Afghan Refugees: Reasons for Flight, Perils of Return

mar 23, 2021

No place in Afghanistan can be considered safe.

William Maley is Emeritus Professor at the College of Asia and the Pacifics, Australian National University i Canberra. He is a recognized authority when it comes to Afghanistan, and he has written many books about Afghanistan, migration and being a refugee. The latest is The Afghanistan Wars, published in 2021. This text is specially written for the project Den onödiga flyktingkrisen (The Unnessecary Refugee Crisis). A Swedish translation was published in Dagens Arena on March 21, 2021.

Photo: Jan-Åke Eriksson. William Maley was the key note speaker at the conference How safe is Afghanistan? in Stockholm on October 4, 2018.

As the case of Afghanistan proves all too clearly, refugees can take flight for many reasons. Some clear-sighted individuals – ‘anticipatory refugees’ – may flee in rational anticipation of the onset of disaster.[i] Others have either experienced direct persecution by virtue of factors recognised in the 1951 Convention Relating to the Status of Refugees, or witnessed people similar to them encountering persecution. Some flee from situations in which deteriorating political conditions or mounting conflict make it likely that persecution will soon materialise – and these situations may themselves have complex aetiologies that require careful analysis.[ii] When questioned as to the reasons for their flight, refugees are unlikely to respond in terms drawn from sociology or international law. Indeed, many run into difficulty because they express their need to flee in terms of hope for a ‘better life’. This can lead bureaucrats to depict them simply as ‘economic migrants’ – although it is worth bearing in mind that this was also the term (Wirtschaftsemigranten) that the Nazis used in the 1930s to describe Jews fleeing Germany.[iii] But for people fleeing Afghanistan, a ‘better life’ can easily mean one that is free of terror, mayhem, and carnage; and even if those in flight also hope that the future will be more prosperous for them than the past, there is little in this that is surprising: who would not entertain such a hope?[iv]

All refugees are individuals: they have their own distinctive stories and experiences. Nonetheless, there are as a rule certain identifiable common factors that come into play in explaining how and why people are unable safely to return to their countries of nationality. They can also be good reasons why people who have initially fled to one country may need to flee to another: all too often, countries of ‘first asylum’ may themselves be repressive or unstable, prompting an onward journey on the part of those desperate for safety or stability in their lives. That said, it is easy to view refugees as passive victims of circumstance, but this should not be at the expense of recognising that refugees very often display courage and initiative which those lucky enough to dwell in stable countries rarely have to exhibit.[v] Their agency can be held against them, for example if they act on their own without bureaucratic approval, but often it means that they can add significant value to the societies in which they settle by virtue of their entrepreneurship and determination. These points are all relevant when one looks at the plight of refugees from Afghanistan.

Some reasons for flight

When looking at reasons why people might flee Afghanistan, it is important at the outset to note that Afghanistan is not a ‘post-conflict state’. On the contrary, the Global Peace Index for 2020 ranked it for the second year running as the least peaceful country in the world.[vi] Whilst the 2001 Bonn Agreement that traced a path for post-Taliban political development created high hopes for future stability, those hopes have not been realised. The drift of US attention to Iraq from 2003 created an opportunity for the Taliban to take shape once again with support from their backers in Pakistan’s Inter-Services Intelligence Directorate (ISI).[vii] UN data show that between 2007 and 2019, some 25,733 Afghan civilians died at the hands of anti-government elements,[viii] with many more suffering injuries of a kind that could seriously blight people’s lives in a country with an underdeveloped health system. Such attacks, combined with the inability of the instrumentalities of the new state to protect against them, have contributed to an atmosphere of fear in diverse parts of Afghanistan, not least because they evoke memories of earlier periods of conflict in the 1980s that cost vast numbers of lives.[ix] The scale of civilian casualties at the hands of anti-government elements is a useful reminder that the persecution which refugees fear is not necessarily from state actors; indeed, the 1951 Convention contains no requirement that states be involved. In Afghanistan at present, ordinary people are gravely at risk of attack by actors such as the Taliban, and by the so-called ‘Islamic State’. Two groups are at particular risk: members of the Hazara ethnic minority; and members of the younger globalised generation that has developed in the last two decades, providing key civil society activists supporting democracy, human rights, and progress for women. These two groups overlap but are not identical.

