews were used to gather data from Syrian nationals settled in territories which fall outside of the Turkish control, to evaluate the relative safety of Turkish safety zones as compared to the rest of the country. This model was also used when collecting data from journalists, researchers and academics. This helped maintain the focus of the interview on the current political and socio-economic conditions in the safety zones of interest.

2.3 Selection of sites and groups

The research sites were selected based on both the levels of accessibility to researchers as well as the current dominant faction in power in respective Syrian governates. Thus, 19 interviews were conducted in three Turkish controlled cities: Tal Abyad (2), Jarabulus (8) and Al Bab (9). The interviews will be referenced in the following manner- the interviews from Tal Abyad will be referred to as interviews 1 and 2; the interviews from Jarabulus will be referred to as interviews 3-10; the interviews from Al Bab will be referred to as interviews 11-19. As a point of contrast, 10 additional interviews were collected from Syrian cities where Turkish presence is not established, among them Tal Rifaat (2), Dayr az Zawr (2), Al-Qamishli (2), As-Suwayda(1), Damascus (1) and Al-Hasakah (2). Interviews from Tal Rifaat will be referred to as interviews 20-21; interviews from Dayr az Zawr will be referred to as interviews 22-23; interviews from Al-Qamishli will be referred to as 24-25; interview from As-Suwayda will be referred to as interview 26, interview from Damascus will be referred to as 27 and finally interviews from Al-Hasakah as interviews 28 and 29. Additionally, this thesis will make reference to an interview with Kourdi (Eyad Kourdi, freelance journalist working with media outlets such as CNN), Kayyali (Sara Kayali, Syria researcher at Human Rights Watch) and Professor David Keen, Professor of Complex Emergencies at London School of Economics (LSE) and a specialist in the field of Safety Zones (Figure 1).

Figure 1.

Numbers indicate number of interviews collected from a given city.

2.3.1 Research strengths

Interviews with Syrian nationals currently residing within the Turkish administered territories humanised the subjects and elicited valuable data regarding the present conditions of safety as well as additional information on the areas socio-economic realities. Interviewing enabled the researcher access to primary sources which could be compared with relevant secondary sources.

2.3.2 Research constraints

This paper may encounter several research constraints. Firstly, unconscious personal biases may impact the format of the research framework and hypothesis. Due to the complexity of the Syrian war and the diverse array of factions involved in the conflict, responses collected may be influenced by the respondent’s personal biases. The interviewee’s political allegiance and beliefs may influence the tone of the interview and their willingness to share a truthful account of their experiences or controversial opinions, for which one can be persecuted during war. Furthermore, the interviewees knowledge and ability to accurately and honestly reconstruct (often traumatic) events may also constitute a research constraint.

Lastly, conducting research during the COVID-19 pandemic meant that interviews were collected by a Syrian volunteer on behalf of the researcher. This has placed significant constraints on the ability to navigate the semi-structured interviews towards the needs of the research. Aside from understanding these research constraints the paper needs to investigate the relevant legal framework underlying this issue.

2.4 Legal framework

2.4.1 The internal protection alternative and the notion’s nexus to the 1951 Geneva convention

The 1951 Geneva Convention does not explicitly refer to ‘internal flight alternative’ and the definition does not establish whether an applicant must establish a fear of persecution countrywide [29]. Instead, independently from the Convention, the IPA was born out of State practice, governmental policy recommendations and academic investigation of more than three decades [30]. In response to these developments, UNHCR revised its stance on the IPA framework and, in a 1995 publication, officially confirmed its recognition of the IPA test as an essential element of the Convention [29].

The presumption that inadequate domestic protection precipitates international protection is the underlying reasoning informing refugee law and is implied in Article 1(A)(2), which defines a refugee as a person who ‘owing to a well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a social group or political opinion is outside their country of origin and is unwilling or unable to avail him or herself of the protection of that country (Geneva Convention 1951, art. 1(A)(2)) [31]. It is not much of a leap to recognise the relationship between IPA and Article 1(A)(2)- the assessment of the availability of an IPA is an essential element in determining whether an applicant is a Convention refugee [32]. Thus, while 1951 Convention does not explicitly refer to the IPA test, it is nevertheless grounded in the notion of ‘surrogate international refugee protection’, a departure from the cornerstone presumption of international law, which associates protection with the obligation of the country of origin [33, 34]. While IPAs are broadly accepted within the status determination procedure, their lack of fixed juridical interpretation has resulted in inconsistent application [29, 32, 35].

Associated with this terminological inconsistency is growing apprehensiveness about the risk of the IPA test placing too high of an evidentiary burden on the applicant [27, 29, 32]. Given that establishing reasonable fear of persecution in a localised area is already challenging, having to provide evidence of persecution countrywide would place too high an evidentiary burden on the applicant. The often-stringent application of the IPA test has been used in Western countries to deny refugee status to an asylum seeker who is unable to present sufficient evidence, advocating for them to seek refuge within the borders of their country of origin [34]. This ultimately contributes to a narrowing interpretation of the definition of ‘refugee’, as seen in the Convention [36]. Given that designated safety zones are often legally rendered, serious concerns arise as to safety zone’s capacity to threaten the institution of asylum.

2.4.2 The IPA test in the refugee status determination procedure

Two main competing analytical frameworks for evaluating the availability of an IPA have emerged in refugee law as a result of State practice- the ‘Reasonableness’ framework (which is the dominant practice) and the ‘Internal Protection’ framework. However, in-depth discussion and comparison of the two competing frameworks are beyond this paper’s scope. Instead, given that, apart from New Zealand, no state has been identified by legal commentators to have adopted the Internal Protection Framework [30], four baseline criteria’s will be derived from the ‘Reasonableness’ framework and discussed. According to the UNHCR Handbook from 1979, the ‘Reasonableness’ framework demands two sets of analysis to establish the availability of the IPA: the relevance analysis and the reasonableness analysis ([37], p. 92). Whereas the former enquires into accessibility and protection from the original cause of persecution, or other forms of equivalent harm, the latter questions the reasonableness of the expectation for a claimant to relocate. Evaluation of the IPA criteria will allow for critical evaluation of the conditions using the historical examples of Srebrenica safety zones, Iraqi Kurdish safety zones, and the safety zones in north-eastern Syria.

2.4.3 Accessibility of the IPA to the applicant

The key issue when assessing the efficacy of an IPA is whether the individual whose claim is under consideration can realistically access the designated region. The IPA must be an achievable alternative, not a speculative option [29], and ‘practically, safely and legally’ accessible to the applicant. Thus, physical barriers, visa and transit regulations, as well as transportation facilities, are to be taken into consideration. Since it is broadly agreed that the refugee definition is forward-looking, the IPA examination should focus on current country conditions [35]. Similarly, the IPA should not be applied where there is a possibility of the applicant being exposed to undue hardship or compelled to hide within it. Where the political situation of the intermediate State changes to satisfy the requirements of Article 1 C(5)-(6) of the Refugee Convention, refugee status may be revoked. The IPA, however, should not apply where the intermediate State makes the IPA inaccessible to the claimant [32].

2.4.4 Protection from the original risk of persecution

Once the availability of access is established, there must exist an IPA where the applicant’s well-founded fear of persecution will be dissipated [27, 29, 32, 35]. The panel must be satisfied on a balance of probabilities that there is no risk of the applicant’s original fear of persecution in that area. The ‘well-founded fear’ test, which measures the ‘reasonable possibility’ of persecution is used to establish the threat within the IPA [21]. Establishing that the original agent of persecution has not yet reached the designated territories for an IPA does not meet the requirements of the test [27]. Instead, it must be found that the fear of persecution is localised and does not extend to the IPA area. A question not resolved in international jurisprudence is whether an IPA can be invoked if the agent of persecution is the government itself. On one hand, the existence of government as the persecutor precludes the relevance of an IPA [38]. This is the same reasoning as the jurisprudence of the Committee against Torture, which held that it is unlikely to find an IPA where the police have targeted the applicant [32] and is consistent with the position in international human rights law [27]. The UNHCR’s recommendations also raise the possibility of the intermediate State’s inability to act throughout the entire territory; allowing for the creation of an internal protection alternative, given that other requirements are satisfied [37]. The least logically consistent application of the IPA is where the State is both the persecutor and can act anywhere within its jurisdiction, a view maintained by the UNHCR [35]. The superior aptitude for a State to inflict harm, as compared to local governments and non-State actors means that, even in times of non-persecution and in a different location, the effectiveness of an IPA as an antidote is diminished [27].