Hazaras have historically experienced significant persecution within Afghanistan,[x] and are vulnerable not only because their East Asian phenotype makes them physically somewhat distinctive, but because the majority of Hazaras are Shiite Muslims, which can lead to their being targeted by Sunni Muslim extremists. In 1998, the Taliban carried out a massacre of Hazaras in Mazar-e Sharif that one informed observer described as ‘genocidal in its ferocity’.[xi] This was unfortunately not an isolated event. In August 2017, there was a major massacre of Hazaras at Mirza Olang in Sar-e Pul.[xii] From late October 2018, Taliban forces undertook coordinated attacks against Hazaras in Khas Uruzgan, Malestan and Jaghori. Many Hazara asylum seekers in western countries originate from these districts. The districts are, however, of no military significance, and the attacks made more sense as a symbolic strike designed to highlight the inability of the Afghan state effectively to protect members of a vulnerable ethnic and sectarian minority, and as punishment for the relatively tolerant and liberal lifestyle of these communities, far removed from the puritanical extremism of the Taliban.[xiii] The dangers faced by Hazaras were again highlighted on 24 November 2020 when a major bombing occurred in the town of Bamiyan,[xiv] where the Taliban in 2001 had deliberately destroyed two famous statues of the Buddha that were amongst the most precious parts of Afghanistan’s rich archaeological heritage.[xv] It is important to note that Hazaras are not targeted only by the Taliban. Islamic State, notoriously hostile to Shia,[xvi] has been responsible for a substantial number of mass-casualty attacks. On 23 July 2016, a peaceful demonstration by Hazaras associated with the so-called ‘Enlightening Movement’ (Jumbesh-e Roshnayi) over the routing of a proposed electricity system was struck by a suicide bombing. Some 85 people were left dead, and 413 injured.[xvii] ISIS claimed responsibility for what it called ‘a “martyrdom attack” on Shiites’.[xviii] And on 17 August 2019, when this writer was in Kabul, a suicide bomber linked to Islamic State attacked a wedding reception in the city, killing a large number of guests, including 14 members of the bride’s family.[xix]

It is also important to note that politics infuses many areas of social life that outside observers unfamiliar with Afghanistan’s complexities might see as apolitical, such as ‘land disputes’.[xx] For example, in 2010, Hazaras’ homes in Wardak were torched by ethnic Pushtun nomads (kuchis) demanding access to lands the Hazaras occupied. In a careful study, Fabrizio Foschini reported eyewitness testimony that the kuchis arrived in Daimirdad district ‘en masse, maybe a thousand of them, a hundred riding horses, thirty to forty on motorcycles, others in pickups’ and that when they reached Tezak valley at 5.20 am, the first thing they did was burn the houses, at which point ‘the whole population evacuated Daimirdad’. Foschini observed that ‘it seems clear that Kuchis never owned, or even seriously claimed to own, land in the district or utilized its territory as pasture’.[xxi] Disputes over ‘land’, or over other resources, are very often really disputes over power and who can exercise it.

Just as alarming have been targeted killings of civil society actors, searingly detailed in a recent United Nations report.[xxii] In the last two decades, civil society has flourished in Afghanistan, which has some of the freest media in its region. A number of the most prominent activists have studied abroad as scholarship holders in Europe, the United States, Japan and Australia, and have made use of the skills that they acquired abroad in promoting modernist values. As a result of this, however, they have recently been targeted with great ferocity, not least because their modernism has been depicted by some observers as a bulwark against the anti-modernist values and ideology of the Taliban. This globalised generation in a real sense holds the key to a positive future for the country, and that has made them acutely vulnerable. Under such circumstances, no matter how strong their commitments, at a certain point the need to survive may force them into exile. If the Taliban were to recover significant state power, the refugee outflows from Afghanistan could dwarf any that have been seen in recent decades, and so grave would be the threat facing those in flight that attempts to deter their movement by punitive measures or the threat of deportation would likely be pointless, since large numbers would indisputably be refugees as defined in the 1951 Convention, or persons in need of complementary protection. As Jane McAdam has put it, ‘In legal terms, “complementary protection” describes protection granted by States on the basis of an international protection need outside the 1951 Convention framework. It may be based on a human rights treaty or on more general humanitarian principles, such as providing assistance to persons fleeing from generalized violence.’[xxiii]

Of course, some refugees from Afghanistan have spent some time in neighbouring states before venturing further afield. Unfortunately, it is often the case that countries adjoining disrupted state have problems of their own that can make them inhospitable to refugees. To start with, countries of ‘first asylum’ may not be parties to the 1951 Convention, something which can severely limit the protections to which refugees have access. Beyond this, the characteristics that may have marked refugees for persecution in their own countries may also be a source of danger in the countries to which they flee: Hazaras in Pakistan, especially in the vicinity of Quetta, have experienced significant sectarian violence in recent years in which Hazara refugees from Afghanistan have found themselves caught.[xxiv] Afghan refugees in Iran have found themselves press-ganged into militias being deployed for Iran’s own nefarious purposes.[xxv] And Afghan refugees who have sought to escape these dismal settings have often found themselves the target for abuse or marginalisation in countries beyond Afghanistan’s immediate neighbourhood.[xxvi] Those who are inclined to see refugee movements as largely the product of ‘pull factors’ would do well to reflect on the potency of the ‘push factors’ that are at play.