2.4.5 No new risk of persecution and indirect refoulement

Simply because an applicant will not be confronted with harm from the original cause of persecution is insufficient evidence to presume an IPA exists. The inquiry into a possibility of persecution must be complemented by an inquiry on whether the protected area would be safe from persecution without a nexus to the Convention grounds [27, 32, 35]. This requirement has been supported by many courts and legal commentators [32, 35]. Broadening the relevant criteria for a lawful IPA beyond persecution for the protected Convention grounds, which gives entitlement to refugee status, might appear counter-intuitive. However, establishing a well-founded fear of persecution for a Convention refugee reason precedes the evaluation of the relevance of an IPA and relocation. The IPA framework of analysis demands a more contextualised assessment of persecution, which is often captured by the reasonableness standard [29]. The reasonableness criterion ultimately poses the question of whether it would be reasonable for an applicant to relocate to the proposed site, and thereby expands the elements in the inquiry to factors such as the applicant’s age, gender, family status and employment experience. If the Court finds that there are serious reasons to believe that the applicant would be exposed to forms of harm equivalent in intensity and gravity to those which prompted him to flee and which were not present in his native region at the time of the flight, the IPA may be denied and refugee status invoked instead. Threats to physical security in the suggested IPA, which would compel an applicant to return to his region of origin where he or she is exposed to the original risk of persecution, amounts to ‘indirect refoulement’ [27, 32]. Storey [29] explains that this notion refers to direct state action, which, directly or indirectly, culminates in the breach of the Article 33(1) of the 1951 Convention, in which the principle of non-refoulement is enshrined. All of these findings illustrate that the protection approach to IPA analysis requires for a more precise legal framework, as the lack of modern regulation allows for interpretation in a manner consistent with their national interests. This can be seen in the examples of several safety zones, particularly those established in Turkey.


3. Safety zones: Case studies

3.1 Safety zones in Iraq: Operation provide comfort

In 1991, against the background of the Iraq uprisings and following the First Gulf War, approximately 400,000 Iraqi Kurds fled towards Iran and Turkey [17]. Due to the historical political tension between the Turkish government and the Kurds, Ankara responded by closing borders to the influx of Kurdish refugees [39] and referring to its geographical limitation to the 1951 Geneva Convention, which means it only recognises refugees fleeing persecution from European events [40]. At this point, over 200,000 Kurds were stuck in miserable conditions in the mountainous border region separating Iraq from Turkey [41]. Turkey’s strategic importance as a member of NATO deflected any potential internarial pressure to receive Iraqi Kurds against its will [5, 42]. Instead, on 5 April 1991, the US-led coalition passed the UN Security Council’s Resolution 688 and moved to protect Kurds by carving out a ‘safety zone’ in northern Iraq [13], justifying it by the danger a ‘massive’ refugee movements ‘across international frontiers and cross border incursions’ represented to ‘international peace and security’. ([43], p. 31). However, since the Resolution was not adopted under Chapter VII, it did not sanction the use of force [5].

The Resolution reaffirmed Iraq’s ‘sovereignty, territorial integrity and political independence’, to alleviate the fear that this intervention would advance the Kurdish quest for self-determination. It was a win-win for the allied powers- it prevented the mass refoulement of refugees at Turkey’s border, which reinforced their commitments to human rights norms and simultaneously weakened Saddam Hussein through the creation of a self-ruling militarised unit within Iraq [42]. Protected by a no-fly zone above the 36th parallel, six Iraqi Kurdish camps were located across 10,000 square kilometres [40] with troops on the ground to ward off further aggression. Thus, the Operation Provide Comfort (OPC) became the first UN-endorsed international intervention into a sovereign state [12].

The literature generally agrees on the success of the Iraqi safety zones [13, 26, 39, 44, 45]. However, the OPC deviates from the traditional model of safety zones where all parties to the conflict agree that a designated location will be exempt from violence ([43], p. 32). The OPC-established zones were accessible to the internally displaced Iraqi Kurds in the northern mountainous border territories. A leading Refugee law scholar, Goodwin-Gill and Hathaway have suggested that sending back refugee applicants to these designated areas is consistent with the 1951 Geneva Convention, whereas Ghiranne’s [26] research points to the conclusion that these territories were free from the persecution linked to the Convention grounds or any new persecution, which would be comparable in intensity.

It must be considered that, potentially, Iraqi safety zones pass the IPA test in jurisdictions with a broader interpretation of the Convention and thus seldom turn to IPA. Some jurisdictions apply the presumption that an IPA is not viable where the persecutor is the State. In OPC that, of course, is the case. Kurdish Iraqi’s fled the systemic violence directed at them by the centralised apparatus of the Iraqi government. However, in this case there is sufficient evidence of the capacity to protect within the suggested IPA. The presence of the coalition forces represented a successful deterrent to Iraqi aggression, signalling the international community’s commitment to ensuring Iraqi Kurds’ safety. What cemented the OCP’s sufficient safety conditions without the Iraqi government’s official recognition was the two parallel, although unassociated, settlements with the Iraqi military leaders. On 18th April 1991b, the UN and the government of Iraq signed a Memorandum of Understanding (MOU) whereby Bagdad attested to several provisions preserving the rights of its citizens, simultaneously sanctioning the UN’s presence on Iraqi territory and its ‘efforts to promote the voluntary return home of Iraqi displaced persons’ (para 2) [46, 47]. This MOU was superseded by an agreement on the 25th May, which allowed for the deployment of 500 UN Guards (UNGCI) to the UN Humanitarian Centres [4]. Although McQueen [4] correctly points out that the arrangements secured by the MOU were substituted for the provision of the safety zones, it was found to be significant on 15 July when coalition troops departed Iraq. After the coalition withdrew on 15th July, continued deterrence was ensured by the no-fly zone over Iraq’s northern territories [13].

However, whilst the OPC has met the criteria for an IPA, citing OPC as a precedent for safety zones should be done with discretion, due to the exceptional circumstances which accompanied its creation. First of all, the decision to establish safety zones was taken against the background of Iraq’s military defeat during the Gulf War, making the coalition’s warnings a credible and effective deterrent. Furthermore, it is worth mentioning the role of state interests, which ensured OPC’s successful enforcement. In the aftermath of the Desert Storm bombing campaign, which forced Iraq to end the occupation of Kuwait, the US was unwilling to diminish this victory by providing inadequate protection to the Iraqi Kurds, a situation, which attracted broad media attention internationally. Put together, this exceptional constellation of factors contributed to the success of the Iraqi safety zones.

3.2 Safety zones in Bosnia and herzegovina

On 29 February 1992 Bosnia held a referendum for independence, with 99.4% voters in favour of breaking away from Yugoslavia [4], plunging Bosnia into a three-way ethnic conflict between Muslim Bosnians and Catholic Croats and Orthodox Serbs [13]. By October 1993, there were 800,000 refugees outside of Bosnia, and 1.2 million internally displaced within the country (United States Department of State, 1994). The encirclement of numerous Muslim-majority towns in western Bosnia and Eastern Serbia by Bosnian Serbs finally provoked a reaction from the UN Security Council, who selected six towns and cities, deployed their military personnel and declared them ‘safe areas’, beginning with Srebrenica [16].