Is peace in the offing?

On 29 February 2020, the Taliban and the US Administration of President Donald J. Trump signed an ‘Agreement for Bringing Peace to Afghanistan’ (Mowafeqatnamah-e awardan-e saleh be Afghanistan).[xxvii] This agreement contained no provision for a permanent and comprehensive ceasefire in the country, merely identifying such a ceasefire as an agenda item for future discussion in what the agreement called ‘intra-Afghan negotiations with Afghan sides’ (para.4). The text provided for a reduction of US troop numbers within 135 days to 8,600, to be followed by ‘withdrawal of all remaining forces from Afghanistan within the remaining nine and a half (9.5) months’ (Part One, A(1), B(1)). This provision was not made conditional on any progress being achieved in ‘intra-Afghan negotiations’, or on any commitment by the Taliban to protect human rights or respect democratic processes. It was conditioned solely on the Taliban honouring Part Two of the Agreement, which dealt only with preventing the use of ‘the soil of Afghanistan to threaten the security of the United States and its allies’. The agreement further provided that up to five thousand Taliban combat and political prisoners held by the Afghan government ‘would be released by March 10, 2020’, with ‘the goal of releasing all the remaining prisoners over the course of the subsequent three months’ (Part One, C). This provision, perhaps naively intended as a ‘confidence-building measure’, was quite remarkable given that the Afghan Government was not a party to the 29 February agreement, and had been excluded from the discussions that led to its negotiation. There was some suggestion that ‘secret annexes’ might have added to the obligations borne by the Taliban, but this was undermined by an explicit statement by US Secretary of State Pompeo on 1 March on CBS Television describing the 29 February text as ‘the complete agreement’. As a result, while the agreement was touted as an element of a ‘peace process’, it was better seen as a ‘withdrawal agreement’ rather than a ‘peace agreement’.

Far from setting Afghanistan on a path to peace, the Agreement instead exposed the dangers involved in negotiating with morally-repugnant actors.[xxviii] Very serious problems of implementation swiftly surfaced. The most obvious related to the prisoner-release issue, which delayed the commencement of discussions between ‘Afghan sides’ for more than six months. The main problem that flowed from prisoner releases, carried out with gritted teeth by the Afghan Government as a result of intense pressure from Washington, was not simply the return to combat activity of released Taliban, alarming as that was; it was even more that the disposition of the US consistently to give in to the Taliban’s interpretation of the agreement sent the signal that the US was an unskilled and enfeebled player that could be taken for granted. In this context, the fierce escalation of attacks against civil society activists, was not remotely surprising. The 29 February prohibition on attacks on US forces virtually invited Taliban attacks on Afghan targets, not only to eliminate potential future opposition, but to create an atmosphere of dismay that might lead even longstanding opponents of the Taliban to consider shifting their loyalties for prudential reasons, as it rarely pays to be on the losing side in Afghanistan.

By 2021, with the advent of the Biden Administration, the upsurge in ferocious violence in many parts of Afghanistan made it increasingly clear that the ‘peace process’ had exhausted any potential it ever had to deliver peace or security. It had actually begun to veer off course even before the signing of the 29 February agreement, at the point in 2019 where the US abandoned a (defensible) ‘nothing is agreed until everything is agreed’ formula[xxix] in favour of a two-stage process which gave the Taliban everything they really wanted – status, prisoner releases, and a US withdrawal timetable – at the first stage, creating no incentive for them to make any concessions during intra-Afghan negotiations. The upshot was strategic stalling on the Taliban’s part, and the illusion of active diplomacy. This left the incoming Biden Administration with horrible dilemmas and no easy options.[xxx] It is not surprising that the prospect of being returned to an Afghanistan in meltdown would fill asylum seekers with fear, and a well-founded fear at that.

Dangers of deportation

A recent study has noted that the ‘deterrent effect of deportation is treated as a given (in spite of a complete absence of any research data to that effect) …’.[xxxi] As long as this conviction – spurious though it may be – remains entrenched, those concerned with the welfare of individual asylum claimants will have their work cut out for them, since the impetus on the part of bureaucratic agencies to deport even in the most discouraging of circumstances is likely to persist. Nonetheless, advocacy is of fundamental importance, not least in highlighting the genuine dangers confronting those targeted for deportation. It is well documented that forced returnees are likely to face severe problems of reintegration, not least because they risk being stigmatised, especially if their initial escape was funded either by loans or by the pooling of resources within lineages to assist the exit of the most vulnerable.[xxxii]

Furthermore, livelihood opportunities can prove highly problematic: serious research in this area highlights the importance of social relations. A study by Kantor and Pain emphasises the centrality of relationships to livelihoods in rural Afghanistan,[xxxiii] and the points they make apply equally to urban areas. The mere fact that there may be people of similar ethnic background living in a potential relocation destination does not overcome this problem, since ethnic identities do not in and of themselves give rise to the ties of personal affinity and reciprocity that arise from family connections. Indeed, one mistake that observers — even Afghan observers — on occasion make is to underestimate the degree of differentiation amongst groups such as the Hazaras, including distinctions between elite and non-elite figures, distinctions based on district of origin and lineage, and distinctions based on values and ideology. A young Hazara male who is returned to a region in which he lacks strong social connections is likely to end up destitute, or be exposed to gross exploitation or criminal predation. Afghans who have been in Europe for more than brief periods may be viewed as gharbzadeh (‘westernised’), and even subtle changes in body language of which they are quite unaware may make them stand out. This can expose them to attack on either ideological or criminal grounds, adding to the dangers that they may already face by virtue of ethnic or social identity. Young Afghans who have spent formative years in a country such as Sweden, which in social terms is radically different from Afghanistan, could find themselves vulnerable as a result.

One final point. It is essential to appreciate that the situation in Afghanistan is extraordinarily fluid. Assessments of the situation made even quite recently do not necessarily provide an accurate picture of a rapidly-evolving situation. ‘Country of Origin’ information reports of the kind that often inform bureaucratic decision-making can become outdated very quickly, since in the absence of strong political and security institutions, even an apparently stable environment can deteriorate sharply and suddenly. No place in Afghanistan can be considered safe.[xxxiv]

More in English


[i]E.F. Kunz, ‘The Refugee in Flight: Kinetic Models and Forms of Displacement’, International Migration Review, vol.7, no.2, Summer 1973, pp.125-146.

[ii]See Antonio Giustozzi with Niamatullah Ibrahimi, Thirty Years of Conflict: Drivers of Anti-Government Mobilisation in Afghanistan 1978-2011 (Kabul: Afghanistan Research and Evaluation Unit, January 2012).

[iii]Gil Loescher, Beyond Charity: International Cooperation and the Global Refugee Crisis (New York: Oxford University Press, 1993) p.17.

[iv]See William Maley, What is a Refugee? (New York: Oxford University Press, 2016) p.8.

[v]See Kate Pincock, Alexander Betts and Evan Easton-Calabria, The Global Governed? Refugees as Providers of Protection and Assistance (Cambridge: Cambridge University Press, 2020).

[vi]Global Peace Index 2020: Measuring Peace in a Complex World (Sydney: Institute for Economics and Peace, 2020).

[vii]See Shaun Gregory, ‘The ISI and the War on Terrorism’, Studies in Conflict and Terrorism, vol.30, no.12, 2007, pp.1013–1031; Matt Waldman, The Sun in the Sky: The Relationship Between Pakistan’s ISI and Afghan Insurgents (London: Discussion Paper no.18, Crisis States Research Centre, London School of Economics and Political Science, June 2010)

[viii]See William Maley, The Afghanistan Wars (London: Red Globe Press/Macmillan, 2021) pp.241, 274.

[ix]See Noor Ahmad Khalidi, ‘Afghanistan: Demographic Consequences of War, 1978-1987’, Central Asian Survey, vol.10, no.3, 1991, pp.101-126.

[x]See Niamatullah Ibrahimi, The Hazaras and the Afghan State: Rebellion, Exclusion and the Struggle for Recognition (London: Hurst & Co., 2017).

[xi]Ahmed Rashid, Taliban: Militant Islam, Oil and Fundamentalism in Central Asia (New Haven: Yale University Press, 2000) p.73.

[xii]Special Report: Attacks in Mirza Olang, Sari Pul Province: 3-5 August 2017 (Kabul: United Nations Assistance Mission in Afghanistan, August 2017).

[xiii]See Rod Nordland, ‘Bodies Pile Up as Taliban Overrun Afghan Haven’, The New York Times, 13 November 2018.

[xiv] Thomas Gibbons-Neff and Nick Cumming-Bruce, ‘Donors Pledge Less Aid to Afghanistan During a Violent Chapter’, The New York Times, 24 November 2020.