Srebrenica proved a disastrous endeavour [17, 26] (Regueiro, 2017) and failed to meet the baseline criteria for a lawful IPA on several grounds. Firstly, it did not satisfy the accessibility criteria, as enemy forces surrounded Srebrenica. UN documents reported about the ‘Serb obstruction of the passage through the UN-protected areas of convoys for humanitarian relief and the United Nations Protection Force’s (UNPROFOR) resupply’. This is supported by evidence that armed conflict prevailed in Srebrenica’s surrounding areas from mid-1993 to mid-1995 [26]. Furthermore, the Srebrenica safety zone did not successfully preclude persecution based on Convention grounds or risk of refoulement. The execution of approximately 7336 Bosnian Muslims ‘by Bosnian Serbs outside the enclave’ during the 6–11 July 1995 offensive is the manifestation of persecution in its purest form.

Safety zones are evaluated against their ability to protect and it is undeniable that Srebrenica fails this test. However, there were several political decisions, as well as deviation from the UN’s safety zone enforcement guidelines on the ground, which paved the way for this failure. Firstly, the designation of Srebrenica as a safety zone was not recognised by all of the parties to the conflict [48]. The possibility of Bosnian Serbs having an aggressive reaction to this was recognised by the UNCHR as it implored the UN to turn to safety zones ‘as the last option’. However, Bosnia’s low strategic importance to the international community, and the strong national security incentives to stem refugees flows from the Balkans, meant this advice was discounted. The geographical proximity meant that European MSs experienced the highest flows of refugees since WWII [26] and created an urgency in the European states to reduce net migration. Thus, during the International Meeting on Humanitarian Aid to the Victims of the Conflict in the Former Yugoslavia on 29 July 1992, the Slovenian delegation proposed ‘safe havens’ in Bosnia ‘to avert new flows of refugees and displaced persons’ [49]. The idea was met with broad enthusiasm [50] and was even endorsed by the UN Special Rapporteur on the Situation of Human Rights in the Territory of Former Yugoslavia, Tadeusz Mazowiecki [51]. Parallel to this, the UN High Commissioner for Refugees, Sadako Ogata alluded to ‘the right to remain’ [52]. This ‘preventing protection’ is informed by the logic that protecting the basic right of the individuals ‘not to become a refugee’ [52] and not to be forced into exile would discourage refugee flows. To shift the attention from the fundamental human rights principle of the ‘right to leave’ (UDHR, 1949, art.13(2)) towards a ‘right to remain’ in a war-torn environment is to weaken the ground for seeking refuge abroad. Thus, it becomes obvious, that Bosnian safety zones served the purpose of an asylum substitute [5, 26, 50, 53] and a threat to the institution of asylum. In practice, this threat was quickly manifested as refugee status was refused due to the availability of a ‘safe haven’ i.e. IPA within Bosnia. States justified their overall unwillingness to accept refugees based on the ‘right to remain’.

In a situation where not all of the parties to the conflict recognise proposed ‘safety zones’, international law prescribes the protection of such areas on the ground [54]. In contrast to Iraq, where national troops and the Iraqi Kurdish peshmerga forces protected the zones [17], the power imbalance between the Bosnian Muslims and the Bosnians Serbs was a direct result of Security Council’s Resolution 713 which led to the imposition of an arms embargo over the entire of former Yugoslavia for the duration of the conflict [4]. This, along with the NATO no-fly zone, prevented Croatia and Bosnia from exercising their right to self-defence.

In Bosnia, these designated areas were under the protection of the UN and UNPROFOR. However, effective coordination was undermined by the fact that the Security Council’s political decisions were taken in New York [4]. This ensured, that the UN troops in Bosnia were slow to react to the developments in Srebrenica. Weak response capacity of UNPROFOR was exacerbated by its inadequate numbers. The UN Security Council decided to ignore the Secretary-General’s and UN Force Commander’s demand for 34,000 troops, opting for the ‘light option’ of 7600 troops (Security Council, 1993). Although, in theory, authorisation to turn to airpower for defence purposes was granted by the UN Security Council, these were considered incompatible with peacekeeping mission by the Force Commanders in Bosnia [1]. Eventually, UNPROFOR’s mandate was extended, allowing it to ‘act in self-defence’ and taking ‘the necessary measures including the use of force’ (para 9.). However, enforcement measures were not provided [26], as UNPROFOR meant to simply ‘deter attacks’ (para 5) directed at the safety zones. Some scholars, such as Posen [16] find it astounding that only two out of six safety zones were taken by the Serbs, given the limited NATO support in these territories.

Other than designating Srebrenica as ‘safe’, the resolution did not elucidate on how to enforce that objective [55]. Instead, the resolution was left open for interpretation by the UNPROFOR who pushed for a demilitarisation agreement in Srebrenica [56] and left the responsibility to ensure the area was ‘safe’ on the shoulders of Serbs and the Muslims [57]. However, while Srebrenica was demilitarised, against the UN’s recommendations, the remaining five Bosnian safe areas were not [58]. The continued military activity within the safety zones meant that the demilitarisation agreement was rendered meaningless immediately. Landgren [5] argues that these unprovoked attacks not only defeated the inherent purpose of ‘safety zones’ but made the Bosnian Serb retaliation and offensive inevitable. Many similar underlying issues detected in Bosnia, particularly underling geopolitical interests, can be seen in the case of Turkish safety zones.

3.2.1 The creation of Turkish safety zones in Syria’s northern territories: Context

Since the outbreak of the civil war in Syria in March 2011, Turkey has been the leading advocate for the establishment of a ‘safety zone’ in northern Syria [59], calling for ‘United Nations […] IDP camps within Syria’, which should have ‘full protection’, and reminding the Council members of ‘the cost of procrastination’ citing Srebrenica, Halabja and Gaza [60]. However, the divided Security Council did not act upon Turkey’s request. In the US Senator McCain proposed a non-binding resolution for safety zones in Syria (S.Res.4242012), but then-president Barack Obama refused to act upon it. Meanwhile, the then-presidential candidate Donald Trump committed himself to the establishment of a ‘beautiful safe zone’ on a ‘big piece of land’ in Syria [61], however, upon his election, his proposed solution failed to materialise. The idea was quietly shelved by the international community, until the concept of safety zones took the form of a ‘de-escalation zone’ agreement, a fruit of the Astana talks on 4th May 2017 between Turkey, Russia and Iran, which sought to curtail the flow of Syrian refugees [62]. However, although de-escalation zones aim to halt hostilities in designated areas, they cannot be equalised with safety zones, due to their lack of protective framework.

Instead, a new safety zone scheme was promogulated by President Erdogan, who presented the earliest official draft during the UN General Assembly meeting in September 2019 [63]. The safety zone, as initially envisioned by Erdogan, was to be a 480-kilometre long and 30-kilometre deep strip of land spreading from the Syrian-Iraq border to the city of Manjib in the west, enclosing governates of Aleppo, Hassakah and al-Raqqa within it [64]. However, with little enthusiasm from the international community, Turkey resorted to threats. In September 2019 Erdogan stated that in the light of ‘handling this burden alone’, Turkey ‘will be forced to open the gates’ [65]. Eventually, blaming the US for delaying the proposal’s execution, Ankara launched a cross-border offensive [66]. On October 9th, 2019, 3 days after President Trump (2019) announced on Twitter that ‘it is time for the US to get out of these ridiculous endless wars’, Turkish Armed Forces (TAF) and Turkey-backed Syrian National Army (SNA) launched Operation Peace Spring (OPS) targeting the American-supported SDF [67]. The fighting between the Turks and the SDF culminated in a ceasefire agreement between Ankara and Washington on October 18th.

The ambitious scale of the project first proposed was moderated under a MOU between President Erdogan and Russian President Vladimir Putin in Sochi on 22nd October 2019 [68]. Following the principles spelt out in the Sochi Agreement, Turkey successfully added 120 kilometres of territory connecting the towns of Ras al-Ayn and Tal Abyad to the pockets of northern Syria it has seized from the Kurdish fighters and the Islamic State (IS) in military operations between 2016 and 2018 [69]. The agreement was hailed as ‘historic’ by Erdogan [70]. It allowed Russia to fill the power vacuum left after the withdrawal of American troops, thereby placing Moscow as a regional power broker in the Middle East. It also ended the Kurds’ domination over a 440-kilometre border as Syrian border guards and Russian military police acquired most of Kurdish-led Syrian Democratic Forces (SDF) frontier posts [71], securing Turkish long-term interest in the region. The joint efforts of Turkey and Syria to assist in the voluntary return of Syrian refugees was sanctioned under Article 8 of the agreement [72].