[xv]See Joseph and Ria Hackin, Bamian: Führer zu den Buddhistichen Höhlenklöstern und Kolossalstatuen (Paris: Les Éditions d’Art et d’Histoire, 1939); Pierre Centlivres, Les Bouddhas d’Afghanistan (Lausanne: Éditions Favre, 2001); Llewelyn Morgan, The Buddhas of Bamiyan (London: Profile Books, 2012).

[xvi]Alissa J. Rubin, ‘Questions Rebels Use to Tell Sunni from Shiite’, The New York Times, 24 June 2014.

[xvii]‘UN Chief in Afghanistan renews Call for Parties to Protect Civilians — UNAMA Releases Civilian Casualty Data for Third Quarter of 2016’ (Kabul: UNAMA, 19 October 2016) p.2.

[xviii]Mujib Mashal and Zahra Nader, ‘ISIS Claims Suicide Bombing of Protest in Kabul, Killing at Least 80’, The New York Times, 24 July 2016.

[xix]Mujib Mashal, ’63 Killed as Explosion Turns Kabul Wedding Into Carnage’, The New York Times, 17 August 2019.

[xx]See Andrea Chiovenda and Melissa Chiovenda, ‘The spectre of the “arrivant”: hauntology of an interethnic conflict in Afghanistan’, Asian Anthropology, vol.17, no.3, 2018, pp.165-184 at pp.173-174.

[xxi]Fabrizio Foschini, The Kuchi-Hazara Conflict, Again (Kabul: Afghanistan Analysts Network, 27 May 2010).

[xxii]Special Report: Killing of Human Rights Defenders, Journalists and Media Workers in Afghanistan 2018-2021 (Kabul: UNAMA, 15 February 2021).

[xxiii]Jane McAdam, Complementary Protection in International Refugee Law (Oxford: Oxford University Press, 2007) p.21.

[xxiv]See, for example, Salman Masood, ‘Armed Attackers Kidnap and Kill 11 Miners in Pakistan’, The New York Times, 3 January 2021.

[xxv]See Ahmad Shuja Jamal, The Fatemiyoun Army: Reintegration into Afghan Society (Washington DC: Special Report no.443, United States Institute of Peace, March 2019).

[xxvi]Christine Roehrs and Khadija Hossaini, Afghan Exodus: Migrant in Turkey left to fend for themselves (Kabul: Afghanistan Analysts Network, 22 December 2020).

[xxvii]For further and more detailed discussion of negotiation with the Taliban, see William Maley, Transition in Afghanistan: Hope, Despair and the Limits of Statebuilding (London: Routledge, 2018) pp.222-239; Niamatullah Ibrahimi and William Maley, Afghanistan: Politics and Economics in a Globalising State (London; Routledge, 2020) pp.162-164; William Maley, ‘Afghanistan: elite tensions, peace negotiations, and the COVID crisis’, Acta Via Serica, vol.5, no.2, December 2020, pp.1-24; and Maley, The Afghanistan Wars, pp.286-291.

[xxviii]See William Maley, Diplomacy, Communication and Peace: Selected Essays (London: Routledge, 2021) pp.236-251. On the Taliban movement as a terrorist network, see William Maley, ‘Terrorism and insurgency in Afghanistan’, in M. Raymond Izarali and Dalbir Ahlawat (eds), Terrorism, Security and Development in South Asia: National, Regional and Global Implications (London: Routledge, 2021) pp.140-156 at pp.148-150.

[xxix]See Mujib Mashal, ‘Confusion over Afghan-Taliban talks further complicates peace process’, The New York Times, 27 July 2019.

[xxx]See Afghanistan Study Group Final Report (Washington DC: United States Institute of Peace, February 2021).

[xxxi]Nassim Majidi and Liza Schuster, ‘Deportation and forced return’, in Alice Bloch and Giorgia Donà (eds), Forced Migration: Current Issues and Debates (London: Routledge, 2018) pp.88-105 at p.94.

[xxxii]See Liza Schuster and Nassim Majidi, ‘What happens post-deportation? The experience of deported Afghans’, Migration Studies, vol.1, no.2, 2013, pp.1-19; Liza Schuster and Nassim Majidi, ‘Deportation Stigma and Re-Migration’, Journal of Ethnic and Migration Studies, vol.41, no.4, 2015, pp.635-652.

[xxxiii]Paula Kantor and Adam Pain, Securing Life and Livelihoods in Afghanistan: The Role of Social Relationships (Kabul: Afghanistan Research and Evaluation Unit, December 2010).

[xxxiv]See “No Safe Place”: Insurgent Attacks on Civilians in Afghanistan (New York: Human Rights Watch, 8 May 2018).