Until now, Turkey’s interventions have successfully established four safety zones, each named after an armed operation during which Ankara annexed tracts of land. The Euphrates Shield Operation between 2016 and 2017 concluded with the Turkish army seizing the northern areas of Aleppo Governate (Al [62]). A year later, Olive Branch Operation secured Turkish control over north-western Aleppo Governate [73]. In 2019, territories east of the Euphrates River were added to the Turkish sphere of influence with the Peace Spring Operation (PSO) [74]. During the final series of cross-border operations in 2020, Turkey expanded the safety zones by sections of Idlib through the Spring Shield Operation [75]. This paper further explores some reasons why Turkish national interests may have been involved in the country’s decision to develop these safety zones.


4. Findings and discussion

The Turkish governed safety zones in northern Syria, similarly to the safety zones in Srebrenica, have been unsuccessful in meeting the minimum criteria for a legal IPA on numerous counts. Several factors contribute to the inaccessibility of the safety zones. At the logistical level, the journey to northern Syria is seriously obstructed by the ‘financial and logical cost of navigating checkpoints under the control of different groups’ (Kayali, 2020). The financial challenge of reaching the safety zone due to financial reasons is reflected in the interviews collected in Al Bab, where two out of seven respondents emphasised the ‘difficult’ (interview 15 2020) and ‘hard’ (interview 16 2020) nature of the flight, citing inflated transport costs as one of the reasons. Whereas respondents in Tal Abyad did not speak of the financial barrier of reaching the designated territories, the statement of an interviewee in Al Bab illustrates the degree of costs in question: ‘I paid everything I had to get here. I had two cars when I was in Maskana, now I have none’ (interview 11 2020). Travel may also be impeded by the quality of roads, due to the underdevelopment of these territories (Kayali, 2020) and their ‘marginal’ importance prior to the establishment of the IDP camps (interview 16 2020). The deserted geography of Jarabulus is well encapsulated by this quote: ‘you have to wait five hours to find a car to take you from here’ (interview 16 2020). Most importantly, however, a journey to Turkish safety zones may prove physically perilous. Out of 19 IDPs interviewed in three different cities falling within the parameters of the Turkish safety zones, 13 respondents revealed that they travelled to safety zones amid bombing. Respondents would recall ‘helicopters throwing bombs’ (interview 13 2020) and ‘gaols, tortures’ (interview 12 2020) on the way to the safe areas. The ‘airstrikes over the road’ (interview 11 2020) would elongate the passage to safety zones, as one respondent recalled that it took him 2 days (interview 2, 2020). The collected data indicates that entering the zones through the eastern border requires passage through the Kurdish Regional Government (KRG), which is met with the possibility of being stranded at the Iraqi/Syrian border, due to the month-long border closures caused by the COVID-19 pandemic (Kourdi, 2020). This falls short of the requirement to be ‘practically, safely and legally’ [76] accessible to the claimant, both internally and from the outside [21] and thus do not meet the requirement of IPA’s ‘accessibility’ standard.

The Syrian safety zones have also failed to extinguish a reasonable chance of persecution for a Convention refugee reason. Out of 19 respondents, seven described the Turkish zones as ‘safe’, while three respondents reflected that their relocation to the safety areas resulted in a marginal improvement in safety, and nine respondents did not engage in the safety evaluation of the territories. A closer analysis of the primary data displays evidence of persecution enumerated in the Geneva Convention. The collected interviews revealed that expressing political opinion that are critical of the political current status quo in Turkish safety zones carries the risk of arrest and potential torture, regardless of the ‘margin of freedom’ to one’s opinions, which exists in these areas (Kourdi, 2020). Furthermore, as previously mentioned in the Iraqi safety zone case, numerous jurisdictions have a strong presumption against finding an IPA where the State is the actor of persecution [77]. The interviews collected from the Turkish safety zones indicate that the respondents fear the retaliation of the State, as opposed to the other Parties to the conflict. In the words of one of the interviewees: ‘We are afraid of the regime only, there are new arrangements we do not know about’ (interview 11 2020). This fear is supported by Kourdi (2020): ‘the regime has nothing to lose and will continue its advancement until they regain the last meter of the country’. However, this presumption is rebuttable if compelling evidence exists to suggest that meaningful protection can be found within the proposed IPA. In the Syrian case, the presence of the Turkish Armed Forces and/or the Turkish-backed military groups such as the Syrian National Army (FSA) (Kourdi, 2020) may in theory deter potential attacks and thus preserve the safety of the IDPs within the zones. However, primary findings suggest that the Turkish military threat has insufficiently prevented aggression. Turkish controlled safety zones are exposed to continuous attacks by both the Kurdish-led forced seeking to re-establish their control over the Turkish annexed land or by the Syrian-Russian military alliance (Kayali, 2020). The on-going contestation over the northern territories, a result how Turkey established dominance over them, lead to a situation whereby Syrian nationals are ‘at risk both inside them and outside the safety zone’ (Kayali, 2020). Considering these findings, it is hard to argue that the Turkish administered safety zones have been established consensually. Thus, the Turkish safety zones have been unsuccessful in ensuring that the underlying cause of flight has been eliminated, contributing further to the assessment that an IPA does not exist.

The safety zones in Syria fail to meet yet another qualification in the IPA test, that is, protection from persecution without a nexus to the Convention grounds. The political instability amid which the safe areas were erected, the territorial struggle which ensued in the aftermath of their creation and the inability to secure their recognition by the central government and the remaining sides to the war, made the safety zones particularly vulnerable to hostilities. Against this backdrop, the Turkish safety zones perform particularly poorly in the previously discussed ‘reasonableness criterion’, which expands the analysis to the provision of basic political, socio-economic and civil human rights and the subjective circumstances of the individual [76]. The on-going in-fighting gives strong grounds to believe that Syrians are re-exposed to other forms of harm of equivalent adversity to those which prompted them to leave the native region. On-going displacement occasioned by violence has been raised as a concern among the respondents. A Syrian proverb has been quoted to capture the ingrained apprehension of future forced migration: ‘the one whose mouth has been burnt by hot milk even blows on yoghurt when eating it’. The interviewee explained the maxim’s relevance: ‘whenever they say armistices are over, we know that we will be displaced again’ (interview 12 2020). All 19 respondents presented a history of displacement, while one interviewee testified that he has been uprooted 50 times (interview 11 2020). It is also important to illuminate the relationship between displacement and penury. Six respondents (interview 3 2020; Interview 3 2020; interview 18 2020; 16 Jarubulus, 2020; interview 17 2020; interview 9 2020) stated that bombing forced them to leave everything they owned behind ‘to survive’ (interview 18 2020) forcing a group of interviewees to abandon their ‘car full of our luggage and […] walk on feet’ (interview 17 2020).

The impoverishment of the Syrian IDPs and the destruction of infrastructure, which accompanied the development of these safety zones (Kayali, 2020) restricts access to livelihood opportunities for those living in protracted displacement- a concern mentioned by all respondents. The state of malaise in the safety zones is broadly reflected in the collected data, as the word ‘nothing’ and ‘bad’ has been used 11 and 8 times respectively to describe the state of services, access to education, humanitarian activity and healthcare in the safety zones. Furthermore, interviewees 3, 5, 6, 12, 14, 15, 16, and 17 (2020) state that neither food nor water wells are provided, whereas one barrel of water costs approximately 1200 Syrian pounds (SYP), and one bag of bread 600 SYP (interview 12 2020). One interviewee commented on this dire situation: ‘those who do not get a job, die from hunger’ (interview 15 2020). However, this is a problematic finding, as in the three cities in which interviews were conducted, jobs appear to be extremely scarce, ‘almost non-existent’ (interview 18 2020), a reality confirmed by all 19 respondents. In Al-Bab it was stated in an interview that a job can be secured 2-3 times in a whole month, for which one would be rewarded circa 10,000 SYP (equivalent of 3,50 GBP), until the surfacing of the next job in 3 months (Interview 11 2020). Due to the absence of employment, children in Al Bab and Jarubulus sell plastic found in waste containers for which they purchase food for the family (Interview 16 in Jarubulus 2020; interview 18 2020). Respondents associated the poor conditions within the safety zones with their lack of recognition in the eyes of their respective local councils (interview 5 in Jarabulus 2020), as one interviewee concluded that Syrian IDPs are seen as ‘ugly ducklings’ (interview 15 2020).

There is little hope for improvement in the near future given that the Security Council failed to adopt two draft resolutions sanctioning cross-border delivery of humanitarian aid in northern Syria [43]. The numerous examples of the lack of decent means of subsistence have been indicated, as they are of relevance to the IPA test when evaluating the possibility of indirect refoulement. Socio-economic destitution may compel IDPs to return an area where he or she would be re-exposed to persecution. An interview with Kourdi (2020) confirms the sporadic occurrence of indirect refoulement, citing an example of a man returning to the Government- controlled areas from the Turkish controlled part of Idleb, where he was subjected to State-inflicted tortures (two broken legs). Looking at these findings, it may be concluded that the current conditions in Al Bab, Jarabulus and Tal Abyad amount to an existence below a bare subsistence level (Existenzminium). The Turkish safety zones, whether in north-western or north-eastern Syria do not satisfy the reasonableness criterion of the IPA inquiry, as it would be ‘unduly harsh’ to relocate there and thus, unreasonable and illegal.

4.1 Key findings—A critical analysis of Turkey’s motives

On October 31st, 2019, President Erdogan applauded Turkey’s ‘lengthy efforts’, which bore the fruits of the ‘most peaceful and liveable places in Syria’- the safety zones [78]. The primary findings enclosed in the above section, however, suggest that Erdogan’s assessment of the zones is Panglossian and evidence to the contrary: they are hardly peaceful or liveable. In fact, the collected data from As-Suwayda, Al-Qamishli, Damascus, Dayr Al Zawr, Al-Hasakah, Tal Rifaat suggest that livelihood opportunities (apart from Dayr Al Zawr), are notably higher than in the Turkish safety zones analysed (with the exception of the more central territories of Al-Bab [Kourdi, 2020]). This assessment is based on collected information on the availability of education and jobs (8/10 respondents indicated unrestricted access to education for children, and 4/10 were unemployed where two respondents being current students and one was a retired military man). However, the evaluation of safety varied across the cities. In Damascus, Al-Hasakah, As-Suwayda and Al-Qamishli 10 respondents unanimously declared the areas as ‘safe’ (interview 24, 25, 26, 27, 29 2020) both respondents from Dayr Al Zawr described feeling ‘unsafe’ (interview 22 and 23 2020) and two respondents in Tal Rifaat expressed fear of possible war and the danger represented by a potential Turkish invasion (interview 20 and 21 2020). It appears, therefore, that even the territories, which permit financing life at levels of minimum subsistence would fail an IPA test. As such, the 2019 Danish Immigration’s Service’s decision to revoke the temporary protection status of three Syrian applicants from Damascus [79] which appealed to the IPA language of Damascus as a ‘safe’ relocation alternative, sets a dangerous precedent in the EU and shows the potential of an incorrectly applied IPA notion in restricting access to asylum. The question, which begs attention, therefore, is if Turkey was truly motivated by altruistic aspirations to help ‘Syrian brothers and sisters’ [80] what explains the geographical designation for the safety zones in northern Syria, which hardly allow for social and economic existence?

Sinem [59] argues that the incentives for the invocation of such zones in fact shifted to match Ankara’s political priorities. Between 2011 and 2014, Turkey’s vision of protected areas south of its borders was primary actuated by its desire to see the demise of the Syrian president Bashar al-Assad, as Ankara encouraged anti-Assad rebels to settle in these territories [81]. The expectation of Bash al-Assad’s prompt defeat compelled Turkey to announce an open-border policy for those fleeing the Syrian civil war. However, this decision quickly proved to be a miscalculated strategic move, which created unforeseen ramifications for Turkey. The paralysis of the Western asylum system and the endurance of Assad’s regime proved to impede a fair international burden sharing in the light of the Syrian refugee crisis. This realisation was reflected in the speech of the then-Turkish foreign minister Ahmet Davutoglu during a Security Council meeting in August 2012: “we feel that the open-door policy of Turkey […] is in fact absorbing any potential international reaction’ [82]. By mid-2018, Turkey hosted 3,584,179 Syrian refugees [18], becoming the largest recipient of Syrian asylum seekers in the world.

The appeal of safety zones as an alternative solution to the unfolding refugee crisis on Turkey’s borders was further fuelled by Ankara’s overstretched reception capacity and the existing legal framework for refugees and asylum seekers. Turkish legal frame emanates from the Settlement Law (1934), the Geneva Convention (1951) and its subsequent Additional Protocol (1967) 81]. However, Turkey retained a geographical limitation to its ratification of the 1951 Convention, whereby in accordance with article 1/3 of the 1967 Protocol only those fleeing events occurring in Europe can be awarded refugee status [83]. Thus, non-European asylum applicants are not permitted to stay, regardless of their successful status determination outcome through UNHCR [81]. Since a category of ‘guest’ is without precedence in international refugee law and in 2011 Turkey devised novel legal and administrative documents including ‘temporary protection regime’ [84] and cemented by the adaptation of the Law on Foreigners and International Protection in 2013 (UN, 2013). Based on the EU Council Directive 2001/55/EC [61, 85], it offers temporary protections to foreigners, and thus Syrians, for those who seek temporary asylum until resettlement. The scale of the exodus towards Turkey added a degree of urgency to finding an alternative solution for the refugee crisis unfolding on its borders was added.

Kourdi (2020) provides a more nuanced and multifaced explanation of Turkey’s political agenda regarding safety zones, which complements the argument derived from the stretched national capacity argument. During the interview, three types of safety zones were discriminated, each with a precise purpose. Type 1 safety zones, encompassing Idlib and northern countryside of Aleppo, is where Turkey would like to implement enforced resettlement programs for IDPs and refugees. However, at the time of writing, the territories hold a strong presence of Syrian rebels and are thus exclusively aspirational. Type 2 zones were inaugurated by Peace Spring Operation, stretching from Tal Abyad to Ras al Ayn and where Ankara intends to complement Type 1 resettlement schemes for Syrian refugees and IDPs, however, its current commitment the execution of this plan is weak, judging by the levels of their development efforts directed at this strip of land (as compared to north-western Turkish controlled cities). Type 3 zones extend from Ad Darbasiyah to Kobani (and excluding the city of Qamishli) with the foremost intent of eliminating Kurdish Syrian Democratic Forces (SDF), elements from the areas, coming into force with the 2019 Sochi Agreement and scaled down in size from being 30-km to 10-km deep). However, Type 3 zones can be more fittingly described as ‘safety zones’ for Turkish national security concerns, intended to shield Turkey from the SDF, led by the People’s Protection Units (YPG), which Ankara regards as an extension of the Kurdistan Worker’s Party (PKK), rather than ‘safety zones’ as a humanitarian form of protection- the focus of this thesis. Thus, ‘buffer zones’ would be a more fitting category for Type 3 zones, as Kourdi (2020) holds that suppression of the growing Kurdish political ambition of the SDF, which risked energising the repressed self-determination aspirations of the Turkish branch, PKK, left other considerations subordinate to this goal- an argument, which finds support in the available literature [86, 87, 88].

Overall, the argument made in this thesis is that humanitarian considerations find little place in the safety zones discussion. Thus, Syrian safety zone realised a number of Ankara’s national interests: it allowed to absorb future waves of Syrian asylum seekers and for a permanent relocation of those already in Turkey, eliminated the threat that ISIS represented to its borders, impede the Kurdish separatism and finally augment Turkey’s influence in the region by offering a logical site for moderate rebel groups. These aspirations represent a clear departure from Turkey’s previously dominant Kemalist identity geared towards a policy of regional non-involvement, towards a Neo-Ottoman vision, characterised by the return of Turkey’s inclination to play an important and active regional role [89]. The establishment of Turkish unsafe safety zones is a dangerous development, an opinion shared by Professor David Keen [17] during an interview, due to the significant leeway the EU grants Ankara in return for its cooperation to curtail the refugee flows into the Union. Thus, although from the outset of the Syrian war, Turkey underscored the ethical and moral facet informing the need for safety zones, it appears that Turkey is cloaking political ambitions under a humanitarian cover [59]. As such it appears that Turkey’s safety zones are not only represent a restrictive refugee policy instrument, but also a geopolitical one and thus, a threat to the institution of asylum.


5. Conclusion

This paper considered whether safety zones represent a complementary policy to international protection or a threat to the institution of asylum. On one hand, safety zones are a relatively low-cost panacea for the shortcomings of contemporary asylum regime. Since those who are unable or unwilling to leave their country of origin in search of asylum are excluded from protection in Refugee Law, it is customarily argued that safety zones correct this lacuna in law by providing immediate in-country protection of civilians from bodily harm. On the other hand, however, pursuing the establishment of safety zones is associated with the risk of impeding refugee protection. The belief in safety zones’ capacity to provide meaningful protection, comparable to that of refugee law, provides justification for host states to tighten their borders and seek their creation without ensuring that all conditions specified in international law are created prior to their inauguration. This, in turn, may result in the provision of inadequate levels of safety to those who are often exposed to persecution for a Convention refugee reason and thus diluting the right to asylum.

Firstly, it has been shown that safety zones may be established in accordance with international law and most importantly, the IPA notion. While the Iraqi safety zones departed from the recommended model of a safety zone as envisioned by international law, due to OPCs inability to secure recognition from all the parties to the conflict, the Iraqi safety zone was accessible, free from the original risk of persecution and persecution without a Convention nexus and facilitated the safe repatriation of Iraqi Kurds. However, it was also shown that OPC was erected amid exceptional circumstances, which conditioned its ability to fulfil its protection objective. However, the success of the Iraqi safety zone has not been reproduced in the case of Bosnia and Syria, both of which failed to satisfy all three baseline conditions of a legal IPA. Furthermore, all three case studies revealed that the creation of safety zones in each of the respective countries was informed by geopolitical interests, as well as the desire to curb migratory movements.

However, former UN Secretary General, Kofi Annan, noted that IDPs are ‘among the most vulnerable of the human family’ [90] and thus safety zones should not be prematurely disqualified as an instrument for their protection. This paper has identified a number of conditions, which have to be met for a safety zones to be effective identified through historical and primary data analysis. That is, safety zones (a) have to be clearly established (b) be in towns with livelihood opportunities (c) not be a contested territory (d) law should be revised to address the problem of unclear accountability for human rights violations that may take place within the safety zones. However, it appears that the most important condition is that safety zones are used as a complementary policy to the institution of asylum and not its alternative.



Interview with Sara Kayyali, Syria researcher in the Middle East and North Africa Division in Human Rights Watch (HRW). 9 August 2020

Interview with Eyad Kourdi, Freelance journalist specialising in the Middle East and working with multiple media outlets such as CNN. 20 August 2020

Interview with Dr. David Keen, Professor of Complex Emergencies at London School Economics and a specialist in the field of Safety Zones. 28 July 2020

Interview with Interviewee 1, conducted by Nour El-Deen Hammadi, Tal Abyad, 30 July 2020

Interview with Interviewee 2, conducted by Nour El-Deen Hammadi,Tal Abyad, 30 July 2020

Interview with Interviewee 3, conducted by Nour El-Deen Hammadi Jarabulus, 4 August 2020

Interview with Interviewee 4, conducted by Nour El-Deen Hammadi Jarabulus, 4 August 2020

Interview with Interviewee 5, conducted by Nour El-Deen Hammadi Jarabulus, 4 August 2020

Interview with Interviewee 6, conducted by Nour El-Deen Hammadi Jarabulus, 4 August 2020

Interview with Interviewee 7, conducted by Nour El-Deen Hammadi Jarabulus, 4 August 2020

Interview with Interviewee 8, conducted by Nour El-Deen Hammadi Jarabulus, 4 August 2020

Interview with Interviewee 9, conducted by Nour El-Deen Hammadi Jarabulus, 4 August 2020

Interview with Interviewee 10, conducted by Nour El-Deen Hammadi Jarabulus, 4 August 2020

Interview with Interviewee 10, conducted by Nour El-Deen Hammadi, Al-Bab, 10 August 2020

Interview with Interviewee 11, conducted by Nour El-Deen Hammadi, Al-Bab, 10 August 2020

Interview with Interviewee 12, conducted by Nour El-Deen Hammadi, Al-Bab, 10 August 2020

Interview with Interviewee 13, conducted by Nour El-Deen Hammadi, Al-Bab, 10 August 2020

Interview with Interviewee 14, conducted by Nour El-Deen Hammadi, Al-Bab, 10 August 2020

Interview with Interviewee 15, conducted by Nour El-Deen Hammadi, Al-Bab, 10 August 2020

Interview with Interviewee 16, conducted by Nour El-Deen Hammadi, Al-Bab, 10 August 2020

Interview with Interviewee 17, conducted by Nour El-Deen Hammadi, Al-Bab, 10 August 2020

Interview with Interviewee 18, conducted by Nour El-Deen Hammadi, Al-Bab, 10 August 2020

Interview with Interviewee 19, conducted by Nour El-Deen Hammadi, Al-Bab, 10 August 2020

Interview with Interviewee 20, conducted by Nour El-Deen Hammadi, Tal Rifaat, 13 August 2020

Interview with Interviewee 21, conducted by Nour El-Deen Hammadi, Tal Rifaat, 13 August 2020

Interview with Interviewee 22, conducted by Nour El-Deen Hammadi, Dayr az Zawr, 14 August 2020

Interview with Interviewee 23, conducted by Nour El-Deen Hammadi, Dayr az Zawr, 14 August 2020

Interview with Interviewee 24, conducted by Nour El-Deen Hammadi, Al-Qamishli, 17 August 2020

Interview with Interviewee 25, conducted by Nour El-Deen Hammadi, Al-Qamishli, 17 August 2020

Interview with Interviewee 26, conducted by Nour El-Deen Hammadi, Al-Suwayda, 18 August 2020

Interview with Interviewee 27, conducted by Nour El-Deen Hammadi, Damascus, 20 August 2020

Interview with Interviewee 28, conducted by Nour El-Deen Hammadi, Al-Hasakah, 21 August 2020

Interview with Interviewee 29, conducted by Nour El-Deen Hammadi, Al-Hasakah, 21 August 2020


  1. 1. Orchard P. Revisiting humanitarian safe areas for civilian protection. Global Governance. 2014;20(1):55-75
  2. 2. Convention Relating to the Status of Refugees. Treaty no. 2545. United Nations Treaty Series. Vol. 1891951. pp. 137-221. Available from: [Accessed: July 4, 2020]
  3. 3. Universal Declaration of Human Rights (UDHR) 1948, (Art.14(2)), 1948
  4. 4. McQueen C. Humanitarian Intervention and Safety Zones: Iraq, Bosnia and Rwanda. New York: Palgrave Macmillan; 2005
  5. 5. Landgren K. Safety zones and international protection: A dark grey area. International Journal of Refugee Law. 1996;7(3):436-458
  6. 6. Sandoz Y. The Establishment of Safety Zones for Persons Displaced within their Country of Origin. In: Al-Naumi N, Meese R, editors. International Legal Issues Arising under the Nations Decade of International Law. The Hague: Martinus Nijohoff; 1995
  7. 7. Long K. In search of sanctuary: Border closures, ‘Safe’ Zones and refugee protection. Journal of Refugee Studies. 2012;26(3):458-476
  8. 8. Gilard EC. Protection of civilians in the conduct of hostilities. In: Livoja R, McCormack T, editors. Routledge Handbook on the Law of Armed Conflict. New York: Routledge; 2016
  9. 9. Henckaerts J-M, Doswald-Beck L, Alvermann C, Dörmann K, Rolle B. Protected zones (Rules 35-37). In: Customary International Humanitarian Law. Cambridge: Cambridge University Press; 2005. pp. 119-126. DOI: 10.1017/CBO9780511804700.015
  10. 10. Brinie R, Welsh J. Displacement, protection and responsibility: A case for safe areas. Global Responsibility to Protect. 2018;10(1):332-361
  11. 11. Sandoz Y. The Establishment of Safety Zones for Persons Displaced within their Country of Origin. Geneva: International Committee of the Red Cross; 1994
  12. 12. Hyndman J. Preventing, palliative, or punitive? Safe spaces in bosnia-herzegovina, somalia, and Sri Lanka. Journal of Refugee Studies. 2003;16(2):167-185
  13. 13. Recchia S. The paradox of safe areas in ethnic civil wars. Global Responsibility to Protect. 2018;10(3):362-386
  14. 14. Outram Q. Cruel wars and safe havens: Humanitarian Aid in Liberia 1989-1996. Disasters. 1997;21(1):189-205
  15. 15. Kaldor M. New and Old Wars: Organised Violence in a Globalised Era. Cambridge: Polity Press; 2012
  16. 16. Posen B. Military responses to refugee disasters. International Security. 1996;21(1):72-111
  17. 17. Keen D. Anything but safe: Problems with the protection of civilians in so-called ‘sade zones’. In: London School of Econimics Working Paper No.17-187. Department of International Development; 2018. Available from: [Accessed: July 23, 2020]
  18. 18. UNHCR, ‘Syria Regional Refugee Response’, Refugees Operational Portal. 2018. Available from: [Accessed: August 20, 2020]
  19. 19. Edwards A. Human rights, refugees, and the right ‘To enjoy’ Asylum. International Journal of Refugee Law. 2005;17(2):293-330
  20. 20. Phuong C. The International Protection of Internally Displaced Person. Cambridge: Cambridge University Press; 2004
  21. 21. UN High Commissioner for Refugees (UNHCR), Guiding Principles on Internal Displacement, 1998. ADM 1.1, PRL 12.1, PRO00/98/109, Available from: [Accessed: July 4, 2020]
  22. 22. Gilbert G, Rüsch AM. Creating Safe Zones and Safe Corridors in Conflict Situations:Providing Protection at Home or Preventing the Search for Asylum? (Policy Brief, Kaldor Centre for International Refugee Law, University of New South Wales. 2017
  23. 23. Chowdhury SR. A response to the refugee problem in post cold war era: Some existing and emerging norms of international law. International Journal of Refugee Law. 1995;7(2):100-118
  24. 24. Clark T. Human Rights and Expulsion: Giving Content to the Concept of Asylum. International Journal of Refugee Law. 1992;4(2):189-204
  25. 25. Chimni BS. The Incarceration of Victims: Deconstructing Safety Zones in N Al-Nauimi and R Meese, International Legal Issues Arising under the United Nations Decade of International Law: Proceedings of the Qatar International Law Conference ’94 (Martinus Nijhoff 1995)
  26. 26. Ghrainee BN. Safe zones and the internal protection alternative. International and Comparative Law Quarterly. 2020;2(1):335-364
  27. 27. Hathaway, James C. Framing refugee protection in the new world disorder. Harvey, CJ. Co-author. Cornell International Law Journal 34, no. 2 2001:257-320
  28. 28. Frederick LA. Psychological Wellness of Refugees: Issues in Qualitative and Quantitative Research. New York: Berghahn Books; 2000
  29. 29. Storey H. The internal flight alternative test: The jurisprudence re-examined. International Journal of Refugee Law. 1998;10(3):499-532
  30. 30. Yi C. New zealand's approach to the internal protection alternative in refugee status determinations. New Zealand Journal of Public and International Law. 2016;14(2):249-284
  31. 31. UN High Commissioner for Refugees (UNHCR). Guidelines on International Protection No.4: Internal Flight or Relocation Alternative’ Within the Context of Article 1A(2) of the 1951. Geneva Convention and/or 1967 Protocol Relating to the Status of Refugees; 2003. Available from: [Accessed: July 12, 2020]
  32. 32. Marx R. The Criteria of applying the internal flight alternative test in national refugee status determination procedures. International Journal of Refugee Law. 2002;14(3):179-218
  33. 33. Hathaway J. The Law of Refugee Status. Toronto: Butterswoths; 1991
  34. 34. Ghrainne B. The international protection alternative inquiry and human rights considerations irrelevant or indispensable. International Journal of Refugee Law. 2015;27(1):29-51
  35. 35. Kelly N. Internal flight/relocation/protection alternative: Is it reasonable. International Journal of Refugee Law. 2002;14(1):4-44
  36. 36. UN High Commissioner for Refugees (UNHCR), An Overview of Protection Issues in Europe: Legislative Trends and Positions Taken by UNHCR, 1995, European Series, 1(2). Available from: [Accessed: July 10, 2020]
  37. 37. UN High Commissioner for Refugees (UNHCR), Handbook and Guidelines on Procedures and Criteria for Determining Refugee Status under the 1951 Convention and the 1967 Protocol Relating to the Status of Refugees, 2011. HCR/1P/4/ENG/REV.3, Available from: [Accessed: July 13, 2020]
  38. 38. Hathaway JC, Foster M. Law of Refugee Status (Cambridge University Press; 2014
  39. 39. Long, K. No Entry! A Review of UNHCR’s Response to Border Closures in Situations of Mass Refugee Influx. Geneva: UNHCR; 2010
  40. 40. Malanczuk P. The Kurdish Crisis and Allied Intervention in the Aftermath of the Second Gulf War. European Journal of International Law. 1991;2(2:114-134
  41. 41. Weller M. Iraq and Kuwait: The Hostilities and Their Aftermath. Cambridge: Cambridge Grotius Publications Ltd 1993; 1993
  42. 42. Frelick B. ‘Preventing Refugee Flows: Protection or Perlil’ World Refugee Survey 1993. Washington DC: US Committee for Refugees; 1993
  43. 43. United Nations Security Council. In Separate Votes, Security Council Fails to Adopt Two Drafts Resolution Renewing Cross-Border Mechanism for Humanitarian Aid Delivery into Syria [Press release]. 2020. Available from: [Accessed: August 28, 2020]
  44. 44. Cotran E. The establishment of a safe haven for the kurds in Iraq. In: Al-Nauimi N, Meese R, editors. International Legal Issues Arising Under the United Nations Decade of International Law. London: Martinus Nijhoff Publishers; 1995
  45. 45. UN High Commissioner for Refugees (UNHCR), The State of the World’s Refugees 2000: Fifty Years of Humanitarian Action, 2000. Available from: [Accessed: July 14, 2020]
  46. 46. United Nations Security Council. Memorandum of Understanding of Understanding between the Government of the Republic of Iraq and the United Nations. 1991
  47. 47. United Nations Security Council. Security Council Resolution 688 (1991)[Iraq], 5 April 1991, S/RES/688. 1991. Available from: [Accessed: July 20, 2020]
  48. 48. Yamashita H. Humanitarian Spaces and International Politics: The Creation of Safe Zones. London: Routledge; 2004
  49. 49. Republic of Slovenia. Ministry of Foreign Affairs. In: Proposal Concerning the Measures for Voluntary Return, Home of the Displaced Persons and Refugees from Bosnia and Hercegovina. 1992
  50. 50. Bartusicki M. EU States and the refugee crisis in the former Yugoslavia. Refuge. 1994;14(3):32-36
  51. 51. United Nations. Commission of Human Rights. Situation of Human Rights in the Territory of Former Yugoslavia. 1993. Available from: [Accessed: August 15, 2020]
  52. 52. UNHCR, 1993. Statement By Mrs. Sadako Ogata, United Nations High Commissioner For Refugees. Geneva: To The Forty-Ninth Session Of The Commission On Human Rights; 1993. p. 3. Available from: [Accessed: July 25, 2020]
  53. 53. Franco L. An examination of safety zones for internally displaced persons as a contribution towards prevention and solution of refugee problems. In: Al-Nauimi N, Meese R, editors. International Legal Issues Arising Under the United Nations Decade of International Law. London: Martinus Nijhoff Publishers; 1995
  54. 54. Gillard E-C. Safe areas’: The international legal framework. International Review of the Red Cross. 2019;99(3):1075-1101
  55. 55. Silber L, Little A. The Death of Yugoslavia. London: Penguin Books; 1996
  56. 56. Rohde D. Endgame: The Betrayal and Fall of Srebrenica: Europe’s Worst Massacre since World War II. Boulder: Westview Press; 1998
  57. 57. Tuathail GO. The ethnic cleansing of a ‘safe area’: The Fall of Srebrenica and the ethics of UN-governmentality. In: Protcor J, Smith D, editors. Geography and Ethics: Journeys in a moral terrain. London: Routlege; 1999
  58. 58. United Nations General Assembly, Report of the Secretary General Pursuant to the General Assembly Resolution 53/35: The Fall of Srebrenica, 1999. A/54/549. Available from: [Accessed: July 14, 2020]
  59. 59. Sinem A. Repatriation to Turkey’s ‘safe zone’ in Northeast Syria: Ankara’s goals and European concerns. (SWP Commment, 1/2020). Berlin: Stifung Wissenschaft und Politik- SWP- Deutsches Institut fur internationale Politik und Sucherheit;
  60. 60. Ministry of Foreign Affairs of Turkey. “Address by Mr. Ahmet Davutoğlu, Minister of Foreign Affairs of Turkey at the 67th United Nations General Assembly, 28 September 2012, New York”. Republic of Turkey, Ministry of Foreign Affairs. Available from: [Accessed: July 20, 2020]
  61. 61. UN High Commissioner for Refugees (UNHCR), UNHCR Position Paper on Relocating Internally as a Reasonable Alternative to Seeking Asylum (The So-Called ‘Internal Flight Alternative’ or Relocation Principle’), 1999. Available from: [Accessed: July 17, 2020]
  62. 62. Al Jazeera. ‘Russia: Syrian Safe Zones Plan Comes into Effect’. 2017. Available from: [Accessed: August 1, 2020]
  63. 63. Wintour P. Recep Tayyip Erdogan proposes ‘safe zone’ for refugees in Syria. The Guardian; 2019. Available from: [Accessed: August 3, 2020]
  64. 64. Maizland L. Who’s who in Northern Syria? In: Council on Foreign Relations. 2019. Available from: [Accessed: August 16, 2020]
  65. 65. MacGregor, M. Turkey threatens to ‘open the gates’, Info Migrants. 2019. Available at: [Accessed: August 13, 2020]
  66. 66. Uras U. In: Jazeera A, editor. Turkey’s Operation Peace Spring in northern Syria: One month on. 2019. Available from: [Accessed: August 8, 2020]
  67. 67. Brussels International Center (no date) Operation Peace Spring: Delusion in Turkey’s Objectives. Available from: [Accessed: August 15, 2020]
  68. 68. Simmons A, Gauthier-Villars D. Russia agrees to help turkey expand buffer zone in syria. Wall Street Journal. 2020. Available from: [Accessed: August 10, 2020]
  69. 69. McKernan B and Borger J, 2019 ‘Turkey and Russia Agree on Deal Over Buffer Zone In Northern Syria’, The Guardian. 2019. Available from: [Accessed: August 10, 2020]
  70. 70. Ant O, Hacaoglu S. Erdogam Says ‘Historic’ Deal With Russia to End Kurdish Threat. Bloomberg; 2019. Available from: [Accessed: August 8, 2020]
  71. 71. Parfitt T, Smith L. Troops Move Into Northern Syria After Deal Struck In Sochi. The Times; 2019. Available from: [Accessed: August 14, 2020]
  72. 72. Al Jazeera. ‘Full Text of Turkey, Russia Agreement on Northeast Syria’. 2019. Available from: [Accessed: August 9, 2020]
  73. 73. Stein, A. Operation Olive Branch: Status update. 2019. Available from:
  74. 74. Turan, B. Turkey’s Military Operation ‘Peace Spring’ and its Aftermath. 2019. Available from: [Accessed: August 25, 2020]
  75. 75. Khaddour K. The Stakes in the Syrian Turkish Border Zone. Carneegie: Middle East Center; 2020. Available from: [Accessed: August 17, 2020]
  76. 76. Hathaway JC, Foster M. International protection/relocation/flight alternative as an aspect of refugee status determination. In: Feller E, Türk V, Nicholson F, editors. Refugee Protection in International Law: UNHCR’s Global Consul- tations on International Protection. Cambridge: Cambridge University Press; 2003. p. 2003
  77. 77. European council on refugee and exiles (ECRE). In: Actors of Protection and the Application of the Internal Protection Alternative: European Comparative Report. 2014. Available from: [Accessed: August 16, 2020]
  78. 78. Mutlu, S. and Orkan, S. Turkey’s Erdogan: ‘Safe Zones in Syria most livable. 2019. Available from: [Accessed: August 16, 2020]
  79. 79. European Council on Refugees and Exiles (ECRE) Denmark: Appeal Board Confirms Rejection of Protection for Three Syrian Nationals. Available from: [Accessed: July 20, 2020]
  80. 80. Gal C. Turkey’s Radical Plan: Send a Million Refugees Back to Syria. New York Times; 2019. Available from: [Accessed: August 21, 2020]
  81. 81. Donelli F. Syrian refugees in Turkey: A security perspective. New England Journal of Public Policy. 2018;30(2):1-9
  82. 82. UNSCOR, 6826th Mtg, UN Doc S/PV.6826 (30 August 2012). Available from: [Accessed: August 20, 2020]
  83. 83. Gular A. Turkey’s geographical limitation: The legal implications of an eventual lifting. International Migration. 2019:1-15. DOI: 10.1111/imig.12683
  84. 84. UN High Commissioner for Refugees. UNHCR Global Report 2012: Turkey. The UN Refugee Agency. 2012. Available from: [Accessed: August 27, 2020]
  85. 85. UN High Commissioner for Refugees (UNHCR). Current Issues in Cessation of Protection Under 1C of the 1951 Refugee Convention and Article 1.4 of the 1969. OAU Convention [Global Consultations on International Protection/Second Track]; 2001. Available from:
  86. 86. Young, M. What are Turkey’s Ultimate Aims in Syria?, Carnegie Middle East Center. 2017. Available from: [Accessed: August 22, 2020]
  87. 87. Conkrar. The Situation in Syria: An Update. Revised Draft Report. 040 GSM 19 E rev.2 NATO Parliamentary Assembly. 2019. Available from: [Accessed: August 23, 2020]
  88. 88. Zandee D. Crisis Alter 1: Turkey and NATO- Living Apart Together. Clingendael Institute; 2019. Available from: [Accessed: August 20, 2020]
  89. 89. Taspinar O. Turkey’s strategic vision and syria. The Washington Quaterly. 2012;35(3):127-140. DOI: 10.1080/0163660X.2012.706519
  90. 90. United Nations High Commissioner for Refugees (UNHCR) Guiding Principles on Internal Displacement, 2004. Available from: [Accessed: August 27, 2020]

Written By

M.A. Natalia Gierowska

Submitted: May 22nd, 2022 Reviewed: June 28th, 2022 Published: October 19th, 2022